Mechanical Machine Components • 3 Couplings, Clutches and Brakes
the thermal capacity of the clutch, whereas Qpc~ are mainly dependent on cooling and heat emission. Values acquired empirically or by means of costly mathematical processing for QE and Qpe~ can be shown as characteristic curves for specific clutches (Fig_ 16). This shows the permissible engagement force Qpe~ (per engaging action) as a function of the engagement frequency Sh' The transitional engagement frequency ShU produces a characteristic value on the curve and is specified by the clutch manufacturer. Using the parameters QE and Shli' the pennissible thermal stress Qpe~ can be specified as a function of the engagement frequency Sh 135.41] ( 4) ~.4.4
Layout Design of Friction Clutches 41,47,48]
(5)
This makes it possible to specify iterdtively the required contact pressure F, the number of frictional surfaces z and the required mean radius of the frictional surfaces rm = (R + 1')/2 (cf. Fig. 14). If, for example, the diameter of the clutch should be small then the number of frictional surfaces or the contact pressure (for maximum permissible contact pressure cf. Appendix F~, Table 1) can be increased. In this way a variety of clutch types can be designed. In Fig. IS the reaction delay t" must be noted at engagement time t w , = til + t ,. To simplify layout design buildup time, tl2 can be disregarded when calculating the slip period t, (cf. Eq. (2»
=A
(woo - w zo) il[K -
ML
.
The surface engagement force at a single engagement qA is then detennined as follows: (8)
where A R• denotes the entire frictional surface of the clutch (9)
moment to be transmitted and the engagement force that produces it, and so the thermal stress is usually the deciding factor for selecting the appropriate size [49, SO]. The moment to be transmitted is governed by the nominal moment of the drive motor and operating machine where allowance must be made for cyclic variations (e.g. piston engines) or the moment of tilt (2 to 3 M N ) in squirrel-cage motors. The engaging and disengaging torque of a clutch M, > M, is generally smaller than the transmittable moment M" which is set up in frictional surfaces which are relatively static in relation to each other. In particular, for wet clutches the sliding friction coefficient I-' is smaller than the static friction coeffiCient 1-'0 (cf. Appendix F~, Table 1). For the practical design of a friction clutch the specified moment MK in Eq. (5) is used:
5
(7)
(~7,
A clutch is basically designed according to the maximum
t
Engagement force Q can be determined as follows:
(6)
Permissible engagement force log 0
OE f--~=---'"
The surface engagement force qA can be compared with the permissible surface engagement force at a single engagement qAE: qA < qAE (cf. Appendix F~, Table 1). It is also possible to make a comparison between the actual and the permissible surface frictional force (qA and qAO) (cf. Appendix F~, Table 1):
where PR denotes the frictional surface pressure, v, the running speed and I-' the sliding friction coefficient. ~.4.S
Size Selection of Friction Clutches
If a dutch to be purchased is being designed for a specific
application the exact requirements must first be specified. Reference [36] provides a questionnaire to assist in the selection of clutches. The required characteristic moment M K of the clutch can be estimated on the basis of the load moment Me. the (reduced) mass moment of inertia]" the angular velocity difference L\.w, the approximate slip period t, and buildup time t" required (cf. F3.4.3, Eq. (l». Using this value a specific clutch is sought from a selected manufacturer. The buildup time t 12 should be apparent from the catalogue. It is then possible, by using Eq. (2) and (3), to accurately determine the slip period t, and the engagement force Q. If a drop in the speed of the drive motor and the mass moment of inertia (with gears) is to be allowed for when the clutch is engaged, then more detailed literature [37, 41 J must be obtained. The calculated engagement force Q can be compared with the permissible values QE (catalogue) for the clutch selected. A larger clutch must be selected when Q > QE' In the case of frequent engaging and disengaging the pennissible engagement force must be determined using Eq. (4) and compared with the actual engagement force. This is only possible if ShU is known for a specific clutch. If only characteristic curves are provided as in Fig. 16, they also can be used to specify the clutch required (cf. [35]).
0.6320E ~.4.6
Selection Criteria (6, S I]
A selection of the types of clutch commercially available
ShU
Frequency of engagement
Figure 16. Pennissible engagement force as in Eq. (4) as a function of the frequency of engagement l35 J
is shown in Fig. 17. A characteristic moment MK = 500 Nm and a speed of n = 1500 min 'were used as a basis for comparison. The chart therefore enables a comparison to be made between the diameter and length of the clutches and between the permissible engagement force QE at a single engagement and the limiting value QEShO at vety high engagement frequency. With the exception of the commercial vehicle clutch (opening clutch)
3.4 Clutches. 3.4.7 Brakes
'"<.>
~
ii!
AI
g'K
~~
l! ~
!.! '-
. "'"-5 al-s
[ij 'l:l
~
.~
jl
~!.!
~ ~
.,,= ~~ ~~ wu
-5~
-s'tl
~ U
~~ AI'" .DE
~
-<.>;-'"
g~
~ DI)'-rur'Vling
Urnllng value of hourly ergagemem fOlce
~. ~
Methods of Operation and Actuating Systems, Properties Single-Surface Clutches. Dry-running friction pairings are
preferred so as not to obtain too great a diameter at a given torque. A closed axial magnetic flux within the clutch is only possible through electromagnetic actuation; rapid reaction with short lifts; low idling moment. Single-, Double-Disc Clutches. Again dry running for
larger torques; all methods of actuation can be used , but hydraulic actuation is usually avoided owing to the danger of leakage (friction linings are oil lubricated); efficient cooling (cooling ribs), rapid reaction , low idling moment, relatively chatter-free (materials with decreasing ILl v, characteristics) . Multiple-Disc Clutches 142, 43 J. Small size even with
large torques, engagement under load, effective cooling but only possible with oil circulation, i.e. wet running; all methods of actuation possible. In the case of magnetic flux controlled discs (electromagnetic actuation) only specific friction pairings can be selected. Rapid reaction can be achieved in wet running by means of thin oil, oil mist or grooves in the discs; low idling moment achieved by corrugated discs. Longer useful life, i.e. little wear in wet running. Cone Clutcb. Suitable for high torques and engagement forces in dry running, actuation usually mechanical or pneumatic.
al
0;
E
-l~
Si "'-5
~:g
"0
'" .~ -s'"
~3 <.> .,
{}<.>
~~ <.>-.;
o~
.~ .~
it
:;I~ "9~
~1
Wel-runnong (Inlernal 011cooling)
al vel)' hlg/1 frequency 01 engagemenl (5,,-0:)
Figure 17. Comparison of types of friction clutches for a characteristic moment MK
they are closing clutches. Also shown in Appendix F3, Fig. 2, is the engaging moment region of various externally actuated friction clutches.
:'i' '5
=
S()() Nm and a speed n
=
I :;00 min ' 1361 .
3.4.7 Brakes Brakes are clutcbes with an idle driven member and 100% slip. Like the clutches there are mechanical, hydraulic, pneumatic and electric brakes (cf. Fig. 2). According to their function they can be classified as locking brakes, stop and regulating brakes as well as dynamometric brakes 137J. Various types of brake are sbown in Fig. 18. The calculations for mechanical, engaging and disengaging friction brakes are carried out much like clutch calculations: characteristic moment M K is replaced by braking moment and accelerating torque M, by decelerating torque. Principles of calculation for drum and disc brakes are contained in Part 1 of DIN 15434.
Types. In principle all rypes of clutches are also available as brakes (cf. Figs 17, 13 and II). Shoe brakes can be divided into internal (Fig. Ilk) and external shoe brakes (Fig. ISb) (vehicles, boists). Band brakes (Fig. ISa) require only low operating forces on account of the selfenergising effect of contact friction. In addition to the type shown in Fig. ISa there are also band brdkes with mUltiple wrdPping as well as internally acting band brakes. Disc brakes (Fig. 18d) have favourable cooling characteristics, especially if they are manufactured with internal air supply. Hydraulic (water, oil) and electric brakes (generators) which easily dissipate the resulting energy are non-wearing dynamometric brakes. Induction brakes (Fig. 18e), by means of a static live coil in the rotor, induce current which causes a force to act counter to the direction of rotation. In these eddy current brakes the braking moment is to a large extent governed by rotational speed [52].
Mechanical Machine Components. 3 Couplings, Clutches and Brakes
a
d
Figure 18. Types of brake (operating force FB partially shown): a band brake [371, b external shoe brake (double) [371, c internal shoe brake (drum brake, simplex), d pneumatically operated disc brakes (Ortinghaus), e induction brakes with fan wheel (Stromag).
e
3.5 Automatic Clutches 3.S.1 Torque-Sensitive Clutches (Slip Clutches) These are safety clutches which protect machinery from damage as they do not exceed a predetermined torque. In this way the unnecessary oversizing of machinery to meet peak moments can also be avoided [53].
Types. In principle, all friction clutches with fixed clutch force can be used as slip clutches. It is important that the normal force should not alter signillcantly with the wear pattern (even spring characteristic) and the clutches should be monitored for permanent slip. Block clutches usually have spring-loaded tapers or pins which disengage when the torque limit is reached. With overload release snap pin and snap ring clutches [54] it must be borne in mind that the ultimate moment may be greatly diffused unless special measures are adopted. Both slip and block clutches may be used to disconnect electromechanical or electric switches in order to switch off the drive motor. 3.S.2: Speed-Sensitive Clutches (Centrifugal Clutches) These are clutches which allow smooth starting so that electric motors or combustion engines accelerate first of all and only then drive the machine. Starting clutches make it possible to design for a reduction in motor size or even electricity supply for machines with a high second moment of inertia or load moment.
Types (Fig. 19). Centrifugal clutches [55] with segments (Fig. 19a) [9] if provided with retaining spring will transmit a moment only after a specillc speed has been reached. Sprag clutches (Fig. 19b) [9] throw powder, balls or rollers against the casing of the driven member by means of a star-shaped rotor so that the moment to be transmitted increases quadratically with the speed of the drive motor. At nominal speed these clutches, unlike hydrodynamic clutches [56], are free of slip and leakage. Starting in an asynchronous motor (characteristic curve
a
Figure 19. Speed-sensitive couplings (see text).
MM in Fig. 2:0) when using a sprag clutch is practically load-free (only M K ) and the machine is at rest up to the intersection I of the clutch characteristic MK and the load characteristic M L . The motor remains at point 2 and accel·
Speed Figure ZO. Characteristic curves of an asynchronous motor MM and a centrifugal dutch M K ; load moment Ml .
3.5 Automatic Clutches. 3.5.3 Directional (One-Way) Clutches, Overrun Clutches
erates the machine up to synchronous running. Subsequently all the pans reach the operating speed at point 3. A disadvantage ofthese clutches compared with springloaded slip clutches is that in practice they only operate on quick-running shafts.
:t.s.:t Directional (One-Way) Clutches, Overrun Clutches The engaging procedure depends on the direction of the relative rotary motion between the driving and driven member: it is prevented in one direction of the relative rotation (locked condition) but not in the other direction (freewheeling condition). Freewheel clutches have the following functions (they cannot usually be distinguished by structural shape) [9, 57, 58]: return stop (for conveyor belts, pumps, automatic gearboxes for motor vehicles, fans); overrun clutch (for multi-motor drives, statting motor drives, bicycle hubs); step-by-step freewheel (for shaping machines, feed mechanisms, ratchet mechanisms).
Types. For simple tasks: Ratchet freewheels (ratchet wheels, ratchet drills) engage the drive positively in one direction of rotation. There are also friction-driven, noiseless ratchets. Grip freewheels [4, 59]' on the other hand, grip noiselessly at each location at greater clutch velocities and with smaller dimensions. Grip roller freewheels (Fig. ~ 1) are often provided with internal spiders with which
Figure :11. Grip roller freewheel with internal spider and individual cushioning.
individual cushioned rollers are pressed into the wedgeshaped pockets. Clamp freewheels [9, 60, 61], which are the same size, transmit more torque but are less robust. They have out-of-centre clamping jaws between circular cylindrical slideways. Wear-reducing additives in the lubricant [63] have the greatest influence on useful life [62] and accuracy of engagement. Wear can be reduced in return stops by centrifugal lift [9]. It is essential to have a perfect radial and axial bearing (there are assemblies with rolling bearings) [4, 9]. Clutches can also be actuated externally: disengaging (full freewheeling condition), re-engaging, full locking, engaging only during a revolution (one-stop clutch [64]). Friction freewheels are friction clutches (discs, cones) which are forced in one direction along a sharp thread. If helical gear clutches are used to transmit torque these are toothed freewheels.
Rolling Bearings H. Peeken, Aachen
Rolling bearings are ready-to-fit machine parts. They consist of rolling elements running on inner and outer race rings and a cage, which keeps the rolling elements apart.
4.1 Fundamentals 4.1.1 Material Stresses and Fatigue in Rolling Contact In rolling contact under load, the effect of "flattening" results in a contact surface the size and stress of which can be calculated according to the Henzian equations. Henzian theory applies to solid and isotropic bodies with material showing elastic behaviour. The bearing surface arising in rolling contact is assumed to be plane and small relative to the dimensions of the elements (for a detailed description of the Henzian equations, see B4). To calculate material stress on the basis of Henzian compression, hypotheses are used for main shear stress, defonnation energy and alternating (onhogonal) shear stress. In Figs 1 and ~ the comparative stresses CTv in the three hypotheses are shown, relative to the maximum Henzian compression Po, for line contact. Consequently the maximum material stress value occurs under the tangential plane. Minimal differences arise in its absolute magnitude and depth. Structural changes in the rolling bearing material, e.g. plastic deformation (shearing strains) or the so-called butterflies, which occur below an angle of apprOximately 45°, indicate that the process of fatigue (fonnation of incipient crack points, cracking,
crack growth, crumbling of material panicles (paring and pitting when lubricated)) is initiated at points of material inhomogeneity owing to the shear stress. In the case of line contact (Fig. ~) the greatest shear stress Tm~ = O.304po occurs at a distance of O.7Bh from the surface at the point x = 0 (where b is half the width of the rectangular bearing-surface area); point contact Tm~ = O.3Ipo, distance 0.47b. With surfaces which are permanently subject to rolling action the pulsating shear stress may be regarded as the dynamic load limit. Since the maximum shear stress is proponional to the Henzian compression, its calculation is sufficient for the evaluation of the state of stress. With mixed friction, as well as with liquid friction, tangential stresses arise from bearing friction in addition to normal load in the contact zone. This results in an increase in the stress maximum migrating to the surface (Fig. :t). 4.1.~
Load Distribution
The load distribution in a loaded rolling bearing is dependent upon the elastic deformations at the contact points of the individual rollers. The calculation of this load distribution and the maximum rolling contact load exen a decisive influence on the determination of the bearing's capacity rating C. Figure 4 shows as an example a loaded single-row inclined ball bearing. The rolling contact forces act in the direction of the pressure angle lX, while the load component F, forms angle f3 with F. If f3 does not exceed a particular magnitude, only part of the race is under load. The load per rolling contact is determined by the elastic deformations at the contact points. According to the Henzian equations (see B4) for point contact Q~/Q= = (ll~/lJm~)'/2; Q~ is the rolling contact load at the point "", Q= the maximum rolling contact load, IJ~
Mechanical Machine Components • 4 Rolling Bearings
F
Po
x
01V ~ ~ ~
0
.:::::
:--...
-+---0-1---- 0.S7
t'4S0, max
~~
)) ) )\ \ "DiY II I \ \ 1.0 ' " ~.~ / ) \ ~ 1.5 / :~0.50 _i / / 2.0 \ 0.5
1060
0.30
-s:55
0.40
2.5
II
I
O'~H 1PO
a
3.0
o
N C ~7 :\ \ ............ )) \'\
I
1.0
I
0.55
\'" '-.!I
45
./
'"
1
2.5 3.0 b
o
1.0
~ 1.5 2.0 2.5
/'
'" ~~ ~,~~/) ./
~
_0.10
/
/
(
015
/
/.
0.10
V
\!--- V
II
I~
V/
-r-
\0.30 0.10
/ /
/
J
Wo. 250
1\
\
05 lO
O'GEH/p I V 0
~~
."
"-
'"
""
\I~~~~ -0.10_ ~~ '--0.15 \1\ 1\ \ "'- I--'I/ \
\
3.0 -2.5 -2.0 -1.5 -1.0 -0.5
x
-':
~
0.40
2.0
.:::-:=::
0.40
l/;I 'r
~
0.5
Q5
Figure 2. Shear stresses under the surface according to the main shear stress and alternating shear stress hypotheses for line contact with Hertzian compression [1].
0.5
O'~SHlpo 1.0
1.5
2.0
1.5 2.0 Z.5
10 -2.5 -2.0 Figure~.
Material stress with line contact, nonnal and tangential
load 121.
2.5
xlb Figure 1. Undimensioned comparative stresses (Tv/Po [IJ; a main shear stress hypothesis, b deformation energy hypothesis, c alternating shear stress hypothesis.
the displacement of the bodies (rolling elements) at point
(e.g. a single-row tapered roller bearing) the load distribution follows Q,/ Qma> = (8,/ 8ma» 108 For e = 0.5 the maximum rolling element load Qma> = 4.06F,/(z cos a). With a = 0° these equations can also be used for singlerow ball and roller bearings that have no play. Together with the material characteristics, information can be derived from them about the load capacities (cf. F4.3.1 and F4.3.2).
+i'" ~
'2 F
Yo'
IlV COS (t
'I'
arna. COS a
Figure 4. Load distribution in single-row inclined ball bearing: a pressure angle, d L race diameter, Fa axial force, Fr radial force, f3 angle of direction of bearing load F, Q.p rolling element load, 1/1 position angle of rolling element, Qmax maximum rolling element load, ed,_extent of rolling race load.
4.1.~
Designations of (Standard) RoDing Bearings
Rolling bearings are designated in accordance with DIN 623 Part 1 by symbolS consisting of prefixes, base symbols and suffixes. Parts of complete rolling bearings are designated by prefixes, e.g. K cage with rolling con-
4.2 Types of Rolling Bearings. 4.2.1 Ball Bearings
Table 1. Base symbol for rolling bearings
Radial and tangential and partly also axial fixing of the bearing. Ease of installation and removal.
Bearing series Symbol for bearing bore See DIN 623
Size series Type of bearing See DIN 623
Width or height series
Diameter series
Sec DIN 616
tacts, L free race, R race with rolling elemems, S stainless steeL The base symbol designates the type and size of the bearing. It consists of two symbols or groups of symbols (Table 1). The dimensions (bore d, external diameter D, width B, edge clearances r 1mln , rZmin ) of rolling bearings are arranged in such a way that several widths and external diameters are allocated to each bearing bore in order to cover a large load range (DIN 616; ISO 104). For radial bearings the grading is by width series (7, 8, 9, 0, 1, 2, 3, 4, 5, 6) and diameter series (7, 8, 9, 0, 1, 2, 3, 4, 5). By combining the two figures (B before D!) the size series is formed (Fig. ;). In addition, size plans are applicable to tapered roller bearings and axial bearings (height series 7,9, 1,2; diameter series 0, 1,2,3,4, 5). For bore diameters of 20 to 480 mm the bore number is given. Except for bearing sizes up to d = 17 mm, bore d in mm is obtained by multiplying the bore number by 5. For example the base symbol 6204 means: grooved ball bearing, single row (bearing series 62), size series 02 (width series 0, diameter series 2), bore d = 5.04 = 20 mm from the width series OB = 14 mm and from the diameter series 2D = 47 mm. In the case of bore diameters below 20 and above 480 mm the bore code number is replaced by the millimetre figure (in part separated by a slash). For tapered roller bearings DIN ISO 355 lays down a new identification. The base identification begins with T for tapered
roller bearing; then follows for contact angle " the angle series (2, 3, 4, 5, 7), the diameter series (B, C, D, E, F, G), the width series (B, C, D, E) and the three-digit hore diameter in mm. The suffixes are used to designate the internal construction, external form, cage design, accuracy, hearing clearance and heat treatment, For more details see DIN 623 Part 1.
4.1.4 Fit and Bearing Clearance The following aspects are of importance in the selection of fit: Support of the hearing rings at their periphery to maintain the full load capacity of the bearing,
o-
w/·
~~
-.trim;, 1
Cr.
8
~
~;;1 -.,L..
=
C ~
, Rl
4). C1 radial bearing clearance smaller than C2, C2 radial bearing clearance smaller than normal (CO), CO normal radial hearing clearance, C3 radial hearing clearance greater than normal (CO), C4 radial hearing clearance greater than C3, C5 radial bearing clearance greater than C4.
4.2 Types of Rolling Bearings 4.2.1 Ball Bearings Types of Construction. For predominantly radial loads: Fig. 6a-g. (a) Single-row grooved ball bearings (DIN 625; ISO 15) can take radial and axial forces and are suitable for high speeds, Their angular adjustability is small. Bearing
ance [3 J
3
Z
Bearing Clearance Groups (according to DIN 620 Sheet
Table 21. Tolerances for shaft and housing for normal bearing clear-
Width series
r-
The first two features require an interference fit. Particularly with greater loads which cause an extension of the rings, and also with shock loads, tight fits are required, The temperature drop between the bearing rings, which occurs in virtually all operating conditions, is also relevant. Tolerances for normal bearing clearance are given in Table 2. As the bearing rings are not very thick, rigid bearing seating and limited shape and running tolerances (straightness, roundness, parallelism and planarity of the shoulders), toleranced tighter than the diameter, are stipulated. The tolerances of rolling bearings are standardised in DIN 620; ISO 492, 199, 5753, 582 Apart from tolerance class PO (normal tolerance) the standard provides for tolerance classes P6, P6X, P5, P4 and P2. Bearings with these tightened tolerances are intended for very accurate shaft guides and very high rotating speeds, A major application of bearings with tightened tolerances is in the operating spindles of machine tools. Bearings for this purpose are manufactured in the tolerance classes SP (special precision), UP (ultra-precision) and HG (high accuracy), in addition to in the standard tolerance classes. Tapered roller bearings dimensioned in inches are found in nonnal tolerance and in tolerance class Q3, The term radial (axial) bearing clearance refers to the distance through which the bearing rings can be moved in a radial (axial) direction from one end position to the other. The bearing clearance should be chosen in such a way that there is no distortion of the bearing rings and surrounding parts. The radial bearing clearance is reduced as a result of the fit, particularly in the case of tighter fits, but also because of the temperature drop, This reduction must be taken into account when choosing the bearing clearance.
:m~:~
~~
----
3 Diameter Z} 0 senes
t I
~
Bore
Figure S. Structure of size plans for radial bearings.
Shaft
Housing
Ball bearings
jS to kS
J6
Roller and needle bearings
k5 to m5
K5
__
Mechanical Machine Components. 4 Rolling Bearings
4.11.11 Roller Bearings
Types of Constnactlon (Fig. Sa-g) Figure 6. a-g. Ball bearing types for predominantly radial loads (see text).
positions that are not in alignment result in additional stresses which reduce the service life of the bearing. Grooved ball bearings are also manufactured with cover or sealing washers. (b) Double-row grooved ball bearings (DIN 625; ISO 15) are manufactured with or without filling grooves. Bearings with filling grooves can therefore convey only slight axial forces. They are not suitable where there are angular errors. (c) Detachable groove ball bearings (DIN 615; ISO 15) are only standardised up to a bore diameter of 30 mm. They have only one shoulder on the outer ring and can therefore be dismantled. Inner and outer ring are fitted separately. A transmission of axial forces is possible in one direction only. (d) Single-row inclined ball bearings (DIN 628; ISO 15) take axial forces in one direction only. They are therefore located against another bearing in an a or X arrangement. Single-row inclined ball bearings cannot be dismantled. (e) Double-row indined ball bearings (DIN 628; ISO 15) take radial and axial loads in both directions as well as instantaneous loads. Their structure is equivalent to a pair of single-row inclined ball bearings in a arrangement. The bearings are delivered with very small amounts of play and so the fits used should not be too tight. ({) Four-point contact bearings (DIN 628; ISO 15) are single-row inclined ball bearings which take axial forces in both directions. In axial section the contour of the raceways of the inner and outer ring consist of arcs which
form ogives. The inner ring of the four-point contact bearing is divided so that a large number of balls can be accommodated. (g) Selj-aligning ball bearings (DIN 630; ISO 15) are double-row bearings with hollow spherical outer raceway which compensate for alignment errors and shaft deflections of up to 4°. Owing to the unfavourable osculation between balls and outer ring the axial load-bearing capacity is less than that of a grooved ball bearing.
Types ofConstnaction. For predominantly axial loads; see Fig. 7.
(a) Cylindrical roller bearings (DIN 5412; ISO 15) can transmit high radial forces but no or only slight axial forces. They can be dismantled; the inner and outer ring can thus be installed separately. The various types are distinguished by the arrangement of the flanges. Types NU and N are used as loose bearings. Type Nj has two flanges on the outer ring and one flange on the inner ring, so that small axial forces in one direction can be absorbed. To take small axial forces in both directions type NUP is used, with its two flanges on the outer ring, a fixed flange and a loose flange disc on the inner ring. The angular adjustability of cylindrical roller bearings is small. Between cylindrical shell surfaces and edge radiusing there is a spherical transition zone. This cylindrical-spherical profile prevents the occurrence of edge stresses and produces a modified line contact with comparatively reduced stress distribution. An almost constant compressive load application without stress peaks is achieved by the so-called logarithmic profile, which does not show any discontinuity in the profile curve. (b) Tapered roller bearings (DIN 720; ISO 355) have a high load-carrying capacity and can take combined loads. They can be dismantled so that the inner and outer ring can be installed separately. Since they can take axial forces in one direction only, it is necessary to fit a second bearing in symmetrical opposition as a countersuppott. The bearing clearance is adjusted on installation. Angular adjustability is slight and therefore attention must be paid to good alignment. (c) Selj-aligning or barrel-shaped roller bearings
(DIN 635; ISO 15) are single-row roller bearings with angular adjustability (up to 4°), which are suitable for high radial loads. Axial loading capacity is small. (d) Selj-altgntng radial roller bearings (DIN 635; ISO 15) are suitable for the heaviest loads. In this bearing two rows of barrel-shaped rollers run on the hollow spherical track of the outer ring. Alignment errors and shaft deflections are compensated for. The rollers are guided on fixed flanges so that axial forces also can be taken. (e) Axial cylindrical roller bearings (DIN 722; ISO 104) take high axial forces in one direction. An axial minimum load is required for kinematically efficient rolling. ({) Axial selj-aligning roller bearings (DIN 728; ISO 104) for high axial forces and relatively high rotational speeds. Owing to their raceways inclined to the bearing axis they can also take radial loads which however must
Axial Grooved Ball Bearings (DIN 711, DIN 715; ISO
104) of the unidirectional type (axial force in one direction only) or the two-directional type take high axial forces. They are not suitable for radial loads. To achieve kinematically efficient rolling even at high rotational speeds a minimum axial load is required.
a
b
f
e
a
b
c:
c:
d
Figure 7. a-cL Axial grooved ball bearings: a unidirectional, b two-directlonal, c unidirectional with spherical housing disc (compensating for angular errors), d two-directional with spherical housing disc.
g Fipre 8. a-g. Types of roller bearing (see text).
d
4.3 Load Capacity, Fatigue Life, Service Life. 4.3.2 Fatigue Life Under Steady Load and Speed
not exceed 55% of the axial force. Because of the hollow spherical raceway the bearings can he adjusted hy up to 2°. To ensure kinematically efficient rolling a minimum axial load is given.
(g) Needle bearings (DIN 617, DIN 618; ISO 1206, 3245) require small radial dimensions owing to limited space. They are particularly suited to shock loads and pivoting motions. Axial forces cannot be taken. They have a higher coefficient of friction than other types of rolling bearing. The needles are kept parallel by the cage. 4.2 ..} Linear Rolling Bearings and Ball Splines Types of Construction (Fig. 9a-c)
(a) Ball guides consist of external bush, cage with halls, and internal hush or shaft. As the cagc only performs the half stroke, the axial movement of stroke is limited. (b) Ball splines contain three or more ball grooves with return. Consequently the stroke is unlimited. Coefficient of friction ~ ~ 0.002 to 0.004. They are suitable only for rectilinear shaft guides. (c) Roller guides as ladder-shaped flat cages or in the form of roller shoes are suitable as flat guides. 4.2.4 Materials Rolling bearing steels; see D3.1.4. The cages are mainly pressed from sheet steel. Brass, light metal (aluminium alloys) and steel are used for the manufacture of solid cages. Solid cages are increasingly heing made now of plastic (glass-fibre-reinforced polyamide PA66).
If a bearing is stationary, swivels or rotates slowly, it is regarded as a statically stressed bearing, for which the static load capacity must be given. This also applies to dynamically stressed bearings which are subject to short sharp shocks. The dynamic load coel1icient C is used with rotating bearings. The terms static and dynamiC do not refer to change; in the exteolal load.
4.'}.1 Static Load Capacity The static load coefficients Cc" for radial loads and Coo for axial loads are static forces which are hased on calculated stresses at the contact point in the centre of the most severely loaded contact point between rolling elements and raceway rated at 460() MPa for selfaligning ball bearings, 4200 MPa for all other radial bearings, 4000 MPa for all radial roller bearings with radial load and 4200 MPa for axial hall bearings, 4000 MPa for all axial roller bearings with axial load. Under these loads a permanent deformation of approximately 0.0001 times the diameter of the rolling element occurs at the contact points between rolling element and raceway. To demonstrate the adequacy of a hearing's load capacity, the static factor J~ ~ Col Po is used. For bearings which are to mn particularly smoothly and easily, a large factor J, is required" The following figures are used: When the reqUirements regarding smoothness and friction characteristics are high. /, = 2 to 2.5; with marked shock loads /, = 1,':; to 2: with normal smoothness demands /, = 0.8 to 1.2; with low smoothness demands and with vibration-free operation/, = 0.5 to O.H, for axial selfaligning roller bearings J~ should be :2: 2. as the
flange of the shaft disc is severely stressed.
4.3 Load Capacity, Fatigue Life, Service Life The bearing size required for any particular bearing is determined on the basis of the load capacity of the hearing in relation to the loads occurring and to the fatigue life and operating safety requirements. As a measure of load capacity the bearing computation uses the static load coefficient Co and the dynamic load coefficient C, which can be computed in accordance with DIN ISO 76 and DIN ISO 281 Part I or taken from the rolling hearing manufacturers' catalogues.
Loads which are composed of a radial and an axial load must be converted into the equivalent static bearing load Po. By this is meant in the case of radial bearings the radial load, and in the case of axial bearings the axial load, that would have caused the same permanent deformations in the hearing as the load actually applied. The equivalent static bearing load is obtained from the following two general t()rmulae
P"
~
X.,F,. + YoF,.
Po
= F,.
The greater of the two values is to be used. F, is the radial component of the greatest static load, F, the axial component of the greatest static load, Ko the radial factor of the bearing and Yo the axial factor of the bearing, which can all he taken from Tables 2 and,} of DIN ISO 76 or from the rolling bearing catalogues. They differ for the various types of bearing.
4 ..}.2 Fatigue Life Under Steady Load and Speed
a Flal guide
q
~k~~ ~
hITi.-Roller shoe
r=mJ
-=
I
c Figure 9. a-c. Linear rolling bearings and ball splines (see tCXL).
The dimensioning of a dynamically loaded rolling bearing is done on the basis of the fatigue life (DIN ISO 281). For an individual bearing it gives the number of revolutions that are executed by a bearing ring or disc in relation to the other bearing ring or disc, before the first sign of material fatigue (pitting) is visible on one of the two rings or discs or on the rolling element. Fatigue life must be distinguished from service life, which means the actually possible operating time of a bearing. It is not possihle to predict exactly the fatigue life of the individual rolling bearing, even with an accurate knowledge of the loading and operating conditions, since fatigue running times vary widely. An assessment can therefore only he made on the basis of statistics from a relatively large number of tests with the same bearings under the salne test conditions.
Mechanical Machine Components • 4 Rolling Bearings
Consequently the concept of nominal life LIO is used. It corresponds to the fatigue life in millions of revolutions reached or exceeded by 90% of a relatively large number of clearly the same bearings. Hence 10% of the bearings can fail earlier. The nominal life is calculated, using the life equation (DIN ISO 281), as LIO
=(sy P,
for radial bearings,
LIO
=(~r
for axial bearings,
LIO in 10" revolutions.
1
(I)
The exponent p has a value of 3 for ball bearings, and a value of 10/3 for roller and needle bearings. The dynamic radial (axial) load coefficient C, (G.) indicates for a rolling bearing the radial (axial) external load, of constant magnitude and direction, which the bearing can take theoretically for a nominal life of 106 revolutions. P, (P,) is the' dynamiCally equivalent radial (axial) load whose magnitude and radial (axial) direction is constant and under the effect of which a rolling bearing would achieve the same nominal life as under the conditions actually prevailing. For radial bearings
P, = XF, + YF, and for axial bearings
P, = FX, + YF,. F, is the radial component of the load and F, the axial
component. The radial factor X and the axial factor Yare established by DIN ISO 281 or the manufacturer's data. If the bearing turns at constant speed, the life in hours may be expressed by (2)
nlis procedure for determining the LIO life is a comparative method whose certainty is the greater, the better the preconditions such as employment of a conventional rolling bearing steel and in practice normal operating conditions (a high degree of separation of the surfaces by the
Table
(3)
or, expressed in hours,
Failure probability factor a I
10
Probability of failure (%)
4
Fatigue running time
0.62
Factor a]
L,
L,
L,
L,
0.53
0.44
0.33
0.22
Life Coefficient a, for Operating Conditions. Coefficient a, is used to take into account the suitability of the lubrication, and the conditions that cause changes in the material characteristics. Here too there are no quantitative estimates for a, in DIN ISO 281. Assuming that no greater probability of survival than the generally accepted figure of 90% is to be applicable, that the bearings are made from materials that were assumed for the specified dynamic load ratings, and that normal operating conditions apply, then a l a, a, I; in that case Eqs (I) and (3) are identical. Beyond DIN ISO 281, the rolling bearing manufacturers offer expanded life calculations in which the coefficients are quantified. Correspondingly, the life coefficients a, for the material and a, for the operating conditions are combined to form a common factor a z, because of their mutual influence. As a function of K = V/VI (v is the operating viscosity of the lubricating oil, VI the reference viscosity as a function of bearing size and speed from Fig. 10), it can be taken from Figs 11 or 12. The influence of the operating temperature on the material is taken into account by the temperature factor.r. as per Table 4. The life coefficient a l is taken unchanged from DIN ISO 281. The expanded life equation then reads
= = =
(5)
(6)
By including the fatigue strength of rolling bearings, the rolling fatigue theory according to Lundberg and Palmgren, from which the classical ISO-standardised equation for calculating L 10 life originates, has been expanded according to [41 to become life Ln,,, such that a fatigue
lubricant, no contamination in the lubrication clearance)
are satisfied. Recommendations in DIN ISO 281 enable improvements in rolling bearing steels and in production methods, and the effect of operating conditions, particularly more precise knowledge of the influence of lubrication on the process of fatigue, to be incorporated in the life calculation. Accordingly the attainable fatigue running time Ln, by the modified life equation is
~.
oj r---.:-,---,--..;:-,----,,-.....
1
B ~--~~--+-~~~--~~--i-_i·_i~
6
~101~--~~~r-~~~~Cf~~t--+-+~ 'EBk---+----'l-ori .~
6
I----"'-..J-----+--+~'" ~r--1""c_+--t"__cl-l
",- 4
(4)
Life Coefficient a] for Probability of Failure. For certain applications it may be desirable to calculate life for failure probabilities other than 10%. For this purpose the factor a l was introduced (Table 3). Life Coefficient a, for the Material. The characteristics of the material have an influence on the life of a rolling bearing. This influence is covered by the coefficient a 2 • Currently, however, under DIN ISO 281 the coefficient cannot be selected on the basis of quantifiable characteristics.
10
4
6 B 10 1 1 dm=iD+dil2 in mm
Figure 10. Kinematic reference viscosity bearing diameter d m and speed n.
"1
as a function of mean
I::BI
4.3 Load Capacity, Fatigue Life, Service Life • 4.3.2 Fatigue Life Under Steady Load and Speed
10
40
--. .f-~ ----li I ___ !..A ~",\ 7<
10 a
I..«:t<\~
~
k-
[\
~
['(
l\\ 10-I 6
/fll/ / If #;VI / I
20
I
VI II
6
("
K\," ,"-.,'-.: ,"-.,.
I
~ 0][~~
]][
~ ~"
810
I
I
~I;<
~ ~/
~V/
I
I
VII / / / +j, / h~ tI~ ~"?I / / / V V V~ A~ / I / ~
81
6
I
z
-
Zone
Fipre 1~. Coefficient
the curve
K
4
6
L
V/
~y
V
t:- I--
~ ~ff-=:
X=V/VI
Figure 11a a z3 graph in accordance with {3): v operating viscosity of lubricant, V t reference viscosity.
Suitable additives in lubricant.
~;~
V ~VVV- V ..--
6 810
I: Transition to fatigue limit. Condition: utmost cleanliness in lubrication clearance and not too high load if fatigue limit is aimed for. U: Good cleanliness in lubrication clearance.
V
/
al0 -I TicIPu IP)
~/ 0,1
4 6
a1
a" for radial ball bearings (4 J. For
K
= 4 is to be used. As TjJPulP) tends towards 0,
tends towards 0.1 for all
K
>4 aZ3
values.
III: Unfavourable operating conditions, contaminants in lubricant,
unsuitable lubricants.
limit load Puis introduced, meaning the limit load up to which no fatigue occurs in the bearing. Values for Pu are given in the bearing tables [4]. Accordingly L= is governed by
10
(7)
8
- - - - 1---"-
/
V
---
-<" (\.
k-
~~ ~['( ~~
I~[\
Zr'
~ ~'0 ~
~ ~t
....
~~-
-
I---r-
The values for au to be inserted in Eq. (7) may be taken from Figs 13 to 16 for the different types of bearing as functions of Tk(Pu/P), with K as parameter. 1)c registers various degrees of contamination (Table S). The figures are based on a general safety factor which depends on the type of bearing and is comparable with
!---I-
1/// / ~/ II
40
- - - !---!-
II
20
~V
10 a 6 6 8
1
6
tit: ~I/ I V
X=V/VI
Fipre
1~. a 23 graph in accordance with [4). Grid surfaces when
I 1
V;rh
810
/,~ V~ V V
~
EP additives are used.
1"7 (/// L ~~ V VV' ~/ ~V ~y ~~ ~ ~~ ~ E:::: ~~ 0.1
~~
Table 4. Temperature factor;; (3J Operating temperature (0C)
Temperature factor It
I
2
150 200 250 300
4 6
0.73
0.42 0.22
al0-I 2 TicIP,!P)
4 6
al
Figure 14. CoeffiCient a" for radial roiler bearings (4 J. For K > 4 the curve K = 4 is to be used. As T/c(Pu/P) tends towards 0, a23 tends towards 0.1 for all
K
values.
Mechanical Machine Components • 4 Rolling Bearings
40
,
1
//~
20 10 6 6 I,
/IlL
I
2
I,
.4:~ ~ >/.0
I
,
~
1;.'.
~~'
~
II
6 610 cL
2
4
Highest cleanliness (contamination particle size of the same order of magnitude as thickness of lubricating film)
~ '/
'IL L /
!~ ~V' ~/
/
§ ~ k.,.---VV ~ ~ ~~ ~ f,=
6 610·'
, 661
2
2
Figure IS. Coefficient a 2:'o for axial ball bearings [4]. For
4
K>
4
the curve K = 4 is to be used. As TJJPul P) tends towards 0, a 2:. tends towards n,t for all K values.
High cleanliness (corresponds to the conditions typical for grease-filled bearings with sealing discs on both sides)
O.S
Normal cleanliness (corresponds to the conditions typical for grease-filled bearings with cover discs on both sides)
0.5
Contamination (corresponds to the conditions typical for bearings without cover or sealing discs; coarse filtering of the lubricant and/or of contamination penetrating from outside)
0.5 to 0.1
Severe contaminationb
o
"-The quoted 'T/c values are only valid for typically solid contamination; life-reducing effects arising from the penetration of water and other fluids into the bearing are not taken into account here. hWith extremely severe contamination the wear predominates; in this case life is well below the calculated value for Inaa.
1.0
zo
Coefficient 7),"
11 'll
!
i
1 6 6
6·
Operating conditions
II/; 'I
4
Table S. Coefficient T/c (guide values) for various grades of contamination [4]
I:
10
6
II
the use of the factor (4 - 3K) is therefore not recommended. If a,,( 4 - 3K) becomes greater than the value of an for K = 1 from the graph, this graphical value for
K
= 1 is to be used.
4.~.~
Dynamic Load Capacity Under Varying Load and Speed
If a rolling bearing runs at varying speeds and varying load P, its fatigue life can be determined from Eq. (1), with
Figure 16. Coefficient a 1 !> for axial roller bearings [4]. For K > 4 the curve K = 4 is to be used. As Tk(PdP) tends towards 0, a 2 ., tends towards 0.1 for alI K values.
the safety factors normal in fatigue strength analyses. 1be graphs are drawn for typical values of this safety factor and are applicable to lubricants without (extreme pressure) EP additives. If lubricants with such additives are used a greater life may if applicable be achieved in the range K < l. The maximum possible life may be estimated by mUltiplying the coefficient a z; (without EP additives) by the factor from (4 - 3K) and inserting this higher factor a z; (with EP additives) in the formula for Ln ". It is however questionahle whether, when there is contamination, a longer life can he achieved by EP additives anyway. If 1) < 0.5
the mean speed nm = q1n l + q2n2 + ... + qnnn and the mean dynamic equivalent load Pm. For this purpose the total time period T under consideration is to he divided into individual periods ti during which the constant loads Pi are effective. qi = t;lTis the proportion of the fractional times ti relative to the overall time T. The relationship for Pm can be derived from the assumption that the "fatigue resistance" IlL used per unit of time is the same as the sum of the individual resistances anILn used per revolution share an:
IlL = a,/L, + a 2 1L, + ... + anlLn where L" L2 ... Ln are the fatigue life values which would have resulted in the respective operating conclitions, e.g. loads P" P z ... P n (the Palmgren-Miner rule). It follows that
Pm
=
(P1'
2: n,qi·lnm + p~. 2: n//j·lnm+"· 1.1
].]
If speed and bearing load in the period T are clearly defined time functions net) and Pet), this gives for nm and Pm
4.3 Load Capacity, Fatigue Life, Service Life. 4.3.6 Limiting Speeds
P
n
P,
way to estimate service life is by comparison with similar installation situations. Eschmann [5] was able to develop a method of estimating wear time from extensive investigations on bearings. Bearing wear V, measured in ~m, which has the effect of increasing play, may be recorded as wear factor fv relative to a constant eo dependent on the bearing bore d, measured in ~m.fv = Vle u (Fig. 18). The wear time is directly dependent on the conditions at the rolling and sliding surfaces and on the running accuracy requirements of the bearing. The conditions in the contact surfaces (e.g. lubrication, dirt) are identified in Fig. 18 by the fields a to k. The empirically permissible wear factors fv and the fields identifying the operating conditions for various installation situations are compiled in
nz
r--- ---,
i
n,
'--i
Pz ~. I
i,
i,
Time element
q Figure 17. Example of dynamic equivalent bearing load and speed varying in increments with time.
Appendix F4, Table 1.
T
nm =
~ JnU) dt
4.3.5 Choice of Required Fatigue Life
o
and
p.
If the required life is known from the conditions of machine operation, Eq. (2) to (7) can be used to select the correct bearing size by determining the required load coefficient. If data on the required fatigue life are not available, however, guide values may be taken from Appendix F4, Table 2.
~ ~W'''''' )l 1P )
J net) dt o
4.3.6 Limiting Speeds
As an example, this gives for the speed and force graph shown in Fig. 17 Pm
= (~ nm
[Pi( (n,q,
+
The rotational speed limit of a rolling bearing for a particular application can only be determined in advance approximately. It depends on type and size of bearing, type of cage, bearing play, accuracy of bearing parts, bearing load and lubrication. Running tests show that the product of limiting speed n, and mean bearing diameter d m = (D + d)/2 is roughly constant for radial bearings up to d m :::; 75 tnm. For larger bearings correction factors as in Fig. 19 are introduced to cover the influences from bearing size f, and bearing load f2' Thus the resulting formulae for estimating limiting speeds are (evaluation Figs 20 and 21): for radial bearings n,dm = fJ.,A, for axial bearings n,'-iDH = f J.,A. (A is the type-dependent coefficient as per Figs 20 and 21, H the height of the axial bearing.) If the load is greater, i.e. lower life values arise, the values read off are multiplied by f,. In Figs 20 and 21 two values for A are given for each type of bearing, the "normal" speed limit, which can be achieved without special action using grease lubrication, and the
n,q,) +
4.3.4 Service Lif"e and Wear Service life is the operating time actually possible during which the bearing fully performs the required function. When external influences are unfavourable, service life can be less than the calculated fatigue life. Thus errors in alignment between shaft and housing, dirt in the bearing,
corrosion, excessive operating temperature or unsuitable lubricants may lead to the premature failure of the bearing through wear or fatigue. Given the multitude of installation and operating conditions it is not possible to determine service life accurately in advance. The most reliable
eo in Jlm 7 8 9 10
910
20
40
60
I
'
I
100
din mm
L, in h
14
20 I
200
30 400
40 600
1
1000
Figure 18. Nomogram for eo, wear time Lv as a function of wear factor ~. and operating conditions a to k.
1:.1
Mechanical Machine Components. 4 Rolling Bearings
;;:~ 4.~00.
104
VOo/})/})
8
r-~~I"""/})Ii?
6 0.6'--..L50----:'::----:-'-c"-"~::------::"
100
25
500
dmin mm
50 100 fOHin mm
150
250
Fipre 1,. Correction factors I, and I, for calculation of limiting speeds. The curves for are valid for bearings of series 511, 2344(00) and 2347(00). For the latter two the value H/2 should he inserted instead of H.
,,<'00000
C c
'"
•
II
10
4
•
I'
'
,¥
l
1j~ ~
~ ~~/:!tII-l, , l l! III1
1530
I
r~tl '
:;;iG
•
I"-
40 50 60 70 I 90 I 120 160 200 I 300 35 45 80 100 140 180 250
Fipre lI1. Approximate speed limits for axial bearings under a load corresponding to a nontinal life of 100000 hours.
,
ft:I~,1 ' •• l
101 s
I'.
1'..
,
,,~~' , '~ll--t ~
10
l'-
fOii' in mm
~-&:
K-Y'
'E
25 30
, --"
r- ,,!'
~-r-$fl
8 6
i
l,
t-: l'
f
1.......
"''SII~
10 3 r-.1 '
10 2 ~~'T
~['
Ii~ 1'11 1'rt:
[IJii
10 5
r:::",
~~171'
'E
f:!~~
..!.-
~
).
"i"
40 50 70 90, 110 1150 1100 I 300 1400 1500 700 900 35 45 60 BO lOll 140 180 150 350 450 500 BOO 1000 dm inmm
Fipre lIO. Approximate speed limits for radial bearings under a load corresponding to a nontinallife of 100000 hours.
"maximum" speed limit. The maximum speed limit can only be achieved with improved cage design, increased bearing clearance and by appropriate lubrication, e.g. oil spray lubrication, and favourable load and cooling conditions.
Deformation of race Hertzian pressure distribution
EHD pressure distribution
4.4 Lubrication of Roiling Bearings Lubricants have the job of reducing friction and wear in rolling bearings by carrying oil into the contact areas in order to achieve a complete separation of the surface by a load bearing lubricating film (ltqutdfrlctton). The state of lubrication is estimated by means of the equations of the EHD theory (elastohydrodynamic lubrication, Fig. 2121) or by determining the viscosity ratio k = v/v, (cf. Section F4.3.2). Furthermore, lubricants protect rolling bearings from corrosion and the ingress of foreign matter. 4.4.1 Choice of Method of Lubricadon Since the type of lubrication influences the development of bearing and seal, the method of lubrication must be decided before the start of the design stage. Rolling bear-
O.l.aD~'(TJ .v) 0.7
hO=(l 1 )0.43 (0)0.13
r;+i'i . T
(
E )0.03 -::-;- In p.m 1
ffi1
Fipre lIli. Elastohydrodynamic lubricating film, example roller/inner ring [6]. b o (measured in ",m) minimum lubricatiog film thickness in rolling contact, a (in mm2 IN) pressure viscosity coefficient, 'I (in mPa s) dynamic viscosity, v (in mls) = (VI + v z)/2velocities, r l (in mm) radius of roller, rz (in mm) radius of inner riog race, Q (in N) roller load, I (in mm) roller length, E (in N/mm') modulus of elasticity = 2.08 X 10' for steel, l/m Poisson's constant = 0.3 for steel.
4.4 Lubrication of Rolling Bearings. 4.4.3 Choice of Grease
Figure 1:-'. Greasing by means of disc with grease holes.
ings are mostly lubricated with grease (approximately 90% of rolling bearings). Expenditure for seals is low; design is simple [7]. Oil lubrication is used if other
machine parts used in the design already require oillubrication, or if oil is necessary for heat transmission. Solid lubrication is reserved for special applications. The choice of lubrication method depends on the operating conditions and the environmental influences. Appendix F4 Table ~ gives common lubrication methods and systems as a function of the speed characteristic value ndm in mm per min. The reliability of the lubricating method also depends on the undisturbed supply of lubricant. The circulation of oil must be monitored and with oil sump lubrication oil level checks are necessary. Grease lubrication is reliable if the regreasing schedules are kept to.
Greasing. The ftlling amount is determined by the speed. The bearing cavities should be permanently smeared with grease so that all functioning surfaces receive lubricant. The housing area on both sides of the bearing should however remain free of grease at high speeds so that the grease expelled by the rolling elements can be contained there. In Fig. 2~ the passage of grease is directed by a disc with grease holes. In bearings with grease volume regulators (Fig. 24) greater quantities can safely be fed in. The grease volume regulator consists in principle of a disc that revolves with the shaft and spins off excess grease into side housing areas. 011 Lubrication. 'lbere are two ways to obtain a low bear-
ing temperature: a sparing or a very copious supply of oil. If cooling at high speeds is necessary, lubrication using
small quantities of oil should always be preferred (drip feed oil lubrication, oil mist lubrication or oil air lubrication) because the friction losses in the bearing are then smaller.
In 011 spray lubrication - for very high speeds - the most important function of the lubricant is to dissipate heat. The oil is sprayed into the bearing between cage and bearing ring by nozzles. Experience has shown that minimum oil jet speeds at IS mls are required. Drain passages are necessary to prevent oil stagnation. Oil bath or oil immersion lubrication is suitable for low speeds. When not in motion the oil level should normally only reach to the middle of the lowest rolling element. A higher oil level may result in frothing and therefore deficient lubrication. The installation of a centrifugal disc intensifies the delivery of oil in the bearing so that higher speeds can be attained. In oil circulation lubrication (pressure lubrication) the oil is fed in most cases along a direct pipe through the bearing. The advantages of circulation lubrication are heat dissipation and the flushing out of wear particles and also low construction cost when exploiting the delivery effect of bearings with asymmetrical cross-sections (roller bearings). Cleanliness in the lubrication clearance, which is necessary for the endurance of the rolling bearing, can be achieved by ftltering the circulating oil.
4.4.2 Choice of Oil For oil lubrication mineral oils subject to minimum requirements under DIN 51 SOl are suitable. Lubricating oils with improved resistance to ageing according to DIN 5 I 517 are preferred. Synthetic oils are reserved for special applications. The behaviour of the oils vis-a-vis sealing materials and plastics is to be checked. Speed, mean bearing diameter, load and temperature determine the choice of oiL With PIC < 0.1 and speeds n < 0.66n,u;1 a lubricating oil with a kinematic operating viscosity of v = 12 mm'/s is sufficient. Figure 2S can be used to establish more precisely the operating viscosity v as a function of bearing type and limiting speed ratio for oil lubrication n,onln; it also gives the nominal viscosity of the oil at 40°C. For bearings subject to very high loads PIC > 0.2 the next higher nominal viscosity grade is chosen. If there are no empirical values, the probable operating temperature is to be estimated. In normal conditions, unblended but preferably inhibited oils
(DIN 51 502, code letter L) can be used. High loads PIC > 0.1, given a viSCOSity ratio vlvj < I andlor high sliding friction elements, demand oils with wear-reducing additives (DIN 51 502, code letter P or EP additives). For oil mist lubrication the atomization capability and oxidation resistance of the oil must be assured. Synthetic oils are used at extremely high or low temperatures. Silicon oils can only be employed with low loads (PIC < 0.025). Guide values for the oil quantity iT as a function of the rolling bearing external diameter D can be taken from Fig. 26. Appendix F4, Table 4 gives characteristic values of various oils. 4.4.~
Figure 24. Rolling bearing with grease volume regulator.
Choice of Grease
For the lubrication of rolling bearings greases of consistency classes I, 2 and 3 (NGLI values) are mainly used. The choice of consistency is determined by bearing type, speed, operating temperature, position, starting torque, sealing effect and ease of delivery. The data in Appendix F4, Table S may be used as guide values. There is only an indirect relationship between the consistency of the greases and the speed characteristic nd=. The decision factor is the basic oil viscosity so that the permissible speed characteristic values with greases of one consistem:y class may fluctuate within broad limits.
Mechanical Machine Components. 4 Rolling Bearings
I--- Normal sphere of application
~-
~ ~'1
~
I. ~-&"",
I
10 1
'EB
!}7t \7»
~,
~i- ~-&W ~",7t I
E
I
/ 10
/
B
/
V V
/
nI
~"", ,,~
!}'IS
'I:-'lS/
T
-~--
20
10
3D
I in 'C
40 50 60 70 80 90100
Figure ZS. Determination of oil viscosity for rolling bearings. Example: radial bearing fully rolling: n",O;l/n = 6, t = 70°C gives v = 70 mm 2 Js and vlscosity class VG 320.
10
the rolling bearing and replaced by fresh grease. It must be ensured that the used grease can leave the site of the bearing. It is an advantage to regrease when the rolling bearing is warm and turning. Guide values for the regrease and grease change times of lithium grease when PIC is :S 0.1 are given in Fig. 27, with the following values to be used for the required coefficients K L :
co
~
~ 10-1 f--- -j,''----1r .".
Din mm Figure 26. a-c. Volume of oil for circulation lubrication: a volume of oil sufficient for lubrication, b upper limit for bearings of symmetrical form, c upper limit for bearings of asymmetrical form.
Appendix F4, Table 6 gives an overview of the composition and characteristics of the most important types of grease. In fully ftlled rolling bearings the necessary quantity of grease is adjusted automatically according to speed. It must be possible for the excess grease to be taken up in spaces in the housing to the side of the bearing. For bearings with cover and sealing discs a filling quantity of around 30% has proved favourable considering service life and friction. The required quantity of grease may be calculated approximately from the following numerical equation: G=
where B inmm.
d,; 40 mm f= 1.5
f· Bdm /1000 in
Type of bearing
Coefficient KL
Radial grooved ball bearing Inclined ball bearing Axial grooved ball bearing Cylindrical roller bearing Axial cylindrical roller bearing Axial <•.'ylindrical roller bearing Needle bearing Tapered roller bearing Self-aligning radial roller bearing Cylindrical roller bearing fully rolling
1.8 1.4
0.6
0.5
f- ZOfIe with increased fa~ure I- probab~ity (grease change)
I
I
CD + d)/2
.,.. r
~
10'
1
,, -
,.. 1-----....
--
Regrease
zme
/;/
40 to 100100 to 130130 to 160 160 to 200 > 200 1.0 I.S 2.0 3.0 4.0
In the case of axial bearings the bearing height H is to be inserted instead of B, Lubricating greases lose their characteristics and must be supplemented and/or renewed. For supplementing, the required period is termed the regrease time (t'n) and for renewal the grease change time (trw). In regreasing, the used grease should as far as poss.ibJe be removed from
1.0 (with cooling) 0.3 1.0 0.8
The values are valid for atmospheric environmental conditions up to temperatures of + 70°C, measured at the outer race ring. At higher temperatures the stress on the
cm 3 ,
bearing width in mm and d m
1.2 1.0
6
I
10
flo"' .... 'X,/n
Figure 1.7. Regrease time and grease change time for lithium soap greases valid for PIC:-:; 0 .1.
4.6 Design of Rolling Bearing Assemblies. 4.6.1 Mounting and Arrdngement of Bearings
greases increases considerably. Shorter greasing times must be expected; for roughly every 15°C increase in temperature (from 70 DC) the greasing time drops to half the initial value.
the dissipation of heat to the surroundings and/or to the lubricant influence the operating temperdture of the bearings. The bearing temperature follows from PR
= cPlI + cPo,
with
4.5 Friction and Heating In rolling bearings the frictional work is composed of the following elements: friction between rolling elements and raceways induding losses by material damping, friction between rolling elements and cage, and between cage and guide surfaces, friction between rolling element end faces and flanges in the case of roller bearings, fIlling resistance of the lubricant, ventilation [windage] losses and resistance from foreign bodies. These influences generdte the friction torque on the rolling bearing, which is dependent not only upon bearing load but also on bearing type, lubrication, load direction and rotational speed. With a load ratio PIC = 0.1, good lubrication and normal operating conditions, the friction torque may be calculated roughly by M
= tlL,Ed.
Figure 28 gives coefficients of friction IL, for rolling bearings with normal bearing clearance and sparing lubrication, as a function of purely radial bearing load [8]. A more accurate calculation is allowed by }J =
1110
+ 111 1 ,
where Mo represents the load-independent friction torque and M, the load-dependent friction torque. Mo
= 10-'/0
(vn)2I'd;, for
Mo = 160· IO- c / ud;" for
vn 2: 2000 mm2/s· min I vn
< 2000 mm'/s· min
I
where Mo = the load-independent friction torque in N mm, /0 = the coefficient, dependent on lubrication and bearing type ([0 = 0.75 to 20), n = the speed in min I, V = the operating viscosity in mm 2 /s. d m = (d + D)/2 = the mean diameter. M,
= j;P,dm ,
where MI = the load-dependent friction torque in min - I, /1 = the coefficient, dependent on bearing type and load ([I <; 0.(02), and P, = the load goveming the friction torque (see [4]). Bearing friction, friction of seals, extemal heating and
and
cPo = Qc/T
A -
Here, cPlI, cPo = the flow of heat dissipated to the surroundings and to the oil respectively, " = the heat transfer coefficient; A = the size of surface transmitting heat, Q = the volume flow of oil, c = the specific heat capacity (1. 7 to 2.4 kJ/(kg K», p = the oil density, TL , TlI , TA , TE = the bearing temperature, ambient temperature, oil outlet temperature and oil inlet temperature respectively. For rolling bearings the heat transfer coetllcient " is some 50% greater than for plain bearings.
4.6 Design of Rolling Bearing Assemblies 4.6.1 Mounting and Arrangement of Bearings Bearings must be mounted, so that there is sufficient bearing clearance between rolling elements and race rings. Assembly is aided by pressure oil (force-feed union) and tapered seating, especially with large bearings. When rolling bearings are being assembled, care must be taken to ensure that the assembly forces are not transmitted through the rolling elements. Bearing arrangement (Fig. 29) for a shan bearing is preferably as fixed bearing and loose bearing. The fixed bearing undertakes the axial guiding by fixing the inner and outer ring, while the loose bearing is freely adjustable. Axial adjustability is by bearing type (e.g. roller bearing type NU) or by movability of the
·m .. m
~~~. I
i
. .
.J
aL
~.
.
1
L
.
bL_.
30
40
50
. - .-.J
I
..--1
.J
.
.---J
~l~
cM:l~jbd:U
0.5 f---!--.-+-~--t-\ Self-aligning ball bearing Cylindrical roller bearing
20
.J
~t% ....
J
10
TE )·
50
Radial bearing load in kN
Figure 28. Friction coefficient 11-, for rolling bearings (medium series) with normal bearing clearance and sparing lubrication in accordance with [81.
Figure 29. a-c. Bearing arrangement: a fixed and loose bearing with peripheral load for inner ring and point load for outer ring (with turning load for outer ring, loose bearing has sliding fit on inner ring); b support bearing (with peripheral load for outer rings,
the latter are flxed axially on both sides and the inner rings move in one direction); (: floating bearing.
~ ~
Mechanical Machine Components. 4 Rolling Bearings
outer ring (point load outer ring) or inner ring (point load inner ring). In a support bearing each bearing position takes on the axial force in one direction. The advantages of the support bearing lie in the simple construction of the bearing seats and accurate guiding in the axial direction, but the re is the risk of axial distortion owing to thermal expansion in operation. With a floating bearing axial guiding is deliberately dispensed with. End-float is limited to approximately 0.5 to I mm. Adjusted bearings (usually constructed with inclined bearings) offer the capability of a precise adjustment of a clearance or initial stress. A distinction is made between O-arrangement (small tilt play) and X-arrangement (small angular mobility). Allowing for the distance between the two rolling bearings, the arrangements are equivalent to the pairings shown in Fig. 30.
a
b
Bearings Prestressed and Installed in Pairs give little play and little resilience as individual bearings. Installation and load distribution possibilities: Fig. 30. Further configuration guidelines [8) and examples of installation from the manufacturers.
c Figure 31. a-c. Non--contact seals for rOlling bearings (sec [ext).
4.6.2 Selection of Fits The fit is to guarantee secure attachment and lmiform support of the bearing rings, as only then can the load capacity be fully utilised. The greater the bearing load, the greater the fit interference that should be chosen. If the fit is too loose there is the danger of the ring shifting when there is radial load (bearing ring revolving relative to the direction of load). Guide values for selecting the fit at the bearing seat positions, as a function of type of load, type of bearing, shaft diameter, load and displaceability, together with data on deviations of shape, may be found in the manufacturers' catalogues. The reduction in the radial clearance owing to an interference fit can be allo wed for in the cho ice of clearance group
for the bearing. In principle, both rings should be rigidly fitted. In loose bearings the fit of a ring must allow some displacement . The ring with point load should always have the loose seating. With point load the load is permanently directed on to the same point of the ring.
4.6.3 Seals Rolling bearings can function without trouble and achieve high service life only if they are protected by effective seals throughout their entire operating time, so that penetration of dirt and loss of lubricant is prevented.
Non-contact Seals Functioning Without Wear. Fig. 31a-c gives examples of seals in decreasing order of effectiveness from the range of non-contact seals functioning without wear. (a) Labyrinth seals, ideally with automatic regreasing and grease nipple regreasing, suitable even at high speeds for grease or oil mist lubrication. (b) Spring sealing washers work without wear after running in. They are best suited to rigid, non-dismantlable bearings.
(c) Gap seal with grease grooves for grease lubrication for high speeds. The sealing effect increases with the length of the gap. The use of the gap seal for circulation lubrication requires spray edges, spray rings or also oil delivery thread.
Grinding Seals Subject to Wear (Fig. 32a- d) (a) Slide ring seal made of synthetic carbon or metal for oil lubrication with automatic adjustment by spring parts suitable for peripheral speeds of up to = 15 m/s. (b) Radial sealing rings are suitable for peripheral speeds of 8 to 12 mls with oil or grease lubrication. The seal is provided by sleeves with sealing lips made of plastic, which are pressed on to the shaft by loop springs. The sealing lip is subject to wear and its life is therefore limited. If the seal is to prevent foreign matter penetrating the bearing, the sealing lip must point to the outside. (e) Felt rings are used fo r peripheral speeds of up to 4 m/s . Their performance deteriorates over time, owing to decreasing ring elasticity and splitting. They are therefore only used where dirt accumulation is low. (d) For designs with limited installation space rolling bearings with cover discs (sufftx Z) and rolling bearings
a
l!gg c
a Figure~.
b
c
a-c. Paired inclineo ball bearings: a O-arrangement , b
Z-arrangement , c tandem arrangement.
b
d
Figure 32. a-d. Grinding seals for rolling bearings: a slide ring seal; b radial sealing rings of various types; c felt sealing ring; d bearing with cover discs, bearing with scaling discs.
4.6 Design of Rolling Bearing Assemblies. 4.6.4 Influence of Bearing Housing Design on Fatigue Life
with sealing discs (suffIx RS) are used. Both seals lie at the bearing periphery. Cover discs may be designated nongrinding seals and sealing discs grinding seals. Bearings with discs and grease ftlling on both sides are considered maintenance-free bearings.
4.6.4 Influence of Bearing Housing Design on Fatigue Life
housing. The load-dependent elastic deformations of housing, bearing components and shaft produce variations in the theoretical load distribution on which the calculation is based. With diminishing housing wall thickness the maximum load on the rolling elements increases and fatigue life decreases [9, 10]. In the housing design a favourable load distribution can be achieved by variation of force introduction and stiffenings.
The distribution of forces on the rolling elements is dependent to a large degree upon the design of the bearing
Plain Bearings H. Peeken, Aachen
5.1 Fundamentals of Plain Bearing Design 5.1.1 Hy....odynamic Lubrication The sound fimctioning of machines demands wear-resistant design and construction of bearings in order that bearing forces are transmitted reliably and at temperatures that are still permissible. Wear-resistance exists if the sliding surfaces are separated from each other by a load-bearing film. In the plain journal bearing, for example, there is a load-bearing ftlm when the shaft is positioned eccentrically. The pump action of the rotating shaft delivers the lubricant into the bearing clearance and with convergent bearing clearance brings about the build-up of oil pressures (Fig. 1). The eccentricity of the shaft adjusts itself in operation in such a way that the integral of the oil pressures of the external bearing load F holds the equilibrium. An interruption in the bearing surface by lubrication grooves in the support zone reduces the load capacity. When the cylindrical plain bearing is loaded, the angular position f3 of the relative eccentricity £ as a function of width ratio B/ D follows a semicircular function.
The oil feed occurs conveniently in the unloaded zone. Ideally, this is the region behind the smallest film thickness hmin' The oil is quickly sucked into the bearing by the
underpressure here and in this way prevents admission of air and consequent foaming.
5.1.2 Friction Regimes in Plain Bearings The possible friction regimes in plain bearings may be explained using the Stribeck curve (friction coefficient [ plotted against the angular speed w for constant bearing temperature it) (Fig. 2). The friction coefficient [ is defined as [= 21>1fl (FD). For w = 0 (point A) there is contact between shaft and bearing shell. Here the law of solid-body friction applies in approximate terms, so that the friction coefficient [0 = JL = Fr/ F is determined by the shaft -shell material pair. The friction coeffIcient drops with increasing rotational speed and reaches the friction minimum at w = Woe (point B). When W> w",[rises again. The point C divides the area of mixed friction (which is associated with wear) in which in addition to liquid friction there is still solid contact, from the area of liquid friction. Only in the liquid friction zone is operation without wear possible, so that the operation point D must always lie to the right of C. The illustrated Stribeck curve in Fig. 2, curve a, is valid for constant bearing temperature and therefore constant viscosity. In practice, regimes with it const often occur. The increasing temperature has the effect, by virtue of the decreasing lubricant vis-
*
COSity
yt, of compensating for the friction value rising with
speed (Fig. 2, curve b) so that there can thus be a roughly constant friction value.
5.2 Calculation of Plain Journal Bearings Under Steady Radial Load 5.2.1 Wear Safety, Conditions for Full Fllm Lubrication The hydrodynamic pressure distribution p f fo A , Hydrodynamic bearing
=[(;P,
z)
F~const *~ const
Figure 1. Plain bearing schematic with pressure distribution. Des-ignations: F bearing load, R bearing shell radius, r shaft radius, [) bearing diameter, B bearing width, p oil pressures in sliding space, p. oil pressures with an oil groove in the load-bearing lone, rp and z coordinates. e eccentricity, h lubrication clearance height, hm;n minimum lubrication clearance height. OJ angular spl'cd of shaft, f3 angle of direction of shaft displacement. C = 2(R - r) operating bearing play, 2e/C = l-: rdative eccentricity, !/J=C/D relative operating bearing play, PI friction force.
*>const
/ "',- Hydrostatic bearing I
W,n
w
Figure 2. Stribeck curve (schematic), f coefficient of friction, Mr
friction torque of bearing, fo friction coefficient of solid body fric~ tion, W,r angular speed at transition to mixed friction, 11 bearing temperature. CUITe a for it = const, curve b for {} #: canst.
Mechanical Machine Components. 5 Plain Bearings
follows from the solution of the Reynolds differential equation
~ (h 3 dP ) + r ~ (h 3 dP) _ 6Ur ~ = o. dcp Tf dcp
dZ Tf dZ
dCP
(1)
Notation is as per Fig. 1 , with U as initial speed at the shaft h = CO + 8 cos cp) 12 as the idealised clearance height without taking into account deformation roughness). The integration of the pressure distribution in non-dimensional representation gives the hydrodynamic load capacity in the form of the Sommerfeld coef fictent So
= PIj?I(Tfw)
(2)
as a function of t~e relative eccentricity 8 and the width ratio BID. (Here P = FI(B· D) = mean surface pressure, 1/1 = C!2R = CI D = relative bearing play.) Figure 3 shows the load capacity according to Eq. 0) in the form of the Sommerfeld coefficient So with variable 8 and BID under conditions of incompressible Newtonian lubricant, laminar flow , absolutely rigid and smooth sliding surfaces, and axial parallel clearance. In order to ensure sound running, the operating point D of the plain bearing must lie within the non-wear zone of liquid friction, without a permissible maximum temperature being exceeded. Freedom from wear accordingly exists if the operational angular velocity W shows an adequate safety margin relative to W,,: W > w.,. It is assumed with generally adequate accuracy that w" defmes the transition to mixed friction. The recommended size of the safety margin between w and w" is given by (3)
where {U) is the numerical value for initial speed in m/s. The angular speed at the transition to mixed friction follows according to [1] for 0.5 < BID < "': (4)
with V = ",D 2B14 as the bearing volume and Tf the viscosity of the lubricant at operating temperature. C" is a
constant described by the clearance geometry C" = 2/(1tl/lhlim ) , in which the smallest tubricant film thickness h lim and the relative bearing play 1/1 appear. If we insert for example 1/1 = 2 . 10- 3 and h lim = 10/3 J.Cm = 10/3· 10- 6 m, it follows that C" = 1 . 108 11m. With this value for C" the calculation remains mainly within the safe margin. With optimal bearing design, low roughness levels and run-in bearings, higher C" values can be achieved. With heavily loaded bearings, whose sliding space is influenced by deformations of shaft and shell, C" is dependent on load [2]. Wear safety may also be derived via the ratio F"I F. F" is the overload of the bearing at the boundary with mixed friction. For 0.5 < BID < "': (5)
With plain bearings under dynamic loads, wear safety is given by the ratio of the smallest clearance to the smallest lubricant ftlm thickness which is still permissible hminl hi""· If, with bearings under static load, deformation influences are disregarded, the minimum clearance h min is generated by hmin = CO = 8)/2 > hum. The relative eccentricity 8 may be determined from Fig. 3 when the Sommerfeld coefficient is known. In order that no contact between the sliding surfaces should occur in the operating point, the condition bmin > hum is to be complied with. bum takes into account the sum of the roughness of bearing and shaft, any shape defects of bearing and shaft, shaft skewing in the event o~ alignment errors and - particularly with rigid bearing construction - shaft deflections. Guide values for h lim may be taken from Fig. 4 [3]. Figure 4 shows that the production roughnesses both of shaft R'J and of bearing R", are dependent on the bearing diameter D. The roughnesses of the harder component (generally the shaft) which take the peripheral load are a determining factor in calculating the minimum permissible lubricant fum thickness at the transition to mixed friction
h1imo
The roughness of the bearing changes more than that of the shaft as it is run in. For this reason the following approximation is also valid:
= 1.5R'J + 0.5R'B + deviations of shape.
h lim
(6)
In establishing w" in accordance with Eq. (3), no allowance was made for the mean surface pressure with which the mixed friction zone is traversed. It does have relevance for the life of a plain bearing whether the mixed friction zone is traversed with a high or low surface pressure. Therefore the limiting value Ps,U" = 25 . lOS W1m2 is inserted as the permissible friction work in the mixed friction zone. U" = w,/J/2 is the peripheral speed at the transition to mixed friction. The permiSSible limiting value
p"
3
~ 0.95 0.96 0.97 0.98 099 1.00
'';'- 1
.s;
20
Relative eccentricity t:--:;;"""',+--.j,..<~jL,.LA'./----j
hr"
I
10
OIB~l E
"-
.5
B i
,
/
-- .--
1
10· 3 ~----;!-:-----;:'-::--f::----::L--f::----::L,--~-L---,L----.J ~ ~ ~ ~ ~
o m
m M
M
W
Relative eccentricity £ Figure 3. Sommerfeld coefficient for fully enclosed journal bearings as a function of BID and e in accordance with 1201.
1
~i--
:z.. -
R~ ,----
- f-
r
1---
//
10
i
e 10 2
o in mm
Figure 4. Lowest lubricating mm thickness h 1im and peak-to-valley height Rt as a function of bearing diameter D [3]; 1 shaft, 2 bearing.
'i.2 Calculation of Plain Journal Bearings Under Steady Radial Load. 5.2.2 Calculation of Bearing Temperature
PS
5.2.2 Calculation of Bearing Temperature The calculation of wear safety in accordance with Eqs (4) and (5) requires the operating viscosity and thereby also the operating temperature to be known. The oil temperature in the bearing is derived from the balance between the friction work and the heat quantities per unit of time which are carried off by convection from the free surfaces A of bearing and shaft: aA (1'1 - tl,mb) or by the lubricant: QdCp( ,'lex - 1'1.). P, ~ M,w
= /FU ~
aA(tl - tl,mh)
-l-
QdCp(tl., -
1'1.). (7)
The relative friction coefficient//I/! of a 360 D bearing may, when B/ D = I, be determined approximately by the following equations: So:5 l:jN
~
K/So;
So
2:
l:/N =
K/ \So.
(8)
With pressure oil lubrication and a loose oil ring (Fig. 6), K ~ 3 is to be inserted; with a tight oil ring (DIN 118) K ~ 4 is to be inserted. More precise values for f! I/! are obtained from integration of the Reynolds equation as a function of e and B/ D in accordance with r1]. (a) Heat Dissipation by Convection. If heat is dissipated only by convection, e.g. in bearings with oil sump and good internal wetting of the bearing housing, the relationships for bearing temperature follow from Eq. (7). For So > I (heavy load range):
1'1- tl,mb
~ [4.25l1!(aA)] yFUB· y'1 ~
w·
(9)
:5
I (rapid action range): 6BU'
1'1 - tl,mb = rxA-;j;' '1 ~ W' ' '1;
At environmental temperatures of tl,mb oF 20 DC the relevant bearing temperature as a fimction of the bearing temperature 1'1'0 determined for ,'lomb ~ 20 DC may be found in Fig. 5 separately for So > 1 and So < 1. The bearing temperature should not exceed a limiting value of '1limi< = 60 to 70°C, as otherwise the oil degrades more rapidly. (High temperatures, e.g. in themtai engines, require special additives in the oil.) The calculated bearing temperature is taken as a mean housing temperature at which the heat arising in the bearing can be dissipated by convection. The temperature which varies over the lubrication clearance differs only slightly from this temperature when there is good equalisation of temperature. Load capacity diagrams [9, 10] give a brief overview of the complete operating range. Exam.ple
6BU' W' ~ aAI/!' (10)
For the heat-emitting surface of the bearing A which is determined from tht' bearing design, in the machine-mounted situation we can put approximately A = 15 to 20 BD. The coefficient of heat transfer a with moving air and adjustable speed w is derived in m/s for
bearing housings of any size from the equation
Since, in addition to temperature, the viscosity is often also unknown, to calculate the two unknowns '1 and 1'1 for the operating regime the temperature dependence of the viscosity of the oil used must be known. For the viscosity-temperature behaviour of lubricating oils we can put
a exp [bl,tl + 95)].
load
F = 10 kN,
n
=
1500
rev
min'- 1 ,
100 mm, B/D ~ 0.8, ISO VG22 DIN 51 519, ,,~20W/(mHC), A ~ 20BD ~ O.16m', U~ 7.85m/s, '" ~ 4-"/U/2.5· 1O.-~ = 1.3· 1O~\ w = 157 S··I, etc = 1· 101i l/rn. In accordance with Eq. (9): W = 826 DCI..jPa s: Appendix FS, Table 1: ttlO = 82°C and TI"p"rntill~ = 0.0052 Pa s. Check for So as per Eq. (2): So = 2.6 > 1, therefore assumption correct; no check necessary for So < 1. Check for transition zone (wear) 'as per Eq. (4): w" ~ 30.6s-', w/w" ~ 5 < ~9 3{U} + {U'} ~ 6.86, no
D
~
wear safety.
(b) Heat Dissipation by tbe Lubricant. If heat is dissipated from the bearing mainly by the lubricant, the Sommerfeld coefficient is determined approximately with the viscosity 71 effective in the clearance, and this is derived from the temperature 1'1 average from between entry temperature tJ{. and exit temperature tlt:x.
in W/(m 1 DC). As there is always air movement in machine rooms w "-;;: 1.2 m/s at least should he inserted; the resulting a value is at least a = 20 (W 1m2 CC),
'1 ~
Bearing
I),m', ~ 20°C,
,'1;
w~ [4.25UI(aA)] ,FUB.
For So
An explicit determination of temperAture it from Eqs (9), (10) and (11) is not possible. It can be done graphically with the coordinate charts for So > 1 and for So < 1 (see Appendix FS, Table 1 and Table 2). These charts illustrate the viscosities (in Pa s) of oils of various viscosity classes = viscosity grades VG as per DIN 51 519 and the straight lines W = const andlor W* = canst dependent on temperature for the ambient temperature ttamh = 20°C. The viscosity grade of an oil is derived from the kinematic viscosity v = Til p (p = mass denSity of oil) at 40°C reference temperature. Every point of intersection of the temperature rise straight lines designated by Wand W* with one of the plotted standard oils is a solution to the set of equations and gives, as intersect coordinates, the desired bearing temperature f)-zo (for 1'tamh = 20 DC) and the associated operating viscosity '1/. For determining the bearing temperature the coordinate chart for So > 1 can be initially used. If the subsequent calculation produces So > 1, the bearing temperature has already been found. if it gives So < 1 the calculation is to be repeated with the coordinate chart So < 1.
100fT!, 90-- -+---j--t---j--t'-7'I ,
80 .~
70
i oof ~
So>1
-+7'1'-7"+,0&"7
~~"o\,
'1\)'<:>",\1
i50~-+~-b~~~7L~~
(II)
According to Rodermund r4], the constants are derived from:
a
~ '1x
exp (-- bI887);
b ~ 159.56 In ('1 4o /'1x);
(n40
'1x ~ 1.8· 10 -4 Pa s
n40 in Pa s
the nominal viscosity at 1'1 ~ 40 CC). The dependence of viscosity on pressure is disregarded for plain bearings.
50
60
70
~
Figure S. Bearing temperature at ambient temperatures itaITlh 20°<;.
80
"*
Mechanical Machine Components • 5 Plain Bearings
With higher eccentricities the development of pressure shifts more to the smallest clearance, so that the mean
temperature ft to be inserted moves more towards the oil exit temperature. If we disregard the heat dissipated by convection in Eq. (7) (adiabatic case), the oil exit temperature is given by ftex = fFU/Qc\Cp) + fte and the mean bearing temperature is ft = (ft.. + fte)/2. (Q"I cooling oil quantity, Cp specific heat, for oil Cp = 1.8.106
With closed-circuit cooling of the oil in an oil radiatof, temperature differences of ltex - it" = 10 to 20°C can be achieved in nonnal designs of radiator. N/m/(m~ DC).)
S.2.3 Required Oil Flow Rate To build up a load-bearing film an oil throughput Q is required, which is given by the difference between the quantities flowing in the widest clearance and in the narrowest clearance
Q=
[1 - 0.223(BID)'] . BUC· e/2.
(12)
If the Sommerfeld coefficient is known, the relative eccentricity e is given by Fig. 3. In the case of pressure oil
lubrication in addition to the oil supply resulting from the turning of the shaft (Eq. (12)) there is the amount delivered through the sliding space by oil pressure p,. If the oil is fed through a bore with diameter d at the point where the clearance is widest, the following applies approximately in accordance with [7] .
7rD'IjJ'p
difference 1} - t1amh = ~ {} < 20°C the operating clearance roughly corresponds to the installation clearance. For bearings with free expansion ability and d tt > 20°C owing to friction heat or heat flux it is accepted approximately that the bore does not change and the heating only has the effect of expanding the shaft.
5.3 Calculation of Plain .Journal Bearings Under Variable Radial Load Here bearing load (according to magnitude and direction) and angular speeds of bearing and shaft are functions of time (e.g. bearings in internal combustion engines). It follows that shaft position, friction, lubricant throughput and load safety of the bearing are dependent on time. If the functions of bearing load and angular speeds are periodic, closed shaft centre paths result. The calculation of these paths starts from Eq. (1), which is extended by the term - l2r' ah/at and is gradually solved. The practical calculation of the displacement paths may be facilitated by the application of approximation functions [11] for the Sommerfeld coefficients SOD and SOy attainable by turning and displacement. With periodically loaded bearings the iteration is to be carried out until closed path curves result.
5.4 Turbulent Film Flow
Qs = 481) In (B;d) (1 + e)'. In the case of pressure oil lubrication and oil feed through a bore the overall amount of cooling oil is governed by Qcl = Q + Q". An increase in Qc\ over 3Q does not bring greater heat dissipation.
S.2.4 Relative Journal Bearing Clearance The relative bearing clearance IjJ = C/D necessary in operation is chosen as a function of the sliding speed U. As a guide the following numerical equation is used: IjJ = 4.,jU/2.5· 10-'; U in m/s. These values may be deviated from by ± 25%. The following infortnation is given with respect to the upper and lower limits for 1jJ:
Operating conditions
Lower '" range for
Upper'" range for
Bearing material
Soft, low modulus of elasticity, white metal
Hard, high
Area load
Relatively high
Relatively low
Bearing width
BID" 0.8
BID~
modulus of elasticity,
At high peripheral speeds the laminar mm flow in plain
bearings changes into a turbulent mm flow after a critical Reynolds number Reeti , is exceeded. Turbulence commences if Re
= pUC/21/
20 ReCti,
= 41.3/
[,fr,
(13)
with p the mass density of the lubricant. With turbulent flow too the plain bearing characteristic values may be determined theoretically by means of a modified Reynolds equation [12]. The equation contains position-dependent correction factors for viscosity, so that in tangential and in lateral direction higher viscosities apparently take effect. Depending on the type of bearing and the eccentricity e the turbulence zone begins above Re = 300 to 1000. With turbulence bearing temperature rises owing to the increased lubricating film friction, but so do the lubricating film pressure and, accordingly, bearing load capacity.
bronzes
0.8
Mounting
Self-adjusting
Rigid
Load transmission
Rotating (peripheral load for bearing shell)
Static (pOint load for bearing sheJl)
Treatment/machining
Very good
Good
Hardness difference between journal and bearing material
2:
100 HB
" 100 HB
With increasing bearing temperature the thermal expansion of shaft and bearing, which reduces bearing play, is to be compensated for by an additional clearance (installation> operating clearance). For bearings with free expansion ability and only small temperature
5.5 Calculation of Plain Thrust Bearings In the case of thrust bearings the convergent clearances needed for hydrodynamic pressure development are produced by machining, tiltable or elastic segments of a rectangular or circular shape (Fig. 6). Sufficiently broad spaces are to be left between the segments so that the
exiting warm oil can be replaced by fresh oil (use scraper if necessary). The choice of bearing type is determined by the operating conditions. If there are high surface pressures, and frequent speed changes under load are expected, bearings with tiltable pads are to be preferred as the optimum wedge inclination with correct support adjusts itself automatically and the wear occurring when the bearing is
5.5 Calculation of Plain Thrust Bearings
Z
-
I
c Figure 6. Design variants for plain thrust bearings. a Solidly worked-in wedge surfaces: Lwe wedge length, Dm mean diameter of thrust bearing and C""t: depth of worked-in wedge surfaces. b Rigid and
elastic support of pads for constant and changing direction of rotation. c: Rigid and elastic support of circular pads for constant and changing direction of rotation; d diameter of circular pad.
starting up or slowing down does not produce any change in clearance geometry. When dimensioning a thrust bearing. the main dimensions (Z load-carrying surfaces and/or pads of breadth B and length L or diameter d) are to be distributed in such a way that with = F/(ZLB) or = 4F/(brd'), U= wD/2, an average effective viscosity TJ and bmin > h 1im the load factor pb~n (1/UB) assumes the appropriate value. In Fig. 7a the load factors cPb:n'n/ (1/UB» and friction factorsj"JpB/(1/U) are plotted as functions of bmln/Cw.for various breadth ratios L/B valid for the lubricating wedge without rest surface. bm'n/Cwe = 0.5 to 1.2 and L/B = 0.7 to 1.5 give broadly optimal conditions. The smaller values for L/B are more favourable for heat dissipation. With tiltable pads, by setting the axis of rotation at a = 0.42L from the run-out edge the optimum ratio bmln/Cw • = 0.8 is obtained, at which the greatest load factor is achieved independent of operating regime. On the other hand, with solidly worked-in wedged surfaces of depth CWe> variations in operating regime (different bm'n) also bring about changes in the load factor. Generally, the most frequently occurring operating regime is chosen as the design point for the bearing. A check is to be made as to whether the wedge angle
p
p
resulting from the load factor via the ratio b mln / Cwo can still be produced. In the case of tilting segment bearings for both directions of rotation the support is concentric. Owing to the occurring thermal and elastic deformations approximately 80% of the load capacities as per Fig. 7a are achieved [I2J. Figure 7b gives load and friction factors for the lubricating wedge with optimum rest surface element (Ww./W = 0.8). In [13J the load capacities for various other clearance shapes are compared. Here the orifice plate bearing with parallel step clearance gives very good results. Bearings with tilting circular blocks are dealt with in [14], and with centrally supported circular pads in [12 J. For accurate calculations the elastic and thermal curvature of the pads is to be taken into account [12 J . Figure 8 shows a combined thrust and journal tilting segment plain bearing for the mounting of a turbo compressor. Particularly at high peripheral speeds (U > 25 m/s) the temperature difference between entry and exit rises with increasing pad length L, so that the mean viscosity in the clearance drops and therefore values smaller than I are chosen for L! B. In this case it is also recommended that there be a reduction in the surface utilisation 25 m/s friction heat can no longer be dissipated to the surroundings by convection alone. In that case return cooling of the oil heated in the bearing is necessary. Heat conduction by the alignment disc and pad is generally disregarded here. The temperature at entry into the load-bearing clearance is then derived from the mixture between the heated oil coming from the previous pad and the oil fed between the pads and coming from the cooler. Owing to the mixing of oil flows the temperature at which the oil enters the lubricating clearance (it,) is always higher than the temperature at which the fresh oil is delivered to the bearing. Stationary temperature conditions are produced in the bearing if the friction heat generated in the bearing per unit of time is removed from the bearing either by convection or by return cooling of the lubricating oil. Therefore with it, as the oil sump temperature, it. the temperature of the return-cooled oil, it=b the ambient temperature and P, the friction work, we have convection: return cooling:
P, P,
= fFU = cr.A( it, - it,mb)' = fFU = !?cICp (it, - it.), (14)
with Qcl as the quantiry of cooling oil. To cover the mix between the hot oil leaving the pad
Mechanical Machine Components. 5 Plain Bearings
I ///{ &//I
.c;.~
F
!
/. I~ -- - -
q
-1 -
-~
-
J •
i I 1'- - --- - '
~
C
10'
j
I
;~
. -~---"'
1
b
,,
~
__
~
____
~~
a 10 10
____
~
____
~
______- J
Ie
a 1 ".,.1[ ••
Figure 7. Load and friction factors lOOP~;n (,."llB) and tJpB( 1JU) for the lubrication wedge: a without rest surface for various width ratios, b with optimal rest surface (LwJ L = 0.8) for various width ratios.
Taking into account the mixing processes the maximum
temperature in the lubricating film is
With the temperature at clearance entry relationship
~, - ~, =
m (1
~,
- ~) (~2 - ~,) ,
from the
(16)
the mean lubricating mm temperature/bearing temperature is given by ~ =
(il, +
~2)/2.
(17)
The entry temperature of the return-cooled oil is given by (18)
Figure 8. Combined axial and radial tilting segment bearing for a turbo-compressor (Demag-Verdichtertechnik, Duisburg), shaft removed.
and the cooled oil supplied the mixing factor m is introduced, for which m = oA to 0.6 is recommended. m = 0 means that the lubricant leaving the lubricating clearance fully enters the input clearance of the following pad. Only the lateral flow {h is replaced by fresh oiL m = 1 means that no mixing takes place. The entire oil supply to the segments is with fresh oiL
The derivation of these equations presupposes that the oil quantity 12" leaving the lubricating wedge at the side only, warms up on average to (~2 + il,)/2. The oil quantity entering the lubricating wedge, 12" and the oil quantity leaving as a lateral flow, 123, may be determined from Fig. 9 . As an example) for pads with optimum support without rest surface (a = OA2L measured from the run-out edge) bm'n/Cw< =' 0.8. The minimum oil quantity required by the bearing for hydrodynamic load transmission, namely the load-bearing oil quantity Qr, is derived from QI
0.. = ZQI'
(19)
Wear safety of the thrust bearing is provided as long as
I[B
5.5 Calculation of Plain Thrust Bearings
1.6,------,----,----.,--,-,-----,--,-,-----, 1.4
1.2
,
.-1
I
~
1.0
BIl.D.\
~
~
.",
1
0.75'
0.8 0.6
0.4 t--------..r~ 0.2~--"?_::_-:::'::-----;;',-;;':-:-;!-;:._'____;;_';;_L:---.-~
0.1
0.4 0.5 0.6
0.8
b
l-
. ~i-I----;--fi-+rtt--i
+-
+-.
OJ4 f---B1-H ---'.S,4,-
I
I
-.-.1
I iii I
1--1 I
0.14 .- - - - --- 0.10
i
I I
D." 2.5 C.6
c
I
,
I
i
1
.----
_
t---I
:.18 - - - - .• -----+---~-:- -. --+-J-,+-_--.-----1
I.S-':
i
l I
I I
OJ
d
:
I
I
G.8
T. I!
i--------"---~:
r--'
I
0.06L-.--------::---::---:--~-'-"-----------'
OJ
.
;.. --
1/
,
2/---_-_~·_-_--·~--~1-·~·----~ 0.2
-----:-:r=-=~ 0.75,
0.14
I
0.1
0.26
~ 0.22
'e>
-1
-
! -~..
G
1
--
OJO I
; ,
Iii
OJ8 .-----.,--1----::--.--.--,-'.,..-1-------,
I
I
I
I
!
'!
i )
0.4 0.5 0.6
J
,
f--j------1
I I
0.8
I 1
,i..,w/Cwl'
Fipre 9. a-d. a Flow factors QdBUCwc (entry to wedge clearance) for the lubrication wedge, without rest surface for various width ratios BIL. b Flow factors QtlBUCwc (entry to wedge clearance) for the lubrication wedge, with optimum rest surface (LweiL = 0.8) for various width ratios. c Flow factors Q3/BUCwr (lateral flow) for the lubrication wedge, without rest surface for various width ratios BIL. d Flow factors fb,/BUCwe (lateral flow) for the lubrication wedge, with optimum rest surface (LwelL = 0.8) for various width ratios.
the minimum film thickness in operation b m.. does not exceed the minimum film thickness at the transition to mixed friction hm,n, U' The ratio b mln/ b mm. rr is a measure of the safety margin considered necessary from the wear limit. b m,n." is determined, in addition to the roughnesses, primarily by production tolerances, assembly errors and elastic and thermal defonnations. As running·in operations are only possible to a limited degree, b ro,n." is generally greater with thrust bearings than with plain journal bearings. Guide values for the selection of b m,n." are given by
~DR./3000
b m,n ." = ) for closed bearings with solidly worked-in (20) wedge surfaces, and b m,n." =
~DRz/12000
for tilting segment bearings (R
calculated for the lubricating wedge with and without rest surface approximately by b m1n = VO.06T/UB/p.
(21)
Guide values for the smaUest pennissible minimum lubricating film thickness for thrust bearings are given by bm'n."m = V5D. 10-' m.
(22)
For the mixed friction zone to be traversed smoothly when the bearing runs out/slows down it is crucial that the peripheral speed on reaching the mixed friction zone does not exceed the limit U" = 1.5 to 2 m/s. In addition to a constant dead load, thrust bearings in many cases also have to bear loads which are dependent on rotational speed (e.g. in turbomachinery). If the thrust bearing only has a constant dead load to bear, the mixed friction zone is only crossed when the bearing is starting up and slowing down. The operating speed of the bearing must therefore be greater than the transition speed by a corresponding safety margin. If on the other hand the load
Mechanical Machine Components. 5 Plain Bearings
on the bearing is only attributable to kinetic forces, e.g. fans with horizontal shaft, or in ships' thrust blocks, the load on the bearing decreases more quickly than the bearing's load capacity, so that there is no lower mixed friction limit. Here however a mixed friction limit is reached in the upper speed range, so the operating speed must keep a sufficiently large safety margin relative to this limit. If the bearing load consists of both dead weight and kinetic forces (e.g. pumps with vertical shaft), mixed friction zones occur in the upper and lower speed range, and snfficient safety margins must be observed relative to them.
10,------,----,-----,-----,-------r----, Shaft and shell rigid
.--l....
shell rigid
5.6 Form Design of Plain Bearings The bearing design creates the preconditions for the hydrodynamic lubrication on which the bearing calculation is based.
S.6.1 Influence oC Design on the Configuration oC SUding Surt"aces In the support of shafts the bending line generally shows an inclined position which can be increased further still by alignment errors, and by offset and angle error of the bearing blocks. The edge loading thereby caused produces a reduction in the bearing's load capacity and requires design measures to adjust the bearing to the distortion of the shaft. This can be done by a one-off adjustment on assembly by means of spherical mounting of the bearing body or by continuous adjustment to shaft inclination by a tilting mounting. With bearings fitted in engines, self-adjustment can be obtained by eccentric support of the bearing body (Fig. lOa) or by the use of an elastic membrane (Fig. lOb). As with inclination, shaft bending in the bearing also results in a reduction of bearing load capacity. The load capacity reductions, which are particularly visible where BID > 0.3, require design measures so that there are as far as possible parallel axial clear-
ances. With bearings of greater breadths satisfactory results are given by resilient design of the bearing support brackets so that optimum adjustment to shaft deformation is achieved. Incorporating the elastic deformations of support brackets and shaft in the bearing calculation [16] it is possible to give, for example for concentrically arranged bearing support brackets, optimal wall thicknesses Sci D as a function of support width biB (Fig. 11). However, the studies [17] show that with cylindrical bracket shapes only a relative optimum in respect of load capacity can be achieved. Further increases in load capacity can only be achieved in operation by running-in procedures. A more precise adjustment of shaft deformation to bearing shell set is produced by a taper cone, but here the bearing
5' - - - - ' < - - - - " - : - - 01 o
0.4
0.6
Figure 11. Load capacities So of cylindrical plain bearings with centre bar of various widths b/ B as a function of the bearing wall thickness SL/ D.
edges must display snfficient rigidity. With thrust bearings the inclined positions of the footstep may be balanced by elastic design of the bearing ring support or by seating the segments on elastic elements (Fig. 6b).
S.6.2 Lubricant Supply Furthermore, the bearing design has to fulfil the requirement for an adequate lubricant supply. With free-standing pedestal bearings, the oil supply is effected by loose or fixed lubricating rings (Fig. 12) or by circulation lubrication (Fig. 13). The operating limits of loose lubricating rings are approximately U = 20 mls and of fixed lubricating rings U = 10 m/s. The lubricant is delivered most effectively in the zero pressure zone Of in the underpressure zone (divergent clearance) in order to prevent the tendency to froth. Outside these areas the resistances opposing bearing oil flow must be taken into account. The distribution of the lubricant within the sliding space is through wellrounded axial oil grooves or oil recesses in the zero pressure zone of 0.7 of the bearing width. Individual oil feed holes are generally not adequate. Annular grooves in the bearing centre give a good oil supply, but divide the bearing into two halves of lesser load capacity. Oil grooves in the load-bearing zone (Fig. 1) disturb the pressure buildup and are to be avoided in rotating, as opposed to oscillating, bearings. In combined thrust-journal bearings the lubricant supply to the thrust bearing is often provided by the oil flowing off from the side of the journal bearing. In the case of bearings under variable load best results have come from annular grooves for the oil supply and axial grooves in the unloaded zone. Oscillating bearings (e.g. piston pin bearings) have for the supply of oil
a Figure 10. Adjustment capability of plain bearings by a elastic deformation with eccentric support, b elastic deformation of a membrane.
0.5
Figure 12. Lubrication variants.
5.6 Form Design of Plain Bearings. 5.6.5 Bearing Shells (Bearing Liners)
,-
,
JI
I
\
I
\
I
-
,... _ 1._,
I
__ L
,-1
~
~--
- - -
- -1 -
I I
L
j
t-
Figure 13. Plain bearing with oil <.:in.:ulation and ring lubrication (ACCi-Telefunkt"n).
several grooves ntmling straight or obliquely in the pin and terminated at the bearing ends , these grooves being supplied with lubricant via a hole . If oil foaming in the bearing cannot he avoided then the frothed-up oil whICh no longer meets the incompressibility requirement must be removed from the bearing by scrapers. At the same time the scrapers prevent the hot oil that is draining out from re-entering the sliding space 120] . 5.6.~
Bearing Cooling
The cooling effect of the lubricant can be improved hy increasing oil throughput, e.g. by partly hollowing out the upper hearing shell. Heat dissipation can also he improved by suitable guiding of the hot oil draining out in the housing of a pedestal bearing. Additional bearing cooling is possible hy the installation of cooling channels or pipes carrying oil or water 118] 5.6.4 Bearing Materials
'Ine bearing material 12 I - 241 must in conJunction with lubricant and shaft material display good sliding and dryrunning properties, and adequate behaviour in respect of wear , running in and bedding in .
As regards sliding properties the wetting power of the luhricant on the sliding surfaces, i.e . the ability to penetrate narrow gaps, is of major importance. This applies
particularly if in the mixed friction range (starting up and slowing down) the generated film pressures are not ye t sufficient to fill the clearance. Running-in bebaviour identifies the ability of plain bearing materials to match surface condition anti configurdtion of running surfaces to each other by abrasion, in such a way that design and production errors, which manifest themselves as alignment errors, deformations and other deviations from specified shape (undulation or roughness) are compensated for. The matching of the bearing surfaces in oper· ation is assisted by the elastic resilience of the hearing materials. With steel as the shaft material, sliding chaf'dcteristics and ntnning-in behaviour decrease in the following order: white metal on lead base , white metal on tin base, lead bronze, gun metal, tin bronze. special hrass . Dry-running property is the capacity of a bearing metal to keep the bearing operable for a short time without great damage even if the lubrication fails . In dry ntnning
residual oil and any solid lubricants present (graphite, molybdenum sulphide) help. Mainly, however, the dryntnning properties are determined by the material chaf'dcteristi" of the metals. Low-melting-point metals with little hardness which melt with local heating and thus reduce friction have the hest dry ntnning properties. Beddingin behalliour identities the ability to embed dirt or wear particles into the surface. The risk of damage to the sliding surfaces can therehy be mitigated. Materials that bed in well (e .g. white metal) in no way ohviate the reqUirement to protect the bearing from the entry of dirt and to keep the luhricant clean by ftlte ring. Wear arises in plain bearings in the area of mixed friction (e.g. during the startingup and slowing-down phases) . Quantitative wear values for bearing metals depending on the opef'dting conditions are only available in limited measure. Wear resistance declines from the bronzes via brass, AlPb bronzes, gun metal , AIZn alloys and cadmium alloys down to the white metals . Plain hearing materials must transmit the forces acting o n the bearing to the surrounding constCu<.:tion for
an adequate lifetime (load capacity). The pressure distribution causes within the bearing material a three-dimensional stress state which can lead to fatigue when there is variable loading. In addition, the temperature grd.dients produce thennal stresses on the plain bearing material. In this respect it should be noted that strength reductions occur above all in bearing materials with low melting point. In the case of statically stressed bearings the material design is undertaken using permissible mean surface pressure. Appendix F5, Tables ~ and 4 give chemical composition, strengths, spheres of application and permissible mean surface pressures for a selection of bearing metals and plastics. With bearings under variahle load the functional reliability of the material is conditional upon the fatigue strength not heing exceeded. Calculation statements regarding the enduf'dnce design of plain bearing material determine from the available three-dintensional stress state, using a suitable strength hypothesis, a comparative stress which is compared with the fatigue strength characteristic value determined experimentally [25] .
5.6.5 Bearing Shells (Bearing Liners) Plain bearing sheUs are furnished with bearing metal by the lining or centrifugal methods or are manufactured in
Mechanical Machine Components • 5 Plain Bearings
a multi-component version. In particular cases the bond between bearing metal and supporting construction is also made by solder or adhesive. For bearings under high dynamic stresses, such as engine bearings, multi-component bearings - generally manufactured by the band lining or roll cladding methods - are used. With a three-component bearing the bearing metal layer (e.g. lead bronze or aluminium bronze) bonded to the steel backing is also provided with an electroplated white metal layer to improve its running behaviour. Since strength increases with declining layer thickness, bearings under high loads require very thin layers. Three-component bearings therefore possess e",remely thin white metal layers, approximately 0.02 mm thick. Since lead bronze or CuAI alloy as a carrier material possesses adequate fatigue strength, the layer thickness is not of such overriding importance. Owing to the low ductility relative to white metal the layer thickness must be between 0.4 and I mm depending on the bl"'ring dimensions. The thinner the bearing metal layer, the less the bearing is able to compensate for example for alignment errors owing to plastic deformations in the bearing metal, and the more important is the correct design of support shell and/or bearing block.
5.6.6 Special Materials for Plain Bearings
Sintered metals made of Fe powder or bronze (Bz) powder pressed and sintered in part with additions have a volume of voids of 0 to 3% or 10 to 45%. Bearings with volumes of voids> 10%, which possess an oil supply through oil saturation, can nm without maintenance. Operation is limited to low sliding speed of 0.3 to 0.5 m/s at up to 600 N/cm'. The stress capaciry of the bearings is limited by heat dissipation, so that at higher speeds the load capacity drops.
P
Graphited Bearing Metals. Fe or Ilz sintered metals impregnated with grdphite (up to 10%) are used as oilfree hearings in the food and textile industries. In the
chemical and electrical industry bearings made of pressed colloidal graphite are also used, often also impregnated with metal powder. Graphite bearings have good sliding properties and are therefore also used as air bearings [I J .
Plastics and Rubber. In contrast with metallic mate!ials, here the tendency to scuffmg/seizing is totally absent. The materials possess low thennal conductivity, low modulus of elasticiry and are susceptible to high temperatures. Rubber is often used in water lubrication. With chemically aggressive media ceramics or carbon are used as bearing materials.
a
Figure 14.
Multiple~sliding~surface
bearings: a for both directions
of rotation (Gleitlagergesellschaft, Gottingen), b with three faces for one direction of rotation with pressure distribution and load capacities for s ::::: 0.6 (Caro Metallwerke, Vienna).
spring constants and the four damping constants, which depend on the static operating state of the bearing, by the solution of the movement equations of the rotor. The calculation of the spring and damping constants is generally based on a linear disturbance formula for the oil pressure components. It should be noted that with multiple sliding surface bearings, as a result of variable load, bearing capacity fluctuations occur which for their part can lead to induced oscillation [30J. In order that the bearing temperature does not become unacceptably high in the case of fully enclosed bearings, relatively large degrees of play are required which however promote the change from laminar to turbulent clearance flow (with resulting increased bearing friction). High friction losses may be reduced with radial tilting segment bearings, in which the running surfaces only partly enclose the shaft. With approximately point support of the segments the bearings are relatively insensitive to shaft skew. With bearings subject to high loads the segments may be produced in differing lengths. Figure 15 shows the construction of a three~segment bearing for a large steam turbine. For the calculation of this bearing taking into account the dependence of viscosity on temperature and pressure. see [31J. For certain applications spiral groove bearings are used as axial and radial guide elements [32, 33 J.
5.8 Bearing Seals The sound functioning of plain bearings requires adequate sealing of the bearing interior, in order to prevent the
5.7 Lobed and Multi.pad Plain Bearings With small Sommerfeld coefficients. bearings show increasingly unstable running behaviour (e.g. high speed steam and gas turbines, turbo-compressors), such that the required quiet running necessitates the use of multiple sliding surface bearings with three or more faces at which the shaft is held by several pressure peaks, whose load capacities add up geometrically (Fig. 14). Multiple sliding surface bearings also include bearings with lobed clearance. Multiple sliding surface bearings are also used for precise and largely load-independent radial shaft mounting. The calculation of load capaciry, friction, spring and damping constants is ascribed to that of a part bearing with arbitrary eccentriciry and angular position 128, 29J. The shaft or rotor movement around the position of equilibrium may be determined with the four
I -r----,-
SedIOO C· D
Figure IS. Tilting segment journal bearing of a large steam turbine (BBe).
5.11 Hydrostatic Bearings. 5.1 1.1 Journal Bearings
ingress of foreign matter. For (ommon types of seal, sec F46.3.
5.9 Dry Bearings Oil-free plain hearings are employed as mainrcnanct:'-free bearings for moderate loads and speeds where for economic reasons expense on lubricants is not possible (household, agricultural and office machines) or where lubricants are undesirahle for reasons of cleanliness (food and textile machines) With low loads air bearings are possible II I. Sintered metal materials, plastics slich as polyamides, polythenes and phenolic resins. and carbon are used as bearing materials. The most favourable running characteristics come together with hardened steel shafts of low roughness. By in<.:orporating metal powder into the plastics their the rmal condu<.:tivity can he improved 134 - 39 J The calculation of dry hearings covers bearing temperature, mechanical load capacity, and thereby life . As the sound operation of these bearings is cntcially determined by heat dbsipation at still permissible ~earing temperature . the operating limits are given by (jJl 0 pnlll values. For bearings with OSCillating swivel movements, in large areas of machine construction ball and socket joints, consistin~ of outer and inner ring with spherical sliding surfaces, are used 1401
5.10 Bearing with Hydrostatic Jacking Systems Hydrostatic starting aids facilitate the transition through the mixed friction zone when large machines are started up. By means of a pump the high-pressure oil is delivered through a nOll-return valve into the recess or groove provided for the purpose. Experience indicates that a pressure of'; to 6 times is require d for the suspension of the shaft . It is only possible to prove suspension hy means of a precise calculation [411 The oil connection in the hearing is to he designed in such a way that penetratioIl of the pressure oil between bearing metal and sllppon shell is prevented 142).
Ii
5.11 Hydrostatic Bearings In hydrostatic bearings pressure oil is pnJouced outside the bearing with a pump and fed to pressure chambers. The lubricant flows off through narrow gaps whose size h is determined by the load . Hydrostatic hearings also function at zero speed, so that the brdnch of the Stribeck curve lying to the leli of the transition point C (Fig. 2) in the area of mixed friction (the dotted line) is missing. However, the ability to operate without wear at any speed has to be achieved at the expense of increased effort for the oil supply by comparison with bearings working hydrodynamically. There are severdl possibilities for generating pressure in the pressure chambers: ( a) a displacemem pump for ea<..:h pressure chamber, (b) common pressurisation for all the pressure chambers, with a throttle conne<..:ted in front of each pressure chamber to stabilise the hearing. Capillary tubes or diaphragms are used as throttle elements. The diaphragnls have the disadvantage of a higher risk of clogging. Oil supply systems that deliver constant amounts of oil per chamber make throttles for stabilisation superfllH)Us.
Figure 16. HydrostatiL· journal bearing (schematic) with oil drain grooves and four recesses.
5.11.1 Journal Bearings In hydrostatic journal bearings [43-45 J bearings with (Fig. 16) and without oil drain grooves (Fig. 17) are used. The hydrostatic journal bearing shown in Fig. 17 is supplied with pressure oil by a common annular groove from all the recesses via built-in capillary throttles. The approximate calculation of the bearings is hased on the statement of now equilibrium at each bearing recess. If there are no oil return grooves the equalising t10ws in tangenrial direction between the recesses are to be taken into a<.Tount. For hearings with oil return grooves a simplified calculation method is given which presupposes a large recess depth I = 5C and requires a relatively small effort (C hearing c1eardnce). The illustrdtion uses the following faClofs: 1I = 8'/(c'A) throttle factor. W = 1'jw/(Pambtl/') speed factor and l' = PZ/P,l(Ub pressure ratio . whe re 8 ~ the diameter of capillary tubes, A = the length of capillary tubes, 1) = the operating viscosity, P,mh = the ambient pressure , pz ::= the feed pressure , c = the rddial hearing clearance, and other designations are as in Fig. 1 . The recess geometry of the bearing with Z recesses is described by the reduction factors i = hIB,} = L,/L" and k = L.!./ L1 as in Fig. 16. With laminar flow in the gaps we ha ve for the mean surface pressure of the bearing and z 2" 4: - ~ -F P
BD
'IT} (I 16
= -
+
k)(l
{I
. + I)Pz -- - -
H,
I }1\ ,
Hz
Figure 17. Hydrostatic joufI1al bearing without oil drain grooves with four recesses at the circumference (Konings, Swalmen. Netherlands) : a hearing lining made of plastic b pressure oil feed . (. capillary tubes, d pressure measurement connection, e pressure oil supply, I c..-onnection for additional cooling circuit.
, •••1.1
Mechanical Machine Components. ; Plain Bearings
with
64 1 + 3'fr'; G(l - e)" H, G
=
64 1 + 3'ITu G(I + e)',
= Z(l +
i)(B/D) + 'frjO + k) . 'frj(l - k) Z(l - i)(B/D)
The factor ,1 which it contains covers the curvature of the recess: A =
;;d~k5 sin [~O + k)
l
PH
For Z = 3 recesses, = (2/3)P applies, In rough terms, (0,2 to 0.3 )pz can be expected for normal e = 0'; to 0.6, Figure 18 shows the mean surface pressure relative to Pz, plotted against the eccentricity e in comparison with measurements. The simultaneous insertion of exact calculations shows that the influence of the bearing instal(the angle between bearing reference line lation angle and load) is of subordinate importance. The illustration in Fig. 18 is based on the optimum throttle characteristic value
P=
P,
0.2
06
04
0.8
Figure 19. Relative load capaCity PIPz of hydrostatic journal bearings as in Fig. 16 as a function of e with various recess numbers Z.
n
U
opt
'fr/G
64) = (,3 -
(I - e')\/l
The relative friction coefficient is, up to e very well by
~=
.
which guarantees the maximum possible load capacity. Figure 19 shows the influence of the recess number Z on the load capacity of the bearing. At higher speeds the bearing's capacity increases due to hydrodynamic influences. Oil consumption is given by
= 0.6, rendered
R, '~ikj, (I + k 'F~-~) ,
[ 1-~+~-'+2]. - e 1+ e Thus the friction work of the bearing P, = IFf! and pump output Pp = pi]",,/ t; may be determined (t; = the efficiency of the pump). Studies on bearing optimisation show that the lowest energy dissipation from friction and pump output is reached at So = 1 and £ = 0.; to 0.6. The friction work declines as web widths decrease; however, web widths should not fall below a particular size, so that bearing force when stationary can be transmitted without damage to the webs (i = k = 0.7 to 0.8, j = OB;). Bearing temperature and thereby operating viSCOSity is deter-
mined on the basis of the heat balance as per Eq. (4).
5.11.2 Thrust Bearings Figure 20 shows a single-acting circular hydrostatic thrust bearing with capillary tubes as throttles. With laminar flow in the bearing the bearing's capacity is given by F =
~ L=_p~
r.' &
.
2In(l/p) 'H'
with
+
64 3u In(l/p)' ~---
p,
Figure 18. Surface pressure relative to feed pressure pz, as a function uf e with the installation angle n as pardmeter. The graph is valid for n = 4, w = 0, U = 40, i = 0.75, j = 0.85, k = 0.64, v = 10,20,50,40.
Figure 20. Single-acting hydrostatic thrust bearing with capillary tubes as throttles. rj recess radius, ra external radius.
'i. 11 Hydrostatic Bearings. S.II.2 Thmst Bearings
u ~
8'
i/i..'
p ~
':MI,II
the undimensioned factor 1fr approximately assumes the numerical value 1.
rJr".
The resulting oil throughput of the hearing is . 7T pTh' Q ~ (, . 1)ln(I/,;) The total energy dissipation P is given hy tlw sum of fric· tion work P r and pump output Pp: P ~ PI -I- P p. where P,
=
(7T/2)(1 - p')1),v2 r:/h, .
Pe =
2In(l/p)
p, . Q/~ ~07T( 1
1'211' If ~
pl)l' 1)r;
The minimum possihle energy dissipation is achieved if
For the clearance h the permissible minimum value hm;nlim is to he inserted. The pressure ratio should be no greater than II =::: 3. Only for high rigidity requirements may greater pressure ratios II be chosen. In any case, two-thirds of maximum rigidity is already obtained with II = 3. The energy dissipation increases with ...JR. If F and ware fixed 1Jf = I then can be achieved by variation of Yj and rJ' Care should be taken to ensure that the viscosity does not fall below a minimum value of I.e; mPa s, so as not to endanger the oil pump. For the calculation of double-acting hydrostatic thrust bearings, see 14':;1,
Belt and Chain Drives H. Mertens, Berlin
6.1 Types, Uses Belt and chain drives are used to convert rotary speeds and torques between two or more non-coaxial shafts, even when there are considerable distances between shafts, at low constmction costs. Traction is provided hy endless flat helts, V·belts, synchronous belts or chains which are wound round the pulleys or sprockets on driving or driven shafts, thereby transmitting peripheral speeds and forces [II.
Frictionally Engaged Belt and Chain Drives. These require a minimum initial tension in order to maintain frictional engagement. The rotary speed is converted when the layout is correct, having a limited. load·dependent slip rate (slip due to elongation) and a near constant (Fig. 1) or infinitely variahle (e.g. Fig. 8c) speed ratio.
Positive Belt and Chain Drives. These too require a type-dependent minimum initial tension to achieve optimum nmning characteristics combined with a long service life and/or to prevent faults in transmission (skipping teeth) (Fig. 2), They then produce a constant speed ratio, leaving aside the usually slight mismatching of the torque transmission with the frequency of the advancing teeth or chain links (polygon effect). As they can he easily twisted, flat helts, V·belts and synchronous helts make it possible to constmct spatial drives with non·parallel shafts (Fig. ~d, e). Steel chains
Figure 1. Friction belts' a tlat helt, b V-belt, c round belt, Each with pulley.
Plate retainer
a~
~ b
~~l
c
Figure 2. Positive belts and chains: a roller on bushing chain on sprocket. b inverted tooth chain on sprocket, c synchronous belt on synchronous pulley
are only suitable for drives hetween parallel shafts, The centrifugal forces increasing as the peripheral speed v of the belt or chain increases reduce the peripheral forces which can be transmitted. Maximum power is therefore transmitted at an optimum peripheral speed vop , of the belt or chain, which however is usually dependent on the minimum size of pulley.
6.2 Flat Belt Drives 6.2.1 Forc:es in Flat Belt Transmissions Peripheral force is transmitted between belt and pulley hy means of shear stresses. Eytelwein's expression for the limiting case for slip throughout the arc of belt contact (shear slip, see Al.l1.2) applies: F;/F; = e~#, where F; and 1'; are the belt surface forces without centrifugal force and the arc of contact {3[rad] = (7T/l80) {3[degrees] , see Fig. 4. During normal operation the belt on each pulley first passes through a static arc {3, in which the belt does not slip on the pulley and then an active arc {3w = {3 - {3,. Shear stresses are transmitted by means of static friction in the static arc and by sliding friction in the active arc [2]. If one leaves aside the shear stress transmission in the static are, then Grashors expression for belt surface force ratio F; / F~ = e M /3 w applies. In design calculations the
'4,.g
Mechanical Machine Components • 6 Belt and Chain Drives
~ ~
b~ Figure S. Forces acting on a pulley.
F:
using the permissible belt surface force of J.l and /3
as a function (4)
Figure~. Planar (a to c) and spatial (d and e) drives: a openbelt drive, b crossed-belt drive, c multiple-shaft drive with flat helt, d spatial flat-belt drive with three guiding idler pulleys L, e spatial synchronous belt drive.
Tbe reduction of yield which occurs as the arc of contact decreases is expressed by the arc factor c~, which is related to /3 = 1T or 1800 Arc factor c~ = k~/k", where J.l = const; where f3 < 1T: ell;o. f3/1T = /3[degrees]/180.
6.2.2 Stresses Single-ply Flat Belts. The stresses for single-ply belts are obtained from the forces and the belt cross-section A = bs. For multiple belts these stresses can only be considered as hypothetical, calculated average values: Belt tensioning
rJ'1
Effective stress
(Tn
Centrifugal stress Figure 4. Designations on open-belt drive with index 1 for the smaller pulley.
full arc of contact f3 of the smaller pulley is assigned to the design load
lTf
= FI/A, rJ', = i'2/A. = Ful A = IT] - lT2 .
= FrlA
=
The bending stress is obtained from the bending elongation in the arc of contact of the smaller pulley. Bending stress rJ'b = Ensb = Ens/dwi (En = modulus of elasticity on bending, en = beJt elongation on bending, s = belt thickness) . Maximum stress:
(5)
(I)
The centrifugal forces acting in the arcs of belt contact to reduce the load pressure are supported by the free side
of the belt, thereby acting as a uniform centrifugal force = pv'A = qv' throughout the belt (p = the average thickness, A = the cross-section of the belt, q = the mass of a belt per unit length). Effective belt surface forces F; = FI - Pr = mF~; F~ = F2 - Fr = F;/m; peripheral force (effective load)
F~
F"
= FI
.- F,
= F:
- F;
= F:
(I -
l/rn),
maximum belt surface force
v2 •
With quarter-tum (crossed in alternate directions) and crossed-belt drives the belt undergoes an additional crosswise tensioning Us at its edges, so that lTma>IC.s = U]
+ Un + IT.,.
Multiple Flat Belts. In multiple flat belts (Fig. 10). which comprise a high-strength, stress-bearing tension ply Z, a friction ply R for transmitting frictional force internally and frequently also an outer ply D or a further mnning ply (for multiple drives) on the outside of the belt, widely varying stresses are produced in the individual plies wben elongation occurs. When bent the position of the neutral bending surface in the belt depends on the thickness and modulus of elasticity of the individual plies. Figure 6
The elastic Jorce oj tbe sbaft Fw which generally does not point towards the bisecting line of the angle of f3, but which does govern the bearing load, as shown in Fig. S, is expressed as Fw
= ,fF'; + F'~
- 2F'1 F' 2 cos f3m.
(2)
The pull Jactor
2m cos /3.
(3)
The yield k denotes the peripheral force FH obtainable
a
b
Figure 6. Stresses and strains in multiple belts. a Under tensile load. b Under bending load (n neutral surface).
6.2 Flat Belt Drives. 6.2.5 Coning Action of Flat Belts, Tensioning
shows the qualitative stress distribution under tensile and bending stress. To simplify matters, in practice only the peripheral force per belt width F;, that is pennissible for the particular type of belt is used as a basis for the design of multiple belts as welL This also takes into account the alternating bending stress that can be withstood in
Jl ---lA, u\~ 0\0
:\\0,\0\\ I
,
The effective running diameter d w of a belt is determined by the position of its neutral bending surface in the arc of contact. To simplify matters when making approximate calculations the pulley diameter d can be used instead of d w . The expression for single-ply belts is: d~ 1 = d , + s: d Wl = d, + s: this is much the same for multiple helts.
Open Belt Drives (Fig. 4) A rc of contact: 2 arccos [(d, - d,)/2el:
\
I
\
\.
\
~ :
\
/
'
/~
.--L. /
,
-"'-"''--...
a
6.2.3 Geometrical Relations
13, =
\
\
_~\
\
,
of miniand the maximunl
tensile stress is calculated using an average modulus of tension (EA'): £ = P'/(EA').
\
\
I
resp(~ct
mum permissible pulley diameter d min permissible bending frequency;;, assigned to it. During bending the neutral fibre is assumed to be at the centre of the thickness of the belt at s/2: the elongation £ under
'41.x,
f3, = 360
0
-
13;:
Belt length (length of neutral fibre when stretched):
Approximate formula for shaft centre distance e at a given belt length
Figure 7. Belt geometry in crossed-belt drive: a crossed at an obtuse angle, b crossed at right angles
6.2.4 Kinematics, Power, Efficiency
Belt SPeeds (6)
As a result of the greater elongation the speed v, of the driving side of the belt must be somewhat greater than the speed Il, of the slack side of the belt in order to maintain steady operation. In practice, the eqUilibrium between the elongations of the driving and slack sides of the belt is ohtained by slip due to elongation in the effective arcs of the driving and driven pulley. Slip due to elongation is produced when JjJ = E, - E2 = (a, a,)/E = ao/E = (v, - V,)/l',. The speed ratio i is there· fore slightly load-dependent in normal operation: (7)
where
In idle mnning the expression i = d. h / d"n applies.
Bending frequency (number of alternate bends per s), and q = (dwl + d wl ) ' /8. The increase ile of the shaft centre distances to prestretch the belt hy £0 = illJ L is obtained from a calculation for Lw and (1 + £,,)L w or ile = (£"L w /2)/sin(f3,/2).
Torques follow from the forces on the side of belt:
Crossed Belt Drives (Fig. 3b with designations as in Fig. 4) Arc of contact:
PouJer:
13, = 13, = 13k, =
360° -
13K,
= (z,1Tdwl n,)/L w
,
f3R = 2 arccos [Cdw ' + d w2 )/(2e)].
The expression e 2: 20b is recommended because of twisting stresses a,. Owing to reverse bending service life shorter than with open belt drive.
Quarter·turn Belt Drive (Fig. 7). Angle of crossing 0°. Length of the centreline of the quarter-tum belt where 0 = 90°:
o*
= 2e + d w ,(1T + y)/2 + d w /1T + rp)/2,
where tan (y/2) = d w ,(2e) and tan (
(8)
z, = number of pulleys. M,
= Fudw,/2:
M, = Pudw,!2.
(9)
where
Length of the crossed belt (centreline):
Lou
fo = z,v/Lw where
Rated power c"Fan where cll is the service factor as in
Table 1 for initial design without calculations for vibration (based on DIN 2218).
EffiCiency l] = P"h/P"n = M,h/(M."i) = (1 -
'41,11
Met'hanical Machine Components. 6 Belt and Chain Drives
Table 1. Service factor (".. for obtaining an approximatioo of the dynamic characteristics of drive and machine and the daily opcf
Service factor e H Mode of operation of the driven machine
Mode of oper.ttion of the driving machine
Near-uniform
lJniform
Uniform Moderate shock
t
wh{'re q ",-, 1.1 and r = I. 0
cO': ()
=- () _'i for r -~ 0 for r = O. L for r = 0.2 for
I
O.24q
1 + O.38q 1 + O.';2q
+
r
+r +r
Heavy shock
+ O.44q 1 + O.62q 1 t- O.7Bq
1 + O.64q + r 1 + O.86q + 1.2r
+r +r + I.2r
1
+
1.06q
+
15r
for positive chain drives,
for synchronous belts
q
1+
r 1 t o 24q + r
Heavy shock
q
o.04q + r
, O.14q+
Moderate shock
d
flat belts and V-belts, an up 10 10 h opcf"dtion per day,
between 10 and 16 h operation per day, over 16 h operation per day
The low q values of e H for flat belts and V·belts are based on the assumption that infrequent short-tenn overloads are partially compensated for by slip action. In the case of positive belt and chain drives it musl be made certain that the iAted power covers maximum peak loads, including moments of inertia and shock! Examples of mode of operation of the driving machine Mode of operation
Driving mal..:hin{·
1Jniform
Elel...·tril...· motors wilh low starting torque (up engines with eight cylinders or more.
Moderate shol:k
Electric motors with medium starting torque (J.5 to 2.5 X nominal torque), internal combustion engines with between four and six l:ylindcrs.
Heavy shock
Electric motors with high starting and braking torque (over 2.5 X nominal lorque), hydraulic engines, internal combustion engines with lip to four cylinders.
10
15 x nominal torque), water and steam turbines, internal combustion
Examples of mode of operation of the driven malhine Mode of opemtion
Driven machine
lJniform
Small masses to be accelerated
Typewriters, light-duty conveyor belts, domestic appliances. Near-uniform
Moderate masses to be accelerated. light-duty ventilation fans. light to medium-duty wookworking machines, belt conveyors for ore, coal and sand, agitators (liquid, semi-liquid), turning, drilling and grinding machines, textile machinery. printing machinery, centrifugal pumps. washing machines
Moderate shock
Moderate masses to be acceler.. ted. Heavy-duty conveyor systems, screw conveyors, mixing machines, industrial ventilation fans, generators and exciters, centrifuges, mbber pro(:essing machines. hammer mills
Heavy shock
Large masses to be accelerated. Reciprocating pumps and t:ompressors with velocity fluctuation < 1 : 80; ball rollers and gravel mills, edge mills, shearing machines, stamping machines, rolling mills for non-ferrous metals, stone cmshers.
virtually bent edgewise (Fig. 8b). Equilibrium is achieved if the bending moment at A caused by this bending strain is compensated by oblique tension acting on the side of the belt (Fig. 8c). Axial misalignment is approximately 0.6 of the belt width, the exact misalignment being obtained after a short running-in period.
Flat Belt Drives with Constant Speed Ratios. The pulleys of conventional open and crossed flat belt drives are made with slightly arched running surfaces to DIN III (ISO/R 100) (Table 2) in order to guide axially the belt that always tends towards the largest diameter of the pulley, In the case of open belt drives with horizontal shafts, the smaller pulley can be made cylindrical where the
speed ratio i > 3. The preconditions for satisfactory running of the belt are: axial parallelism of both shafts, concentrically running pnlleys, alignment of the largest diameters of convex pulleys in one plane, edges of the belt within the breadth of the pulley b, > b, smooth pulley running surfaces to DIN 11 I. "Non-slip", porous or rippled surfaces or adhesive anti-stripping agents prevent natural slip due to elongation in the effective arc, increase wear and may excite longitudinal vibrations by means of stickslip effects, Tbree-dimensional Belt Drives (Fig. ~d, e) accept cylindrical pulleys. To provide a secure belt guide in quarterturn belt drives (0 = 90°) the following is recommended:
6 .2 Flat Belt
• 6.2. ~ Coning Action of Flat Relts, Tensioning
[)riv~s
the pulleys. When the distance between centres is adjust· able (e.g. drive motor to slide rails) tensioning can also be produced after winding on by increasing the distance between centres. If the distance between centres is fixed, the belt length remains constant in all operating conditions. The belt surface forces F' and the shaft elastic forces I'~ are therefore reduced by centrifugal force . The degree of stretching must therefore be specified as larger in tenns of (If in order to ensure the required frictional engagement for the operating speed. The shaft load rises slightly with increased torque and is determined by the exact distribution of the elongation (2). As the stretch is to be maintained for long periods of operation , this tensioning process is particularly suitable for belts with high dimensional stability, e.g. multiple belts with polyamide or polyester fabric plies; it is the tensioning process that is mainly used for this purpose.
Figure 8. a Belt running axiaUy on to the larger diameter. b Equilibrium when belt runs tangentially on to cone pulley. c Drive with two cone pulleys for continuously variable transmission.
Table 2. Recommended curve heights b as in DIN 1 I ) h in mm
for bs
up to
I~ h
- b
s
2')0 mm
0 ..\ 0.4 0 .'
0.3 04 0 .)
0.6
0.6
up to .\55 up to 500 up to 710 up to 1000 up to 14!X)
0.8 1.0
0.8 1.0
1.2
1.2
1.2
1.5 2.0
up
[0
up to 1000
1.5 1.8
(b) Tensioning Shaft. The shaft load Fw is applie d to the traversing shaft by means of weights or (flexihle) springs, suitable for belts with time-dependent elongation under load. Note, however, that in addition to the greater expense involved there is also the danger of vihration.
b in rum for b s > 2'50 mm
112 140 180 224
up to lip to
(c) Tensioning Idler at the Slack Side of the Belt. The movable spring-loaded or weighted tensioning idler produces constant belt surface force 1-'1 in all operating conditions. If the tensioning idler is used on the outer side of the belt, th~ arc of contact {3 is increased and at the same time the arc factor c/3 is improved. The additional tensioning idler does, however, increase bending frequency, thereby lowering the permissible useful stress at higher belt speeds. Its diameter should be greater than d I . min to take the service life of the belt into consideration and its nmning surface should always be cylindrical. As this tensioning process results in rather low loads on the side of the belt and the shaft with small torques, it is suitahle t,)r drives that operate mainly underdriving and belts with time-dependent elongation. lIere too , the risk of vibrations must be noted. If a fixed (adjustable) tenSioning idler is used on the slack side of the belt to adjust the stretch and also to increase (3, the same operational characteristics arise as in the tensioning process shown in
2.)
breadth of pulley b, = 2b, centre distance of the central plane of the pulley from its respective counter-rotating wheel e" e1 = (0.2 to O.~)b (Fig. 7b), djd, c, I to 2.5, 20b.
e""
Production of Tensioning. The minimum required shaft load Fw to ensure frictional engagement can be produced using the processes shown in Fig. 9a-d by: (a)
I."tet
Fig.9a.
Belt Stretch with Fixed Distance Between Cen-
Se!flensioning
(d)
witb
Double
Tensioning
Idler
tres. The belt length is dimensioned so that the belt is
[3). The tensioning idlers in the driving and slack side of
tensioned by elastic elongation when it is wound on to
the belt have a fixed (adjustable) distance between
~ ~
~ 120 7
Fw.o
100
Fw
~
Fw"'Fw.o
E
..,:
I
.J.
b O 20
40
F; 60
'----'-.-'---' fu c
20
40
60
20
40
Figure 9. Dependence of tangential force ... and ... haft load Fw on the peripheral force Fu al conslam speed with different tensioninR methods 13l -: : lHO 0). Index 0 : Forces when stationary.
a to d (for f31 =
':M',a
Mechanical Machine Components _ 6 Belt and Chain Drives
Figure 10. Co nstruction of multiple belts: a single-ply fabric belt, b multiple-ply fabric belt, c rayon cord belt. d band belt with wide tension bands, most commonly used type: D o uter layer, Z tension member, L running layer.
Table 3-. Construction and use of be lts as shown in Fig. 10 (for standard values . see manufacturers' data)
b
Belts
d
Tension member
PA, B
B, PA, E
Running layer"
PU
G or balala
G or CJ{
G or CH
Manufacture
Endless. in lengths
Cut to size from roiL
Endless, in length s
Cut to size from roll,
PA
endless, vulcanised at joint
endless, cemented at joint
High speeds, grinding spindles
Use
Robust, for low power ratings
Multiple pulley drives,
Robust, most common
maximum speed up to
type, up to 6000 kW for twin- and multiple-shaft
1000 kW
drives "m ~ x
20 to 50
70
(ru / s )
150
100
70
20
63
d1.min(mm)
15
Ip"fllaxat d min (I/s)
10 to 20(50)"
10 to 20
30(100)"
30(80)"
F:.
10
30
48
48(110)"
m"
(N/mm)
Maximum e longation
E
2 to 4
in
1,8
operation ( %)
- 20 to +70
-20to +70
Ambient temper.tture rdnge
-40 to +80
-20 to + 80
ee) apA polyamide, E polyester, B cotton, CH chrome leather, PlI pOlyurethane, G elastomer (mbher). hValues in brackets only after consultation with manufacturer.
centres, they are guided, with little friction, on a circuit round a pulley bearing and must therefore each have the same axial loads in each operating condition, This however reqUires differing angles of contact for the tensioning idlers and results in sclf-tensioning, In the case of a partload low belt surface forces F; and I'; are set up, and with them low shaft loads, The belt surface forces and shaft loads rise with increasing peripheral force F", Careful matching of the tensioning idler geometry with distance between centres, pulley diameter and belt elasticity is required in the case of this tensioning process which is also effective for alternating drive direction (brakes), The relatively simple design is suitable for drives up to very high power with predominantly part-load operation and belts without significant time-dependent elongation,
although here too there is a greater tendency for transverse vibrations than in designs with fixed tensioners.
6.2.6 Materials Owing to their limited strength, short service life and considef'dble elongation during operation, leather belts, which at one time were in common use, have been replaced by synthetic multiple belts (composite belts), These belts are either manufactured in endless form to a suitable length or heat-honded at the point of application at their ends, which have been cut diagonally and bevelled to make them endless, Figure 10 and Table ~ show the construction and materials of common types of belt and Table 4 the characteristic values of the materials used in flat belt plies,
Table 4. Material characteristics of flat-belt tension members Material
Rayon cord Polyamide band Leather, high-grade Leather, normal
F...,.."" .. (N/ mm.l)
P (kg/m ')
Ultimate tensile strength
Coefficient of friction on grey
(N/ mm' )
( %)
cast iron and steel
900 500 30 to 50 20 to 30
700(K)0 150000 30n to 500 100 to 300
1400 1140 900 1000
15 20 to 25 30 30
0.3 to 0.7 0.3 to 0 .7
Rm
6.3 V-Belts. 6.3.1 Uses and Characteristics
6.2.7 Calc:ulation The limits on permissible stress for belts art' not set by their tensile strength but by disintegration (failure due to fatigue) and, in the case of inadequate tensioning, by wear. In flat belts the tensile strength Rm therefore amounts to 10 to 20 times the permissible operating tension un- Damage to belts is accelerated by higher temperatures and greater flexing work, i.e. by greater bending frequencies and smaller bending radii. The permissible operating load is determined by experiment. The rough design of an open flat belt drive of the most frequent type (as shown in Fig. IOd) is based on the pennissible (index') nominal peripheral force 1':" relative to 1 mm belt width, with an allocated minimum pennissible pulley diameter d l . min of the smaller pulley, as shown in Appendix F6, Table I. The belt speed Um,x and the bending frequency !R.m= as in Table ~ should not be exceeded. Where d l is the diameter of the smallest pulley, f3, the arc of contact, c~ the arc factor, b the belt width and n"n the drive speed, the following applies to belts as shown in Fig. IOd on the basis of manufacturers' data [4]: Permissible relative peripheral force F:. pcnn = c/3F:' N ' (2 - dlmin/dl):
Refinements can be expected in the calculation in Eq. (11) in respect of V-belts. If a belt drive is specified with fixed distance between centres as in Fig. 9a, the belt must be mounted using elastic seating stretch. If when operating with F~.perm the sum (F; + F;) = kvE~,p~rmb is selected and allowance is made for the centrifugal force during operation as in Fig. 9a, the stretch £, can be calculated as follows:
s,
=
tJ.LI I.
=
En
+
Sf
= I(kj2) F:po,m +
Fil/(EA'),
where Pi = p'u'; (EA') and p' are as in Appendix F6, Table I. Reference values for k, = (m + 1)/(m - I) with m as in Eq. (1), e.g. for f31 = 1T and p = 0.51: K, = (5 + 1)/(5 - I) = 1.5 or p = 0.4: k, = 1.8. Belt length slack, i.e. smaller by the amount of stretch: L
=
l.w/(i
+
e,).
Shaft load produced by tensioning when stationary witb additional load F~ and F,
=
Fwo
F,
=
[(k,!2)F:'penn
+ P~lb =
e,(EA')b )
= FI \2(1 - cos (31) = 21'1 sin (f31/2).
( 10)
Compare the bending frequency J~ witb the permissible bending frequency /H,max for minimum pulley diameter dl. min given in manufacturer's data. Example Open-belt drive fur rotary piston blower, where d, = 315 mm, d z = 800 mm. e = 870 mm, n 1 = 14.,0 min ··1, PI = p . .n = 90 kW. If d w = d, then according to Eq. (6) v = 23.9 m/s, i = d,o/d,," = d,ld, = 254: {3, ~ 147.6": {3, = 214.4°; Lw == 3S59.5mrn. Belt type fordl.mil1
where d 1 ,min == 200 mm and larger belt widths. With this smaller type of belt the pulley diameters d] and d 2 can be reduced and higher shaft loads withstood. Vibration characteristics of the belt drive must also be taken into considerJ.tion when making a final decision using the calculations provided in DIN 740 for resilient shaft couplings and stn'ng vibration. Belt manufacturers should always be consulted in respect of particular cases; see manufacturers' literature [11].
6.3 V·Belts 6.~.1
Uses and Characteristics
V-belts (Fig. Ib) make use of the frictional transmission of motion and power over medium shaft centre distances 151. They are guided finnly in sheaves in all working positions, as well as over short stretches and in angular drives. Almost all types are also suitable for coupling (tensioning the V-belt while the drive wheel is running by means of a radially moving shaft or tensioning idler). Dimensions for the basic types have been standardised internationally; see Appendix F6, Table 2. For further types for special purposes, see Fig. 12. Tbe frictional transmission of peripheral force takes place solely through the tapered sides of the belt crosssection. Direct contact with the bottom of the groove results in reduction of transmittable peripheral force, shear slip and damage due to overheating. The shaft centre distance must be adjustable by x amounts as in ISO 155 or manufacturers' data; as a rough guide x ;e: + 0.03Lw is sufficient for tensioning and retensioning the belt and Ixl ;e: O.015Lw for installing the belt over the rim of the sheave without using force. The effective diameters d w (Fig. Ib) and corresponding effective widths b w (Fig. 12a and Appendix F6, Table 2) of the belt and the sheave denote the position of the non-flexing tensile ply in the V-belt cross-section. lbey should as far as possible agree with the corresponding reference diameter d r and the reference width b, of the sheave (this does not apply to V-ribbed belts to DIN 7867). The wrap angle a is specified as a function of d" owing to the lateral strain in the helt. Frequent (fR) and large (1/ d w ) bending strains increase the internal heating of the belt and reduce its power rating over the same service life (Fig. 11a). Preconditions for a long service life are: continuous maintenance (inspection) of the correct tension, accurate alignment of the sheaves and smooth surfaces of the sheaves, d w . min and shaft centre distance e no smaller than necessary. Avoid reverse hending (back bend idler). Where tbey are unavoidable, idlers should be designed as sheaves, where d w > d w . mi 11'
SP8
20
SPC
SPB !i SPC ~15 ~
c.
Fw " == 9.293]\; Bending frequency /~ 2·24000/3';59.5 = 13.5 s- I < l;\,max: reference valuej~,n"''' = 50 s I for d l .min for belt calculation. The task may also he accomplished by belt type 28
'M ..t'
lOo
a
d,,)d w, min
b
10
v in m/s
40
Figure 11. Power rating of narrow V-belts to DIN 7753 with the same service Iik [), 8]. a Encased V-belt. b Relation to power rating P", of raw-edged narrow V-belt to power rating Pun1 of encased narrow V-belt. dW,tnin as in Appendix F6, Table 2.
1.1.1:.
Mechanical Machine Components • 6 Belt and Chain Drives
Operating Limits. Ambient temperatures = -" 30 to 80
0 (;
(dw '
("-
55
+ d wl );
to 70 0 ( : ) ; im.x = 10; e = (0"7 to 2) Fw = (15 to L5)Fu ; power ratings up to
Po", > 1000 kW (up to 35 parallel belts), 11m" = 0"97 for single belts; l1m= up to 0"95 for V"ribbed belts. 6.~.2
Types and Sizes
The various types are distinguished by the geometrical shapes of the belt cross-section and the sizes by their internal construction" Figure 12a-i shows the most common types of V-belts: (a) Endless V-Belts to DIN 2215 (also classic Vbelt). bol h = 15 to 16; cross-sections designated according to width b o ; for V-belted sheave dimensions and materials see DIN 2211 and DIN 2217. Examples of use: types 5 and 6 for laboratory equipment, precision mechanics; 8 and 10 for domestic appliances; 13 to 22 for machine construction (medium rev Imin) and farm machinery; 25 to 40 for heavy-duty drive mechanisms with large shaft centre distances, large sheave diameters, low speeds and severe operating conditions. (b) Open-Ended V-Belts to DIN 2216" Cut lengths, strong fahric plies, with holes pre-punched for belt connector, for medium peripheral speeds" P mox up to 15% lower, d w .min up to 15% greater than with endless belts to DIN 2215 having the same cross-section. Greater residual elongation so frequent retensioning or shortening required" Belt ends are joined by belt connector after seating, simple installation on awkward drives where endless belts cannot be installed" Can be easily stored" (c) Endless Narrow V-Belts to DIN 7753, bolh = 12 to 1.4 with narrow sheaves to DIN 2211 (dimensions and materials)" These transmit greater power than V-belts of the same effective width as described in DIN 2215. They constitute the most commonly used type of belL (d) Endless Broad V-Belts for industrial variable-speed drives to DIN 7719. bolh = 2.8 to 3"25. Groove angle 0: = 24 to 30 0 Smaller angles produce a greater regulating range as well as a risk of self-locking (the V-belt wedging in the sheave groove). Approximately 20% less power is transmitted than with V-belts of the same height in crosssection as described in DIN 2215" DIN 7719 does not apply to variable-speed drives in motor vehicles or farm machinery" Regulating range im.,limin = 4 to 12 possible with two adjustable sheaves"
@j b
C
(e) Cogged V-Belts. V-belts as shown in a and d, with transverse grooves cut in the inner face of the cross+section to increase flexibility, make it possible to have smaller sheave diameters and take up less space, with power being only slightly reduced. The grooves do however cause periodical running-on shocks and noise, unless the transverse grooves are unequally spaced. (f) Endless Hexagonal V-Belt< for farm machinery (twin V-belts) to DIN 7722. bmoxlh = 13" For planar multipleshaft drives with counter-rotating sheaves. Power transmitted as in V-belts to DIN 2215 with equal maximum cross-sectional width. Used in medium-duty drives (combine harvesters) down to light equipment (gardening equipment, street-sweeping machines)" (g) Raw-edged V-Belts. Cross-sections as in DIN 2215 and DIN 7753 Part 1 They have only one outer covering ply of fabric. but - unlike the other types of belt - no fabric shell on the bearing edges or the "cogged" inner face. The belt substructure, a polychloroprene-rubber mixture, is very flexible and elastic and is reinforced for increased expansion forces (preliminary tension) with supporting planes arranged at right angles to the running direction" They transmit greater power using small sheave sizes and high speeds (Fig. lIb), cope with smaller sheave diameters (approximately 0.7 to OB dw"min as in Appendix F6, Table 2) than encased V-belts (therefore also require less working 'pace for the same power) and are less susceptible to oil, heat, slippage and abrasion"
(h) Multiple Narrow V-Belts (power belts) 'l11ese comprise up to five narrow or classic V-belts of equal length (within the set), which are banded tightly together by shrouding. Shrouding prevents twisting or significant vibration of individual belts within the set Grooved belts to DIN ISO 5290" (i) Ribbed V-belts to DIN 7867. Further development of multiple V·belts towards flat belts. Five cross-sections with
space between ribs in mm: PH L60; PJ L~4: PK 3.56; PL 4" 70; PM 9AO. PK preferred for motor vehicle construction, PJ, PL and PM preferred for industrial belt drives and PH for special applications" Up to 60 polychloroprene ribs wide without casing which completely fill the grooves in the appropriate pulleys. Tension cord made of low-stretch rayon cord" Refer to manufacturers' data for power rating with additional speed ratio per rib. Peripheral speeds up to v = 60 mis, depending on the crosssection" The smaller sheave diameters and higher speed ratios per step than in V-belts reduce working space required, whilst there is quieter running and greater uni" formity of motion; deflection possible" 6.~.~
Calculation
In order to calculate the rated power P N of open V-belt drives as a function of service life, a numerical equation obtained from test results, as given in ISO 5292, is being increasingly used. When characteristic quantities are introduced this equation may be formulated more clearly: PN =
e
d
f
g
h Figure 12. Types of V-belt. For a to i, see text.
c~o
1) . -v . [ I + K, (dW"min I - ---,Vo
dWI
Ki
where the arc factor c~ = L25 . (1 - 5 - ~,!180); arc of contact /31 of the smaller pulley; rated power Po at peripheral speed Vo for minimum pulley diameter d w .Olin at speed ratio i = 1 (/3, = 180") and belt length Lo; rated power
6.5 Chain Drives. 6.5 . 1 Types, Characteristics and Uses
P N at peripheral speed v for effective diameter of tht: 0 ({3, '" 180' ) and smaller pulley d w ' at speed ratio i belt length Lw, K, = 1.124 .- 0 . 124 exp [- 3(; - I») and i 20 O. An evaluation of details provided in a manufac-
*
turer's catalogue is given in Appendix F6, Table 2. as an initial gUide. As only deviations from the manufactuft~rs ' information arise in a particular case, the information provided by the V-belt manufacturers must be used when checking calculations. Standards DIN 2218 and DIN 7753 may also be used as a guide. The correct dimensioning of a belt drive is dependent on a series of factors and environmental conditions. It is therefore recommended, especially where there are difficult problems with drives, that the experience of the manufacturers in this sector, i.e. V-belt and drive manufacturers. be called upon; see manufacturers' literature [1 I) The rated power cBPall :S zP':'.' for z belts running in parallel is determined according to Table 1 USing estimated values for c. so that tht: rt:quired number of belts is Z 20 cHP. o / PN . Calculation of all other operating dimensions as per flat belts.
6.4 Synchronous Belts 6.4.1 Design, Chararteristiu and Uses Synchronous belts (Fig. I~) have teeth on one or hoth sides, with which they transmit peripheral force positively without slip (Fig. 2c). The body of the helt consists of neoprene or polyurethane with the pulling strands made of high-tensile glass fibres or steel. kevlar or rayon cord which are wound helically in the case of bdts which are manufactured in standard lengths. normally in endless form. The pulling str.lI1d dett:rmines the neutral bending plane and its length is also the effective length Lw of the belt, it runs on the effective diameters d W12 = ZI.lPh/'Tr round the synchronous pulleys (crowned rims) where the tooth numbers are z l ' Z l and the tooth pitch is Ph" Synchronous belts (toothed belts) are maintenance-free if correctly set , no lubrication re quired . At greater speeds, power ratings, initial tensions and belt widths meshing noises occur, basic frequency 10, = n,z, Owing to their positive transmission of motion synchronous belts are suit-
able for drives which maintain the speed ratio (e.g. tinting gears) , and in the cast of toothing on both sides. for multiple-shaft drives with counter-rotating pulleys. and in the case of greater axial distances, for spatial drives (Fig. ~e). Standards. DIN 7721
and D[N/ISO 5296 regarding dimensioning and measurelnent of effective length.
6.4.2 Design Hints In planar drives the synchronous belts must be guided axially by lateral rims on both sides of at least one crown pulley or alternate sides of two crown pulleys. To permit installation and initial tensioning there should be a shaft
,
....,
•••
... ....
.~. b
a
C
Figure 1~. Cross-sectional shapes to toothed belts: a , b single and double toothed to DIN 7721 with metric increments and DIN / ISO 5296 with inch increments, c:: HTI) (high torque drive)
cross-sec..-tion .
'AI_g
or belt tightener which can be moved radially . Where the shaft centre distance is fixed the crown pulleys are mounted together with the installed belt. As far as possible tensioning idlers should be designed as crown pulleys (dw > d w ' ) and arranged internally so as to avoid deflection at the slack side of the belt, but not spring-mounted as there should be no stretching of the belt if it is laid out correctly. Recommended limiting values: e = (0.5 to 2) (d w ' + d w , ) , d,/b 20 I. In the case of spatial belt drives the straight line between the winding-on and running-off point must at the same time be the line of intersection of both wheel planes so that the belt is only twisted and not pulled off to the side (see Fig. ~e); lateral rims may be dispensed with; shaft centre distance per 90° of twist 12b.
e9() ?
Operating Limits. Ambient temperature = - 40 to
90 °C; P",~ = 400 kW; v",,, = 40 (Type T20) to 80 (1'5) m/s. fH.",~ = 100 s ' ; i m. , = 12; 1)m~ = 0.98. 6.4.~
Calculations
Calculation of Lw (approximate). e and v as for flat-helt drive: exact: Lw = Pt7b' where Zb = number of helt teeth; number of engaging teeth Z< , = z,{3,/360° (rounded off to whole number) ; speed ratio i = zjz.; selection of belt according to rated power and number of teeth Zl > z l .min with power data for reference width b,,, as in Appendix F6, Table ~ and width factor kw = (bJb ,,,) , H as in [SO ';29'; and initial load factor k, = I for z., 20 6 or k, = I - 02(6 - z .. ) for z", < 6. With the power rating CHP,n :=;
k,P"
~ IJ-. {15 (~.')"" Vo
bsl\
bo;()
_ 0.';
(;~,,)2} v ,
and v = n IZ1Pb = n lZ1.Pb, the minimum required belt width b s is obtained. Maximum belt widths b s . lllu :::::: (4 to IO)Pb ' Recommended shaft preload Fwn = f<;, . The service factor en must be increased in accordance with manufacturers' data rather than Table 1 for step-up transmissions for I /i = 1.24. Higher power ratings can be transmitted using HTD (high torque drive) belts [6] and RPP belts (belts with a paraboliC cross-section) [91 as a further development of trapezoidal toothed belts and AT belts I]() I as strengthened T types . An additional deciding factor for the selection of belts. particularly in automohile construction, is that the development of noise should be kept to a minimum . This is what is aimed for by modifying the shape of trapezoidal teeth . See manufacturers' literature III)
6.5 Chain Drives 6.S.1 Types, Characteristics and Uses Chain drives (Fig. 2a, b) transmit power loads up to 200 kW per single chain positively and without slip using low peripheral speeds between parallel shafts and where there are more than two shafts also by counter-rotation. Power loads of more than 500 kW are possible using multiple-strand chains (made with up to 12 strands but mostly with up to 3 strands) . Where the smaller sprocket has a small number of teeth. the rotary transmission becomes uneven owing to the rhythmically variable run-off or nip point of the chain, or the pOlygon effect as it is known. Resulting from this are periodically variable rim speeds, excitation of vibration and noise at higher chain speeds. This is moderated by increaSing the number of teeth and
I Mil.,
Mechanical Machine Components • 6 Belt and Chain Drives
~
lldllJI 1
1
a
1
1
1d]JJJo b
Figure 14. Transmission chains: a single bushing chain, b single roller chain; I roller link plate with bushings held in place, 2 pinlink plate with pins, 3 movable roller.
reducing the spacing. On the other hand, by reason of its longitudinal elasticity the chain moderates starting shocks. The service life of a chain is limited by the maximum wear-induced elongation it can withstand and is reduced by inadequate lubrication, dirt and stress due to shock and vibration. The most common types are the bushing chain (Fig. 14a) to DIN 8154 (inside a closed transmission casing with very good lubrication), the roller chain (Fig. 14b) to DIN 8187 and DIN 8188 (the most frequently used type; the lubricated roller reduces wear and noise) and the inverted tooth chain (Fig. 2b) (to DIN 8190) (silent running at high peripheral speeds). For other steel link chains see DIN 8194 which covers structural shapes and nomenclature (German, English, French).
6.5.2 Design Hints Where possible, measure shaft centre distances to an exact number of chain links (pitch P) in order to avoid bent links. The centre distance should be such that the arc of contact is at least 120° to pinion, nonnal: e = 30 to SOp. The sag on the slack side of the chain should be approximately 1% of the centre distance. The maximum allowable wear-induced elongation of the chain III should generally not exceed 3% of the original chain length /, in sprockets with more than 67 teeth only 111/1 ~ 200/z.z as
a percentage. However. with fIXed shaft centre distance without a tensioning device, only IlIII:5 (0.6 to 1/5)% compensation for wear is allowable on the chain using transversely movable shafts. For a ftxed shaft centre distance, a cylindrical tensioning idler (up to v = I m/s) or a tensioning wheel is used, both on the slack side, lightly loaded by springs or weight. Owing to the polygon effect only sprockets with at least 17 teeth should be selected. For medium to high speeds or maximum permissible load, the pinion should have hardened teeth and if possible 21 teeth. Sprockets should nonnally have a maximum of 150 teeth. The preferred number of teeth is 17, 19, 21, 23, 25, 38, 57, 76, 95 or 114. If the chain drive is positioned at an inclination towards the horizontal greater than 60°,
then the required chain tension is provided by tensioning idlers, tensioning wheels or other suitable devices. In tensioning and deflection wheels at least three teeth should be engaged at anyone time. A speed ratio i of 3 to 7 times is considered good, but ratios above 10 times are possible. The amount of lubrication required depends on the type of chain and the chain speed; Jor notes on roller chains see DIN 8195: for a DIN 8188'{)8A-l chain with a pitchp of 12.7 mm, for example, the following applies: oil applied by oil can or brush, up to v = 0.7 mls (uncertain, at least once daily); drip-feed lubrication, up to v = 3.9 mls (drip-fed oil for each series of link plates with 2 to 6 drops each per minute); oil bath (maximum oil level up to lowest roller centre) or centrifugal lubrication up to v = 8.4 m/s; pressure circulating lubrication, if necessary with filter and oil cooler, up to vm" = 19 mls (with constant oil flow on the inner face of the slack side as well as the tight side of the chain; also for cooling the chain). After a single lubrication, efftciency drops off quickly with increasing operating time; '1m= < 0.97. The maximum power transmission at vopt = nOzlP. For reference values for no where ZI = 19 and 15000 operating hours with speed ratio i = 3 with 100 chain links as in DIN 8195 see Appendix F6, Table 4.
6.5.3 Calculations Chain speed v = nlzlP = n,zzP, geometrical diameter (roller centres) d w' .z = plsin (180° Iz,.z), chain length 1= Xp, where X is the number of links (whole, exact number). X 2" Xo where Xo = 2elP + (z, + zz)/2 + P(z, - z,)z/(4e1T Z ), centre distance
zz) + J(x-~' ;zz)' -2(zZ :z'n
_ p [( X -z,-+e-4 2
The pitch p of the roller chains as in DIN 8187 (European type, Code B) and DIN 8188 (American type, Code A) is standardised in inch increments; see Appendix F6, Table 4. The power rating Po corresponds to the speed no; in accordance with DIN 8195 the following applies to n 1 :S no, i ::s 7:
P = Po (~r9 N0 97 (~r73 WO'8 (4~p)"'6, N
with rated power csE'," :5 P N , by which means the service factor can be estimated from Table 1 or even DIN 8195; N = 1 for single-strand chain, N = 2 for double-strand chain, N = 3 for triple-strand chain. See manufacturers' literature [11].
Friction Drives H. Peeken, Aachen
7.1 Mode of Operation, Definitions Friction drives, also known as ratio traction drives, are unifonn-motion, frictionally engaged transmissions [1] in
which, in contrast to belt drives, contact does not take place over a large area but more or less in the fonn of points or lines. The area of the contact surfaces caused by flattening and the distribution of pressure can be determined by applying Hertz'S law (see B4). The theory of Stribeck's contact pressure applies in the case of nonrigid, non-metallic materials. The transmission of moments
7.2 Types, Exanlples • 7.2.2 Continuously Variable Traction Drives
'Mill
2
2
~ F,
I
a
F,< pJ.
d,
Figure 1. Forces and transmission in friction wheels: a with parallel axes, b with intersecting axes, withom sliding friction, c \vith intersecting axes, with slip in the conlan line
is brought about by means of peripheral forces F" which act betwt:en tht: rotationally symmetrical wheels under contact pressure Fn (Fig. la). An effective coefficient of fricti(J/1 /L, = F,/ f:, (Table 2) is ddined which is always smaller than the actual coefficient of friction /L. The tangential energy storage efficiency factor is therefore v, = /L,/ /L. The axes of rotation are usually situated in one plane in order to avoid the skewing that occurs with inclined axes, [n the case of variable-speed drives, however, a certain sliding motion (see F7.'.!) must be taken into account. A purely rolling motion is only possible if the tips of both pitch cones coincide (Fig. Ib). The transmission is defined by the speed ratio of driving shaft (index I) and driven shaft (index 2): i = n,/n, = d,/d,.
the literature the reciprocal i = n Z/n 1, signed if necessary. is also to be found for the speed ratio. especially that of variable-speed drives. 111e speed of the driving motor, n L which in practice is often constant, serves as the reference quantity with the result that (l1 z = 0) does not become j = ex when the driven shaft is idle. [0
7.2 Types, Examples In their simplest fonn, friction drives consist of two rotating bodies directly attached to the driving and driven shafts. In order to reduce the high conta<..~t pressure, which in this case must be fully absorbed by the bearings, pairings with higher coefficients o(friction (Fig. 2) are preferred. Special characteristics can be obtained by means of designs using connecting links, which, although having the disadvantage of the serial mounting of two contact points in the power flow. does allow the parallel mounting of a number of connecting links. In this way
a
b
c
Figure 2. Frictiun wheels with friction coatings, where B > b: a hard organiC friction coaling, b vulcanised rubber friction ring, c tensioned rubber friction ring.
Figure 3. Device for producing a torque-dependent axial force = F, tan a = (Al/r) tan a
F~
the power can be increased and the bearing load reduced (e .g. planetary arrangement to reduce radial forces). In variable drives, the driving and driven shaft can then be fixed spatially, and the sliding motion can be minimised over the whole range of adjustment. Contact pressure Fll is either produced by elastic force, in which case it is generally constant and slippage is made possible in the event of overload, or it grows as the load incrt~ases. Conditional on the principle used, the force is load-dependent (Fig. 5b, d) or else it is strongly influenced by torque-dependent clamping devices as shown in Fig. 3. In this way, the speed ratio changes only slightly as the load varies, the drive being" torsionally stiff'''.
7.2.1 Friction Drives with Fixed Ratio In aU applications that do not require synchronous nmning, friction drives with tlxed ratio tlnd themselves in direct competition \vith positive drive types such as toothed gear drives. They are distinguished by their simple construction, which allows econontic designs and may at the same time perform the task of an overload clutch. They also per/(lrm a dual function in the bearing arrangelllcnt and driving of large tubular vessels, As the geometry of the contact area does not alter with time, there is no fear of periodical excitenlent of vibrations (contact shock, fluctuations due to teeth sticking), in contrast to toothed gear drives. It is therefore possible to produce drives with very low noise-levels (Fig. 4) and very higb speeds (e.g. lip to 16000 s -, in texturing machines) are also attainable at the speed increasing ratio.
7.2.2 Continuously Variable Traction Drives The lack of positive locking in traction drives makes possible a continuous variation of their speed ratio within the linnts i min and i m ;.!". This propelty is characterised by the adjustlnent ratio if = imax/imill" By cOIubining this with a planetary drive to make an adjustable coupling drive (see F8.9) the adjustment ratio can be enlarged or narrowed down as desired, so that for example it is possible to reverse the direction of rotation with every type.
Figure 4. Planetary friction drive as in II I J: I drive shaft for split sun wheeL.! fixed outer ring, j spherical planet wheels, 4 device for wrsionally adapting tbe two axially movable and rotatable halves of tht· planet wheel to shaft I (d. Fig. S) . s; planet carrier as main drive pinion
'M'''.1
Mechanical Machine Components. 7 Friction Drives
Variable-speed drives are often supplied as complete drive units with mounted asynchronous motors by which the range can be extended via pole-changing control. In most cases, drive-side step-down gear systems can be mounted which are used to obtain any desired speed range. Figure S shows a selection of common operating principles. (Drives in Fig. Sa are dry-running with plastic friction ring, all others with lubricated steel rolling elements.) The great variety is due to the various demand, that are made of friction drives, such as economic efficiency (price, operating efficiency, service life), idle adjustment, adjustment up to n 2 = 0, etc. The selection of a suitable variable-speed drive for a specific' application is made on the assumption that the drive must meet the torque requirement of the machine over the entir(" speed range. The course of the driven moment, designated as the driven centreline, over the speed nl. is therefore an iIIlportant characteristic of the variable-speed drive. At constant drive speed n" the characteristics of the types shown in Fig. S through various ranges (Table 1) of the schematic driven centreline as shown in Fig. 6 can be demonstrated. The steady torque which in many types is constant over a certain
range of adjustment II often may not be transmitted at extreme speed ratios (ranges II and III), as for example the permissible pressures according to Hertz's law are exceeded by smaller radii of curvature or the drilling motion results in increased wear. Moreover, the often hyperbolic torque decrease over range III is caused by the limited driving power.
7.3 Principles of Calculation 7.3.1 Slicfutg Motion In order to calculate the relative motion in the contact area the friction wheels involved are replaced by cones tangential to the contact surface, which is assumed to be flat. In general, the tips of these pitch cones do not meet at a point in the contact plane, as shown in Fig. 7. The peripheral speeds are only identical at point P, the difference between them increasing along the surface line. This motion, superimposed on the purely rolling motion, can be described by a relative rotation at angular velocity "'" normal to the contact plane. Generally speaking, the rela-
I~ -': I
.
1
a
e
d
f
g
Figure S. Schematic drAwing of some traction drives (cr. Table I): 1 drive mechanism, 2 main drive pinion, 3 connecting member, 4 device for torsional1y adapting the rollers.
IM'd
7.3 Principles of Calculation. 7.3.2 Slip Rate
Table 1. Ratings of traction drives (Fig. S) in accordance with manufacturers' catalogues (as at 1989). Values for largest and smallest of each type with flange-mounted drive motor, n, = 24 l/s
Fig. Description (manufacturer) no.
Sa Cone/friction-ring drive (SEW, StOber, Fiender-Himmelwerke)
0.15b 0.05
50 Cone-disc drive (Unicurn)
103 0.15
5d "H-drive" ring-sheave drive (Heynau) 5e
"Beier drive" cone pulley-ring drive (Sumitomo)
Sf
Ball-ring drive (Planctroll, Neuweg)
5j
a
nt
10.4 0.14
"Disco" planetary sheave/ring drive (Lenze)
.=
12.5
0.61 0.55
11,111
0.86/0.43 = 2 2.40/0.2 = 12
0.92 0.92
I1,III
3/0.33 = 9 3/0.33 = 9
0.79 0.79
I1,III
1.3/0.33 = 4 0.8/0.2 =4
0.8 0.8
III
ISO 1.2
0/0.39 = 0/0.39 =
0.77 0.7
I,ll, III
13.4 2.0
1.2/0 = 1.2/0 =
0.79 0.72
III
1.2/0.2 = 6 1.2/0.12 = 10
0.9 0.9
I,III
2.21/0.29 = 7.75 2.14/0.21 = 10
0.95 0.8
III
0.67/0.13 = 5 0.67/0.11 = 6
0.86 0.85
III
1407 3.8
3440 3.2
68"
Toroidal drive (Arter)
2/0.22 = 6 3/0.33 = 9
0.6 0.36
120' 0.2
0.8'
;i
11,111
43 3.0
2.36b 0.086<
5h Double cone-ring drive (Kopp)
0.9 0.7
75 2.4
3.2 0.2
5.76 0.02
5g Ball-disc drive (PIV, Riemers)
\.25/0.25 = 5 1.1/0.22 = 5
Nm
0.08
5b Hollow cone-ball drive (Heynau)
1Jmax
M2m~x
kW
10
18.6 0.12
P,
(n 2 /n')max cp=--(n 2 /n')mln
P2ma,.
1200 IB 120 2
300
00 00
00 00
=
~
Characteristic ranges
1/s.
n , = 47 I/s. cwith transmission. b
II
I
7 .~.2 Slip Rate
1II
---'----------_._-
n,/n,
Figure 6. Schematic driven curve for traction drives as shown in Fig. S. The existing ranges for the individual types are given in Table I.
tive motion between rolling elements 2 and 1 is obtained using the vector equation W rel = W, -- w,. By sectioning vertically parallel to the contact surface the sliding and rolling speeds being sought can be determined:
with the values
A plus sign is used when P is between 5, and 5" Ww = j~ cos a 2 :::t: WI cos alj
and a minus sign when a pitch cone is a hollow cone. The sliding/rolling ratio w,,/ww denotes the extent of the sliding motion and its associated losses. It is determined by the type and varies over the range of adjustment (e.g. 0 to 15 (Fig. Sa) and 0 to 0.5 (Fig. Si».
The size and shape, i.e. the semiaxes a and b of Hertz's contact ellipse, are detennined, among other things, by the main radii of curvature of the rolling elements at the contact point. In the planes fixed by the axes of rotation these are the radii P, and p,. The plane perpendicular to them and to the contact surface produces conic sections with radii of curvature p; and p;, at the contact point. Where sliding motion is present, the peripheral speeds of the rolling elements are identical only at one point, the centre of rotation P. Consequently, its position determines the relative speed ratio. During idle running, P is at the centre M of the contact ellipse (Fig. 7a), by means of which the speed ratio wo,/wm = rodro, is fixed. Frictional forces which, although producing a moment about P, fail to produce any resulting peripheral force for reasons of symmetry, are set up in the direction of the slipping speeds. When moments are being transmitted and the pOSition of the contact surface is invariable, the centre of rotation must as a consequence lie outside the centre M [2). The integral action of the frictional forces JJP dA towards the circumference then produces the desired tangential force F" Further, a sliding moment Mb is set up about P. These sectional reactions can be combined into a resultant force F, whose line of application passes through the hypothetical point of application of force K. This is expressed as Mb = F,IN' In order to minimise the sliding moment the contact surface should be as small as possible. Point contact is therefore preferred where sliding motion is present. The peripheral speeds of both rolling elements again coincide at P to produce the speed ratio under load
1.111
Mechanical Machine Components. 7 Friction Drives
Driving side
I
''''--j',. I
P.
~~.-.--. . '", 'i
I
.
c "'1
b
d Figure 7. RoUer contact with sliding motion: a idle nmning, b under load, c magnified contact ellipse with frictional forces in the direction of the running speed, displacement of the centre of rotation P by I on the occurrence of a peripheral force FlO d Upturned sectional representation of a with main radii of curvature p; and p;,.
The relative change in speed ratio compared with idle running is described as rolling slip Sw
s =
W02 /WOI -
W,/Wl
= 1 _ r l /r2
rOd r 02
WOl/WOl
w
= 1
erOI - I sin CY(1)/er02 + I sin CY.,) r Ol /r02 erOI - I sin CY.,)/rOl (ro,
+ I sin
CY( 2 )/rO ,'
In the event of constant contact pressure Pn and invariable coefficient of friction IL the slippage becomes greater with increasing load, i.e. increasing excursion from the centre I. Large wheel diameters and small cone angles CY. have a beneficial effect on slip characteristics. Calculation methods for determining the peripheral forces to be transmitted and the length I which determines the kinematics usually assume a geometry unaffected by tangential forces and a pressure distribution in the Hertz contact surface. For the simplest case of a constant coefficient of friction, phase diagrams [2, 3 J are available which show graphically the mutual dependency of the influencing variables I, IN' a, b and V,. In the case of traction drives without sliding motion (e.g. Fig. la, b) the basic theory cannot be applied. A displacement of the centre of rotation continuously increasing with increasing load cannot be signalled,
because it has not been defined in idle ruuning on account of the peripheral speeds that are constant over the whole contact surface. These cases can only be worked out if a coefficient of friction is assumed which varies with running speed. Accordingly, the ruuning speed not only determines the direction of frictional forces but also their variable quantity. More recent theories [4 J take this influence into con-
sideration, especially for the most common case of lubricated Hertz contact surfaces. The simultaneous calculation of elastic deformations and hydrodynamic processes characterises these EHD (elastohydrodynamic) contacts. The course of the pressure in the contact area resembles the Hertz pressure distribution with maximum values of
some 1000 N/mm2. This radically alters the properties of the lubricant in the gap. In particular, special friction wheel oils known as traction fluids [5 J solidify, thereby enabling the surfaces to separate (gap width < I fLm [6]) when the maximum permissible shear stress in the 'Order of T = 100 N/mm2 is attained at the same time. Figure 8 shows measured friction curves for a conventional mineral oil with a suitable high naphthene content and a synthetic traction fluid at various sliding/rolling ratios. Independently of the rolling slippage under examination here, an alteration occurs to the speed ratio caused by an alteration of the friction radii resulting from loaddependent elastic deformations. It is possible to conceive of designs in which rolling slip may even be fully compensated for in this way. Exceptionally, the slip values Sw of finished variablespeed drives are between L 5 and 5% above this when under nominal load.
7.3.3 Power Rating and Efficiency Power ratings of the types of drive shown in Fig. S in accordance with manufacturers' catalogues for the largest and smallest of each type are shown in Table I. The power shown is the mechanical power P2 available at the driven shaft and the total efficiency derived from it is calculated on the basis of absorbed electric power Pel' Besides the Hertzian stress as delimited by material strength and frictional wear, temperatures which rise as a result of inefficient heat elimination with larger size determine the power limit of traction drives.
I.·.
t
7.4 Hints on Use and Operation
Plate~22 lX'..
mately equal over the contact surface and has at the contact point in idle running the value
Test rolier
Plate 1 The
Cone angle lX, =lXZ
30'
Sliding/rolling ratio rob/row
- ---
0,577
Speed 2v,=16.8m/s Compressive stress =7Z6N/mmz 0,1 0
"
0.Q9
.
V-
I V/
-0"
rr
0,08
11/ II/II 11j f I
0.0 7 0,06
0,05 ~
003
~fr
0,02
i--A-
I
If ~
0,04
1'-.._
-----
F-
"
..t.
DC.l ;:"'1
I~r;
.. =0' • = 10'
" = 20' 0=30'
..
~.
-0.01
0,01
0,02
power
is
therefore
P, = tJ.v/J-NFn
The associated coefficients of friction and slip values /J-N and Sw are obtained, for example, from available friction curves or roughly calculated using the data provided in Table 2. If sliding motion is present, the frictional power can be estimated as in [8] in the following manner. First, the slip rate associated with the existing power ratio /J-N = FI/ Fn from the friction curve for sliding motion and placed in the above equation, The effective coefficient of friction for this slip rate is, however, selected from the slip curve without sliding motion, Only part of this high coefficient of friction is utilised for the transmission of peripheral force when sliding motion is present, the rest being attributed to sliding friction losses, A more accurate method of calculation is to be found, for example, in [4],
7.3.4 Combinations of Materials itt Use
IU
0.0 1
frictional
= w1ro1SwJ.IFn·
I
0.03 Slip rate
0.04
0,05
0,06
Sw
Figure 8. Friction curves as in (7J of a naphtha-based mineral oil and a synthetic friction wheel oil (higher JLN values) at various sliding'rolling ratios.
When they are of equal weight, and thus shaft and bearing capacity are approximately equal, the power rating of traction drives is smaller by an order of magnitude than that of toothed wheel drives (Fig. 9) as these are capable of bearing the full normal force Fn when their contact surfaces are under an equal amount of stress, whereas friction drives can only bear f-Ll'n as a peripheral force. Power losses occur especially in the bearings and in frictional contact itself Only in roller pairings without sliding motion can the frictional power be given directly. The difference between the peripheral speeds is approxi-
Figure 9. Power-weight ratio comparison of traction drives.
Table 2 shows a selection of friction wheel materials in use together with standard values for calculations. In the case of metallic materials the permissible Hertzian stress PH pe= is given, otherwise it is the Stribeck contact pressure kpe= = Fn/(bdl), cf. Fig. 2b or k;e= = Fn/(d,p) where do = dld,/(d l + d,) (Fig. 2a). The effective coefficients of friction /J-N indicated contain a certain, common margin of safety. Details are in accordance with [8], otherwise sources are as indicated. The demands made on friction pairings in respect of high resistance to rolling and wear with a simultaneously high coefficient of friction cannot be met simultaneously to best advantage. Because of the favourable point contact in variable-speed drives, it is almost exclUSively all-steel drives that are to be found there. Friction drives with a fixed speed ratio, on the other hand, usually have line contact and can be cheaply designed with elastomer friction wheels, because any shaft and bearing loads that occur are light. Lubricants and dirt must be kept away from the running surfaces so that a high coefficient of friction can be guaranteed.
7.4 Hints on Use and Operation Friction drives with fixed speed raUos are often used in precision drives to transmit low power ratings. If the wheels are removed they have the effect of a clutch (tape recorders). If provided with a flexible rubber friction coating they are particularly quiet. With hardened, precisionground and lubricated steel friction wheels they are silent, but with fast-running, dry metallic friction pairings they are noisy. Variable-speed friction drives are used to drive machinery and equipment whose driving speed is intended to be continuously variable (agitators, smooth-running belt conveyors) as well as to maintain a constant speed by manual adjustment or automatic control of the speed ratio. The range of adjustment should be specified as small as possible so that full use can be made of it. In this way localised wear. i.e. the formation of tracks during pro-longed running, is avoided by means of a uniform speed ratio. An exception to this is the drive shown in Fig. Sf', as the ball tracks also vary with each rotation when the speed ratio is uniform [9]. With slow-running drives it is usually better to have a small size with an additional over-
'4'M
Mechanical Machine Components • 7 Friction Drives
Table l. Characteristics of some material pairings Pairing
Lubricant
P Hpcrm • k:enn, kpcnn (N/mm')
=
I
Naphtha-based friction wheel oil
= =
0 I
Synthetic friction wheel lubricant
=10
Grey cast iron-steel GG 26-St 70 Grey cast iron-steel GG ~I-St 70 GG 18-St 50
PH..".. = 2500 to 3000 P Hpcnn = 2000 to 2500
P Hpcnn
=10
w"lww
Sw
0.03 to 0.05 0.025 to 0.045 0.015 to 0.03
0.5 to 2 I to 2 4 to 7
0.05 to 0.08 0.04 to 0.07 0.02 to 0.04
o to
0.02 to 0.04
I to 3
0.1 to 0.15
0.5 to 1.5
(%)
Point contact
Hardened steel Hardened steel for sliding/rolling ratio w"lww = 0
Associated slip rate
Effective coefficient of friction #LN
Paraffin-based friction wheel oil Dry
=
300 to 800
= 2500 to 3000 PH""~ = 2000 to 2500 P Hpenn = 300 to 800 PH",,~
1
I to 3 3 to 5
Line contact =
PHp<:rm
450
line contact PHpe~
= 320 to 390
(Crane wheels, DIN 15070)
Rubber friction wheels to
Dry
line contact v < 1 m/s: k~rm = 0.48
0.6 to 0.8
6 to 8
Coating vulcanised onto St [12]
v = I to 30 m/s: k~ = O.48/d17"j
Coating pressed on
< 0.6 m/s: k;.~ = 0.48 0.6 to 0.8 v = 0.6 to 30 m/s: k;'~ = 0.33/v"·75
6 to 8
DIN 8220
Organic friction material
V
Dry
line contact k purn = 0.8 to 1.4
drive transmission at the front and underdrive transmission at the rear than a heavy construction without an auxiliary drive, because the efficiency of friction drives increases in proponion to the speed range [10]. If only a small adjustment ratio is required for preciSion control, a planetary adjustable coupler drive should be used (see F8.9.6) in which the variable-speed drive has to transmit only a pan of the total power and can therefore be specified smail.
0.3 to 0.6
2 to 5
In most manufactured drives, contact pressure increases, either as determined by the type of construction or as a result of torque-dependent pressing devices, with increasing load. Over the range of pattial loads the rolling elements can be released, thereby avoiding considerable wear due to slippage when overloaded. In order to reduce the risk of failure in the event of a heavy overload, many manufacturers supply their drives with auxiliary slip clutches.
Gearing H. Winter, Munich (Section 8.9 by H. W. MUller, Darmstadt)
Advantages. Transmission
of motion (precision equipment) and power (up to 85000 kW in one pairing) without backlash. Relatively compact systems. High efficiency (restrictions for worm gears and spiral gears should be respected).
Disadvantages. Rigid power transfer (flexible coupling provided if applicable), vibrations due to meshing, e.g. chatter marks in machining processes. Counter-measures: finer tooth quality, helical gearing, belt-driven stage, etc. Gear Pairs (Fig. 1). Parallel shafts: spur gears, simplest to manufacture, most reliably controlled, up to highest power levels and speeds (internal gear dearer, limited
manufacturing possibilities, in some circumstances "flying pinion", mainly for epicyclic drive trains). Intersecting shafts (usually less than 90°): bevel gears. Small pinion offset: hypotd gears, extreme-pressure lubricants required because of longitudinal sliding in point contact [1]. Large pinion offset (centre distance): crossed helical gears, for small forces (point contact) other than for smaU intersection angles. Worm gears for high bearing capacity (line contact) with lower transmission levels; self-locking in cenain circumstances when power flow is reversed.
Noise Behanour (see F3). High slidingjractions are advantageous: worm gears (up to 10 dB lower noise level attainable than for spur gear systems), hypoid gears. Noise levels for fine-quality highly-stressed spur gear systems can be decisively reduced only by going over from spur toothing to helical gearing (total contact ratio > 2.5). For low-stressed gears (precision equipment), the influence
'M·tl
8.1 Spur and Helical (;ears - Gear Tooth Geometry. 8.1.2 Transmission Ratio, Gear Ratio, Torque Ratio
External gear pair Internal gear pair Spur gears (cylinder gears) L~~~_J
Bevelled gear pair
I
Axes Intersect
Axes parallel
-b~1
7he contact normal must always pass through the pitch point.
E
I
i-//
Hypoid gear pair
the contact point - pitch point C. Instead of rotating around 0 1 and 0" gear 2 (pitch circle 2) is allowed to roll over the stationary gear 1 (pitch circle I). Thus every point on gear 2 - even the instantaneous contact point, Y, - executes a rotary movement around the corresponding instantaneous centre of rotation - pitch point C. To ensure that flank 2 neither rises above flank 1 nor cuts into it while this is happening, common tangents TT in Y must also be tangents to the circle with a radius CY around C; i.e., TT must be vertical to YC - for every pitch position:
Skew gear pair
Worrn gear set I
Transverse axes Figure 1. Gear pairs.
SPatial Gearing. The motion is accordingly transferred in the same way, even if the rule of the common normal is adhered to for only one transverse engagement position and the contact point moves around over the width during the rotary movement. For helical gearing with overlap ratio, Eq. (13) e~m> I. Wildhaher-Novikov gearing (see F8.1.8). 8.1.2 Transmission Ratio, Gear Ratio, Torque Ratio Transmission
of tooth accuracy is predominant. For low power levels, plastic gears (metal pinions). noise reduction lip to 6 dB; plastic~plastic pair, up to 12 dB as against steel-steel 120]
Efficiency, lJ. Under full load, including Planck's losses, storage losses and seal losses for oil lubrication: single-stage spur gear system with roller bearings approximately 98% (1% loss each shaft), for best quality (turbo gears) up to 99%, slow-running, grease-lubricated spur gear stage, cast 7J = 93%, milled 95%; bevel gearing 97%; hypoid gears 85 to 96%, worm gear 30 to 96% (see FS.8.3). Coefficient of friction for oil-lubricated tooth flanks /-L"I = 0.03 to 0.07. Overall efficiency 11 = 111> '1']1 to where 1)1 is the efficiency of the first stage, etc. Efficiency considerably lower for partial load and start-up (lower temperature)
8.1 Spur and Helical Gears - Gear Tooth Geometry A pair of gears must transmit rotary motion uniformly from shaft to shaft b: 'u,/ w" = const. This happens when two imaginary pitch cylinders roll over each other (Fig. 2). The tooth profiles must he produced in such a way that this condition is adhered to.
a
i
= w,/Wf, = n,/nh = rh/r,
(Fig. 2).
(I)
Entire reduction ratio i = ii, i 2 ··-. where il is the transmission in the first stage, etc.
Gear Ratio (for spur gears
=
radius ratio) (2)
u required for calculation of replacement radii of curvature (see F8.1.7). Transmission into low (gear 1 driving): i Transmission into high (gear 2 driving): i
= u. = l/u.
Pitch point C therefore divides centre distance a in inverse ratio to the angular velocities; see Eq. (6). For gears with transmission that is not constant (e.g. elliptical gear wheels), C must change its position on the centreline 0,-0 2 as per Eq. (I).
Moment Ratio (3)
In practice, for high-efficiency power gears iM not for many clock gears (see F8.1.8).
= i, but
8.1.1 Rule of the Common NonnaI Figure 3 applies to flat gears (see F9.3.2): the circumfer ential speeds of the two pitch circles must be identical at
Figure 2. Pitch cylinders with jOint pitch plane' 1 axis of small gear (pinion); .2 axis of large gear (wheel): pinion driving: WI ~ w", lJh = %; wheel driving: lVl '"'" W", (UI == Wf,; CC instantaneous axis = axis of instantaneous rotation
Figure 3>. For basic requirement of gear tooth system.
'4":'
Mechanical Machine Components • S Gearing
Left flank Right flank Reference cylinder (Pitch cylinder) Right flank line
Figure 6. Description and dimensions of spur gearing.
Figure 4. Point-for-point determination of path of contact and conjugate tooth profile.
8.1.~
Geometric Construction for Path of Contact and Conjugate Tooth Profile
Flank I and pitch circle given (Fig. 4). Normal at point Y, intersects pitch circle 1 at C ,. If gear 1 is rotated through the triangle Y,C,O, until C, coincides with C, then Y is a point on the path of contact (geometrical location of all contact points), since YC is the flank normal. Rotating the triangle YC0 2 backwards around the segment CC 2 = CC, leads Y to the point on the conjugate tooth profile, Y2, allocated to Y,.
8.1.4 Tooth Traces and Tooth Profiles Tooth Traces (Fig. 5). Spur toothing for low circumferential speeds; advantage: no axial forces, simple manufacture, suitable for sliding gears; disadvantage: runs less quietly. Helical toothing for higher bearing capacity and circumferential speed due to more uniform transmission under load, runs quietly; disadvantage: axial forces. Douhie helical toothing makes it possible to neutralise the axial forces. Disadvantage: gap for tool to run out, load distribution not always reliable, axial vibrations in certain circumstances. Note: transverse rolling motion and sliding motion take place even with helical toothing.
Toothing for Change Gears. Profile and counter-profile (rack tool for gear and mating gear) of crown toothing are identical here, so that only one tool is needed to manufacture all the gears, which can also all mesh with one another, if the profile centreline = the rolling curve during manufacture. Involute change gears [4]. 8.1.5 General Relationships for All Tooth Profiles Figures 6 and 7. The equations also apply to helical gears (henceforth referred to as I I Hel. ./ /). Transverse values (Fig. 5) are denoted by the index t and normal values by n. For spur toothing, the indices t and n can be dispensed with. For specifications for internal gears, see F8.1.7. Pitch, p. Distance between two adjacent flanks circle. If p is determined by standardised m = phr, the relevant circle will be described as ated circle. (For involute gear teeth, graduated applicable, pitch circle.)
*
P
=
TIdlz
=
on pitch module a graducircle, if
TIm,
IIHel: Pn = p, cos (3 = TImn;p, = TIm,1 I.
)e 4)
Pitches of pinion and gear must coincide.
Single Toothing. Simple tooth profile of one gear preset. Profile of mating gear to be designed as per FS.1.3, or given profile to be reproduced by tool in hobbing operation [1]. Toothing Pair. Generation of toothing by hobbing a common crown toothing reference profile: for spur toothing, this is for a flat plate - i.e. a rack (e.g. Fig. 10); for bevel toothing for a flat gear - a crown gear. Reference profile and counter-profile are not identical, two tools are required [1].
I
Normal section ~ Front sedion
a
Normal-! sectIOn
I
-.
1 i- Front section
--""'pl---
b
c
Figure S. Spur gears: a spur gear, b helical gear, c double helical gear.
Figure 7. Gearing dimensions of spur gear pair (involute gear): B single internal contact points, the leading pair of teeth just out of contact (point E); D single external contact point: subsequent pair of teeth just in contact; B the single external contact point for gear 2.
'iM'lI
8.1 Spur and Helical Gears - Gear Tooth Geometry. 8.1.6 Sliding and Rolling Motion
Angle of Action, I. Travel from start to end of meshing, A, to E I , on pitch circle (Fig. 7).
Graduated Circle Diameter (S)
Centre Distance (Fig. 2)
+ r, =
a = r,
m(z,
+
z,)/2 = mz, (1
+
u)/2
)
(6)
IIHel.: with m = m,ll. For involute gear teeth, see Fq. (50, 33). For internal gears z]., d 2 , a is negative (see };8.1.7).
Modulus, m. Important standard values. Addenda and dedenda are usually selected in relation to m. To limit the number of tools, m" should be selected from standard range. Table 1 I I Hel.: m, = m,,1 cos {3/1. (In the UK and USA, diametral pitch is usually Po = zld. With d in inches: m in nun = 25.4IP".)
Depths of Tooth. Addendum h, (normal dum, h,
=
(normal
m), deden-
=
l.l m to l.3m). IIIfel.: with m =m"1 I-
~
h"
+ 11,,2'
+ 2ha
=
2d -
Pressure Angle, ex. Angle between tangents to pitch circle at C and corresponding contact normals YC (Figs 4 and 7); for a in involute gears, see FS.1.7. IIHel.: tan a, = tan a,,/cos /31 I. Contact profile, active profile (Fig. 7). The part of tooth prot1le AK used for contact.
Additional Variables for Helical Gears Spread (for Helical Gears), U. Distance between end points of a tooth brace over the width, measured on the graduated circle arc. U = b tan fJ (Fig. 8).
/3 positive: left-hand: /3 negative. For extemal toothing, tooth alignments of pinion and gear must be opposed; for intemal gears they must be the saIne.
Overlap Ratio
Tip Diameter da = d
Eu. Ratio of angle of action to pitch. For uniform transmission of motion with spur toothing, 8" = liP> 1 required; usually 1.1 to 1.25 needed (even for helical gearing). For 8ex in involute gears, see FRl. '7.
Tooth Alignment. Right-hand:
en
= b, + h,.
Depth of tooth b
Working depth of teeth b w
Transverse Contact Ratio,
dflTlcetillf!. wheel -
(8)
2c.
Root Diameter d, = d - 2h,.
(9)
Tip Clearance, c. Distance between addendum circle and dedendum circle of mating gear (normal O.3m). IIHe!.: where m = m"II,
= O.lm
to
8~ =
Ulp,
=
b sin /3/(m"1T).
(13)
Uniform transmission of motion possible, even for small depths of tooth (limiting case zero), if 8{l > 1.
Total Contact Ratio ey = eo
+
(14)
e~.
8.1.6 Sliding and Rolling Motion According to law of motion (see A21.2), absolute velocity in direction of contact tangents, TT (Fig. 9) Wa = WaPa
=
(vt/ra) (ra sin
0'
+= Ky)
= v, (sin 0' ::;: gyira), C2 =
h, - hw = a - (d",
(15)
( 10)
+ d n )/2.
Tooth Thickness in graduated circle s
=
p - e
for space width
e
(II )
and s 2 are made smaJ1er than the nominal dimension by the toot/} thickness nUlrgin, As- This gives rotary ba<.:klash
Upper sign for contact point on dedendum flank a or tip lower sign on addendum flank -;; or root b of the tooth.
b,
SI
j(-
= fJ -
"1 --
Sl'
normal backlash
(12)
in = it
cos a; shortest distance between non-working flanks; required in order to avoid jamming from heating, from swelling (plastics!), or as a result of manufacturing tolerances. IINel.: .in = it cos all' cos {3//. Reference values for A, as per Table 4.
Transverse Path of Contact, g". For contact of utilised part of line of action. Normally bounded by addendum circle, or earlier with undercuts (Figs 7, 11).
+ at tip
(a or b); - at root (a or b).
Cumulative Speed. This is important for lubrication pressure (sec F8.3): 1J, =
U',
+
U'h =
v, [2 sin a :;: g, (llr, g, (1 + Ili)lr,]
+
I/r,,)] (16)
= l', [2 sin a:;: ~inus
sign at root a or tip b; plus sign at root b or tip
a. Cumulative Factor K,
=
1',111, = [2 sin a :;: g, (1
+
l/i)lr,].
(16a)
Table 1. Module series (DIN 780 and ISO standard '54-1977), With· out signs: distortion range L with signs) (: range II Module m (mm)
I )1.125( 12'5 )137<;( 1.5
)l.75(
)3<;(
2 )22';(
2.<; )275(
>4.'5( '5 )'5'5( (,
)7(
)14(
2<;
)4'5(
8
16 )IS( 20
)28( 52 ):\6(
';0
)22(
40
)9( 10 )II( 12
Figure 8. Overlap length, U, and helix angle, gear (DIN 39(0).
/3,
on a helical
'M@_.
Mechanical Machine Components. S Gearing
Figure 9. Speeds on tooth flanks. a Dimensions for calculation: index : driving;
b
points during engagement.
Sliding Speed. This is important for heating, seizing stress (see FS.5.l): g
-V "
I
(17)
Sliding Factor, Kg Kg
= Vg/v, = :;: gy (1/r, + l/r,,) = :;: gy
+ 1/1)/r,.
(1
(1S)
Minus sign at root a or b, plus sign at tip a or b. The sign indicates the direction of the friction force (Fig. 9b). S.1.7 Involute Teeth
Used almost exclusively in machine-building. Simple precise manufacture is possible in small cut processes (straight-flanked reference profile; Fig. 10). Change gear characteristics; unifonn transmission of motion is possible with the same tool even with centre distance variations, different tooth fonns and centre distances, by means of profile offset. The direction and size of the tooth normal p",,·m sp opl2-,;-ep opl2
force (bearing force) is constant during contact (see K5.2).
Characteristics of Involute Gears. The path of contact is a straight line with an angle of pressure a; the actual profiles of tooth flanks are circular involutes, the tooth flanks of crown gears (racks) being straight, while those of external gears are convex and those of in~ernal geared wheels concave. Circular involutes are described by the points on a straight line, the "generatrix", which rolls over a circle, the "base circle". The straight-flanked reference profile is standardised for machine manufacture in DIN S67 (Fig. lOa); for corresponding tool profiles I and II for finish machining, together with III and IV for the pre-machining of gears, see DIN 3972. Suitable and properly proportioned gears are available for most applications in this area. For a reference profile for light engineering, see DIN '58400.
SPecial Cases. The protuberance profile (Fig. lOb) cuts the tooth root cleanly, in order to avoid notches due to gear grinding. A greater depth of tooth (b~ = 25m instead of 2m) is used for particularly quiet-running gears (with high gearing, beware risk of seizing!). Angle of pressure 15° with adjustable centre distances (higher transverse contact ratio).
Profile reference line
p ' " Effective flank -~
Root line
ISO and abroad. Standards: ISO 53; AGMA 20 J.()2 , 207.06; BS 436.
Involute Function. For the calculation of numerous variables of involute gears, e.g. the tooth thickness at a particular point, it is advantageous to use the involute function, "inv a" (referred to in speech as "involute a"), which can be obtained from tables as a function of a. inva
a
Root rounding
GOOO":;?: Tooth b J
Workl
C
= tan a-a.
(19)
Gear Variables for Involute Gears. The general relationships in FS.1.5 apply. For further dimensions, see Fig. 7.
Base Circle: //Hel.:
rh =
rcos a,//.
(20)
Base Pitch: p, = p cos a = Ph' Base Circle Pitch: / / Hel.: transverse base pitch, p"
Figure 10. Basic profile of involute gear: a basic rack as per
Pr cos at"
DIN 867: b protuberance tool as per
Normal Pressure Distribution:
[49]' "p"'o
a
b: b speeds of flank contact
= (0.3 to 0.6)0;,
(addendum, baPo of basic tool profile corresponds to dedcndum, h tp of basic gear profile): c tooth flank generated by b.
=
(21)
'Mp...,
8.1 Spur and Helical Gears - Gear Tooth Geometry. 8.1. 7 Involute Teeth
Radii of Curvature. II Hel· transversell as per Figs 7 and 9a: PC!
TIC
=
Pu = CT,!
=
O.Sdh1 tan
=
UPCl' PA2 =-=-
=
CY w
= BT2 =
Q'sin
I .
all
AT} = O.S(d;! -- d~2)!'
=
PEl = O.S(d~l -_. d~2)" PB1
PBz
O.,)d 1 sin
T1R = PEl --
Pt"t'
(22)
Psi
O:w -
"w
where d h ~ 2rh .d. (Fig. 6), ~ the effective pressure angle, II He!.: = "w,!l. (For internal gears, P with index 2 is negative!)
"w
with root contact path I:
Gearing with Modified Profile (noonal case of invol-
g,=AC= PAl- p" and tip contact path I:
ga
= CE
ul(d,,/dh , ) '
=
PEl
-
i]' - tan
-
Pu,
ii' +
-
"w lu + Ii),
"w = "wi/.
IIHet.:
TranslJe,'se Contact Ratio· (24)
= 2r\
Tooth Thicknesses at geometrical radius, r:
+ inva -- inv O'y)
(s/21'
ute gears). During manufacture, tool reference profile backed off from graduated circle (radius r) by an amount xm (profile offset ~ + xm) or moved forward (-- xm) and rolled at this dimension. Base circle radii rb = r cos a remain unaltered. Thus undercutting can be avoided, larger r-ddii of curvature, thicker tooth roots, and the maintenance of specific centre distant·es are pOSSible with standardised modulus. Contact ratio usually lower, radial force greater as a result of larger effective pressure angle. Only small change in tooth foon with large numbers of teeth.
Dimensions of Gears with Modified Profiles
Tooth Thickness at radius r, (transverse value) Sy
where
Moving the tool off (positive profile offset x), reducing h Napo or using a helical gear can therefore avoid undercutting, i.e. reduce the maximum number of teeth (Figs 13 and 14)
Contact Path:
g" = 05dh , I (d"ld h , ) '
The path of the rounded rack tip (relative tooth crest track) cuts Ushaped involutes while rolling; corresponding point on path of contact: U Undercutting can reduce the contact ratio (Fig. 11; "harmfur undercut) and weakens the tooth root. The maximum number of teeth follows from the condition that II coincides with T,.
s
(2';)
given s and a at radius r At tip:
s," > O.2m"
m'7r'2 + 2x tan ,,) + A,
=
with (negative) tooth thickness margin, A,; reference values for A" Table 4 (see F82);
(Figs
1~
II1Iet.:
and 14)
Shaft Pitch a, from tooth thickness with zero-play contact (transverse value):
inv ", with
51
~
inv"
0'
from
(26)
+ [z,(s, + s,) - 27rr,]/12r, Iz, + z,l1
at radius
1'1' S2
and r2 (Eq. (27».
Q'
with r l and
+ 2x tan
On)
+ Amll. (27)
=
d + 2xm - 2h,",
(28)
IIHe/: with m·= m"ll. Tip Diameter:
=d
Y/.o
Undercut (Fig. 11). For small numbers of teeth, the addendum flank of the rack undercuts the tooth root of the gear when the point of intersection H lies below T,.
m n (7r/2
Root Diameter: d,
with a as in Eq. (6) and
fl =
cos
sn = 5,
4-
2xm + 2h"p + 2km,
IIHet.· with m = mnll, hfp, h,p, c km, addendum height alteration
(29)
(Fig. lOa).
(= telescoping,
Fig.
12), Eq. (32), to maintain the negative values of the tip
clearance for external gear pairs (positive for internal gear pairs, usually set to zero).
Centre DiMance:
a = O. ';m(z, + z,) cos "/cos
= ad cos a/cos
"w
Ct'w,
(30)
Centre Distance of Zero Gear. Manufacturing tolerance
(± centre distance variation, Am =- Am1 + A m2 ) increases or reduces backlash. For reference values for Am" Am" see Table 4 (see F8.2).
Lffectil'e Pressure Angle iov
"w
~
inv "
+
"u-
2 tan ,,(x,
This is made up of
+ x,)/(z, + zz), (31)
Figure 11. Undercut contact cannot begin before (i': for the remaining transverse path of contact g" is as per [II.
II1Iet.: inv
"WI
= inv ", + 2 tan ''"(x, + x,)/(z, + z,)II.
1."14
Mechanical Machine Components • 8 Gearing
Figure 12. Gearing with modified profile (staggered teeth), Left: gearing for gear and mating gear with joint basic profile (note: no flank contact!), Right: operating position of gearing after pushing together and alteration of addendum, k· m (note: no joint basic generating prome).
Addendum Height Alteration: (32) where ad (centre distance of zero gear) is as per Eq. (33). For reference profile as per DIN 687: '" = 20°, COS" = 0.940, tan" = 0.364, inv" = 0.0149.
Zero Gear: Xl
a
= Xz =
0,
=
0.5m(z,
ad
=
(Y'W
=
lX,
+ z,),
(33)
/ /Hel.: <>W, = ",/ /.
Additional Specifications for Involute Internal Gears. All the gear geometry equations can be used unaltered if the tooth number of the internal geared wheel, z" is made negative. All calculated values of the diameter thus become negative, as do the gear ratio and the centre distance of an internal gear pair. (But the absolute values are to be shown in the drawings!) Profile offset towards the tip - so with internal gears inwards is described as positive. Only the diameter of the root circle can be obtained from the generating tool: where a o = the centre distance during the cutting of the teeth and d. o = the gear shaper cutter tip diameter.
V Zero Gear:
Selection of Profile Offset. Favourable: gear pair with ref-
For elimination of undercutting and strengthening of pinion at expense of gear when u '" 1.
V-Gears:
Many suitable profile offset systems [4, 5].
necessary to check meshing interference, tip thicknesses
Recommendations Flexible rule, which leads to balanced gear systems, in DIN 3992, Selection from (XI + X 2 ) as per Fig. l.Ja; a as per Eqs (31) and (30), round up if necessary and determine accompanying (XI + x 2 ) as per Eqs (30) and (31) or from given a; distribute (XI + x 2 ) as per pairing line (Fig. l.Jb or c), Simple rule: OS gear as per DIN 3994 and 3995, XI = X 2 = + 0.5; <:hange gears; a = F(zi + z2)m; number F established through (ZI + Z2) and {3; F and important gear data ('an be obtained from DIN 3995.
Additional Specifications for Involute Helical Gears. Here too, the lines of contact are straight lines, but nm obliquely over the tooth flanks and wander over the tooth width during contact. The profile offset is given in multiples of the normal plane pitch; for selection of profile offset, see F8.1.7, corresponding virtual number of teeth znx as per Eq. (34). The normal tooth form of an involute spur gear with a virtual number of teeth, znxo is similar: Znx = z/(cos'
f3b cos f3) =
z/cos' f3,
(34)
and is used in the selection of profile offsets, for determining the geometrical limits (e.g. tip thickness) and for calculating the strength.
Back Helix Angle f3h
f3h = tan f3 cos "" (35) or sin {3., = sin f3 cos "n' For special gears with pinion tooth numbers 1 to 4, see from tan
[6].
erence centre distance, x = :+: 0.5 to 0.65. For z, < - 40 (extreme - 26), z, 2: 14 (extreme 12) and z, + Zz :$ - 10, respect conditions for manufacture and mounting (radial assembly). For other gear pairs with reference centre distance, see DIN 3993. Staggered teeth do not essentially increase bearing capacity, but allow for greater freedom in configuration, and of course make it and space widths (Fig. 14). For planet gears, select planet tooth number, Zp, 0.5 to 1.5 smaller than results for zero gear from z, (sun gear) and z" (internal geared wheel). x, and Xp are determined using Eqs (30) and (31) and distributed, for example as per Fig. 13; the aim should be to obtain xp + XII :5 O. For direction of pitch with helical gear, see F8.1.5. Full account of geometrical relationships: DIN 3993 [8-10].
8.1.8 Other Tooth Profiles (Besides Involute) and Gears for Non·uniform Transmission Cycloid Gears. Flank forms originate from rolling of two pitch circles on the circles of contact. Scarcely used now except for vane-type pumps, since they are difficult to manufacture accurately (individual hob for each number of teeth), sensitive to centre distance variations and not true to moment (see F8.1.2). Cit-care Gears. Higher standards in precision equipment manufacture (high degree of transmission, constant moment ratio - Eq. (3), small bearing forces, hardly any lubrication, large amount of backlash, i.e. no jamming due to dilt; though not conformal transmission, high contact ratio), can be attained by using circarc gears and involute
gears (DIN 58405) than by using cycloids. Since not conformal, transverse contact ratio always 1. Teeth thickness on pinion (generally steel) usually smaller than on wheel (generally brass or plastic). Bending stress at root of gear generaliy decisive for bearing capacity. Manufacture from metals usually by means of hobbing, sometimes copy-milling; from plastics by means of injection moulding [11].
. r.
8.1 Spur and Helical Gears - Gear Tooth Geometry. 8.1.8 Other Tooth Profiles
'at..tel
f--j---+ . .--, 27,;~1~ ~i~~~~~;;~J/,/, ,;/J;~ ~~ ,;; ~;~;~ ;~~ --::;
2.0 r----i]-----,18 1-1----1-16 -.---- - --For special cases - - f---- ---r----.. ..l--- - ~ 14 1 1.0 "085 ~ :L(O /.,; , For high tooth root and flank load-bearing capacity /;,~/ 1//.;: V/.;: P7; 08 l;//;: , }X/Y,///Y///1-->/ '/1//- /"/;: V';''; V//- I///- P5;'--: 0.6 1--- ~ ~ ~~ ~ \'':< '. ,<,~'-1;::'-.\.~"'-.-'~~'-."\J'.~'-."'\l'-.'~. ~" ..~~ --04 ~"'~ ~ r>;..~ :<:: 0:« ,,,,~. ," "::-< P~ ' '. : For well-balanced gears " 02 ::\.'-::-..: tx' " 1::-<' -..:"",!-:\.-';'-.'\..,,'-.'-::1."::'::< ,"'~ P3.C:: o ://;/ //y /. :,;,/;..V; //,V// ///- / ./~ / / r/>/ ~///. P2:!. -0. 2 ._--- - \IV >~ . For high degree of overlap ~/ /~ /~ /,;.~ ~/,~/ /h Pl;/, ~ -0 4~ \ For Special cases-+-_ -- f - - f-- 06 --
~
.-+.+:::~
i
., ,
".,'
~
:J..
'.-'"""
"'':'<
_-+ __ ±JJ. -.Lt+ .j-t--- ---t
-08 -1 O~....l......20 40.1 60 80 2",+Z"I"S4 a
J// / --t-.~;;;(
1.0
0.9 0.8 I--- -'- "," < < ~ 07 0.5
03 02 01
-0
0.458
'0'0-
;q- __
.
~
/
ft ~- ---- ~
-- -._-
- - -, - ---
!
~~i
;
l~eL ndercut
o
-'j
10 / Znx1 ~ 16,1
----
240
---
f-
~
;:::::::::: ----
260
280
,
A
'I
-
--
-.
----
--= ;:::=: c----
-
\ 1'---
---
20
/
30
~f---+--'-
~xl+Z.nQ= 27
2
r---... i
40,5050708090
2;'1"37,9
Tooth number z lor2,,1
100
1-"-"-
- -~
--- f--- -
1--55 55 54
-t--- t--- ForspfJa~~
-
300
------ ~ ---:::
Zero gearing
]-+--+1 I
~
;::;:;---
.-.-~
~-
.:-J~~
-0 4 -0 5
c
S\
lV.....- ---
-0. 21imit- ~ -0
\-
'-"'k~7~···· "7~--+-*~r---
0.5 ~;~5 x,= 04
--I-_~
100 120 140 160 180 200 220 Cumulative tooth number 12,+2/1 or (l", -lnd
",,,"'.'....
53 52 51
f------
110
----- t---f----
120
130
140
150
Figure 1~. Profile offset selection (DIN 3992). a Recommendations for profile offset factors sum. b and c Recommendations for total profile offset distribution: b transmission into slow, c transmission into rapid. For {"xample, see text. GTey area: danger of contact interference. For distribution of -'"I + Xl with Zl > 1SO, Zl = 1SO can be used.
l.t4'
Mechanical Machine Components. 8 Gearing
./".
/
-1.5 L--LL:J----L----"----'_~-L-___'_._L.____'___'''_>...>.OI -60 -50 -40 -30 -10 -10 10 10 30 40 50 60 Zn
Figure 14. Practicable profile offsets for internal gear pairs with basic profile (DIN 867). E: XI> E: x 2 : recommended range for zero gear pairs with reference centre distances. limits: 1 through minimum pinion tooth tip thickness (see Fig. 13),2 through undercut, 3 and 4 through minimum tip diameter, 5 through minimum tooth root space width of hollow gear (DIN 3960).
Transition curve Figure lfi. W-N gearing. Pinion flank convex, gear flank concave (left: basic form; right: practical execution, P2 > PI'
than PI - point contact. In rotary transmission the contact point wanders over the tooth width. Individual tools (for each modulus and helix angle) reqUired for pinion and gear for gearing with convex tip profile and concave root profile [1, 12, 13].
Bearing Capadty. Hertzian ellipticity surface is spherical surface. Corresponding expansion is greater than that in lateral sense, owing to close fitting. Pressure areas wandering over tooth width favourable for formation of lubrication pressure; friction effect smaU. Transverse sliding speed same for all flank contact points. Wear consequently uniform (favourable for running-in lapping).
Flank Bearing Capadty (from comparison of Hertzian compression values), torque approximately two to three times as high as for involute gears. Fipre IS. Mangle gear. Design for path of contact and tooth flank, dimensions.
Mangle Gears. Used for heavy-duty turntables with large diameters, rack-and-pinion jacks (Fig. IS). For roIling of W, on WI, M describes curve Z; equidistant with bolt radius gives pinion flank. Refereace V"ues~ Lowest pinion tooth number min. ZI = 8 to 12 for circumferential speed VI = 0.2 to 1.0 m/s; pin diameter d. = 1.7m; addendum, b m = m(l + 0.03z,); tooth width b = 3.3m, average bearing length of pin I = b + m + 5 mm; space radius, TL = 0.5dB ; distance, a l• = 0.15m; backlash, it = O.04m. Bearing capacity according to practical experience: Table 2:.
Wlldhaber-NoYikov (W-N) Gears
Tooth Forms. In their basic form, pinion flank consists of convex arc and gear-tooth flank of concave arc, with radius PI = P2 around pitch point C, Fig. 16. Contact on entire arc only in this meshing position, i.e. no transverse contact ratio present. Uniform transmission of motion possible only using helical gear with intermittent contact ratio B ~ > I. In order to avoid bearing edges at tip or root due to centre distance variations, p, is made rather larger
Table:l. Guide data for mangle gear of crane slewing gears with pinions made of St 70 and bolts made of St 60 for heavy duty operation (51] Peripheral force
(kN)
20
Pinion tooth number Z I Module m Tooth width b Bolt diameter dB
9
9
9
(mm) (mm) (mm)
21 80
2S 90 4S
30 110 SO
35
30
40
Tooth Root Bearing Capadty about the same as for involute gears. Point application of force entails risk of comer breakages at B ~ = 1 and breaks in the tooth centre (single tooth contact) at B~ > 1.2. Operating Peiformance. Favourable noise and osciUation behaviour with precise, rigid construction. Pitch and flank line variations lead to jerking when meshing begins. Under certain circumstances, centre distance variations (to which deformation can also contribute) bring about considerable displacement of the meshing at the tip or root, i.e. an increase in the flank and root stress, together with greater running noise. Eccentric Gears. [38-42]. Non-drcular Gears [53-57].
8.2 Tooth Errors and Tolerances, Backlash Tooth accuracy to be specified in terms of grade as per DIN 3961 to 3967! Grade 1; highest precision, grade 12 lowest. Examples; master gears Q2 to 4; marine gears and turbo-gears Q4 to 6; heavy machine construction, Q6 to 7; smaller industrial gears, crane control gears and belt gears, Q6 to 8; slow, open gears, QlO to 12; slewing rings Q9 (cast> QI2). For large tooth widths, additional specification of a contact pattern required (no interchangeable
manufacture!). Flank line corrections or profile corrections if necessary, i.e. known variations for balancing deformations effective [1]. Tolerances for individual errors (profile, pitch, concentriciry, flank lines); DIN 3962; for composite errors (tangential composite error and radial composite error): DIN 3963. Checking tangential composite error or radial composite error is frequently sufficient for acceptance of
8.3 Lubrication and Cooling
Table 3. Estimating flank lines angular deviation, J;{fj: for precise values, see DIN 3%1: hi/3=H", 4.I6bn 11; tables: DIN 3962
6
Quality gntde
7
8
9
LO
II
Table 4. Recommendations" for upper tooth thickness deviations A~n<:' upper tooth thickness tolerances Tsn (DIN .3967) and centre distance deviations Aa (DIN 39(4)
12 1~,,series~'
Factor H",
0.'570.76 I
1.32 1.85 2.59 4.01 6.22 9.63 14.9
gears. For tolerances for centre distances: DIN 3964; for tooth thicknesses: DIN 3967. For f,1~' see Table ~. For precision levels attainable using various manufacturing and heat treatnlent processes, and a comparison of the DIN grades with the ISO and AGMA grades, see Fig. 17. Recommendations for sdection of tooth thickness margins, Asnt" , tooth thickness tolerances, Tsn and centre distance margins, A,: Table 4. Thus theoretical backlash: (36)
max. ie where A,n
~
A,n< -- 1:" and + A.,
min. ie where A,n = A,,,< and - Am·
lbeoretical rotary backlash, i, = i,,/ cos
I . t....
/3.
Acceptance backlash usually smaller due to manufacturing variations.
Cast turntables DIN> 12
2·a
Turntables (normal play) Turntables, converters
29(:10)
Centre distance deviations A,
to
28
be
26
9(8)
Plastic machines
c to cd
25
7
Locomotive drives
cd
2';
7
General machine construction, heavy-duty machine construction, nonreversing
b
26
(little play)
Vehicles
d
26
7
General machine constmction. heavy-duty machine constmction, reverSing, shearing, mnoing gear
c to e
24 to 2'5
6 to
6
Operational backlash, e.g. during startup, due to gears
Machine tools
24
wanning up more rapidly than housing, under certain circumstances significantly smaller than ie.,; see Table 4, footnote a.
Agricultural tractor~, combine-harvesters
1.7to.288
-Ii ,
,
Ground
~~~---~~l
:- Shaved
~
I I
Hobbed
FInished after heat treatment
to 2'5
f
"In case of lack of experience, check as per DIN ;967. Appendix A, in particular temperature influence As per proposals by A. Seifried, Friedrichshafen bCondition: for each gear iAM'l·i : : : rAJ 'Condition. T,,, > .2 tooth thickness variation R, as per DIN 3962; check!
8.3 Lubrication and Cooling Lubricating Film Thickness. The minimum lubricating ftlm thickness at the pitch point is a suitable indicator of the lubrication condition, as per EHD theory. For steel gears, as per Oster on the basis of [141, with the gear ratio U, the numerical value equation h,
= 0 . 003
[(au)/(u
+ 1)'1"'·
(1',,1',)"". (Pc/ 840 )
,I
Finis~ed
before heat treatment
I Small dimensions I I Large quantities I Vehicle manufactUring ,,I Induction hardeo,ng , Hobbed, etc -: or Individual tooth ~ (SB€ above) Shaved , I fIa~ hardening
(Pc
J
'Hobbed etc (see above) Co Profileo etc (see above) - - - - - - - - - i--H~bed etc'; ; !Inductlon hardening or (~ee above), Profiled, etc. ___.. ' rotary. flaCle hardening I (SB€ above) -~~ - - - - - - - - - - Case-hardened Small dimenSions Single-part product'on
"2(,
in fLm (37)
Figure 17. Gearing qualIty and manufacruring proce:;:;e~ (approximate allocation of DIN, ISO and AGMA grades as per adjacem pitch error, m = 6, d = 7'5 to 150 mm). For manufacturing processes, see S5.2.
-.JF,U+l alb" ---;;-- for Eq. (48) )
~ ZHZE
is valid as an approximation. The lubricating ftlm material viscosity, "'" in mm'/s, is obtained from the bulk temperature
Here a = the centre distance and b = the width in mm, = the circumferential speed in mis, ;;.. = the oil temperature in oc, P" = the tooth power loss, from Eq. (38), in kW, P, = the Hertzian compression in N/mm' (see Eq. (37) and eo = the transverse contact f'atio. The specific
v,
luhricating film thickness,
..
,
Mechanical Machine Components. 8 Gearing
can be used for qualitative evaluation.
PKG~aA(ilG-"1.)
A> 2: mainly hydrodynamic lubrication, hardly any wear;
A < 0.7: applications in many sectors of industry, marginal lubrication predominates. Check risk of micro-pitting!
with
a~15to25W/(m'K)
(39)
with static air and unimpaired convection (lower limits: higher levels of dirt and dust, low speeds, large gears). For fans on rapidrunning shafts, a is increased by a factor of /K: spur gear with one fan,fK = 1.4; two fanS,/K = 2.5; bevel gear with one fan'/K = 2.0. Influence of wind speed and insolation considerable.
Lubricant and Lubrication Method Instructions for selection: see Table S.
Lubricant Viscosity (DIN 51502) or worked penetration (DIN 51804), dependent on temperdture: manual application; NLGI class 1 to 3 adhesive lubricant (NLGI = National Lubricating Grease Institute). Central lubrication system: NLGI I to 2 lubricating grease (transportable); spray coating: NGLI 00-0 liquid grease (sprayable); splash lubrication: NLGI 000-0 liquid grease (free-flowing); lubricating oil viscosity: reference values as per Fig. 18. (Influence of roughness, temperature, type of operation [I]. EP additives where danger of seizing exists; synthetic oils (low coefficient of friction, high viscosity index, expensive) under extreme operating conditions. For Lubricating Devices, housing connections: see F8.lO.4.
Thermal Economy. Power loss P v should not exceed cooling capacity PK • For small to medium-sized gears, air cooling through housing walls (cooling area, A, in m') and temperature difference between housing and ambient air of ~G - ~" in K, normally sufficient. Remove excess power loss by water cooling. (38) Roughly Gear losses Pvz = O. S to ) % of nominal power for each stage (with v > 20 ms, gear losses independent of load are also to be taken into account [I J). Storagelosses PVI. (see F5.5 and F6.1.2).
other losses, seals (see FS.6 ..3 and F6.6.7). Cooling capacity (heat emission) of housing:
PVf)
8.4 Materials and Heat Treatment Gear Manufacture (For worm gears, see F8.8.) For bearing capacity of materials and corresponding quality requirements, see Table 14. In addition, costs for materials and heat treatment, machinability and/or workability, noise behaviour, number of units (manufacturing process) are decisive (in many cases the only important factor) in the selection.
Typical Examples from Various Applications Gears for Tadde, Instruments, Domestic Appliances, etc. (i.e. for the trdnsmission of motion or small forces). Alloys of Zn, Mg, AI. Thermoplastics (injection moulding); automatic steels, structural steels; malleable alloys of AI, Zn, Cu, laminated plastic, thermoplastics (extruding machines, cold drawing, presses or punches, or millers); sintered metals (final sintering). Vehicle Gears. Alloyed carburising steels - milled or shaped, shaved - case-hardened - (if necessary, ground instead of shaved); low-alloy heat-treatable steels - milled or spliced, shaved - carbonitrided. Turbo-gears, Marine Gears. Alloyed heat-treatable steels - milled, ground if necessary; AI-free nitriding steels - milled, shaved (or ground) - gas-nitrided (ground if necessary); alloyed carburising steels - milled - casehardened - ground.
Table S. Selection of lubricants and type of lubrication Circumferential speed (m/s)
lubricant
Up to 2.5
Adhesion lubricants
Up to 4 (if necessary 6) Up to 8 (if necessary 10)
Type of lubrication
Apply by hand Spray lubrication
}
Gear format
Special features
Open' }
Provide for covering hood wherever possibleh Note c
}) FlUId grease )
Note d
Up to 15 Splash lubrication"
Up to 25 (if necessary 30) Over 25 (if necessary 30) Up to 40
Od
Closed
With perforated sheet walls, splash lubrication possible, cooling fins
Injection lubrication
Notef
Mist lubrication
For low-stress intermittent service
>lFor example, cement mills, rotary kilns. hSpeaying amounts, spraying times ll]; lubricate bearings separately. <'Especially for intermittent service, partwly for lifetime filling (no oil tightness required!). dlnunersion depths in operating condition, .3 to max. 6m at v > 12 mIs, absolutely necessary to check c for low values. ~Speaying oil adequate for above gears and bearings if v1 in (m/~)J./d in m > 5, otherwise wiper needed. Oil quantities approx. 5 to 101/10sskW; for oil level check ventilation, see F8.10.4. Large-scale operations, friction bearing operations, vertical operations - usually injection lubrication here too. IInjection lubrication: if oil pressure varies upwards or downwards, up to t' = 25 m/s in front of contact, above 25 mls in front 0/ and behind engagement. Injection quantities as per heat balance, roughly: Qt: = 0.8 to 1.0 l/min per cm tooth width. Total anlount of oil, Q = Qt: (0.5 to 2.5 min) with external tank, Q = Q., (4 to 30 min). Oil pressure in front of nozzle, approx. O.S to 1.0 bar, higher for turbo-gears; for fittingS, see FS.lO.4; for coolers and filters, see [1].
'41'1'
8. 4 Materials and Heat Treatment - Gear Manufacture
8,10 1
11m I l!
I Insufficient oil - - ------f----j---j--t--t-+_ I+-IH-+--t--t- High externa,llernperalure " -----I Roller beanngs [5 10 3 1------+---+_-_--+ Worrn gears -0" Increasing slress -'" 9 ~----Jerky operallon ~ f------,',! I i . .lInjection § Unalloyed oils. __ ~ - - - ('{!~I!l.~lera,-_ -.,"'", ~~s~ ~~~~Iubrication ~ Increasing splashing ~:!! ~ or abovel I " (worrn I 3l and crushing losses 140'2 u 40 r--------t-------j---r--, ~be,low) --t-- - .c;,,'" I I I ' " ' :I - Decreasing splashing and CrUShingV-"" ~- ~ '( 5 -E $For very high, transient stress--!' ,_ , '['- I :il losses ~ - v, 30 E E c E 1 0-1000 I i . II "' t'- '" ~>iBlended oils I T c a~ 700 nectlon u ncatlon Srnooth operation u --> 3 a~ 500 ~--J""''' f- Decreasing stress P;;! 20 1ii Fricllon beanngs--- ~ 90 1ii 220 a~ 100 , ~",,," Low external tern perature " C
I II
1
8
,
= ~
= ~
'w 1 0
;;u
Ol '>
'-'
10
C
'w
§1O 1---_ ',> '-'
1
9
~
~
7
~ 9 E 0
'"E
5
'" 10 2 ~ z
C<
co
0
z
--- For srnall aXial distances, fnction
8 1-- ______
-~
6
~
,""-J',:
f10 - I 1
~~
fill
"'"
beanngs, cenlrallubricatlon systern
_ _ _
I
-L "" '""" Adequale oil supply ~-"'- 'M~J-"-,--j--t--t-+-+,_-t__-+1 -50t
""'1- -"~~
~ __
h:l ~ ,;:\ ..;L__~_ ~,,,,\
~-~-----+-t-t-±l~-f-I
--l-+- I_~_
~--
++-I-,--~
J - ~~ g~~~~~~~ol~~ --I--t-~-tt+ ++-l ~-H~-I- --t--~ _L-~~_~+- '~ S" "", . . "" ...;- . . .
T
""
1-'
--'
I
-
-- i-
,II
~
I
t-r--t-+ I I: I
--
I1
1-: II! I
--
I
I
I
I
I I
I
I
I
I f~
6 7 8 9 10 12 Crrcurnferenllal speed
~ a> 1000
Injection lubncatlon
:
i )-
15 V,
10 in m/s
I"
F
"]1
700 75 f---f-- Norrnalrange ! 0
<
'-,- 1[I 30
40
,,,
50 60 70 80 90
Figure 18. Selection of lubrication oil \,iscosity for ~pur gears, bt'Vei gears and worm gears. Shown is the approximate allocation of ISO and SAE viscosity classes: distortion classes appear hatched. Splash lubrication is also possible at higher l't if centrifuged oil is fed to the tooth entrance through fins or oil guide plates.
Large-Toothed Gears, Turntables. Alloyed cast steel (attention should be paid to scrdp risk due to cavities), alloyed carburising steds (rolled) - milled - induction or flame-hardened if necessary.
Gearing for General and Construction Industry. Alloyed and unalloyed cast steel - hohhed, shaped, or planed_ Alloy carhurising steeL hobbed or similar case-hardened - ground (if necessary, Hnish milled using carbide metal hohs, honed if necessary), AI-free nitriding steels - hohbed or similar (if necessary, shaved or ground, lapped if necessary) - gas-nitrided, Non-alloy and alloyed heat-treatahle steels - hohhed or similar, shaved - bathnitrided_ Non-alloy and alloyed heat-treatable steels hobbed or similar - induction-hardened - or rotary flamehardened_ Non-alloy and alloyed heaHreatable steels hobbed or similar - induction-hardened or individual teeth flame-hardened - (ground if necessary).
Materials and Heat Treatment
Principles of Selection.
GG grey iron, GGG spheroidal
graphite iron, GS cast steel - for instructions sec Table 14. Special cast iron with suitahle heat treatment eqUivalent to heat-treatable steels (attention should be paid to machinability!) l1S1.
Heat-treatable Steels - llnhardened. Gears - and drive units larger, heavier and more expensive than with gear-cutting processes involying hardening. However: heat tn:atment (before gear cucting) frex from risk. no change in dimensions aper gear cUlting, usually no gear grinding necessary: relatively soft material compensates for design and manufacture faults earlier during running in: manual
dressing of tooth flanks pOSSible; safety against fracture usually more than suffiCient
CariJUrisfng Steels ~ Case-llardened. Expensive, but root and flank strength levels controllable for small to medium-sized gears up to highest hardness range (HRC = ';8 to 62). Distortion due to hardening requires gear grinding in single-part production (normal: d ~ 900 mm; In:s. 2'; mOl; extreme d up to 5000 mOl; m up to 56 mm). For coarser grades unground (see Fig. 17) (usually d.:s. 2';0 mm. m :'S 6 mm; with limitation d.:s. ';00 rum, m .:s.lOmm). Quenching cracks certainly avoidahle ~ reliable hardening process ~ Rotary Hardening (flame or induction hardening). Cost-effective for small to medium-Sized gears (normaL d .:s. 200 mm, m.:s. 6 mm: extreme, d up to 1';00 mm, m up to 18mm), certainly controllahle in medium hardnes~ range (HRC = 45 to 56), higher risk of cracking thereafter. Uniform gear quality only with constant material values and heat treatment kept constam [1 S1.
lIeat-Treatable Steels
Heat-Treatable Steels - Individual/'ooth Ilardening
~
Twill-Hank
lIardening (flame or induction hardening). Cost-effective for large gears (d up to about 5000 mm, In > H mOl); controllahle in medium hardness range (HRC = 45 to S6). Careful preparation (hardness sampling). constant, i.e. continuously monitored hardness regI..ilating data required. Little distortion, g(:ar grinding usuaUy not necessary. Tooth base unhardened, reduced root strength [16J. Rejininp SteeL., - Individual Tooth Hardeninp,
~
,Space Ilardening
(flame or inducation hardening). Tooth base abo hardened. Costeffective for large gears in medium hardness range (as for twinflank hardening, but flame only at In :.> 16 mm) (HRC == 4~ to .,2, pOSSibly .,6). Low hardness risk (quenching cracks) only with
1:41'4
Mechanical Machine Components _ B Gearing
appropriate preparation and monitoring, long years of experience, suitable materials and optimum hardness conditions (hardness sampling). little distortion, but frequent pitch errors at starr of hard-
ening; gear grinding often required [t 5). AI/ree IVitriding Steels, Ileat-Treatable Steels, Carburising Steels -
nitrided (long-Duration Gas-Nitrided). Low distortion, difficult process. Normal: nitriding depth, nhd, = 0.3 mm, d In :5 6 mm; more difficult: nhd = 0.6 mm, d
< 300 mm, < 600 mm,
m < 1() mm. Lower strength values should be specified for larger d and m with nitriding steels! Here, and with thin-walled gears, gear grinding usually needed after nitriding due to distortion. High
strength attainable with certainty only with special grade of materia!, long years of experience, optimum manufacturing and inspection equipment. Otherwise, strength can vary widely. Nitriding steels especially sensitive to impacts and bearing edges. < 15 j.Lm white layer should be aimed for. Heat-Treatable Sleels - Nilro-carburised (Short-Duration Gas/\litrided). New low·distortion process. avoids many short-duration bath nitriding problems 1171, has largely replaced it. Low overload capacity. Heal-treatable Steels - Nilro-carburised (Short-duration Bath-
process. Normal: d < 300 mm, m < 6 mm; more diff1cult: d up to 600 mm, m up to 10 mm. Practically no diffusion zone. i.e. reduced load capacity, if white layer « 30 j.Lm thick) worn. nitrided). Low-distortion
lIeat-It'eatable ."leels - Carbo-nitrided. Hardness penetrdtion depths (nitrogenous martensite layers) 0.2 to 0.6 mm. Highest core strength possible to support thin hardened case Suitable for small gears made in large quantities.
Pitting. Crumbling, especially between dedendum and pitch circles, from excess flank pressure. Initial pitting ends heat-treatable steel local overloads and stops - and so is harmless. Progressive pitting destroys tooth flanks. Remedy: Large radii of curvature (proftle offset), surface hardening (especially case-hardening) (Fig. 19), more viscous oils, precise gear-cutting, low flank roughness. Micropitting. Numerous microscopically small incipient cracks and breaks, optical impression of a grey fleck. Remedy through improved lubrication conditions (also influence of additive) [57]. Hot Seizing. Grooves and seizing marks in high sliding speed range as a result of limiting temperature conditioned by material and lubricant. Remedy through lower moduli, tip and root relief, nitriding, low flank roughness (running-in), especially effective: extreme-pressure oils (oils with chemically active additives). Cold Seizing. Groove wear with removal of large amounts of material at low circumferential speeds. Remedy through better gear precision, smoother tooth flanks, more viscous lubricants, tip relief. Abrasion Wear. Laminar removal of material, especially at tip and root, often decisive at low circumferential
8.5 Load Capacity of Spur and Helical Gears
speeds (II < 0.5 m/s) as a result of insufficient lubrication pressure being generated. Can be remedied through high lubricant viscosity, certain synthetic lubricants, many extreme pressure additives, MoS suspension, surface hardening or nitriding. Important: Equal tlank hardness on pinion and gear.
8.5.1 Types of Tooth Damage and Remedies
8.5.2 Checklist
For definitions and origins, see DIN 3979, ct, Fig. 19.
Before design work begins, all requirements for and influences on the functioning of the gear are to be listed. Often decisive for success or failure. Instructions: Table 6.
Forced Rupture. Usually from accidents, jamming or similar; difficult to estimate forces invoLved. Remedy: overload protection, breaking pieces. Endurance Failure. Fatigue fracture after comparatively long running periods above endurance limit, usually from notches, quenching cracks, faulty material or faulty hardness treatment in tooth root. Remedies: higher moduli, effective pressure angle (profile
()ffse~),
root rounding
(avoid grinding notches), surface hardening (especially case-hardening), shot-blasting, precise gear-cutting, tooth end-relief or crowning to relieve ends of teeth.
8.S.~
Guide Data for Gear Rating
Gear data (tmnsmission, modulus, centre distance, diam-
eter, contact ratio (see FB.1.2, F8.1.4, F8.1.5, F8.1.7).
Pinion Diameter tI.. From the simplified characteristic value for the rolling pressure, K' = [F, (u + 1)/(bd,u)]. it follows that d
J
>i
,-
2M, K' (bid,)
U
+I
~----
U
.
(40)
In contrdst to DIN 3990 and other gear standards, the torque is described as M instead of T, in order to maintain unifonnity for all areas of application. Experimental values for K' as per gears manufactured; examples are given in Table 7. For the selection of material and heat treatment see FB.4. For heat-treatable steels, the hardness of the pinion material selected should be approximately HB = 40 above that of the gear material.
Tooth Width" (as per guide data for bid" Table 8). For larger widths, flank line corrections required to balance out deformations. Overlap ratio: attention should be paid to Eq. (13).
a
0.3
Cucumferenllal speed '\ in m/s Fiprc 19. Main load-bearing capacity limits of gears: a heaHreatable steel, b carburising steel; I wear limit, 2 tooth break limit, 3 seizing limit (hot seizing), 4 pitting limit, 5 micro-pitting limit.
Number of Teeth and Modulus. Standard pinion tooth numbers - Table 9, with modulus determined using Eq. (5). Minimum moduli conditioned by risk of tooth comer break, Table 10. Standardised modulus range: Table 1. When the modulus has been determined, check
8. 'i Load Capacity of Spur and Helical Gears. 8. 5.3 Guide Data for Gear Rating
1."4'
Table 6. Checklist for gear drives (plus sketch with connection dimensions)
EJJect on: sealing A, application factor B, manufacture F, gear model G, housing H, construction, cooling/heating K, bearing L, lubrication S, gearing V, permissible stress Z I. Main Junctiuns required for draft calculation o Drive/power takeoff speeds (transmission
o functions: type of gear (spur gears, bevel gears, etc.), type of fitting (stationary, slipon, flange gear, etc.), other (application factor, multi-motor drive, flywheels, drive/power takeoff left-hand/righto hand/optional) (see also 2.4) .................. K
o Customer specifications for main
constant, switch step tolerance); direction of rotation constant/alternating... Z o Type of operating machine, type of drive machine. .. ........... B
2. Other functions required for draft, calculation and design 2.1 OPerating data 2.3 Forces on gear o Number of machine startups... . .... B 0 Axial forces on drive and power takeoff o Consequences of a damage incident (endangering human life, loss of production) . . .... Z
0
shafts (e.g. toothed-gear type coupling).. .. H, L, V Forces on housing .............................. H, L
o Overturning, starting and switching-off o Radial forces on drive and power takeoff shafts (e.g. chain sprocket, belt pulley) .. moments of machine, level, quantity and duration of shocks in operation, peak ............. H, L moment, catastrophe moment.. . . ... B o Return barrier .. ....... s o Running time per day (% operating time) Z
Position of working machine to drive machine (position of drive shaft to power take-off shaft of gear, variable position, limits), type of gear, if applicable centre distance .. K Power, permanent operational moment, nOminal moment of working/drive machine, maximum moment, starting moment or similar ... .. .................. Z
2.5 Lubrication Heating (for startup)
0 0
COOling (fresh water, salt water, brackish water or air, temperature); central cooling system or individual cooling
o Lubricant freely selectable/specifications o Provision through central lubrication system (lubricant, viscosity, pressure) or individual gear lubrication
o Overload safety, switching-off moment.. . B o Reversal of direction of force (reverse operation)..
.Z
2.2 Manufacturing data
2.4 Customer requirements: specifications, 2.6 Environment, location acceptance conditions o Restrictions on materials selection o Type of couplings on drive and power o Location (hall, covered, in open air) .. (machinability, delivery time) .................. Z takeoff.. .. ....... L, V ............. A, S, K o Dimensions and weight restrictions due to machine tools, furnace dimensions, .. ........ F, Z hardening devices ..
o Calculation specifications (e.g. classification societies, factory specifications) ...... ............ ........... ............ Z
o Tools available.. . .............................. F, V o Form of shaft journals on drive and power takeoff (flaoge forged on - hole circle, adjustment spring or similar corrected for oil pressure unit) o Noise, efficiency, guarantee (type of test run).. .. .......... V, H, F
o Restrictions for assembly, installation, space, weight, transportation, dirt, dust, foreign bodies, spray water, water vapour .. .. ............... A, H, F o Foundation (e.g. steel frame, rigid concrete): separate (joint with drive and power takeoff) . ................................. H o Temperature (max., min.), insulation ... K, S
o Design (forged, welded, shrunken tooth rims; shaft-hub connection; cast, welded housing).. ......................................... K, H o Accident prevention specifications ..... K, Z
whether sufficient rim thickness is available under the tooth root with the pinion mounted (adjusting spring or similar) (see Fig. 49), and whether the remaining shaft section is adequate with a geared shaft.
advantages are obtained through helical gear. For coarser grades and material for casting also used with unhardened steels (can withstand running-in, surplus safety against fracture).
Spur GeariAg - Helical GeariAg. For properties, see F8.1.4. If low-noise running is required, and for jerky operation, it is preferable to go over to helical gears and fmer quality. For average conditions:
Up to v, = 5 mis, Q8-9 unhardened, Q7 .......... drdened,
Spur: Up to v, Q6-7.
= 1 mls Q)()-12,
up to 5 mls Q8-9, up to 20 mls
Helical or Double Helical: Required for Q8 hardened gears or more precise, otherwise increased danger of tooth corner breaks and no
Up to v, = 2 mls QIO-12 unhardened, Q7-8 hardened, Up to v, = 20 mis, Q6-7, above v, = 40 mls with Q4-5.
Angle. Single helical gear f3 = 6 to 15° (limitation of axial force). Check overlap ratio, Eq. (13): up to v, = 20 m/s; Be ~ 1.0(0.9); By 2:; 2,2; above 40 m/s: Be 2:; 1.2, By 2:; 2,5. Double helical gear only if single helical gear too wide or axial forces too great: f3 = 20 to 30° Note: Only fix one shaft axially and check whether axial Helix
'MICJ,'
Mechanical Machine Components • S Gearing
Table 7. K* factors of manufactured spur gears (for nominal power, unless otherwise stated) as per company specifications and [1, 2, 47, 481. Material: steel (unless otherwise stated), Heat treatment: v, quenched and subsequently drawn; ch, case-hardened; 0, nitrided. Machining: f, milled, planbed, slotted; s, shaved; g, ground Application Drive/power takeoff
v
K* factor
Gear
Pinion
Material Heat trcament Machining
Hardness
Material Heat treatment Machining
Hardness
v, f
v, f n, S
eh, g
225 HB >60 HRC >58 HRC
180 HB >60 HRC >58 HRC
0.80 2.0 2.8
E-Motorl
v, f
210 HB
Industrial operation
v. f eh. g
350 HB >58 HRC
v, f v, f
eh, g
180 HB 300 HB >58 HRC
1.2 2.0 4.4
v, f v, f eh, g
210 HB 350 HB >58 HRC
eh, g
180 HB 300 HB >58 HRC
4.0
v, f v, f
225 HB 260 HB
v, f v, f
210 HB
0.6 1.0
7.5
eh, g
>58 HRC
v, f
320 HB
1.5
0.,
v, f
260 HB
GS. f
180 HB
1.3
Turbine/Generator
>20 >20 >20
24-hoUf operation
10
E-Motor/largc-scale operation (lifts, rotary furnaces,
<5
Comments
(N/mm')
(m/s)
il, S
eh, g
v, f v, f
180 HB
1.0 1.8
mills)
K,= 1.6
Converter (for maxil'l'lum moment)
E-motorl Machine tools Hobbing machines
Not
catastrophe moment 22
03
eh. g eh, g
>58 HRC >58 HRC
eh, g eh, g
>58 HRC >58 HRC
3.0 9.0
For peak moment
seldom arising
Milling machines (spindle 22 head)
eh, g
>58 HRC
Cast polyamide 12 g, f
75 Shore hardness
0.70
(SH)D
E-motor/Crane hoisting unit (for max. weightlifting capacity and continuous operation)
10 to 14 4 to 8 2 to 4 0'; to 2
v, v, v, v,
E-motor/Gripper hoisting unit (for max. gripper closing moment)
12 6
eh, g
E-MotorlSmall industrial operation E-Motor/Small units
f f f f
190/230 190/230 190/230 190/230
eh, g
>58 HRC >58 HRC >58 HRC
eh, g eh, g eh, g
>58 HRC >58 HRC >58 HRC
<5
v, f v, f
350 HB 350 HB
laminated plastic Polyamide
0.53 0.35
<5
v, f 200 HB v, f 200 HB Brass, aluminium
Zinc die casting Brass, aluminium Brass, aluminium
0.20
<3 <3
f f f f
eh, g
230/280 230/280 230/280 230/280
HB HB HB HB
v, v, v, v,
HB
1.1
HB HB HB
1.3
1.6 1.8 7.0
11.0 15.0
Stage 1 Stage 2 Stage 3 Stage 4 Stage 1 Stage 2 Stage 3
0.20 0.10
aApplication factor for calculation.
forces are being introduced from outside (causing nonuniform force distribution!). In general, arrowhead should mn behind. Attention should be paid to limits of manufacture (e.g. hob exit) (see FS.1O.3).
Basic Profile
See Fig. 10.
Profile Offset
See F8. I. 7.
Bearing Forces (Fig. 20). Tooth normal force, O'W{ acts as lateral force, axial force, Fx = Ft tan f3 on lever arm r on shaft. Radial and thmst bearing forces at A and B should be determined from this in accordance
Fr/cos
8. 'j Load Capacity of Spur and Helical Gears. 8.5.4 Evaluation of Load Capaciry
Table 8. Maximum foundation"
\'alue~
for bid) of fixed spur gears with rigid
Straight and helical gears: twin-sid('d, symmetrical mounting: Normalised (HB ::::::: 180) bid] -s; 1.6 Quenched and subsequently dra.wn bld l S~ 1.1 (HB 2> 200)
Case-hardened or surface-layer hardened
bid)
Table 10. Minimum values for mn D[l'\ gear quality
Bearings
min mil or tnt
11 to 12
Steel constmction, light housing
b/lO to b/1,
Steel construction or floating pUlion
b/l S to /,/2,
1.1
S
8to')
Nitrided
Double hdical gear
Bld l
Us limes bld t
'$
values ahove, for H see Fig. 5 Twin-sided, asymmetrical bearings
80% of values above
Same size pinion and gears (broadfaced steel gears and i == I)
120'.!-;, of valucs above
/-<1oaling bearings
';0% of values above
1.111
6to'7
Well supported in housing
b/20 to b/30
6 to 7
Exactly parallel. rigid housing
b/2'; to h/3';
';to6
bld , ::;: 1, exactly paralleL rigid mouming
b/40 to b/60
.!For lighter models on steel frame approx. 60'!";) of values
Table 9. Standard pinion tooth numbers. spt'ects, n < 1000 min I. upper range for n
ZI"
Lower range for
;> :~OOO
min
Transmission
I
Figure 20. Tooth force forces,
H
Quenched and suhsequently drawn {() 230 liB Above ';00 HB (and hard/quenched and subsequently dmwn)
';2 to 60
29 to S-;
2':; to ':;0
22 to 4':;
';0 to ")0
27 to 45
23 10 40
20 to 3';
Cast iron
26 to '-1'5
23 to 40
21 to';';
18
Nitrided
24 to 40
21 to .;';
19 [()';I
16 to 26
(0
30
components for Gdculating bearing
rigid gear. average lubrication conditions, and strength values which have been determined using standard reference test gears under standard test conditions. Different conditions are present in reality: external additional forces due to jerky starting and load variations; internal additional forces due to tooth forming errors and deformation effects: model influence, lubrication (circumferential speed; viscosity, rouglmess); root rounding, etc. The effect of these deviations is determined through intluence factors.
Input
Quantities. See
diagram
with
example
(p. FI36). Cas(:-hardcned (or surfacehardened)
z
= 12
z
= ~
z
="
z
= I to
.:I
21 to :\2
19 to 29
16 to 2';
14 to 22
Lowest practical tooth number for power gear (mating tooth number 2 23) Lowest tooth number for transmission of motion for reference profile as per DIN 86"', straight gear Lowest tooth number for transmission of motion for reference profile as per DIl'\' ':;8400 (light engineering), straight gear For transmission of motion. possible with staggered gears or helical gears, E" < 1 [7 J
with bearing distances. Attention should be paid to overturning moment of axial forces in calculating radial forces!
8.5.4 Evaluation of Load Capadty A check should be carried out to see whether the gear has adequate theoretical margins of safety against all failure limits, in so far as a rough estimate as per Eq. (40) and Table 7 is not :':tufficient.
Basic Concepts. Calculation is based on the nominal circumferential force exerted on the tooth of a faultless
Peripheral force
F,
=
2M/d
=
2P/(dw);
(41)
Peripheral velociry
v,
=
O.5dw
=
TIdn.
(42)
Application for simplified calculation for industrial = 20°, h,o/m = 1.25 gears: DIN 867 hasic profile: :+: 0.05, p,,,/m = 0.25 :+: 0.05. Number of teeth in pinion: 20"" z, "" 50. Average to high load: KAF,/b 2: 200 N/mm tooth width. Operation in subcritical range: see Fig. 21. Transverse contact ratio: 1.2 < e u < 1.9. v, > 1 m/s. Roughness in root rounding R, < 16 f-lm. Lubricant as per Table 5 and Fig. 18. Continuously operating gears. For helical gears, EO I~ ?: 1. For divergent conditions, calculations as per DIN ;)990 [1].
"0
Force Factors (These determine the controlling force per mm of tooth width, valid for all limits of stress.) The factors depend on the controlling circumferential force; they are calculated approximately as follows: K with grade of gear cutting and KII/3 or KI'/3 with circumferential force FtKAKJb. Many force factors become 1 with small errors and high external cirCllnlferential forces.
Application Factor K A • This takes into consideration the additional forces introduced by the drive or the power
'41Cg
Mechanical Machine Components. 8 Gearing
takeoff. For guide data, see Table 11. If the maximum moment is being used for the calculation (see Table llc), then KA = 1 should be used.
Dynamic Factor. This takes into account internal dynamic additional forces: Fig. :21. Width Factor K Hp , (flank). = KF~ (root) takes into account influence of manufacturing tolerances, 1m" and cumulative deformation, .f.,hg, on the force distribution over the tooth width:
Table 11. Continued c Examples of method of operation of driven machines
Method of operation
Driven machines
Uniform
Current generators, uniformly loaded belt conveyors or plate conveyors, worm conveyors, light lifts, machine tool feed drives, ventilators, turbo-compressors, agitators and mixers for substances of uniform density, punches, when laid out in accordance with maximum cutting moment
Moderate shocks
Non-uniformly loaded belt conveyors or plate conveyors, machine tool main drives, heavy lifts, crane slewing gears, heavy duty centrifuges, agitators and mixers for substances of non-uniform density, metering pumps, piston pumps with several cylinders
Average shocks
Mixers for interrupted openltion with rubber and plastics, light ball mills, woodworking, single-cylinder piston pumps
(43)
is determined, and KH~ (= KF~) is derived from Fig. 22. For x~, see Table 12: .f.,hg = 1. 33.f.,h' Set 1m, = IH~ of a gear as per Table 3 or as per special specification. I mg as per approved gears, Table 13; the construction should be made suitably rigid. In case of any doubt, deformation should be checked - especially of pinion shaft. Check can be carried out under load, in accordance with bearing pattern, using bearing pattern lacquer insoluble in oil (DIN 3990).
Face Factors K Ha (nank) and K Fa (Root) take into account non-uniform distribution of circumferential force on engaged pairs of teeth as a result of pitch and form deviations. For rough estimates or coarse gears under low stress:
St~ong
Table 11. Application factors for gear drives a For industrial gears (n < 3600 min-I, (zlv/lOO)· [u 2)(1 u 2 )]t < 10 with VI in rols) Method of operation of drive machine (for examples, see Table llb)
Method of operation of driven machine
L1niform
Moderate
Average
shocks
shocks
Strong shocksa
1.00
1.25
1.50
1.75
Ught shocks
1.\0
1.35
1.60
1.85
Moderate shocks
1.25
1.50
1.75
2.0
Strong shocks
1.50
1.75
2.0
Voifann
+
2.25
or above
shocks
Bucket chain drives, sieve drives, dipper dredgers, heavy-duty ball mills, rubber kneaders, smelting equipment, heavy-duty metering pumps, rotary boring equipment, pan grinders
o The table values apply only to the nominal moment of the working machine. Thus, the nominal moment of the drive motor can be used in its place, provided this corresponds to the moment requirement of the working machine. o The values apply only for gears that are not operating in the resonance range, and only with unifonn power consumption levels. In applications involving unusually severe stresses, motors with high starting torque, intermittent service or operation under extreme repeated sudden loads, gears must be checked for static strength and time strength. o If special application factors are required for specific gears then they are to be used. o Torque values resulting from inertia moments are to be respected in a brake. Moreover, these are decisive for the maximum gear stress. o For a hydraulic coupling between motor and gear, the KA values for moderate, average and strong shocks are reduced if the characteristic of the coupling permits this.
aNitrided gears not suitable in general
b Examples of method of operation of drive machines ) (44) Method of operation
Drive machine
Uniform
Electric motor, steam turbine, gas turbine in unifonn operation (low start-up moments seldom arising)
Ught shocks
Steam turbines, gas turbines, hydraulic motors, electric motors (larger startup moments arising more frequently)
Moderate shocks Strong shocks
Multi-cylinder combustion motor Single-cylinder combustion motor
KFo
= IIYe
KFo.
=
1.2; ) (45) 1.4. We are calculating Ze (see Eq. (50» and Y e (see Eq. (54») on the safe side here. For gears under normal load (fatigue failure margin of safety, SF:s 2, pitting margin of safety, SG:S 1.3), DIN grade 8 or fmer with spur gears, or 7 or fmer with helical gears: 2:
eo..n ;:::
(46)
Safety Against Pitting The flank compression (Hertzian compression; see B4.2) at the pitch point must be smaller than the permissible compression, with the condition:
, . ,• •
8.5 Load Capacity of Spur and Helical Gears. 8.5.4 Evaluation of Load Capacity
Table 121. Run-in characteristics for Eq. (43) Material
xl
UHlim
N/mm' Cast iron Heat-treatable steel HeaHreatable steel Heat-treatable steel case-hardened or nitrided
0.45h 0.20b 0.6Ob 0.73b 0.85
400 800 1200
aApplies to any Fftx = (fma + !shg) at v ~ 5 mIs, for F(J~ < 80 tJ.m at 5 m/s :os;: VI < 10 mis, for F(Jx < 40 J.Ul1. at VI ~ 10 m/s. At larger see DIN 3990 [II bIf necessary, carry out linear interpolation, and also where rack and pinion are made of different materials
F._
Table 13. Guide data for permissible flank line variations due to cumulative deformation, t..., in ILlIl (for gear pair in gear unit) Tooth width (mm)
Above Above Above Above Above Above up to
20
b
e.
e.
Figure :u. Dynamic factor K. (DIN 3990/ISO/DIS 6336): a spur gears. b helical gears with 2" I (for < I, see DIN 3990, [II).
1.05 1.06 1.0B 1.10 ~
""
1.15
!.120
>if'
""" " " f'- "'-
"
"'- "-
f'-
I'\. i>@
"'- "'-"
"
" "
_
f\.. "
"
,;,
1~1S
~~ "-Oa~---> 17 -
G?00~~~
~~~'''\ ~ 1'f'f-a ~ ",,-
"" ~'f'~~~"'-,,'""" rz -~ "'- "'" "'-
I'-., 1.30 I'-., 1.40
"-"",,", r-c'"
<'a "- " ~o~
~.?J'O~/OOO ~
"-00
I
""" I
"-
~ ""
"'-
20 40 100 200 315 560 up to up to up to up to up to 40 100 200 560 315
Very rigid gear (e.g. stationary turb01lear)
6.5
7
8
10
12
16
Average rigidiry (usually industrial gears)
6
7
8
II
14
18
24
Flexible gears
10
I3
18
25
30
38
50
For pliant, flexible gears (e.g. welded single-web gears and small helix angles, with small hub diameters and small hub widths), use f shg from column 2 for calculation.
"-
"" "
""'-"""
""- ""1.60 ~~ f'. """"~s 1.BO 2.00 5 15 20 25 30 67B91O F~y in flm
40
"'-
50 60
Pc is the Hertzian compression at pitch point; Zx is the amplttude factor for pitting strength (Fig. :Z~). ZH' the zone factor, determines the curvature at the pitch point:
Figure :1:1. Width factor, KKy (= K,.) (DIN 3990/ISO).
2 cos f3. cos <>W, cos2 0\ sin awt .
(49)
Z., the elasticity factor: St/St: Z. = 190 ,IN/mm2, St/GG: Z. = 165 ,IN/mm2, GG/GG: ZE = 145 ,IN/mm2. where |
Z., the contact ratio factor; Z~, the helix factor: Z. = ,J (4 Z. =
is the nominal value offlank compression:
Z~ (48)
8 a) /
3 for spur gearing,
~ for helical gearing (8~
= Jcos
2:
1),
(50)
/3.
u, the gear ratio, Z2/Z" is negative for internal gear pairs. For lubricants and viscosity values other than those as per
I.KlI
Mechanical Machine Components. 8 Gearing
Table 14. Standard gear materials, application, strength 8"
No. Type, treatment
Application, properties
HB Flanks
O"Hhm
(N/mm.l)!\
Grey iron DIN 1691
GG 20 GG 25
6 7
8 9 ]()
11 12
80"
220
340"
110b
For small dimensions, properties between GG and GS
150 220
350" 440"
310"
Spheroidal graphite iron DIN 1693
GGG 40 GGG 60
12%
For large dimensions also; properties between GG and GS, flame and induction hardening also possible
180 250 350
390 to 470" 490 to 570" 700
280 to 370" 330 to 430" 520
160 180
320" 340"
250" 270"
120 160 190
320" 370" 430"
2BO"
Unalloyed cast sted DIN 1681
GGG lOOk
GS 52.1 GD 60.1
Standard structural steels DIN 17 LOO
St 37 5t 50 St 60
4% 4%
18% 1;%
25% } 20%
More cost-effective for large dimensions than rolled or forged gears - difficult to cast (cavities, casting stresses)
Easily welded, no defined join
15%
200
Ck 45
no 980 lOBO
lL90
500
1000 2500
250" 300"
680 650 880 BOO 960 B70 1050 940 lL60 1050
5000 10000 190 270 300 310 320 1 120(l 1100 350
700 740 790 800;1200' 1000' 1300f
34 CrMo 4
Rotary hardening (individual tooth hardening), individual tooth hardening, not sensitive to cracks, for high core strength at unhardened tooth root
Heat-treatable steels and carburising steels, nitrided
42 CrMo 4 V
Nitriding steels, nitrided
31 CrMo V 9 V
16 MnCr 5 V
24
14 CrMo V 6.9 V
26 Heat-treatable steels and 27 carburising steels, 28 nitro-carburised
C 45 N 16 MoCr 5N
430 530 580 590
to 530"·m to 7l0"·m to 770"·m to 780"·m 600 to 790h .m 650 to 840h .m
Rotary hardening, small dimensions, b < 20 Rotary or individual tooth hardening
42 CrMo 4 34 CrNiMo 6
Carbo-nitrided
280 h
Rm in N/mm.l for heat treatment crosssection b as per
Heat-treatable steels, flame18 hardened or inductionhardened 19 20 17
29
JOO"
12% 3%
Ck 45 N" 34 CrMo 4 V" 42 CrMo 4 VO 34 CrNiMo 6 V 30 CrNiMo 8 V 34 NiCrMo 12.8 V
23
180
GTS 35 GT565
13 14 15 16 16A 16B
22
For complicated gear forms, costeffective, easily machinable, silencing - susceptible to shock
Black malleable iron DIN 1692
DIN 17 200 heattreatable steels (also as cast steel m )
21
a Fl" (N/mm2)!l
}
320 430 450 460 470 490
to 400"·m
to 580"m to 620"·m to 620"·m to 640"·m to 650h . m
Root also hardened
50 to 55 HRC
1000 to 1230
500 to 750 Root not hardened 300 to 450
Nht < 0.6; Rm > 800; m < 16; Rather suitable for running-in, less edge-sensitive than 31 erMa V 9 Nht < 0.6; 8 m > 700; m < 10
48 to 57
780 to 1000
520
Standard steel Nht < 0.6; Rm > 900; m < 16; Edge-sensitive for Nht > 0.6; Rm > 900; til < 16
60 to 63 HRC
Low distortion, more favourable price: d < 300; m < 6
42 CrMo 4 V
Higher core strength and surface hardness: d < 600; m < 10
34 Cr 4 V
Core strength up to 45 HRC, vehicle gears
}
1120 (m 1250
(til
to
740 HRC
< 16) < 10) 560 to 840
42 to 45 HRC
650 to 760
460 to 600
52 to 55 HRC
650 to 800
460 to 640
55 to 60 HRC
1100
to
1350
600 to 900
8.5 Load Capacity of Spur and Helical Gears. 8.5.4 Evaluation of Load Capacity
1.lCpI
Table 14. Continued No. Type, treatment
Application. properties
HB Flanks
ITIIHm
(N/mm2)/o::
30 Carhurising steels
16 MnCr 5
Standard steel: normal up to
31 32
15 CrNi 6 17 CrNiMo 6
m = 20 For large dimensions above In lJ oder sudden load above m =
DIN 17210
Case-hardened
58 to 62 HRC =
1300 to 1500
620 to 1000
16;
')
aBreaking elongation as dimension for viscosity. bFor lower third or scatter range. "Budget-priced, easily machinable; for advantageous smoothabk black-white structures,
ITlllim
to 700.
steel for medium-sized and large gears.
fObtainable.
limiting values for lTHhm and (TFE for quality industrial gears (steel production monitoring, high degree of purity, ground tooth flanks, acceptance as per factory certificate, many years of experience with carefully supervised heat treatment, comprehensive inspection of surface hardness, progression of the hardening process, structure, etc.) Lower limiting values and values given without scatter range safely obtainable. They apply to materials taken from stock and involve limited checks on the main material data and heat treatment data. hCarry out linear interpolation in case of hardness variations in groups nos. 1/2, 5/4, '; to 7, 8/9, 10 to 12, 13 to 16 B. kAustempered. m UHhm and U FE approx. 80 ~/mm2 lower for GS
fl.Upper
Table 15. Guide data for saf!:ty factors
Real pitch module m,
Damage limits
Continuous strength
Load acceptance
Maximum momenth
Nominal moment X Application factor
(a) - (b) - (e)
(a)
(b)
(e)
Pitting safety SH min
0.'; to 0.7
1.0 to 1.2
1.3 to 1.6
Tooth break safety
0.7 to 1.0
1.4 to 1.';'
1.6 to 3.0'
aI" a H Structural and heat-treatable steels, }
GGG perlite, GTS perlite
continuous strength
b F , bl! Edge-hardened steels
Grey iron, GGG ferr. d F, d H All materials under static stress e F , eH Nitrided steels
CF , (.'H
Figure 2:3. Size factor for tooth root strength (index F). Size factor for pitting resistance (index H) as per DIN 3990, ISO/DIS 6336,
Table '; and Fig. 18, take influence of viscosity and speed as per DIN 3900 into account. Use 85% a lllim for milled tooth profiles (roughness influence). For hardened ground mating gears, a HHm of heat-treated gears can be increased by the mating of material factor,
(a) For calculations with maximum moment against continuous strength (e.g. shearing, pressing, converters, hoisting units): values apply to heat-treated or case-hardened gears (avoid nitriding). (b) Normal case (usually industrial gears): plant gears to fulfil higher specifications; values in upper range. (c) High reliability, critical cases (very high endurance, high risk of damage, high consequent costs, no spare parts, no overload safety devices - e.g. large-scale gears, turbo-gears, marine gears. aircmft gears). "Adequate safety (approx. 1.';) against maximum moment provided (e.g. start-up shocks). bTheoretical working moment.
Zw: Zw
=
1.2 - (HB - 130)/1700
(51)
using the HB of the heat-treated gear. Equation (48) applies to helical gears where 8~ 20 I; it can also be used for helical gears where 8~= < I and spur gears, provided profile offset values are selected as per Fig. 13. Otherwise, see DIN 3990. Convert to specific internal point of application, B (see Fig. 7) (DIN 3390) [1], at z". < 20: a liD .
Minimum margin of safety Table 15.
SlImi":
for guide data, see
Micro-pitting (see [58]) approximately: Ani< = 0.7. At A > Actio according to experience so far, micro-pitting should not be expected. (A: see F8.3.) Margin of Safety Against Fatigue Failure The local stress arising at the tooth root (if the notch effect is taken into account) must be less than the permissible stress. So the condition arises:
SF
=
a FE YX/(aFO KAKvKF~KFcJ 2::: SF min-
(52)
Here O"FE = O"Flim' 2,0; O"Flim is the nominal flexural fatigue strength of the standard reference test gear with stress correction factor (~ notch configuration number) = 2.0; guide data for a FE as per test bench experiments (see Table 14).
Yx is the size factor for tooth root strength; see Fig. aFO
2~.
is the nominal value of the basic stress: (53)
Yes, the tip factor, determines tooth form, including notch form, when force is applied to the tip. The basic profile is as per DIN 867; see Fig. 24. Yt;, the contact ratio factor, determines the conversion
I.il,
Mechanical Machine Components • 8 Gearing
4.2
0.7
!
I
4.1f-----t~--+-----'--
X=~B--
i
J--
t'1 I
i:
-
-
-.
PIe
-0,5 -0.2 +++-+t+tI----t--I
Iii
4,O'-----~-~_-"--_'_~_'___'___~_LLLL___'___'___U_'_L.LLL"__'__"
10
7
II
12 Z
14
or
16 18 20
30 4050
when a force is applied at the specific external point of application (for helical gears. Eq. (34) applies for the normal replacement gear). Y~ is the helix factor. Ye
= 0.25 + 0.75/£om' Y~
100 400
Znx
= 1 - f3 "/1202: 0.75.
('\4)
For large root rounding-off values, the stress concentration index must be taken into account (DIN 3990) [ 1]. Note the influence of greater roughness, grinding notches, shotblasting, grinding out notches [18].
Margin of Safety Against Hot Seizing and Cold Seizing Subsequent remedial measures are often possible (see F8.5.1) [1]. For calculation, see [11 and DIN 3990, ISO/DIS 6336.
Margin of Safety Against Sliding Abrasion This is necessary for speeds below 0.5 m/s. According to [19]' increased wear should be expected if theoretical minimum lubricating film thickness (Eq. (37» is less than 0.1 ,."m (maximum wear occurs at approximately 0.0 I to 0.02,."m). For remedial measures, see F8.5.1. On calculation, see [1].
Figure 24. Tip factor (DIN 3990, ISO/DIS 6336). Y" (= YFa • Y~a) for basic profile: an = 20 0, halmn = 1, haolmn = 1.2S,Paolmn = 0.25; for rack, YFS = 4.62; for internal spur gears with PF = Pau/2: YFS = 5.79.
Geometry:
Gear 2
Gear I
o o
Normal pressure angle, an Real pitch module, mn U c.entre distance, a [j Tooth width, /J o Number of teeth. Z - Gear ratio. u o Helix angle, !3 o Profile offset factor, x lJ Reference diameter, d. Eq. (5) o Root number. dj, Eq. (28) [j Tip diameter. d" Eq. (29) where hIP = 1.2'5 m;
o o o [J
o o o o
20 3S 180 53
Unit
mm mm mm
63
36 1.7S 12 0.5 128.815 1235 139.0
0.3686225.426 mm 219.2 mm mm 23·1.7
c= O.25m
Real pressure angle, an (see F8. 1. '5. pressure angle) 120.728 Base circle, db, Eq. (20) Contact pitch, Pet> Eq. (21) Effective pressure angle awn Eq.
20.4103 211.274 mm mm 10.535 22.7462
(31)
Transverse path of contact, g". Eq. (23) Transverse contact ratio. Bo> Eq. (24) Overlap ratio, ".' Eq. (13) Total contact ratio, By, Eq. (14)
IS.9
mm
I.S1 1.00 2.51
For the cakuJation of timing gears, gears with load peaks that rarely occur on or with unified loads, see [L 191.
Calculation of Load-Bearing Capacity Example Calculation of load-bearing capacity of first spur gear stage of a stirring machine. Drive: electric motor. [J = value on diagram.
Given. Motor speed n] = looO min I; power PI = '51 kW; quieter running required (see also Table 3). Centre distance (l preset [J Gear grade 6 as per DIN 3962 (see also Table ~): FH " = 10 fLm. n Basic profile as per DIN 867, an = 20 ° (Fig. 10). n Gear material: pinion 16 Mner S (Table 14, no. 30), gear 42 erMo 4 Y (Table 14, no. IS) [J Hardness: pinion 60 HRC, gear 300 HB LJ Flank machining (roughness): ground, R3 = 0.5 V-m (corresponding to R2 """ .3 V-m) o Roughness at tooth root: Ra:S: 2 V-m (corresponding to K~ 12fLm.).
Circumferential force. Eq. (41), F, K'factor, Eq. (40)
=
7561 N.
= 1.74 as per Table 7, adequately dimensioned.
Circumferential speed, Eq. (42):
VI =
6.7 m/s.
Lubricating oil viscosity at 40°C, Fig. 18: ISO·YG 220.
V'IO
= 1..3.10 2 mm 2 /s,
Force Factors Application/actor: KA Dynamic factor'
= 1.3
used (see aJso Table 11).
K, = 1.08
(l\ ,zl/lOO)· ru 2/(1
+
as
per
U2)Jl/l = 2.1
Fig.
lib,
where
'4.
B.6 Bevel Gears. 8.6.1 Straight Bevel Gears
Width factor, Initial
KH/'l ( .-::":
Ktl .):
characteristic
value-
as
Table
per
12
for
= 750 N/mmol /inclusive , going to Xli = 0 ..':;510 .85, J~>r" = fU (J = to J..Lm (gear grade 6, see above), flank line devialion UUlim
through cumulative deformation: f~hlJ. p., ~ 12.6 ~11l .
=
8 J..Lrn as per Table 13.
With Eq. (43):
Face factor . Kilo and KF<>: helical gear, DIN gradl'
~
7, Eq. (46):
K Hn = KF" = I,
Margin of Safety Against Pitting Zone factor , Eq. (49) with {3h as per Eq. (3S), at' aWl: ZII Elasticity factor . for St/Sr: Zr
=
(THO ::::
value of 466 N/rnm2
flank
Geat Dimensions (Fig. 25). Dimensions on external reference cone (back cone): index e. The tooth form (on the back cone, RB) is approximately the same as that of a spur gear with the radii r\' 1 and rvl. on the generatrices of the back cone. Axial angle I
= 0.8 .
compression
= 0, +
0, usually I
=
900.
Reference cone angle, 0, from tan Ii, = sin l/(u + cos I) ,
190 -JN/mm
l
Contact and helL-,;factors (Eq. (SO» : Z •. Z(3
Nominal
= 2..1.
cutting teeth of octoid form r31, and therefore to proftle offset only for gear pairs with reference centre distances (see FB. I .7). Moreover, reinforcement of the pinion at the cost of the gear by means of tooth thickness alteration is possible (lateral proftle offset) , as well as differing flank angles on the front and rear flanks or tooth height alter· ation with identical tools (s~parate planing tools for the two flanks'). Tooth height generally diminishing towards apex of cone [50 I.
(Eq
(48» :
Size factor (Fig. 23): /..,, = 1
(56)
= u.
(57)
External cone distance R. = 0.5 d./sin Ii:
(5B)
For I = 90°: tan Ii,
=
I/u,
tan Ii,
External reference diameter
strength (Table 14) , fixed , for pinion 1500 N/mml , for .300 HB gear all I i", = -:':;0 N/mml .
Pitting fatigue (TUhn,
=
iWatingofmaterialfa("/or(gear) (Eq. (51»: Zw'= 1.1. Safety factor for pining (Eq. (47» : pinion SH As per Table IS sufficient
=
(60)
with modulus on back cone, me' Gear ratio (61 )
1.1. gear .511 = I.l-
for I = 90°: see Eq. (57). Tip diameter
Margin of Safety Against Fatigue Fracture Tip /aclor (Fig. 24): Y,." = 4;\2 . Y", = 4.;\5 (with Eq. (34»: Znl :::: .38.3 , Inl = 67) Contact and beli-.:factors (Eq. (54» ' Y"Y IJ
(55)
(63)
normally: h"., = m. (1 + x h ): h .. , = m (1 - x h ).
= 0.67.
Nominal value Of basi<' stress (Eq. Cj3» : ami O"F02 = 158 N/mm 2 .
(62)
-=;
157 N/mml,
(64)
Dimensions on internal reference cone : index i instead of e.
Alternative Spur Gears, based on mean tooth width Basic' fatigue strength, fixed as per Table 14. for pinion O"t+ = 900 N/mm 2. for gear (TFf = 600 N/mml
Size/actor (Fig.
Z~):
Y,
~
decisive for load-bearing capacity calculation (Fig. 2;).
I.
Safety fa<:tor for fatigue failure (Eq. (52» : pinion SI'1 SFl = 2.2. As per Table IS sufficient .
=
3.4, Gear
8.6 Bevel Gears Greater efficiency as against worm gears, and more cost· effective at higher power levels (as straight bevel gear pair). More complicated as against spur gears (high offset, shaft angle deviations. axial position of gear and pinion, deflection with flying pinion). Counter·measures: Iimi· tation of tooth width, crown gearing, lapping and pairing pinion and gear together, axial setting of pinion, roller bearings (little bearing play) , rigid housing (see FB.6 .5). 8.6.1 Straight Bevel Gears Nornlal up to
lJ
= 6 m/s, ground at up to 50 m/s
(aircraft construction).
Tooth Form. Straight·sided basic crown gear, manufac· tured using planing tool with straight cutting - leads to
Figure :IS. Bevel gear pair and alternative spur gears for calculat, ing loadbearing c.-apability. I heel, 2 toe.
I.'t:.
Mechanical Machine Components. S Gearing
Table lei. Guide data for selection of pinion tooth number"", tooth width and profile offset factorl' in bevel gears u
z,
18 to 40 0.212 0
bId, Xh
Limiting values: blR,
S
1.12
1.25
1.6
2
2.5
18 to 38 0.226 0.03
17 to 36 0.240 0.06
16 to 34 0.284 0.12
15 to 30 0.336 0.18
13 to 26 0.404 0.24
0.3; bId,
S
0.7S; blm
S
4 12 to 23 0.474 0.28
\0 to 18 0.615 0.36
6 8 to 14 0.75 0.42
7 to 11 0.75 0.45
10. For helical and spiral-toothed bevel gears e.2: I.S.
aFar spiral·toothed, hardened bevel gears, select Z I more on lower limit, for spur-toothed, unhardened bevel gears more on upper limit. bFor spur-toothed bevel bears with staggered O-toothing (Xhl = - Xh2) and nonnal tooth height (bgP = bfp = m, Fig. 10). Profile offset for helical or spiral-toothed gears about 85% of these values.
d m,
= del - b sin Il"
Ilm2
= udml ,
(65)
for
for
(68)
= Zvl
(69)
dvm2/Zv2.
= Zl '>/(u 2 + 1)/u 2 , Zv2 = Zvl • u 2 .
(70)
Recoounendations for selection of number of teeth, modulus, tooth width, prome offset; see Table Hi; backlash: Table 17. Basic profile: see Fig. 10, ISO 677. Load-Bearing Capacity for alternative spur gears to be deter· mined as per Eqs (65) to (70), using F, ~ 2M,/dm ,. Guide data for K(ja = (KH~Fa) as per Eqs (47) and (52), owing to greater uncertainties for bevel gears and the restrk."ted contact pattern (crown gear): Kf1a = 2.0 when pinion and gear supported on both sides;
K(3a sides;
=
2.2 with floating pinion and ring gear supported on both
Kf1a = 2.5 with floating supported pinion and ring gear.
Check: in no operating condition should contact pattern lie at end of tooth (see F8.6.5).
BeariAg Forces. See FS.6.4. 8.6.2 Helical and Spiral Bevel Gears Low-noise running; milled or planed and lapped up to v = 40 m/s; ground at up to 80 m/s (maximum 130 m/s); pay attention to axial forces! The relationships in (FS.6.1) apply to the gear geometry for the transverse values of the bevel gears and alternative spur gears, Le. m = m, = mn/cos {3.
Table 17. Nonnal flank play values for bevel gears and worm gears
Module m
Flank play
Up to 1.6
Above 1.6 Up to 5
Above S Up to 16
(0.08 to
(O.OS to O.03)m
(0.04 to
(0.03 to
O.03)m
O.02)m
O.04)m
Ahove 16
Figure 216. Crown wheel gear with axial offset, Taschenbuch fur Maschinenbau, 13th edition.
0:,
as per Dubbel,
Helical Gears. Tooth height dimensions such as b'e> b'e> are given dependent on the real pitcb module, but the profile offset is frequently given in multiples of the transverse module. For selection of number of teeth, profile offset factor, X h and remaining gear variables, see Table
16.
Spiral Bevel Gears. Helix angle variabie over Width, flank line course, tooth heights, helix angle largely conditioned by manufacturing process (see K5.2). So layout and calculation as per machine manufacturer's specifications. 8.6.3 Special Gears
Crown Gears (Fig. 26). Pinion is spur wheel or helical gear, crown gear is generated by shaping, using piniontype cutter; pinion offset also possible. Load-bearing capacity low [2].
Bevelled Spur Gears (Fig. 27). Spur wheels or helical gears with profile offset variable over width. As per Fig. 27a, suitable for setting for contact free from back-
Gear I
, Gear
a
b
Figure 217. Tapered spur gears: a as spur gearing (parallel axes); b as bevel gear pair (axial angle I).
'4i!!1
S.S Wonn Gears
Table 18. Calculation of tooth force components on bevel gear Spiral direction and direction of rotation h of Axial force driving gear
Radial force
Right-hand spiral, rotating to right
Driving gear
Driving gear F" =
or Left-hand spiral rotating to left
cos
sin
f3 cos 8)
~ (tan
an
sin 8 - sin
f3 cos 8)
Driving gear F .. = co~
Drin~n
F"
=
Fr =
~ (tan an cos £, cos
f3
sin
f3 sin
(5)
Driven gear
cos f3
F
or Left-hand spiral, rotating to right
f3
Driven gear Fx =
Right-hand spiral, rotating to left
~ (tan an sin 8 +
p
Fr =
cost f3 (tan an cos 8
+ sin f3
sin 8)
+ sin f3
sin 8)
Driving gear
f3 (tan
an
sin 8 - sin
f3 cos 8)
gear F
F
f'.. = cos!
f3 (tan an cos
8
Driven gear
cost f3 (tan an sin B + sin
f3 cos 8)
F.. =
~ (tan an cos 8
cos f3
-
sin
f3 sin
8)
aUse values of angles f3, a and if for which the stresses are specified. hSpir.tl direction and direction of rotation seen from bevel tip outwards.
lash, as per Fig. 27b, for small pitch angles that cannot be set on bevel gear cutting machines [21- 23].
Hypoid Bevel Gears. Bevel gears with intersecting axes (Fig. 1). Execution with spiral toothing throughout, as per machine manufacturer's specifications [25-2SI. 8.6.4 Bearing Loads Calculation of force components as per Table 18 and Fig. 28. Attention should be paid to ovenurning moments of axial forces in calculating radial bearing loads.
8.6.5 Design Hints for Bevel Gears For pinions mounted on shaft: gear rim thickness under toe 2: 2 m (take groove into account if applicable). Bearing interval as per Fig. 28: I, = (1.2 to 2)d" where u = 1 to 2; I, = (2 to 2.5)d, where u = 3 to 6; one bearing as close as possible to pinion tip; 12 > 0.7d,. Contact pattern under full load approximately 0.S5b (tooth ends free) with high-precision teeth and housing and rigid construction, otherwise smaller (approximately 0.7b). Select direction of skew in such a way that axial force pushes pinion away from contact (guarding against face backlash). Mounting must permit axial shifting of pinion and gear (setting of contact pattern and face backlash).
~'
Tooth widths of pinion and gear as similar as possible (approach edges!). Ensure oil feed to rear pinion bearing. Hubs and gear-cutting tolerances DIN 3965, gear specifications in drawings DIN 3966.
8.7 Crossed Helical Gears Characteristics (see FS, introduction), application: tacho drives, small units, textile machines, centrifuges and similar [1,43-49].
8.8 Worm Gears Cbaracteristics (see FS, introduction): standard transmission in a stage 5 to 70 into low, 5 to 15 into high. Automatic locking with driving gear (i.e. 1)' ,-c; 0) conditions efficiency, 1) < 50% with driving worm' Any alteration in worm requires alterations in tool (for gear pairing, see F8.1.4). Main application (economic efficiency) up to centre distance of a = 160 mm, n up to 3000 min -', practicable up to a = 2 m and 1000 kW power. Low-play duplex worms for dividing gear [29]. Types of Pairing (Fig. 29). On most customary cylindrical worm gear (Fig. 29a). For double enveloping wonn gear pair, see [29]; cone drive worm gear see [30].
o
r,
Fe
Figure 2:8. Tooth force components for calculating bearing forces.
a
b
c
Figure 29. Types of worm gear pairs: a cylindrical worm gear (cylinder worm - globoid worm wheel), b contrate worm gear (enveloping worm - spur gear); c double enveloping worm gear pair (enveloping worm - globoid worm wheel).
'M'le.
Mechanical Machine Components. B Gearing
Figure 30. Defining quantities of a cylindrical wonn gear.
Flank Form follows from manufacture (see K5.2). ZA,
ZN, ZK and Zl worms differ only slightly in efficiency and flank load-bearing capacity. ZC (hollow flank) worms are therefore rather more favourable , but more sensitive to load variations (worm deflection) . 8.8.1 Cylindrical Worm Gear Geometry For axial angle I = 90°: initial sizes are average worm diameter d m, and tooth profile in axial section (Fig. ~O). With other axial angles, the analogous relationships for cylindrical helical gears apply (see F8.7). Equations follow from the relationships between rack Proftie of worm (in axial section) and worm gear (sign: Z) or from considering worm as helical gear (sign: S) or as threaded spindle (sign: G).
T I
Main Dimensions and Gear Data Transmission: I = n,/n b
(71)
s (with driving worm = n,/n,).
1
Gear ratio: u
I
= Z1,/ZI
(with driving worm = 0. Centre distance: a = (dm , + d m ,)/2 = (dml + d, + 2xm)!2.
j
TPitch angle: S I Z
(B2)
1
tan Ym
= [(2a/d ml )
Sliding speed: v.
-
11z,/(z,
+
2x).
= 7rdm,n,/cos Ym'
For Zl worms, the relationships for involute helical gears also apply (see FB.!. 7) with f3m = 90° - Ym' Lines of Contact (C-lines) Contact points and tooth form of gear can be calculated or constructed from given axial section profile, A, of worm for given pitch circle (= graduated circle) of gear in accordance with basic requirement of gear tooth system (see F8. 1.1). The same applies for evety section, P, parallel to the wonn axial section. C-Unes are thus obtained; example,
(72)
see Fig. ~O . Since the tooth profile of the worm in section P differs from that in the axial section, here too there is another counter-profile. For construction, see [1], for calculation [32, 331.
(73)
8.8.2 Tooth Loads, Bearing Loads
Z Profile offset, x: Since a rack (= axial section of worm) Z is not altered by profile offset, only the worm gear can have a profile offset, x = x" as a result of which the pitch lines of the rack are displaced, while the pitch circle (= graduated circle) of the gear remains unaltered. For selection of profile offset, see F8.B.4. Axial pitch modulus:
Calculation of circumferential force F, from torque M and power P. For relationships see Figs ~O and ~l. Tooth
T
m = m xl = rna = Px/'f'( = P,I/(7rZ,) = d m, tan Ym/ZI'
G
..L
r~"= d ml
d'l
Z
d.,
=
2a - d mb
= dm , +
2m (1
For normal worm profile, 2m is the
Tip clearance usually c,
(76)
usual common tooth
+ x), height.
d 2 = Z2m = d m2 - 2xm , de = d a2 + m, d n = d m , - 2(m + c,), d n = d m 2 - 2(m + c,).
1
(74)
(75)
= dml + 2m ,
= c, =
(graduated circle = pitch circle) See note on Eqs (76) and (77). 0.2m .
(B3) (B4)
(77)
.--(.,
FllIIl~ - F.ml
(ml
(7B)
(79) (BO) (Bl) Figure ~1. Tooth forces on a wonn gear.
8.8 Worm Gears. 8.8.4 Rating and Evaluation of Load Capacity
Z = (7
loads of gears with offset promes are also specified for d m [I].
F,m' = F,m2 tan(Ym
+
p,) = -Fxm2'
(85)
To calculate the bearing loads, it is sufficient to insen p, as always being positive, with a value of between 3 and 50 (see F8.8.3). (86)
-Fxml •
f~m1. = Frml
=
Frm2
=
Ftm2
tan
(87)
(lx.
Bearing loads can be derived from these load components, radii and distance between bearings (Fig. ~I). In this context, attention should be paid to ovenurning moments: MK, = F,m#m,/2, MK2 = F,m,dmz/2.
(88)
Similarly, any external transverse forces on input or output shafts should be taken into account. 8.8.~
Effidency
For guide data, see Table 19. Preferences based on scatter ranges specified there: gear material Cu-Sn-bronze more favourable than grey, iron, AI-bronze, brass; hardened ground worm more favourable than hardened, milled worms; ZC worms more favourable than other tooth forms; high viscosity, suitable synthetic oils more favourable than low viscosity, mineral oils (attention should be paid to running-in characteristics); large pitch (multiple and thin worms) (attention should be paid to bending) more favourable than small pitch (single thick worms).
8.8.4 Rating and Evaluation of Load Capadty To begin with, all requirements and influences with regard to stresses and functions should be carefully clarified. Compare check list for spur gears (Table 6). Dimensions are detennined and the margins of safety, SH and SF and the worm deflection, ii, are checked, together, if applicable, at high speeds, with the margins of safety for temperature, ST, and wear, Sw, as per [I], and the values obtained are corrected if necessary.
Centre Distance ", T......mission i and Power p. Specified Select number oj teetb, z" according to experience [BS 721] (a in mm). Numerical value equation:
'4e.1
+ 2.4a')/u,
(89)
Round number of teeth, z" up or down to next whole number; then Z2 as per Eq. (72). Note. The Z2/Z, ratio, which is not a whole number, makes it easier to manufacture the gear with a fly cutter and reduces the harmful effect of pitch fluctuations. The running noise decreases as the number of gear teeth increases; Z2 2: 30 as far as possible with ax = 20 0 and nonnal tooth height. Selection oj Diameter/Centre Distance Ratio, dm,/a (Fig. ~~). Pay attention to tendencies of SH' ii and 1/. Thus, with a view to as high a degree of efficiency as possible, a low value for dm,/a is aimed at, but attention should be paid to deflection, owing to the risk of worm shaft breakage. Then dm' = a (dml/a) and tan Ym as per Eq. (82). Finally, a check should be made to detennine whether available tools (especially hobs) can be used. This usually also detennines the tooth form. Recommendation Jor Profile Offset Factor, x.
ZI worms: - 0.5 :s x:s + 0.5, preferably: x = 0; ZC worms: 0 :s x :s 1.0, preferably: x = 0.5. Additional Variables. m as per Eq. (74), d 2 in accordance with Eq. (78), dm' as per Eq. (76), am2 as per Eq. (77), d n as per Eq. (SO), dl2 as per Eq. (81), am2 in accordance with Eq. (75). Guide data for additional dimensions (see Fig. ~): de = d'2 + m, b, = 2m(Z2
+
1)',
(90) b 2 = 2m[1
Worm
+ (dm,/m +
1)'].
Interesting if hobs are available to cut the gear. It should also be noted that one worm (i.e. one hob as well) can be used for various transmissions and provides for different centre distances to this end. First detennme Z2 as per Eq. (72) and select x 2 , dm2 as per Eq. (78) and a as per Eq. (73). Then continue as described above.
Gear Moment M .. Speed .... Transmission I, Given Calculate centre distance, a, from Eq. (91) and the variables given there. Round up a to next highest value of range as per DIN 3976. Then continue as described above.
Table 19. Cumulative efficiency 1) in % of cylindrical worm gears (average value), roller bearings, standard ntineraJ oil (Italic figures: self-locking or possible self-locking) Worm speed
min-I
Transmission
10
20
40
70
15
77 to 82 68 to 73 60 to 65 47 to 52 30 to 35
150
84 to 89 76 to 81 72 to 77 58 to 63 43 to 48
1500
91 to % 88 to 92 81 to 87 75 to 80 64 to 69
Fi......, ~:I. am.!a ratio. Solid curves as per Eq. (83) for x = 0; dotted lines define field of industrially made worm gears; 1increases,
1 decreases.
11111'1
Mechanical Machine Components. 8 Gearing
Table ZO. Material characteristic values for worm gears
Standard
Worm gear material
Rm
Rpo.2mln
(N(mm')
(N(mm')
HB
8,. (%)
Elastic modulus
(N(mm') DIN 1705
DIN 1709
)
(THUrn
(].,irn
(N(mm')
140 150
260 280
80 95
12 5
88300 88300
147 147
265 425
115 190
G-CuSn 12 Ni GZ-CuSn 12 Ni
SK 12 Ni
160 180
280 300
90 100
14 8
98100 98100
152 152
310 520
140 225
G-CuSn 10 Zn GZ·CuSn 10 Zn
130 150
260 270
75 85
15 7
98100 98100
152 152
350 450
165 190
GZ-CuSn 14
200
300
115
4
92700
150
370
180
450 480
750 750
180 190
8 5
107900 107900
157 157
500 550
565 605
320 400
680
750
170 185
122600 122600
164 164
250 265
402 502
300
700
160
122600
164
660
377
120 500
300 790
250 260
98100 175000
152 182
350 490
150 628
SoMs
GZ·CuAl 10 Ni
W1A
GG-25d
GG-70e,f
13
5.5
b
(N(mm')
SK 12
G-CuAl 11 Nid.e GZ-CuAl II Nid.e
DIN 1691 DIN 1693
Z," ~N(mm'
G-CUSn 12 GZ-CuSn 12
G-CuZn 25 AI 5 GZ·CuZn 25 AI 5
DIN 1714
(
"Applies to case-hardened worms (ground, HRC 6Q.± 2); for heat-treated, unground worms: values for (THlirn multiplied by 0.75; for grey iron worms, multiply values for (THlirn by 0.5. bApplies to am = 20 0 , for an = 25 0 multiply values by 1.2, for alternating stress multiply values by 0.7. CFor steel worms. For grey iron worms: ZE = [2.86 (E. + E) (E,E)r i with E, for grey iron, E as per table. dDriven only with mineral oil, eFor low sli~ng speeds (manual operation). fpearlitic.
Calculation of Margin of Safety Against Pitting, SB
Numerical value equation: SH
=
"Hllm
ZhZn/(Z.Zp -VlOOOM,K,la' 2:
SHHm
(91)
(for units, see tables and M, in N m). For "HHm, rolling strength, and ZE' elasticity factor, see Table ZOo Service life factor, Zh = (25 OOO( Lh) 1/6 :S 1.6 with Lh in h. Load cycle factor, Zn = [1/(n,/8 + I)J 1/8; contact factor, Zp: for guide data, see Fig. ~~. Application factor, K A : see Table 11. SH min> depending on reliability of specifications and consequences in case of damage, = 1 to 1.3.
15
3.2
8
""-
~
""""
b<..ZIWorm
ZCWorr;>-
4
2.0 0.2
----- r--
04
OJ
Figure 3030. Contact
SF
= Ulim
mb,l (F"KA )
2:
SFmin;
(92)
for U lim , limiting value of U factor, see Table ZO; for K" application factor, see Table 11; SF"';n = SHmin.
Chec:kiag Worm Deflection The deflection, IJ, of the worm must be limited in order to avoid problems with contact (for failure to meet basic requirement of gear tooth system, see FRI.1) and increased contact pattern dislocation.
-
0.5
Diameter/centre distance ratio dm/a factor
Zp
for
from Fnnl as per Eq. (85), F=I as per Eq. (87) and MKI as per Eq. (88). limiting value, IJlim = ml100 to m/250, depending on individual initial capability of material pairing and requirement for efficiency.
ZI
8.8.S Embodiment Design, Materials, Bearings, Accuracy, Lubrication, Assembly
0.6 wonns
20', - 1 '" x '" 0.5) (also approximately, for ZK, ZA and ZN worms) and ZC worms (a ~ 24',0.3'" x '" 1.2) [1, 331. (ao ~
Calculation of Margia of Safety Against Tooth Breakage, SF
Assumptions. Shafts with diameter d ml on two supports with intervals as per Fig. ~ 1 under load from net force
'",,-
,
Calculation with variable load and speed, short·time operation [1 J. Wear load-bearing capacity [1, 34].
Embodiment Design for housings, shafts, seals: see F8.9. For example, see Fig. ~. Position of worm for splash lubrication, as low as possible, with VI < 10 mis, also at side, for v, < 5 m/s also above; for injection lubrication, any position.
8.9 Epicyclic Gear Arrangemems • 8.9.1 Kinematic Fundamenta.ls, Tenninology
/
R,ght·handed WOlm
Rotaling to rig
Len·handed worm
7
6
~
ROlating to left
••eg
~
Figure34. Worm gears (Flender, Bocholt). Rated power, 24.5 kW, III =- 1'500 min I, i = 20. I ZC worm, 16 MnCr S, case-hardened, ground; 2 gear rim, GZ - CuSn 12 Ni; 3 hub St 37; 4 housing erG 20 with horizontal fins: 5 ventilator; 6 oil drain; 7 inspection cover with ventilation; 8 radial scaling ring (sealing inwards); variable sealing of worm shaft shown; 9 additional sealing rings; 10 cemrifugal disc ; J 1 oil return (shown offset); 12 separable ball bearing (for light duty) ; J3 t;lpereu roller hearing (for heavy duty); 14 shim rings for axial adjustment of gear. Dimensions in mm.
Optimum Worm made of carburising steel (58 to 62 HRC) or heat·treatable alloy steel. carburised surface hard· ened (HRC < 56), at < 3 mls also unhardened. For power gears usually as right·hand solid wonn, Fig. ~. For cost-effective gear under low load, also slip·on hollow worm. Bearing interval as small as possible (deflection'):
"0
1 = (1.3 to J.;)a
(93)
Worm Gear Rim for power gears, bronze moulded by centrifugal action (GZ·CuSN 12 or GZ·CuSn 12 Ni) most suitable, having initial capability and with little tendency to be corroded. AI·bronze, special brass for low gliding speeds only (danger of seizing, increased sliding abrasion). Grey iron for low speeds only (manual operation), relatively favourable for pairings with grey iron worms. Gear Rim usually screwed to hub using dowel screws; for press-on or cast gear rims, see [I J.
Bearing Intemal of gear shaft not too small (tipping risk!): 12 = '" + II) = (0.5 to 0.7)d, (see Fig. ~1). Mounting to be on roller bearings throughout , sliding bearings for low running noise only (e.g. for lifts).
Worm Shaft. For small and medium·sized dimensions, mountings with separable or angular ball bearings or tap· ered roller bearings, range 313. For large dimensions, fast and loose mountings. for example with double·row angu· lar ball bearings.
Gear Shaft. Grooved ball bearings, range 63, or tapered roller bearings, range 302 , 322.
Accuracy, Flank Backlash. As grades not standard· ised to date, guidelines as per DIN 3961 to 3964, individ· ual and cumulative deviations: grade 4 to 5 for precise dividing gears, sighting mechanisms and similar: grade 5 to 6 for lifts and smooth-nmning gears with v, < -; m/s; grade 8 to 9 for normal industrial sectors; grade 10 to 12 for auxiliary drives , manual drives and similar with
v, < 3 m/s.
Set contact pattern on outgoing side (lubrication wedge').
Running·in (with nominal moment , low speed , thin oil) increases efficiency and flank load·bearing capacity, but is economically possible only in special cases.
Flank Backlash is approximately as per Table 17. For low·play gears see [281 .
Lubrication. For a guide to the selection of oil viscosity and type of lubrication, see Fig. 18. Grease lubrication only at VI < 0 .8 m/s or for intermittent service (heat removal), abrasion in grease , attention should be paid to difficult grease changes' Mineral oils with mild extreme pressure additives simplify rurming-in. SynthetiC oils make low coefficients of friction possible, i.e. high efficiency and high linliting heat output; nmning-in behaviour usually less favourdble. Oil change after running·in, then after approximately 3()()O hours, then approximately once a year [35J .
8.9 Epicyclic Gear Arrangements H. W. Miiller, Darmstadt
8.9.1 Kinematic Fundamentals, Terminology Epicyclic gear arrangements differ essentially in only one aspect from the usual simple transmission gearing. While with trdnsmission gearing the housing, together with the gears housed within. is rigidly connected to a base, with epicyclic gear arrangements it is mounted on the base so that it is able to rotate, and is provided with an additional shaft (Fig. ~S). SO, from positive transmission gearing with a degree of operation F = I, there emerges a differential ur superimposed gear arrangement with a degree of operation F = 2. (The degree of operation of a gear lays down how many movements of any type can and should be assumed for it in order to determine its state of motion clearly.) Here the original hOllsing is shrunk onto
1.lel
Mechanical Machine Components. 8 Gearing
~ :J fi 'f
a Figure 3S.. Development of an epicyc1k gear from a transmission gear.
a support that carries only the gear mountings. Protection and oil tightness are ensured by a new housing. Kinematically, however, trus is a part of the base. The torque of the new support shaft, s, is identical with the supporting moment of the original gear housing. In this way, epicyclic gear arrangements can be developed from toothed-wheel and friction gears (epiLyclic gears), hydrostatic gears, traction mechanism gears, linkages having only turning and sliding pairs and other gears [58]. Epicyclic gears (for the most frequent fottnats, see Fig. ~6) are also described as ''Planet gears" and their gearwheels, with circulating axes, as ''Planet gearwheels" or ''Planets''. If the new support shaft, s, is momentarily or pettnanently secured or designed to be stationary, then the epicyclic gear arrangement reverts to "stationary transmission" with the "stationary gear ratio", i'2' using any of its "stationary transmission shafts", numbered 1 and 2 and the "stationary efficiency", 1) in both possible directions of the power flow (pfl) during the exchange from drive to power takeoff:
= ("-')
Stationary gear ratio, i 12 Stationary efficiency eoff on 2. Stationary efficiency takeoff on 1.
1)"
n2
(n~.o.
b
41iWI
1 c
d
er:f lW
.
P,
0)
for drive on shaft 2, power
The indices 1, 2 and s each correspond to the shafts so described with their gears or the support. For speed ratios or speed transmissions, the sequence of indices indicates: first index numerator, second index denominator: for efficiency values; first index drive shaft, second index power takeoff shaft. Planet gears are described using p and the index of the gear that they mesh with in each case - Fig. ~6. Trus individual indexing of the epicyclic gear shafts using 1, 2 and s simplifies calculation and makes it easier, and makes it possible, for example, to analyse the operating behaviour of all models of epicyclic gear arrangements using a single simple computer program [59,60]. In a stationary transmission gear, the power is transferred through the mesh exclusively as "rolling power", Pw' during the meshing of the gears with their "rolling speeds", n w ' and n w2 ' While trus is happening, the tooth friction power loss P" disappears as lost heat. Should only the support be set in motion, at the speed n m , with the transmission gear stationary, the entire gear arrangement rotates, including the two stationary transmission shafts, at trus speed, like a coupling, with no internal relative motion. This can involve the transmission of "coupling power", P.. without any loss, at the "coupling speed", n,. Should a coupling speed, n" be superimposed on an operating transmission gear, then the typical operating condition arises of a planet gear with three running shafts with speeds of n" n, = n wl + n, and
• P, PI
i
for drive on shaft I, power tak-
1121
.
f
g
I 2 lJ.,.U
:J
2
-1.2 ... - 11.3
A:
i11~
B:'I
i'1~ Z1 1z,
[;
0:
A: B:'I
c:
~'1'" ~1' '" 0.985 (lz,I+lz,lllq~g
i:l~
-054 ... -1...-53
i'l = (Zllzpl) (Zp, Iz,) ~'1"'1)11 '" 0.985
0:
(lzp,z,1 +Iz, zp,l)lql ~ g
A:
i'l~
B:
i'1~ - zl lz,
c:
1)'1'"
-1
~1'
0.98
'"
0:
(lz,1 + Iz,l)lq ~g
A:
i11~
B:
i'1~ (z, Izp,) (zp,lz,)
c:
l...41 0.98
1)'1'" 1)1' '"
0:
IIzp,z,l-lz, zp,I)/ql ~ g
A:
i'l ~ 1... 2.7
B:,I
c:
i'l ~ (Z/Zpl) (zp';z,) 1)11'" 1J1, '" 0.99
0:
(Iz p, z,l-lz, zp,I)/ql
g
1.2 ... 17. 6
A:
;'1~
B:'I
;'1=- z,lz,
c:
1),1'"1]11 '"
0:
~
I
0.975
IIz,I-lz,IJlq=g
I=p
A:
ill = -0.2...-176
1I+-r
B:
ill =-z,lz,
c:
1)11 "'~11 "'
0:
Z, . Zl variable
0.99
., tooth numbers of hollow gears are negative, see DIN 3960 Figure 3>6. Most frequent types of planet gear: a-c minus gears, d-fplusgears, gopen planet gear. Z = tooth numbers. A = possible range of stationary gear ratio for q = 3 planets/planet sets on periphery, approximately equal tooth root stress for all gears, Zmin = 17, zmax = 300. B = stationary gear ratio. C = 7112 = 'TJZl with 71wa = 0.99 for a spur gear stage, 11wi = 0.995 for a hollow gear stage. D = tooth number conditions for uniform arrangement of q planets/planet sets on periphery with ± g = whole number, t = largest common divisor of Zpl and Zpl of a stage planet.
n 2 = n w2 + n,. Here the rolling power, Pw , wruch can be transferred only between gear shafts 1 and 2, and the coupling power, Pk , transferred without loss between all three shafts, also overlap at the same time. Vice versa, the rolling speeds of a gear operating with three shafts can be derived as n w ' = n, - n, and n w2 = n 2 - n s , as also can the stationary gear ratio (94)
Rearranged, trus haslc speed equation, valid for all types of epicyclic gear, can be simplified to give
8.9 Epicyclic Gear Arrangements. 8.9.3 Sign Conventions
While the gear ratio, i 12 , of a positive transmission gear is unchangingly ftxed by its geometrical data, e.g. the gear diameter, any two speeds can be preset for a three-shaft epicyclic gear and can determine its state of motion. The speed ratios arising as a result of such speeds can no longer be described in terms of format-dependent "gear ratio", i, but are called ':tree speed ratios", k. Particular attention should be paid to this difference, because both variables can appear in one equation. Thus. for example, for any preset free speeds, n, and n" Eq. (9S) gives:
However. should one of the three shafts be fixed, e.g. = 0, or n, = 0, then the gear arrangement becomes positive again, and Eq. (9S) gives the "epicyclic gear
n,
ratios" i" = 1 - i",
i"
= I - I Ii"
(96)
as well as their reciprocal values, which describe the rotation of the support and the planet gears. The shaft not referred to in the index of a specifiC gear ratio, i, is not moving.
power values and efficiency values are also valid for any compound gear trains, provided these have a basic ratio of F = 2, with three external connecting shafts, a, band c, and provided their speeds and torque values are not reciprocally dependent on one another, as sometimes happens with hydrodynamic converters. It is of no importance here which three members or shafts, out of the many elements present in the gear arrangement, are selected as external connecting shafts. For gear arrangements with non-uniform transmission, e.g. linkages having only turning and sliding pairs, the equations are valid, in each case, for only one relative arrangement of the elements in their positive kinematic chain [61] and the associated instantaneous gear ratio between two of the three connecting shafts. The complete interchangeability of the indices is the basis for a principle that is useful in the building up of gears:
Should any stationary or epicyclic gear ratio of an epicyclic gear coincide with any stationary or epicyclic gear ratio of another epicyclic gear, then the two gears are kinematically at the same level, i.e. both have the same six gear ratios. but as a rule they differ in effiCiency. For an example of kinematically equivalent gears, see Fig. 37.
8.9.2 Generalisation of Calculations The two stationary transmission shafts and the support shaft of an epicyclic gear are kinematically at the same level. Thus, Eq. (9S) can also be written in a general form [58]:
where a, band c can be replaced by I, 2 or s in any arrangement. From this follows Table 21, showing the direct calculation of any free speed ratio, k, or Ilk, assuming that any stationary gear ratio or epicyclic gear ratio, i, and any free speed ratio, k, or Ilk, of the gear arrangement are known. But it also follows, as a wider consequence of this, that the equations for all operating data, and thus also for torque values, power values and efficiency values, remain valid if the indices of the shafts are exchanged in any way, provided that it is the same change for alJ equations.
Thus. the equations given below for simple epicyclic gear arrangements for the calculation of torque values,
Table 11. Generally valid conversion of free speed ratios. k or 11k, of gear with a known stationary or epicyclic gear ratio, iarr For a and b, use indices of known gear ratio. for c index of remaining shaft DeSired
I.e..,
8.9.3 Sign Conventions The following sign conventions apply in the analysis and building up of epicyclic gears: Speeds. All speeds of parallel shafts with the same direction of rotation have the same signs. The positive direction of rotation (n > 0) can be attributed to any shaft. Speeds with the opposite direction of rotation are then negative. It follows from this that
gear ratios i and free speed ratios k are positive if Shafts are running in the same direction (i, k > 0), and negative if shafts are running in opposite directions (I, k
< 0).
The direction of rotation of required speeds is then determined in accordance with the same rule from their signs, obtained as per Eg. (95), Eq. (97) or Table 21. Torque Values. A torque value is positive (M > 0) if the torque is acting on the gear in the direction of rotation defined as positive; if it is acting in the opposite direction, it is negative (M < 0). Power Values. It follows from the above definitions that driving power fed to a gear is always positive
s -HtH-s -Hffi
Keys
In relation to free speed ratio
1 2
1 2
Ge neral Example
+f* 1 s 2
~ ~1,: :{ffi4 ~ 7
k ch ( I - iah) I i ah
kac . ia~~ ka~ - I -+ tab
iOb {bO
k",.
~
1 - iah kah
k"
~
1 -
-
k" + iah
l/k('h
(h
tah
1(k a,
iabkha
I - iab
I
-
-
I
ioe
3
=
[iJ
is1
113 - 2
=
'21
i 15
I:
;'5
;52
+ +
;:: I ~ ;~~ I"
+
312
=
=
f~l
iZs
i,s i51
[iJ i21
i2s
[iJ
is2
f~Z
iZ1
izs
lab(1 - k",)
Figure ~7. Example of three kinematically eqUivalent planet gcars
'M'g
Mechanical Machine Components. 8 Gearing
(P,n = 2'lTMd ,11", > 0), because a drive shaft always assumes the direction of rotation in the sense of rotation of the driving torque. Power takeoff values, in contrast, are negative (P,b < 0), because the external power takeoff moment acting as a brake on the gear is operating in the opposite direction to the power takeoff sense of rotation. Power losses are negative, as they constitute a power takeoff (Pv < 0).
8.9.4 Torques, Powers, Efficiencies Torques. The ratio of the torques is determined through the stationary gear ratio ilZ and the stationary efficiency, 1/" and 1/". It does not alter if any coupling speeds n, are superimposed (without loss) on an operating transmission gear. From the equilibrium conditions there follows the moment equilibrium MI
+ M, + M,
= O.
(98)
For the transmission gear, it follows from the power balance that, in the cwo power flow directions: Drive on 1: M 2 n 2 = - M 1n 1 1J1l' Drive on 2: Mznz=- M,n,/1/zl' If the cwo efficiencies are combined in the expression 1/';;', the torque ratios can be formulated, independently of the power flow: (99) Equations (98) and (99) give Ms _
wI
M) - t121Jo - 1,
(100)
1.
(101)
Since the power balance of the transmission gear is identical to the balance of the rolling power if the support is circulating, these equations are also valid for epicyclic gears. Here the exponent wi follows from the sign of the rolling power P wI of the shaft 1: If Pw, > 0, the rolling power flows from shaft 1 to shaft 2, if P w, < 0, from 2 to 1. From that follows the definition of 1/';;' for the calculation:
M~
1st Pw, = >O: wi = + 1 ~ 1/;;" = 1/" 2 (n, - n,) ''IT { <0: wi = -1 ~ 1/;;'1 = 1/1/21 (102)
where ~ is the preset torque, or is calculated, if Mz or M, is preset, from Eq. (99) or (100), using 1/';;' = 1. Equations (99) and (100) indicate that the ratios of the three shaft moments to one another are determined only by the stationary gear ratio, i lZ , and the stationary efficiencies, 1/w~, and are thus constant for either of the two rolling power flows.
still free, results from the characteristics M = I( n) of the three machines [58 J. Should this fail to bring about a stable condition, the equipment begins to race or stops. Should one of the torques M equal zero (e.g. the machine be disconnected), then the remaining moments also equal zero, as per Eq. (103), the gear runs idle, and power transmission is not possible. For a list of torque equations see Table 22. In accordance with Eq. (98), one of the three shaft moments must have the opposite sign to the other cwo and must be equal to the sum of the other cwo. This shaft is called the cumulative shaft, the other cwo being the differential shafts. For epicyclic gears with negative stationary gear ratios (minus gears), the support shaft is always the cumulative shaft, and with positive stationary gear ratios (Plus gears) it is the slower-running stationary transmission shaft. If the cumulative shaft is stopped, negative transmission always occurs on the two differential shafts running, owing to their torques being in the same direction, and positive transmission occurs if a differential shaft is stopped. Thus any simple epicyclic gear can generate two reciprocal negative transmissions and four reciprocal positive transmissions in pairs.
Powers. With M in N m (kN m) and n in s-', the shaft outputs and the power loss, P" become: PI = M,n,2'lT W (kW),
(104)
P, = M zn,2'lT W (kW),
(105)
P, = M,n,2'lT W (kW),
(106)
P, = M,(n, - n,) 2'lT(I - 1];;") W (kW).
(107)
A characteristic feature of the epicyclic gear is that the shaft outputs, P, and P z, are the sums (superpositions) of the rolling and coupling powers. With w = 2'lTn, the
following equations are obtained:
Shaft output
=
rolling power + coupling power
P z = Pwz
+ PkZ
= M, (w, -
w,)
+ M,w"
Ps = Pks = Mswso Depending on the speeds selected, the rolling and coupling powers can have the same sign or opposing signs, i.e. have power flows which move in the same direction or in opposite directions. The shaft outputs, P, and P z , can thus constitute the sum of these cwo partial outputs or the difference becween them. In the first case, the dissipative rolling power remains smaller than the shaft output, and the total efficiency becomes greater than the stationary efficiency. But with partial power flows going in opposite directions, the rolling power can be as much greater than the shaft power as is desired; the total efficiency then becomes correspondingly lower than the stationary efficiency. It can even become negative, and thus lead to the automatic locking of the gear - see F8.9.5. This con-
(103) This equation, characteristic for differential gears, is valid irrespective of the individual speeds involved, even when one shaft is stationary. If power is transferred becween three machines through the three shafts of an epicyclic gear, then Eq. (95) must be fulfilled for the speeds, and Eqs (98) and (103) for the torque values. This arrangement leads to an operating condition in which the reciprocal coordination of speeds and torques, though
Table 2:2:. Fonnulae for torque values With wl=+ 1: 11'0 1 ::;; TJl2orwl=- 1: 11'0 1 = 1/1121> wI from Table 2:3 for transmission gears, from Table 2:4 for superposition gears, or from Eq. (102)
1.lti
8.9 Epicyclic Gear Arrangements. 8.9.5 Self Locking and Partial Locking
sideration of the partial outputs gives an insight into the operating behaviour of a simple planet gear, but it is not required for the calculation of the operating data. By the superposition of any rolling and coupling powers in any planet gear, any of the six possible power flows can be generated: three each with either 1, 2 or s as the only drive shaft and with two power takeoff shafts (power division), or with 1, 2 or s as the only power takeoff shaft and with two drive shafts (power totalisation). Which of these is to be the only drive or power takeoff shaft (total output shaft, TOS) is decided by the stationary gear ratio, i 12, and a completely free choice of speed ratio, k (Table 24). If the TOS of a superposition gear is preset by a layout having a single drive shaft (motor) and two power takeoff shafts (machines), then the speed ratios, k,lo k" and k,,, must lie within the range given for them in Table 24. If the speeds are preset for a superposition gear with two motors and one machine connected, and if in addition the power takeoff shaft is also the TOS, then this is a power totalisation system. However, if one of the two motors is connected to the TOS, then it alone is driving the gear, while the other motor, together with the machine, must form a power takeoff and must be operated as a hypersynchronous brake - cf. FB.9.7.
Efficiency. With the general definitions, 1)
=-
(P. b / p.n )
- "
(lOB)
Loss'
= -
(PJP. n ) = I - 1),
(109)
Efficiency
=I
the total efficiency of a planet gear with two or three shafts running
'YJtot
=
I
+~ IPan
= I
+ M,(n,
Table 23. Efficiem.:y levels for epk-yclic transmission gears (For simple toothed wheel planet gears, 7112 = 1}21. for coupled planet gears, determine 11111 and TJu I separately; first index drive shaft, second power takeoff shaft.) <0
1
;121}12 -
i ll
wI
>1
Otol
1
-
+1
il:1/7121 - I ;12 - I
+1
-I
~ i 12 /Tl2l
-
i'27112- 1 i 12 - 1
~ i12/1/21 - 1
1
wi
-I
+1
-I
wi
-I
-I
+1
~
~
il.l. - 1/1112
wi
+1
i. 2
+1
-
1/1112
-I
- n,)2'IT(\ - 1)0")
!Pan
(110) is displayed, with P v as per Eq. (107), and one or two shaft outputs as per Eq. (104), which are shown as drive outputs by their positive sign. With a self-locking gear, however, a power takeoff shaft, the torque of which takes on a positive sign only as a consequence of self-locking (see FB.9.5), may not be taken into consideration. The minus signs in the definition equations (Eqs (lOB) and (109» are needed so that 1) and, should take on a positive value, as usual, in spite of the negative quotients (P" P,b < 0). The efficiency of transmission gears can be expressed using the stationary gear ratio and the stationary efficiency alone. For superposition gears, a free speed ratio must be added, e.g. k 12 , which characterises the power flow (Tables 2~, 24) [;B], with the relevant equation depending on the associated power flow in each case. Simple planet gears, representing stationary transmission, are practically loss-symmetrical, like standard gear train gears, i.e. 1)" = 1)". For epicyclic gears, however, especially for plus gears, the efficiency values in the two power flow directions can be very different, owing to the superposition of the rolling and coupling powers. For minus gears, the epicyclic efficiency values are always higher than the stationary efficiency values. In Eq. (110) and Tables 2~ and 24 - as also in the other literature - it is assumed that, with a circulating support, the load-<:ontrolled tooth friction losses and planet bearing losses in the transmission of the rolling power Pw are as large as for stationary transmission. These losses alone form the basis of the calculation. Additional churning losses and ventilation losses arising when the support is also circulating, losses due to sealing ring friction and
influences due to lubricating oil guidance can, if necessary, be taken into account after the calculation of 1)"", In the determination of the stationary efficiency, only the load-controUed losses referred to may be brought into play. Should more precise specifications not be available, then for practical calculations it is sufficient to assume a rolling efficiency 1)w. = 0.99 for a pair of external helical
gears and 1)wi = 0.995 for a hollow gear stage with internal gearing. The stationary efficiency, 1)1l, is determined from this as the product of the rolling efficiency of the individual gear stages - cf. Fig. ~6; for the more precise determination of the degree of effect, see [69J.
8.9.5 Self Locking and Partial Locking Self locking (sl) means that a gear cannot be moved, whatever the size of the drive moments, but instead remains stationary, internally blocked, because its friction power loss, Pv , would be larger than the driving power in the state of motion. However, it does operate if the additional power required to overcome the friction, or the "release moment" needed to release the jamming is fed into it through the power takeoff shaft in the power takeoff direction of rotation. A self-locking gear made to run in this way must be driven on all running shafts and produces only friction power loss. Example Self-locking hoist mechanisms must be driven on what are actually the power takeoff shafts to reduce a (driving) load.
As a rule, self locking occurs only in one area of the possible power flows of a self-locking gear. For power flows in self locking systems, the torque M j of a locked
1.11:. Table
~4:.
[58J; PF
Mechanical Machine Components. 8 Gearing
Efficiency levels for superposition gears and allocation of ranges of k 12•
TOS
I"
k"
k"
k,.
<0
>/1s
<0
i,z to 0
o to I
>1
o to I <0
o to ill
112 to 1
>1
>1
kts
and k25 to positions of totalisation output shaft, TOS
power flow
<0
o to I
1 to ill
>i 1z
<0
o to I
1 to ib
o to lIS
<0
>1
i l5 to 1
ils to 0
o to I
>1
1<
>lz5
1 to IZ5
2<
s
o to I
i~
to 0
s
<0
1- 112
I
(kl2 - 1121)1l)(1 - 112)
I
s<2
I
1<
1<
s
2
I
2<
2<
1<
(kl2 - 112 1)12)(1 - 112 ) (kl2 - i 12 )(1 - i 12 1)I2)
+1
+
kl2 - i 12
1)ZIO - k 12)
I s
I
1)21 (kl2 - 112 )
(k'I21)Zl -
(kl2
2
-I
- k 12 )
+1
-I
i 12 )(1 - i 12 ) 112 )
-I
'I1I2(k 12 -
i 12 ) + 1 - kl2
+1
k12 - 112
+ 1121(1 I i 12
- k 12 )
power takeoff shaft j, and thus the takeoff output Pi' reverse their signs, by comparison with a frictionless drive ('1'2 = '12' = 1). But the power takeoff output, PI' which thus becomes positive, does not become a "real" drive output. Thus, for example, the bearing flanks remain the same as if j were a power takeoff shaft. They do not change over to the other side, which is the bearing side for a "real" drive. The "power takeoff output" which has become positive may therefore not be used as p.n , but only as P ab in Eqs (108) to (110)! A negative effldency value for the running condition in the locked power flow direction is thus established as the criterion for self locking. Self locking occurs as reciprocal blocking of all three connecting shafts if the full release moment required for
+1
1)12(k ll - i 12 )
k 12 )
+1
i 12 )
+
1
k"
i 12 )(1
I
(kl2 - 112 )(1 - I 121)lz)
2>S
I >s 2 I
>2
2 >1
2 >1
I
2>S
I
I
>2
>2
2 >1 s
-I
(kl2 - i 12)(1)21 - / 12)
(k,,'/2. - /,,)(1
1)120 -
k
(2
+1
i 12 )
ku
-I
1)zl(1 - 112)
112 )
1)zl(k 12
i 12 )
I,,) + I
1)12(k12
+ i 12 (1
k 12 )
+1
k 12 (1 - 1)12)
ku
112
+
+1
i 12 )
/ 12 1)12)(1
(kl2
-I
112 )
(k 12 1)zl
s
+ 1121112(1 - R12) Ru(1 - 112 )
kl2 - 112
-I
-I
k IZ1)zl{1 - 112 ) + IIl{1
1)zl(k12 - 112 )
(kl2 - 112 )( 1)ZI - i 12 )
I
s
1)12(1 - ilz)
wi
1)101
2>s
s
II'lJ
Efficiency level
1)12(1 -
>2
s
;12)('1121
s
s
+ i 12 (1
k 121)21(1
I
S
kl2 - liZ + 1121)12(1 - k ll) k,,(1 112 )
s<2
+1
-I
2
s
-I
(k 121)zl - 112 )0 - 112) i 12) (kl2 / 12)(1)21
1- 112
2
2 >1
1121)12)
s
s
+1
s
i,,)(1
(kl2
PF
s
+ 1)z,{1 - k 12)
kl2 - i l2
wi
- ill)
k l2 0
s
s<2
2<
o to I
>1
I
1)101
+ I 121)I2{1 - k 12)
kl2 - 112
s
I
>1
i21< to 1
2
s<2
o to i25
Efficiency level
PF
IIl1)I2(1
k 12 )
(kl2 - i 12 )(1 - 112 "(112) 112 ) /12'1112)(1 (kL!
+1
1 - i 12
-I
kl2 - i 12
+
'I1IZ(1 1)12(kll
k 12)
11zl(1
ill)
+1
i 12 )
+
1
ku
-I
k 121J2I(1 - 112 ) 1121(k12 - i 12 )
+ i ll(1
R12 )
the locked power takeoff shaft to run is not supplied from outside. If the locked shaft here is the single power takeoff Shaft, with two or three shafts connected, then the gear is completely blocked and stationary (self locking, '1 < 0). If it is one of two power takeoff shafts, then the gear rotates, internally blocked, and transfers power, like a coupling, through the unlocked power takeoff shaft, with 1= 1, '1 = 1 (panial locking). Partial locking is possible only in a gear capable of self locking, with three connecting shafts, and characterised by '1 > 0, as is described for funher operational examples in [58,60]. Epicyclic gears are capable of self locking only if either 1. Their stationary transmission is self locking in one or
1>.9 Epicyclic Gear Arrangements _ 8.9.7 Design of Simple Epicyclic Drive Trains
both power flow directions (e .g. in a vehicle differential with self locking worm gears [60], or Their stationary gear ratio lies in the range ')J, < i"
2.
<
1/1)".
In case (I), the shafts I and/ or 2 are power takeoff shafts capable of self locking. Self locking or partial locking occurs if the rolling power flow of the gear coincides with the locked power flow of the stationary transmission. In case (2), the support shaft is the power takeoff shaft capable of self locking. Self locking or partial locking occurs only if it forms a power takeoff shaft for the gear. The change in the sign of its torque as a result of friction also occurs for the stationary transmission if the support shaft remains stationary, with only the direction of effect of its support moment being reversed.
8.9.6 Hints for Design Planet gears display some special design features by comparison with simple transmission gears [62]. By means of torque division through q, on the periphery of mounted planet gears or planet gear sets, the transferable power of planet gears or similarly constructed stationary transmission gears, known as branching gears or star gears, can be increased by the factor q if uniform support is ensured for all gears by such a statically over-determined arrangement, for example by making the elastic flexibiliry in the gear area greater than the actual dimensional differences here. When q = 3 (sets of) planets, the gear is statically determinate at the periphery, if one of the three gear elements, I, 2 or s, as often happens, is centred under load without support in the gear housing, through meshing alone. Nevertheless, additional dynamiC stresses are present - see [63]. None of the above calculations is influenced by the number, q, of planets/planet sets involved. Uniform distribution of several planets on the periphery is geometrically possible only if the tooth number conditions as per Fig. 36 are integrally fulfilled (for other gear shapes, see [58]). For "stage planets" (Fig. 36b, d, e) precise reciprocal pOSitional coordination of the two planet gear rims and marking of the pair of teeth meshing in the mounting position are required. Gears with simple planets are thus easier to manufacture. When the life of
the planet bearings is being calculated, th" c"ntrifugal forces of the planets are to be taken into account, and their relative speeds, as against the support. are to be used as basic data [64]. For gears as per Fig. 36. these are (n p \
-
ns)
(n l
nJzJ/zpl
-
(n/. - nS)Z2/zp2' For gears as in Zpl
=
Zp2
= zp and
n)ll
Fig.
(np2 ~6a,
f;,
n~)
=
f, insert
= n p1 = np'
8.9.7 Design of Simple Epiq'clic Drive Trains Transmission Gearing
'Meg
thus z, ~
liz,1 -
z,)/Z ~ 17. (z,
+ Iz,I)/
~ (34
+ 68)/3~34
integrally. The installation condition is fulfilled; if it is not fulfilled, Zmin and approximate centre distances should be varied using proftle offset - see FS.l.7. Finally, the calculation of the modulus as per F8.5 and the draft design should be carried out, with the centrifugal forces acting on the planet gear bearing being taken into account. Superpos~onGears
For each superposition gear, in addition to its stationary and two speeds, n, or a free speed ratio, k, gear ratio, the total output shaft is determined and moreover the power flow (PF) and the total efficiency, 1)<00 are established using a torque. Thus a desired PF can be coordinated with preset speeds only in limited ranges of free speed ratios, k - see Table 24. The range limits are characterised in each case by one shaft remaining stationary during stationary or epicyclic transmission, or by the "coupling point" (n, = n, = nJ.
i,,,
Speeds Constant. Should three constant speeds, n" nb and no be preset, then the stationary gear ratio required for this, i12 = itheoretical, is obtained from Eq. (94). If n a , nb and nc are used in the six possible combinations as n" n, and n" then the same three pairs of reciprocal stationary gear ratios are obtained, i.e. the same three gears as they are obtained from each of these individual stationary gear ratios by the variation of itheorctical, as in transmission gears and as per Fig. 37. It follows from the kinematic eqUivalence of these three gears that in each of them the shaft with the same speed, n" n h or nc, is the total output shaft. Thus, if three speeds are preset, the power distribution between the associated shafts is fixed, and is indeed independent of where and how these shafts are arranged in the gear format fmally selected - see Table 24. Example n a , n h , n~. = IS, 9, 12 S-I. With, for example, n l = 9, n l = 12, n~ = 18, using Eq. (94) it follows that: ill = 1.5, k\l. = 9112, and in Table 24 under ill> 1 and kll = 0 to 1 - TOS is shaft 2. i.e. the shaft with n = 12 S-I.
If two constant speed ratios are preset, e.g. k'b' k cb , itheoretical = iab is calculated from Table Zl, and three stationary gear ratios, together with suitable formats, as
then
for transmission gears.
Speeds Infinitely Adjustable. For a superposition gear with infmitely variable speeds, calculations are carried out for each of its two speed setting limits, one identified with 0 and the other with' (the identifications are interchangeable), as with constant speeds. In a layout as per Fig. 38, the following speeds are permanently associated with the gear shafts a, band c: n" variable power takeoff speed, adjustment ratio
Example i1h"",..,{kal = + J, smallest number of teeth, z" = 19, q = ;) planets on periphery. There are three possible stationary or epicyclic gear ratios as per Eq. (96), each with suitable formats as per Fig. ~6: ithe"n.t,cal
= i 12
+ J, formats
d. f
i 'h t"ort:tkal = i h : i 12 = I - Ii- = I - J = - 2. formats a, b
itht:oretical = ill> i,].> i l , give the same gears, with the designations 1 and 2 exchanged. Suitable format: gears as per Fig. ~a, with ill. = - 2, lead to simplest construction (see Fig. 37). Determination of tooth numbers: equations Band [) as per Fig. 36a must initially be fulfilled, and a,p = al. p must apply for the centre distances. For a zero gear (.\"1 = Xl = 0), it follows that:
Zl.
=
ilcZl
(lz,1 -
zp)/2:
(-
2)~4=- 6S.
alp
= all'
Figure 38. Symbol of a superposition gear with infinitely variable output speed. H constant-speed main motor; N auxiliary motor with infinitely adjustable speed; A, C possible supplementary gear positions.
'4"1"
Mechanical Machine Components • B Gearing
(111)
nb
= preset constant
(main motor, H)
n c ' adjustably preset (auxiliary motor, N) (112) For a speed reversal within an adjustment range, 'P < O. Either of the minimum and maximum ratios k can be classified as 0 or ., as desired; there are thus four possible combinations: a 'P" selected as desired, with two reciprocal 'Pc's selected as desired (gives two solutions) and the reciprocal 'P, with the same 'Pc (gives the same two solutions). From each, by using Eq. (97) on the adjustment limits, 0 and', two equations can be derived for the determination of the stationary or epicyclic gear ratio, i b • (with nc = 0) of a suitable planet gear for either a given b or a given k~b:
k:
(113a)
~it~~rr~-0.5
~~~-w~_1.0j-IO~ -2.0 -0.5
~
-2.0
-1.0 -0.5
(114a)
The speed ratio, k, not preset in either case, is obtained for both limits, 0 and' (as well as for any intermediate speeds selected), from the adjustment function valid for the entire speed range: (113b)
or kah,c
=
- k cb ( I - i b .)
(l14b)
iba
The limiting speed ratios,
k~h,a, k~b.a' Of k~b,'-" k:h,c, calculated in this way certainly give the preset adjustment ratio 'h or 'P" but do not usually give the desired speed ratios kthcoJ"eticalo An adjustment to shaft a using one additional
transmission gear, A, is thus required, if Eq. (113) has been used for calculation, or to c through C, using Eq. (114) - see Fig. 38. The gear ratio of such a supplementary gear, depending on position, is
The A or C position of such a supplementary gear influences the absolute speed in the gear, so that, after the detailed calculation of all possibilities, it is placed at the point that leads to the most favourable speeds and torques in the entire installation. The most suitable planet gear is selected from the four solutions for iba = i[hc.:orCtical' with three possible solutions in each case, as in F8.9.7. The distribution of the driving power to the main motor Hand the auxiliary motor N can be calculated for the solutions found as per FB.9.4. It is the same for several solutions, right down to the influence of the efficiency, for, according to [65]' if operation is considered as being free of losses (1) = I), it is dependent only on the adjustment ratios 'P. and 'Pc: with the power of the auxiliary motor N with reference to the total drive power being
apply at the adjustment limits. The favourable combinations of 'P. and 'Pc can be estimated using Eo = - 0.5 ... + 0.5 before work begins on
>
0.5
'Po
- 0.5 -10 1l'Po
bSSl onlyletl
or
0
"""::L..<"-LLL.JLLL..!:.....'-.LL.L'--LLLL_..L_-'-_-'" - CD -00 1.0 2.0
1.0
0.5 1/'P,
0.5 [22JonIYle~H5 Dietl and Ic~
I
<
0.5
Figure '9. Power ratio, eo, plotted against combination of adjustment ratio, CPa and '1',-, of a superposition gear as per Fig. 308.
the layout (Fig. 39). For Eo < 0, the auxiliary motor, N, runs as a generator, Pc < O. Example with solutions 1 and 2: reqUired na::::: 66 to 40 S-I, nh::::: 2S S-l, TJd = 33 to - 50s- I , n~.l = SO to 33 S-l. Whence CPa = n~/n: ~ 66/40 ~ 1.65.
-
8.9.8 Compound Planetary Trains Gear Symbols and Shaft Denotation Gear symbOlS as per Fig. 40, though they contain only the information required for the calculation (position of shafts and their couplings) make it easier to keep track and simplify the analysis and synthesis of compound planetary trains considerably. The shafts of all sections of a compound planetary train are henceforth referred to as 1, 2 and s, with the addition of one dash for the shafts of the second gear (1', 2', s') and two dashes (1", 2", s")
7tcf e
f
Figure 40. SymbolS for epicyclic gears: a With support shaft in any or unknown position, b Shaft 2 designed to be stationary, C shafts 2 and s can be connected or disconnected or flXed by braking, d epicycliC adjustment gears with infinitely variable stationary gear ratio, e.g. hydrostatic epicyclic adjustment gears, e simple adjustment gears with stationary housing and two connecting shafts, specified by figures> 2, f simple adjustment gears with infinitely variable gear ratio, stationary housing and shaft specifications> 2, e.g. V-belt adjustment gears.
1 ...1.
S.9 Epicyclic Gear Arrangements. S.9.S Compound Planetary Trains
JWJ@h~B a
I Z
I' Z'
I" Z"
B
A
b
Figure 41. Example of three-stage series planet gear' a Diagram, b symbol with shaft specifications transferred from a, here l"'B = ib iI', iI''''';
'71AR
= YJh 1J1'~' 111",". 1JlIA = T/s"l" 11,'1' T}sl'
for the shafts of any third planet gear present, etc. (Figs 41, 42). Thus all equations given previously. including those for Tables 21 to 24, or an existing computer program [60] can be directly used for every section. The dashes added to identify the sections are ignored during calculation in each case and then restored.
Formats of Compound Planetary Trains
Series Planetary Trains. (Fig. 41) are planet transmission gears connected in series, each with one fixed shaft in order to bring about higher gear ratios with high efficiency. A reduction in the space rcquireUlcnt and the best total efficiency are obtained with minus gears as per Fig. 36a, b. Calculation of entire reduction ratio and total efficient')' analogous to simple multi-stage transmission gears (see FS, Introduction and FS.1.2).
Coupled Planet Gears. (Fig. 42) consist of two planet gears, which are connected to one another by two shafts each. As transmission or superposition gears, such trains provide for particularly low unit weights and unit volunles for gear ratios up to i> 1501 [59, 661. If provided with external connecting shafts, I. If and S. in accordance with Fig. 42b to d, coupled planet gears have three connecting shafts with the degree of freedom F = 2. like a simple planet gear. Thus, as a set of gears it also has the same operating behaviour, and can be calculated precisely in the same way, using the sanle equations and Tables 21 to 24, if the indices I, 2 and s are used instead of the
silmL=ru
S~~J
lE 1
b
analogous shaft denotations I, II and S [5S]. If the connected coupling shaft S is fIxed, then the set of gears acts as a series train like a stationary tf'dnsmission, and its "series gear ratio" (analogous to a stationary gear ratio), i lll , together with its series efficiency (analogous to stationary efficiency) 711 [J and 71u [can be determined as for series trains (Fig. 41 - see example). If coupled planet gears are operating as a superposition gear, then their two sections are equivalent in their functions. If one of its individual shafts, e.g. shaft II, Fig. 42b, C, is fixed, then the associated section acts as a transmission gear and can be formed by a planet gear with a stationary shaft or by a simple transmission gear with a stationary housing. Here, as the "auxiliary gear", N, its only task is to preset the speed ratio, k" = in' of the "main gear" H, which is connected to the external connecting shafts. The external gear ratio of the coupled planet gears, i" = k<", can be calculated using Table 21. If the function-oriented denotations as per Fig. 42d are replaced by the general denotations (see F8.9.2), e.g. e ~ a, a -~ b,f ~ c, then, in Table 21, fIrst column, the speed ratio desired becomes kea = kab = k d ) (I - fab) + i ab, and converts back to the original denotations as per Fig. 42d: (ll7)
where tea is the gear ratio of the main gear when the shaft f is intended to be stationary. Example For gears as per Fig. 42' i l2 = _. 4.:1, i l T = ~ 0.36. So in the previous equation k ra = kl. = in = 1/~ 0.36 = - 2.778 and (." = i;, = 1·-il2=1+4.5=5.3, and using Eq. (117) ir~=k<"~= - 2.:-:'8(1 -- 5.3) + 5.:1 17.24. The same result is obtained, together with the efficiency, if the coupled planet gears are generated analogously to a simple planetary train: as per Fig. 42d and c and Eq. (96). S
o
i,,'
(1 -
l/i, ,.)
~
-- 4.3· (1 - 1/- 0.36)
16.24.
Whence. using Eq. (96), i lS = 1
~
ill = 1 - (- 16.24) = 17.24.
Series efficiency: 1]1 II =, T/d' T/f',·'
= T/Il. . T/2',6 = 0.985·0.989 = 0.9"74.
with
as per 'I'able 23, and with
S \Xlhence, using Table 23, under iu. < 0:
1
c
~
S
d
I~ ~U
Figure 42. Example of t:oupJed planet gear as turho-prop. Redut:tion gear 1681. a Sectional drawing. b Diagram with shaft specifications. c Gear symbol with correctly positioned shaft specifications transferred from b. d Symbol of a coupled planet gear with
(- 16.24·0.974 - 1)/(- 16.24 - 1)
0.976.
driving power of auxiliary gear Co
= driving powe;-~f'~;~pli'~g gear [5S[ applies
function-oriented specif1cations for its shaft as per its position: u,
a' connected coupling shaft;!,/, free coupling shaft: e , e' individual shafts; t, ll, 5 external connecting shafts specified as for simple epicyclic gear,
~
If friction is neglected, the power !lowing through the auxiliary gear is dependent only on the gear ratios, and can easily be estimated using denotations as per Fig. 42d: if the power ratio is defIned as
"" = (I
- I/i,,)/(l -- 1/i",.).
(lIS)
I ...t.l
Mechanical Machine Components. 8 Gearing
For the example in relation to Fig. 42, eo these equations.
Adjustable Coupled Gears (Fig.
=
00
0.693 using
are coupled planet gears which, as auxiliary gears, contain an adjustable gear with an infinitely adjustable gear ratio, if"" and thus also offer an infinitely adjustable overall gear ratio, i". Their adjustment ratio 'I' (adjustment range) is infmitely selectable for a specific adjustment ratio '1" of the auxiliary gear N with a suitable layout for the main gear H. As a rule, a commercially available adjustable gear is used as the auxiliary gear, the housing of which, being a ftxed "suppon shaft" corresponds to the individual shaft e' of the auxiliary gear_ As with coupled planet gears with constant gear ratios, a separate calculation is carried out for the two limiting gear ratios of the adjustment range. Here all variables associated with one anothcr at a selected limiting gear ratio are identifted using 0, while the corresponding variables at the other end are identifted using'. Thus the adjustment ratios 'I' for the coupling gear and '1" for the auxiliary gear, are deftned as follows:
e.
Z.O
05;:0
1.0
1.0
4~)
OS Oe.
-05 -1.0 J-1.0 Z.:J
-00
-2.0
-10 -0.5
-0.5
-1.0
11'1'
0.5
1.0 1.0
2.0
-Q5~
J 00
0.5 119'
(119) If the speed is reversed within an adjustment range, 'I' and/or '1" is negative. The load on the auxiliary gear (which it is desired to keep low) can already be estimated from the adjustment ratios (Fig. 44) for operation conceived as beingfriction-jree (index 0): at the adjustment limits, with So as per Eq. (118): s~
=
('I' - 1)/('1" - 1),
sg =
e~
=
s~
Figure 44. Power ratio f:o plotted against combination of adjustment ratios tp of coupled adjustment gear and cp' of its infinitely adjustable auxiliary gear.
'1" /'1'.
To obtain the preset adjustment ratios 'I' and '1" , a planet gear should be designed with the gear ratio ie, between the shafts e and a, with the shaft f designed to remain stationary. Depending on whether the limiting gear ratio is assumed to be iis or i~.,., the result obtained is either (120) or
a rule, it differs from the desired theoretical gear mtio itheoretical. so that supplementary gears 1I1 and/or V are required, as per Fig. 4~a and b in a layout using Eq. (120) or as per Fig. 4~c and d using Eq. (121). The correspondence between i ls 'and ira' is closer for any operating point within the adjustment range using Eq. (117) - see [70].
Reduced Coupled Planet Gears are coupled planet gears I/i"
= 1 + (1 -
rp)/[r,.,.(rp - '1")].
(121)
Whichever limiting gear ratio was not selected, t;.'a' or 115, it is then obtained from kf' = if"" using Eq. (117). As
f fl 3 1II • IY
a
b
Figure 430. Coupled adjustment gear with infinitely adjustable vbelt gear (70). a Symbol of format with supplementary gears, III and V, belonging to auxiliary gear. b Gear diagram of coupled adjustment gear as per a with supplementary gear, fIl. c Symbolic representation with external supplementary gears, 111 and V. d Gear diagram of gear as per c with supplementary gear. V. and additional two-stage gear with i = t tor centre distance bridging.
in which the suppons form the free coupling shafts, ff' , of the two gear sections (Fig. 42d) and can thus be combined into one element. Moreover, the gears of the two gear sections mounted on the connected coupling shaft and the planet gears meshing with them are of the same size; they can thus be reduced to a single gear pair [58, 67] - Fig. 45. However, a given reduced coupling gear can be expanded into three different coupled planet gears, depending whether shaft A, B or C is considered as the coupling shaft, 5, connected to the gear. All three have the same speed behaviour in relation to the shafts A, B and C, and are thus kinematically equivalent. However, their efficiency values can differ conSiderably from one another. The only simple coupling gear which "has the same effect" as the reduced coupling gear is the one for which one of the individual shafts, I and II, is a differential shaft for their gear section, while the other is a totalisation shaft (F8.94) [58, 60J. At the same time, this has the highest efficiency. It can be determined using a simple formulation [60]: if a stationary gear ratio, i xy > I, then y is a totalisation shaft. Otherwise, i.e. even with a negative gear ratio, y is a differential shaft. Let the shaft 5 in Fig. 4Sb to d be identified as x and the pan gears I, Il and Ill, each connected to I and 11, be identified, each in succession, as y. Then i12 or i21 becomes i..,. In Fig. 45, the totalisation shafts are marked by double dashes in the symbols. The combination in Fig. 45c proves to be the coupled planet gear with the same effect, which is now analysed, taking the place of the reduced coupling gear, as was described in connection with Fig. 42.
I ...i l
8.10 Design of Geared Transmissions. 8.10.1 Types
5
~
I~
1'
JI Z'JI
5
5'
2
a
Z'
A<~r~~c:,
b
Figure 47. Gear with coaxial drive and power takeoff: a without, b with torque division; B large overall length. T long intennediate shaft, A torsionally elastic shafts.
f~!/J·
b
a 5
I
1 ~"I"? 1
I
II 5
2"
IIIJI
I
5
5"
Coaxial Drive and Power Takeoff First stage badly used in format as per Fig. 47a. Smaller and lighter gears obtained using torque division, for example as per Fig. 47b (but more complicated, load balancing elements required); planet gears (see F8.9). Distribution of total gear ratio for condition: minimum total gear volume, free selection of bid or bla (check as per Table 8); index J, fIrst stage, etc. For (THlirn values, see Table 14. Two-stage gears: (122)
Three-stage gears:
u. = O.6u 417 un
Figure -is. Reduced coupled planet gear. a Diagram of reduced coupling gear. b-d Schematic representation and symbols (with double dash for totalisation shaft) of three kinematkallyequivalent simple coupled planet gears derivable from it with I: as the one with the same effect.
8.10 Design of Geared Transmissions The rules and guide data given here are the bases for many constructions used in the machine building industry for
= 1.1u2!7
(UHlimdaHlirnn)217 (UHlimiuHlimll)4r'
(123)
(UHlirnn/aHliml)417 (aHlirnU/UHlimlII)217.
(124)
Total
(125)
Straight Bevel Gear Pairs
More rigid and more cost-effective than bevel gear pairs for i > (3 to 5) as per Fig. 48 (large ring gears, thin pinion shafts). Usually bevel gears in fIrst stage (for larger moments spur gears are more cost-effective and less sensitive in second and third stages); exception - rapid-running gears with high noise requirements [ 1] or modular gears 135].
average ratios. It is useful to round up the dimensions so
obtained. Other dimensions are known from experience to be within specifIc ranges, or are useful or necessary
for individual pieces of research. If possible, strength and rigidity should be calculated. 8.10.1 Types Spur Gears
Normal Format (as in Fig. 46a and b). Simple, reliable in operation, with easy access. For larger, multi-stage gears, symmetrical format as per Fig. 46c: - larger total tooth width, compact.
a Figure 46. Gear with laterally offset drive and power take-off. a
Single-stage for i < 6(8) b Two-stage for 6 < i < 25(35). pinion of first stage mounted so that torsion and bending work against each other. C Heavy-duty gear
Fiprc 48. Straight bevel gear pair (Lohmann & Stolterfohlt, Witten). Rated power P = 280 kW, splash lubrication, 351 oil. weight without oil 495 kg, oil level check with dipstick, spur gears case-hardened and ground, bevel gears case-hardened and lapped; 1 spring rings as stop for coupling hubs, 2 shaft nut, locking ring and washer, 3 depth of fit for cover adjusted at fitting, 4 oil feed from catch pocket, 5 Nur bearing in stator bore H7, 6 adjusting pin, 7 axial bearing, 8 radial bearing, 9 spring ring with angular shim, [0 H6/u6 shrink fit, 11 cover sealed with sealing paste.
1."1'
Mechanical Machine Components. 8 Gearing
Worm Spur Gears Economical from I> 12, depending on size. Wonn gears in first stage as far as possible (efficiency, noise, size); exception - if spur pinion mounted directly on motor shaft, e.g. in gear motors (no coupling, no separate pinion mounting needed). 8.10.2 Connection to Drinng and Driven Machine Electric motor often flanged directly on gear for gear motors up to 50 kW (usually 0.4 to 4 kW) (no coupling, no separate setting up, no alignment). For higher power levels, usually separate setting up, connection to motor and machine through compensating couplings (see F4). Considerable forces can be introduced through transverse misalignment and offset angles or protruding couplings, axial movements of the motor armature and the power takeoff - in spite of compensating couplings - (pay attention to two half-arches in dimensioning bearings, housing, shafts and force distribution!). TItis does not apply to plug (slip-on) gears for the power takeoff shaft, and with flanged motors it does not apply to the drive side either. The gear power takeoff shaft is pennanently connected to the shaft of the machine and the gear rides on it. Gear weight and transverse forces from the support moment must be absorbed by this shaft and a torque support. 8.10.~
Detail Design and Measures of Gears
Finished cast gears - including gear-cutting - (also injection cast gears) for small dimensions, light loads and large quantities of components, if necessary with cast-on cams, dogs, etc., for heavy loads also finish-forged (e.g. differential bevel gears). In machine construction for small and medium-sized dimensions, usually fully-turned or contour-turned disc gears; for larger dimensions welded gears have largely forced out cast, shrinkage and dust constructions (even for alloy steels up to 300 HB or possibly 340 HB) (see F8.4).
Gear Types For d < 500 mm and series - drop-forged, for individual manufacture solid discs or web gears (light construction) made of forged round stock; for 500 < d < 1200 mm, disc gears or web gears, free-fonn forged and/or contour-turned if necessary, even for larger dimensions with severe safety requirements; for d> 700 mm usually welded (b/d < 0.15 to 0.20: singledisc gears, above that twin-disc gears, b > 1000 to 1500 mm: triple-disc gears). Transition at lower values under heavy load, thick bandage, vertical shaft, if high axial rigidity reqUired (large f3), for finer gear quality (rigidity during gear-cutting!).
General Embodiment Design Rules (Fig. 4,). If hR undershoots the limiting value given here, the gear must be cut into the shaft. With shrunk-on, thin tooth rims, pay attention to shrinkage stress and tooth root stress [37]. Always check whether clamping is possible for gear cutting and gear grinding. For solid gears and disc gears see Fig. SO; for welded gears see Fig. SI. For spedftcations for gears and gear body dimensions in drawings, see DIN 3%6 and DIN 7184. 8.10.4 Embodiment Design of Gear Cases Usually overall housing as bearing construction - for examples see Figs 48 and ~.
a
Figure 49. Gear body dimensions - general. a for relief of tooth ends: at b> 10m: b, = m, at b
< 10m: b, = I + O.lm.
PI locating faces (internal or external) for gears which cannot be generated on shafts or tensioning spindles, from diameter approximately 700: b p .... 0.1 mm, b p = 10 nun. 2, locating face, P2, at b> 500mm.
Axial runout: Nat vt :s 25 ms, Tat vt > 25 m/s. Sprocket boles, clamping bores and ligbtening boles: Quantity n: d, 300 < d,
< 300: - (clamping througb bore), < 500: n = 4,
500 < d, < 1500: n
=
5,
1500 < d, < 3000: n = 6, d,
> 3000:
n = 8;
(Check clamping facilities of work site) - no bores in high-speed gears; threaded blind holes, G, for transportation of solid disc gears heavier than IS kg.
Hub diameter. d N = (1.2 to 1.6)dw (depending on material, shrinkage: small values for large d w ); hub width, b N ' " d w and b N ' " d,/6 (for helical gears, check tipping through neutralisation of play or shrink fit slipping). Avoid V-shaped projection of hub (cf. Fig. 50). b For protection against transportation damage: Edge break, a
= 0.5 + O.Oldw '
Tip edge break, k = 0.2
+ 0.045m.
Face edge break, t = 3k. Edge rounding with radius = k or t for strictest requirements (e.g. aircraft gears) and nitrided gears (see also F8.4).
c Basic hub tbickness: Unhardened or nitrided, b R > 205m. Case hardening, flame hardening, induction hardening, flank hardening or space hardening, b R > 3.5m. Rotary flame hardening or rotary induction hardening, b R > 6m (pay attention to pOSition of adjusting spring and to shrinkage strain). Foe surface hardening, specify which areas must remain soft, e.g.
threaded holes, possibly bores).
For larger gears, from time to time rigid lower cases with mounted bearing upper sections. Upper sections then have only protective function, and must be easily inspected [1].
8.10 Design of Geared Transmissions. 8.10.4 Embodiment Design of Gear Cases
...
, .,.,
General Embodiment Design Rules Cast Housings with more than three sections preferably made of GG 20, large gears GG 18 (easily castable, little shrinkage or distortion, easily machinable), GGG 40, (is 38.1 (weldable!) (higher strength, more difficult to process). Pay attention to greater heat expansion and lower rigidity for light metals.
a
Welded Housings make it possible to save weight (rigidity through fins or profile); suitable for individual manufacture and impact stress. Material usually St 37-1 or 2 (high-stress: St 52-3).
b
Figure SO~ Dimensions of turned or forged and turned gear bodies. Avoid V-shaped projection of hub (machining of plane surface, stacked mounting).
Undivided Housings preferred for small gears; instal· lation through lateral apertures. Incidentally, horizontal joints in shaft plane favourable for sealing, installation, inspection.
a Nonnal fonn (provided there is no limitation on weight) made from forged round steel: lateral boring because of machining ('05t5, and to ensure support from chuck only at (d) - d N )/2 > 25 mm. Cost·effective for unhardened and hardened gears (low machining volume, little distortion due to hardening) bJ
;;:::
3m; b A = 0.5
+ O.lm,
Bearing Bolts should be designed in accordance with static tooth root load-bearing capacity. Tighten to 70 to 80% Re. Provide for at least two alignment pins (d = 0.8 flange screw diameter), for larger gears, others near the bearings. Secure screws in inside of gear with wire. Provide for at least two opposing threads for fordng screws.
max. 2 mm.
Transverse bores (quantity: Fig. 4,): d M = 0.55(dN + dJ ). d n = d a/20 === 30 mm, edge distance between bores ;,::: O.8dIl • d N ; see Fig. 4'. b Light construction format (e.g. aircraft and spacecraft, small working load, as per prototype test): b H > 2r,; d H ~ (0.1 to 0.2)d,. Number of bores: see Fig. 4,; b J ~ b R ~ 1m (as per Fig. 4,<); r, = t (as per Fig. 4,b); d N , see Fig. 49; d M • see Fig. SO•.
Unhardened, b,
~
l.5m
+ O.lb.
nitrided,
case-hardened
(very
Foot screws should be calculated from support moment of gear. For steel frame fOlUldations alignment pins and adjusting screws (with fine threads) useful in gear root.
light):
Flame-hardened or induction-hardened. case-hardened (less light): b J = b, as per Fig. 4,,,; b, ~ 2m + O.15b. Drop-forged or free-form forged: generatrices, K:
~
Distance between gears and housing walls large enough to avoid jamming of fragments and high pumping of oil. Distance between gears, and between gears and housing walls, laterally and on diameter as per numerical value equation,
to 10° tapered.
'ti' -~
~~~~~
bB
~I
I
I,
Relief notch I r~bs/3
a
Figure S1. Dimensions of welded gear bodies. Web plates and fins are usually made from St 37.1 or 37.2, hubs from St 52.3. Welded joints are formed as per stress and manufacturing possibilities. h.1 = hR as per Fig. 49c; d ll , d M , hI! as per Fig. SOb; quantity of bores or pipes, see Fig. 49. a Single-disc gear: b,. = complicated.
Without lateral fins, if
O.12d~
/3 <
+ (~ to 10 mm), depending on difficulty of operation, thicker if need be, if precise manufacture (clamping)
10'.
With lateral fins, in general, if f3 > 10°; thickness of lateral fins
=
O.6bs ;
b~
= 1. Sb,.; rs
= I';b~
(at least 10 mm).
Quantity of fins:
< f3 < 20° = quantity of bores. f3 > 20° = double quantity of bores.
at 10° at
b, " Twin. 40 mm; d R = (0.12 to 0.20) (dJ - d N ), at least SO mm: SR = (0.:\ to O. ~ )bs for smaU to large pipe diameters. Stiffening fins between pipes approx. O.Sbs thick; b v = 2b~; rs and quantity of fins as in a. For other dimensions and quantity of stiffening fins see Fig. 49.
E vent hole, approx. 06 mm; after stress-free annealing weld up or close with screw. b Format for d~ < 2000 mm. hR > 40 mm. Form bI for hub jutting forward or back (dashed). Then support to cut teeth on gear rim and pipe. Safe (costly) welded connections for high dynamiC stresses. (Also useful for types in Fig. Sla and c.) c Format for d s > 2000 mm. Smaller pipe near tooth rim (h~ = 40 mm; as small as possible) to let clamping screw through; larger pipe to let damping mushroom head through. Other dimensions as b.
'M"
Mechanical Machine Components. 8 Gearing
Table 2S. Guide data for dimensions of gear housings (L
=
Component
Designation
Wall thickness for bottom box (a) Unhardened gearing (b) Hardened gearing
longest housing length in mm) Cast construction b
Welded construction
0,007L + 6" 0.005L + 4 O.OIOL + 6" 0.007L + 4 GG, GGG: 8; GS: 12
0.0()4L
+4
0.005L
+4
W",a
GG GGG, GS GG GGG, GS minimal maximal
50
4 25 0.8ww 0.5ww
Stressed top box, bearing cover Unstressed hood
0.8ww 0.5ww
Reinforcement and coaling flns
0.7 x Thickness of walls to be reinforced
Flange thickness Flange width (projecting part)
w," b,
3ww (Wall thickness ww)
Continuous base strip with recess Continuous base strip without recess Continuous transverse base strip Base strip width (projecting part)
3.5ww
External diameter of bearing housing
1.2 x External bearing diameter
Bearing bolt diameter Flange bolt diameterf! Distance between flange bolts Foundation boltsi Inspection cover bolt
2ww J.2ww
3.5ww
J.8ww 1.5w~
+
15 mm
(6 to lO)d," 1.6ww
0.8ww
1.5wl. 4.5ww + 15 mm
3ww 1.5ww (6 to IO)d,"
2ww 6ww
aFar gears from approx. L == 3000 mm bottom box often double-walled with approx. 70% of above wall thickness. blifting taper approx. 3°, ~'For turbo-gears: + approx. 10 mm (vibration and noise damping). dThicker if necessary. in accordance with noise level required. "For bolts and nuts. fAs close as possible to bearing. sLitting screws of same thickness. hDepending on density requirements. iQuantity """ 2 X quantity of bearing screws.
SA
B
= 2 + 3m + B with
= 0.65
(v, - 25)
20
0 (v, in m/s)
to ground about 2sA , provided oil supply adequate. For injection lubrication large drain aperture important: diameter approximately (3 to 4)SA' For splash lubrication, oil drain plug (with magnetic spark plug if necessary, see below) at lowest point. Gradient of gear base toward drain aperture 5 to 10%.
Locating Faces for larger gears on narrow sides of bottom flange, jutting out approximately 120 X 40 mm, for large gears at external bearings also. Contact pattern can be repeatedly set using spirit level. Machining of Flange Surfaces, R, = 25 fl-m, bearing points and bearing faces, R, = 16 fl-m, inspection cover, root surfaces, R, = 100 fl-m. Inspection Cover should pennit inspection of entire meshing over entire tooth width, together with lubricating oil supply. If danger of loosening exists, provide hinged cover and hinged bolt (e.g. on crane control gears). Through.holes should be avoided to inside of housing (oil tightness)_
Lifting Lugs, Eye-bolts or similar should be provided to allow the removal of the top box and to lift the gear (on the lower box). Ventilation for pressure balance with filter (against dirt and moisture) at highest point (pay attention to direction of splashing!). For splash lubrication, sight glass or dipstick required. The dipstick can be provided with a magnetic sparking plug (wear check). For injection lubrication, connections for monitoring oil pressure, rate of flow and temperature [1]. Housing Dimensions are determined by inherent stability (not strength). For guide data, see Table ::liS. 8.10.S Bearings
Roller bearings are preferred throughout. Friction bearIngs are only used for high-speed gears (about v, > 30 m/s), very large dimensions or specially quiet running. Bearings need to be as tight as possible to gears (for minimum distance see F8. 10.4) , but the minimum distance between bearings should be 0.7d, (effect of centre distance variations, bearing rigidity, overturning moment due to axial force).
8.10 Design of Geared Transmissions. 8.10.5 Bearings
Avoid overhanging. If necessary, select distance between bearings approximately two to three times overhang, shaft diameter > overhang. For double helical gears, ftx only one shaft axially, in general the gear shaft (with larger dimensions; larger axial forces often introduced from outside through them). For small gears, grooved ball bearings, fast and loose bearings usually economic, for medium sizes grooved ball bearings as fIXed bearings, cylindrical roller bearings as movable bearings or tapered roller bearings in 0 arrangement (provided distance between bearings not too large). For spur or helical gears with F,/F, "50.3, cylindrical roller bearings possible. Take up high axial forces using separdte thrust bearings:
Four-point contact bearings (also when axial force is reversed).
Self-aligning roller bearings up to F,/ F, = 0.55. Note: At F,I F, > 0.1 to 0.25 centre bearing, but not below this;
'iM..t'
if applicable, pay attention to angular deviation when axial force reversed and axial play relatively large. Double-row tapered roller bearings suitable for high axial forces and changes of direction, Fig • .J4.
Adjustable bearings, e.g. with eccentric cases, used to set contact pattern for large and high-speed gears.
Bearing lubrication for series gears using oil spray or oil collection pockets, out of which oil or bores (d = 0.01 X external bearing diameter, minimum 3 nun) run behind bearing. For large and high-speed gears, usually injection lubrication (oil nozzle diameter 2: 2.5 nun owing to danger of stopping up, corresponding to approximately 3I/min); ensure oil return from cavity behind bearing through bore (d = 0.03 X external bearing diameter, at least 10 nun or several bores) (at height of lower roll body, thus providing oil supply for startup).
Kinematics H. Kerle, Brunswick
9.1 Systematics of Mechanisms 9.1.1 Fundamentals Definition of Mechanisms. Mechanisms are systems for converting or transmitting movements and forces (torque). They consist of at least three links, one of which must be deftned as the fixed link [IJ. For the sake of completeness, a distinction is drawn between kinematic chain, mechanism and motor mechanism. The chain becomes a mechanism when one of the links of the former is selected as the fixed link. The mechanism becomes a motor mechanism when one or more links of the former is driven.
Mechanism Structure. Structural studies of the type,
Points on links of planar mechanisms describe paths in mutually parallel planes; points on links of (generally) spatial mechanisms describe spatial curves or paths in mutually non-parallel planes; spherical mechanisms are special spatial mechanisms with point paths on concentric spheres (Fig. 2). A kinematic pair comprising two contiguous links (or sections thereot) determines the joint. Planar mechanisms require planar joints with up to two degrees of freedom (translation and rotation), while spatial mechanisms on the other hand very often also require spatial joints with up to ftve degrees of freedom in addition to planar joints (Fig. .J). For example, turning and tuming-and-sliding pairs are characterised by a shaft and bore, sliding pairs by hollow and solid prisms, screw joints by a nut and bolt, ball-and-socket joints by the ball and socket. Lower kinematic pairs or sliding joints are in surface contact (e.g. shaft and bore), higher pairs in linear (e.g. cam plate and
number and conftguration of links and the joints which connect them usually begin with the kinematic chain. There are open and closed kinematic chains, with or without multiple power-transmission paths (Fig. 1).
a
c
LbO 9 II-
d
Fipre 1. Kinematic chains: a open, b closed, It open with multiple-power transmission paths, d closed with multiple-power transmission paths.
b
c Figure 2.. Sample mechanisms: a planar, b general spatial (shaft coupling); c spherical: 1 fixed link.
1 ...1:.
Mechanical Machine Components • 9 Kinematics
JOint
V
~
,
Planar
/\
~ ~ Y
Turning pair
~ Sliding pair
0-------0
Single: 1
~
Double: 2
e
d
c
f
Figure 4. link symbols for planar mechanisms: a binary (n 1 ) link with two turning elements, b binary (n 2 ) link with two sliding dements; c ternary (n~) link with three turning elements, d quaternary (n , ) link with four turning elements, e quaternary (n 1 ) link with two fuming and two sliding elements, f fixed link.
-'-
Cam jOint
• b
a
IT
~ 8 8
+
c~
Degree of freedom
Symbol Spatial
Spatial: 5
ing to move the entire nlechanism (e .g. a roller with a rotary bearing on a cam plate), F is reduced by these
Planar: 2
identical degrees offreedom. GrUbler's equatiun
~
F
4!
~
~
A
Ball-and-socket Joint
(@
3(11 .- 1) - 2g
(2)
applies to planar mechanisms that only have turning-andsliding joints with f = 1. F = 1 means constrained motion, e.g. for the four-bar linkage (Fig. Sa) where n = 4 and g = 4. For a five-bar linkage (Fig. Sb) where n = 5 and g = S, F = 2 applies. The degree of freedom of a mechanism indicates the minimum number of drives or input inlpulses a mechanism must receive in order to fulft1 a function which is calculable in advance. Where F = 2, motion must be induced at two independent locations (e .g. main and regulating drives) or two independent ti)rces or moments act as input impulses (differential gear mechanism or self-adjusting mechanism). Correspondingly higher minimum requirements apply in the case of F> 2.
Screw Joint
Turnlng-and-slidlng pair
=
9.1.2 Types of Planar Mechanism
~
r-
Planar joint Figure 3. Joints and joint symbols.
roller) or point contact (e.g. ball on a plate). Form-closed joints ensure contact between the elements by means of an appropriate form, while force-closed joints require one or more additional forces to maintain permanent contact. In the case of planar mechanisms which usually have turning-and-sliding pairs it is expedient to divide the mechanism links according to the number of element sections into binary (n z-), ternary (n,-) and quaternary (n 4 -) links (Fig. 4), especially as in addition a planar cam joint can be replaced kinematically by a binary link (cf. F9.1.2).
Four-Bar Turning-Pair Linkages. A four-bar linkage is capable of rotation if Grasbofs criteriun is met: the sum of the lengths of the shortest and longest links must be less than the sum of the lengths of the other two links. There can only be one "shortest" link (lmin) but up to three "longest" (i.e. identical lengths). Depending on which of the four lengths a, b, c, d (Fig. Sa) is I min , the resulting mechanism is either a crank-and-rocker (lmin = a, c), a douhle crank (lmin = b) or a double rocker with rotating coupler (lmin = d). Four-bar linkages that cannot rotate are designated non-rotatable double rockers. All relative rocking motions occur symmetrically to the adjacent link. There are internal and external rockers. Non-rotatable four-bar linkages can only contain one "longest" element but up to three "shortest" [2J. The third group comprises folding mechanisms where each of two link pairs are identical in length, e.g. parallel-crank mechanisms [3J.
Degree of Mechanism Freedom. The degree of freedom F of a mechanism is a function of the number n of links (including the fixed link), the number g of joints with the respective degree of joint freedom f and the free-
dom of motion b: F
= ben
- 1) -
±
(b - It).
(I)
For mechanisms which are generally spatial b = 6 is inserted, and for spherical and planar mechanisms b = 3. If, in addition, individual links can be moved without hav-
I I \
/
,
a Figure S. Planar turning-pair linkages: a four-bar linkage (F = 1), b five-bar linkage (F = 2).
9.1 Systematics of Mechanisms. 9.1.2 Types of Planar Mechanism
Klnemalic chain
Equal angular ; velocities
Mechanism
:a
I .....
Example
b
re b
2
III
o ]
I
-
Figure 6. Four-bar sliding-pair linkages: a inverted slider crank, b slider crank, c cllipsograph, d scotch yoke, e rotating scotch yoke (Oldham coupling), f sliding-pair loop.
Four·Bar Sliding.Pair Linkages. If turning pairs are replaced by sliding pairs this results in sliding-pair chains and mechanisms. Loop motions occur if the sliding joint connects two moving elements. Three chains (Fig. 6) are derived from the four-bar linkage (kinematic chain of each four-link mechanism): chain I with one sliding jOint, chain II with two adjacent sliding joints and chain/II with two diagonal sliding joints. As a result of kinematic inversion (link inversion and change of fixed link), the three chains yield six four-har sliding-pair linkages. Each sliding pair causes equal angular velocities, regardless of the mechanism dimensions. e.g. in the case of chain I, w 12 = Wn and W ,4 = W , •. The following is generally applicable: w ij = Wjl is the angular velOCity of the link i relative to the link j. Sliding-pair mechanisms are therefore sometinIes transmissions which transform speed uniformly
(constant transmission ratios).
Multi·1ink Mechanisms. For each group of kinematic chains with the same numher of links and the same degree of freedom there is a clearly definable number of different chains and mechanisms. Figure 7 shows six-link con-
Kinematic chain
I Mechanism ,
variants
strained chains (F = I) based on Watt's and Stephenson's chains (variants as a result of changing the fIxed link) with a number of prdctical examples. If douhle joints are used a further five different chains can be added. The equations of synthesis (Fig. 8) yield eight-link constrained chains with two quaternary and six binary links, with one quaternary, two ternary and five binary links and with four ternary and four binary links. If multi-joint links are also considered there are according to Hain 60 different eight-link constrained chains, yielding a total of 330 mechanisms as a result of kinematic inversion. Cam Mechanisms. The standard cam mechanisms are three-link cam mechanisms consisting of the cam, follower (rocker) and frame. The cam and follower touch at the cam jOint (point of contact K) - in many cases an additional tracer element, e.g. a follower-mounted roller
with an identical degree of freedom, improves the operational properties without changing the kinematics: the frame links the cam and follower [4 J. Generally, the frame is the fixed link 1, the cam is the input memher 2, and the follower is the output member 3.
Examples
Figure 7. Six-link constrained kinematic chains and sample mechanisms (l Watt chain, 11 Stephenson chain),
,.,#"
Mechanical Machine Components. 9 Kinematics
Kinematic chains
Examples
2· n4 ~ 8 6.n2~12
820 I. n4 ~ 4 2.nJ ~ 6
5.n 2 ~IO
820
4.nJ~12 4.n2~
8
8 20
Figure 8. Six-link constrained kinematic chains and sample mechanisms.
All three-link cam mechanisms can be derived by means of a /ixed-llnk change from the three-link cam-pair chain with tuming-and-sliding pairs, which in tum originates from a corresponding four-link chain (equivalent chain) (Fig. 9) [5]. In this equivalent chain a binary link connects the centres of curvature of the cam and follower or tracer element which are currently correlated at the point of contact K. The "three-pole theorem" that is well known
I
~
II
~~~/-23
L,
1
12
f
I
\
13
II!
"~ i/ _£J' 23 {"i'-
t7"-- ...- \
.--"'-
in kinematics states that the relative motions of three links i,}, k (random link numbers) are determined by the three instantaneous centres of velocity ij, ik and}k (double and multiple joints represent degenerated pole line sections in one point) that lie on a line (pole line). This theorem is of particular significance for cam mechanisms, both for systematic classification (equivalent mechanisms, sliding and rolling cam mechanisms), for analysis (velocity
""y
1
12
~ l'
3
-
=
(
/
T
13=
1z:t:'L 1 ~
_--.L
3
.....'Co
I
-
P\ G)\ ~ r;;I 5 ~ bj)s hl~ k0' cIJ ~~ a
d
g
j
m
77777'""/7/771/
f
7?7?/T/777?7;
n
e
c
p
i
I
0
~§~ ~ ~ ~ ~ ~ l~ Figure'. Systematics of three-link cam mechanisms with tuming·and·sliding pairs.
9.2 Analysis of Mechanisms. 9.2.1 Kinematic Analysis of Planar Mechanisms
determination) and for synthesis (determination of principal dimensions). In general, each chain with turning pairs and at least four links produces cam-pair chains if a binary link is replaced by a cam jOint. If the connecting joint between this binary link and the adjacent link is a rotary joint [6, 7] the concomitant cam is a dosed cam about which full rotation takes place; if an oscillating link is present, only a partially tracked cam (slotted link) can be provided which permits ahead and reverse rotation of the follower in the slotted link. The interchangeability between chains and mechanisms with turning and cam pairs (equivalent mechanism theory) is applicable as far as the acceleration stage of kinematic calculation methods (see F9.2). In general, there is sliding and rolling of the contiguous links in accordance with the two degrees of freedom with a (planar) cam joint; most cam mechanisms are therefore sliding cam mechanisms. In the special case of roller cam mechanisms pure rolling occurs in the cam joint because the instantaneous centre 23 in a three-link cam mechanism (Fig. 9) is coincident with the point of contact K. Tooth gear mechanisms with two meshing cam flanks are easily classified as sliding cam mechanisms.
9.2 Analysis of Mechanisms 9.2.1 Kinematic Analysis of Planar Mechanisms Graphic-Computational Method
Correlation of Locations. With linkages in general and four-bar linkages in particular. it is important to specify certain relative locations for two mechanism links. This coordination is known as a "zero-order transfer function ". In the case of a slider crank with kinematic offset e, the instantaneous location of the sliding block c as the output link is correlated with the location of the crank a as the input link as a function of the cam angle cp (Fig. lOa): cos cp + \b 2
-
(a
sin cp -
e)2
0)
For the inverted slider crank (Fig. lOb), the location '" of the sliding bar c characterises the relationship to the location of the crank a
'" = "" + arccos(e/m').
( 4)
The following applies in the case of a four-bar turning-pair linkage as shown in Fig. 10c:
5
mol + c2 - h 2 ) '" = "'. - arccos ( - - 2m*c ---.
(5)
The following apply for Eqs (4) and (5):
"" = 180 0
-
arccos (
d - a cos cp)
m*
and
m*
= 'la' + d'
- 2ad cos cpo
Velocity State as First-Order Transfer Function. In the case of the slider crank (Fig. lOa) the sign-oriented (directional) "radius of translation" m represents the velociry v. of the sliding block relative to the crank's angular velociry w, (6)
The radius of translation as the first-order transfer function (UF I) of the sliding block can be measured perpendicular to the sliding direction as the distance between the relative pole Q and the crank pivot Ao. For the inverted slider crank (Fig. lOb) and the fourbar turning-pair linkage (Fig. lOe) ahd UFI of the link c is expressed by the angular velocity ratio w,) Wa or reciprocal transmission ratio l/i with the pole pitches q, and qb: UFl = wjw, = d",/dcp = l/i = q,/qb'
(7)
The pole Q corresponds to the pitch point of two meshed gearwheels and can lie both inside (external gearing) and outside (internal gearing) the path AoBo.
Transfer Functions of Four-Bar Linkages
s = a
1.ld'
Acceleration State as Second-Order Transfer Function. The second-order transfer function (UF2) can be determined with the aid of the collineation angle A and UF!. Kinematic derivation is based on the law that the velocity of the relative pole Q on the linear frame extension AoBo is a measure of the acceleration of the output link c. The following equation applies for the sliding block of the slider crank (Fig. lOa) with A as the angle between the coupler b (on the inverted slider crank between the normal to the sliding direction) and the collineation axis k as the link between the two instantaneous centres P and Q:
DF2
= d 2s/dcp' = UFi/tan A.
(8)
In the case of an inverted slider crank and a four-bar turning-pair linkage, the following applies for the UF2 of the link c (Fig. lOb and c): UF2 = d 2 "'/dcp2 = UF1(l - UFl)/tan A.
----.----1
a Figure 10. Geometric principles of transfer functions: a slider-crank; b inverted slider-crank, c four-bar turning-pair linkage.
(9)
'MItAl
Mechanical Machine Components. 9 Kinematics
With the aid of the transfer functions it is possible in turn to determine the acceleration an of the sliding block or the angular acceleration etc of the link c of an inverted slider crank and four-bar turning-pair linkage (10)
The angular acceleration of the crank a is designated a,. The rotating inverted slider crank and the rotating fourbar turning-pair linkage can be used for two different principal motions, i.e. to generate oscillating and rotating drive motions. Oscillating (d> a + e) and rotating (d < a + e) inverted slider cranks and four-bar turningpair linkages are available as crank-and-rocker and doublecrank mechanisms. Oscillating inverted slider cranks and crank-and-rocker mechanisms are used for reciprocating motion, and rotating inverted slider cranks and double cranks for non-uniform rotation, e.g. as compound mechanisms [3,8].
mechanism with n links and g joints with a degree of freedomf= 1 to be [10]
p =g For
p=
'P,n
= 'P2 = 'P;
x + iy
=
rexp(i'P), i = v-I,
The deviations dr, and/or d'Pj of these estimated values from the exact values are calculated iteratively as unknowns in a linear equation system until they no longer exceed a specified positive value. Then ran or 'Pan is inl:reased by an increment, where the previously iterated location of the mechanism serves as the new estimated location etc. [9]. The iteration is based on the '"closed conditions'" of the polygons or loops replacing the mechanism, from the complex number Zj:
L
(Zj) =
L
[r, exp(i'Pj)] = 0; k = l(l)P (13)
;=-\
(summation of m joint spacings). Equation (3) must be repeated p times. The number p of independent loops is derived independently of the degree of freedom F of a
B"
y
(drive equation),
+ r, exp (i'P,) - r8 exp (i'P8) - Ir, - r6
r7 exp (lip-.)
+ r, exp (I'P,) - r 4 exp (i'P4) - Ir, - r6 = O.
= 0,
With the angles f3, and f34 constant, 'P- = 'P2 + f3, and 'P8 = 'P4 + f3•. The lengths r, are also constant apart from r6 and are specified like 'P.n' The closed conditions mean that 2p (real and imaginary parts) transcendental equations are available to detertnine the same number of parameters of location. A Taylor's series expansion for (IS)
which only takes account of the first-order series terms, results in the following iterative specification after insertion in Eq. (13): dr,n or d'P.n
(II)
(12)
becomes
r, exp (i'P,)
k
1Il
} = I
this
Z; = Zj + dz"
then describes the connecting line between two pivot points. Initially a specified original position for the driving member(s) - r = r. for a reciprocating cam as drive element and 'P - 'Pan for a crank as drive element - is postulated together with appropriately estimated parameters of location (paths rj and/or angIes 'Pj (expressed in ("'adians) It)r the other links from
ek =
(14)
I).
the mechanism in Fig. 11 7 - (6 - I) = 2 and consequently
Loop Iteration Method The structure of the mechanism to be examined is inserted in the complex number plane (Fig. 11). The complex number Z =
(n -
=
=0
(drive equation)
(16a)
(16b)
1 (1)p.
A linear equation system is derived in this way from the real and imaginary parts of Eq. (16a) and (l6b). (17) with a (2p + 1) • (2P + I) matrix of coefficients K for the components of the correction vector de which contains the deviations drj and/or dcpj' where j = 1(I)m. After each iteration step there is an improvement in the initial approximation b - cOfllprising the real and imaginary parts of the complex sums e k in Eq. (13) - in accordance with Eq. (12). The sums e k (control option and termination criterion) disappear for the accurately calculated mechanism location. The value of the determinant of the matrix of coefficients K must be continuously monitored. If Eq. (17) cannot be resolved, either a closed condition has been violated or a special position of the mechanism with poor transmission properties in respect of motions and forces has been reached. A change of determinant sign indicates a change in the installation location. To determine the velocities and accelerations the closed conditions - Eq. (13) - are derived once or twice as a function of time. This generates two further linear equation systems with the known matrix of coefficients K which now only have to be resolved once (18) and (19)
e
The vectors e and contain the velocities rj and/or CPj' and the aC<.:derations r, and/or ---- ------- [5 - - - Figure 11. Six-bar linkage with multiple power transmission paths (F = 1): 2 input crJ.nk, 6 output redprocating cam.
Module Method This method has shown itself to be particularly user-friendly for mechanisms comprising "double hinges" (two
I.'M
9.2 Analysis of Mechanisms. 9.2.1 Kinematic Analysis of Planar Mechanisms
jointed binary links) with tuming-and-sliding pairs. A further condition for this is that the drive variables (travel or rotational angle, usually relative to the ftxed link) are available as fimctions of time. The structure of a constrained eight-bar linkage (double press) shown schematically in Fig. 12 contains the rdatively simple kinematic nlodules "input crank" AoA "double hinge with three turning pairs" A'CCo , CoCC, A'AoA" and "douhl~ I,
Yj
0
1~~''11----__
a
hinge with output reciprocating cam" CD, A"B [11-13]. The output variables A (coordinates x, y of a link point P and angle w of a link with temporal derivatives) of a module are either variable inputs EV for the subsequent module or final results. Constant input variables EK represent, for example, link point pitches I, crank radii r, static offsets v and location parameters K. A tenlary link with three turning pairs (links 2 and 6 in Fig. 12) can be seen as being derived from a double hinge with three turning pairs.
~
n
x
o
Yl.A-\K"~ ~
b
K"~
..
.....
_---x EK: l,v,K,P],A, EV : P,
"Input Crank" Module. Calculation of the coordinates x, Y in m with temporal derivatives .Y, _V in Ill/s and X, y in Ill/s' of the link point P when the following are given: coordinates X o, Yo in m of the tixed link point Po, the crank radius r in m, the angular location w in degrees or fad, the angular velocity u) in rad/s and [he angular acceleration U' in rad/s 2 ; see Fig. l~a. r, Po IX",Yol: W[u.',
Input Output
Pix, y, x,
."i:
=
X
= (Yo
Y=
(Yo - Y )U'. j'
=
Y EK : EV :
P, ,PI A: W
_v, X, yl
+ ,. cos(w). y = Yo + r sin(u').
= Xo
A: P
u. , u']
Calculation method:
x
EK (,po EV: W A P
(20) (21 )
(x - xo)U'.
- Y)U'
T
(Xo - x)li".
(22a)
(x - x,,)u'
T
(Yo - y)u."'.
(22b)
Hi
d
Figure 13. Planar kinematic modules: a input crank, b double hinge with three turning pairs, c double hinge with output reciprocating cam, d link with two turning elements.
"Double Hinge with Three Turning Pairs" Mod· ule. Calculation of the coordinates x, y in m with temporal derivatives x, yin 111/S and x. yin m/s2 of the link point P when the following are given: coordinates Xl')'l' in m with temporal derivatives '\"I')'!. j:2,)'2 in m/s and Xl, YI. 'Y2, Y2 in m/s.!, the pitches II, Ii. in III of the ftxed link points P,. P, and the location paramet"r K (K = + 1) if the order of the points P,P,P is oriented such that it is mathematiLally positive, otherwise K = - 1); see Fig. l~b.
X 2.Yl
Input:
Pix, y, .i:, .Y, .1:.
Output:
H"
=
(H,I/" - H,li,)/(2H,),
He
~
(H,H, + H,H,)/(2H,),
x=x,+H",y=y,+He, HI;', = X2
i'l' H9 = Yl - jll>
H\()
=
H o '- III'
1/"
=
H,fi",
H14
= H,H 12 /HI.">' H],;
_v]
Calculation method:
-
H]6
= X2
2
HII
= H~
- H2•
+ I/,j-f",
=
-116HI2/HI!,'
+ HI·p.Y = .i'l + HI';'
X = XI -
(23)
'\'1' HI"'
(24)
= Y2 - Y].
+ H"I/". X=
6
x,
+ (H"I/" + HeH'9 - H"I/'8)/1I",
y = y, -
(1/"1/,,
(25a)
+ H'P'9 + H,,J!'8)/1I,,· (25b)
"Double Hinge with Output Reciprocating Cam" Module. Calculation of the coordinates x, y in m with Figure 12. Eight-bar linkage (F = L) assembled from simpk modules. 2 input crank. 4 and 8 output reciprocating cams.
temporal derivatives X, y in m/s and X, y in Ill/s' of the link point P when the follo\\ing are given: coordinates XI!
1.G11
Mechanical Machine Components. 9 Kinematics
y,
in m with temporal derivatives X" y, in m/s and x" y, in m/s2 , the sliding lines and direction with the coordinates x" y, and x" y, in m of two points P" P" the pitch I in m of the link points P" P, the static offset v in m (v > 0 if the order of the points P;P.P is oriented such that it is mathematically positive, otherwise v < 0) and the location parameter K (K = + 1 if the order of the points P,P is oriented such that it is mathematically positive, otherwise K = - 1); see Fig. 13c.
Input:
I, v, K, P, [x" y,], P, [x" y,];
P, [x"y"x"y"x"y,] P [x, y, x, y, X,
Output:
y]
Calculation method: H,
= x, - x" H2 = y, - y"
H,
= Hi + m, H, = H';~H3'
H.
= H,H7
H9 = H~
HlO X
-
H,H6
+ H~
- v2
x
+ v, -
t' + (2vH.) ,
= H,H6 + H,H7 , HI1 = K ~Hio
- H9 - H lO ,
= x, + H.Hll - vH" Y = y, + H,HI1 + vH,,(26)
+ H,H 13 , = (X,H'2 + y,H13 )/H14,
H14 = H 4 H'2 HIS
links, i.e. in a sliding pair the joint force acts perpendicular to the sliding direction, in a cam pair in the direction of the normal at the point of contact. It is further postulated that the input member moves at a constant velocity or angular velocity D. The input force or input torque required for this can be determined. The joint forces in the joint }k between two link elements} and k are always revealed in pairs by means of a section through the joint }k. If G;k represents the joint force from the link} to the link k, G;k = - Gkj applies both to the direction of the joint force as vector and to the components X;k and ljk in the x and y directions; see Fig. 14. In accordance with the three conditions of planar statics, the joint forces on one link k are in eqUilibrium with the other forces and moments acting on link k. These also include the inertia force - in components - mkxk and - m.jik - at the centre of gravity Sk (mass m k in kg), which moves in the x and y directions respectively with the accelerations k and Yk, and the moment of inertia - Ik'Pk (mass moment of inertia]. in kg m 2 relative to the centre of gravity) of the link rotating in the x-y plane with the instantaneous angular acceleration 'Pk' For a ternary input link labelled link number 2, mounted such that it can rotate in fIxed link 1 and connected to links I and m by means of rotating pairs and on which act, in addition to the inertia effects (centrifugal force only in this case), the input moment M,n, a supplementary moment M2 and an external force F2 at point P" the conditions of equilibrium for 'P2 = 'P,n = Dt (time t) as per Fig. ISa are:
X"
+ X 12 + X m2 = - m 2r2D2 cos
('P,
+
Y2)
- F2 cos (7,), (27)
Y"
+ Y" +
Ym ,
H'6 = (x,H, - y,H,)/H14,
- F2 sin
= (Hi6/1 - X,H'2 - y,H13 )/H14 , X = -H,H17 ,y = -H,H17 .
- X 12/21 sin ('P2) -
H17
"Unk with Two Rotary Elements" Auxiliary Module. Calculation of the angular location w in degrees or rad, the angular velocity W in rad/s and angular acceleration in rad/s 2 of the mechanism link when the following are given: coordinates x"y" X2,Y2 in m with temporal derivatives X" (.Y,), x2 , (.Y,) in m/s and X" (ji,), x2 , (ji2) in mis' of the link pOints P" P2 (value in () as alternatives); see Fig. 13d.
Y2)
(33)
(72 ),
x m,z2m sin ('P2 + /3,)
+ Y 12/ 21 cos ('P2) + Y m,z2m cos = - F2P2 sin (72 - 'P2 -
(28)
(32)
= -m 2r2D2 sin ('P2 +
('P2
+ /32) + Man
8 2) -
M 2·
(34)
w
Input:
P, [x"
y" x"
(y,),
x"
j
(y,)],
P 2 [X"Y2' x" (Y2), x" (Y2)]
Output:
k
W[w,w,w]
Calculation method: 1= ~(X2 - X,)2 W
= arccos
+ (Y2
[(x 2 - x,)/I] sign (Y2 - y,),
- 180° :s w :s + 180°, W
= (x2 -
a
- y,)2,
x,)/(y, - Y2)
= (y,
n I
(29)
- y,)/(x, - X2), (30)
w = [x,-x2 + w2(X,-X2)1I(Y2-Y') = Oi, - Y2 + w2 (y, - Y2)]/(X, - X2)'
(31)
9.2.2 Kittetostatic Analyses of Planar Mechanisms No account is taken initially of friction in calculating the forces transmitted in the joints between the mechanism
Figure 14. Forces in a frictionless jOint. a Turning pair, b sliding pair (sliding direction t), c cam joint (normal direction n).
9.2 Analysis of Mechanisms. 9.2.3 Kinematic Analysis of Spatial Mechanisms
I.Rlet
nents and the input variables, with the matrix of coefficients A, which can be reduced to a "core matrix" by deleting those columns that only contain one element other than zero and the related lines, and with the vector r, which largely comprises the known (given) forces and moments. 9.2.~
a
Kinematic Analysis of Spatial Mechanisms
A closed analytical representation of the kinematics of spatial mechanisms is only possible in individual cases [1417]. An iterative method - cf. F9.2.1 - based on spherical coordinates (spatial polar coordinates r i , ai' b i ) for each mechanism link} [10, 18, 19] in the vector form
yl
(40a) Xmk
with the length r i and the unit vector
ei
COS
(a;l
= [ cos
(ai)
(40b)
sin (a i )
b Figure IS. Forces and moments on ternary links with turning and sliding elements: a input link, b link in general motion.
is therefore recommended (Fig. 16a). The description of the spatial mechanism structure (example in Fig. 16<:) is based on the "vectorial equivalent system", Fig. 16d. The constant coordinates are the sizes, while the variable coordinates are the mechanism's location- and time-dependent kinetic quantities with temporal derivatives (velOCities and accelerations) which have to be calculated; the time-dependent input values ran or (tan or {jan' which have to be given in accordance with the degree of freedom F (Eq. (1», are also variable. The closed condition ( 41)
For a ternary link k subjected to general motion and connected to links} and m by means of turning pairs, and to which link I is connected by a sliding pair, the following equilibrium applies (Fig. ISh): X jk
+
(G[~ - G;~)
sin
= m,xk - Fk cos
('Pk
+
Q'k)
+ X mk
must be evaluated p times (P as per Eq. (14». The location of the kinetic axes (e.g. rotary, sliding and helical axes) relative to each other, which must be constantly maintained during motion, can be expressed either by scalar products
(35)
(Tk ).
(42)
(36)
- Xmklkm sin ('Pk
= mkr,
[jik cos ('Pk
- FJ>. sin (T, -
+
13k)
+
+
)'k) -
Ymklkm cos ('Pk
+
rj
13k)
y
xk sin ('Pk + )'k)]
'Pk - Sk) -
M, + ikifk'
(37)
In general, the angles and lengths cited are constant, with the exception of '1'"
z
= 2g, + g, + F
Allowing for Gkj = - G jk , X kj = - X jk and Ykj linear equation system
Yjk'
C
(39)
x
d
y
the
is derived for each mechanism position with the unknown vector x, which contains the joint forces or their compo-
:~
z
-.{
7777;
*r" I
----#f--
(38)
= -
x
a
X
Figure 16. Kinematic analysis of spatial mechanisms: a spherical coordinates, b basic vectors of crossing and intersecting axes of motion, c shaft coupling as example, d vectorial equivalent system for «:.
'M.
Mechanical Machine Components. 9 Kinematics
or by vector products
23
(43) where the shaft angle Ajl = const. (Fig. 16b). To this end, either the r vectors already defined in Eq. (14) are used or new ones are introduced, e.g. r 7 , in Fig. 16d. The evaluation of Eqs (41) to (43) is carried out iteratively with the aid of the Taylor's series expansions terminated after the first-order terms ej
= s. + B j ,u.1aj + ej,t3~f3i'
e j ."
= ae/a"j' e;.~ = ae/af3j,
J
2 12 /
to
9'1
a ( 44a) (44b)
If the exact values and are inserted in Eqs (41) to (43), a linear equation system can be constructed for the corrections 6,rj , 6,"j and 6,f3j of the estimated values e j and rj' The process is begun with an initial position for the input link and relevant estimates for the mechanism's kinetic quantities in accordance with the drawing or approximate calculation; the sufficiently precise iterated location provides the estimated values for the next location after incrementation of the input variable, etc. The values of the velocity and acceleration stage can be determined from the temporal derivatives of Eqs (41) to (43) which have been carried out once or twice.
9.2.4 Running Quality of Mechanisms The running quality of the mechanisms depends on the geometric and kinematic variables, design and material properties of the links and joints and on the interplay of forces and power flow in the mechanism [20, 21]. For planar mechanisms at least, the transmission angle and the dynamic properties are important parameters in respect of power flow.
Transmission Angle The divergence from the optimum 90° transmission angle is an indicator of the quality of motion transmission from drive to drive. The transmission angle is defmed as the angle I" between the tangent t, to the absolute path of the pivot point of the joint guide link [22] (output links mounted with bearings in the fixed link are always joint guide links) and the tangent t, to the relative path of the (transmission) link driving the joint guide link with respect to the input link. In the case of four-bar turningpair linkages this is also the angle 1"34 between the links 3 and 4 (Fig. 17a) when link 2 is the input link; in the case of a slider crank the direction 1434 is replaced by the normal to the sliding direction. Excessively small I" values indicate that there is a danger of jamming. Account may have to be taken of a number of transmission angles, which can only be determined with knowledge of the instantaneous centre configuration, in the case of multi-link mechanisms with multiple powertransmission paths. With the six-bar linkage (F = I) represented schematically in Fig. 7b I" \71 applies for motion transmission from the input link 2 to the output link 6; in the reverse direction, however, with link 6 as the input link and 2 as the output link the angles I" l'!?, I" \(~ and I"~(~ apply.
23
b Figure 17. Transmission angle: a four-bar linkage, b six-bar linkage with multiple power-transmission paths.
linkages, therefore, the transmission angle only has a limited significance in respect of the term "opposite fixed angle". High-speed linkages should be evaluated on the basis of their dynamic properties, taking account of the influence of both inertia and external loading [23,24].
9.3 Synthesis of Mechanisms The synthesis of mechanisms (dimensional synthesis) is used to identify suitable mechanisms for the given transmission and guidance tasks of points and link locations
[25]. This process uses an analysis of the systematic and of the geometric-kinematic properties of the mechanisms. Appropriate analytical programs are also employed alongside certain synthesis processes to determine suitable mechanisms using the "synthesis by means of iterative analysis" method.
9.3.1 Four-Bar Linkages Favourable Oscillatin", Motions for Transmission and Acceleration A four-bar linkage (Fig. 18) in the form of a crank-androcker mechanism converts rotaty motion to oscillating
80
Dynamic Properties The power flow during a motion period in a mechanism under load can change its direction continuously; with
Figure 18. Geometric principles of AIt's dead-centre positions for crank-and-rocker mechanisms.
9.3 Synthesis of Mechanisms. 9.3.1 Four-Bar Linkages
I.M
or five positions if two such curves intersect. Simpler procedures are available if the special positions are used [30]. Programmable calculators permit the calculation of the dimenSionally syntheSised curves without application of Burmester's theorem by means of automatic iterations [31, 32]. Further options are offered by point location
reductions [33].
/
/
/
/
Figure 19a Synthesis of a four-bar turning-pair linkage for given angular positions.
Example The three angles 'PIll 'PI~ and 'P14, are to be correlated with the angles !", !" and !" (Fig_ 1,). The angles A.U 'A. 11 • of which A 2 • 1 is coincident with AI' The circle drawn through the three points AI = A.l. " A.i", At .• yields as its centre point the pivot pOSition B I and thus all the dimensions of the mechanism sought in its position /. In the initial design phase
motion. The crank angle '1'0 is correlated with the oscillating angle 0/10' An infinite number of crank-and-rocker mechanisms is available for '1'0 and 0/10 [26]. 0/10/2 and '1',,/2 applied to d(AoBo = d) at Bo and Ao respectively give the intersection point R. The perpendicular bisector of A"R at M. intercepts BuR at M b . Circles with M,R = r, and MbR = rb are geometric loci for crank positions A, and rocker positions B, of a crank-and-rocker mechanism at the bottom dead centre position A"A,B,B" with an arbitrary angle (3. If d is assumed, the dimensions are
a
2r, cos (180 0
-
(3 - '1',,/2),
(45)
b = 2rb cos (1800
-
S - (3 - '1',,/2)
(46)
=
and
c = ~tP + (a + b)2 - 2d(a + b) cos (3.
(47)
(3 enables the determination of the optimum crank-androcker mechanism for transmission [27], the optimum adjustment means for transmission with variable '1'" and 0/10, the optimum six-bar series linkage for transmission and the optimum crank-and-rocker mechanism for acceleration with the smallest maximum acceleration for outward or return travel. There are a similar design and appropriate results for the slider crank in respect of the optimum dimensions for transmission and acceleration [27- 29 J.
Correlation of Angles The Burmester cirde point and centre point curves allow four (homologous) positions in a plane to be governed
b
x
Figure ~O. Synthesis of a four-bar turning-pair linkage for given coupler point positions.
Figure ~1. Notation in planar three-link mechanisms: a motion plan, b mechanism with follower roller, c mechanism with lever roller.
I.Gf:·
Mechanical Machine Components _ 9 Kinematics
a pivot point Bll i.e. a link length BoBI> can be selected instead of A \. together with other correlated initial angle pairs. With six-bar linkages it is possible to defme six or, under certain conditions, even eight correlated angle pairs with appropriately extended point position redul'tion.
Generation of Given PI.... ar Curves In theory a given planar curve can he generated in nine points exactly by the coupler curve of a four-bar turningpair linkage. To date, practical procedures for general locations are only known, as in the example below, for seven points. Example If five points E] to Ee, are given on a curve (Fig. 20) the perpendicular bisectors of the paths E JE4 and E2E3 intersect at Bo , from which an arbitrary ray Xt) emanates. The rays Xl> XJ. are marked on this such that with Xo they include the angles t/114/2 and "'H/2, which are formed by the perpendicular bisectors and BoBl and BoB2' EIAI = E~J. can be marked off with AI on XI and A2 on X 2 at an arbitrary equal length. The perpendicular bisector of AlA]. intersects xt) at Ao, and the circle about Ao can be drawn through Al and AJ. on which A,\, A 4 , A" are located as intersections of the circles about E3 , E4 , B", with EIAI as the radius. The points B02 and Bo" are found with M.A.B n2 = tlE2"12BU' MIAIBo" = tlEJJJJo. The equivalent would occur with the points A3 and A4 to give B03 = B02 and B04 = Bo as point position reductions. The circle through the three points Bo = B 04 , B02 = B03 and B o" yields its centre point as the point position Bland thus the mechanism sought in its position J. Other E points can also be paired initially, thereby achieving a different intersection Bo. Since the ray Xo and the lengths EIAI can be arbitrarily postulated the coupler curve can be made to pass through seven E points also. 9.~.2
Cam Mechanisms
Three-link cam mechanisms with the frame as the ftxed link [4 J are generally used to generdte periodic motions with dwells (output link standstills) and transitions with favourable acceleration properties. The task assigned within the overall machine system determines the "motion plan" (Fig. 21a) of a cam mechanism with motion broken down into individual steps ik. Thus the drive motion 5 for a follower roller (Fig. 21b) or 1/1 for a lever roller is a function of the drive motion 'P (rotary angle of the cam plate). A lever's rotary angle 1/1 expressed in radians can be converted to a follower stroke with the aid of the relation 5 = 11/1 (lever length I = B"B). With the exception of the dwells, a "motion law" is assigned to each motion step in a standardised wording, i.e. relative to
the partial stroke 5ik = 5k rotational angle
(5 - 5J/5ik
= J;k
or
5i
'Pi [34, 35]
[('P -
'/fik
= I/Ik -
-
'Pi)/
I/Ii and partial
= fez).
(48)
The functions fez) are mainly exponential functions fez) = Ao = A7+ A27 + ... + A.z2 or trigonometric functionsf(z) = A cos (vz) + B sin (vz) or combinations of the two. The boundary values of the derivatives according to the rotational angle 'P or the first-order or secondorder transfer functions at the locations i and k determine the type of motion task, and it is essential that they be assimilated shock-free (no jump from 5' or 1/1') and jerkfree (no jump from 5" or 1/1"). Further quality criteria arise from the maximum values of the following derivatives of the standardised laws in accordance with z:
C = max (f'l) C, = max (If']) Jerk coefftcient Cj = max ([("'I) Static torque coefftcient CHstal = C,,' Dynamic torque coefftcient CHdyn = max (f]'I). Velocity coefftcient
Acceleration coefftcient
The smallest values of the selected function fez) are the optimum in each case. There is an infmitely large number of cam profiles for a specifted motion task, of which the optimum for transmission purposes (minimum deviation of transmission angle m from 90 0) can be identifted. The "principal dimensions" for this are, in the case of a cam mechanism with roller follower, the offset e and the radius of the "basic circle" R Gmin of the roller centre-point path or the "basic stroke" SGmin or, in the case of a cam mechanism with a roller lever, the lever length I and the "basic angle" I/IGmin or 1/1' between the lever and the ftxed link.
9.4 Special Mechanisms For details of special mechanisms to achieve particular
motion tasks, sometimes with unusual design boundary conditions, please refer to the specialist literature and the relevant VOl Codes: spatial linkages and cardan shafts [36-40], spatial cam mechanisms [ 41, 42], indexing mechanisms [43-46, 49J and wheel crank mechanisms which are combinations of linkages and at least two wheels for rotary-dwell motion and pilgrim-step motion [47-49J.
Crank Mechanisms K.-K. Kuttner, Berlin
A crank mechanism (Fig. I), referred to in kinematics as a slider crank (see F9, Fig. 6a), converts the oscillating motion of a piston J with a gudgeon pin B via the connecting rod 2 into rotary motion of the crank 3 with the webs, crankpin and crankshaft journal K and M, or vice versa. Its task is power transmission - when it is usually called a drive mechanism - and in control applications which also require special designs. Such drive mechanisms are used in piston engines, in presses and in hydraulic and pneumatic drives.
Designs The crosshead drive mechanism (Fig. la), rated power :5 1800 kW and speed :5 1000 min- I, comprises the cross-
head 4 for particular guidance with journals, piston rod 5 and piston J. The trunk piston system (Fig. Ic) is used for rated powers of :5 420 kW and speeds of :5 10 000 min - J. Special piston forms illustrated are the eccentric (Fig. Ie) for control systems, which was developed from an extension of the crankpin because of the small eccentricity, and the equivalent cam mechanism (Fig. Id) used for refrigerator compressors.
10.1 Kinematics The dimensions of the crank mechanism (Fig. 2) are determined by the crank radius r and the connecting rod length I or by the stroke 5 = 2r and the connecting rod ratio A = r / I. The stroke is the piston displacement
I.INI
10.1 IUnematics. 10 1.3 Piston Acceleration
Vz
Figure 1. a-e. Crank mechanism forms (broken line marks positions in operation)
Figure 2. Kinematics of crank mechanism: A = 1/3 withdidwn. A- x broken line: scalesforr=lmandw=ls 1
between the top (TDC) and bottom (BOC) dead centres. The motions proceed from the piston displacement x and the crank and connecting rod angles
10.1.2 Piston Velocity
The piston velocity, assuming 'I' = wit and Eq. (5), is
v
em, crankpin velocit), v/ and acceleration at,: n
= liT = wl(21T). =
Vz
em
(3), (4)
V
\1 - A' sin' 'P)/Aj.
(5)
=
(6)
+ (A'/l6) sin" 'I' + ... j.
Approximate values arc yielded if only the first three or two terms of Eq, (6) are used. XK
=
r[l - cos rp + (A/2) sin 2 'Pl,
XK~ =
r(l - cos
=
rw [ sin
Ai 15A') + ( -A+ -+-
2
8
=
sin
26'
3A'
+ 256 sin 6'1'
2
1 + ....
(10)
are yielded by Eqs (7) and (8)
rw (sin 'P + (A/2) sin 2'Pl. v KS = rw sin 'P.
(11),(12)
The errors
l' - v K = rw/207 and {J - VK~ = rwl6 for A = 113 decrease as A decreases. The velocity changes its sign at the dead centres and i~ a function of [he crankpin velocity V7 = rw. Its extreme values VnLlX = rw ,it + AI. are approximately f3 = 56.';0·A.
Piston Acceleration
The piston acceleration, assuming
a
=
w·c\zI = rw' [cos dIP
(
cos
+J32 +
f3
~jl cos' '£] sin 'I' cos'
f3
rp)
(7), (8)
Equation (8) applies exactly for the crosshead crank (see F9, Fig. 6d) or for A = 0 or without the term /( 1 - cos (3) in Eq. (5). The errors x = X K = rl200 and x - XK~ = rl6 for a large A = 1/3
decrease as A decrease~
llK
10.1.~
r[ I - cos 'I' + (Al2) sin' 'I' t (A'/8) sin4 'I'
(9)
2 ~I - A' sin'
Approximate values
r(1 - cos '1') + I( I - cos (3)
where A = rll, sin f3 = A sin 'I' or cos f3 = -II - A' sin' 'I' from the triangles BKL and LKlvl. If the expression under the radical is expanded in accordance with Taylor's ser· ies, then x
A sin 2'1' ) rw ( sin 'I' + - ,,~~~=~
Ai 3A') - ( 16 + 64 sin 4'1'
As shown in Fig. 2. the piston displacement is
'F + (1
rw-~'--
d'P
It follows from Eq. (6), taking account of the goniometric equations, that
10.1.1 Piston Displacement
= r[ I .- cos
sine 'I' + (3) cos f3
dx
= w- =
(1),(2)
wr = TIns,
=
dl
=
= 2slT = 2sn,
The crank displacement, velocity and acceleration have the period T = lin. act on the centreline and are positive from B to M.
x
dx
= -
, [
=rw
cos'P+A
co,s 2'1' + A' sin' 'I' ] , . Ij(\ - A' sin' '1')'
(13)
From Eq. (10). harmonic analysis of the acceleration yields
'Mt4,' a
Mechanical Machine Components • 10 Crank Mechanisms
= rw' [cos
cp
(~ + ~f) cos 4cp + ;~~ cos 6cp ~ ..
-
Approximate values aK
A' 15A') + ( A +4 + 128 cos 2cp
= rw 2
(cos
.J
(20)
(14)
Oscillating Forces (Fig. ~). Substituting the acceleration a from Eq. (14). the forces of the first, second. fourth. sixth etc. orders are given for the multiples of the crank angle cp.
are yielded by Eqs (11) and (12).
+ A cos
aK;. '--- rw 2 cos '1'.(1"), (16)
Zq:),
The maximum errors arc a -- a K = r(,l/50 and a - aK~ = rwl/2.83 for A ~ 1/3. Curve (Fig. 2:). Accelerations achieve the exact values according to Eq. (1.» at the TDC (OT) ('P ('P~ 180'. f3 ~ 180').
~
0'.
f3
~
0') and H])e (liT)
where a K .OT is always the maximum, a K . trT is only the minimum, however. for A::s l/4. Where A> 1/4 there are two minima laK mini> a K . lIT symmetrically to BDC. Acceleration is a function of the crankpin acceleration in accordance with Eq. (4). Its zero point is at the maximum velocity.
Fo
= m" rw'
2: [(A) cos kcp = 2: F k= I
k
= 1, 2, 4, 6, 8,
k•
k=]
.... n; F,
= m"rw2 cos
cp:
A' 15A') F" = mJw' ( A + '4 + Us cos 2cp;
FYI
=
(21)
9A'
128 m.,rw' cos 6cp.
These periodic forces counteract the piston acceleration in the centreline. They are positive from the crankshaft to the piston [2]. Sufficiently precise values are yielded by a K as per Eq. (15).
10.2 Dynamics Periodically fluctuating fluid pressure. inertia. weight and frictional forces occur in crank mechanisms. The fluid pressure or primary forces depend on the working process involved. with whose period T, = aTln they change. and on the piston area. The number of cycles is a, = 2 for four-stroke motors. otherwise aT = 1. These forces are transferred through the machine frame and drive mechanism and transmitted as a rotary force at the crankpin K and as a torque to the crank MK (Fig. la). The inertia forces of the drive mechanism or the secondary forces are proportional to the square of the angular velociry. have the period T = lin and are transmitted from the drive mechanism to the hase and the environs, where they act as vibration exciters. They are therefore of great importanct". although their mean value per revolution is zero.
The weight forces are negligihle at high speeds. The frictional forces depend on many influences, e.g. working pressures, drive mechanism masses and condition, finish and lubrication of bearings. Experiments must therefore be conducted to identify them.
10.2.1 nuid Pressure Forces With trunk pistons or simple disc pistons (Fig. Ib) of areaA K • the fluid enclosed by the cylinder. case and piston exerting pressure p and the constant air pressure p, on the rear of the piston generate the fluid pressure force (18) This acts on the piston and drive mechanism. An opposite force of equal magnitude is exerted via the case on the cylinder and frame. With double-action machines (Fig. la) the forces
Fo FH
= F, + F,,; F, = morw' cos cp; = Amorw2. cos 2'P.
I
Although the deviation of the approximation is only 0.46% for A = 1/3 and cp = 0, the more precise values are significant in the case of resonance of poorly damped oscillations [3]. The amplitudes or the magnitudes of the vectors rotating at angles cp or 2cp are
Their projections to the centreline yield the forces F, = P, cos cp and F" = P" cos 2cp. The extreme values of F, are ± PI with 'P = 0° or 180°, i.e. at the top or bottom dead centres, while those of FH are ±P" with cp = 0°, 90°, 180 0 , 2 7 0 0 and 360 0 . F, forces are zero at 900 and 270'. and F" at 45', 135', 225 0 and 315'. The Fo forces reach the maxinmm P, + P" at TOC and the value - P, + P" at BOC which is at a minimum at A < 1/3.8.
Rotary Forces. F, = m,rw' are centrifugal forces at the rotating crank of constant magnitude but changing angle 'P. Their locus curve is a circle, and their components are i'~ cos cp ~ F, in the centreline and F, sin cp perpendicular to this. Equilibrium (Fig. 4b) is achieved by means of balance weights on the crank webs opposite the crankpin with the centre of graviry SG' the radius r G and the mass me; = m,rlr" [1. 4] (on multi-cylinder machines. see Fl.l).
BDC
TDC
thus in total i'~
=
(Pos - PKS) Ans
+
(PKS -- P,) As.
(19)
act in addition to the piston rod force Fs , = p,As< on the case or crank side DS and KS respectively.
10.2.2 Inertia Forces
Forces. Where there is oscillating and rotary motion in the drive mechanism with masses mo and m, the following forces occur:
(22)
...
'PI-I I
12
4 6
a
~t:
360' 6 18
'FJ 10 12 14 16 18 20 22 24 b
10 14 8,12,16
Figure 30. Inertia forces (oscillating): a vectors, b time curve, locus curve.
c C
10.2 Dynamics. 10.2.4 Forces in Drive Mechanism Components
l.t4.
Piston. The piston force acts at the centreline (Fig. Sa) FK = Fs - Fo·
(27)
The direction from B to M is positive, For simplicity's sake, the piston absorbs the entire inertia force Fo; frictional and weight forces are disregarded, Its curve, especially the extreme values, is a function of the working process, the speed and the Fo,m;,,1 Fs,m= ratio, Zero positions are at Fs = f:,.
Gudgeon Pin. The piston force FK is divided here into the connecting rod force F" at its centreline and the normal force FK perpendicular to the cylinder centreline.
Figure 4. a and h. Masses and balance weights,
f~,
Masses Connecting Rod (Fig. 4a). This moves with oscillating motion at the gudgeon pin B and with rotary motion at the crankpin K With the rod length I and the distance r" between the point K and the centre of gravity S through which its mass m" passes, then
Crankshaft (Fig. 4b). To simplifY the centrifugal force calculation, the crank webs with the mass mw and the distance rw of their centre of gravity Sw from the rotational axis M are reduced to the crank radius r. Since the centrifugal force cannot change during this, then (24)
Osdllating Mass. Assuming the masses m K for the piston, mKS for the piston rod, m K , for the crosshead and m o ," for the connecting rod as per Eq, (23), then (25) for the crosshead or trunk piston system.
Rotating Mass. With the crankpin mass m z, masses
(28), (29)
Crankpin. The connecting rod force is divided here into the radial force FK in the crank direction and the tangential force I';, perpendicular to this, In accordance with Eqs (28) and (29), then
(23) follows approximately as in the calculation of the bearing forces, For rods of the same design A = r I I and r,,1 I are constant. Under the usual conditions of A = 114 and r,,11 = 1/3 then m o ," = m s,/3 and m,·s, = 2ms,/3 apply.
= FK/cos f3, FN = FK tan f3
Zero positions (Fig. Ga): for FK = 0, F" = FK = 0; furthermore, FK = 0 at f3 = 0" or ep = 0°, 180 0 and 360°,
Fr
= FK sin
(ep
+ f3)/cos f3,
(30)
FR
= FK cos
(ep
+ f3)/cos f3,
(31)
The positive tangential force acts in the direction of rotation.
Torque. The changes in Md = Fl'r cause major speed fluctuations which are balanced by flywheels (see J1.1), There are zero positions at FK = 0 and at ep = 0°, 180° and 360" etc., thus at f3 = 0°, These are the TDC and BDC in which an engine does not start up automatically, The points where f~ = 0 and FK = Fs (in Fig. Gb at ep = 76.4° and 283.55° for A = 1/3.8) and the mean torque are independent of rotational speed since the inertia forces, thus also their mean tangential force, are zero (on multi-cylinder machines see J1.1 ). Housing (Fig. Sb). This absorbs the drive mechanism forces, On the case, Fs acts on the cylinder and F. on the guideway. In the frame the bearing absorbs F" = as per Eqs (27) and (29). The moment Md = F~a = FTr is absorbed by the frame anchor bolts or used for torque
10.2.3 Resultant Fon:es
measurements in motors with self-aligning bearings. At the cylinder centreline the force Fs on the case is countered by the force F., To ensure eqUilibrium, the base must absorb the rest, i.e. inertia forces r This also applies to Fn which is not contained in F K •
The resultant of the fluid pressure and inertia forces f~ and Fo has the period aTln, It passes through the drive mechanism and housing, i,e. through the case, cylinder and frame; see Fig. Sa, b.
Although the inertia forces relieve the drive mechanism components at TDC at high rotational speeds, they are
m red and m r . Sf as
mr
p~r
Eqs (23) and (24), then
= mz +
rnw(r/rw )
+
mr.~t·
(26)
f:
10.2.4 Fon:es in Drive Mechanism Components
~ST
[,KW r"
FrSI
~ZFKl C
a
Figure S. Forces in piston engine: a drive mechanism, b housing, c individual components.
'4"..1
Mechanical Machine Components a 10 Crank Mechanisms
oI::*. Expansioo+
Load change
r--,"Combustion
+ CompressionI
o
I
I
0'\
I
I
__
/
o~~~~:~~ ~i~__++-HI I I Fk I II I o I II
+ Ff..
F.L = -F" = -~Fi
(33)
At the dead centres F.L = FK , since FN = O. At mc, F.L = morW' (1 + A) - Fs, and BDC FBL = m orw 2 (1 - A). In addition to F" or FI' and FR , the big end also absorbs F,." = m,."rW'. Thus
FKL = ~Fj
+ (FR -
At the dead centres here FT (30) and (31). Thus at mc
=0
F,.,,)2.
and FR
FKL = FR - Fr.;, = Fs - morW' (1
(34)
= FK as
+ A)
per Eqs
- m,." rW'
and at BDC, when Fs = 0 FKL = morW' (1 - A) - m,.St rW'.
Crankshaft. The forces - FKL and F,.I<. W = m K •W rw2 act on the crankpin. Where F, = F,.w + F,.s" it follows that (35)
3.0 2.5 r-
The crankshaft journal absorbs the opposed force which is
I
40 a
~A
n=O
I:
i
I
I
FM = - Fs
~ 1500 min·I-t---+--+--t-+--+-~
It,I/)OOO
I I
+ morW'
(1
+
A)
+ m,rw2 at mc and
FM = - morW' (1 - A) - m,rW' at BDC.
I
10.3 Components of Crank Mechanism
([\
I\
There is a very wide range of embodiments of the individ· ual components. Their designs are determined by the machine size, working process, medium, rotational speed, bearings, loads, materials and production processes [1, 6-8). 10.3.1 Crankahafts
0.5
\\,
: : \ 1/ 't) 1.0 I-H--+--t-+--f-\-l'~+-+--I
\
-( I
I
1.5 t-TI-t---+--+:-t---t---4I:\-+il / -f+---+---l 2.0 b O·
Y
I
90'
180' rp
270'
360'
Figure co. Force and torque curves of a two-stroke diesel engine. a Forces at n = 4500 min -', b torque with changing rotational speed.
strongly present at BDC where the fluid pressure forces are usually negligible (FSK - 0). Account is also taken, for this reason, of the masses attributed to the individual components (Fig. Sc) [5]. Piston. The resultant of the forces Fs, FN acting on the gudgeon pin as per Eq. (29) and FO.K = mKFo/mo, where m K is the piston mass, is (32)
At the dead centres, '" = fJ = 0 or '" = 180° and fJ = 0°, thus FN = 0 as per Eq. (29). At TDC F. = Fs - mKrW' (1 + A), and at BDC, if Fs = 0, then F. = m K rw 2 (1 - A).
Connecting RoeL The = mo.sr.Folmo act on Po = Fo,K + Fo.st • then
Fo,St
forces the little
+ F. end.
and When
The crankshaft (Fig-. 7) consists of the crank throws with the journals 1 running in the main bearings, the crankpins 2 for the connecting rod and the webs 3 to connect the pins and hold the balance weights 4. The coupling 5 transmits the torque while the shaft extension 6 drives the auxiliary systems. The mean distance between two throws is called the cylinder pitch a = (1.2 to 1.6)D where D is the piston diameter. Designs. With in·line diesel engines (Fig. 7a), which large forces bear, there is a journal between each pair of crankpins, thus with z cylinders there are z + 1 main bear· ings. With an even number of cylinders and small forces (Fig. 7b) oblique webs connect two crankpins each, resulting in 1 + z/2 main bearings. With fan-type machines (Fig. 7d), up to five connecting rods are positioned adjacent to each other on a crankpin, while with V-engines (Fig. 7c) there is also one crank throw per connecting rod. Throws with attached pins (Fig. 7e), Le. for two strokes, are also found with mechanical com· pressors. Overhung cranks (Fig. 7C) - used with very small high-speed machines - require very strong webs and two bearings because of their free design. Crankshafts (Fig. 7 a) made of case·hardening or tempering steel, usu· ally forged in several drop forges, often have hardened journals. With cast shafts (Fig. 7b) of malleable cast iron or cast steel the low material strength is balanced out by the design because of the ease of deformability during casting. Crankshafts which are too large even for hammer forging are fabricated, i.e. pins and webs are joined by shrink fitting. Plain bearings are easily split, as a result of
. l 'sw
10.3 Components of Crank Mechanism. 10.3.3 Pistons
'Mtd
0--1
~ . -'
a
......
, .,
- ,+
43 --1 2
a
b
~.
e~ lfYJ
Figure 7. a-f. Crankshaft designs for two-cylinder machines.
which the crankshaft (Fig. 7a) can comprise a single component. Roller bearings are usually not split because
of heavy wear at the die line. For this reason, the journals are extended beyond the crankpins or the shafts are fabri· cated. lO.~.2
Connecting Rods
Connecting rods (Fig. 8) which form the connection between the crosshead or piston and the crAnkshaft, consist of the little and big ends J and 2 containing the bearings 3 and 4 and the shaft 5. They transmit the rod force and are made of case-hardened or cast steel or light alloy. Because of their low tensile strength, grey iron rods are only admissible in single-action two-stroke engines. The outer points of the rod describe a violin·shaped curve when in motion. This curve defines the space which must be left for the engine unit in the frame and cylinder.
Designs. The simplest form with undivided ends (Fig. 8a) requires overhung cranks or fabricated crankshafts. Split big ends enable one-piece crankshafts to be used. Oblique splits (Fig. 8c) are found with the strong big ends needed in diesel engines because of their high ignition forces to enable the connecting rod to be with· drawn through the cyLinder bore. The big or little ends can be branched (Fig. 8d) with crosshead or V-engines. If the inenia forces are of major proponion, i.e. with highspeed units, the shaft is designed with an H or J cross-
d
c
e
Figure 8 . a-e. Connecting rod designs and forces.
section (Figs 8b and c) , the former of which ensures better merging into the big and little ends. Otherwise a rectangular, circular or oval cross-section, which is also easier to manufacture, is sufficient, although the necessary moment of inenia requires a relatively large mass.
Bearings. These are usually plain bearings comprising a suppon shell approximately 2 to 5 mrn thick with a coating 0.25 to 0.5 mrn thick. They are made of gunmetal for the gudgeon pin and white metal or high-lead bronze for the crankpins. Roller bearings for the gudgeon pin (usually needle bearings, with cylindrical roller bearings for crankpins) require undivided connecting rods with large ends, even if the rolling elements run directly on the hardened pins or journals or in the connecting rod ends. The bearings are lubricated through bores in the rod or crankshaft or, in the case of gudgeon pin bearings, oil is also injected. lO.~.~
Pistons
Pistons are made of grey iron, cast steel or light alloy. Their crown absorbs the fluid pressure forces . The skin ensures straight movement of the piston and provides a seal for the medium and lubricant. With gaseous fluids the piston is heated Significantly, i.e. it is subjected to severe mechanical and thermal stresses. The heat is dissipated by means of the piston rings. Because the piston is heated more than the cylinder, the piston skin must be ground to the appropriate shape or contain expansion-regulating elements to ensure that there is play in operation. Contact between the piston and the case is prevented by the axial play.
Designs Trunk Piston (Fig. 9a). With motors, the crown 1 often accommodates valve cavities or a combustion chamber or,
I.N'
•
Mechanical Machine Components . 10 Crank Mechanisms
I ..
8
9' . '6
~
~ '"
a
3
, ·4
II ig. ilE 2
2
0200-
I
b
c
.
I
00
a
II .. r-
080
lOP
[§Ijj[J
e
c
~ f
Figure 10. a-f. Piston rings.
d
e
Action. The compression ring (Fig. . lOa) is pressed via its internal face against the cylinder wall and via its flank against the opposite side of the groove by the medium, thereby ensuring a seal. The oil scraper-ring (Fig. 10e) distributes the injected oil to the wall from the inside of the piston via bores J and slots 2, while the edge 3 scrapes off the used oil which drains away via the bore 4.
Figure 9. a-e. Piston designs.
10.3.4 Strength Calculation with large motors, ducts for the oil cooling system. The skirt 2 is ground to an appropriate shape or contains the expansion-regulating elements to ensure that there is play in operation. It houses the floating gudgeon pin 3 held by circlips 4 in its bosses 6 reinforced Witil ribs 5. The piston rings 7 and the scraper rings 8 with their drainage bores 9 are located around the skirt circumference. The material used is, for example, aluminium with silicon, manganese and nickel admistures of low density ( = 2.85 g/cm», good thennal conductivity (= 1.25 W/(cmK)) and high alternate bending strength ( = 80 N/mm'). The high coefficient of thennal expansion ( = 2.5 per °C), however, requires particular design measures. Disc Piston (Fig. 9b). Its boss 1 accommodates the piston rod 2. It is designed for double-action machines and has two sets of piston rings. Several discs are screwed
together to fabricate pistons in order to utilise one-piece sealing elements such as carbon rings or grooved sleeves. In hydraulic pistons (Fig. 9c) , grooved sleeves 1 with back-up rings 2 seal the cylinder while the rubber ring 3 seals the rod.
Differential Piston (Fig. 9d) . This has up to three stages for compresso rs. With the exception of one stage, the piston surfaces are of annular design. The trunk-piston embodiment shown here is for a two-stage compressor with the first and second stages 1 and 2 [41.
Plunger (Fig. ge). The skirt 1 slides past the guide bush 2 of the cylinder 3. The packing 5 which is adjustable from the o utside by means of the gland 4 seals the plunger. The size of the seal area makes this design long and heavy, and it can be used only with self-lubricating and slow-running machines such as hydraulic pumps [41. Piston Rings. The special grey iron, rectangular cross-section rings (Fig. lOb) are slotted and sit in the piston grooves. Their contact face. either parallel or oblique, determines the leakage losses but must acco mmodate the expansion of the entire ring without seizing. Their axial play is negligible so that the grooves do not deflect. The rings are subjected to the greatest bending stress (= 400 N/mm2) when running over the cylinder surface. Their hardness is less than that of the cylinder walls in order that , as the easier to replace, they wear more rapidly. DIN 24 909 specifies compression rings for sealing purpo ses, with tapered rings (Fig. 10d) used when running the motor in and scraper rings (Fig. lOe and f) to remove excess oil. Their dimensions are standardised. The legend " TOP ~ on the ring must be positioned towards (he piston crown.
Calculations are carried out using the simple strength-ofmaterials equations for certain cross-sections during the design phase or using fmite elements (FEM) for the whole assembly in a concluding calculation (see B8).
Approximate Calculation. This refers to the locations of greatest stress and potentially dangerous cross-sections and is helpful in the initial design stage.
Crankshafts (Fig. 11). The crankpins and journals 1 and 3 are subjected to torsional stresses, while the webs 2 must also withstand tensile and compressive loads. The greatest stress comes from alternate bending in the fillet between the web and the crankpin towards the rotational axis. The load acting here is F = FR + Fe> where the radial and rotary forces can be c alculated with the aid of Eqs (20) and (31). The pitch line for bending caused by the
tangential force is also located here. The substantial fatigue notch effects can be reduced if the following conditions are observed for the radius of curvature p, pin/journal diameter a, web width b and height b: pia 2: 30.05, b/a = 1.2 to 1.8 and h/a = 0.3 to 0.5. Oil bores should be sited well away from the areas of curvature . Further improvements can be ensured by relieving notches, tapers, enlargements and ftlleted bores (see D 1. 5) . At a load F = FR + F" bearing pitch I and pitch e of the dangerous cross-section and the stress concentration factor <> (see DI.5), the bending stress is (36)
Connecting Roas (Fig. 8e). The ends are subjected to bending stresses like a beam mounted on two supports. In accordance with Eqs (33) and (34) , the maximum load
at TOC in trunk piston machines is FBI. = rnorw' (1
+ A) - Fs and
a Figu.re 11. Crank throw: a construction, b bending moments.
[0 ..) Components of Crank Mechanism. [0.3.4 Strength Calculation
Figure 12. Crankshaft half-throw of a car with divlsion into elements and lines of constant relative torsjonal stress (Daimler Benz AG).
These forces act on the ends in the direction of the shaft when morw' (l + ,\) < F,. In double-action machines the maximum for the crank side KS (see Fig. 1a) is at HDC. Here, however, the masses and tluid pressure forces expressed in FRL and FKL have the sanle sign." They are thus cumulative and are directed away fro111 the shaft. This results in larger ends and cases. Their lever arms are a 1 = d o/2 and a 2 = dj2 where do and do arc the internal diameters of the little end and big end respectively. The case tension bolts are tightened to ::0::: 2F and = 70\!-h of the yield point where their shaft diameter is = 80lYo of the thread root diameter. In the case of roller bearings. the cubic mean of FKL applks ['i I.
Pistons. The crown represents a plate subjected to bending forces with a uniformly distributed load from the pressure. This plate is damped at the edges in the case of a trunk piston and in the centre in the case of a disc piston. The skirt is subjected to intemal pressure stress like a pipe. The initial design of pistons for thermal machines is based on empirical values because the stresses involved are difficult to quantit)". Trials are then conducted to improve the performance of the finished pistons.
Finite Elements (FEM). Finite element analysis is used to calculate the tensile , compressive, bending and torsional stresses on drive mechanism components and their resulting dcfornlation, and to deternline their vibration properties. It also records the interaction of
TnICkress ao IMl
a
b
Figure 13. Connl:cting rod of a crosshead compressor (Borsig AG, Deutsche Babcock): a curves of equal stress, b division into dements
these stresses with the crankcase via the bearings, Several thousand volume elements with nodes are frequently necessary for each cOlnponent. However, this enables curves of equal stress and deformation to be derived for the entire conlponent. Because the method is time-consuming and expensive, specialised computer progranls such as "Nastrans" are available. and a combination ofFEM and CAD is advantageous for visualising the result. Simpler methods, such as the integral equation method, are also in usc. The elements and curves in Figs 12 to 14 afe greatly simplified.
Crankshafts. The torsional stresses in the half throw (Fig. 12) were calculated using boundary element modelling. The results are then transferred to the entire crankshali by a special program. nle relative torsional stress curves shown indicate a marked increase in the crankpin fillet. The interaction between the shaft, bearings and
0.486
0.126 0133
a
b
c
Figure 14. Solid-shaft piston, diameter, 115 mm, for a diesel engine: a diviSion into spatial elements, b mean stress s~ in N/mm~ with warm pistons, c wall expansion in mm (Mahle GmbH)
I ••
Mechanical Machine Components. 10 Crank Mechanisms
of the piston crown and = 150 °C on the underside of the skirt. The solid-shaft piston (Fig. 14) is divided into about 700 spatial elements with 4200 nodes for fmiteelement analysis. The gudgeon pin boss is the most severely stressed part during operation. The mean load at the lower ring groove is am = - 61 N(mm' with deflection a. = - 58 N(mm'; for its inner edge am = - 68 N(mm' and a, = - 55 N(mm2 In the piston hollow, am = - 30 N(mm' and a. = - 20 N(mm 2 The upper and lower edges expand by 0.486 mm and 0.133 mm, respectively. FEM enables very complicated piston calculations to be carried out, but is not suitable in the event of plastic deformation. Further aids are the strain gauge technique, photoelasticiry and holography, which only record mechanical loads.
crankcase with non-linear lubricating fLlm properties is the subject of further study.
Connecting Rods. The effects of fluid pressure and inertia forces are calculated. Peak stresses of 100 N(mm' were found on the tension bolt seats and 70 N(mm' at the upper bore of a large compressor connecting rod cut from sheet metal (Fig. 1~). Pistons. The stresses and deformations caused by fluid pressure and inertia forces as per Eqs (18) and (20) and by temperature are at their greatest on diesel engine trunk pistons. With peak pressures of up to 150 bar, mean piston speeds of 9 m(s and power outputs of 35 kW per cylinder, mean temperatures of = 400°C occur in the centre
Appendix F: Diagrams and Tables Appendix Fl
Appendix Fl Table 1 .. Basic, additional, and explanatory symbols for the description of welds according to DIN 1912
b AC
I Aangeweld 1 ....-eId
J
V·weId
l- I--
,
Semi V-wekI
5
Y-weid
Ii
Semi V-weld
"'",,10'''' "
51mbO!
~ V I~
{;;Q
Y
. _-
r'
~ I-'
9 Bull weld
~
=
FJetweid
~
b,.
SIotweld
Il Spol weld
u
~ V
8 Semi~
/I
51 m b~
I;
~ y
IQ
1I1us.l rol ,~1I
f
~ .JI..... I ~I ~ €f Q;;? I' Linear Se,nr' ~I
~
7
No DesignalOO
~-
~
Q iQ IQ
n
I Sleep V·weId ~S'BeP
15
Iii
Face wakl
,
! I
!
!
-
17 Su~~-eId
{;
JL
11 !-l
"'
19 FQId wekI
m
,0--
AdciIionaIS)1I1boI
-.........,-
Plane
--
C
~
c
E,~,na'Of\'
-
(e.g. MelI'O!id)
-ct
Bas~
b. _
l-
Moo.mlm9 weIib
,nd addiIionaI symbols ("-"",,pies) Illustration
Weld typo
DoubJe.Vwekl IX·weldl with
FjIot ~-eId
with
conca",su~
f Bon V-wekI wfh
'--:-:-
ExpIanaIoJy symOOl
Welds aI aroond
OO
e
,
symbols
Run and typo 01 w.~
l\i1h plane mace
Q
18 OO'lQue wekI
Iorm
V·~-eId
--
-0
,§'
V·weId
,
~ '. .... ~
Su~"",
Concave
plano surfa
. " Tho iliustrabCo is CI1iy III expt.m the f"", of the doutIe-sided cuMld we~ sun""", II Also r_ seam weld 'OIlere ajlpIlIIJIiate (•.g. bead weld)
"m••,
~I v
~ i -x ...
\~
~ ~ (lj) .'
'S.'
t1~
:IT
Mechanical Machine Components. 11 Appendix F: Diagrams and Tables
I.t"
Appendix Fl Table 2. Rules and standards for welds Area of application 1. Engineering
General calculation base
Weld calculation
Rules for the classification and building of
Section:
machinery
te(:hniques"~
~Welded
connections and welding
Also recommended:
DV 952: see under vehicle manufacture DIN 15018: see under materials handling 2. Vehicle manufacture
Loading and safety of railway vehicles
L
DV 952: rules for welding in private factories
(welded vehicles, machines and
equipmentb ) 3. ShipbUilding
Rules for the classification and bUilding of steel seagoing ships 1980 edition, vol. III, Materials - welding: chapter 1980 edition. vol. I. Classification rules for seagoing ships, building rules for hulls" 7. section I, Shipbuilding - welding rules a Classification rules for steel seagoing ships, refrigeration eqUipment. steel inland water ships, 1980 Edition
4. Tank construction
DIN 4119: Surface cylindrical sheet tank building from metallic materials
DIN 18800 T1: Steel construction
5. Pressure vessel construction
DIN 3396: Surface high-pressure gas containers: recommendations for construction, equipmt'nt, installation, testing, commissioning, and operation
DIN 18800 Tl: Steel construction
TRD (Technische Regeln Druckgase - Technical rules for high-pressure gases)d Instruction leaflet of the Pressure Vessel Research Group (AD~notes)d Section: ~Welded connections and processing of Rules for the classification and building of welds"a machinery for seagoing ships. section. containers and apparatus under pressure" 6. Boiler and boiler-tube construction
7.
Containers for inflanunable liquids
Technical rules for steam boilersd Rules for the classification and building of steel seagoing ships. 19HO edition: volume II, Machinery, electricaL refrigeration a volume Ill. Materials - welding: chapter 7, section 2, Engineering - Welding Rules a DIN 6608: Horizontal steel containers for underground storage of liquid mineral oil products DIN 6616· Horizontal steel containers for surface
AD notes
storage of liqUid mineral oil products
DIN 6618: Vertical steel containers for surfa<..'e storage of liqUid mineral oil products DIN 6625: Site-assembled steel containers for surface storage of heating and diesel oils Rules for inflammable liquids (VbF): see National Law 1970 p. 689. Technical rules for inflammable liquids (TRbF) 8. Gas Mains
9. Building Construction
DIN 2470 Part 1: Gas mains from steel tube with operating pressure up to 16 bar Part 2' Gas mains from steel tube with operating pressure over 16 bar DIN 1050: Structural steel: calculations and structural design DIN 4112: Portable buildings DIN 4114: Steelwork, stability (cracking, tipping. buckling) DIN 18808: Steel construction, frameworks of hollow sections under direct static loading
DIN 4] 13: Aluminium structures under direct static loading
DIN 2413: Steel tubes, calculation of waH thickness in tenns of internal pressure DIN 18800 Tl: Steelwork DIN 18801: Steel structures
DIN 18800 T1: Steelwork DIN 18808: Steel construction, frameworks of hollow sections under direct static loading Recommendations of the Institute for Building Construction. Notices from the Institute for Building Construction, 1972; 3(5): 1-6
'4w:'
Mechanical Machine Components. 11 Appendix F: Diagrams and Tables
Appendix Fl Table 2. Continued Area of application
General calculation base
Weld calculation
10. Bridge construction
DIN 1073 (DIN 18809): Steel road bridges: basics for calculation DS 804: Rules for railway bridges and other engineering structures (VEI)b
DIN 4101: Welded steel road bridges (additional DIN 4100 and DS 804) DS 804: Rules tor steel railway bridges and other engineering structures (VEI)b
11. Materials handling
DIN 4118: Materials handling equipment for mining
DIN 18800 Tl: Steel structures
DIN 4132: Craneways, steel frames: basics of calculation, structural design and eqUipment DIN 15018: Cranes, basics of steel frames DIN 4420: Working and safety scaffolding (in connection with DIN 18800 Tl, DIN 18808)
12. Scaffolding
Effective: DIN 18800 Tl: Steel structures, and DIN 18808: Steel structures, frameworks of hollow sections under direct static loading
aGemIa11ischer lloyd. bGerman State Railways. c[jght gauge railway design 1970; 14: additional vol. 2. ctyereinigung der Technischen Uberwachungvereine (Association of Supervisors).
Appendix Fl Table 3. Weld seams of various quality levels: DIN 15018 Weld type
Butt weld
Weld quality
Extra quality
Nonnal quality
Welding method
Root cleaned out, top buttwelded, sheet processed in stress direction, no end-crater
Root cleaned out, top buttwelded, no end-crater
Symbolic example
ci>lP
cl>
::i X
Testing and error-free finish Test method
Abbreviation
Distortion-free weld test on 100% of weld length. e.g. X-ray
P 100
As for extra qUality but with only pull-test with maximum CIz 2:: 0.8 x perm CIzD in range for pulsating tensile stress with maximum CIz 2:: 0.8 X perm CIzO in alternating range with maximum CIz 2:: 0.8 X perm CIzl) or with maximum CId 2: 0.8 X perm
P 100
CIdD
Double semi-Vweld with double fillet
Filler weld
Extra quality
Root cleaned out, throughwelded, kerf-free weld-run, processed as necessary
Normal quality
Width of remaining joint in the root to 3 mm or to 0.2 times thickness of welded component. Smaller value to apply.
Extra quality
Weld-ntn kerf-free, processed as necessary
Normal quality
:fJsLI::
"cL re~ &~
Distortion-free test of most important usual welds in samples of at least 10% of the length of each weld, e.g. X-ray
P
Distortion-free test for lamination and grain distortion in the weld region of the appropriate sheet transverse to tension plane, e.g. X-ray
D
Mechanical Machine Components • 11 Appendix F: Diagrams and Tables
,*.
Appendix Fl Table 4. Metlic ISO threads, coarse and fine series (as DIN 13 Pan 12, Part 12 Additional Sheets, and Part 28) Nominal diameter d
Coarse thread Pitch p
Fine thread
Core section Stressed section
A~
Pitch p
Fine thread (extra fine)
Core section Stressed section
A-,\
A, (mm)
5
6 8 10 12 (14) 16 (18) 20 (22) 24 (27) 30 (33) 36 (39) 42 (45) 48 (52) 56 (60) 64 (68)
Nominal diameter d
A,
(mm)
0.7 0.8
125 1.5 1.75 2
(mm)
7.7'5 12.69 F.89 32.84 52.:10 76.25 104.7
144.1
25 2.5 2.5
:\
3 35 3.5
45 4.5
, 5
5.5 5.5
6 6
]75.1 225.2 281.5 324.3 427.1 519 647.2 759.:1 913 1045 1224
1.\T' 1652 1905 2221 2520 2888
8.78 H.2 20.1
(0.;)
(0.7'i) (0.";)
36.6 'i8.0 84.3 1I'i 1'i" 19:' 24'i
:103 3'\:\
1.25 1.2';
1.'; 15 1.5 1.'; 15
4'i9 '561
694 81" 9"76 1121 1.\06 14"3 1 ",8 20:10 2362 26"'76
:100;<;
4
9.01 13.07 20.27 36.03 56.29 86.03 116.1 157.5 205.1 259.0 319.2 364.6 473.2 596.0 732.8 820.4 979.7 115) 1341 1 ;43 1834 2050 2384 2743 :) 127
Fine thread (fine 2)
Pitch p
p
Core section Stressed section
A~
Pitch
72 (76) 80 (85) 90 (95) 100 (105) 110 (11 ';) (120) 125 (130) 140
6
3287 3700 4144 47)4
6
;;,64
6 6
60:12
6740
6
7488
~-:'60
6 6 6
827.\ 9100 996; 10869 11813 1:1818
8';60 9400 10300 11200 12100 14200
(, (,
6
(0.35) (0.5) (05) (0.75) 0.75
333
384 496 621 761 865 1028 1206 1398 1604 1900 2144 2485 2851 3242
(mm')
1.5 1.5 1.5 1.5 1.5
1.5 1.5 1.5 1.5
10.02 15.12 22.79 39.37 64.75 91.15 128.1 171.4 221.0 276.8 338.9 385.7 497.2 622.8 762.6 916.5 1085 1267 1463 1674 1928 2252 2601 2975 3374
Core section Stressed section
Pitch p
4 4
134 178 229 285 348 401 514 642 784 940 1110 1294 1492 1705 1973 2301 2653 3031 3434
Core sectiun Stressed section
A_~
A,
4
10.6 16.1 24.0 41.8 679 96.1
Fine thread (extra fine)
A, (mm)
(mm)
.\463 3889 4344 4945 'i';90 6nO "000
6
6 6
9.79 14.5 22.0 39.2 61.2 92.1 125 167 216 272
A_~
A, (mm)
Core section Stressed section A,
A~
(mm)
Fine thread (fine I)
(mm)
Pitch p
3536 3970 4429 5038 5687 6375 7102 7869 8674 9519 10404 11327 12290 14334
3568 4100 4566 5190 5840 6540 7280 8050 8870 9720 10600 11 ';00 12500 14600
3799 4248 4723 0352 6020 6727 7473 8259 9084 9948 10852
11795
12777 14859
3862 4315 4794 4530 6100 6810
7560 8350 9180 10 100 11000 11900 12900 15000
'4':t,.
Mechanical Machine Components. 11 Appendix F: Diagrams and Tables
Appendix Fl Table S. Nominal values for metric ISO trapezoidal threads (selection) as DIN 103 Part 4 (pitch as preferred series DIN 103 Part 2) Thread Pitch nominal P diameter d (mm)
10 12 16 20 24 28 (30) 32 36 40 44 48 (50) 52 (55) 60 (65) 70 (75) 80
90 100
(mm)
2 3 4 4 5 6 6 6 7 7 8 8 8 9 9 10 10 10 10 12 12
Flank Nut diameter external d z = Dz diameter D,
Bolt root Nut root Bolt core diameter diameter section d, D, 1rdU4
(mm)
(mm)
(mm)
(mm)
(mm')
9.0 10.5 14.0 18.0 21.5 25.5 27.0 29.0 33.0 36.5 40.5 44.0 46.0 48.0 50.5 55.5 60.0 65.0 70.0 75.0 84.0 94.0
10.5 12.5 16.5 20.5 24.5 28.5 31.0 33.0 37.0 41.0 45.0 49.0 51.0 53.0 56.0 61.0 66.0 71.0 76.0 81.0 91.0 101.0
7.5 8.5 11.5 15.5 18.5 22.5 23.0 25.0 29.0 32.0 36.0 39.0 41.0 43.0 45.0 50.0 54.0 59.0 64.0 69.0 77.0 87.0
8.0 9.0 12.0 16.0 19.0 23.0 24.0 26.0 30.0 33.0 37.0 40.0 42.0 44.0 46.0 51.0 55.0 60.0 65.0 70.0 78.0 88.0
44 57 104 189 269 398 415 491 661 804 1018 1195 1320 1452 1590 1964 2290 2734 3217 3739 4657 5945
Appendix Fl Table 6. Minimum thread depths in blind-hole threads [78] Recommended single thread depth for class 8.8 Thread rmeness d/ P
<9
Nut material hard AI alloy, AlCuMgl Grey iron GG 25 Steel St 37, C15M Steel St 50, C35M Steel cast with
1.1d I.Od I.Od 0.9d 0.8d
Rm > 800
[!,]
8.8
10.9
10.9
<9
lAd 1.25d 1.25d I.Od 0.9d
12.9
<9
lAd l.4d 1.2d I.Od
Mechanical Machine Components. 11 Appendix F: Diagrams and Tables
1.1:11
Appendix Fl Table 7. Tensile force Fsp and tightening torque Msp for bolts and screws with metric ISO coarse threads as DIN 13, Sheet 13 and heads as DIN 912 or 931, for friction coefficient J.4." = fJ.K = 0.12 over 90% of length (according to VDI Guideline 2230) Dimension
M,p (N m)
F" (N)
8.8'
10.9'
12.9'
8.8'
10.9'
12.9'
4050 6600 9400 13700 17200 27500 40000 5';000 75000 94000 121000 152000 175000 230000 280000
6000 9700 13700 20100 25000 40000 59000 80000 111000 135000 173000 216000 249000 330000 400000
7000 11400 16100 23500 29500 47000 69000 94000 130000 157000 202000 250000 290000 385000 465000
2.8 5.5 95 15.5 23.0 46.0 79.0 125.0 195.0 280.0 390.0 530.0 670.0 1000.0 1350.0
4.1 8.1 14.0 23.0 34.0 68.0 117.0 185.0 280.0 390.0 560.0 750.0 960.0 1400.0 1900.0
4.8 9.5 165 27.0 40.0 79.0 135.0 215.0 330.0 460.0 650.0 880.0 1120.0 1650.0 2250.0
4500 6300 9500 11800 18900 27500 38000 53000 66000 86000 109000 124000 166000 200000
6600 9300 14000 17300 27'500 40500 56000
7800 10900 16400 20200 32500 47500 65000 92000 110000 144000 182000 207000 275000 335000
3.8 6.5 10.9 16.0 32.0 55.0 88.0 135.0 195.0 280.0 380.0 480.0 720.0 970.0
5.5 9.5 16.0 23.0 47.0 81.0 130.0 200.0 280.0 400.0 540.0 680.0 1020.0 1400.0
6.5 11.1 18.5 27.1 55.0 95.0 150.0 235.0 320.0 460.0 630.0 800.0 1190.0 1600.0
Bolt
M4 M5 M6 (M 7) M8 MIO M 12 M 14 M 16 M 18 M 20 M 22 M 24 M 27 M 30 Screw (dT M5 M6 (M 7) M8 MIO M 12 M 14 M 16 M 18 M 20 M 22 M 24 M 27 M 30
= 0.9 . d,)
aClasses according to DIN ISO 898 Tl.
79000
94000 123000 155000 177000 236000 285000
I.I:M
Mechanical Machine Components. 11 Appendix F: Diagrams and Tables Appendix FI Table 8. Surface pressure p" (N/mm') for compressed components of various materials (according to VOl RichtliDie 2230)
Pull stress
Material
Rrn (N/mm')
Surface pressure PG (N/mm2)
St 37 St 50 C 45 42 CrMo 4 30 CrNiMo 8 X 5 CrNiMo 18 lOb XIO CrNiMo 18 9 b Stainless dispersion hardened materials Titanium, unalloyed Ti-6Al-4V GG 15 GG 25 GG 35 GG40 GGG 35.3 GDMgAl9 GKMgAl9 GKAlSi6Cu 4 AlZnMgCu 0.5 AI 99 Glass-fibre-reinforced materials Glass-fibre-reinforced materials
260 370 500 420 700 800 850 1000 750 1200 210 500 to 700 220 500 to 750 1200 to 1500 1000 to 1250 390 to 540 1100 150 250 350 400 350 300 (200) 200 (300) 450 160
300 1000
600 BOO 900 1100 480 220 (140) 140 (220) 200 370 140 120 140
"With mechanised tightening surface pressure can be up to 25% less. t.with cold fastened materials surface pressure is considerably higher.
Appendix F3
Appendix F3 Table I. Notes on common friction pairings [36J Dry
Wet Friction pairs
Sinter bronze- Sinter ironsteel steel
Paper-steel
Hardened steelhardened steel
Sinter bronze- Organic coatings-grey steel iron
Nitrided steelnitrided steel
Coefficient of friction Sliding I" Static JLo Ratio JLoII"
0.05 to 0.10 0.12 to 0.14 1.4 to 2
0.07 to 0.10 0.10 to 0.12 0.10 to 0.14 0.08 to 0.10 1.2 to 1.5 0.8 to 1
0.05 to 0.08 0.08 to 0.12 1.4 to 1.6
0.15 to 0.30 0.2 to 0.4 1.25 to 1.6
0.3 to 0.4 0.3 to 0.5 1.0 to 1.3
0.3 to 0.4 0.4 to 0.6 1.2 to 1.5
Technical data (standard values) Max. sliding speed VR
40
20
20
25
40
25
4
4
0.5 to 1
1 to 2
30
(m/s)
Max. friction surface pressure p. (N/mm2) Permissible surface work for individual
0.5
0.5
1 to 2
0.5 to 1
0.8 to 1.5
0.3 to 0.5
1 to 1.5
2 to 4
1.5 to 2.5
0.7 to 1.2
1 to 2
0.4 to 0.8
1.5 to 2.0
3 to 6
q.. Olmm') Pennissible surface friction power C:Uo (w/mm') (see VOl 2241, Sheet 1, Section 3.2.2)
I.I:JCI
Mechanical Machine Components. II Appendix F: Diagrams and Tables
""
'=
<=
"", c
<:
~
a;
!
'5
§
u.J
'"
~
~
~
Z
~./
I I
10
b 101
'
/ [ /'
/~/
I
"/~ I , we~hl ..(
G
Y ;/ ;;'7 1 ] -+-1--;-1--"-1t - - - - - - , - - - +I - +---'-I--+-~I'-H I /.j' i I ll , 1 1 ' : I I 10 8
10'
8
10'
1
Nomonallorque 11" In Nm
,
'10 5
,
8
10'
Figure 1. Coefficients for non-switchable couplings l5]: a speed n or external diameter D~ ; b weight g or length La as in catalogue; I douhletooth couplings. 2 membrane and spring shackle couplings, 3 metal-elastic (rotary elastic) couplings, 4 medium elasticity elastomeric couplings, 5 high elasticity eiastomeric couplings, a high-speed, b medium-speed types.
1.1:11
Mechanical Machine Components. 11 Appendix F: Diagrams and Tables
~ ~ W :011I
Multi-disc couplings Multi-surface couplings
~ ~~~
Double-surtace couplings (single disc) -I Double-surtace couplings (hose couplings) I Sfngle-surtace couplings (tooth control) I Single-surtace couplings (membrane control)
WW W W I I
I
I
~ ~ ~ ~ ~~ I
m
~W~
~ r%j
i
Pole-face couplings (tooth control)
~~ ~ ~ ~~
Pole-face couplings (membrane control)
~r:% W W ~
Cone couplings (Single acting) Cone couplings (double acting)
~~ ~
Cylindrical couplings I
Air-friction couplings
~~ ~~~ [~ ~ 0a
Screw-band couplings Frtctlon rtng couplings
~ ~~
WW~
Fricllon pad couplings
10
2
10
1
10
10 2
10'
Switching moment Ms in Nm
10'
lOS 106
Figure 2. Switching moment r-.lOges for types of remote control friction-switched couplings [6 J.
Appendix F4
Appendix F4 Table 2. Practical values for necessary service-life h
Appendix F4 Table 1. Operating ratios a to k and operating factors Jv for various assemblies
Assembly
Vehicle clutches Vehicle wheel bearings Trucks Railway freight cars Railway rolling stock E-series motors Large motors Turning and milling spindles Machine tools Drive motors Large-scale machinery Blowers Circumferential pumps Compressors Jaw crushers Papermaking machinery
Operating ratio fields as in Fig. 18
Wear factor Jv
g-k h-; f-h c-d c-d c-d b-d a-b
5 to 8 4to6 12 to 15
c-d d-e c-d f-h d-f d-f f-g b-c
8 to 12
3to6 3 to 5 3 to 5 0.5 to 1.5 3 3 6 5 3 3 8
to to to to
8 8 10 8 to 5 to 5 to 12
7 to 10
Road vehicles (fuUload) Passenger cars Trucks and huses Rail vehicles Cargo truck axle bearings Tramway vehicles Railway carriages Locomotives Gearing for railway rolling stock
Agricultural machinery Constructional machinery Electric motors for household equipment Series motors Large motors Machine tools Gearing for general engineering Large gears Ventilators, blowers Gear pumps Crushers, mills etc, Paper and print machinery Textile machinery
900 to 1700 to
10000 30000 20000 30 000 1'5000
1600 9000
to to to to
34000 50000 34000 100000 70000
2000 to 1000 to
5000 5000
1500 20000 '50000 15000 4000 20000 12000 500 12000 50000 10000
to
to to
to to to to to to to to to
4000 40000 10000U 80000 20000 80000 80000 8000 50000 200000 50000
Mechanical Machine Components. 11 Appendix F: Diagrams and Tables
Appendix F4
Table~.
Lubricant
Lubrication
'4':'"
Choice of lubricants [61 Lubrication equipment
Constructional details
Attainable speed
Type of bearing used
coefficient n . drn (min - I . mm)a
Solid lubricant For-life lubrication
~
1500
Mainly deep-groove ball bearings
For-life lubrication
~
0.5 ·10"
Regreasing
special greases, period as Fig. 27
All bearing types, except axial spherical roller bearings, depending on speed and grease type
Regreasing
Grease
= 1 . 106 for certain
Spray lubrication
Hand pressure, grease guns
Commercial
lubrication equipment"
Oil (large amounts)
Oil sump lubrication
Dipstick or standpipe bevel control
Oil circulation lubrication through self-lubrication of the bearing or bearing elements
OJ] (minimal amount)
Input via grease
nipples or grease
controllers, exit space for used grease Lower friction and better noise properties with
Input via tubes or bores, exit space for used grease
special greases
High-volume housing. Overflow pipes, fitting for control eqUipment
0.5 . 10"
Oil inlet bores, highvolume bearing housing. Supply components appropriate to circumferential velocity and oil viscosity. Check supply action of bearing
Must be determined
All bearing types, sound damping depending on oil viscosity, higher bearing friction due to oil splash losses, good cooling. Removal of wear particles by circumferential and spray lubrication
1·10"
Oil circulation lubrication
Circulation lubrication equipmenrt'
Yery large bores for oil inlet and outlet
~
Oil injection lubrication
Circulation lubrication equipment with spray nozzles
Oil inlet through directed nozzles, exit through sufficiently large bores
Tested to 4 . 106
Oil impulse lubrication, oil drop lubrication
Commercial lubrication equipment,h drop generator, oil spray lubrication eqUipment
Outlet bores
~ I. 5 . 10" depending on type of bearing, oil viscosity, oil quantity or type of construction
Oil-mist lubrication
Oil-mist generator,' or oil collector
Extraction eqUipment if necessary
Oil-air lubrication
Oil-air lubrication equipmentd
if necessary
Al) bearing types. Noise damping depending on oil viscosity, friction depending on oil quantity and oil viscosity
Extrdction equipment
aDepending on type of bearing and constructional ratios. hCentral lubrication eqUipment consisting of pumps, container, filter, piping, valves, throttles. Circulation equipment with oil return, possibly with cooler. Conunercial eqUipment with time controlled dose valves for small quantities (5 to 10 mm3/hub). "Oil-mist eqUipment consisting of container. mist generator, leads, back-pressure jets, control, compressed air eqUipment. dOil-air lubrication equipment consisting of pumps, containers, leads. volumetric oil-air dose distributor, nozzles, control, compressed air equipment.
1.I:d
Mechanical Machine Components. 11 Appendix F: Diagrams and Tables
Appendix F4 Table 4. Coefficients for various oils [6]
Viscosity at 40 ° (mml/s) Use for oil sump temperature
Mineral oil
Polyalphaolefin
Polyglycol (water insoluble)
Ester
Silicone oil
Alkoxyfluoro-oil
2 to 4500 100
15 to 1200 150
20 to 2000 100 to 150
7 to 4000 150
4 to 100000 150to200
20 to 650 150 to 220
150
200
150 to 200
200
250
240
-20b 220 Moderate Good
-40b 230 to 260b Low Good
-40
-60 b 220 to 260 Low Moderate to goodb
-60b 300b Low Good
Moderate to badh Good
Good 1.5 to 4.5 ·10'
Very good Moderate to good 1.0 to 3.0· 10' 2.5 to 4.4 . 10'
eC) to Use for oil exit temperature eCl to Pourpoint in °C Ignition point (OC) Evaporation loss Water resistance
Tar content Pressure-viscosity coeffiCient
200 to 260 Moderate to high Good, b difficult to separate because of similar density Moderate Moderate to good Good 1.1 to 3.5 . 10" 1.5 to 2.2 . 10' 1.2 to 3.2· 10"
(ml/NY
High-temperature suitability (= 150°C) High-load suitability Compatibility with elastomers Price ratios
Moderate
Good
Moderate to goexlb Goodb
Very good
Very good
Very good3 Good
Very good" Goodb
Very good" Moderate, tested on coatings 4 to 10
Good Moderate to bad
Badb Very good
Good Good
4 to 10
40 to 100
200 to SOO
6
3With EP-additives. bDependent on oil type. <-Measured to 200 bar. Height dependent on oil type and viSCOSity.
Mechanical Machine Components. 11 Appendix F: Diagrams and Tahles
1.1:*1
Appendix F4 Table S. Criteria for selection of greases [61 Criteria for selection of greases
Properties of selected greases
Operating conditions
Grease selection according to Appendix F4 Table 6
Speed coefficient n . d m Load coefficient P / C
Running requirements Low friction, even at starr
Grease of consistency classes 1 to 2 with synthetic lOW-viscosity oil base.
Low and constant friction under continuous running, but higher starting friction permissible
Grease of consistent classes 3 to 4, grease amount less than 30% of free-bearing space, or grease of consistcn<;y classes 2 to 3, grease amount less than 20% of free-bearing space
Low running noise
Filtered grease (high purity) of consistency class 2, with especially high provision for low noise, very well filtered grease of consistency classes 1 to 2 with high-Viscosity oil base
Constructional ratios Position of bearing axis transverse or vertical
Outer ring rotating. inner ring stationary or operated by bearing centrifugal force
Adhesive grease of consistency classes 2 to 3 Grease of consistency classes 3 to 4 with high thickening agent content
Evaluation Frequent regreasing
'111in grease of consistency classes I to 2
Occasional regreasing, for-life lubrication
Squeeze-stable grease of consistency classes 2 to 3, service temperature considerably higher than operating temperature
Environmental ratios High tempentture. for-life lubrication
Temperature-stable grease with synthetic base oil and with temperature-stable (or synthetiC) thickening agent
High temperature, regreasing
Grease that does not build up sediment at high temperature
Low temperature
Grease with thin synthetic oil base and suitable thickening agent. consistency classes 1 to 2
Dusty conditions
Firm grease of consistency class 3
Condenser water
Emulsified grease, e.g. sodium or lithium soap
Water spray
Water shedding grease, e.g. calcium soap
Aggressive media (acids, alkalis etc.)
Special grease recommended by roller bearing or lubricant supplier
Ionising radiation
Up to energy levels 2 . 1{)4 J/kg, roller bearing grease to DIN 51825 Up to energy levels 2· 10' J/kg, as recommended by bearing manufacturer
Vibratory stresses
Lithium EP grease of consistency class 2, frequent regreasing Under moderate vibration conditions, barium-complex soap of consistency class 2 with firm lubrication properties, or lithium soap of consistency class 3.
Vacuum
To 10- '; mbar. roller bearing grease to DIN 51 825; to higher vacuums as recommended by manufacturer.
1.1:1:1
Mechanical Machine Components. 11 Appenilix F: Diagrams and Tables
Appendix F4 Table 6. Rolling bearing greases and their properties No. Thickener
Base oil
Operating temperature
Water resistance
Remarks
0c" Sodium soap
Mineral oil
-20to+100
Not resistant
Emulsified with water otherwise possibly fluid
Lithium soapb
Mineral oil
- 20 to + 130
Resistant to 90°C
Emulsified with small amount of water, thinner with larger amounts, multi-use grease
Lithium-complex soap
Mineral oil
-30to+150
Resistant
Multi-use grease with high temperature resistance
Calcium soapb
Mineral oil
- 20 to
+ 50
Very resistant
Good sealing against water, penetrating water not taken up
+
Resistant
Good sealing against water
Aluminium soap
6 Sodium-complex soap
Mineral oil
- 20 to
Mineral oil
- 20 to + 130
Resistant to approx. 80°C Suitable for high temperatures and loads
70
7
Calcium-complex soaph
Mineral oil
- 20 to + 130
Very resistant
Multi-use grease suitable for high temperatures and loads
8
Barium-complex soaph
Mineral oil
- 20 to + 150
Resistant
Suitable for high temperatures and loads and also speeds (depending on base oil viscosity); resistant to evaporation
9
Polycarbamideb
Mineral oil
- 20 to + 150
Resistant
Suitable for high temperatures, loads and speeds
Mineral oil
-20to+150
Resistant
Suitable for high temperatures and loads as well as speeds (depending on base oil viscosity)
Mineral oil -20to+150 ami/or Ester oil
Resistant
Gel grease, suitable for high temperatures at low speeds
10 Aluminium-complcx soaph
II
Bentonite
12 Lithium soapb
Ester oil
-60to+130
Resistant
Suitable for low temperatures and high speeds
13 Lithium-complex soap
Ester oil
-50to+220
Resistant
Multi-range lubricant for wide temperature application
14 Barium-complex soap
Ester oil
- 60 to + 130
Resistant
Suitable for high speeds and low temperatures, resistant to evaporation
15 lithium soap
Silicone oil
- 40 to + 170
Very resistant
Suitable for high and low temperatures at low loads up to medium speeds
aDepending on type of bearing and lubricant. Through selection of appropriate mineral oils the cold operating condition of the grease can be improved 1 to 10 (e.g. - 30°C in special cases down to - 55 °C). bAlso with EP additives.
Mechanical Machine Components. 11 Appendix F: Diagrams and Tables
I.t=ijl
Appendix FS
Appendix FS Table 1. Graph for determination of bearing temperature ~2() to ~aah == 22°C for So < 1
10
f----t-----+-----t---l-t---r r__
~~~ ~ __
'"
~
-t ______ ____r-__
_____
~_V_G_f_or_e_'to._gg_/_cm_l-t ~ ~-t____~ __
Win
'1m' 'C'/Ns
32 22
Bearing temperature 111fJ in'C for
0.,."
20'C
____
.MUI..
Mechanical Machine Components. 11 Appendix F: Diagrams and Tables
Appendix FS Table 2:. Graph for determination of bearing temperature
P.w to Paah == 22°C for 50 < 1
10
VG for e ~ 0,9 g/cml
\
W*in 'C/Pos
Bearing temperature 810 in 'C for
a-Qmb ~ 20 'C
70
80
100
120
Mechanical Machine Components. II Appendix F: Diagrams and Tables
'Mml
Appendix F; Table 3. Maximum permissible loadings of friction bar (N/mm:l) (lower limit for unhardened shaft and materials high sliding speed, upper limit for hardened shaft and low sliding speed)
p
Bearing material
p
Brirrell hardness HB eN/mm')
Lg Pb Sn IO Zinc alloy Aluminium alloy Tin bronze, guarnetal G Ph Az 25
Special alloys
eN/mm')
230
5 5 5 8 6 7
700 750 700 300 950
to to to to
25 28 SO 55 to 80 to SO
Appendix FS Table 4. Chemical and technological properties of bearing materials Material
Chemical composition
(%)
LgPbSb 14
LgPbSn 6 (WM 5)
LgSn 80
LgSn 80 F GCdj';i
SnPbBz 15 SnPbSz 13 SnPbSz II
838 Pb; 14 Sb: I So;1 As;
7,.8 Pb: 15 Sb: 6 Sn; 0.5 As; 1.2 Cu;1 Cd; 0., Ni
2 Ph; 12 Sb; 80 Sn; 6 Cu
0.5 Pb; II Sb; 80 Sn; 9 Cu
98.4 Cd: 1.6 Ni
I, Pb; 2.5 Sn: 79., Cu; 3 Ni
274 232 133 73
278 237
80
256 210 142 81
340 289 197 115
27
29
61.8
66.7
7(}
';H
89.3
76.5
55700 61.8 37.3
0.2 Cu
20 DC
Hardness and hot hardness HB
50°C 100°C 150°C
(Nimmo.!)
Apparent yield point Tensile strength Rm
Rpo
2
(N/mm2)
(N/mm~)
Modulus of elasticity E (N/mm")
180 160 1.)0
29';00
29900
151 76
78.5
513 491 466 445 84.4
13 Pb: , Sn; 79 Cu;
3 Ni
II Ph; 8 Sn; 77.5 Cu; 3.5 Ni
658 649 626
863 803 786 769
120
163
67'5
129
136
192
209
';7900
54200
81500
84000
85100
67.7 42.2
69.7 50
76.5 64.8
109 952
116
Crushing yield point, a uo " (N/mm2)
20°C 100°C
34 2\
47 27
Crushing resistance (N/mm")
20°C ISO °C
HI
137 85
189
95
121
157 102
285 226
515 420
661 524
701 666
60
115
167
132
241
277
.\51
348
104
1,5
196
216
(.TuB
flexurAl resistance,
ah!;
eN/mm")
Bonding resistance according to Chalmers eN/mm') (steel CHI) Fatigue strength under reverse bending stress, Uh, (N/mm2)
30
Thermal conductivity, 20 to 100°C (kJ/(m hOC)) Avenlge coefficient of expansion. 20 to 100°C (10 /,/oe)
,6.9
39.2
451
28
27.5
30.4
;12.4
60.8
76.,
87.3
137.2
141.6
250.2
185.4
177.9
188.5
22
22
19
18
18
18
76.9
25
U8
24
I.M
Mechanical Machine Components. 11 Appendix F: Diagrams and Tables
Appendix F6
Appendix F6 Table 1. Flat belts (Siegling, Hannover) Extremultllr SO/SSG (elastomer) or L (chrome-leather). Tensile modulus (EA *), belt thickness s, mass per unit length p', ... refers to belt thickness approximately 1 mm 10
Type no.
dl,rnin
F~N
EA'
~
(mm)
63
(N/mm) 8 (N/mm) 500
14
20
28
40
54
80
100 140 200 280 385 540 12.5 17.5 25 48.5 67.5 35 700 1000 1400 2000 2700 4000
P'le
p'
G
(mm) (mm)
1.5 2.2
1.7 2.7
2.5 3.0
2.9 3.7
35 45
4.3 5.7
5.7 7.5
G L
(kg/m') 1.5 (kg/m') 2.1
11.7 2.4
2.7 3.1
3.1 3.6
3.8 4.5
4.7 6.1
6.1 7.4
Appendix F6 Table 2. V-belt dimensions (selection) and belt coefficients for assessment of pennissible nominal power PN from Eq. (11) and manufacturers' data [6, 7], valid for speeds of smaUer disc n::S; nl,max and v:S; Vmax . Profile designation according to DIN 7753 Tl (corresponding ISO) or DIN 2215 (number) or ISO (chapter)
Proftle designation DIN
ISO
Working width b_
Draw length Lo
(mm)
Power
Speed
Speed
Rotational
Diameter
Power factors
d_. mm (mm)
K,
speed
Po (kW)
"
Vo
vm~
(m/s)
(m/s)
n maxd
(min-')
Spzo SPAO SPBo SPCC 6,b
W OC 13'0 17'0 22'0 32'0 40"0
SPZ SPA SPB SPC y
Z A B C
D E
8.5 11
14 19 5.3 8.5 11 14 19 27 32
1600 2500 3550 5600 315 822 1730 2283 3802 6375 7182
1.90 3.57 9.43 21.54 0.21 0.53 1.04 2.99 8.28 21.45 3018
19.50 20.73 25.95 28.15 12.22 13.02 11.52 16.42 21.02 24.54 24.62
~ Agrees in general with maximum width bo according to Fig. 12a. bRough cut, lWith woven cover. dUpper limit of manufacturers' data.
44.0 44.0 41.9 44.5 35.2 32.7 33.5 33.4 33.4 34.2 33.5
8000 6000 5000 3500 12000 6000 6000 4000 2850 1450 1200
63 90 140 224 20 45 71 112 180 315 450
4.610 4.268 2.832 2.339 4.730 4.725 5.950 4.113 2.725 1.994 1.713
K,
K, =(1
~
0.250 0.270 0.330 0.353 0.200 0.250 0.160 0.240 0.300 0.330 0.350
K,)/2
0.500 0.460 0.340 0.294 0.600 0.500 0.680 0.520 0.400 0.340 0.300
Mechanical Machine Components. 11 Appendix F: Diagrams and Tables
'Mme.
Appen.d.b. F6 Table.J. Values for commonly used synchronous belts for estimating according to manufacturers' data with glass-fibre strands Gf (6] and steel strands 5t (10]
Type
Fibre
Pitchp'o (nun)
P"
v"
(kW)
(m/s)
b"" (nun)
Gf
5.080
3.61
29.85
25.4
Gf
9.525
4.72
28.%
25.4
H
Gf
12.700
16.33
38.94
25.4
XH
Gf
22.225
16.94
29.85
25.4
XXH
Gf
31.750
20.31
29.21
25.4
T 2.5 T5 TIO T 20
St SI SI 5t
2.5 5.0 10 20
25.00 86.16 65.98 50.18
25.4 25.4 25.4 25.4
Xl
O.9Sh
9.38b 16.32'0
22.9}0
.aHigher minimum number of teeth with comra·bending. bCalcuiation for 6 load-bearing teeth (manufacturers' data z~ma.x
zl.nlUl for n l (min')
vrna><
nm~
(m/s)
(min ')
25.4
10000
46
6000
(n )'''
61
6000
20
(n950r~
50
4400
22
(n950)""
50
3000
(25) 80
15000 15000 15000 6000
(n950
10 ~
to
(n950)"'0
12 ~
16 ~ 950
------...!....
------...!....
10/18' 10/15' 12/20' 15/25'
= 15).
Appendix F6 Table 4. Standard roller chains (selection) DIN 8187
DIN 8195
Chain
no.
P" kW
min
06 B 08 B lOB 12 B 16 B 20 B 24 B 28 B 32 B 40 B 48 B 56 B 64B 72B
3.5 7.5 11.0 14.7 32.0 47.5 68.0 78.0 92.0 120 140 160 160 124
1700 1400 1200 1050 680 500 350 300 250 180 125 80 54 30
no
DIN 8188 DIN 8195 Chain
no.
Po kW
no
08 A lOA 12 A 16A 20 A 24 A 28 A 32 A 40 A 48 A
8.5 1950 14.8 1550 19.0 1300 34.2 980 54.0 720 70.0 550 85.0 440 320 105 120 205 100 100
min- l
p nun
9.525 12.7 15.875 19.05 25.4 31.75 381 44.45 50.8 63.5 76.2 88.9 101.6 114.3
60 45
'MPI'
Mechanical Machine Components. 12 References
References 1'1 Connections. II]
DIN 7190: Pressverbande, Berechnungsgrundlagen und Gestaltungsregeln. Berlin: Beuth 1988. - [2] DIN 4768, T I mit Beiblatt I: Ermittlung der Rauheitsmessgr6ssen Ra , Rp Rmax mit elektrischen Tastschnittgeriiten: Grundlagen und Urnrechnung der Messgrossen R. in R, und umgekehrt. - [3] DIN 7154 Pt I and Pt 2: ISO-Passungen fur Einheitsbohrung; Toleranzfelder, Abmasse und Passtoleranzen, Spiele und Dbermasse. - [4] DIN 18800 Pt I: Stahlbauten, Bemessung und Konstruktion, Marz 1981. - [5] DIN 254: Kegel, Begriffe und Vorzugswerte. - [6] DIN 1488: Kegelige Wellenenden mit Aussengewinde, Abmessungen. [7] DIN 1449: KegeJige Wellenenden mit Innengewinde, Abmessungen. - [81 DDR-Standard TGL 19340 Pt 4: Maschinenbauteile; Dauerschwingfestigkeit; Kerbwirkungszahlen 13k fur Achsen und Wellen. - [9] VDI-Richtlinie 2029: Presspassungen in der Feinwerktechnik, October 1958. - [10] Kollmann F G. Welle-Nabe-Verbindungen. KonstruktionsbUchcr, vol. 32. Springer, Berlin; 1984. [II] Leidich F. Beanspruchung von Pressverbindungen im elastischen Bereich und Auslegung gegen Dauerbruch. Diss., TH. Darmstadt, 1983. - [12] Seefluth K. Dauerfestigkeit an Wellen-Naben-Verbindungen. Diss., ro, Berlin, 1970. - [13] Galle G. Tl"~gfahigkeit von Querpressverbanden. Schriftenrcihe Konstruktionstechnik (ed. Beitz W), no. 4, ro, Berlin, 1981. - [14] Kreitner L. Die Ausbreitung von Reibkorrosion und von Reibdauerbeanspruchung auf die Dauerhaltbarkeit zusammengesetzter Maschinenenteile. Diss., TH Darmstadt, 1976. - [15] Hausler N. Zum Mechanismus der BiegemomentUbertragung in Schrumpfverbindungen. Diss., TH Darmstadt, 1974. [16] MUller H W. Drehmoment-Obertragung in Pressverbindungen. Konstruktion 1962; 14: 47-52,112-15. - [17] Lundberg G. Die Festigkeit von Presssitzen. Kugellager 1944; 19: 1/2, 1-11. - [18] SKG-Zeitschrift: DerDruckOlverband. Schweinfurt, 1977. - [19] Kollmann F G. Die Aus!egung elastisch-plastisch beanspruchter Querpressverbande. Forsch Ingcnieurwes 1978; 28: I-II. - [20] Gamer U, Kollmann F G. A theory of rotating elastoplastic shrink fits. Ing Arch 1986; 56: 254-64. - [21] Schmid EA. Thearetische und experimentelle Ontersuehung des Mechanismus der DrehmomentUbertragung von Kegelpressverbindungen. VOI-Fortsehr Ber series 1, no. 16, 1969. - [22] MichIigk T. Statisch Uberbestimmte Flanschverbindungen mit gleichzeitigem Reib- und Formschiuss. Diss., ro Berlin (1988). - [23] Roland G. Beistrag zur Konstruktion von Fordertrommeln mit durchgefuhrter Welle in Spannsatzausfuhrung. VDI-Fortschr Ber series 13, no. 24, 1984. [24] BIKON-Technik: Welle-Nabe-Verbindungen. Grevenbroich, 1989. - [25] BIKON-Technik: Grundwissen, Hinweise Lieferprogramm. Grevenbroich, n.d. - [26] Deutsche Star: Toleranzringe. Schweinfurt, 1988. - [27] Fenner. Taper-Lock-Spannbuchsen. Nettetal-Breyell, 1988. - [28] Hochereuter & Baum. DOKO Spannelemente. Ansbach (ohne Jahr). [29] Ringfeder. Spannsatze, Krefeld, 1988. - [30] Ringfeder. Spannelement. Krefeld, 1988. - [31] Ringfeder. Sehrumpfscheiben. Krefeld, 1988. - [32] Ringspann. TOLLOK Konus-Spanne!emente. Bad Homburg, 1989. - [33] Ringspann. Sternseheiben und Spannscheiben fur Welle-Nabe-Verbindungen. Bad Homburg, 1989. - [34] SKF Kugellagerfabriken. Druckolverband. Schweinfurt, 1977. - [35] Spieth-Maschinenelemente. Druckhiilsen. Esslingen, n.d. - [36] SpiethMaschinenelemente. Fabrikations-Programm. Esslingen, n.d. - [37] StUwe. Schrumpfscheiben-Verbindung. Hat-
[38] Lenze, SUdtechnik. ETPtingen, 1989. Spannbuchsen fur Wellen-Nabenverbindungen. Waiblingen, n.d. - [39] Handbuch Antriebstechnik. Tabellenwerte-Uber Lieferanten und Prodnktdaten. Krausskopf: erscheint jahrlich. - [41] DIN-Taschenbuch 43: Mechanische Verbindungselemente 2, Bolzen, Stifte, Niete, Keile, Stellringe, Sicherungsringe. Beuth, Berlin, 1988. - [42J (Normen, DIN-Taschenbucb 69: Stahihochbau Richtlinien). Beuth, Berlin; 1986. - [43] DIN 1073: StabIerne Stf'dssenbriicken. Berechnungsgrundlagen. - [44] DIN 4113 TI: Aluminiumkonstruktionen unter vorwiegend ruhender Belastung; Berechnung und bauliche Durchbildung 1980. - [45] DIN 15018 Pt I: Krane; Grundsatze fur Stahitragwerke; Berechnung. DIN 15018 Pt 2: Krane; Stahitragwerke; Grundsatze fur die bauliche Durchbildung und Ausfuhrung. DIN 15018 Pt 3: Krane, Grundsatze fur Stahltl"dgwerke; Berechnung von Fahrzeugkranen. - [46] DIN 18800 Pt I: Stahlbauten, Bemessung und Konstruktion. - [47] LN 29730: Nietrechnungswerte bei statischer Beanspruchung fur Universal-Nietverbindungen. - [48] LN 29731: Nietrechnungswerte bei statischer Beanspruchung fur Senknietverbindungen. - [49] DASt Bau-Richtlinien fur Verbindungen mit Schliessringbolzen im Anwendungsbereich des Stahlliochbaus mit vorwiegend ruhender Belastung. Deutscher Ausschuss fur Stahlbau, Cologne, 1970. [50] Vorlaufige Richtlinien fur Berechnung, Ausfuhrung und bauliche Durchbildung von gleitfesten Schraubenverbindungen (HV-Verbindungen). Stahlbau-Verlag, Cologne. - [51] Dampfkesselbestimmungen. III. Techn. Vorschriften, Pt 3, TUV, Essen, 3rd edn, Cologne. - [52] Stahl im Hochbau. Verein Deutscher EisenhUttenleute, DUsseldorf. - [53] Aluminium-Taschenbuch, 14th edn (Aluminium-Zentrale DUsseldorf), Aluminium-Verlag, DUsseldorf, 1988. - [54] Niemann G. Maschinenelemente, vol. I, 2nd coo, Springer, Berlin, 1981. [55] Kollmann F G. Welle-Nabe-Verbindungen, Springer, Berlin, 1984. - [56] Heide W. Untersuchungen an Kerbstiften und Kerbstiftverbindungen. Diss., ro Hannover, 1969. - [57] Hoffmann G. Technologische Probleme der Nietung und ihre Auswirkung auf die Dauerfestigkeit. Luftfahrttechnik 1962; 8: 90-8. - [58] Hummel 0 H. Nieten. KEM 1977; 14: 90-2. - [59] Militzer O. Rechenmodell fur die Auslegung von Wellen-Naben-Passfederverbindungen. Diss., ro, Berlin, 1975. - [60] Munz D. Bruchmechanikkonzepte fur Zeitfestigkeitsberechnungen. In: VOl Ber 661 - Dauerfestigkeit und Zeitfestigkeit, VOl-Verlag, DUsseldorf, 1988. - [61] Netz H. Dampfkessel, 7th edn, Stuttgart, 1972/3. - [62] Neuber H. Ober die Beriicksichtigung der Spannungskonzentration bei Festigkeitsberechnung. Konstruktion 1%8; 20: 245-51. - [63] Sollmann H. Ein Beitrag zur Elastizitat der Bolzen-LaschenVerbindung. Wiss Z ro Dresden 1965; 14: 1417-24. [64] Steinhardt 0, Valtinat G. Hochfeste, vorgespannte Schliessringbolzen im Stahlbau. Maschinenmarkt 1970; 76: H. 44. - [65] Valtinat G. Untersuchung zur Festlegung zulassiger Spannungen und Kriifte bei Niet, Bolzen- und HV-Verbindungen aus Aluminium-Legierungen. Aluminium 1971; 47: 735-40. - [66] Willms V. Auslegung von Bolzenverbindungen mit minimalem Bolzengewicht. Konstruktion 1982; 34: 63-70. - [67] Firmendruckschriften zu Stiftverbindungen: W. Hedtmann KG, HagenKabel (Spannbillsen) 5800 Hagen I. - Kerb-Konus Ges., Dr C. Eibes & Co., 8454 Schnaittach/Opf. (Kerbstifte). W. Prym, 5190 Stolberg (Spiralspannstifte). - C. Vogelsang GmbH, 5800 Hagen 5 (Spannhiilsen). - [68] Firmen-
Mechanical Machine Components _ 12 References
druckschrift zu POlygon-Verbindungen: Fortuna-Werke Maschinenfabrik: Fortuna-Polygon-Verbindungen, Stuttgart-Bad Cannstatt Cohne Jahr). - [69] Firmendruckschrift zu Axiale Sicherungselemente: Seeger-Orbis GmbH, 6240 Konigstein. - [701 Pahl G, lIeinrich ). Sicherungsringverbindungen Berechnung von Forrnzahlen, Dauerfestigkeit, Ringverhalten. Konstruktion
1987: 39: 1-6. - [71] Beitz W, Pfeiffer B. Einfluss von Sicherungsringverbindungen auf die Dauerfestigkeit dynamisch belasteten Wellen. Konstruktion 1987; 39: 7-13. [72] Firmendruckschriften zu Nietverbindungen: Gebr. Titgemeier, Gesellschaft fur Befestigungstechnik, 4500 Osnabruck (leaflets on HUCK bolts with locking ring; HUCK blind rivets. POP and POP-blind cup-rivets, blind riveting nuts and bolts; GETO expanding rivets in plastics). - Gesipa - Blindniettechnik GmbH, 6082 Morlc elden-Walldorf (leaflets on blind rivets in steel, copper and aluminium alloys). - Honsel, Alfred: Nieten- und Metallwarenfabrik, 5758 Frondenberg/Ruhr (leatlets on tubular and semi-tubular rivets, brake and clutch-lining rivets). - [73] DIN Taschenbuch 10: Mechanische Verbindungselemente (Schrauben, Massnormen), Beuth, Berlin, 1985. - [74] DIN-Taschenbuch 45: Gewindenormen. Beuth, Berlin, 1985. - [75 J DIN-Taschenbuch 55: Mechanische Verbindungsclemente 3 (Technische Liderbedingungen fur Schrauben und Muttem), Beuth, Berlin; 1985. - [76] VDI-Richtlinie 2230 Paper 1: Systematische hochbeanspruchter SchraubenverbinBerechnung dungen - Zylindrische Einschraubenverbindungen. VD!EKV-Ausschuss Schraubenverbindungen, Beuth, Berlin, 1986. - [77] DIN 18800 Pt 1: Stahlbauten, Bemessung und Konstruktion (auch gleitfester Verbindungen mit hochfesten (HV-) Schrauben). Beuth, Berlin, March 1981. - [781 Blume 0, )llgner K H. Schrauhen-Vademecum, 7th edn. Bauer & Schaurte Karcher GmbH Neuss/Rbein, 1988. - [79] Wiegand H. et al. Schraubenverbindungen, 4th edn, Konstruktionsbiicher, vol. 5. Springer, Berlin, 1988. - 180] Kubler K H, Mages W. Handbuch der hochfesten Schrauben, Girardet. Essen, 1986. [81] Kober A. Schaden an grossen Schraubverbindungen - Spannungsanalyse - Bruchmechanik - Abhilfemassnahmen. Maschinenschaden 1986; 59: 1-9. - [821 Agatonovic P. Verhalten von Schraubenverbindungen bei
zusammengesetztcr Betriebsbeanspruchung. Diss.. TU Berlin, 1973. - [83] Bauer C D. Ungenugende Dauerhaltbarkeit mitverspannter fedemder Elemente. Konstruktion 1986; 38: 59-62. - [84] Birger) A. Die Stauchung zusammengeschraubter Platten oder Flansche (russ.). Russ. Eng. J. 1961: no. 5: pp. 35-8. Extract in: Konstmktion 1963; 15: 160. - [85] Esser J. Verriegelungsrippen an Sicherungsschraubcn und Muttem. Ingenieurdienst no. 34, Bauer & Schaurte Karcher GmbH, Neuss/Rbein, 1986. - [86] Galwelat M. Rechnerunterstlitze Gestaltung von Schraubenverbindungen. Diss., TU Berlin, 1979. [87] Galwelat M. Programmsystem zum Auslegen von Schraubenverbindungen. Konstruktion 1979; 31: 27582. - [88] )ende S, Knackstedt R. Warum Dehnschrauben? VOl-Z 1986; 128: 111-19. - [89] Kober A. Zum betriebsfesten Dimensionieren grosser Schraubenverbindungen unter schwingender Beanspruchung mit besonderem Bezug auf den Abmessungsbereich M 220 DIN 13. Maschinenschaden 1987; 60: 1-8. - [90] Koenigsmann W, Vogt G. Dauerfestigkeit von Schraubenverbindungen grosser Nenndurchmesser. Konstruktion 1981; 33: 219-31. - [91] Michligk T. Statisch uberbestimmte Flanschverbindungen mit gleichzeitigem Reib- und Formschluss. Di"., TU Berlin, 1988. - [92 J Neuendorf K. Ein Balkenmodell flir die Berechnung des elastostatischen Verhaltens hochbean-
..
,
spmchter Schraubenverbindungen. Diss., TU Berlin, 1975. - [93] Spiess D. Das Steifigkeits- und Reibungsverhalten unterschiedlich gestalteter Kugelschraubtriebe mit vorgespannten und nicht vorgespannten Muttersystemen. Diss., TU Berlin, 1970. - [95] Thomala W. Beitrag zur Dauerhaltbarkeit von Schraubenverbindungen. Diss., TH Darmstadt, 1978. - [951 Yakushev AJ. Effect of manufacturing technology and basic thread parameters on the strength of threaded connections, Pergamon Press, New York, 1964. - [96] VDI-Richtlinie 2232: Methodische Auswahl fester Verbindungen. Systematik, Konstruktionskataloge, Arbeitshilfen. VDI-EKV-Ausschuss Verbindungstechnik, Beuth, Berlin, 1990. F2 Elastic Connections (Springs) [1] Almen J 0, Laszlo
A. The uniform-section disk spring. Trans ASME 1936; 58: 305-14. - [2] Baumgartl E et al. Zur Dauerfestigkeit vernickelter Schraubendruckfedem. Draht 1967; 18: 58291. - 13] Betz A. Federgelenke. VDI-Z 1966; 108: 51-4, 93-6. - [4] Brunn H. Vergleich der Beanspmchung von Gse und Wickelkorper bei der Schraubenzugfeder mit angebogenen Gsen. Draht 1969; 20: 661-3. - [5] Busse L. Schwingungen zylindrischer Schraubenfedem. Konstruktion 1974; 26: 171-6. - [6] Busskn, Automobiltechnisches Handbuch, Cram, Berlin, 1965. - [7] Buttler K. Reibungsfedem Bauart Ringfeder im Maschinenbau. Konstmktion 1970; 22: 149-53. - [8] Buhl P. Zur Berechnung von Tellerfedem mit Auflagetlachen. Draht 1966; 17: 7537. - [9 J Buhl P. Zur Spannungsberechnung von Tellerfedem. Draht 1971; 22: 760-3. - 110] v. Estorff H-E. Einheitsparabelfedem flir Kraftfahrzeug-Anhanger. BriininghausInformation no. 2, 1973. - [11 J v. Estorff H-E. Technische Daten Fahrleugfedem. Pt 1, Drehfedem. Stahlwerke Bruninghaus Werdohl, 1973. - [12] v. Estorff H-E. Technische Daten Fahrzeugfedem. Pt 3. Stabilisatoren. Stahlwerke Bruninghaus Werdohl, 1969. - [13] v. EstorffH-E. Parabelfedem flir Guterwagen. Techo Mitt Krupp 1979; 37: 10915. [ 141 Fedem K. Beherrschung und Ausnutzung von Schwingungen als Konstruktionsaufgabe. VOl-Z 1958; 100: 1220-32. - [15J Fedem K. Dampfung elastischer Kupplungen (Wesen, Frequenz- und Temperaturabhangigkeit. Ermittlung). VDI-Ber 1977; 299: 47-61. - [16] Fedem K. Fedemde Verbindungen (Fedem). In: Dubbel: 16th edn, Springer, Berlin, 1987. - [17J Fischer F, Vondracek H. Warmgeformte Fedem, Konstruktion und Fertigung. Hoesch AG, Hohenlimburg, 1987. - [18] Friedrichs J. Die Uerdinger Ringfeder (R). Draht 1964; 15: 539-42. - [19J Go G D. Problematik der AusJegung von Schraubendruckfedem unter Beriicksichtigung des Abwalzverfahrens. [20] Gross S. Automobil Ind 1982; 3: 359-67. Berechnung und Gestaltung von Metallfedem. Springer, Berlin, 1960. - [21] Gross S. Zylindrische Schrdubenfedem mit unveranderlichem Verhaltnis von Einheitskraft zu Belastung. Draht 1963; 14: 483-7. - [22] Gross S. Drehschwingungen zylindrischer Schraubenfedem. Draht 1964; 15: 530-4. - [23] Hegemann F. Dber die dynamischen Festigkeitseigenschaften von Blattfedem fur Nutzfahrzeuge. Diss., Til Aachen 1970. - [24] Hempel M. Untersuchungen an Eindraht-Schraubenfedem. (Pt 1: Werkstolfe und Fertigung. Prlifung und FederkrJft, Bruchhaufigkeit und Federverformung. Pt II: Dauerfestigkeitspriifungen an Federdrahten und Fedem). Konstruktion 1953; 5: 335-43 und 1954; 6: 60-9. - [25] Hertzer K H. Ober die Dauerfestigkeit und das Setzen von Tellerfedem. Diss., TH Brunswick, 1959. - [26] Heuer po). Entwicklungstendenzen bei Fahrzeugfedem. ATZ 1966; 68: 241. - [27] Hoesch, Stahlfeder-Handbuch. - [28] Hoesch-Fedem, Technische Blatter der Hoesch Werke HohenlimburgSchwerte. - [29] Hoff H. Federstahle. Werkstoff-Hand-
'MP'i
Mechanical Machine Components. 12 References
buch Stahl u. Eisen, Q 11. Verlag Stahleisen, Dtisseldorf. [30] Huhnen J. Entwicklungen auf dem Fedemgebiet. Drabt 1966; 17: 669-81 and 1967; 18: 592-612. - [31] HuhnenJ. UnmOgliche 'Schraubenzugfedern" jetzt verwirklicht. Drabt 1975; 26: 595-9. - [32] Huhnen J. Schraubenzugfedem, Vorschllige zur Weiterentwicklung der DIN 2089 Paper 2. Drabt 1987; 38: 218-22. - [33] Hubner W. Deformationen und Spannungen bei Tellerfedern. Konstruktion 1982; 34: 387-92. - [34] Kaiser B. Einfluss des Kugelsttahlens auf die Schwingfestigkeit von Federelementen. Drabt 1987; 38: 116-20. - [35] Kaiser B. Dauerfestigkeitsuntersuchungen an biegebeanspruchten Federn aus FederbandstaW. Drabt 1987; 38: 675-SO. - [36] Keding H. MOglichkeiten zur Erhilhung der Schwingfestigkeit von Fahrzeugblattfedern aus Federstahl 55SiMn7. Kraftfahrzeugtechnik 1970; 104. - (37) Keil E. et al. Versuche mit Federstliben bei statischer und schwingender Verdrehbeanspruchung. Konstruktion 1969; 21: 61-8. [38] Keitel H. Die Rollfeder - ein federndes Maschinenelement mit horizontaler Kennlinie. Drabt 1964; 15: 534-8. [391 Kranz A. Beitrag zur Beschreibung der Eigenschaften geschichteter Trapez- und Parabelfedern, Diss., Uni Hannover 1983. - (40) Kreutzer A. Warmsetzen von Schraubendruckfedem. Drabt 1984; 35: 386-9. - (41) Kuhn P R. Ober den Einfluss der Endwindungen auf das Verhalten schraubenfOrmiger Druckfedern. Drabt 1969; 20: 20612. - (42) Loeper B. Nicht-zylindrische Schraubenfedern im Automobilbau und deren Berechnung. Autom.-techn. Z 1974; 76: 385-90. - [431 Lutz O. Zur Berechnung der Tellerfeder. Konstruktion 1960; 12: 57-9. - (44) Lutz D. Auswahi, Konstruktion und Erprobung von Federungssystemen fUr die Niveauhalrung von Kraftfahrzeugen. Diss., TU, Munich, 1973. - [45] Manual on design and manufacture of coned disk springs or Belleville springs. Special Publications Dept, SAE SP-63. - (46) Merkblatt Stahl no. 394: Fabrgestellfedern (Tragfedern fUr Strassenfahrzeuge und ihre Berechnung). Beratungsstelle fUr Stahlverwertung, DUsseldorf, 1974. - [47) Mubr und Bender. Mubea Tellerfedern-Handbuch. Attendorn, 1987. - (48) Mubr KH, Niepage P. Zur Berechnung von Tellerfedern mit rechteckigem Querschnitt und Autlagetllichen. Konstruktion 1966; 18: 24-7. - [49] Mubr K-H, Niepage P. Eine Methode zur schnellen und einfachen Berechnung von Tellerfedern mit Autlaget1lichen. Konstruktion 1967; 19: 109-11. - [50] Muhr K-H. Einfluss von Eigenspannungen auf das Verhalten von Federn aus Stahl unter schwingender Beanspruchung. Stahl u Eisen 1968; 88: 1449-55 and 1970; 90: 631-6. - [51] Mubr K-H, Niepage P. Ober die Reduzierung der Reibung in Tellerfedernsliulen. Konstruktion 1968; 20: 414-17. - [52) Mubr K-H et al. Warmvorsetzen vermindert die Relaxation von Tellerfedern. Konstruktion 1975; 27: 468-71. - [53) Murasch p-J. Kreuzfederelemente als Gelenke fUr kleine Schwenkbewegungen, ihr kinematsches Verhalten und ihre Belastbarkeit. Diss., TU Berlin, 1972. - [54] Niepage P. Beitrag zur Frase des Ausknickens axial belasteter Schraubendruckfedern. Konstruktion 1971; 23: 19-24. - [55) Niepage P. Zur rechnerischen Abschlitzung der Lastspannungen in angebogenen Schraubenzugfeder-Osen. Drabt 1977; 28: 9-14, 101-8. - [56] Niepage P, Mubr K-H. Nutzwerte der Tellerfeder im Vergleich mit den Nutzwerten anderer Federarten. Konstruktion 1967; 19: 126-33. - [57] Ott) D. Schwingungen mechanischer Systeme mit Strukturdlimpfung. VDI-Forschungsheft no. 603. VDI-Verlag, Dtisseldorf, 1981. - [58) Palm J, Thomas K. Berechnung gekrilmmter Biegefedern. VDI-Z 1959; 101: 301-08. - [59) PalmJ. Gekriimmte Trapez-Biegefedern. VDI-Z 1960; 102: 653-59. - (60) Schremmer G. Ober die dynamische Festig-
keit von Tellerfedern. Diss., TU Brunswick, 1965. - (61) Schremmer G. Dynamische Festigkeit von Tellerfedern. Konstruktion 1965; 17: 473-79. - [62] Schremmer G. Die gescWitzte Tellerfeder. Konstruktion 1972; 24: 226-29. (63) Svoboda Z. Zur Berechnung von Federslitzen. Drabt 1969; 20: 592-98. - (64) Tautenhahn W, Otto P. Parahelfeder mit progressiver Krilmmung fUr Nutzfahrzeuge. ATZ 1983; 85: H. 7/8. - [65] U1brichtJ. Progressive Schraubendruckfeder mit verlinderlichem Drabtdurchmesser fUr den Fahrzeugbau. ATZ 1969; 71: H.6. - (66) Ulbricht J. Volumennutzungsgrad von warmgeformten Federn. Estelberichte 1974; 9: H. 3; see also lnd Anz 1974; 96: 16638. - [67] Wahl AM. Mechanische Federn. Triltsch, Dusseldorf, 1966. - (68) Walz K. Werkstoffprobleme bei Federn. Drabt 1968; 19: 604-12, 783-97. - [69] Walz K. Tellerfedem aus Kunststoff. KEM 1969; 6: 58, 61-3. - [70] Walz K. Korrosionsprobleme bei Federn. Technica 1974; no. 25. - [71] Walz K. Stahlfedern und Werkstoffprobleme. Drabt 1976; 27: 91-8. - [72] Walz K. Gestalrung von Tellerfedersliulen fUr den Werkzeug-, Vorrichtungs- und Maschinenbau. Bleche-Rohre-Prome 1976; 23: 134-9. - (73) Wanke K. Korrosionsschutz von Fedem. Drabt 1964; 15: 103-13. - [74] Wanke K. Beitrag zum Vorsetzen (Voreinrichten) von Schraubenfedern bei Raumtemperatur bzw. erhOhten Temperaturen (Warmsetzen). Drabt 1964; 15: 309-17. - [75] WernitzW. Die Tellerfeder. Konstruktion 1954; 6: 361-76. - [76] Wernitz W. Neuere Erkenntnisse tiber Eigenschaften von Tellerfedern. TH Brunswick: JMF-Bericht 1960; 41. - [77] Wiesecker-Krieg J. Federstlihle, ROchling-Burbach Techn Mitt 1972; 31. [SO] Battermann W, KOhler R. Elastomere Federung - Elastische Lagerungen. Ernst & Sohn, Berlin, 1982. - [81] Becker G. W. et al. E1astische und viskose Eigenschaften von Werkstoffen. Deutscher Verband fUr Materialpriifung (VDM). Beuth, Berlin, 1963. - [82] Benz W. E1astische Lagerung auf geneigt angeordneten Gummipuffern. MTZ 1967; 28: 28-34. - (83) Gamer U. Genaue Berechnung der Gummi-Torsionsfeder. Forsch Ing Wes 1973; 39: 1316. - (84) GObel E F. Gummifedern, Berechnung und Gestaltung, 3m edn. Konstruktions-Bucher, vol. 7. Springer, Berlin, 1969. - [85] GObel E F. Gummifedern als moderne Konstruktionselemente. Konstruktion 1970; 22: 402-6. [86] Joos R. Obersicht tiber verschiedene Dlimpfungsmechanismen. Feinwerktechnik/Messtechnik 1976; 84: 21928. - [87] JOrn R. Lang G. Gununi-Metall-E1emente zur elastischen Lagerung von Motoren. MTZ 1968; 29: 2528. - [88] Kiimmlee H. Ein Verfahren zur Vorhersage des nichtlinearen Steifigkeits- und Dlimpfungsverhaltens sowie der Erwlirmung drehelastischer Gummikupplungen bei stationiirem Betrieb. Diss., TU Berlin, 1985 und VDIFortschrittsber 11136, VDI-Verlag, Dusseldorf, 1986. [89] lipinski J. Fundamente und Tragkonstruktion fUr Maschinen. Bauverlag, Wiesbaden, 1972. - [90] Malter G, JentzschJ. Zur Abhlingigkeit des E- bzw. des G-Moduls von der Beanspruchung. Plaste Kautschuk 1975; 22: 30-2. [91] Malter G, Jentzsch J. Gummifedern als Konstruktionselement. Maschinenbautechnik 1976; 25: 109-12, 121, 225-8. - [92] Pinnekamp W, JOrn R. Neue Drehfederelemente aus Gununi fUr elastische Kupplungen. MTZ 1964; 25: 130-5. - [93] Walz K. Tellerfedern aus Kuntstoff. KEM 1969; 6: 58,61-3. - [94] Continental-Schwingmetall: Katalog der Continental Gununi-Werke AG, D·3OO0 Hannover. - (95) Gimetall: Katalog der Firma MetzeIer AG, D8990 lindau. - (96) Phoenix-Metallgununi: Katalog der Phoenix Gununiwerke AG, D-21oo Hamburg 90. - [97] Simrit. Catalogue no. 400, January 1976. (Werkstoffbeschreibungen) and Katalog Megulastik der Firma Freudenberg, D-6940 Weinheim. - [98] Werkstoff-
Mechanical Machine Components • 12 References
Gummi. Yeroffentlichung der Datwyler AG, Schweizerische Kabel, Gummi- und Kunststoffwerke, CH-6460 Altdorf-Uri. - [100] Hansen J. Faserverbundwerkstoffe, vol. 3. Dokumentation des BMFI'. Springer, Berlin, 1986. [101] Mallik PK. Static mechanical performance of composite elliptic springs. Trans ASME J Eng Mater Technol 1987; 109: 22-6. - [102] Ophey L. Faser-Kunststoff-Yerbundwerkstoffe. YDI-Z 1986; 128: 817-24. - [103] Schiitz D. et al. Werkstoffmechanik (Faserverbundwerkstoffe). In: LBF-Bericht no. TB-I08, 1988. - [104] Publication of GKN Yandervell Ltd, London SWIA IDB, 1987. - [105] Publication on Composite Springs, Hoesch-Iscar Faserverstarkte Fedem GmbH, Eisenstadt/Osterreich, 1987. [106] Kunststoff.Fedem (GFK), Krupp Briininghaus GmbH, Werdohl, 1987. - [110] Behles F. Zur Beurteilung der Gasfederung. ATZ 1961; 63: 311-14. - [111] Die Gasfeder. Technical information from the Stabilus GmbH, Koblenz, 1983. - [112] Harnaekers A. Entkoppelte Hydrolager als Losung des Zielkonflikts bei der Auslegung von Motorlagem. Automobil Ind 1985; 5: 553-60. - [113] Reimpell J. C. Fahrwerktechnik, vol. 2. Vogel, Wiirzburg, 1975, p. 207. - [114] SpurkJ H. Andra R Theorie des Hydrolagers. Automobillnd 1985; 5: 553-60. Standards and Guidelines. DIN-Taschenbuch 29: Fedem,
Normen. Beuth, Berlin, 1985. - DDR-Standard TGL 39249: Ermiidungsfestigkeit; Schraubenfedem; Blattfedem; Kennwerte und Diagramme. - DIN-YDE-Taschenbuch 47: Kautschuk und Elastomere. Physikalische Priifverfahren, 5th edn, Beuth, Berlin, 1988. - DIN 740 Pt 2: Nachgiebige Wellenkupplungen: Begriffe und Berechnungsunterlagen, August 1986. Beuth, Berlin. - DIN 53440: Priifung von Kunststoffen und von s('hwingungsgedampften geschicht· eten Systemen - Biegeschwingungsversuch. Pt I: Allgem· eine Grundlagen zur Bestimmung der dynamisch-elastischen Eigenschaften stab· und streifenformiger Probekorper; Pt 2: Bestimmung des komplexen Elastizit· atsmodul; Pt 3: Bestimmung von Kenngrossen schwin· gungsgedampfter Mehrschichtsysteme. January 1984. Beuth, Berlin. - DIN 53445: Priifung von polymeren Werkstoffen: Torsionsschwingungsversuch. August 1986. Beuth, Berlin. - DIN 53505: Priifung von Kautschuk, Elas· tomeren und Kunststoffen; Hatepriifung nach Shore A und
D. June 1987. Beuth, Berlin. - DIN 53 513: Priifung von Kautschuk und Elastomeren. Bestimmung von visko-elastischen Eigenschaften von Elastomeren bei erzwungenen Schwingungen ausserhalb der Resonanz. January 1983. Beuth, Berlin. - DIN 53531 Paper I: Priifung von Elasto· meren; Trennversuch an Elastomer-Metall-Yerbindungen; Priifung an einer Metallplatte. December 1972; Pt 2: Prii· fung von Kautschuk und Elastomeren; Trennversuch an Elastomer-Metall-Bindungen. June 1981. Beuth, Berlin. DIN 53 533: Priifung von Elastomeren; Priifung der Warmebildung und des Zermiirbungswiderstandes im Dauerschwingversuch (Flexometerpriifung). Pt 1: Grundlagen; Pt 2: Rotationsflexometer; Pt 3: Kompressions-Flexometer. August 1975, Beuth, Berlin. - ISO 2856-1975 (E): Elastomers - General requirements for dynamic testing. International Organization for Standardization, Geneva, 1975. - ISO/TC 108-DB 5405: Nomenclature for Specifying Damping Properties of Materials. ISO/TC 108 (Secr. 108) 185. August 1975. - ISO/TC 61/WG 2 - Draft Proposal: Plastics - Terminology for Characterizing the Damping Properties of Solid Polymers. American National Standards Institute (ANSI). September 1975. - YDI-Richtlinie 2062: Schwingungsisolierung; Paper 1: Begriffe und Methoden; Paper 2: Isolierelemente. January 1976. Beuth, Berlin.
'MIi'
F3 Couplings, Clutches and Brakes. [1] VOI-Richtiinie
2240: Wellenkupplungen. YDI-Yerlag, Diisseldorf, 1971. [2] Peeken H, Troeder C. Elastische Kupplungen. Springer, Berlin, 1986. - [3] Hinz R Yerbindnngselemente: Achsen, Wellen, Lager Kupplnngen. YEB Fachbuchverlag, Leipzig, 1984. - [4] Schalitz A. Kupplnngsatlas, 4th edn, Ludwigsburg: AGT-Yerlag 1975. - [5] Ehrlenspiel K, Henkel G. Membrankupplnngen als drehstarre, biegenachgiebige Ganzmetallkupplungen. VOlBerichte 299, 1977. - [6] Buschhaus D. Rechnergestiitzte Auswahl von schaltbaren Wellenkupplungen. Diss., TU Berlin, 1976. - [7] Tochtermann W, Bodenstein F. Konstruktionselemente des Maschinenbaus, Pt 2. Springer, Berlin, 1979. - [8] YDI-Richtlinie 2722: Homokinetische Kreuzgelenkgetriebe einschliesslich Gelenkwellen. YDIVerlag, Diisseldorf, 1982. - [9] Dittrich 0, Schumann R. Anwendungen der Antriebstechnik, vol. II: Kupplnngen. Krausskopf, Mainz, 1974. - [10] Schmelz F, Grafv. SeherrThoss H-C, Aucktor E. Gelenke und Gelenkwellen. Springer, Berlin, 1988. - [11] Hartz H. Antriebe mit Kreuzgelenkwellen, Pt 1: Kinematische nnd dynamische Zusammenhange. Antriebstechnik 1985; 24: 72-5. - [12] Hartz H. Antriebe mit Kreuzgelenkwellen. Pt 2: Probleme und ihre LOsungen. Antriebstechnik 1985; 24: 61-9. - [13] Schutz K H. Gleichlauf-Kugelgelenke fur Kraftfahrzeugantriebe. Antriebstechnik 1971; 10: 437-40. - [14] Benkler H. Zur Auslegung bogenverzahnter Zahnkupplnngen. Konstruktion 1972; 24: 326-33. - [15] Fleiss R, Pahl G. Radial- und Axialkrafte beim Betrieb von Zahnkupplnngen. VOI-Berichte 1977; 299 - 116] Heinz R. Untersuchnng der Zahnkraftnnd Reibungsverhiiltnisse in Zahnkupplnngen. Konstruktion 1978; 30: 483-92. - [17] Pahl G, Strauss E, Bauer H P. Fresslastgrenze nichtgemrteter Zahnkupplungen. Konstruktion 1985; 37: 109-16. - [18] Pahl G, Muller N. Temperaturverhalten olgefiillter Zahnkupplungen. VOI-Berichte 1987; 649: 157-77. - [19] Stotko H. Moderne Entwicklungen bei Bogenzahn-Kupplungen. Konstruktion 1984; 36: 433-7. - [20] Basedow G. Zahnkupplungen fur hohe Drehzahlen. Antriebstechnik 1984; 23: 18-21. - [21] JarchowF, SturmathR. Tragfahigkeit und Federsteifen von Wellenkupplungen mit fedemden Laschengelenken. Konstruktion 1979; 31: 33-40. [22] Henkel G. Membrankupplnngen - Theoretische und experimentelle Untersuchnng ebener nnd konzentrisch gewellter Kreisringmembranen. Diss., Univ. Hannover 1980. - 123] Roper R, Japs D. Bestimmung der statischen Momentkennlinie elastischer Wellenkupplungen. Antriebstechnik 1980; 19: 403-7. - [24] Beitz W. Untersuchungen der elastischen und dampfenden Eigenschaften drehelastischer Kupplungen nnd ihre Dauerfestigkeit. Diss., TU Berlin, 1961. - [25] Klingenberg R. Experimentelle und analytische Untersuchungen des dynamischen Yerhaltens drehnachgiebiger Kupplungen. Diss., TU Berlin, 1977. - [26] Gnilke W. Zur Grossenauswahl drehnachgiebiger Kupplungen. Maschinenbautechnik 1982; 31: 537-40. - [27] Benner J. Experimentelle Untersuchnngen des mechanischen Yerhaltens drehnachgiebiger Wellenkupplungen und Entwicklung eines Ersatzmodells. Diss. RWTH Aachen, 1984. - [28] Kiimmiee H. Ein Yerfahren zur Yorhersage des nichtlinearen Steifigkeits- und Dampfungsverhaltens sowie der Erwarmnng drehelastischer Gummikupplungen bei stationarem Betrieb. Fortschritt-Berichte YDI series I, no. 136. VOI-Yerlag, Diisseldorf, 1986. - [29] Peeken H et al. Angeniiherte Bestimmung des Temperaturfeldes in elastischen Reifenkupplungen. Konstruktion 1986; 38: 485-9. - [30] Troeder C et al. Berechnungsverfahren von Antriebssyste.. men mit drehelastischer Kupplnng. YDI-Berichte 1987;
141ij:·
Mechanical Machine Components • 12 References
649: 41-68. - [31] Hartz H. Anwendungskriterien fur hochdrehelastische Kupplungen. Pt 1: Antriebsarlen und deren Besonderheiten. Antriebstechnik 1986; 25: 47-52. [32] Peeken H et al. Beanspruchung elastischer Kupplungen in Antriebssystemen mit Asynchron-Motoren. Antriebstechnik 1979; 18: 484-9. - [33] Peeken H, Troeder C. Auswirkungen des Wellenversatzes bei elastischen Kupplungen. VDI-Berichte 1977; 299. - [34J Heyer R, Mollers W. Ruckstellkriifte und -momente nachgiebiger Kupplungen bei Wellenverlagerungen. Antriebstechnik 1987; 26: 43-50. - [35J VDI-Richtlinie 2241 Paper 1: Schaltbare fremdbetatigte Reibkupplungen und -bremsen. VDI-Verlag, Dusseldorf, 1982. - [36J VDI-Richtlinie 2241 Paper 2: Schaltbare fremdbetatigte Reibkupplungen und bremsen. VDI-Verlag, Dusseldorf, 1984. - [37J Niemann G, Winter H. Maschinenelemente, vol. III, 2nd edn. Springer, Berlin, 1983. - [38J Winkelmann S. K1assenmerkmale, Anwendungsfelder und Trends bei schaltbaren, mechanischen Kupplungen. VDI-Berichte 1987; 649: 27387. - [39J Orthwein W. Clutches and brakes. Dekker, New York, 1986. - [40] Appelhoff H. ElektromagnetZahnkupplungen und Periflex-Wellenkupplungen fur die Hutten- und Schwermaschinenindustrie. Antriebstechnik 1986; 25: 31-4. - [41J Winkelmann S, Harmuth H. Schaltbare Reibkupplungen. Springer, Berlin, 1985. - [42J Fedem K, Beisel W. Betriebsverhalten nasslaufender Lamellenkupplungen. Antriebstechnik 1986; 25: 47-52. [431 Korte W. Betriebs-- und Leerlaufverhalten von nasslaufenden Lamellenkupplungen. VDI-Berichte 1987; 649: 335-58. - [44 J Korte W, Ruggen W. Magnetpulverkupplungen. asr-digest fur angewandte Antriebstechnik 1979; 3: 47-9. - [45J Hasselgruber H. Der Schaltvorgang einer Trockenreibungskupplung bei kleinster Erwarmung. Konstruktion 1963; 15: 41-5. - [46J Duminy J. Beurteilung des Betriebsverhaltens schaltbarer Reibkupplungen. Diss., TU, Berlin, 1979. - [47J SteinhilperW. Derzeitliche Temperaturverlauf in schnell geschalteten Reibungskupplungen und -bremsen. Diss., TH Karlsruhe, 1963. - [48] Steinhilper W. Dcr Kraftfluss in unter Last geschalteten Lamellen-Kupplungen und das ubertragbare Drehmoment. Konstruktion 1967; 7: 262-7. - [491 PaW G, Zhang Z. Dynamische und thermische Ahnlichkeit in Baureihen von Schaltkupplungen. Konstruktion 1984; 36: 421-6. - [50J Pahl G, Oedekoven A. Kennzahlen zum Temperaturverhalten von trockenlaufenden Reibungskupplungen bei Einzelschaltung. VDI-Berichte 1987; 649: 289-306. - [51 J Ernst L, Ruggen W. Richtige Auswahl von Kupplungen und Bremsen. Antriebstechnik 1982; 21: 616-19. - [52] Schneider R. Elektromagnetische Hystcresekupplung. VDIBerichte 1987; 649: 435-47. - [53J Hoppe F. DasAbschaIt- und Betriebsverhalten von mechanischen Sieherheitskupplungen. Diss., TU Munich, 1986. - [54] Rettig H, Hoppe F. Sicherheitskupplung mit Brechringen fur Schwermaschinenantriebe. Antriebstechnik 1986; 25: 4853. - [55 J Fleissig M. Untersuchungen wm Drehmomentverhalten von Fliehkraftkupplungen. VDI-Z 1984; 126: 869-72. - [56 J Korber E. Hydrodynamische Anlaufund Rutschkupplung mit konstanter Filliung. VDI-Berichte 1979; 299: 171-7. - [57J Stolzle K, Hart S. Freilaufkupplungen. Springer, Berlin, 1960. - [58] Timtner K. Freilaufkupplungen fur zukunftsorientierte Anwendungen. Antriebstechnik 1986; 25: 31-5. - [59J Jorden W. Gebrauchsdauer von K1emmfreilaufkupplungen. Konstruktion 1972; 24: 485-91. - [60J Timtner K. Die Berechnung der Drehfederkennlinien und zulassigen Drehmomente bei Freilaufkupplungen mit K1emmkorpem. Diss., TH, Darmstadt, 1974. - [61 J Peeken H, Hinzen H. Funktionsfahigkeit und Gebrauchsdauer von K1emmkorperfreilaufen im
Schaltbetrieb. Antriebstechnik 1986; 25: 35-40. - [62J Schlattmann J. Lebensdauerermittlung von K1emmrollenfreilaufen aufgrund von Werkstoff-Verformung, -Ermudung und Wiilzverschleiss. Fortschritt-Berichte VDI Reihe 5 no. 200. VDI-Verlag, Dusseldorf, 1986. - [63J Tonsmann A. Verschleiss und Funktion Der Einfluss des Schaltverschleisses auf die Schaltgenauigkeit von K1emmrollenfreilaufen. Diss., Univ. Paderbom, 1989. [64 J Bollmann E. Die Eintouren-Rollenkupplung - ein vielseitiges Schaltelement. Antriebstechnik 1973; 12: 101-06.
Standards
and Guidelines. DIN 115: Schalenkupplungen. - DIN 116: Scheibenkupplungen. - DIN 740: Nachgiebige Wellcnkupplungen. - DIN 15431-15437: Trommel- und Scheibenbremsen. - DIN 42955: Toleranzen fur Befestigungsflansche fur elektrische Maschinen, zulassige Lageabweichungen. - DIN 71751-71 752: Gabelgelenke. - DIN 71 802; 71 803; 71805: Winkelgelenke. AGMA Standards: 510.03: Nomenclature of flexible couplings. - 516.01: Metric dimensions for gear coupling flanges. - 900O-C90: Flexible couplings - potential unbalance classification. - 9001-A86: Lubrication of flexible couplings. - 9002-A86: Bores and keyways for flexible couplings (inch series). - 9003-A91: Flexible couplings keyless fits. F4 Rolling Bearings. [1 J Schlieht H et al. Ermudung bei Wiilzlagem und deren Beeinflussung durch Werkstoffeigenschaften. FAG-Walzlagertechnik; 1987-1. - [2] Stocklein W. Aussagekraftigc Berechnungsmethode zur Dimensionierung von Walzlagern. Walzlagertechnik. Pt 2: Berechnung von Lagerungen und Gehausen in der Antriebstechnik. Kontakt und Studium; B. 248. Expert-Verlag, Grafenau, 1988. - [3 J FAG Standardprogramm. Katalog WL 41510/2 DB. 1987. - [4] SKF Hauptkatalog. Katalog 4000 T. 1989. [5J Eschmann P. Das Leistungsvermogen der Walzlager. Springer, Berlin, 1964. - [6J FAG Kugelfischer George Schafer: Schmierung von Walzlagem. Pub!. no. WL 81115 DA; 1985 edition. [7 J Druckschrift SKF (Werkzeugmaschinenlager). - [8 J JOrgensmeyer: Gestaltung von Walzlagerungen. Springer, Berlin, 1953. - [9J Miinnich H. Auswirkungen e1astischer Verformungen auf die Kraiteinleitung in Walzlagern. Kugellager-Z. no. 155, 1'1'.3-12. - [IOJ Sommerfeld H, Schimion W. Leichtbau von Lagcrgehausen durch gunstige Krafteinleitung. Z. Leichtbau der Verkehrsfahrzeuge 1969; H. 3: 3-7.
Company documents: FAG, Schweinfurt. - INA, Herzogenaurach. - NSK, Ratingen. - NTN, Erkrath-Unterfeldhaus. - SKF, Schweinfurt. - SNR, Stuttgart. Standards and Guidelines: DIN-Taschenbuch no. 24: Wiilzlager, 5th edn, Beuth, Berlin, 1985. - DIN 611: Vbersiehl uber das Gebiet der Wiilzlager. - DIN 615: Schulterkugellager; ISO 15. - DIN 616: Massplane; ISO 104. DIN 617: Nadellager mit Kafig; ISO 15. - DIN 618: Nadelhii1sen-Nadelbuchsen; ISO 3245. - DIN 620: Toleranzen; ISO 15. - DIN 622: Tragfahigkeit von Walzlagem. DIN 623: Bezeichnungen. - DIN 625: Rillenkugellager. DIN 628: Schragkugellager. - DIN 630: Pendelkugellager; ISO 15. - DIN 635: Tonnenlager-Pendelrollenlager; ISO 15. - DIN 711: Axial-Rillenkugellager; ISO 104. - DIN 715: zweiseitige Axial-Rillenkugellager; ISO 104. - DIN 720: Kegelrollenlager; ISO 355. - DIN 722: Axial-Zylinderrollenlager; ISO 104. - DIN 728: Axial-Pendelrollenlager; ISO 104. - DIN 736-739: Stehlagergehause fur Wiilzlager; ISO 113-2. - DIN 981: Nutmuttern; ISO 2982. - DIN 4515: Spannhiilsen; ISO 1012. - DIN 5401: Kugeln; ISO 3290. DIN 5402: Zylinderrollen-Walzen-Nadeln; ISO 3096. DIN 5404: Axial-Nade1kranze. - DIN 5405: Radial-Nadelk-
Mechanical Machine Components • 12 References
ranze; ISO 3030, 3031. - DIN 5406: Sichenmgsbleche; ISO 2982. - DIN 5407: Walzenkranze. - DIN 5412: Zylinderrollentager; ISO 15. - DIN 5416: Abziehhiilsen; ISO 113-1. DIN 5417: Sprengringe; ISO 464. - DIN 5418: Anschlussmasse. - DIN 5419: Filzringe-Ringnuten fur Walzlagergehause. - DIN 5425: Passungen flir den Einbau. DIN 51825: Wiilzlagerfette. - DIN-ISO 76: Statische Tragzahlen. - DIN-ISO 281: Dynamische Tragzahlen. - DIN ISO 355: Metrische Kegelrollenlager.
F5 Plain Bearings. [IJ Vogelpohl G. Betriebssichere Gleitlager. Springer, Berlin, 1967. - [2J Spiegel K. Ober den Einfluss elastischer Deformationen auf die Tragfahigkeit von Radialgleitlagem. Schmiertechnik und Tribologie 1973; 20: 3-9. - [3J VDI-Richtlinien 2204 E Papers 1-4: Diisseldorf, 1990. - [4J Rodermund H. Berechnung der Temperaturabhangigkeit der Viskositiit von Mineraliilen aus dem Viskositatsgrad. Schmiertechnik und Tribologie 1978; 25: 56-7. - [5 J Hakansson B The journal bearing considering variable viscosity. Report no. 25. Inst. of Machine Elements, Chalmers Univ. of Technology, Gotesborg, Sweden, 1964. - [6J BMFT Projektleitung Materialund Rohstofforschung (ed.): Tribologie: Reibung Verschleiss - Schmienmg. B. 1 to 12. Springer, Berlin, 1988. - [7J Roemer E. Oldurchsatz, Oltemperatur und Lagerspiel von Gleitlagem mit Druckschmierung. VDI-Z 1961; 103: 743-7 and 790-4. - [8] Roemer E. Der Eintluss der Temperatur auf das Lagerspiel eines Gleitlagers. Konstruktion 1961; 13: 262-7. - [9] Noack G. Berechnung hydrodynamisch geschmierter Gleitlager dargestellt am Beispiel der Radiallager. Gleitlagertechnik I. Tribotechnik vol. 49. Expert-Verlag, Grafenau, 1981. - [10] Peeken H. Zustandsschaubild flir Gleitlager. Konstruktion 1968; 20: 169-76. - [11] Butenschon H ]. Das hydrodynamische, zylindrische Gleitlager endlicher Breite unter instationarer Belastung. Diss., Univ. Karlsruhe, 1976. - [12] Glienecke ], Han 0 C. Gleitlager-Turbulenz. Forschungsber. der Forschungsvereinigung Verbrennungskraftmaschinen (FVV). H. 265. Frankfurt, 1983. - [13] Fricke]. Das Axiallager mit kippbeweglichen Kreisgleitschuhen. VDI-Forschungsheft 567. Diisseldorf, 1975. [14] Rost U. Die Berechnung des ebenen Kreisgleitschuhs. Ing.-Arch. 1969; 38: 1-14. - [15] Fricke J. Berechnung und Auslegung von hydrodynamischen Axialgleitlagem. Gleitlagertechnik 2, Tribotechnik, vol. 163. Expert-Verlag, Gf'dfenau, 1986. [16J Peeken H, Knoll G. Zylindrische Gleitlager unter elastohydrodynamischen Bedingungen. Konstruktion 1975; 27: 176-81. - [17] Peeken H. Rechnenmterstiitzte Konstruktion von Maschinengehiiusen zur Optimierung von Steifigkeit, Festigkeit und Betriebssicherheit hydrodynamischer Gleitlager. Konstruktion 1982; 34: 229-38. - [18J Droste K. Schmienmgsgerechte Konstruktion. VDI-Ber 1%6; 111: 15-19. - [19] VDI Richtlinie 2201: Diisseldorf. - [20] Lang 0 R, Steinhilper W. Gleitlager. Springer, Berlin, 1978. - [21 J Hilgers W. Abhangigkeit der Lagerwerkstoffeigenschaften in Verbundlagem von der Stiitzkorperkonstruktion. VDI-Ber 1975; 248: 149-58. - [221 Roemer E. Lagerschalen aus Bandmaterial flir die Motorenindustrie. Auto-lndustrie 1961; 3-7. - [23] Hilgers W. Erkennung der Ursache von Schiiden an dickwandigen Verbundlagem. Goldschmidt informiert 1978; 3: H. 45, 7089. [24] Hilgers W. Lagermetalle. Goldschmidt informiert 1970; 2: H. 11, 2-24. - [25] Peeken H, Salm T. Drittschicht-Dauerfestigkeit. Forschungsber. der Forschungsvereinigung Verbrennungskraftmaschinen (FVV). H. 403. Frankfurt, 1987. - [26] VDI-Richtlinie 2203: Dusseldorf, 1970. - [27] Hilgers W. Lagerwerkstoffe flir hohere Forderungen. VDI-Nachrichten 1975; no. 38. [28] Pollmann E. Das Mehrgleittlachenlager unter Beriick-
'iMP@'
sichtigung der veranderlichen Olviskositat. Konstruktion 1%9; 21: 85-97. - [29] Frossel W. Berechnung von Gleitlagem mit radialen Gleittliichen. Konstruktion 1962; 14: 169-80. - [30] Gersdorfer o. Tragkraft und Anwendungsbereich von MehriIiichenlagem. Konstruktion 1962; 14: 181-8. - [31] Ott H H. Kippsegment-Radiallager mit Schmiermittel von veranderlicher Viskositat. VDI-Ber 1975; 248. - [32J Muyderman E A. Constructions with spiral-groove bearings. Wear 1966; 9: 18-141. - [33] Hiibner W, Hallstedt G. Berechnung und Anwendung von Spiralrillen-Kalottenlager. Konstruktion 1972; 24: 393-7. [34] Hentschel G. Hochbelastbare Trockengleitlager. Antriebstechnik 1976; 15: 522-8. - [35] Erhard G, Strickle E. Gleitelemente aus thermoplastischen Kunststoffen. Z. Kunststoffe 1973; 63: 4-5. - [36] Hachmann H, Strickle E. Polyamide als Gleitlagerwerkstoffe. Konstruktion 1964; 16: 4. - [37J Detter H, Holocek K. Der Reibwiderstand und die Beanspruchung von feinmechanischen Lagem im Trockenlauf bei kleinen Gleitgeschwindigkeiten. Feinwerktechnik 1970: 74: H. 11. - [38] BASF: Kunststoffe in der Priifung. Werkstoffblatt 3110, 1 October 1975. - [39] Kayser H D. Hinweise zur Dimensionierung von Gleitlagem aus Kunststoff oder Kunstpressstoffen im Trockenlauf. VDI-Ber 1975; 248. - [40] Hentschel G. Wartungsfreie Gelenk- und Schwenklager (Oszillationslager). VDI-Ber 1975; 248: 137-42. - [41] Dietz R, Herfeld]. Die Berechnung hydrostalischer Radiallager flir die Drehzahl n = 0 unter Beriicksichtigung veranderlicher Spalthohe. Maschinenmarkt 1964: H. 3, 18-24. - [42] Rasmus W. Hydrostatische Anfahrhilfe. Goldschmidt-Mitt. 1974; H. 30, 46/47. - [43] Peeken H, Benner ]. Berechnung von hydrostatischen Radial- und Axiallagem. Goldschmidt informiert, Gleitlagertechnik 1984; 2: H. 61, 42-148. [44] Peeken H, Heil M. Das optimale hydrostatische Axiallager. Konstruktion 1972; 24: 381-6. - [45] Heil M. Die Auslegung optimaler hydrostatischer Axiallager flir hohe Tragiasten. Konstruktion 1974; 26: 227-31.
Standards and Guidelines: DIN 38: Lagermetallausguss in Gleitlagem. - DIN 118: Stehgleitlager mit Ringschmienmg. - DIN 149: gerollte Buchsen flir Gleitlager. DIN 322: Schmierringe. - DIN 502/3: Flanschlager. DIN 504: Augenlager. - DIN 505/6: Deckellager. DIN 648: Gelenklager: ISO 7124. - DIN 1591: Schmierlocher - Schmiemuten - Schmiertaschen: ISO 12 128. DIN/ISO 4384: Hatepriifung an Lagermetallen. DIN/ISO 4386: Priifung der Bindung metallischer Verbundgleitlager. - DIN/ISO 6279, 4281/2/3: Lagerwerkstoffe. - DIN 7473174: Gleitlager ungeteilt/geteilt mit Lagermetallausguss. - DIN 7477: dazu Schmiertaschen. DIN 8221: Buchsen flir Gleitlager nach DIN 502/3/4. DIN 31 651: Gleitlagerkurzzeichen und Benennungen; ISO 7904-1. - DIN 3 I 652: Berechnung von hydrodynamischen Radial-Gleitlagem; ISO 7902-1. - DIN 31 654: Hydrodynamische Axial-Gleitlager im stationaren Betrieb. DIN 31661: Schiiden. - DIN 31670: Qualitatssicherung von Gleitlagem. - DIN 31 690: Gehausegleitlager. DIN 31 692: Schmierung. - DIN 31 696: Segmentaxiallager. - DIN 31 697: Ring-Axiallager. - DIN 31 698: Gleitlager-Passungen; ISO 12129-1. - DIN 50282: Gleitverhalten von Werkstoffen. DIN 71 420/24: Zentralschmierung. F6 Belt and Chain Drives. [11 Dittrich 0 et al. Anwendungen der Antriebstechnik, vol. III. Krausskopf, Mainz, 1974. - [2] Halbmann W. Zum Schlupf kraftschliissiger Umschlingungsgetriebe. VDI-Fortschrittsber series 1, no. 145. VDI-Vcrlag, Diisseldorf, 1986. - [3] Neu K. Untersuchungen zum Betriebsverhalten offener und sdbstspannender Flachriemengetriebe. Diss., Dniv. Stuttgart, 1979. -
,.,1,.
Mechanical Machine Components. 12 References
[ 4] Siegling: 3000 Hannover 1 (publications on highstrength flat belts, transfer and process drives, spindle drives, corrugated and conveyor belts). - [5] Milller H W. Anwendungsbereiche der Keilriemen in der Antriebstechnlk. In: Amtz-OptibelHiruppe Hoxter: Keilriemen. Heyer, Essen, 1972. - [6] Continental: 3000 Hannover 1 (publications on V-belts, V-ribbed belts, toothed belts and HTD belts). - [7] Optibelt: 3740 Hoxter 1 (publications on transmission components, ribbed belts). - [8] Milller H W. Zugmittelgetriebe. In: Dubbel: 16th edn. Springer, Berlin, 1987. [9] PireUi: 8752 Kleinostheim (publications on toothed belts). - [10] Mulco: 3000 Hannover 1 (publications on toothed belts). - [11] Handbuch Antriebstechnlk: Table of manufacturers and product data. Krausskopf: appears annuaUy. F7 Friction Drives. [I] VOI-Richtlinie 2155: Gleichformig iibersetzende Reibschlussgetriebe, Bauatten und
Kennzeichen. VOl-Verlag, Diisseldorf, 1977. See also "Note on Standards" at the end of this section. - [2] Lutz O. Grundsatzliches iiber stufenios versteUbare Wiilzgetriebe. Konstruktion 1955; 7: 330-5, 1957; 9: 169-71, 1958; 10: 425-7. - [3] Overlach H, Severin D. Berechnung von Wiilzgetriebepaarungen mit ellipsenformigen Beriihrungsflachen und ihr Verhalten unter hydrodynamischer Schmierung. Konstruktion 1966; 18: 357-67. - [4] Gaggermeier H. Untersuchungen zur Reibkraftiibertragung in RegelReibradgetrieben im Bereich elasto-hydrodynamischer Schmierung. Diss., TV Munich, 1977. - [5] Matzat N. Einsatz und Entwickiung von Traktionsfliissigkeiten. Synthetische Schmierstoffe und Arbeitslliissigkeiten. 4th Int. KoU., Technische Akademie Esslingen, January 1984, 16.116.26, paper no. 16. - [6] Johnson K. L., Tevaarwerk]. L. Proc Roy Soc A (1977). - [7] Winter H., Gaggermeier H. Versuche zur Kraftiibertragung in VersteU-Reibradgetrieben im Bereich elasto-hydrodynamischer Schmierung. Konstruktion 1979; 31: 2-6, 55-62. - [8] Niemann G., Winter H. Maschinenelemente, vol. III, 2nd edn, Springer, Berlin, 1983. - [91 Basedow G. Stufenlose Nullgetriebe schiitzen vor Oberlast und Anfahrsttissen. Antriebstechnlk 1986; 25: 20-5. - [10] Scbroebler W. Praktische Erfahrungen mit speziellen Reibradgetrieben. Tech Mitt 1968; 61: 411-14. - [11] Hewko L. O. Roller traction drive unit for extremely quiet power transmission. J Hydronautics 1968; 2: 160-7. - [12] Bauerfeind E. Zur Kraftfibertragung mit Gununiwiilzriidem. Antriebstechnlk 1966; 5: 383-91. F8 Gearing. [I] Niemann G., Winter H. Maschinenele-
mente, vols IT and III, 2nd edn. Springer, Berlin, 1989/6. [2] Dudley D. W., Winter H. Zahnriider. Springer, Berlin, 1961. - [3] Keck K. F. Die Zahnradpraxis, parts 1 and 2. Oldenbourg, Munich, 1956 and 1958. - [4] Winter H. Die tragfiihigste Evolventen-Geradverzahnung. Vieweg, Brunswick, 1954. - [5] Dudley D. W. Gear handbook. McGraw-Hill, New York, 1962. - [6] Richter W. Auslegung profiJverschobener Aussenverzahnungen. Konstruktion 1962; 14: 189-96. - [7] Roth K. Evolventenverzabnungen fiir parallele Achsen mit Ritze1ziiltnezahlen von 1 bis 7. VOI-Z 1965; 107: 275-84. - [8] Piepka E. Eingriffsstorungen bei Evolventen-lnnerverzahnung. VOI-Z 1970; 112: 215-22. - [9] ClarenbachJ et aI. Geometrische Auslegung von zylindrischen lnnenradpaaren - Ertauterung zum Normentwurf DIN 3993. Antriebstechnlk 1975; 651-8. - [10] Erney G. Auslegung von Evolventenlnnenverzahnungen. Antriebstechnlk 1975; 14: 625-9. [II] Naville R. Die Theorie derVerzahnung der Uhtwerktechnlk. Microtechnlc 1967; XXI: 506-9, 587-90. - [12] Niemann G. Novikov-Verzahnung und andere Sonderverzahnungen fiir hohe Tragfiihigkeit. VOI-Ber 1961; 47: 512. - [13] Shotter B A. Experiences with Confomtal/WN-
gearing. World Congress on Gearing, Paris, 1977, vol. I, p. 527. - [14] Dowson D, Higginson G R. Elasto-hydrodynamic lubrication. Pergamon Press, Oxford, 1966. - [15] Johansson M et aI. Austinitisches-bainitisches Gusseisen aIs Konstruktionswerkstoff im Getriebebau. Antriebstechnlk 1976; 15: 593-600. - [16] Winter H, Weiss T. Tragfiihigkeitsuntersuchungen an induktions- und flamrngehiirteten Zahnriidern, parts I and IT. AntriebstechnJk 1988; 27: 4550, 57-62. - [17] Walzel H. Kann das Nikotrierverfahren das Badnitriren ersetzen? TZ fiir prakt Metallbearb 1976; 70: 291-4. - [18] Winter H, Wirth X. Einfluss von Schleifkerben auf die Zahnfussdauertragfiihigkeit oberlliichengebatteter Zahnriider. Antriebstechnik 1978; 17: 37-41. [19] Rhenius K T. Betriebsfestigkeitsrechnungen von Maschinenelementen in Ackerschleppern mit Hilfe von LastkoUektiven. Konstruktion 1977; 29: 85-93. - [20] Rettig H, Plewe H-J. Lebensdauer und Verschleissverhalten langsam laufender Zahnriider. Antriebstechnlk 1977; 16: 357-61. - [21] Krause W. Untersuchungen zur Geriiuschverhalten evolventenverzahnter Geradstirnriider der Feinwerktechnlk. VOI-Ber 1967; 105. - [22] Gavrilenko V A., Bezrukov V I. The geometrical design of gear transmissions comprising involute bevel gears. Russ EngJ 1976; 56: 34-8. - [23] Beam AS. Beveloid gearing. Mach Design 1954; 220-38. - [24] Hiersig H M. Zylinderrader mit Rechts- und Linksllanken von ungteicher Steigung. Konstruktion 1979; 31: 7-11. - [25] Keck K F. Die Bestimmung der Verzahnungsabmessung bei kegeligen Schraubgetrieben mit 90° Achswinkel. ATZ 1953; 55: 302-8. - [26] Coleman W. Hypoidgetriebe mit beliebigen Achswinkeln. Automotive Ind. June 1974. - [27] Richter M. Der Verzahnungswirkungsgrad und die Fresstragfithigkeit von Hypoi
Mechanical Machine Components. 12 References
9. - [45J Jacobsen U A J. Crossed helical gears for high speed automotive applications. Inst Mech Engng Proc. Autom Div 1961/2; 359-84. [46J Rohonyi C. Berechnung profilverschobener. zylindrischer Schraubenrader. Konstruktion 1963; 15: 453-5. [47J Henriot G. Engrenages. Dunod. Paris. 1980. - [48J Seifried A. Ober die Auslegung von Stimradgetrieben. VDI-Z 1967; 109: 236-4 L - [49 J Sicfried A, Burkle R. Die Bertihrung der Zahnflanken von Evolventenschraubenradem, Werkst Betr 1968; 101: 183-7. - [50J Maag-Taschenbuch. MAAG AG, Zurich, 1985. - [51 J Pohl F. Betriebshutte, vol. I, Abschn Kegelradbearbeitung und Maschinen fur Kegelradbearbeitung. Ernst & Sohn, Berlin, 1957. - [52J Ernst H. Die Hebezeuge, vol. I. Vieweg, Brunswick, 1973. - [53 J Cbironis N. P. Gear design and application. McGraw-Hill, New York, 1967; includes papers from: Bloomfield B. Noncircular gears, pp. 158-63; Rappaport S. Elliptical gears of cyclic speed variations, pp. 166-68; Miano S. V. Twin eccentric gears, pp.169-73. [54J Cunningham F, Cunningham D. Rediscovering the noncircular gear. Mach Design 1973; 45: HO-5. - [55J Ludwig F. Verwendung eines Koppelgetriebes zum Herstellen walzverzahnter Ellipsenriider. VOI-ller 1956; 12: 139-44. - [56J Ferguson R.]. et al. The design of a stepless transmission using noncircular gears. Mech Mach Theory 1975; 10: 467-78. [57J Yokoyama Y et al. Dynamic characteristic of the noncircular planetary gear mechanisms with nonunifortn motion. Bull ISME 1974; 17 149-56. - [58J Winter H, Schonnenbeck G. Graufleckigkeit an einsatzgeharteten Zahnriidem: Ertnudung der Werkstoftrandschicht mit moglicherweise schweren Foigeschiiden. Antriebstechnik 1985; 24: 53-61. - [59J MUlIer H W. Die Umlaufgetriebe, Berechnung, Anwendung, Auslegung. Springer, Berlin, 1971. - [6oJ Millier H W. Einheitliche Berechnung von Planetengetrieben. Antriebstechnik 1976; 15: 11-17,859, 145-9. - [61 J MUlIer H W. Programmierte Analyse von Planetengetrieben. Antriebstechnik 1989; 28: [62J Millier H W. Ungleichmiissig ubersetzende Umlaufgetriebe. VDI Fortschrittsber. Series I, 1988; 159: 49-64. - [63J Jarchow F. Entwicklungsstand bei Planetengetrieben. VDIBer 1988; 672: 15-44. - [64J Winkelmann L. Lastverteilung in Planetengetrieben. VDI-Ber 1988; 672: 45-74. [65 J Potthoff H. Anwendungsgrenzen vollrolliger Planetenrad-Wiilzlager. VDI-Ber 1988; 672: 245-64. - [66J MOller H W. Oberlagerungssysteme. VDI-Iler 1986; 618: 59-78. [67J Dreher K. Rechnergestutze Optimierung von Planeten-Koppelgetrieben. Diss., Dartnstadt, 1983. [68J Schnetz K. Reduzierte Planeten-Koppelgetriebe. Diss., Dartnstadt, 1976. - [69J Brass E A. Two stage planetary arrangements for the 15: 1 turboprop reduction gear. ASME Paper 60-SA-l (1%0). - [70J Schoo A. Verz.hnungsverlustleistungen in Planetenradgetrieben. VDI-Ber 1988; 627: 121-40. - [71J MUlIer H W. Anpassung stufenloser Getriebe an die Kennlinie einer Maschine. Also: Optimierung der Grundanordnung stufenloser Stellgetriebe. Maschinenmarkt 1981; 90: 1%8-71, 2183-5. ISO Standards: ISO 53: Bezugsprofil fur Stirnriider fur den allgemeinen Maschinenbau und den Schwertnaschinenbau. - ISO 677: Bezugsprofil fur geradverzahnte Kegelriider fur den allgemeinen Maschinenbau und den SchwerISO 701: Intemationale maschinenbau. Verzahnungstertn- inologie: Symbole fur geometrische Grossen. - ISO/R 1122: Vokabular fur Zahnriider; Geometrische Begriffe. - ISO/R 1122, Add. 2: Vokabular fur Zahncider; Geometrische Begriffe, Schneckengetriebe. ISO 1328: Stimriider mit Evolventenverzahnung - ISO Genauigkeitssystem. - ISO 1340: Stirnriider; Angaben fur die Bestellung. - ISO 1341: Geradverzahnte Kegelriider,
'M.II
Angaben fur die Bestellung. - ISO 2203: Zeichnungen; Darstellung von Zahnriidem. DIN Standards: DIN Taschenbuch 106. Antriebstechnik 1. Nortnen Ober die Verzahnungsterminologie. Beuth, Berlin and Koln: 1981. - DIN 37: Zeichnungen; Darstellung von Zahnriidem. - DIN 780: Modulreihe fur Zahnriider; Moduln fur Stimriider und Zylinderschneckengetriebe. - DIN 783: Wellenenden fur Zahnradgetriebe mit Wiilzlagem. - DIN 867: Bezugsprofil fur Stimriider (Zylinderrader) mit Evolventenverzahnung fur den allgemeinen Maschinenbau und den Schwertnaschinenbau. DIN 868: Allgemeine Begriffe und Bestimmungsgrossen fur Zahnriider, Zahnradpaare und Zahnradgetriebe. DIN 3960: Ilegriffe und Bestimmungsgrossen fur Stirnriider (Zylinderrader) und Stimradpaare (Zylinderradpaare) mit Evolventenverzahnung. DIN 3961: Toleranzen fur Stimradverzahnungen; Grundlagen. - DIN 3962: Toleranzcn fur Stimradverzahnungen; Abweichungen einzelner BestimmungsZuliissige grossen. - DIN 3963: Toieranzen fur Stimradverzahnungen; Zuliissige Wiilzabweichungen. - DIN 3964: Toleraozen fur Stimradverzahnungen; Gehiiuse-Toleranzen. DIN 3966: Angaben flir Verlahnungen in Zeichnungen; Angaben fur Stimrad-(Zylinderrad-) Evolventenzerzahnungen und Geradzahn-Kegelradverzahnungen. DIN 3967: Getriebe-Passsystem; Flankenspiel, Zahndickenabmasse und Zahndickentoleranzen. - DIN 3970; Lehrzahnriider zum Prtifen von Stimriidem. - DIN 3971: Verzahnungen; Bestimmungsgrossen und Fehler an KegeJriidem. - DIN 3972: Bezugsprofile von Verzahnwerkzeugen fur Evolventenverzahnungen nach DIN 867. DIN 3975: Begriffe und Bestimmungsgrossen fur Zylinderschneckengetriebe mit Achsenwinkel 90"' - DIN 3976: Zylinderschnecken; Abmessungen, Zuordnung von Achsabstaoden und Obersetzungen in Schneckengetrieben. DIN 3978: Schriiglmgswinkel fur Stimradverzahnungen. DIN 3979: Zahnschiiden an Zahnradgetrieben; Bezeichnung, Merkmale, Ursachen. - DIN 3990: Tragfahigkeitsberechnung von Stimriidem. - DIN 3991: Tragfahigkeitsberechnung von Kegelriidem. DIN 3992: Profilverschiebung bei Stimriidem mit Aussenverzahnung. - DIN 3993: Geometrische Auslegung von zylindrischen Innenradpaaren. - DIN 3994: Profilverschiebung bei geradverzahnten Stirnriidem mit 05-Verzahnung, Einfuhrung. - DIN 3995: Geradverzahnte Aussen-Stirnriider mit 05-Verzahnung. - DIN 3998: Benennungen an Zahnriidem und Zahnradpaaren. - DIN 3999: Kurzzeichen fur Verzahnlmgen. - DIN 58400: Bezugsprofll fur Stimriider mit Evolventenverzahnung fur die Feinwerktechnik. DIN 58405: Stimradgetriebe der Feinwerktechnik. DIN 58420: Lehrzahnriider zum Priifen von Stimriidem der Feinwerktechnik. - DIN 58425: Kreisbogenverzahnungen fur die Feinwerktechnik. - DIN 45635 T 23: Geriiuschmessung an Maschinengetrieben. VDI Guidelines: VDI-Richtlinie 2060: Beurteilungsmassstiibe fur den Auswuchtzustand rotierender starrer Korper. VDI-Richtlinie 2159: Getriebegeriiusche; Messverfahren - Beurteilung - Messen und Auswerten, Zahlenbeispiele. - VDI-Richtlinie 2546: Zahnriider aus thertnoplastischen Kunststoften. - VDI-Richtlinie 3720: Liirtnartn konstruieren. F9 Mechanism Engineering, Kinematics. [1 J VOI-Richtlinie 2127 (Entwurf): Getriebetechnische Grundlagen: Begriffsbestimmungen der Getriebe (1988). - [2 J Braune R. Die Bedeutung des kOrzestcn und des liingsten Gliedes fur die systematische Betrachtung ebener viergliedriger kinematischer Ketten. Ind-Anz 1971; 93: 2258-60. - [3 J
'M.g
Mechanical Machine Components. 12 References
VDI-Richtlinie 2145: Ebene viergliedrige Getriebe mit Dreh- und Schubgelenken: Begriffserklarungen und Systematik (1980). - [4] VDI-Richtlinie 2147: Ebene Kurvengetriebe: Begriffserklarungen (1962). - [5] Volmer J. (ed.). Getriebetechnik; Leitfaden, 3rd edn. Vieweg, Brunswick, 1989. - [6] Hain K. Ermittlung der Umlauf- und Schwingbewegungen in durchlauffiihigen sechsgliedrigen Getrieben. Grundl Landtechn 1966; 16: 129-39. - [7] Hain K. Systematik und Umlauffiihigkeit drei- und mehrgliedriger Kurvengetriebe. Konstruktion 1967; 19: 37988. - [8] Hain K. Rechenprogramme fur beschleunigungsgleiche Getriebe mit unterschiedlichen Hauptbewegungen. Werkstatt Betrieb 1976; 109: 73-80. - [9] Shigley]. E., Uicker ]. J. jr. Theory of machines and mechanisms. McGraw-Hill, Tokyo, 1980. - [10] Lohe R. Beeinflussung der Laufeigenschaften durch Massenverteilung bei ebenen und riiumlichen Getrieben. VDI-Ber 1980; 374: 135-45. - [11] Kerle H et aI. Berechnung und Optimierung schnellaufender Gelenk- und Kurvengetriebe. Expert Verlag, Grafenau, 1981. - [12] Braune R. Entwurf Richtlinie VDI 2729: Modulare kinematische Analyse ebener Gelenkgetriebe mit Dreh- und Schubgelenken (1989). - [13] Liitgert A, Braune R. KAMOS - ein interaktives Entwicklungswerkzeug zur Analyse, komplexer Koppelgetriebe. VDI-Ber 1989; 736: 119-50. - [14] VDI-Richtlinie 2138: Raumliche Kurbelgetriebe: Umformung von Drehbewegung in Schwingschubbewegung (1959). [15] VDI-Richtlinie 2139: Raumliche Kurbelgetriebe: Umformung von Drehbewegung in umlaufende Drehschubbewegung (1959). - [16] VDI-Richtlinie 2723: Vektorielle Methode zur Berechnung der Kinematik riiumlicher Getriebe (1982). - [17] VDI-Richtlinie 2724: Berechnung der Kinematik viergliedriger Getriebe: Ein Rechenprogramm (1986). - [18] Lohe R. Berechnung und Ausgleich von Kriiften in riiumlichen Mechanismen. Fortschr-BerVDI-Z, series 1, no. 103 (1983). - [19] Ahlers W. Zur Bestimmung der Lagefiihrungsgriissen von Manipulatoren am Beispiel einer Operationsleuchte. Konstruktion
1986; 38: 81-6. - [20] Matthaei H. Ober den Leistungsfluss in Kurbelgetrieben. Konstruktion 1966; 18: 45-9. [21] Marx U. Ein Beitrag zur kinetischen Analyse ebener viergliedriger Gelenkgetriebe unter dem Aspekt Bewegungsgiite. Fortschr-Ber VDI-Z series 1, no. 144 (1986). - [22] Miiller H W. Beurteilung periodischer Getriebe mit Hilfe des "Ubertragungswinkels". Konstruktion 1985; 37: 431-6. - [23] Stiindel D. Das dynamische Laufkriterium bei Geriitemechanismen. Feingeriitetech 1974; 23: 507-9. - [24] Kerle H. Dynamische Maschinenanalyse mit Hilfe programmierbarer Tischrechner. Forsch Ing-Wes 1980; 46: 149-53. - [2S] Dittrich G. Systematik der Bewegungsaufgaben und drundsatzliche Liisungsmiiglichkeiten. VDI-Ber 1985; 576: 1-20. - [26] Alt H. Das Konstruieren von Gelenkvierecken unter Benutzung einer Kurventafel. VDI-Z 1941; 85: 69-72. - [27] VDI-Richtlinie 2130: Getriebe fur Hub- und Schwingbewegungen: Konstruktion und Berechnung viergliedriger ebener Gelenkgetriebe fur gegebene Totlagen (1984). - [28] VDI-Ricbtlinie 2125: Ebene Gelenkgetriebe: Obertragungsgiinstigste Umwandlung einer Schubschwing- in eine Drehschwingbewegung (1987). - [29] VDI-Richtlinie 2126 (Entwurf): Ebene Gelenkgetriebe: UbertragungsgOnstigste Umwandlung einer Drehschwing- in eine Schubbewegung (1986). - [30] Kracke J. Massbestimmung ebener viergliedriger Kurbelgetriebe fur die Sonderfille von vier Ubereinstimmungen. Diss., TU Brunswick, 1972. - [31] Braune R. Ein Beitrag zur Masssynthese ebener viergliedriger Kurbelgetriebe. Diss., RWTH Aachen, 1980. - [32] Hain K. Konstruktionsdaten-Auswahl fur das Gelenkviereck durch
Computer Dialog. technica 1976; 25: 791-8. - [33] Hain K. PunktJagenreduktion als getriebesynthetisches Hilfsmit-
tel. Maschinenbau/Betrieb, Beil Getriebetechn 1943; 11: 29-31. - [34] VDI-Richtlinie 2143, paper 1: Bewegungsgesetze fur Kurvengetriebe: Theoretische Grundlagen (1980). - [35] VDI-Richtlinie 2143, paper 2: Bewegungsgesetze fiir Kurvengetriebe: Praktische Anwendung (1987). - [36] Dudit7.a F. Kardangelenkgetriebe und ihre Anwendungen. VDI-Veriag, Diisseldorf, 1973. - [37] Hiller M. Analytisch-nurnerische Verfahren zur Behandlung riiumlicher Ubertragungsmechanismen. Fortschr-Ber VDIZ series I, no. 76 (1981). - [38] VDI-Richtlinie 2156: Eiofache riiumliche Kurbelgetriebe: Systematik und Begriffsbestinunungen (1975). - [39] VDI-Richtlinie 2154: Spharische viergliedrige Kurbelgetriebe: BegrifIserklarungen und Systematik (1971). - [40] VDI-Richtlinie 2722: Homokinematische Kreuzgelenkgetriebe einschliesslich Gelenkwellen (1982). - [41] Hain K. Entwurf iibertragungsgiinstigster Zylinderkurven- und Kegelkurvengetriebe. Werkstatt Betrieb 1978; 111: 93-8. - [42] Zakel H. Geometrie, Kinematik und Kinetostatik des Kurvengelenks riiumlicher Kurvengetriebe. Diss., RWTH Aachen, 1983. - [43] Uchtwitz O. Getriebe fur aussetzende Bewegungen. Springer, Berlin, 1953. - [44] Eckerle R. Optimale Auslegung von Malteser-Schaltwerken. Feinwerktechn 1969; 73: 482-7. - [45] Hain K. Erzeugung von Schrittbewegungen durch Planeten-KurvenGetriebe. Antriebstech 1973; 12: 315-22. - [46] VDIRichtlinie 2721: Schrittgetriebe: Begriffsbestimmungen, Systematik, Bauarten (1980). - [47] Hain K. Die Erzeugung gegebener Kurven mit Hilfe von Raderkurbelgetrieben. Feinwerktech 1949; 53: 81-9. - [48] Volmer]. Raderkurbelgetriebe. VDI-Forsch 1957; 461: 52-5. - [49] Neumann R., Watzlawik P. Synthese von Raderkoppelschrittgetrieben mit Hilfe von Kurventafeln, Maschinenbautech 1974; 23: 52-9. FlO Crank Mechanism. [I] Bensinger W 0, Meier A.
Kolben, Pleuel und Kurbelwelle bei schnellaufenden Verbrennungsmotoren, 2nd edn. Springer, Beriin, 1961. - [2] Biezeno C. B., Grammel R. Technische Dynamik, vol. 2. Springer, Berlin, 1971. - [3] Haffner K E, Mass H. Torsionsschwingungen in der Verbrennungskraftmaschine. Springer, Vienna, 1985. - [4] Kiittner K H. Kolbenmaschinen, 5th edn. Teubner, Stuttgart, 1984. - [5] Kiihler G, Riignitz H. Maschinenteile, vol. 2, 7th edn. Teubner, Stuttgart, 1986. - [6] Lang 0 R. Triebwerke schnellaufender Verbrennungsmotoren. Springer, Berlin, 1966. [7] Mayr F. Ortsfeste Dieselmotoren und Schiffsdieselmotoren, 3rd edn. Springer, Vienna, 1960. - [8] Sass F. Bau und Betrieb von Dieselmaschinen, vol. 1, 2nd edn. Springer, Berlin, 1948. - [9] Maas H, Klier H. Kriifte Momente und deren Ausgleich in der Verbrennungskraftmaschine. Springer, Vienna, 1981. and Guidelines: Kolbenbolzen fiir fiir Dieselmotoren, Kraftfahrzeugbau DIN 73 124 DIN 73 125 fur Ottomotoren. Kolbenringe (the fIrst numbers given apply to mechanical engineering, the secondnamed numbers apply to automotive engineering) DIN 34 109 and DIN 70909 Obersicht, Aligemeines; DIN 34 110 and DIN ISO 6620 Rechteckringe; DIN 34 III and DIN ISO 6620 Minutenringe; DIN 6662 Trapezringe; DIN 34130 and DIN ISO 6624 Nasenringe; DIN 34146 and DIN ISO 6625 Olschlitzringe; DIN 34147 and DIN ISO 6625 Dachfasenringe; DIN 34 148 and DIN 70948 Gleichfasenringe.
Standards
Note on Standards
Note on Standards FI.l.2, page F8: see also the following related (R) or equivalent (E) publications. For DINs 1913,8554,8555,8556: ISOs 544 (R), 636 (R), 2560 (R), 3580 (R), 3581 (R); for DIN 1913: ISOIDIS 2560 (R), prEN 20544 (E); for DINs 8554, 8555, 8556: prEN 20544 (R); for DIN 8556: ISO/DIS 11837 (R); for DIN 8559: ISOs 636 (R), 864 (R), prEN 440 (E). FI.l.2, page F9: see also the following related publications. For DINs 1732, 1733, 1736: ISOs 544, 636, 2560, 3580, 3581; for DIN 1732: also prEN 20544; for DIN 8573 TI: ISO 1071. Fl.l.4, page FIl: see also the following related (R) or eqUivalent (E) publications pertaining to DIN 1912. TI: ISO 2553 (R); T2: ISO 6947 (R); T4: no equivalent: T5: ISO 2553 DAD I (R), ISO 2553 (R). Publications pertaining to DIN 8563. TI: ISO 6213 (E); T2: ISO 6213 (R), lSI/DIS 3834 (R); n: ISO 5817, DIN EN 25817; no: ISO 10042 (R), prEN 630 (E); Tl04: prEN 719 (E); Tl05: prEN 2885 (E), ISOIDIS 9956-5 (E); Tl06: prEN 288-6 (E), ISOIDIS 9956-6 (E); Tl07: prEN 288-7 (E), ISO/DIS 9956-7 (El; Tl08: prEN 288-8 (El, ISO/DIS 9956-8 (E); TI 10: prEN 729-1 (E), ISO/DIS 3834-1 (E); Till: prEN 729-2 (El,
'M.@.
ISOIDIS 3834-2 (E); Tl12: prEN 729-3 (E), ISOIDIS 38343 (E); TI 13: prEN 729-4 (E), ISO/DIS 3834-4 (E); Tl20: prEN 1011 (E). F/.J.5, page F12: see also the following related publications. For DIN 1745: ISO 6361-2, ISO/DIS 6361-2 DAD I; for DIN 1746 Part I and DIN 1747: ISO 6362-2 and ISO/DIS 6362-2; for DIN 1746 Part 2 and DIN 1748 Part I: ISOs 6362-1 and 6363-1; for DIN 1748 Part 4: ISO 6362-4. FI.l.5,page P/3: see also the equivalent DIN V ENV 1993 (Eurocode 3). DIN 15 o18-related publications are: Part 1: ISO 4301-4, ISO 8686-1, ISO/DIS 8686-1, ISO 4301-5, ISO 8686-5, ISO/DIS 8686-3; Part 2: no equivalent; Part 3: ISO 8686-1, ISOIDIS 8686-1, ISO 8686-5, ISO/DIS 8686-3. P7, page 200: see the following related British Standards: BS 3092: 1973 (1988): Specification for main friction clutches, maio power take-off assemblies and associated attachments for internal combustion engines. BS 3170: 1972 (1991): Specification for flexible couplings for power transmission. BS 6613: 1985 (1981): Methods for specifying characteristics of resilient shaft couplings. BS AU 203a: 1988: Specification for dimensions of couplings between power take-offs and ancillary driven units on commercial road vehicles.
Hydraulic and Pneumatic Power Transmission R. ROper, Dortmund
_ _ Fundamentals of Fluid Power Transmission Systems_ 1.1 The Flow Process The specific energy of a moving fluid (liquid or gas) is described by the Bernoulli equation: Y, = Elm = h
f at
au. ds. + -u + gz + 2
2
The continuity equation applies for the steady·flow con· dition:
m = const. = pAu. In the special case of incompressible fluids,
it = Au. Fluids
can transfer signals by means of an energy state (e.g. a given pressure), or flow strength and energy by transfer of specific energy as mass flow. The conversion from mechanical to fluid energy takes place in a steady flow process, as in Fig. 1 (throughput direction is positive):
Y m = Pmlm = Y" - Y n = h2 - hI
+
(u~ - uD/2
+ g(Z2 - ZI)'
With h2 - hI = Ab 12 = (Ab')12 + P v •12 lm (index s = isentropic), the irreversible component of the specific work is
giving the irreversible loss as Pv •12 lm. For energy transfer in a fluid drive, the specific enthalpy h of the remaining components becomes negligible. The specific work is then simplified as or
Ym = (dh,)'2 '1;'1 (+ 1 motor, -1 generator). i.e. the fluid work is the differential of work done by out· put and input shear forces. The power transfer then fol· lows for constant flow strength m or V
P = dEldt = V·
dP'2 (uniform flow).
Mean working velocities up to 5 mls are found.
Pressure range
Application
Low pressure
30 to 50 bar
Machine tools (feed mechanisms)
Medium pressure
Up to 170 bar
Transport equipment, building machinery. traversing gear
High pressure
200 to 450 bar
Presses, tensioners, aircraft hydraulics
Under unsteady conditions, the compressibility of the working fluid must be taken into account. Mean values for the compressibility of hydraulic oils are:
J3
= - dVIVdp = (7 to 4.5) . 10- 5 bar-I
at 250 to 20 bar oil pressure and 80 to 20°C oil tempera· ture. For unsteady conditions, calculations are more complicated, since masses, resistances and spring characteristics are distributed, not concentrated. Differential equations are non-linear. Linearisation gives approximations to within ± 20",0 of the critical frequency, or accurate results with matrix methods [1]. By change of state, it can also be expressed as
f dp = (f 2
(dh,),2 = (du 2 )12/2 + g . .:1%12,
Y m = Pmlm = h2 - hI = (Ab')12 + Pv . 12 lm,
Designation
.:lb., =
2
v
I
v
dp), +
P v •12lm
I
(see G:I Fig. :I).
Nomenclature: A flow cross-sectional area; E fluid energy; g acceleration due to gravity; h specific enthalpy; m mass; m mass flow; p pressure; Pm mechanical power; Pv power loss; s length of flow path; u flow velocity; v specific volume; V volumetric flow; Y, specific hydraulic energy (work); Ym specific mechanical work; Z geodetic altitude; p density of fluid; '1. overall efficiency. By using high energy density (pressure), large forces or moments in linear or rotary motion may be obtained with small dimensions. This gives low specific weight « 1 kglkW for the complete apparatus) and moment of inertia (rapid reversing). The lower volumetric flow resulting from the use of higher pressures allows the use of smaller dimensions in control units, and produces low transmission losses (loss - V or V"). Losses in the flow process also occur as pressure losses. The energy loss causes the temperature of the fluid to rise.
1.1.1 Hydraulic Power Transmission
FJaure 1.
Schematic of open-flOW process.
The various oils and other fluids used in hydraulic equipment are virtually incompressible. The equation of state is therefore identical with the isochore
Hydraulic and Pneumatic Power Transmission. 1 Fundamentals of Fluid Power Transmission Systems
f
aqueous solutions of polymers, preferably polyalkylene-
2
(db')!2
=
11
dp
= dp!2Ip.
!
1.1.2 Pneumatic Power Transmission Gases are extremely compressible, and working velocities are therefore uneven due to variations in expansion. Gases are mainly used for power transmission only for subordinate purposes (small machine tools), partly at polytropic conditions pu" = canst., where n = 1.3 to 1.35, and frequently without expansion at full pressure operation. Accurate positioning can only be achieved via mechanical equipment, and the main applications are in tensioning and press tools. Working velocities are very high. Singlestage compression limits the available pressure range, particularly as compression to higher pressures causes a large temperature rise. Designation
Pressure range Application
Low pressure
Up to 1 bar
Control equipment
High pressure
6 to lObar
Presses, teosioners, transport and machine tools
For precise velocity, input to pneumatic equipment on machine and hand tools, parallel or post hydraulic control (pneumo-hydf'dulic), or electronically operated throttle controls may be used.
1.2 Hydraulic Fluids Mineral oils, water-based fluids, or water-free synthetic fluids are used as working fluids in hydraulic systems. The DIN 51 524 or ISO TC 131 HL and HLP fluids are mineral oils defined in DIN 51 519 or ISO 3448 for viSCOSity classes v = 7, 10, 15,22, 32,46 and 68.10- 6 mls at 40°C. Mineral oil HL contains additives to improve resistance to high thermal loadings and to increase cortosion protection. The HPL oils have extra additives to improve properties in regions of mixed friction, while the HLPD oils have other additives to counter the effect of entrained water. Also, HD, SI, and S3 engine oils are sometimes used as hydraulic fluids. For viscosity and temperature characteristics of hydraulic oils see Appendix G, Fig. 1. The variation of viscosity with pressure is negligible at pressures below p = 200 bar. The selection of oil is made according to the operational viscosity needed for the components (manufacturer's recommendation) and the mean operating temperature, with the temperature of the oil being likely to exceed that of the surroundings by 30 to SO K. An additional selection criterion is a good starting capability after an extended period at low temperature. High viscosity can lead to suction difficulties. Nonnal operating viscosities range between 20 and 60· 10-6 m 2 /s. New developments in oils with polymer additives (high-VI oils) show less dependence of viscosity on temperature, allowing cold-start temperatures to be lower by about 10 0c. In situations where oil leaks may be ignited (forges, foundries, coal-mines), non-flammable fluids have to be used. Fluids containing water can prevent sudden ignition by the buildup of a layer of steam. These include the following: HFA - oil-water emulsions (see G6); HFB water-oil emulsions with water contents up to 60%; HFC:
glycol-waler solutions with up to 60% water. When water-
containing fluids are used, loadings must be reduced because of lower wear resistance, especially in the case of rolling friction, and special filters must also be used. Category HFD comprises water-free fluids, non-flammable because of their chemical composition, e.g. only phosphate esters. They exhibit good wear resistance, but before use, toxicity and effect on sealing materials should be tested.
1.3 Systematology 1.~.1
structure and Operation of" Fluid Transmissions
In fluid drives, generators (pumps, compressors), motors, and control units are connected in a circuit in which the working fluid circulates (Fig. 2). Because of the high pressures involved, only positive-displacement machines are suitable. Because of this, and the incompressibility of the hydraulic fluid, there is in hydrostatic drives a volumetric link between the drive and the drive components, i.e. the transmission of hydraulic drive is virtually independent of load (parallel characteristic). In contrast, the compressibility of air under pressure is very much greater. Pneumatic drives exhibit series characteristics. The pressure medium is transferred by pipes, which feature allows considerable latitude in the location of drive, motor and control unit. Distances can be up to 30 m with hydraulic drive, and as much as 150 m with pneumatic drive. The control system regulates the transmission of work and limits system loadings. It can work directly on the fluid flow by switching flow channels, by changing or splitting the direction of flow, or indirectly by changing the geometry of pumps and motors. The function may be either engaged (pressure control, position contral), or
disengaged. Control systems also work by cylinder cutoff. They can operate either directly or indirectly. There are thus good possibilities for remote control and automation by a combination of electrical and electronic control media, 1.~.2
Classification of" Fluid Transmissions
Energy transmission using compressible gas or liquid flow gives an almost unlimited variation of energy flowing through the system to transmit force or torque and linear or angular velocity. A classification can be made according to the following external constraints.
Transmission Power Transmissions transfer input power over the widest possible range to produce required force moments Generator (pump)
Control (valve)
Motor rotary/linear
Pmech ~MM"WM ~FMVM
-
Power flow
I Set point Figu:re 2. Block diagram of fluid drive.
Signal flow
1.3 Systematology • 1.3.4 Symbols
and tbe desired linear or angular velocity at the work location. Because of their high power, high efficiencies are possible.
Position Transmissions transmit accurately input signals and errors to the work location for control and regulation. Accuracy of information is most important; efficiency can be neglected. Operation Depending on function, piping components can have varying significance:
r--------·l
Power Mechanisms transfer power from generator to work location. It is important to have good efficiency over long ranges (e.g. traversing gear).
i
Force Mechanisms deliver large forces or moments to the work location; efficiency is less important (e.g. presses, shears, clamps).
I
c
Feed Mechanisms usually act against only small forces to transmit advance motion with high accuracy in botb position and velocity. Efficiency is mostly unimportant (e.g. machine tool feeds, copying controls). Type of Output Motion In hydraulic drives, similar and different types of machine may be arbitrarily combined. Systems can therefore be classified as follows, according to the more important output motion designed for individual cases:
Rotary. Where the angle of rotation of the output shaft is infmite.
Figure 3. Classification of fluid drives: a remote-controI stationary transmission (/ internal rotor, 2 housing, 3 fixed point (comrol),
4 subassembly, 5 fluid piping), b compact hydrostatic circulating drive, c combined drive with hydrostatic main and mechanical auxiliary drive, d remote-controUed c.lrive assembly.
Reciprocating. Where tbe angle of rotation of the output shaft is limited to a given value.
Linear. Where output motion is in a straight line. Function Self-Controlled Systems. These amplify or distribute forces, or transmit force to remote locations. Input force is
the muscular force of the operator (e.g. hydraulic brakes).
Remote-Controlled Systems. These are tbe more usual type of hydraulic and pneumatic drive. Mechanical energy is input from some external source, transmitted as fluid pressure, and released at tbe work location. Only control actions such as switching involve operator commands. Servo Systems. Here input force (simple mechanical force, or forces from measurement) is amplified analogically by the adtlition of input energy (e.g. turbine governors, power steering and braking systems). 1.~.~
Transmission Layouts
The internal construction of a hydraulic drive is determined by application, operating conditions, and consequent requirements of pumps and motors. Positive-displacement machines consist of a main stator and rotor, the function of which then determines the necessity for the other components (internal! external rotor). With interconnected drives, pumps and motors may be separated, whereas in compact drives they will be located within the same casing (only hydrostatic drives). Remote transmissions are, as seen in Fig. ~, possible
only as fixed transnusslOns, i.e. one component of the machine is fixed relative to the foundation (frame). Epicyclic transmissions occur by mechanical connection of machine components in compact transmissions. Coupled transmissions are the combination of hydrostatic
transmissions with mechanical (geared) transmissions, and are usually compact. Their construction results from a three-shaft (epicyclic) transmission and a twin-shaft parallel transmission, with which two of the shafts of the epicyclic transmission are connected. Depending on requirements, the main or parallel transmission may be hydrostatic (see VOl Guideline 2151). Because of the connection between the machines, power can be transmitted between tbe required components only via fluid flow: collected transmissions are supplied by several pumps connected in parallel; in distributed transmissions power is transmitted from several motors in series or parallel (differential). 1.~.4
Symbols
In drawings of hydraulic and pneumatic transmissions and control systems, components are represented by graphical symbols. These ignore actual shape and represent only function. The representation in the circuit is at tbe rest contlition of the mechanism, or, if this is not possible, in tbe home position of the control system. Shape, meaning and use of symbols are given in ISO 1219. Further symbols, especially for piping (DIN 2429, ISO 4067-1), can also be used. For the most important symbols in use, see Appendix S, Fig. 2.
BII
Hydraulic and Pneumatic Power Transmission. 2 Components of Hydrostatic Transmissions
_
Components of Hydrostatic Transmissions_
2.1 Pumps
service life) is determined by pump type and consequent component loading, etc . The second important criterion is the form of chamber, i.e. the magnitude of the lift volume relative to the size of the machine. With the generally adopted rectangular cell section of rotary pumps, tolerances are more difficult to determine. The pressure-dependent internal leakage losses limit the range of application of low and medium pressure machines. Since cylindrical fits are simpler to manufacture, linear piston machines are
2.1.1 Synopsis Hydraulic pumps are either rotary pumps (rotary piston) or lift pumps (linear piston) with fixed or adjustable displacement volume (Fig. 1 and Table 1). In practice, application determines the type of pump used. Permissible operating pressure (in economic use and over full
~.....
~c
Rotary machinery
i~.j~ r------------.------------------------1~ Description Schematic representation j"*~
Linear machinery Schematic representation
Description
7 linear piston pumps
Gear pumps With extemat teeth
2
8 Radial piston pumps With internal piston operation
With internal teeth
9
3 Screw pumps With external piston operation
10 Vane pumps
Axial piston pumps
Single-acting
Swashplate pumps
5
• .
Multi-acting
7.
.
11 ' Port-plate pumps :
'.
" z
/,
12
6 Rotary piston pumps Tilting-head pumps
Figure 1. Overview of commercial hydraulic pumps.
•
2.1 Pumps. 2.1.2 Characteristics and Power Rating
Table 1. Common operating values for hydrdulic pumps (system numbers as Fig. 1) Displacement volume
No.
cm·~/rev
0.4 to 1200
1.2
2 :\0
800 800 500 8 1000 800 0.4 to t5000 \.5 to 3600
5
6 7
8,9 10, II, 12
to
to to to
Pressure range bar
Speed rev/min
Preferred oil viscosity 10-- 6 ml/s
lip to 200 Internal gear pump to 550 lip to 200 lip to 100 Up to 160 (200) lip to 160 lip to 400 lip to 630 lip to 400
1500 to :\000 (lip to :\500) 1000 to 5000 500 to 1500 500 to 3000 500 to 1500 1000 to 2000 1000 to 2000 500 to 3000
40 to 80
more suitable for the higher pressure ranges. In contrast, application has little effect on rotational speed. The general rule is that permissible speed is determined by physical size.
Rotary Pumps These deliver the working fluid by uniform rotation in a chamber, the volume of which is cyclically altered through the form of the confining walls or the movement force of a vane. The processes of suction and delivery occur concurrently rather than consecutively as would be the case in a piston pump. Adjustable displacement volume is only possible with single-acting vane pumps.
Lift Pumps
80 30 30 30 20 20 30
to 200 to 50
to 50 to 50 to 50 to 50 to 50
= Mw, because of friction in the mechanism and between compressor elements is reduced by the friction power loss Pv,r = Mrw to
Pm
Displacement Power P u = (M - M,)w. This is transmitted by the displaced volumetric flow and may be divided into the displacement power P,h for tlp and the hydraulic power loss P v .h
= Vth.:lPh = Mhw,
This brings together the flow losses and the (very small) compression work. Thus we have Pm = P[h + Pv,r + Pv,h
or M = Mth + Mr + Mh ·
Both types of loss arise in the mechanism and, since they are dimensionally similar, are jointly expressed in the
Hydro-mechanical Efficiency
These are characterised by the separation of the drive shaft from the pumping cylinder, whose volume is cyclically altered by the action of a linear piston. The delivery volume can be adjusted by the driving gear geometry, or in the control system. Because of the internal flow reversal of the fluid, valves are necessary between the cylinder and the intake and delivery pipes.
The pressure drop tlp causes a leakage flow Vv through the gap, so that the displacement volumetric flow is reduced to the
2.1.2 Characteristics and Power Rating
Actual Supply Flow
The displacement volume, which is equal to the lift volume, Vb is determined by the geometric properties of the
and thus affects the power loss Pv.v = Vvtlp = P,h - Ph'
machine, and is usually expressed in cm~ frey. Assuming complete filling of the lift volume at suction, we have:
Theoretical Supply V'h = n V, =
= rotational speed; w =
W Vo
= basic volume = V;/21T). For pressure drop between inlet (I) and outlet (0) tlp = P, - Po, supply gives a theoretical pump
(n
27m; V;)
moment
M,h = tlpV;/21T
=
1/hm
= 1/11 +
(Pv.,
V=
+
pv.h)/Pm
+ Pv.h)/P,h]·
V'h -
Vv
T/v = Phi Pth = 1 - Pv.vl P[h = 1 - Vvl V[h'
The balance of the transfer of mechanical drive power in the hydraulic pump power Ph = V tlp is brought together in
Total Efficiency 1/,
=
Ph/Pm
= (1
=
I - rpv/Pm
- Pv.,h/Pm) [1 - pv.vI(Pm - PV.'h)]
= 1/hm1/v·
Magnitudes depend on operating conditions, and they are usually expressed as characteristic curves as in Fig. 3.
1.0 0.8
~06 E • ~
..: 0.4
~
0.2
-~ V ""-- 'r-Ti,'" 'fJ.~ ~
7' f
o o
'7t
+-
n=cons!
~---+-
0.2
0.4
I
i
0.6
0.8
!!.pl!!.Pmox
a
Figure 2. Power flow diagram for a hydraulic pump (see Section G1.1 for an explanation of the nomenclature),
- (Pv.,
Volumetric Efficiency. This is given by the equation
tlpVo·
The actual relationships between loss-related power transmission are shown in Fig. 2. Mechanical drive power
= P'h/Pm = I
0.4
0.6
nln max b
0.8
1.0
Figure 30. Typical curves of efficiency fot" constant-flow pump, in relation to a openning pressure, b speed.
Hydraulic and Pneumatic Power Transmission _ 2 Components of Hydrostatic Transmissions
ure compensation using pressure fields made up of movable bearing elements. Efficiency in continuous use is 7), = 0.85 to 0.75 (normal performance) and up to 7), > 0.9 with pressure charging.
Figure 4. Simple gear pump with plate construction (Robert Bosch, Stuttgart).
2.1.~
Gear·Type Pumps
Gear·type pumps (rotary displacement machines) consist of at least two intermeshing rotors which cause displacement via tooth meshing. Two subsidiary types are gear pumps and screw pumps.
Internal Gear Pumps. These have better mesh ratios and thus an irregularity in delivery flow of only {j = 3 to 5%. Extending the inlet and outlet zones over a greater range of angles ensures good fIlling and delivery ratios and low noise . Ring pumps consist of two wheels, where the number of teeth on the internal gear wheel is 1 greater than on the pinion. Operating pressure is below 100 bar. High-pressure pumps have a Sickle-shaped seal between the rotors with operating pressure < 350 bar.
Screw Pumps. These have pulsation-free and low-noise characteristics (normal construction: see Fig. 1), and run at higher speeds and pressures. They are used in lift machinery and fme-work machine tools. Problems are manufacturing cost, relatively lower volumetric efficiency « 0.8%) , and delivery of high-viscosity fluids.
Gear Pumps
2.1.4 Vane·Type Pumps
These may be further subdivided into external gear pumps, in which at least two externally toothed wheels mesh, and internal gear pumps in which one internally toothed gear wheel meshes with at least one externally toothed gear wheel (Fig. 1). There is a further division into simple pumps with one pair of wheels, and multiple units in series (one central drive gear, several secondary wheels) or parallel (several sets of wheels on the same shaft). Multiple units are used to supply several separate circuits (multiflow pumps) , or as cut-off pumps. The displacement volume is
With vane pumps (typical construction: see Fig. 1), the compression stroke is divided by a number of vanes that can retract into the rotor (vane pump) or the stator (rotary piston pump). The compression stroke is formed by the relative motion between rotor and stator. Their advantages over gear pumps are: small pulsation, low noise, low specific weight (0.4 to 0.6 kg/kW), and higher operating speed.
V;
=
-rrb
4
- (I
[d~,
+ d;, . Z,/Z2
-
d;",
(1
+ Z2/Z ,)
+ z, /zz)-rr 2 m 2 cos 2 O'p/3l
(d. external diameter, d w pitch circle diameter, z number of teeth, m modulus, b width, flank angle of the basic
"p
tooth profile) . The supply flow will have a pulsation rate dependent on the number of teeth. The irregularity {j = (Vm= - Vmin)/V depends mainly on the number of teeth. Flow pulsation causes pressure changes in the pressure strokes and is the main source of noise. External Gear Pumps. These usually have involute teeth, with a contact ratio greater than unity. The volume pressurised by the meshing teeth is discharged via grooves in the compression space. There are normally two equalsized wheels. The irregularity {j = 25 to 10% for numbers of teeth z between 9 and 20. Figure 4 illustrates plate construction. Usually simple journal bearings are used, but roller bearings may be used in larger pumps. Operating pressure (normally less than 100 bar) can be increased to more than 200 bar by press-
Vane Pumps
These operate on the prinCiple of a rotor eccentrically mounted within a casing and provided with radial slots in which vanes slide (Fig. ;). These vanes are compressed against the outer wall by centrifugal force, spring pressure, or the working pressure, to form cells that expand
and contract with the movement of the rotor. It follows that the compression volume is
V; = 4-rrrmeb. Adjustable pumps are constructed so that the eccentricity e can be altered while in operation, and the supply for a given pump speed and direction varied or even reversed. Multi-vane, single-stage pumps work up to a maximum operating pressure of 160 (250) bar. By arranging two opposing inlet and outlet chambers, radial force on the rotor can be cancelled out. Multiple-vane pumps reach efficiencies of 0.85 to 0.9 or more, while single-vane pumps are around 0.6 to 0.85.
Rotary Piston Pumps These work on the prinCiple of stationary vanes and a rotating cam. The rotor has two curved, polished cams set at 90°, which rotate in two opposed chamber rings,
Figure S. Vane pumps: a flow diagram , b example of construction.
2.1 Pumps • 2.1. 5 Piston Pumps
separated by a partition. Operating pressure is 175 bar, and efficiency> 0.9.
2.1.5 Piston Pumps Piston pumps (positive-displacement machines) have advantages over rotary machines in their lower leakage losses owing to good cylinder sealing, and the possibilities for use as variable-displacement pumps with high operating pressures by altering the drive geometry. Because of the small displacement, the pistons are usually driven by cams (eccentrics). The drive can be applied to either the central shaft or the cylinder block. Control is by slide or conventional valves. Slide valves are commonly used in rotary block designs; otherwise axial and cylindrical designs have rotary slide valves or eccentrically driven longitudinal slide valves. Disadvantages are the leakage losses and the higher noise levels due to compressive shock with the necessary connection of the cylinders with the pressure channels. If on the other hand the self-controlled valve opens when the pressures are equal, the pumps nm more quietly, especially at higher pressures and have a higher volumetric efficiency because of the positive valve seating. However, it is not possible to reverse the flow, and at higher speeds valve overlap leads to poorer volumetric efficiency. The flow from a singlecylinder pump is basically sinusoidal. The equal division of supply from a multi-cylinder pump
iT'h = n V;
(n
= rotational speed)
is overlaid by a flow pulsation, the irregularity of which (il) depends on the number of cylinders: i 8%
14
32.5
6 14
25
8 7.8
9 1.5
The displaced volume is V; = iAH (i number of pistons, A piston area, H piston lift). Pumps are therefore designed with odd numbers of cylinders. The most important types are radial and axial pumps. Series piston pumps are not widespread; they are however found as variable-displacement pumps with swashplate control (Bosch Presspump, ct. diesel injection equipment) in test machinery.
Radial Piston Pumps In this type the pistons move along radii of a circle, the centre of which is a rotating shaft. Pumps with internal
eccentrics, driven by an eccentric shaft, and with a fIXed piston block, have valve or slide valve control. This is the preferred structure for constant-displacement pumps, but some very large pumps are still controlled by adjustment of the eccentric drive (Exzentra, Stuttgart). The use of twin cylinder blocks side-by-side driven by 180 0 opposed eccentrics facilitates drive-shaft balance. Drives with external eccentrics are used exclusively with rotating cylinder blocks and internal control (port control).
Constant-Displat:ement Radial Piston Pumps With these the eccentric is enclosed in a roller bearing, the outer ring of which acts on the pistons by friction during the pressure stroke. The kinematically limited sliding occurs only during the suction stroke. Depending on the form of the load ring, there can be up to eight equalsized cylinder units in one plane, connected together via a peripheral channel or in groups (multi-flow pump) leading to the output. Control is by spring-loaded needle valves on the suction side, and ball valves on the pressure side. As the casing also defines the suction area, a smaller supply is necessary. Operating pressures of up to 600 bar are employed.
Variable-Displat:ement Pumps. Figure 6 shows the main type of construction, in which the cylinder block rotates about a fixed central shaft. This shaft has twin charmels bored through it for the oil supply, and forms a control slide valve in the working plane. The pistons are transversely loaded by a crosshead drive, and are held by gudgeon pins and slides against the outer race, which is driven by the roller bearing and rotates with it. The external eccentric can move relative to the casing pins, to give stepless adjustment of the eccentricity from + e to - e, and with it the magnitude and direction of the supply for any given rotational speed. The rotational speed is limited because of the considerable inertial forces, which are additional to the pressure forces acting on the race. Working pressures range up to 450 bar, with a typical mid-range efficiency of 90%. Axial Piston Pumps The pistons of the axial piston pump are arranged in a
circle parallel to the axis of the cylinder drum and are driven by the motion of an inclined plate (adjustable in the case of variable-displacement machinery) relative to the cylinder (see Fig. 1). The following layouts are possible:
Figure 6. Adjustable radial piston pump with external piston operation and internal control (Wepuko Hydraulik, Metzingen).
Hydraulic and Pneumatic Power Transmission. 2 Components of Hydrostatic Transmissions
Swashplate Pumps. Driveshaft and cylinders are coaxial; the cylinders are fixed and the shaft transmits motion via a swashplate (wobbleplate). These pumps are usually valve-controlled, in which case the flow direction is irreversible, but the shaft can turn in either direction. Mainly used as constant-displacement pump. Port-Plate Pumps. The driveshaft and the cylinder are coaxial, with the cylinder driven by the driveshaft. The inclined port-plate is fixed. Valve control is by the cylinder; flow direction is reversible without reversing direction of rotation. Tilting-Head Pumps. The driveshaft is tilted relative to the cylinder, with the cylinder and bearing plate driven. Valve control is by cylinder, with flow direction reversible (see Fig. 7 a, b). Operating pressures vary from 180 to 220 bar, with peak pressure over 400 bar, and drive speed up to 3500 rev/min, depending on size. Typical performance and efficiency curves are shown in Fig. 7c. The displacement volume is given by
V;
=
rotation) or linear (cylinders). Rotary motors follow all the constructional principles of rotary compressors and valve-controlled piston pumps detailed in Section G2.1. They usually have constant-displacement volume, and are only used as variable-displacement machines in exceptional cases. The power balance for a hydrauliC motor (G2.1.2) is shown thus: The hydraulic power PH = V dp is reduced by the leakage losses p v.v = Vv flp from the theoretical power P'h = V'h flp = (WVoM,h)/VO• Therefore: Volumetric effiCiency: 1)v = P,h/Ph = 1 - (Vv/h. The hydraulic power loss p v.h = V,h dPh = wMh, and the mechanical power loss pv., = M,w can be combined as Pv,rh = P vr + Pv,h' Mechanical motor power PH = P'h - Pv.,h = P,h1)hm =Mw. Overall effiCiency 1), = Pm/h = 1)v 1)hm· For the distribution of losses and actuating variables, see the notes in Section G2.1.2.
Gear Motors
lAB = iTr(cP /4)D sin a
(d piston diameter, D diameter of piston assembly, i number of pistons, a tilt angle).
2.2 Hydraulic Motors Depending on output motion, hydrauliC motors can be classified as rotary reciprocating (limited angle of
These have poor starting characteristics under load and their range of application is limited to higher speeds. For slow-running applications, gear motors with flanged-gearwheel reduction gears (iG = 6 to 18) are acceptable. Internally toothed gears without separators are better. The inner rotor has one tooth fewer than those on the outer ring. When both rotate, control is by fIxed, sickie-shaped grooves (Gerotor). With a fIxed outer ring, the inner wheel performs additional peripheral motion, controlled by rotary slide valves (Orbit).
Vane Motors Vane-type motors can be used as high-speed units, possible with reduction gearing, or as slow-running units with multiple admission.
Piston Motors All types of slide-valve controlled axial and radial piston pumps are equally good as hydraulic motors. They can effectively be classifIed in speed ranges:
_. _ Reciprocating angle a =const
o - - Efficiency .. =const
\11 - (T
/\
.
\
0
~o \
N°
t "- l\ast""'\ ~ ~ 80
J "-\ ...... \' ........ '
c
o
20
40
Supply Ilow
Figure 7. Adjustable
91 } ' 91
i""-- r-... f(I
l'/V..... in %
tilting-drum
\
ao
I
High-speed
n = 300 to 3000 (6000) rev/min
i
10 to 750 rev/min
"
Reciprocating Motors These work over limited angles (maximum 720°), and the reciprocating motion is either produced directly (vane motors, with a moving vane in a divided annular cylinder, working angle < 300°), or indirectly from a linear piston via gearing (for rack and pinion drive see Fig. 8).
100
axial-piston
pump
(Mannesmann-Rexroth GmbH, Horb): a zero cutting, b flow con-
ditions at tilt angle ao set at 90° (1 piston, 2 piston rod, 3 cylinder block, 4 shaft with drive flange, 5 control surface, 6 bearing flange, 7 axial cylinder rOller-bearing, 8 cylinder housing, 9 reciprocating
bearing), c curves of flow and efficiency.
n
~
r---. t:~
l-
n = 1 to 150 rev/min
Mediumspeed
For high speeds, axial piston motors are preferable, with reduction gears for lower output speeds. Slow-speed motors are usually radial. For a given drive torque they have a lower moment of inertia and therefore have better dynamic propenies than geared motors. Some irregularity is noted at lower speeds. Pressures and efficiencies are as for pumps.
./ i
90
Slow-speed
Linear Motors These can be either single- or double-acting, depending on cylinder design. Single-acting cylinders, as their name implies, can exen force only in one direction. Piston rod
2.3 Valves. 2.3.1 Directional Control Valves
Figure 8. Reciprocating motor with straigl1t~line working piston motion and rack-and-pinion operation.
and piston are one unit and sealed in the piston rod guides. Stroke is about 2.5 X piston rod diameter, and return force has to be provided by external means, or by a spring. Double-acting cylinders can be pressurised on either side to provide motion in both directions. Piston area on the piston-rod side is smaller by the area of the piston rod As< than the piston area A" so that different forces are produced in either direction, or different speeds are obtained for the same pressure and suction flow. The area ratio is cp = A,I (Ak - As<).
Calculations: Piston force: stroke FIJ = 7),PA k/cp.
= 7)DpA k,
return
F,
Speed:
Vo
= VIA k
V,
= VI(A k -
As,)
= vocp·
For rapid traverse (bottom and piston-rod side strokes equally pressurised, effective area therefore equal to that of the piston rod): l!Jo
= VIA" = v"cp/(cp -
1).
Losses due to pressure-dependent sealing friction F, and inflow pressure losses J.Ph are incorporated into the cylinder efficiency (A = operational area): 7)
=
[(P - L1Ph)A -
F,]/pA.
In
double-acting cylinders, for outward stroke (A = Ak )7)D = 0.9 to 0.95, for return stroke (A = Ak/cp) 7)2 = 0.85 to 0.9, and for rapid traverse 0.2 to 0.4. For stroke terminal velocities greater than 0.1 mis, thrust buffers are necessary. Main cylinder dimensions are defined by piston diameter d, = (12 to 400) mm as well as cp values of 1.25; 1.6; 2; 2.5; and 5.
Structural Guidelines. Cylinders must not be loadbearing, and must experience no bending moments or transverse forces. The load should be supported by the shortest path compatible with function so as to avoid deflection. Examples of correct mounting are shown in Fig. 9.
2.3 Valves These are switching devices in the flow of hydraulic power between pumps and motors with either on-off or adjustable action.
Figure 9. Examples of cylinder layouts.
Classification by Function: directional control valves (directing oil flow); shut-off valves (stopping flow); pressure control valves (regulating pressure); flow control valves (regulating flow). Classification by Principle of Operation: seat valve (sealing element can be a ball, cone, or plate; leakproof sealing) or slide valve (rotary or longitudinal slide; versatile as shutoff valves). Classification by Type As single valves in piping systems, as block valves with equal size and continuous main channels running through them fOrming a single block, as valve blocks with several valves in the same casing, and as plate valves. Plate valves are widely used, as their simple construction and interconnection permit their use in circuit blocks (see Fig. 11). 2.~.1
Directional Control Valves
Into this category come all valves that transfer external pOSitional movements between the connections and thus determine the course and direction of the oil flow. In the majority of cases they have pure switching functions (onoff). It is possible to control the flowrate through throttling (static position function), but because of the associated losses this is only suitable for low-power equipment. Valves can be designated according to the number of switched connections and the number of switch positions (e.g. 4 connections with 3 switch positions gives a "4/3 valve"). Valve connections are designated as follows: P pressure connection; L leakage connection; A, B, work connections; R, T discharge connections; Z, Y, X, control connections.
Seat Valves. These are insensitive to type of fluid used and to contamination, very reliable, and suitable for high pressures. Their disadvantages are high operating forces, and the necessity for individual operation of the closing components so as to ensure proper closing. With directly controlled valves inlet, nominal diameter is limited to < 4 and switching function is simple (2/2 and 3/2 directional control valves). Large sections (often up to nominal diameter 100) are possible with indirect operation. Modular construction (cartridges) is possible with 2/2 directional function (Fig. 10) , allowing extensive circuits within a
Figure 10. Modular 2/2 distribution valves. A, B working connections; X control connection.
'fljl.,
Hydraulic and Pnewnatic Power Transmission. 2 Components of Hydrostatic Transmissions
single , suitably bored-out block. Operation is by an external control valve with X-connections.
Slide Valves. These are very common because the pistons can switch several ways simultaneously and can therefore permit different switching arrangements depending on design. Their construction principle is illustrated in Fig. 11. Piping connections are made by means of drilled or cast channels in the annular slot. For each position, the slotted slide piston makes a connection between different pipe fittings. Static pressures are equalised by equal areas in the slide chamber, flow forces by slot design. Opening characteristics are influenced by indents on the piston shoulder, while alteration of the switching characteristic is possible by chamfering the piston (e.g. continuous two-way connection). Opposed location of the control edges of the slide and casing slots (overlap) affects switching characteristics. With negative overlap, more strokes are briefly interconnected, i.e. there is a danger of unintentional activation of the motor, but there is also better sensitiviry to flow control and building of pressure peaks in switching moving masses. Positive overlap gives better protection against leakage losses. Operating pressures of slide valves range up to 350 bar. Leakage losses at high pressures are noticeable, so that, e.g. running motors are safer with shuttle valves. Some exhibit greater flow resistance (nominal 3 to 8 bar, according to manufacturers' data). Operating Methods for Directional Control Valves Valves are provided with and without preferred switching positions; the so-called impulse valves remain in the switched condition after removal of the control command (memory fimction), otherwise they move back to the rest position under spring pressure, or, in large installations, hydrauliC pressure. Valves are switched manually or mechanically, by hydraulic
Of
pneumatic pressure, or electro-
magnetically. Direct electromagnetic operation is limited to about 3 kW hydraulic power because of the reiatively small magnetic switching force. Larger valves are operated by oil pressure via small flanged pilot valves taking oil either from the working circuit (self-controlled) or from a distant source (remote-controlled). Operating pressures are up to = 4 bar. Magnetic switches can be either dry (operating against oil pressure), or wet (submerged in oil), for either AC or DC. Voltages are usually 24, 48, 180, or 220 V. Maximum switching power is about 100 W.
2.3.2 Shuttle Valves Shuttle valves allow flow in only one direction. Construction is on the seat-valve principle, the simplest form being a spring-loaded ball valve. Because of their leakproof closure, shuttle valves are often used as stop valves for cylinders under load. In such cases, pressure relief is via auxiliary pistons, controlled by auxiliary cones for large crosssectional areas (see Fig. 12). Opening pressure against spring can vary between 0.5 and 3 bar. In feeder valves, which require very low opening pressures, operation occurs by the weight of the cone itself, and they can therefore only work in the vertical position.
2.3.3 Pressure Control Valves The essential characteristic of these valves is a special fimction whereby switching occurs when a predetermined pressure is reached, usually against spring resistance. The connection can be continuous (via change of throttle cross-section) or discontinuous (switching).
Pressure-Limiting Valve. With this rype of valve, oil is allowed to flow freely into the tank up to a predetermined pressure, after which any slight increase in pressure rapidly opens the throttle and so limits system pressure. With directly controlled pressure-limiting valves, oil pressure lifts the cone off its seat against a spring force.
Pressure increases sharply with flow, giving an operational limit above the "saturation" point, depending on the preset pressure. Alternating static and dynamic forces acting on the cone tend to produce flutter, but this is overcome by damping.
Preset Valves. For large flows, these are constructed as shown in Fig. 13. The main cone is held in the "closed" pOSition by a weak spring and backpressure, until a small, directly controlled pressure-limiting valve opens and pressurises the rear face (with throttle between admission and rear face). When the rear face is pressurised via an additional 2/2 directional control valve, the preset valve can fimction as a rotary valve. Remote control can be effected via connection of further preset and loading valves at X. Pressure Control Valves. These maintain constant downstream pressure independent of higher upstream
b
Figure 11. Preset 4/3 distribution valve with electromagnetic operation (Mannesmann-Rexroth, Lohr): a design, b symboL
a
A
2 J
~---
b
Figure 12. Rescuable nOll-return valve (Mannesmann-Rexroth, Lohr): a design; A, B working connections, X control connection; 1 release piston, 2 main cone, 3 preset cone; b symbol.
2.3 Valves. 2.3.5 Proportional Valves
liW"
~ I
._J___ ____
. . -
a
b
--
-
IF
_~-,
1 -
tp
Figure IS. Schematic of flow regulators: a 2-way design , b 3-way design; P pressure connector, F working connector, T outlet connector. I measurement throttle, 2 throttle piston, 3 spring.
Figure 13. Preset pressure limiting valve for plate co nstruction (Mannesmann-Rexroth, Lohr) : a design (I preset valve , 2 throttle.
3 main cone) , b symbol.
pressure by throttling, if necessary by additional release of outflow ( 3-way pressure control valve).
Pressure Switching Valves. With these valves, new paths for the working fluid are opened when the predetermined pressure is reached. The self-controlled type continue to switch in a subordinate circuit whi1e maintaining pressure in the main circuit (emergency valves, servo valves). Remote-controlled valves, depending on pressure, switch to another working Circuit, or allow depressurised free circulation in it (cutoff valves, load-storage valves).
2.3.4 Flow Control Valves Flow control valves are the simplest way of controlling the rate of motion in a hydraulic drive. In principle, they are continuously operating throttle valves with adjustable throttle area. Depending on this area, and on the square of the flow velocity, a pressure difference arises in the valve that is part of the pressure drop in the system.
Single Throttle Valves. Control is from outside and is therefore unrestricted (see Fig. 14). With fixed pressure level and given motor load the residual pressure dro p over the throttle remains at a level equivalent to a fixed pressure flow. Changes in circuit pressure or motor load cause changes in flow. Throttling should occur over the shortest
possible section (aperture), otherwise there is a strong oil viSCOSity (temperature) effect. When it is important to maintain constant working speed, flow regulators should be used. Here, with flow as the measured variable, there is a limited adjustment of the throttle section , and hence a readjustment of the valve pressure drop for constant flow.
em1
il -tVE-+ ttffi Figure 14. C.orrunon throttle designs.
TWO-Way (Main) Flow Regulators. Construction is as shown in Fig. 15. The flow-measuring orifice 1, adjustable for flow strength, generates a pressure drop I1Pm (approx . 3 bar) , balanced by throttle piston 2 against spring 3 . This is adjusted for a pressure drop I1p, ~ (pp - I1Pm) - PI" Motor load fluctuations first produce small changes of V. The measurement orifice pressure drop altered by 111]';' thus moves 2 to a new throttle position I1p~, which then makes I1Pm ~ const., i.e. V ~ const. The pump supply excess then flows back into the tank through the pressure-limiting valve. Only unidirectional flow is possible in a flow regulator; if flow control in both directions is required, non-return valves must be used. With large motor fluctuations, the controller should be on the discharge side. Three-Way Flow ControUers. These provide constant motor inlet flow by diverting excess flow from the pump (see Fig. 15) . Construction is mainly as described above , except that the throttle piston 2 opens an additional outlet. It is only possible to use these on the motor inlet side. Accumcy is 2 to 5%. Flow Distributors are constructed on similar prinCiples. The pressure drops in two parallel measurement orifices are balanced by a single throttle piston , which then throttles both or if necessary the differentially loaded motor branches. A very good flow distribution for both flow directions is possible by parallel circuits of two-gear motors with mechanically coupled shafts.
2.3.5 Proportional Valves Magnetically operated distribution valves act only with a discrete switching function (i.e. digitally). With proportional valves, analogue c onversion of electrical signals
into hydraulic actions is pOSSible. The magnetic force determined by the feed c urrent is balanced against the effect of pressure or a pressure drop, and thus a proportionality between the electrical input pammeter and the output pressure or flo w strength is established. There is no feedback between mechanical components (cost-
Figure 16. Proportional DDS pressure valve ( MannesmannRexroth , l ohr): / magnetic system, 2 baffle plate, 3 nozzle , 4 preset valvt', 5 main piMon.
Hydraulic and Pneumatic Power Transmission. 3 Structure and Function of Hydraulic Transmissions
effective compared with servotechnology). Reproducible accuracy is :::t: 2%. Construction and function are as shown
in Fig. 16. The magnetic force operates on the nozzledeflector-plate system and produces a control flow proportional to chamber pressure (i.e. system pressure). Function is affected by flow strength, thereby limiting application, e.g. as a precontrol unit for a pressure-limiting system.
2.4 Hydraulic Equipment For connecting pumps, motors, and valves (hydraulic circuits), tubing of steel (DIN 2391, ISO 3304, DIN 2413, ISO IDIS 10 400), or synthetic rubber with fibre reinforcement, pipe couplings (DIN 2367), oil containers (VT ~ (3 to 5) capacity per min), ftlters (fineness 10 to 30 to 60 fLm), and hydraulic accumulators are necessary [3].
~ Structure and Function of Hydraulic Transmissions_ 3.1 Hydraulic Circuits The circulation of pressurised fluid in a hydrostatic drive is referred to as a circuit, which may be open or closed, and with or without a feed pump. A circuit should be protected against overload by at least one pressure-limiting valve, while closed motor circuits with small permissible loadings should have their own pressure-limiting valves downstream of the distribution valve.
3.1.1 Open Cirarits (Fig. la) Open circuits incorporate an oil reservoir. The pumps always feed in the same direction, and the outlet flow from the motor back to the tank is unpressurised. Changes in the working direction of the motor are brought about by diverting the flow through a four-way distribution valve. Hydraulic circuits with fixed-displacement pumps mayor may not have unpressurised circulation under idle conditions, but variable-displacement pumps usually revert to zero supply. Advantages of open circuits are the way in which excess heat is carried away by the flow, as well as cooling and cleaning of the oil in the tank. The
main disadvantage is the unidirectional flow. The motor braking power, which is developed when for example load falls off, can only be offset by throttling at outlet (exit flow throttle valve, or, with high flow rates, a speCial brake valve).
3.1.2 Closed Circuits (Fig. Ib) In a closed circuit, low-pressure oil from the motor outlet is led back directly to the pump suction side. Oil can flow in either direction; that is to say, not only is energy supplied to the motor by the pump, but the motor can also act as a brake on the pump and so feed back to the prime mover. Thermal loading is conSiderably lower than in open circuits for this reason. Change of motor direction is either by distribution valve (unidirectional circuits) or by changing the direction of pump supply with a reversing mechanism (circuits with variable flow direction). Closed circuits have a feed pump that pressurises the appropriate low-pressure circuits to 3 to 8 bar. This is to avoid cavitation on the suction side of the main pump, to reduce leakage losses in the main circuit, and to exchange feed-pump excess flow with that in the main circuit for cooling and cleaning (without a highpressure cooler and ftlter, a flush valve is needed; 5 in Fig. Ib). Feed flow is about 10% of main flow. Depending on flow direction, the circuit is protected by one or two
cross~connected
pressure-limiting valves. For
idle flow, the main pump circulation valve is used (with variable machines for about 4% of Vm ,,).
3.1.3 Semi-closed Cirarits
a
With closed circuits incorporating hydraulic cylinders, the different bottom and shaft-side volumes have to be taken into account. Depending on working direction and area ratio 'P, large differential flows may have to be introduced or removed from the circuit. Flush valve and feed pump are calculated accordingly, and if necessary a back-pressure valve 2 is provided for the main pump (Fig. Ib).
3.2 Operation of Hydraulic Transmissions 3.2.1 StartLmgProcess b Figure 1. a Open circuit with rotary motor and flow control hydraulic cylinder connected in parallel; unpressured circulation in idle condition. b Closed circuit with variable flow direction with rotary motor and hydraulic cylinder; I feed pump, 2 return valve,
3 circulating valve, 4 motor distribution valve, 5 flushing valve, 6 preset valve, 7 cooier, 8 filter.
The pressure in the working circuit is controlled primarily by motor load: load pressure is flpF' To this must be added the pressure associated with overcoming mechanical friction J.p" flow losses in pipes and valves J.Ph, and, where flow velocity is not constant, with accelerating masses flp,. The pump pressure J.p; = J.PF + flp, + J.Ph + flp, is limited to apPLY by the pressure-limiting valve. A greatly simplified representation of the starting conditions is
3.3 Control. 3.3.1 Variable-Speed Drive Units
ItM-
show that the gear ratio is affected (also during operation) by two factolli: (a) (b)
Changing VOP/VOM = variable-speed drive, Changing Vv/Vp = throttle-controlled drive.
3.3 Control Zeit f
3.3.1 Variable-Speed Drive Units
fa
Figure:l. Start·up of hydraulic drive. t, acceleration time; PPLV pressure-limiting valve adjustment; p, load pressure; p, friction pressure, Ph hydraulic losses, Pa acceleration pressure; Pp, P; pressures at the pump; v velocity.
Depending on type of control, these may be c1assif.ed as:
Primary Coatrol. Pump adjustable from zero to maximum feeds a constant-displacement motor.
Secondary Control. Pump delivelli a constant supply to a variable-speed motor. given in Fic. :I. As long as the motor has not reached the speed corresponding to the pump supply flow, the excess flow is diverted to the pressure-limiting valve, and the pressure in the circuit is tlpPLV' The flow losses tlph increase with supply flow .1 - V or - V'; motor acceleration arises from the pressure tlp. available after subtracting the usable loading pressure tlpF and the pressure used in overcoming friction tlp" up to the Iintiting pressure tlpPLV' A more accurate consideration takes account of the energy involved in oil compression and in elastic deformation of circuit components, especially in circuits with large oil contents or with elastic components (reservoirs, hoses), which lead to transient effects and increased acceleration times. Once the motor has reached its final speed, pump pressure falls back to .1pp . When the motor is suddenly started from idle, a pressure peak P'P occurs before the pressure-Iintiting valve can respond because of the mass effect of the oil flow, the pump and the drive motor. This peak can be reduced by elastic components (hoses, reservoirs), and rapidly reacting pressure-Iintiting valves, but is most safely counteracted by the rise-delay time associated with the opening characteristics of the circulating valve.
3.:1.:1 Formal Description Under steady conditions, the motor generates mechanical power p.b = PmM = MM .... , or FMvM to overcome working and friction conditions, and therefore a hydraulic power P bM = VM.1PM = PmM/1J
VM
=
Vp - Vv
= Vp(l -
Vv/Vp)
=
Compound Control. Both components are controllable; control can be series or parallel. Primary-control drives are very widespread, though the use of primary-secondary drives is increasing. The work· ing diagram is shown in Fig. 3. During the primary phase, motor torque MM and circuit pressure tlp remain constant. From startup, the pump, running at fixed speed, delivelli a flow Vp, which increases by adjustment, and the drive speed n Mand power P increase proportionally. At n" the pump is at maximum, and further increase of output speed is possible only by secondary control, i.e. by reducing the motor volume VOM . When maximum pressure tlpmaa, and hence constant power P, is reached, this adjustment produces a hyperbolic decrease in motor torque MM' If the drive power is less than the "gear-angle power" P E = .1Pmaa Vmaa/ 1Jtl', then there is a power Iintit on primary control, which only happens with a corresponding pressure revellial. Secondary control takes place via speed-controlled or positionally controlled hydraulic motolli, fed from pressure mains (as comparable electric motOIli are fed from electric mains). Pressure-controlled pumps and if necessary hydraulic accumulatolli keep circuit pressure con· stant. Load torque adjustment at constant speed is achi· eved by adjustment of motor swept volume, i.e. the power change produces different values of motor displacement. Advantages over the flow controlled machinery described above are: better time response, especially over greater distances; parallel connection of several units possible without mutual interaction; braking energy fed back into circuit.
VP1Jv<;,
.1PM = tlpp - tlphL = tlpp 1Jh.·
Overall E.ffictency 1Jt = Pab/P= =
VM ' tlpM1JtM1JtP/ Vp' tlpp
= 1JvU1Jbo1Jtl'1JtM·
The heat generated by flow friction losses Qa = P, (l - 1Jt) must be removed by convection in the pipes or the tw, or if necessary by an additional cooler. The allowable excess temperature of the oil over ambient is around 30 to 50 K. The definitions
Speed Ratio v
= ..../Wp = (VOP/VOM) (l
-
Vv/Vp) 1JvP1JvM'
Torque Ratto Figure~.
control.
Characteristics of a drive with primary-secondary
I#WII
Hydraulic and Pneumatic Power Transmission. 3 Structure and Function of Hydraulic Transmissions
\
./-'v<:;,-.. . v.....---.......'\ Pressure cutoff
1
Power ratio PIPm"
I
Hyperbola p' V= canst
I I I I I
.~.
Figure 4. Throttle-controlled drive, circuit and drive characteristics: a bypass throttle drive; b drive characteristic for a; c main flow throttle drive. MM load moment, 1 pump, 2 pressure-limiting valve, 3 throttle, 4 motor. ~.~.2
Throttie-ControUed Drives
For low powers « 5 kW), hydraulic drives with low-cost fixed-displacement pumps and motors are more economic, though of course their working speeds have to be controlled by diverting flow from the main circuit. With bypass throttled circuits (Fig. 4a), the diverted flow is led from the throttle into the tank. As motor loading MM increases, the circuit working pressure also increases and
the diverted flow, or in other words the output speed n M , is reduced. Flexibility is affected by throttle setting, and maximum torque by the setting of the pressure-limiting valve. With primaty-flow throttling as in Fig. 4.:, the fraction VM of the pump supply flow reaches motor 4 via throttle 3, and the outlet flow Vv returns to the tank via pressurelimiting valve 2. Pump pressure ApP' kept constant by adjusting 2, is reduced in the throttle to about ilPD, of the pressure ilpM required by the motor loading. Motor volumetric flow VM and hence speed n M are therefore controllable by adjusting throttle area AD" since VM - AD,
'l/ilPD,'
As load moment MM increases, working speed nM falls as a consequence of shift of pressure drop ilpn, - ilp/" = App - tlp,;. Maximum load moments are preset by adjustment of the pressure-limiting valve. Both types of drive are only usable when there is little requirement for speed constancy (saws, woodworking machinery). Their main advantage is overload protection because of their great flexibility under load [4]. It is also possible to have parallel connection of several units on one load-sensing pressure main (modified throttle control). ~.~.~
Automatic Control of VariableDisplacement Pumps
Since adjustment of flow volume depends mainly on loading (pressure), the term "controller" for the adjusttnent device is not really accurate.
I
esidual flow %
ill
m~ w~
~
Vp/Vpo in %
m 00
~
m
Figure S. Schematic and mode of operation of a power-regulating valve with pressure cutoff: 1 control piston, 2 control pin, 3 spring guide, 4\ to 4~ spring series, 5 outlet valve, 6 pressure piston, 7 control pump.
Zero-now Control. This is the simplest type for the situation where a given pressure has to be attained, but no more flow is available (presses etc.). The pumps are maintained at maximum via spring pressure. Once this pressure is exceeded, the pressurised control piston of the pump moves back to lower supply. To ensure cooling and lubrication under maximum pressure, minimum flow is maintained at 4% of main flow. Pressure increases linearly and the supply is reduced, i.e. the power curve is parabolic. With the preset type, the pressure characteristic is horizontal, and power falls off linearly (constant pressure control). Constant-Power Control. By setting up a number of springs so that the force of the maximum total pressure is equal to the spring force of the next set, or where extra springs are stepped in after a predetermined interval, a hyperbolic adjustment P = Ap V = const. (precise load limitation) is obtained via a series of tangents (see Fig. S). Preset types provide better dynamiCS, and allow additional functions, e.g. controlled starting, to be fulfilled. Pressure Cutoff. If no more flow is available after reaching a given high pressure on the hyperbolic power curve, the controller can be combined with a pressure cutoff device (Fig. S). The spring packs 4, to 43 of the power controller act against an oil-pressure loaded piston 6. When the maximum pressure is reached, cutoff valve 5 opens and releases the load on the piston. The pump then reverts to idle condition because of the pressure on control piston 1.
4.2 Design of Hydraulic Circuits
IfW'"
_ _ Configuration and Design of Hydraulic Transmissions_ AB
4.1 Hydraulic Circuit Arrangements 4.1.1 Remote Drive Transmissions If feed motion only is required with a plunger cylinder, then circuits with 3/2 distribution valves will suffice (Fig. la). Flow is led from the pump into the cylinder, but, at idle, pump and cylinder are switched into the return flow circuit. If the cylinder has to stop at each intermediate position, a 3/3 distribution valve with a closed central position is required (Fig. Ib). Normal switching of the motor results with a 4/3 valve as shown in G3 Fig. 1. Valve arrangements with crossswitching either A- B-T or A- B-P (for rotary motors only A-B) at the middle position permit external motor adjustment, e.g. to control units (floating installations). Rapid feed of machine tools is most often provided by cutoff pumps (Fig. :la). The increasing pressure at the beginning of the working stroke switches in the fastaction pump with its large supply to the low-pressure circuit. With cylinders, the different values of the bottom and shaft-side areas of the piston face are used. The fastaction valve 4 in Fig. :lb initially opens both cylinder connections, and the piston rapidly moves forward; since the pump flow is only acting on the shaft side, V E = VIA". The pressure increase then switches the fast -action valve out, so that only the bottom side of the piston is exposed to the action of the supply flow (step function corresponding to cylinder area ratio in valve needed). With multi-motor drives in parallel, it must be noted that if several machines are switched on simultaneously only the motor with the smallest loading will advance, and that non-return valves must be provided to prevent the other motors reversing. Simultaneous feed of all motors in such cases can be achieved by including flow regulators. The pump circulating valve is so arranged that it closes under either hydraulic or mechanical control upon operation of a given distribution valve. Series Circuits are possible when the distribution valve can be loaded on the back-pressure side with the full working pressure (connection T); the circulating valve is not needed. Simultaneous
Switching of several motors is not permisSible, otherwise opposing pressure and flow conditions would prevail (see Fig. 3). Use of 6/3 valves (Fig. 4.) prevents possible error Switching, since the operation of a valve shuts off the downstream motor from the flow. Arranging these valves in blocks pennits freely interchangeable series-parallel circuits (Fig. 4b). Because of the problems associated with it, reference should be made to the literature [5) for applications of synchronised switching.
4.1.:1 Variable-Speed Drive Units In compact drives, pumps and one or more motors are
built into a single housing. Closed circuits are generally used. The housing contains all necessary auxiliaries, such as feed pumps, pressure-limiting and flushing valves, and positional and other controls; it also acts as an oil reservoir.
4.2 Design of Hydraulic Circuits The design of hydraulic circuits is carried out as follows: (a) (b) (c)
(d) (e)
(g) b
(h)
Record and set in chronological order mechanism function; establish cycle time and working speed. Evaluate working principle (linear, reciprocating, or rotary motion). Record forces (torques) as a function of time; establish pressure range. These are selected so that the permissible working ranges of available hydraulic series components are used so far as possible (cost minimum), while allowing a safety factor for overload (around 10 to 15%). Select motors as (a), (b) and (c). Calculate flow strengths by drawing a volume-time diagram using data from (a) and (d); Select pumps for pressure range and supply (dimensions, number, fixed or variable), decisions on storage units. Select control type (manual, partly or fully automatic), or controllers. With decisions (e) and (g), select valves (dimensions, type of circuit, operation), establish
AB
a
2
:t'iaure Z. Rapid-feed switching: a with rapid-feed switching pump, b using piston area ratios with rapid-action valve; 1 rapid-
action pump, 2 working feed pump, 3 cut-off valve, 4 rapid·feed valve.
~XI~
Figure 3. Circuit series switching; multiple motors with 4/3 distribution valves.
(f)
Fipre 1. • Simple plunger cylinder Switching with circulation, end-stop motion. b Valve design with intermediate position.
AB
AB
:: s~~: I!J.'
.
~·~'!1f--tl--1~
11 L---JL-----..J4++ tIL-l
J:XIO]:II~
L
~
Figure 4. a Safety circuit series switching with 6/3 distribution valves. b Extension for series-parallel switChing.
ItjlGI
(i)
Hydraulic and Pneumatic Power Transmission. 5 Pneumatic Installations
pipe diameters (permissible oil velocities up to 1. 5 m/ s in suction piping, 3 to 6 m/ s in pressure piping in pressure range 100 to 400 bar). Calculate output power and losses: drive power; establish loss balance including heat removal.
2
After this, considerations of economy, ease of assembly and maintenance, operational safety, and other factors (climatic conditions, service and maintenance staff) must be taken into account. The most important interpretative aid is the volumetime diagram for the cycle time tT (tT with recurring working strokes: time for one working cycle), in which the displacement flow of the motor is additive (as below) (see Fig. S).
Time I
12
15
Time t, to t2: pressurising with two cylinders 1, simultaneous loading of pressure reservoir 2. Displacement volume 2V, + V2 (V2 from storage diagram). Time t3 to t.: rapid feed of working cylinder 3 from ! displacement, volumes V3/3. Time t. to t,: working feed cylinder 3 in idle stroke, volumes 2V,/3. Time t6 to t7: rapid-return cylinder 3 for full stroke, volumes V,/cp. Time t8 to t9: depressurisation, volumes = 0, pressure cylinder return via spring force. Time t9 to tT: change workpiece, volumes = O. VM/(t,+, - t,) = V = flow strength = required pump flow. since motor displacement flow + opposed pump supply flow = O. The diagram is espeCially suitable for determining the storage circuit flow (cycle time tT large compared with motor working time). Pump supply is then Vp = ~VM/(O.9tT)' i.e. the pump can be made as
• • • • • • • • •11 Pneumatic
Figure S. Switching diagram and displacement diagram for press circuits: 1 pressure cylinder, 2 pressure reservoir, 3 working cylinder, 4 working pump, 5 rapid-action pump.
small as possible and storage volume = difference (displacement line - supply line) from the zero line .
Installations • • •
Properties of pneumatic drives are as follows:
5.1 Pneumatic Components
Advantages
Compressors. Pneumatic equipment is almost exclusively fed from compressed air mains with a centralised compressor unit.
Rapid operation because of high flow speed (up to 40 m/s in piping) and low mass of working medium. High switching frequency (hammers, etc.). Highly compressible working medium (air), with virtually constant pressure forces even with change of position (pressure cylinders, shock absorbers). Insensitivity to temperature changes, although there is the danger of condensed water freezing in control valves in open-air plant. Low piping costs, because air is simply blown off after giving up its energy to the control valves. Also small leakages are unimportant, and there is no danger of contaminating sensitive materials (foodstuffs, etc.). Usually cheap to install, since in many cases there is already a compressed-air main which can be used.
Disadvantages Because of the compressibility of the working medium, use is restricted to mechanisms with mechanically or otherwise limited end stops. Pneumatic drives are only suitable for relatively low powers because of low working pressure.
Motors. Rotaty motors are usually of vane or gear type. As they do not use the expansion work of the compressed air (full-pressure machinery), efficiency is low. They are also usually very noisy, making outlet silencers necessary. Oscillating motors are predominantly of the toothed-rack type. Linear motors (cylinders) are in prinCiple identical in structure to the hydraulic cylinder, except that they are rather lighter owing to the lower pressures involved. For small strokes, membrane cylinders are suitable, in which the piston is replaced by a rubber or plastics membrane loaded and supported by an intermediate piston rod and cylinder jacket. For larger strokes a roller membrane is used. For cutting, press, and die work, in which the working volume is ftlled for only short periods, percussion cylinders (which use expansion energy of the compressed air) are more economical than full-pressure cylinders. Here, the compressed air is fed into a pre-chamber and on starting led through a large opening in the cylinder bottom. During this expansion, it encounters the largearea piston face and imparts high kinetic energy to the system.
Valves. In construction, function and operation, pneumatic valves are generally similar to those used in
5.2 Circuits
hydraulic units. However, because of the lower pressures and higher flow velocities iovolved, they can have smaller dimensions and iocorporate lighter materials such as aluminium and plastics. There is increasiog use of seat valves, because of greater operational safety and little requirement for lubrication. For complex circuits, disc-type distribution valves ace
used. For small units, piston slide valves set io control bores, or runniog io elastomeric sealiog elements, are used. Larger units usually have O-ring seals on the piston or io the housing, because lappiog-io of the matiog parts would be prohibitively expensive.
Series Installations. Compressed air for pneumatic units should be ftltered free from particles of dust and scale, should be dry, and should carty in mist form the lubricant necessary for the equipment drive. Furthermore, its pressure should remaio constant at the correct level independent of the maios pressure. Installations are therefore provided with what is usually called a "service unit", consistiog of a combioed ftlter, pressure regulator and oil mist generator.
Filter. This consists maioly of a combination of a vortex chamber (to centrifuge out large particles and droplets), with a metal-mesh, textile, or sioter ftlter downstream. Dirt and condensed water are collected in a transparent plastics contaioer, which enables the level of contamination to be checked.
Pressure Regulator. This adjusts the pressure downstream of the throttle area by means of a membrane held agaiost spring resistance. Increasiog secondary pressure reduces the throughput cross-section; additional equalisation pistons to compensate for primary pressure give even more accurate control.
Oil Mist Generator. This works on the carburettor principle: pressure drop across a nozzle draws oil from a transparent storage tank and iojects a fme mist into the airflow. Adjustment of the oil-air ratio is made via throttles io the air nozzle and the oil pipe. Some units use a so-called "micro-oiler", by which oil droplets that are too large are removed from the airflow by diverting it ioside the oiler.
5.2 Circuits Automated units with sequence control are cheaper to produce than those with individual initiation of the work-
'tW-
,~~l~ I
I
I L _____ ~----"
Figure 1. Simple sequence switching of two pneumatic cylinders with impulse valves: 1 start-valve; 2, 3 4/2 distribution impulse valves; 4, 5 cylinders; 6, 7, 8 push-button valves, roUer-operated.
iog cycle; they are also safer in function, because the ioitiation of the next stage is connected to the completion of the previous one (sequential circuit). This type of equipment can either be electrically controlled, or can be fully pneumatic, with push-button valves to trigger impulse distribution valves. The latter type has the advantage that the whole unit takes compressed air from a siogle source and so is less subject to disturbances (Fig. 1). The starter valve 1 is pressed and this puts the impulse valve 2 ioto the state that allows the piston io cylioder 4 to advance. At the end of its stroke, the piston reaches contact 7, which switches impulse valve 3 to operate cylioder 5. This in tum reaches contact 8 at the end of its workiog stroke, which sets cylioder 4 ioto reverse via valve 2, and the piston, when it reaches contact 6, gives the feed command to cylinder 5 via valve 3. The unit thus returns to its ioitial position, from which another operation of starting valve 1 will initiate a new workiog cycle. Storage units in pneumatic circuits act like capacitors, allowiog time-dependent operations to be iocluded. The disadvantage of unequal feeds io pneumatic drives with varying loads, as well as the sometimes excessive feed velocities, can be countered through a combination with hydraulic control equipment. Here the compressed air provides the feed force, but the hydraulics provide the feed velOCity (hydropneumatic feed units). Constant feed and if necessary position control are also possible usiog electronically controlled proportional throttle valves.
Water Hydraulic Systems Water has certaio advantages as a workiog fluid that favour its use io hydraulic systems. Applications include units in which for safety, or because of environmental considerations, the use of mineral oil is illegal (e.g. miniog), or where the inherent safety qualities of water (fire protection, waste management), or its economics, favour its use. There are two distioct areas of application, dependiog on basic design. What may be termed water hydraulic systems are those specially developed to use water; they tend to be large and expensive, such as in slow-runniog plunger pumps, large presses, liftiog equipment, or miniog construction. The other category, industrial water hydraulics, contaios simply oil-hydraulic equipment adapted to use water. The workiog medium is HFA fluid with a water content
> 95%. Oil-water emulsions need to be carefully checked for, e.g., micro-organisms, pH value, or separation. Sioglephase solutions of synthetic concentrates do not give problems. Additives are required for corrosion protection, especially in steam condensation areas. Properties are essentially determioed by the water content: low contaminant absorption (high demands on ftlter); operatiog temperature limited to + 2 to 50°C (ice formation, cavitation); viscosity only slightly (50%) greater than water; extremely poor lubrication properties. Use in commercial applications imposes considerable restrictions. Workiog pressure and speeds must be reduced by about 40%, and the service life of rolling bearings can fall by a factor of 20. Where slidiog beariogs are used, there is little hydrostatic support. Leakage losses are
ItW':'
Hydraulic and Pneumatic Power Transmission. 7 Appendix G: Diagrams and Tables
3 to 10 times higher than with oil, requiring larger pumps. Distribution valves above nominal size 10 may only be seat valves (cartridge-V; see 62, Fig. 10), but conullercial pressure-limiting, pressure-regulating, and throttle valves can readily be used. Units with commercial components are relatively problem-free up to about 70 bar with HFA fluid. There is no difficulty with sealing, but the use of zinc and cadmium, or paper (mters) are not permissible. Newly developed pumps for higher operating pressures have encapsulated, self-lubricated bearings or pressurised hydrostatic bearings. Price in relation to standard pumps
is 1.5 to 2, with much the same ratios for unit and energy costs. Acquisition costs for HFA are about 25% of those of oil, but the additional costs of maintaining the working fluid need to be taken into account. No general recommendations can therefore be made, except that individual decisions must be taken on a cost basis in which the primary definitive costings for acquisition and operation are offset against secondary advantages (e.g. reduced environmental damage).
Appendix G: Diagrams and Tables • • • • • • •
I
2000a
l
1000a
\
500a 200 0 1000 500
'-
\
~~
\
20 0
\.
\
I'\.'"
.-~ ",-' " r\"\
,,"\ "-
1
~~
0 ;\~
1\ ~"'"l\ ~ ~ ~ I'~ "
10a
~
a
.'"
I~
~, ~~
f"-
\ 1\ 1\"" ~ "" "\ ~~ ~ \ " l~ 1\ H~ 1\ '\~ ~ '\
~
~
I' ~\
H~
f\.
a
o
\
5
f\
D-
1\1\
\
"6-
~\
"
-20
-10
a
10
20
30
I_a
"" ""'"
~\ ~ \ 1\I\.'\.
i-
I' 1\
1\1
1
2 -30
.~1b
~
50
60
100'r Temperature II Figure 1. Viscosity-temperature diagram for hydr.mlic fluids.
70
80
90 'C 10 I 210'F
Hydraulic and Pneumatic Power Transmission. 7 Appendix G: Diagrams and Tables
I
I Meamng and explanation
Symbol Hydraulic pumps
~
With adjustable-displacement volume a. with one, b. with two directions of flow
Hydraulic motors
lBllJl
c)
-------
00= be>
IRotary motors with constant-displacement
volume: a. with one, b. With two directions of flow
=D=
-~
Oscillating motors
rn=
linear motors (cylinders)
tJ§=
Single-acllng
Double-acting With tWO-Sided, adjustable damping
GITh
---
Drives for unldirecllonal outlet rotation with adjustable pump and constant-speed motor -
~ S I: II! If:'-.'~Zlnlll~
j
Compact hydrautic drives
----
?
-----------------
.-.------
,--L,
Working plplllg, piping for energy transfer
-----
Control piping, piping for signal transfer
----- ---
Overtlow oil lines
Pressure control vatves
"------"
Hoses
0ttJ
0
a. Crossed pipes, b. Connected pipes
Hydraulic receiver
Within each field arrows show the direction of flow, closed connections shown by break If the change of pOSition of a sWitched route IS attached to a connection, the arrow 011 th, side is broken Operallng symbols are shown perpendicular
I to the connecllons outside the rectangle
2/2 valves, Idle position closed, hand-operated With lever 3/3 valves, idle position closed, spring-centred. pressure-operated
I
4/3 valves, low-pressure pump circulation at mid-position spring-centred, magnetic or pre-set operation
Shut-off valves
---
0+ b+
Valves are represented by a rectangle Number of divisions = switching states piping shown attached to the Idle condillon
Distribution vatves
Double-acting with piston rod at one side
lID§=
HydrauliC p"plng etc
Meaning and explanation
Hydraulic valves, general Pumps With constant-displacement volume
00= be!> o~ b45=
v!J:x
Symbol
'tWO
~---
bOJ
~J
1fJ b[~
a
Spring-load non-return valves Controlled non-return valves, pressure-opera ted closure
Pressure control valves (general)' a. with open idle position, b. with closed idle position
IPressure-limiting valves Pressure control valves' a. without outlet opening, b. double-edged valve with exit opening
Flow valves
-I~
Throttle valve, adjustable Three-way flow control valve, adjustable
Figure 1. Symbols for hydraulics and pneumatics according to DIN ISO 1219 (selection).
'M"
Hydraulic and Pneumatic Power Transmission. 8 References
References General. Pippenger and Koff. Fluid-Power Controls. McGraw-Hill, 1959. - Streeter and Wylie. Fluid Mechanics. McGraw-Hill, 1985. - White FM. Fluid Mechanics. McGraw-Hill, 1986. - Boxer. Fluid Mechanics, Work Out Series - Macmillan, London, 1988. GI to G6. [1) Feldman DG. Untersuchung des dynamischen Verhaltens hydrostatischer Antriebe. Konstruktion 1971; 23: 420-8. - [2) Schlosser WMJ. Ober den Gesamtwirkungsgrad von Verdriingerpumpen. 0 + P 1968; 12: 415-20. - [3) Roper P. Die Dynamik des HydroSpeicherkreislaufes. Konstruktion 1968; 20: 3419. - [4) Roper P. Hydrogetriebe mit Stromteilung Sammelschrift. V. Konferenz lIber Hydraulische Antriebe. Ces-
koslovenska Vedecko-Technicka Spolecnost, Prague, 1981. - [5) Zoebl H. Schaltplllne der Olhydraulik. Krausskopf, Mainz, 1973. - [6) Rechten AW. Fluidik. Springer, Berlin, 1976.
Standards and SpecificaUons. ISO 1219: Schaltzeichen (replacement for DIN 24300). - DIN 24312: Druck, genormte Druckwerte, Begriffe. - DIN 20021/22: Schlliuche mit Geflecht-Einlage. - DIN 24334: Hydrozylinder, Hauptmasse. DIN 24335: Pneumatikzylinder. DIN 24340: Hydroventile, Lochbilder. - DIN 24346: Fluidtechnik, Ausftlhrungsgrundlagen. - VDMA 23417: Schwerentflammbare Druckfillssigkeiten, Richtlinien. - VDMA 24320 (DIN-E 24320): Schwerentflammbare Druckfillssigkeiten, HFA. - VD12152: Hydrostatische Getriebe.
Components of Thermal Apparatus H. Gelbe, Berlin
Fundamentals 1.1 Heat Exchanger Characteristics Heat exchangers are devices that transfer heat between two or more fluid flows in the direction of the temperature gradient. Their aim is to change the state of these fluids (cooling, heating, changing the state of aggregation and/ or other physical properties), and to assist in making processes economic (waste heat utilisation). Their distinguishing characteristics are:
Mode of Operation. A distinction is made between continuous-flow (recuperative) and discontinuous-flow (regenerative) heat exchangers. Heat Transfer. This can take place directly ('without wall', or heat-transfer contact) or indirectly (transfer by thermal conduction through separating walls). Examples of direct heat transfer include injection condensers, separating stages for the thermal separation of mixtnres of materials, and solar-powered distillation plant. Boilers, piping systems or vessels are heated directly by flames or flue gases, occasionally with the aid of a heat-transfer medium (organic heat-transfer medium, molten salts or metals). State of Aggregation of Fluids. A distinction is made between pieces of apparatus with flow and without phase change (preheaters, air coolers, flue-gas-driven superheat-
ers and others) and those with phase change (condensers, evaporator equipment. evaporator coolers and others). Calculations become more difficult if phase changes have to be anticipated on both sides (as e.g. in an evaporator-condenser).
Temperature and Pressure. Depending on their use, heat exchangers may be classified as either low-temperature (down to -100°C), normal (50 to 500 DC) or hightemperature (up to = 1400 DC, e.g. waste-heat boilers in petrochemistry), as well as vacuum, low-pressure (a few bar), high-pressure (100 to 500 bar), or ultra-high-pressure (some 10' bar).
1.2 Thermodynamic and Fluid Dynamic Design
12/
The aim is to obtain a large thermal power transfer A under conditions of optimum or maximum permissible pressure loss (see H1.2.3) and where the sum of the costs for the equipment, the required building and the energy including generation and transfer (pumps, pipelines) is to be minimised.
1.2.1 Thermodynamic Design of Recuperators The thermodynamic design is based on the transfer equ= k ' A, MM' where is calculated ation (see C10.2) from the balancing equations = m, cp' (t, - t 2 ) = m2 cp2 (I; - t;) (see Fig. 1).
12
12
12
Heat Transfer Coefficient. Because 1/k = 1/ a, + o/A + l/a, (see ClO.2), k is always smaller than the lowest value of a. Hence this smallest value must be improved by influencing the flow (crossflow, generation of turbulence) by increasing the velocity (increasing the number of tubular or shell ducts or deflection plates increases the pressure loss) or by insertion of fms or lamellae (especially in the case of gases with low avalues). Determination of the heat-transfer coefficient a as a function of the phase condition and the mode of flow
of the fluids as well as of the surface geometry (plate, interior or exterior of the pipe, smooth, undulating, fmned) and the position of the apparatus (horizontal, vertical) is carried out with the help of power series of non-dimensional characteristic numbers (see CIO.4 and II]). Leakage and bypass flow as well as non-uniform flow in tube bundles can be taken into account by means of correction factors II]. Estimated k-values are given in
Mode of Construction. Equipment featuring tube bundles (plain tnbes, hairpin tubes, field tubes and tnbe registers) is among the most widely used. Further subdivisions are according to the method of attachment (tnbe plates, headers) and the guiding of the bundles (spiral tube, wrapped bundle). Apart from these, plate, spiral, double-shell and lamellar heat exchangers are also employed. Size. Compact heat exchangers are those with heat-transfer areas exceeding 700 m' per nt' of volumetric structure (space vehicles, aircraft).
a
Exchange s,uriace area
b
Exchange surtace area
Figure 1. Temperature dependence in both media. a Counterflow. b Parallel flow.
Components of Thennal Apparatus. 1 Fundamentals
Table 1. Attention must be paid to the degradation of heat transfer due to accumulations of dirt.
I,----@
Temperature Calculation. The distribution of the fluid temperature and the mean integral temperature difference litM depend on the mode of flow (parallel flow, counterflow, crossflow, crossmixing, combinations of parallel flow and counterflow in multi-pass and coupled apparatus) as well as on the intensity of the heat transfer. (Transfer units N; k . A/mcp ') If lit., and 1it. 1 are the large and the small temperature difference for parallel flow and counterflow (Fig. 1) then the following formula applies: t.
t.tg , - 1it.1
(5)-----11
t;-@ @--Il @--I,
w
;_1 t _ -_ t2; _ _ GI
'. _
t;
w+W
t; ; w +ww [wIV + exp
tl-t; BGI ; tl _
(1 1) ]]
[ l-exp [ - -+- kA w W
,
[(1 + IV1) ]] -
;;
kA
(2)
,
(5)-----11
I' I;
"-ru I', I'1
~ I',
' , - '1
I,
®--®
I;----®
+-2A III-Ill 1;=1; + 1=1 111-1;1
I,-©
11=1,-+ 11,-1;)
®--'; C0--©
/;=1,-+11,-1;1
@--Il I, =1,
1 '1
t2.-.--t{
I, =11 + A~B 11;-11)
@----I z I, =11- ~=~ 11;-11)
~, I,
=', -A 11,-1,1
'1
, 1 IH,-BII,-I,I I'----®
~'l I;
I;--@
C0--©
I,----@
1 11
I, I;
I,-©
To check the calculations for a given heat exchanger (k . A known), two temperatures are sufficient to determine all the others with the help of Fig. 2 (required temperatures circled; given temperatures not). The quantities A and B for parallel flow and counterflow respectively (subscripts Gl and ct respectively) are given by A
~
I,
@--I,
(1)
tM ; In(t.tg,/ t.tkl
Parallel flow
Crossflow
Countertlow
I, =1; + 1~B 11;-1;)
t;----t z I1--I',+ l::A.(t' 1-B 1-I') ,
t
1 , - 11 ,;;,,-+11,-, 1)
1 11
, 1 ®--®
I,
,;=,,-+11,-, 1)
Figure 2:. Temperature distribution for counterflow, crossflow and parallel flow (Plauk). (3) t} - t~
B ;-C(
'I -
t;
(4)
where Wand ware the thermal capacities of the fluid flows with the higher and the lower temperature respectively. Here w or W; mc p , where m is the mass flow and cp the thennal capacity. Further, tlo t2 and t;, are the inlet and outlet temperatures of the hotter and the colder medium respectively. Flows that differ from parallel flow or counterflow result in smaller values for l1tM
t;
Table 1. Estimated k-values in Wf(rn! K) for heat exchangers of tube-bundle (shell aud tube) type (VDI-Wiirmeatlas, paragraph
The correction factor 8 is normalised to the mean temperature t.tM. f for counterflow, which has been calculated for the same starting and final temperatures. In Fig. ;t e is plotted as a function of the operating characteristic 5 ; (t; - t;)/(tl - t;) for cross-counterflow, with R; w/W = (tl - t2 )/(t; - t;) as a parameter. Here, fluid Fl
1.0
R~4b\~~ "--....:;:: ~ R-~
o9
30
;io 8
20\
~ ~1\1\\
\
J
CI-4)
" o. 7
Gas (~ 1 bar) against gas (~ 1 bar) Gas, high pressure (200 to 300 bar) } surrounding the tubes Gas, high pressure (200 to 300 bar) in the tubes Fluid against gas (~ 1 bar) Fluid against fluid Heating vapour surrounding the tubes, fluid in the tubes Evaporator: heating steam surrounding the tubes (a) Natural circulation, depending on Viscosity
(b) Forced circulation Condensers: } Cooling water through tubes Organic vapours and NH:\ surrounding tubes Steam turbine condenser (pure H2 0 steam, brass tubes)
5 to
35
1SO to
500
o51- R.U I' t'
o. 5
r,
0.1
0.2
03
0.4
0.5
0.6
\
1\8\05~ 4 0t
1 ltlII I II \1
0.7
0.8
0.9
10
S.lli-I,')!IH)
15 to 70 1SO to 1200 300 to 1200
300 to 1700 900 to 3000
1,-
300 to 1200
1500 to 4000
Figure 3. Mean temperature difference for the speCial case of cross-counterflow (Plank).
1.2 Thermodynamic and Fluid Dynamic Design. 1.2.3 Calculation of Pressure Drop
[ 1 1 (I 1)8 ]
llk.,=(T,+T2 )--T-+--T-+ r+r A
1
Ci:: 2m
2
I
2
B
where, with the usual units shown in brackets:
k., (W 1m2 K) = heat-transfer coefficient for fundamental
" 0.7 f---1f--It-+~
oscillation;
06 f---+----1+-f+--++ 0.1
~(
It{ r
0.2
OJ
04
0.5
0.6
Sclti-til/lt:-til
O}
08
A (m 2 )
total heat transfer area of a regenerator;
T" 1',
duration of hot or cold period respectively;
(5)
0910
2,4 or more
logarithmic temperature difference using the inlet temperatures t~, ( or outlet temperatures ~~, I': respectively; = heat-transfer coefficients referred to mean temperature of storage mass;
Ct'lm, Ct'2m
(W 1m' K)
Internal passes
8 (m)
thickness of a storage mass dement (i.e. diameter if shape is cylindrical or spherical);
Figure 4. Correction factors £ for a 1,2-heat exchanger and the technically acceptable region 121.
is not thoroughly mixed, while fluid FI' is. Figure 4 shows an analogous diagram for a \,2 heat exchanger (P = 1 pass through shell, r = 2 passes through tubes). For R = I, efficiency = operating characteristic 5 cannot exceed 0.57. The dashed line connects points of constant slope and separates the region where the curves have steep slopes (high sensitivity to perturbation of the operating conditions) from the region of technically acceptable values of e [2J (see H1.3)'
1.2.2 Thennodynamic Design of Regenerators Heat transfer takes place during two periods (the heating and cooling periods). The changeover sequence is shown in Fig. S.
A" (W/m K)
thermal conductivity of storage mass;
auxiliary function as in Fig. 6;
a (m'/s)
thermal diffu5ivity of storage mass.
1.2.3 Calculation of Pressure Drop The size of a heat exchanger is decisively determined by the pressure drop. For this reason, calculation of the pressure drop is one of the first steps in design because it determines the geometry of the tube bundle (diameter and length of the bundle and the tubes, flow cross-sections) required for the thennodynamic design. The lrictional pressure loss in fully fonned tuhular flow contains contributions /:, from the inlet and the outlet pressure losses and from losses due to internal stmctures and deflections (see A6.2). The total pressure drop in the tubes for one tubular pass is given by
Mode of Construction with static or moving storage
(6)
mass (Ljungstr6m-type) [\,3J. Further distinguishing characteristics are type and structure of the storage mass as well as changeover time. The storage mass tenlperature is subject to periodic fluctuations. The calculation of heat trAnsferred during a full period in one regenerator is carried out [1, j] in accordance with rdation
For values of A(Re, dlk) see A6 Fig. 8; for reference values for {, see A6.2.4. If the number of tubular passes is nT, then for constant volume flow and if w ~ n T the pressure drop in terms of that occurring in one tubular pass is given by
(5)
(7)
with kl ko = \ for the fundamental oscillation (approximation) and klko < \ when taking harmonics into consideration as in r 1. 4] and with
To calculate the pressure drop for Ilow through multiple-row tube bundles in cross-flow (external space of
01 8
/,
0.1 ~---.-----41_ UV
01 ~
00
:~:
8~
00 4 20
I---
1---1-
1-
r-...
40
-1-
60
80
6(1 1) 2(j-r,+r;
100
120
140
160
2
Figure S. Circuit diagram for regenerator~ (shO\vn for gas). Indices: 'inlet; " outlet; H, period of heating: K, period of cooling. FI 1 and Fl 2, cold and hot flo~' of medium. lJV. Changeover valve.
Figure 6. Auxiliary functon (jJ for the calculation of the heat-transfer coefficient (Hausen): I, plate: 11, cylinder; ill. sphere: 8, thickness of plate or diameter: a, thermal diffusivity.
Components of Thermal Apparatus. 1 Fundamentals
I,
I I,
.=10
Iy 12
1t2
I;
I,'
a
~ Ix
I,
.= 0.81
Such a comparison is shown in Fig. 7. The change in temperature 5 = 0.67, which pennits only 1,1 apparatus in counterflow, is distributed over two 1,2 apparatuses, each with 51 = 52 = 0.5: s = 0.81 for R = l. Apparatuses with more than p = 2 shell passes are barely capable of economic manufacture and operation. If the transfer equation and the balance equations (see H l. 2.1) are divided by the greatest temperature difference t1 - t; and by mI' Cpl, or, where appropriate, ril]. . Cpb six non-dimensional characteristic quantities, SI. 5" N I , N" Rand e, are obtained. For a given flow arrangement, two of these characteristics determine the other four. The following applies to the operating characteristic: 5 1 =N I ·e=5,·Nt/N,=52 ·R,
I; (11)
I:
I12
(10)
where (similarly for 520 N,)
The effect of the flow configuration on operating characteristics is shown in Fig. 8. A detailed description of the effect of different flow configurations and changeover alternatives on power transfer and calculation of temperature behaviour is given by Marrin [6]; tables in [1].
f,
b Figure 7. Series arrangement of a two reverse-current (2,2) and b two 1,2 apparatuses to fonn one 2,4 heat exchanger.
1.4 Efficiency, Exergy Losses 1.4.1 Efficiencies heat exchanger with deflection baffles) reference should be made to the literature [1]. In Eqs (6) and (7) A is the frictional coefficient of the tube, /;v the drag coeffiCient, w the fluid velocity, p the fluid density, Land d the length and inside diameter, respectively, of the flow channel. The influence of frictional pressure drop on the heat-transfer
The degrec of reversibility is a measure of the thermodynamic perfection of an apparatus or process [5].
coefficient a of a smooth tube or tube bundle carrying
swned exergy flows EO'., In order to assess the effect of
flow along its length is given approximately by Grassmann [5] for Re numbers> 6000 as
(12)
is the ratio of the exergy flow Ew leaving the balance region for the environment ( useful power) to the con-
the heat exchanger on the process, the overall process quality index
(8)
where K is the medium constant, d h the hydraulic diameter of the flow channel, it the volume and A the area of heat transfer. From the transfer equation Q= kA !1tM and from Eq. (8) and using k' = kla = const. (k' depends on the resistances of both fluids and on the roughness) Eq. (9) is obtained:
( Q ) 1.41 (itdgl79 !1p)041
A = k'K!1tM
5
(9)
'Ibis allows an estimation of the area A if the pressure drop is given.
1.3 Heat Exchanger Flow Arrangements and Operating Characteristics If the correction factors of Eq. (4) become too small (see Fig. 4), a series arrangement of 1,2, 1,4 or 2,4 heat
exchangers may be considered. This makes possible an increase of the effective temperature gradient up to values in close proximity to the optimum. For an economic assessment, the costs of this solution (compact method of construction, improved a-values, smaller s-values) must be compared with the cost of a counterflow apparatus.
Figure 8. Effect of the flow configuration on power t1"3llsfer for flows of equal capacity iRi = I (COntinuous line) and R = 0 (dashed line): 1, counterflow; 2, ideal; 3, single-sided; 4, crossflow, mixing
both sides; 5, parallel flow (6].
2.1 Basis for Design Calculations
(13)
O.~ ro-~,,---,,--,-----,--,---,
Tu =290 K
given by Glaser [7] can be used. for which the sum of the consumed exergy flows of all equipment with lossfree heat exchangers has to be determined.
0.3 I-+--t---f--\ '<:::1
.:} 0,2 f--'1--+t--1r--+--1--+----j
1.4.2 Exergy Losses Exergy losses are mainly caused by the following pro· cesses: fmite temperature differences. thermal conduction or inverse mixing, pressure drops, and heat exchange with the surroundings (insulation losses). Detailed examples for heat exchangers can be found in [8].
0.1
r--+--f---''cl----t--~-+__-_I
200
100
300 T in K
Losses Due to Finite Temperature Differences If a quantity of heat Q flows from the absolute tempera· ture T to the temperature T', the corresponding exergy
~OO
500
600
Figure 9. Exergy losses due to finite temperature differences.
Losses Due to Friction. The specific exergy loss is
loss is given by Evlf2=(T,,·6.T)/[T(T-/:;.T)]
where /:;.T = T - T'. For heat exchangers, approximate cal· culations using the mean logarithmic values (isobar, con· stant values of cp) can be carried out: (15) Similarly for T'. SubSCripts 1 and 2 relate to the inlet and outlet temperatures respectively. Equation (14) is illustrated in Fig. 9. The losses increase steeply with decreasing temperature (low-temperature technology). In order to limit these, small temperature differences are required.
Thermal Conduction or Inverse Mixing Owing to the usually high flow velocities, losses due to molecular or turbulent axial transpon processes are small. For thermal inverse mixing as a result of designs featuring multiple passes, see H1.3.
f 2
(14) ev
= -Tu
(vlT) dp.
(16)
1
The losses increase with the specific volume v of the fluid and with decreasing temperature. For determination of the losses [5], the pump power can be used as an approximation. An accurate analysis is necessary [8], because under cenain circumstances a pan of the lost energy can be regained as heat.
Insulation Losses. If T' = Tu and Q = (AlIl)A(T- Tu) are substituted in Eq. (14) (where II. is the thermal con· ductivity and Il the thickness of the insulation), Eq. (17) is obtained [5]: ~_(T-Tu)2 (AlIl)'AT
(17)
At low temperatures, good insulation becomes cost-effec· tive.
Apparatus and Piping Components 2.1 Basis for Design Calculations Permissible Operating Pressure PD' This is from 10 to 20% higher than the maximum operating pressure that can occur under the least favourable operating conditions. This determines the acceptance conditions in accordance with the Pressure Vessel Regulations, the aSSignment to test groups, the calculating pressure, the test pressure and the threshold pressure of the safety valve. Calculating Pressure p. In general, this is the permiss· ible operating pressure. Static pressures due to the feed media are taken into account only if they increase the stress on the components by more than 5%. In general, calculations should not use the pressure difference between internal and external pressures, but both pressures separately (the exception is in cases of panial vacuum). Calculating Temperature. This is the highest wall temperature to be reached (at least 20 DC) under normal conditions as indicated in Table 1. Characteristic Strength Value. This is the lower of the following two values (at the calculating temperature): tensile yield point Rm or 0.2 x limit Rp()21~ and long-term creep strength Rp!1()()()oo~ for 100000 h (see 02.2).
Safety Factor. This takes care mainly of uncenainties in the calculation assumptions and guarantees that during water pressure tests at the normal pressure of 1. 3PB, a sufficient margin is maintained relative to the character· istic strength value at 20 DC. This amounts to S = 1.1 at the test pressure for rolled and forged steels and to S = 1.5 at the calculating temperature and pressure.
Table 1. Calculating temperature III Type of heating
Calculating temperature
None By gases, vapours or liquids Fire, flue gas or electrical heating
Highest temperature of the feed media Highest temperature of the heating medium Highest temperature of the feed medium +20 °C for screened wall +50 °c for direct contact with wall, however at least 250°C
Components of Thermal Apparatus. 2 Apparatus and Piping Components
Allowances. C I for shortfall in the wall thickness resulting from the manufacturing process; to be determined in each individual case; c" wear and corrosion allowance (at least 1 mm - not required if adequate protection against the effects of the feed medium exists, when Se "" 30 mm or for heat exchanger tubes).
2.2 Cylindrical Shells and Tubes Under Internal Pressure D.lD, < 1.21. The required wall thickness is given by D,p S
= 2v(K/S)
(la)
+ P + c i + c"
where D, and D, are outer and inner diameters, K is the pennitted material stress, p the calculating pressure, S the safety factor and C I and c, the thickness allowances for the wall. The reduction correction factor is v = 1.0 for seamless welded shells and 0.8 to 1.0 for welded joints in the shell, depending on the rating of the welded seam in accordance with the AD Code of Practice HPO.
1.21 < Da/D,
:5
1.S. The wall thickness is given by D,p
S
= 2.3(K/S) _ P
+ C I + c, .
(lb)
Tubes D.lD, < 1.7. The tube wall thickness is determined by internal pressure, ease of handling during transport and installation, sag between supports, possibilities of external damage (mechanical, corrosion), type of tube joint, working load and restrained thennal expansion. For D, < 200 mm, wall thickness under internal and external pressure is calculated in accordance with the AD Code of Practice Bl and Eq. (la). The calculation for steel tubes with regard to internal pressure distinguishes three regions as DIN 2413: I, mainly static stress up to 120°C; II, mainly static stress above 120°C; and III, dynamic stress. For region I, the required wall thickness is given by
D,p s= 2v (K/S) +C, +c,.
(2a)
The safety factor S depends on the elongation at rupture. For /), = 20%, S = 1.6 with an acceptance certificate in accordance with DIN 50 049, and S= 1.75 without. In the case of seamless tubes with special-quality specifications, the safety rating of the welding seam v lies between 0.5 and 1.0, depending on quality. For region II, the calculation is carried out in accordance with S=
D,p (2K/S-P)v+2p +c l +C2 ·
(2b)
The safety factor S with respect to the thermal yield point (tensile) = 1.5 (in accordance with DIN 50 049), otherwise S = 1.7.
Thermal Stresses D.lD, < 1.7. The change in length III caused by the temperature differences it - ito between operating and installation temperature is given by
=E· a(it- ito)
- '!. ~ it _ it 3D, + D, '; - 2 1 - v ( , ,) 2(D, + D,) ,
0'
0",
D,+3D, a E = 2"l="V (it, - itl ) 2(D, + D,}
(4)
From these, the maximum internal and external stationary stresses can be calculated approximately:
P(D, +Se)
UV'i=~+Uitl'
(5)
Here, se is the actual wall thickness, v the lateral contraction coefficient, and it the temperatures. These approximate fonnulas are sufficiently accurate in practice as long as the greater of the two comparative stresses O'v, I and O'v., is always considered, i.e. as long as the following holds true:
(6) Uito
~
P(D, - 3se ) -
4· Se
.
All equations apply to undamped cylinders without additional external or bearing forces. Under any shortfall of the conditions expressed by Eq. (6), or in the presence of additional axial stresses, the comparative stresses must be calculated from the previously sununed main stresses. Stationary thermal stresses may exceed K/ S, provided they occur alone (O'v, _ :5 2K/ S). According to Eq. (5), superimposed stresses arising from pressure and temperature differences will lead to large peak stresses at the internal axis if the gradients are opposed (PI> p" it, < it,) (disadvantageous!), but to more unifonn stress distributions if the gradients have the same sense (test whether under certain conditions O'v., > O'v.,). According to Eq. (4), thermal stresses increase with increasing wall thickness at constant temperature difference it, - it,. At a given quantity of heat Q and length of tube 10 , the temperature difference must also increase with increasing wall thickness, owing to the increasing resistance of thermal conduction:
it, - it, = 2 Q l ' In '1Tol\
~ . D
(7)
j
The thennal stresses increase in logarithmic fonn while the pressure stresses decrease. The summed comparative stresses fonn pronounced maxima, which are displaced in the direction of lower wall thicknesses as the thennal stresses increase.
2.3 Cylindrical Shells Under External Pressure
If the change in length is restrained, an axial stress 0'.
elasticity. If pressure forces exist, attention must be given to buckling of the tube. If temperature differences occur in the wall owing to heating or cooling, tangential and axial stresses will occur at the inner and outer axes (subscripts i and a respectively), which are of equal magnitude in both cases, positive at lower temperatures and negative at higher ones:
(3b)
will occur; 10 is the installation length, a the coefficient of thermal expansion (see C3.1.2) and E the modulus of
For cylindrical shells under external pressure with D, "" 200 mm, calculations have to be carried out with
respect to denting and plastic defonnation if D,/D, :5 1.2.
2.5 Domed End Closures
Table 2. Safety correction factor against elastic buckling in the presence of external pressure [1] 0.1
0.01
0.005
0.003
0.001
3.0
3.5
3.7
4.0
5.5
In fixed-tube apparatus (see H3), the most frequent form of application, the tube plates are mutually reinforced by tubes that have been welded or rolled in.
Wall Thickness. For flat end closures and tube plates this is calculated from s= CD
Elastic Indentation. If the length of the shell is L then the maximum permissible operating pressure (notation: see Eq. Ia) is given by _~[ 2.0 se - c 1 - c, P - S (n' - I)A 2 D,
+ 0.733
2)5] . (n2 _1+ n'A-2 - 1.3) (s< - D, c C 1 -
n
Plastic Deformation. If the length of the shell is L then the highest permissible operating pressure for DJ L :5 5 (notation: see Eq. Ia) is given by CI -
C,/
P = 2.0 5 -~D~,-
[1+ 0.OI5u (I - 0.2~) _~D~,_] . L
(9)
Sc - C1 - C2
The usual value u of the circularity error is 1.5. For D,/ L > 5 the maximum permissible operating pressure is the lesser of the two following values:
P=2.0~SC-Cl~C2 S
D,
and P = 3.0 K ~ S
Kv
(II)
(notation: see Eq. la). If t is a tube division the reduction correction for plates with re-entrant tubes (V-tubes) is given by v
v=(t-d,)/t v = (t - 0.833d,)/t
(8)
n
K s. -
pS
-+c,
= (t-d,)/t,
and for plates with fully supporting tubes (ftxed plate, floating head) by
where S = 3 and A = 1 + (nD, / 2L)'. It is required that 2: 2 and 2: D,/2L, where is the number of indentation waves that can occur along the circumference on failure (Table 2).
n
J
(s -c-c)' _<_ _1_ _ '
L
(10)
with the safety factor S = 1.6 for rolled and forged steels.
2.4 Flat End Closures and Tube Plates Flat plates are always used if the pressures or pressure differences are small, or when it is necessary to have a flat separating face. This is the case in the majority of tube bundle apparatuses and for the covers and closures of high-pressure vessels. Where no flatness requirement exists, it should be determined whether the separating or closure function can be fulfIlled by domed components. These permit a more economic use of material. Tube plates are used in the range of sizes from 100 to 4500 mm. As flat and hole-free closures of very large vessels or apparatus, they are also used up to 8000 mm. The thickness of tube plates varies from a lower limit (membrane closure) of a few millimetres to an upper limit of 650 mm for steam generation in nuclear power stations. Tube divisions lie between 1.2d and 1.5d and the number of holes between the limits of 10 to 10' (the latter in steam generators of nuclear power stations). Besides the circular plate (most frequently used), rectangular or elliptic plates, annular plates or flat closures with flanged rims are also employed. Flat-walled components can, in principle, be implemented unreinforced or reinforced by profIling or by tie-rods. The plate thickness may vary in the radial direction.
for for
d,/d;:5 1.2 and d,/d l > 1.2,
where d, is the external diameter of the hole or tube and d the internal diameter. j
Calculation Correction and Diameter. The values of these, C and D, depend on the type of end closure, its connection to the casing, and the arrangement of the tubes. If no additional edge moment is present, then for full flat plates and uniformly perforated plates with reentrant tubes (hairpin tubes) C depends on the plate support. C lies between 0.32 and 0.35 for plates with ftxed support and between 0.40 and 0.45 for plates with loose support. For welded-in plates see [I]. If, in the case of fIXed-plate apparatus, the supporting action of the tubes is to be taken into account, the permissible buckling force for the tubes must be observed. If an edge moment acting in the same sense is present, some of the correction values will increase considerably. In the case of rolled-in or welded-in tubes, the tensile force F exerted on an individual tube must be transmitted to the tube plate. The formula 1OF/(awd,) 2: Sw 2: F/(aw (d, - d,)) applies to the standard rolled length sw, where 12 mm :5 Sw :5 40 mm. The permissible stress awof the rolled joint is 150 N/mm2 if the rolled joint is smooth, 300 N/mm2 if it is grooved, and 400 N mm 2 if it is flanged. In the case of welded-in tubes, the thickness of the welding seam g at the shear cross-section must be at least g = O.4F!(d,K/S).
2.5 Domed End Closures The shapes of domed end closures lie between the limiting cases of flat and hemispherical closures. In Central Europe, torispheric closures predominate, consisting of a spherical cap (radius R) and a rim (radius r) (Fig. 1). Well-known constructions are the dished end (R = D" r = O.ID,) and the three-centre arch end closure (R = 0.8D" r = 0.154D,). In general, R :5 D" r = 2: O.ID, or r 2:30s applies, where s is the required wall thickness of the domed closure. The height of the rim hi should be not less than 3.5s for dished and 3.0s for three-centre arch
Figure 1. Domed end closure.
Components of Thermal Apparatus. 2 Apparatus and Piping Components
closures. For closures that consist of a flange part and a spherical cap part that have been welded together, a minimum distance x must be maintained between the welding seam and the flange. For dished end closure, x ~ 3.5s and for three-centre arch closures x ~ 3. Os. but must be at least 100 mm. In the Anglophone countries, the ellipsoidal shape predominates, usually with a 2 : I ratio of the axes. In all cases, dome end closures allow better utilisation of material than flat closures. In comparison with hemispherical closures, they offer the advantage of smaller height and, frequently, better access. The dimensions lie between the limits of 50 and 12 000 mm. Connection between end closures and neighbouring components should, if possible, be implemented in the form of butt joints. Transitions in cross-section must be implemented in conical form. The rules for calculation apply to domed end closures with a cap radius R <; D., a rim knuckle radius r'" O.ID. and a wall thickness Se '" O.OOID, (se '" 2 mm). In the presence of external pressure, the safety factor must be increased by 20%.
Required Wall Thickness. With the calculation correction factor f3 and other factors, as in Eq. (la), the required wall thickness is given by
D,pf3
(12)
S ~ 4vKIS + c 1 + c,.
Calculation (SC -
Factors. With x d;/ D., the following applies to
Correction
c1 - c,) I D. and y
~
domed end closures: Dished. f3 ~ max (1.9 + 0.03251x 07 + x; 1.9 + 0.933yl'I'X), Tbree-Centre Arch f3 ~ max (1.55 + 0.0255Ix o62S; 1.55 + 0.866YI'I'X), where 0.001 <; x <; 0.1 and 0 <; y <; 0.6. The formulas apply to cutouts with diameter d, in the region of the flange and outside the crown region of 0.6Do of the cap.
Buckling. In the presence of internal pressure, the end closures are dimensioned adequately against buckling in the flange region if the following apply: Dished end closures, p
<;
41.6E[(sc -
Tbree-centre arch,
<;
33.3E[se -
p
C1 -
C1 -
c,)ID.l'14 c,)ID.l' 34
(13)
where E is the modulus of elasticity (see BI.2). In the presence of external pressure, additional safety against elastic indentation must be provided:
p
<;
0366
~ (Se -
Sk
C1- c,)2 .
(14)
R
Cylinders
vA~2/(2+x)
+ 2.52y l66/(10+x);
Spherical shells VA ~ 2/(2 + x) + 1. 28y P5 I( 4.5 + x);
with 0
<;
x
<;
8 and 0
<; y <;
2.
Width and Thickness. 'Ine width of a disc-shaped reinforcement must be at least b '" .J(D, + SA - c 1 - C,)
(sA - C1 -
C,)
and The thickness of the reinforcement must not exceed the actual thickness Se of the end closure. The length I, of a tube-shaped reinforcement must be at least 16~ 1.25 .JD,+sS-C I
-C 2 )
(ss-c 1
-c2 ) ·
The ratio of the wall thicknesses should be (Ss - c 1 - c,)/(sA -
C1 -
c2 )
'"
2.
Mutual influence of two cutouts can be neglected if the distance between their closest pOints I '" 2 .J(D, + SA - c 1 - C,) (sA - C 1 - C2 ).
2.7 Flange Joints 2.7.1 Bolts At operating temperatures above 300°C or operating pressures exceeding 40 bar, tension bolts must be used. Here, only those bolts will be rated as tension bolts whose shank diameter d, <; 0.9dK or whose dimensions comply with DIN 2510. Bolts with a full-length thread are rated as stud bolts. If pOSSible, bolts smaller than MIO should not be used. The number of bolts should be as large as possible (ratio of interval between bolts to bolt hole diameter, tid!. <; 5).
Stress Conditions. As seen in Fig. 2, the following forces are applied to a flange: FR , longitudinal force; Fp, force on annular cross~section due to internal pressure; FD , jointing force; Fs force arising from bolting. The flange must withstand these forces. Forces that may occur due to a bending moment in connected piping systems are, as a rule, not taken into account. In compliance with AD Code of Practice B7 and also if test pressures exceed 1.3PR, the forces at the bolts must be evaluated in accordance with the operating and installation conditions before pressure is applied. At the permissible material stress K and the safety cor-
The safety correction factor Sk can be obtained from Table 2.
2.6 Cutouts The equations apply to vessels with internal pressure and for DoD, <; 1.2, with a distinction between disc-shaped and tube-shaped reinforcements. Within limits, these are mutually interchangeable, or they may be used simultaneously.
Reduction Correction Factor. For a wall thickness Ss of the connection tube of diameter d" required wall thickness SA at the edge of the cutout and with x ~ d, I .JV, + SA -
y
~
(ss -
C 1 - C,)
C1
-
(SA -
C2 ) C1 -
(SA c2)
C1 -
c,)
the reduction correction factor is given as follows:
} (15)
and Figure 2. Forces applied to a fixed flange.
2.7 Flange Joints. 2.7.2 Flanges
Table 3. Safety correction factor S for bolted connections [11 Installation and test condition
1.5
l.l
1.8
1.3
50
3.0
For tension bolts
Materials of known UTS and safety in proportion to UTS or CTHIlO()()(]()
e.g. to DIN 25 \0
For stud bolts e.g. to DIN 2509 or DIN 931 Materials of unknown UTS with safety
in proportion to
Operating condition
flat metallic gaskets, k I ~ b D + 5 mm and for other types of metallic gasket 5 mm ,; k I , ; 6 mm. For the effective width under installation conditions, gasket O.8bo ,; ku ,; b D applies to flat metallic gaskets, and 0.16,; ku/kl ,; 0.33 to other metallic gaskets depending on shape (see Appendix H2, Table 1).
urs
2.7.2 Flanges The bolting force to be taken up by the flange (see Fig. 2) is Fs :0- F~o during initial compression and Fs :0FD + FR + Fp during pressure testing and operation. Weakening of the flange by the bolt hole is accounted for in the calculation hy a calculating diameter d; Creduced bolt-hole e1iameter') (Fig. 4).
rection factor 5 (Table ~), the thread diameter d k of a stud bolt and the shank diameter d, of a tension bolt in a connection employing n bolts are both given by: d,
and
dk~
J
Loose Flange (Fig. ~a). The resistance moment with b ~ d, - d, - 2d~ and with the reduced bolt-hole diameter d{ (Fig. 4) must be at least W ~ O.7874h~b
451',
- - ' +c 'ITcpKn
:0-
5 Fs K (d, - d 4 )/2 .
(17)
(16)
For a support surface created by a metal-removing process or an equivalent surface, the quality rdting factor cp can be put at 1.0, otherwise at 0.75. The design allowance c, for the operating condition, must be c ~ 3 mm up to M24 and c ~ I mm from M52 or corresponding diameter onwards. Linear interpolation must be used in the intermediate region, where c ~ 0 mm for tension bolts. The temperature used for calculation is around 30°C below the highest temperature of the working medium for connections of loose flange to loose flange, around 25 °C for fIxed flange to loose flange and around IS °C for fIxed flange to fIxed flange.
Jointing Forces. These depend on the shape, the jointing material and the operating conditions (pressure and temperature). If ko and k I are the effective width of the gasket during installation and operation respectively, dl) the mean diameter of the gasket, p the pressure to be sealed off and K o ." the deformation stress of the gasket (see Table 4), then the required sealing force is P D = 7rdoklPSf) with Su ~ 1.2 for operating conditions and I ,; 50 ,; 1.2 for test conditions. The required initial compression force is Ibo = Trd D koKD_ 20. For gaskets of soft
Welded Flange (Fig. ~b). Here,
applies, with b ~ d, - d 1 - 2d{.; a ~ 0.5(d, - d, - SI) fortest and operating conditions and a ~ O.5(d, - d) for installation conditions.
- - - do -d;
a
materials with pD ,; lObar· m, F/'x, ~ 0.2F~o +
n.s '>I(0.25'ITd'p + FD)F' 1)0 ,
applies if F~ < F,;o- In the case of metal gaskets Fm= ~ 'ITdokokD. ".
Width of Gasket. Under operating conditions for liquids, this is O.Sb D , ; kl ,; I.lbu if the gaskets are of soft material or soft metal material and 0.5b D , ; kl ,; 1.8bo for gases and vapours, depending on shape and material. For Table 4. Deformation stress
KlJ{j
of metallic sealing materials III
- - - do - - -~
Sealing material
Aluminium, soft Copper Soft iron Steel St 35
13 CrMo 44 Austenitic steel
20
100
200
300
100 200 350 400 450
40 180 :110 :180 450 480
20 130 260 330 420 450
(5) 100 210 260
5(~)
390 420
400
(40) 170 190 330 390
(80) ( 120) 280 350
Figure~.
Flange types. a Loose backing flange. b Welded flange.
c Welded collar. d Weldneck flange.
I:
I'..
Components of Thermal Apparatus. 2 Apparatus and Piping Components
100 0.75
Pressure drops are made up from losses in straight lengths of pipe and losses occurring in bends and fittings (individual resistances); see A6.2 and [I]. For pressuredrops in steel pipe See Appendix H2, Fig. 1, and for fittings see H2.9.1 and Appendix H2, Fig. 2.
d;=v·d l
"r-.,
0.50
1"\
2.8.~
I
500
Types, Standards, Materials
General dj In mm
Figure 4. Reduced diameter of bolt hole.
Welded Collar (Fig. ~c). Equation (18) applies where do is used in place of d, and de = 0 is entered.
WeldAeck nange (Fig. ~d). Here the section moduli Wfor cross-section A-A and B-B must be verified by calculation. D,:S lO00mm Section A-A: W= 0.7874(h~h + (d, + SF)S~ 2: F,(S/K)a, Section B-B: W= 0.7874(h~h/B2 + 0.75(d, + s,)S;) 2: Fs (5/ K)a, b+(SF-S,)B, . where B = b + (SF + s,)B,(B, + 2) WIth B, = (hA - hF)/hp , b = do - d, - 2d~ and a = 0.5(d, - d, - s,) under pressure testing and operating conditions and a = 0.5(d, - do) under installation conditions.
1000 mm :S D, :S 3600 mm Section A-A: W= 0.943(h~b + (d, + SF)SF(O.8sF + O.lhF) 2: F,(s/ K)a, Section B-B W= O.943(h~b/B2 + I.S(d, + s,)sf)"'= Fs(S/K)a
with B, b and a as for D, :S 1000 mm. However, hA - hF must be 2: 0.6 hF and SF - s, must be 2: 0.25hF'
2.8 Piping 2.8.1 Pipe Diameter The internal pipe diameter d is calculated from the continuity equation with volume flow jr and pipe cross-section corresponding to the chosen flow velocity v as d = v4V/(1rv). If jr is given, v must be chosen so that the operating and piping system costs are low and d corresponds to standard values. Large v implies a small pipe diameter, small fittings, low costs of insulation and painting, but with, on the other hand, high pressure-drops (greater costs of pumps, higher operating costs) and higher noise levels. The economic diameter is that leading to the lowest sum of plant and operating costs, taking into account the efficiency of utilisation (= operating time/ (operating time + down time». Standard velocity values are to be found in [6, 7] and in Appendix H2, Table 2. 2.8.2 now Losses Incompressible fluids cause pressure drops; compressible fluids (gases) cause pressure drops, volume expansion and accelerations. Heat exchange with the surroundings depends on insulation.
Important standards and regulations for piping systems: DIN 2400 Piping systems; summary of standards for planning, design and materials. DIN 2401 Tl (Part I), Components subject to internal and external pressure stress; pressure and temperature data, concepts, standard pressure ranges. T2 (Part 2), Piping systems; pressure ranges, permissible operating pressures for piping components of ferrous materials. DIN 2402 Piping systems; nominal sizes; grading. DIN 2406 Piping systems; letter symbols; pipe categories. DIN 2408 Tl (Part I), Piping systems for process plant; planning and design data. DIN 2413 Steel pipes; calculation of wall thickness with respect to internal pressure. DIN 4279 Tl (Part I), Tests for internal pressure of pressure piping systems for water; general data. T2 to TIO (Parts 2 to 10), Tests for internal pressure of pressure piping systems for water, various materials. ISO 4200 Seamless and welded pipe; summary of dimensions. Pressure Vessels V, Regulations relating to pressure vessels, May 1989. VdTiiV Code of practice sheets relating to various test procedures for piping systems. Maxmilian-Veriag, Herford. DVGW Worksheets for the construction of gas and water pipeline systems. ZfGW-Veriag Frankfurt a.M.
Nominal Pressure PN. This is the pressure in standard piping components based on a given starting material mentioned in the relevant standards and designed for a temperature of 20°C. It corresponds to the permissible operating pressure at the lowest operating temperature of a material. Staging of the nominal pressures PN in bars (units not given): 1.6
2.5
4
6
16 20 40 50 10 12.5 25 32 60 80 160 200 250 320 400500 600 BOO 100 125 1000 2500 4000 1600 6000
Nominal Size DN. This is the reference parameter (characteristic indicator) for parts fitting each other, e.g. pipes with formed parts, or fittings. The nominal size is stated without units; it corresponds approximately to the internal size in mm. Summary of Types of Pipe (DIN 2410). General data relating to welded pipe of unalloyed or low-alloy steels
DIN 1626: Commercial grade: For general requirements of piping systems and vessels as well as of equipment manufacture. Up to 120°C: For liquids up to 25 bar, for air and non-hazardous gases up to an operating pressure of 10 bar; up to 180 °C: For saturated steam up to 10 bar. Materials: St 33, St 37, St 42. With Regulations for commercial grades: For more demanding applications, suitable for bending, flanging and similar; up to 120°C: up to 64 bar, over 120 and up to 300 °C, also up to an operating pressure of 64 bar if wall temperature in °C multiplied by pressure in bar :S 7200; with special acceptance without specified limit. Specially tested pipe with quality specifications:
2.8 Piping. 2.8.4 Pipe Fittings
For especially demanding requirements; up to )00°(; without specified limitation of the operating pressure. General data regarding seamless tube of non-alloyed or low-alloy steel DIN 1629: Applications and materials similar to DIN 1626.
Precision Steel Tube. Seamless (DIN 2391, for all pressures, 4 to 120 mm external diameter), welded (DIN 2393, for all pressures, 4 to 120 mm external diameter), welded and single cold-drawn (DIN 2394, up to PN 100. 6 to 120 mm external diameter) for high-precision applications, especially surface conditions, low wall thicknesses. Designation and material: tube 30 x 2 DIN 2391 St 35 (or St 45 or St 5';) hright drawn, soft. hard, soft annealed, etc.
Threaded Tube. This is seamless or welded, mediumheavy (DIN 2440) and heavy (DIN 2441) of St 33-1 or St 33-2. Seamless Steel Tube (DIN 2445, 2448, 2449. 2450, 2451. 2456 and 2457). This is made of different steels St 00 to 5t 52 (corresponds to DIN 1(29) with 10.2 to 558.5 mm external diameter. At the same external diameter lower wall thicknesses than DIN 2240, e.g. f(,r outside diameter = 60.3 mm, according to DIN 2448, wall thickness = 2.9 mm standard (however, a large choice is possihle), as compared with wall thickness = 3.6'; mm according to DIN 2440. It is availahle up to PN 100. hence suitable for the most diverse purposes in mechanical engineering and equipment manufacture.
Welded Steel Tube (DIN 2458). 'Ibis is made of St 3) to St 52-3 steels for all nominal pressures with 10.2 to 1016 mm outside diameter and even smaller wall thicknesses than DIN 2448. e.g. at do = 60.3 mm. S = 2.3 mm standard (however, an equally large choice as with DIN 2448, hence extended field of application).
1:111
28516 (DN 40 to DN 1200) and TYTON sleeves DIN 28516 (DN 50 to DN 6(0).
Pressure Pipe of Ductile Iron (DIN 28(10) with threaded sockets (water up to PN 40, DN 80 to DN (00), stuffing box sleeves (water up to PN 25, DN 500 to DN 1200), or TYTON sleeves (water up to PN 40, DN 80 to DN 6(0), for gas up to PN 1.
Other Pipe Materials
Copper. DIN 1754 for external diameter 3 mm (wall thickness 1 mm max.) up to 419 mm (wall thickness 4 mm max.); material: copper DIN 17671 with strength specification F20 (
Polyvinylchloride
(PVC) Hard (unplasticised PVC (uPVC» for drainage installations, vent ducting, water and gas piping systems. For general quality specifications see DIN 8061; for dimensions see DIN 8062: External diameter 5 mm (wall thickness 1 mm max.) up to 1000 mm (wall thickness 29.2 mm max.). For standards relating to chemical stability see DIN 16929. Other Plastics [8]. DIN 8072 soft polyethylene tube. DIN 8074 high-density polyethylene tube. DIN 8077 polypropylene tuhe. DIN 16868 and DIN 16869 Tl (Part 1), glass-fibrereinforced polyester resin tube. DIN 16870 and DIN 16871 Tl (Part 1), glass-fibrereinf(lrced epoxy resin tuhe.
Steel Tube for Gas and Water Piping Systems. This is seamless (DIN 2460) and made of various steels: St 00 for gas up to PN 1 and water up to PN 25, St 35 for gas up to PN 100 and water up to PN 64; 88.9 up to 508 mm external diameter. Welded tuhe (DIN 24(1) is of St 33 for gas up to PN 1 and water up to PN 20, St )7-2 for gas up to PN 80 and water up to PN 64; 88.9 to 2020 mm external diameter. Surface protection: external protection is by hituminous materials with glass-fibre matting and whitewashed; internal protection is by bitumen coating, linseed oil, cement mortar or other materials forming protective coatings. Applications are gas Of water piping systems external to buildings in or above the ground.
Steel Tube for Long-Distance Pipelines. This is employed for flammable liquids and gases (DIN 17 172) ftlr all pressures from 100 mm external diameter upwards.
2.8.4 Pipe Fittings For Steel Pipe /'langed Connections (Fig. 3
and Fig. S). These are preferred for high-pressure lines and becanse they are easily dismantled. DIN 2500 gives a summary for steel and cast iron. For mating dimensions, see DIN 2501.
Standards J'Jr Flange Shapes Fig. Sa and b: DIN 2558 (PN 6; DN 6 to 100), DIN 2561 (10, 16; 6 to 40), DIN 2566 (lO, 16; 6 to 100).
Fig. Sc: Cast-Iron Pipe
Pressure Pipe. For water up to PN 16 threaded sockets DIN 28 511 (DN 40 to DN 6(0), stuffing box sleeves DIN 28 512 (DN 500 to DN 1200), lead-joint sockets DIN 28513 (DN 40 to DN 1200), flanges DIN 28514 and
DIN 2573 (6; 10 to 5(0), DIN 2576 (10; 10 to 5(0).
Fig. Sd:
GG: DIN 2530 (1; 10 to 4(00), DIN 2531 (6; 10 to 36(0),
Figure 5. Flange forms. a Screw-on flange, oval, smooth. b Screw-on flange with boss, round. c Smooth-face flange for soldering or welding. d Flange of GGL, GS or G(~G. e Weldneck flang{". f Loose backing flange. g Blind flange.
I:
I""
DIN DIN DIN DIN DIN DIN DIN DIN DIN DIN DIN DIN DIN DIN DIN
2532 2533 2534 2535
Components of Thermal Apparatus. 2 Apparatus aud Piping Components
(10; (16; (25; (40;
10 10 10 10
to 3000), to 1000), to 500), to 400); GGG: DIN 28 604 (10; 40 to 1200), 28 605 (16; 40 to 1200), 28 606 (25; 40 to 600), 28 607 (40; 40 to 400); GS: DIN 2543 (16; 10 to 2200), 2544 (25; 10 to 2000), 2545 (40; 10 to 1600), 2546 (64; 10 to 1200), 2547 (100; 10 to 700), 2548 (160; 10 to 300), 2549 (250; 10 to 300), 2550 (320; 10 to 250), 2551 (400; 10 to 200).
Figure 6. Screwed connectors for pipes. a 1 Steel pipe. 2 Union nut. 3 Female seating cone. 4 Male seating cone. b DIN 3930 mechanical compression fitting.
Fig. Se: DIN DIN DIN DIN DIN DIN DIN DIN DIN DIN DIN DIN
2630 2631 2632 2633 2634 2635 2636 2637 2638 2628 2629 2627
(1; 10 to 4000), (6; 10 to 3600), (10; 10 to 3000), (16; 10 to 2000), (25; 10 to 1000), (40; 10 to 500), (64; 10 to 400), (100; 10 to 350), (160; 10 to 300), (250; 10 to 300), (320; 10 to 250), (400; 10 to 200).
Fig. Sf: DIN 2641 (6; 10 to 1200), DIN 2642 (10; 10 to 800), DIN 2655 (16; 10 to 500), DIN 2656 (25; 10 to 400).
For Welded-on or Formed Collar. DIN DIN DIN DIN DIN DIN DIN
2673 2674 2675 2676 2667 2668 2669
(10; 10 to 1200), (16; 10 to 500), (25; 10 to 500), (40; 10 to 500), (160; 100 to 250), (250; 100 to 250), (320; 100 to 250).
Fig. Sg: DIN 2527 (6 to 100; 10 to 500).
cross-section of every type are also made from tubular material. Modern piping systems mostly feature only flanged or screwed connections for their fittings. If welding is not carried out carefully in pipes of small nominal sizes (approximately below ON 50), reduction of crosssection could occur and hence increase in flow resistance.
Processes. Gas Welding (for unalloyed or low-alloy steels with wall thicknesses of up to about 3 mm); arc welding (for wall thicknesses above 3 mm); inert-gas welding or submerged-arc welding (for automatic welding of large piping systems); see DIN 8564: welding of piping systems; steel piping systems, fabrication, testing of welded seams. Additional standards, guidelines and Regulations must be observed [8]. DIN 2559 T1 (Part 1), weld preparation for the seam, guidelines for end preparation forms. DIN 8558 T1 (Part 1), guidelines for welded connections to steam boilers, vessels and piping systems; examples
of implementation. DIN 8560 certification of welders. DIN 8563 T1 (Part l), quality control of welding work; T3 (Part 3) fusion-welded connections for steel.
For Cast-Iron Pipe
Socket Connections (Fig. 7). These are preferred for cast iron (GG) aud cast iron globular (GGG). Direction of flow from the socket end to the plain end of the pipe. Advantage: quick installation; disadvantages: exact pipe installation length required, sensitive to longitudinal forces.
Screwed Connections. Steel fittings for the chemical and shipbuilding industries: see DIN 2980 and 2982. Detachable connections for connecting to apparatus in frequent need of repair or possible modifications using flat gaskets (Klingerite gaskets) or conical sealing surfaces metal-tometal contact, Fig. 6a. See also DIN 2353 and DIN 3930: Solderless screw connections with olives; Fig. 6b. Advantage of this screw connection: high-pressure loading (up to DN 630, easy installation, small space requirement, suitable for various tube qualities.
Welded Connections. Welded tubes have the advantages of unchanged fluid-tightness (hence the testing of the welded seam for fluid-tightness by X-rays or ultrasound in the case of importaut long-distance pipelines) and - in contrast to flanged connections - of low heat loss. Branch components and those providing change of direction or
~
~ )~ a
b
c Figure 7. Socket connections. a Stuffing box. 1 Stuffing box ring foUower. 2 Sealing ring. 3 T-bolt with nut. 4 Stuffing box socket. b Plug socket. c Screwed socket. 1 Screwed ring follower. 2 Sealing ring. 3 Screwed socket.
2.8 Piping. 2.8.5 Expansion Compensators
1:,11
1.6 .-----,-----r--.-'T-c~
~ 141--t-+-l--->'f--1
g12
Figure 8. Connections for plastics pipe.
1!l
810
Figure'. Screwed connections for pvc
~
~ 08
pipe. I Screwed sleeve.
~ ~ 06
2 Union nut of uPVC or
annealed cast iron (GlW) or Cu-Zn alloy. 3 Flat ring seal. 4 Nipple sleeve, glued on.
0.4 LLL"--'---...L...--'----,-' Temperature in 'C Figure 11. Correction factor for temperature conversion.
Fig. ~), between two fIXed points. From the constructional point of view, they may be classified as follows: Compensation by Piping Layout (without additional fittings; Fig. 10). Anchors should be at fittings where possible. If large temperature differences are present, piping must be installed prestressed to counteract thermal expansion (e.g. for compressive stresses when piping heats up, install under tension - cold-set). Customary prestress equals 50% of expected stress [9]. Length I of piping leg for steel with pipe outside diameter D and change of length M is 1= 0.0065 -.fiJFJ; for copper 1= 0.0032 -.fiJFJ; for calculation see [11]. Expansion
Figure 10. Simple expansion compensators. a Pipe leg. b Z-bend. c V-bend.
Approximate Calculation of the Anchor Forces. This is carried out with numerical equations for St 35 at a temperature of 400°C with 50% prestressing and a radius of curvature R = 5d. For conversion to other temperatures and materials, see Fig. 11.
For Copper Pipe Flanged and Screwed Connections. These are similar to those for steel pipe, but with different pressure ranges (strength).
Return Bends. Fu = 10IM/(rG) in N, where M is the total expansion between the anchors in cm, I is the axial geometrical moment of inertia of the pipe in cm4 and C is the correction factor as in Fig. l::1a.
Welded Connections. These are very extensively used in equipment manufacture.
Piping Legs. Fx
For PVC and Other Plastics Pipe
=
Flanged Connections. See DIN 8063 for large diameters with loose backing flanges (mostly metal; Fig. 8). Screwed Connections (Fig. 9). See DIN 8063.
Expansion Compensation Through Special Componeuts [10].
Welded and Adbeslve Connections. For processes, see DIN 19 533. PVC is mostly hot-air welded using filler rod; polyethylene is welded by fusion jointing. PVC can be glued using preformed or glued-on adhesive sleeves (similar to soldering sleeves). Adhesives are usually of the soluble rype (tetrahydrofuran). Polyethylene cannot be glued.
Expansion Loops (Fig. l~a) are, like return bends, very safe in operation, requiring no maintenance. They do, however, take up a great deal of space. They are suitable for outside use. Implementation may be effected as plain, corrugated or folded pipes. They should be arranged, if possible, so that the apex of the loop itself does not move, but is restrained by a sliding support. The anchor force is as for return bends.
2.8.5 Expansion Compensators Expansion compensators take up thermally induced changes in length (see Eq. (3) and Appendix H2,
~u co
<.:>
il .l!l
2.0
E
~ 0
1.5
~
.ti
1.0
~
30
.l!l
§
0.5
1.0
0
u
100
200
300
Diameter in mm
400
\
b
,/.,
I
----Z-bend
~"
\
\ \ ,,
;;
"'1.0
500
b},
I
15 -----1 _--Pipeleg
.~ 0.5
0
a
Prestress 50%
St 34 ~20 r--- _ Steel HOO'C
co
:§
2.5
= blI!I', Fy =b 21/12 in N. =
Z-Bends. Fx b,j/12, Fy bJ/12 in N. I in cm4 applies to both, I = Ix + Iy in m is the total length of the piping legs. Correction values b l to b 4 are as in Fig.l::1b.
-~
-;/
~
--- --
2.0
Ratio 1,//,
/
b,
b,._ i-bJ _I: -~ 3..0
4.0
Figure Ill. Correction factor for the calculation of axial pipe forces. a V-bend. b Z-bend and pipe leg.
1:11.
Components of Thermal Apparatus. 2 Apparatus and Piping Components
Figure 13. Expansion compensators. a Expansion loop. b Axial compensator with internal pipe (bellows compensator). c Telescopic compensator. d Rubber compensator.
Bellows Compensators are expansion compensators, require no maintenance and take up the least space_ Lenticular compensators with few but high corrugations are usual for very large diameters (about DN 5000). Single- or multi-layer bellows (Fig. l~b) with many low corrugations, made up of single or multi-layer cold-worked sheet steel with large expansion capabilities, are used for high pressures (DN 600: PN 100; DN 250: PN 250). Rubber Compensators (Fig.
of various kinds are used for DN 40 to DN 400 and for temperatures of 100 °C at PN 10.
moments. The force acting on a fixed point is usually the resultant of forces acting in different directions.
Guides fulfilling the function of guide points supplement the anchors and allow axiaI movement and in part also torsional movement; see Fig. 14 [10].
2.9 Shutoff and Control Valves
l~d)
Telescopic Compensators (Fig. l~c) are prefabricated. The inner tube is smoothed and sometimes hard-chromed so as to reduce frictional resistance. Packing Materials: Permanently elastic Buna seals require no maintenance and can be used for almost all media. Plastic seals (adipose hemp for water, lead lamellae-asbestos for gas) must be after-sealed. Articulatory Compensators, apart from axial expansion, also take up lateral deformation. Attention must be paid to axial forces during installation!
;1.8.6 Pipe Supports With reference to the pipe and surroundings (e.g. buildings), the purpose of pipe supports is to fix free pip-
Function Valves (pipe closers) in piping systems serve as follows:
Shutoff elements prevent the flow of a fluid. They must close tightly against the fluid flow and, in order to avoid percussive stresses, in such a manner that the velocity is not reduced to zero suddenly (exception: quick-action gate valves).
Control Elements (globe valves) are intended to affect the volume flow as a function of parameter to be controlled.
Safety Elements open a cross-section for pressure relief if pressure becomes excessive.
ing systems to make them operationally secure.
Types of Construction (Sununary)
Hangers are intended to carry the pipes. The drop must
Fittings (DIN 3211) used are as follows:
be adjusted accurately and a certain movement allowed. Designs go as far as 'constant hangers' where the forces caused by expansion are kept constant by means of preset springs and articulated levers.
Valves. A shutoff element (plate, cone, piston, ball) is lifted in the direction of flow to open a cylindrical annular cross-section as a free-flow cross-section (Fig. ISa). Valve-like shutoff elements, where, because of particularly favourable flow conditions or particular corrosiveness of
Supports fulfil the same function as hangers, with the difference that the forces are restrained from below (Fig. 14).
0::::. -:·::0
Anchors
serve to fix the direction of expansion unequivocally; they are capable of resisting forces and
1tl?~ --
+
-
7,////$'7$$/'?
"
b
~b
•
&tW$~'wAY,0 ~
Z
(Sto ~
~ ~ d
'I
e
Fi......, 14. Pipe supports. a Pipe trolley. b Roller support. c U-bolt. d Cylinder bearer. e Dome head.
-
.~
c
~j>-e
f
g Fi......, IS. Basic types of shut-<>ff element. a Valve. b Slide valve. c Cock (or ballvalve). d Rotary disc in pipe. e Flap valve. f Blank swing disc. g Diaphragm valve. h Teardrop clack valve.
2.9 Shutoff and Control Valves. 2.9.2 Valves
the fluid. a membrane is depressed. are the diaphragm valve (Fig. lSg) and the cylindrical piston valve (Fig. ISh) in which the tlow configuration has circular symmetry.
Slide Valves. The shutoff element (a circular plate with faces arranged either in parallel or to form a wedge) is moved at right-angles to the direction of flow to open up a crescent-shaped or circular free-flow cross-section (Fig.15b) Cocks or Rotary Slide Valves. The shutoff element (a ground truncated cone or a sphere with lateral bore) is rotated about its axis at right-angles to the direction of flow and opens up a lenticular or circular cross-section (Fig.1Sc). Flaps or Butterfly Valves. A disc, initially normal to the direction of tlow, is turned about a hinge or about its own axis, which positions it parallel to the pipe axis in the pipeway; or a disc swings out of the pipeway on one flange bolt, thereby opening up the entire pipe cross-section (Figs lSd-f) Slide valves and cocks that open up the full pipe crosssection are suitable when pull-through elements (pigs) are used to separate fluids conveyed in different forms, Of for cleaning. See Table S for a comparison of the advantages/ disadvantages of these construction forms.
Materials Body materials are selected on the basis of tlow medium requirements (erosio11, corrosion), operating temperature (heat resistance), and operating pressure (strength, possibly resistance against hulging). For the sdection of metalic materials see DIN ):\:\9. About 80% of all bodies are cast, mainly from grey cast iron but also from cast steel and non-ferrous castabk materials (brass and guometal in installation engineering). In the chemical and water-treatmCIlt industries there has been a sharp increase in castings made of plastics (usually injection-moulded). Some parts of valves made from steel are produced by drop-forging (at high pressure).
Grey
1:1..,
Steel. C 20 for drop-forged bodies, yokes and flap screws, weldable; 50 CrY 4 for tlanges, spindles. bolts and nuts up to 520°C, conditionally weldable; X 20 Cr 13 for parts of valves subjected to high mechanical stresses, barely weldable; X 10 CrNiTi 18.9 with very good chemical stability (organic and mineral acids), weldable; X 10 CrNiMoTi 18. 10 in the presence of highly aggressive acids and for higher temperatures. also for cold valves down to -200°C, weldable.
Nonferrous Metals. G-Cu64Zn, G-CuSnIO, G-CuSnS Zn7, G-AlMg3 and others, potable-water valves, physiologically acceptable, AI alloys, seawater proof (shipbuilding), also for the chemical industry. Plastics and Others. Hard PVC (unplasticised PVC uPVC), polyamide, PTFE, and silicones as well as ceramics for the chemical and hygiene industries.
Hydraulic Properties In the case of sharp changes of direction, fittings (valves) cause large pressure drops, a desirable property where they are used as control elements. The flow-resistance coefficient (R and velocity l' are referred to the ,cross-sectional area of the connection A R . The volume V is given by V = A. ~2!1P/ p~", where !1p = ~RPV2 /2 is the flow pressure drop. At large Reynolds numbers (Re> 10'), changes only slightly (for ~. values see Appendix H2, Fig. 2). For completely open shutoff elements, ,. of 0.2 to 0.:\ can he assumed [12]. The value of k, as defined in VDI/VDE Guidelines 2173 for control valves and in VDI/VDE Guidelines 2176 for control flaps, is impurtant in control engineering as are the basic forms of control-valve characteristics [13]. Here the valve characteristics at constant !l.p across the valve must be distinguished from the operating characteristics, which are affected by the ratio of the pressure drop across the valve to the tOlal pressure drop in the pipe as a function of the tlow [14].
'R
2.9.2 Valves Regardless of their function, valves are manufactured as straight-seat, slanted-seat or angle valves.
cast Iron
For water, steam, oil and gas. lined with rubber or enamel for aggressive media (GGL denotes cast iron alloy; GGG cast iron globular); GGL-20 to PN 16 at 120 'C, GGL-2, to PN 16 (25) at 300 "C; GGG-4'i to 70 for feed water and live steam up to PN 40 at 4'i0 0c,
Cast Steel (GS). GS-C2'i for steam, water and hot oil up to PN 320 at 450°C, easy to weld; GS-20 MoV 84 for steam and hot oil up to PN 400 at 5'i0 °c, weldable: GSX 12 CrNiTi 18.9 for acid-proof and hot valves.
Straight-seat Vall'es (Fig. 16). These provide the most advantageous arrangement in piping systems, with easy operation and maintenance and unifonn stressing of the valve elements but entail a high pressure-drop.
Slanted-seat Valves (Fig. 17). These possess a low resistance coefficient {n'
Angle Valves. These can be advantageous if the additional function of an elbow is required, but mean higher pressure-drops. For dimensions see DIN 3202.
Table S. Advantages and disadvantages of the various forms of construction Property
Valves
Sliders
Cocks
Flaps
Flow resistance Opening/ closing time Wear-rate of seal Suitability for changing direction of flow InstaUed length lnstalled height Range of use up to
moderate medium good moderate large medium medium Dj\: maximum PN very good
low long moderAte good small large maximum DN
low short poor good medium small medium ON medium P]\; moderately good
moderate medium moderate poor
Suitability for throttling
medium PN
poor
~mall
small maximum DN only small PN good
I:'.
Components of Thermal Apparatus. 2 Apparatus and Piping Components
Figure 1,. Axial pressure~reduction valve (Samson). J Coupling nipple. 2 Set-point adjuster. 3 Spring. 4 Metal sealing bellows. 5 Cone. 6 Working diaphragm. 7 Seat. 8 Connection nipple.
Figure 16. Straight-seat valve
a. Erhard).
in one piece; the stem is located in a self-sealing cover, the shape of the body is advantageous for flow, and the stem nut is rotated (handwheel height is constant)_
Valve Types for Various Purposes Changeover Valve. This is employed for fluid flow that is to be directed through alternative piping systems. Non-return Valve (prevents backflow). Fluid flow possible only against the force of a spring or a weight.
Pressure-Reducing Valve. Inlet pressure is reduced to an Figure 17. Canted-seat block and bleed valve.
Valve Components. Figure 16 shows: 1, valve body (cast, forged, welded or moulded manufacture); 2, valve disc with seat rings (plate shape, conical or parabolic); seat rings of rubber, cast iron (GG), copper alloys, high alloy steels, stellite or nitride steel depending on fluid, pressure and temperature; 3, valve stem, and 4, nut; 5, stuffing box for sealing off the valve stem; 6, valve or stem drive (handwheel; electromotive, hydraulic, pneumatic or electromagnetic drive with
adjustable outlet pressure (lower pressure) which is kept constant with great accuracy independent of the inlet pressure and the flow rate. For example (Fig. 19), if outlet pressure falls owing to increased flow or falling inlet pressure, the diaphragm (6) with seat (7) moves to the right, thereby opening a larger orifice.
Float Valve. Here a hinged float raises or lowers the valve stem or the valve plate.
Steam Trap (Fig. 20). This drains off the liquid phase (e.g. condensate from saturated-steam equipment) , float
remote control).
drainers , thecmal drainers, thermodynamic drainers.
For large-seat cross-sections, a prelift valve (block-andbleed valve) which reduces the opening force may be useful (Fig. 17). Figure 18 shows a high-pressure control valve. This is a forged valve, with control cone and stem
Safety Valve. This prevents an increase in the operating pressure to values above the permissible pressure (see AD Code of practice A2). The threshold pressure equals the permissible operating pressure. The valve may be weightloaded (very accurate) or spring-loaded (for a compression spring, the valve force increases as the valve lifts).
2
Figure 18. High-pressure forged control valve (SempeU). I Control cone. 2 Stem gUide. 3 Cover, self-sealing. 4 Uhde-Bredtschneidec seal with: 5 Divided ring. 6 Valve stem register to prevent stem turning. 7 Rotatable stem nut.
3
Fipre 10. Thermal action steam trap with regulating diaphragm (GESTRA AG). I Body. 2 Regulating diaphragm. 3 Bonnet . 4 Backstroke cone. 5 Dirt collector. 6 Sieve holder/support.
2.9 Shutoff and Control Valves a 2.9.4 Cocks (Rotary Gate Valves)
Figure 21. Shut-off gate valve. 1 Sealing wedge. 2 Body 3 Bonnet. 4 Stem. 5 Lock nut. 6 Stem nut. 7 Dust ring. 8 Slip ring. 9 Hexagonal
bolt. ]0 to 12 O-rings. 13 Grooved cylinder dowel
Quick-Action Valve. This shuts off the system in the case of pipe fracture or similar damage. Direct shutoff movement is by means of a spring, a weight or a pneumatic force (principle of the closed-circuit current).
2.9.3 Gate Valves Field of Application. Large nominal sizes, high flowvelocities, low to medium nominal pressures, small installed lengths (see DIN 3202). A survey is given in DIN 3200.
Components. With the exception of seats and seals these are similar to valve components (see Fig. 16). Figure 21 shows a simpk shutoff gate valve with an internal stem nut (which is liable to seize up owing to dirt and high temperatures), and O-ring seals in place of a stuffing box. Forms of Construction (Fig. 22). Depending on the shape of the bonnet flange, gate valves may be defined as round-body (large installed length. high pressure-strength of the bonnet connector), oval (shorter construction, low pressure-strength or increased wall thicknesses) or flat valves (further reduction in length, frequent strengthening of the cover connector by means of rihbing, particu-
larly for large nominal sizes). For a summary of materials and limiting conditions of application of gate valves, see DIN 3352 and [12]. In contrast to other valves, gate valves are always suit-
a
cI
Figure 22. Fanns of gate sealing. a Plate gate. b Spectacle shutoff gate. c Wedge gate. d Parallel double-plate gate. e Double-plate wedge gate.
1:111
able for both directions of flow; however, they can be used only as shutoff elements. Generally they are implemented as straight-through devices (there are no angle types). The type of sealing is most important, since the stem force does not act directly on the sealing faces. Figure 22a: Construction is simple; in the closed position a plate is seated by the line pressure. Sealing action is low during opening movement, and wear may occur owing to sliding friction. Used in long-distance gas pipelines. Figure 22b: Raising the spectacle plate exposes the orifice. Where necessary, sealing is by spring-loading. Used in gas and oil systems (and also where dust contamination is liable to occur). Figure 22c: In this common construction, shutoff takes place by pushing a rigid, wedge-shaped valve disc between the seats of the valve body. The stem force enhances the sealing action. Used frequently in the low and medium pressure ranges. Figure 22d: Two parallel moving sealing discs are forced onto their seats at the end of the closing movement by the action of toggle levers or wedges. This results in a greatly reduced sliding action and hence reduced wear. Figure 22e: In this improved form of the wedge gate valve, two wedge-shaped sealing discs, capable of moving relative to each other, are pushed by means of a hemispherically shaped pressure piece onto the seat faces with great force. This is a robust form of construction with high sealing capability and low wear - up to PN 400. The gates are actuated by hand or else via transmission gearing, on electric motor with a gearbox or hydraulic or pneumatic actuators.
Standards DIN 3204 Flat wedge gate valves (PN 4; DN 40 to 300). DIN 3216 Flat wedge gate valves in cast iron (GG) (1.6 to 10; 350 to 1600). DIN 3226 Round wedge gate valves in cast iron (GG) (PN 16). DIN 3228 Flat wedge gate valves in cast steel (GS) (PN 10) DIN 3229 Oval wedge gate valves in cast steel (GS) (PN 16).
2.9.4 Cocks (Rotary Gate Valves) Advantages are simple and robust construction, low space requirement, possibility of rapid shutoff and changeover, low flow losses, possibiliry of conversion to mUlti-way cock with several connections. Disadvantages are large sealing faces which slide on each other, causing wear. The frictional forces, depending on the preloading of the plug, manufacturing quality of the sealing faces, lubricants, and on the type and temperature of the fluid, are relatively high. Further included in the group of taper-plug cocks are the gland cock used, in particular, in the chemical industry for poisonous media (body closed off below, plug sealed and held by packing and follower), the lubricated cock for aggressive viscous and contaminated media in coking plant and in the petrochemical industry (in this case the plug is lubricated via a groove and lubricating chamber), the easy-turn cock for viscous media like latex (in this case the plug is slightly raised before turning and pressed back into its seat after turning), and the mult/way cock, e.g. a three-way or four-way cock, for switching the flow to various directions. An important technical onward development is the ballvalve (Fig. 23). The sealing component in this case is a ball with a cylindrical bore for straight-through flow.
1:11:.
Components of Thermal Apparatus. 2 Apparatus and Piping Components
Figure 13. Ballvalve for large diameter pipelines.
This is practically free of flow resistance (resistance coefficient ~R = 0.03 when the ballvalve is fully open and is approximately equal to the resistance of a similar length of pipe). Such ballvalves are produced in sizes from DN 80 to DN 1400 and for PN 10 to PN 64.
2.9.S Flap Valves (Butterfly and Clack Valves Flap (butterfly) valves, with a construction similar to that of Fig. 24, are used as shutoff valves, throttle (control) valves and, more rarely, as safety valves in the water supply industry (pumping stations and fIltration plant), in power stations (cooling circuits), in the chemical industry (service water, also acidic and alkaline media) and in sewage treatment (purification plant , pumping stations) . They are used in increasing measure in potable water dis-
tributio n and in long-distance gas pipelines where they take the place of oval gate valves. Like gate valves, flap (butterfly) valves dose off tightly against liquids . Flap (butterfly) valves are produced in the larger nominal sizes (DN 5300), generally for PN 4 to DN 2400 and for PN 16 up to DN 1200. Their space requirement is not much larger than the pipe cross-section . The flap (disc) can be actuated by hand, by an electric motor through a spurwheel segment or wonn drive, or by means of a hydraulic piston and, where necessary, a dead weight to amplify the action or to balance the flow forces. In general, the disc is so arranged that the half of the disc pointing upstream moves downward on closing (so that the dosing fo rce is amplified by hydraulic action). Non-return flap (clack) valves serve as safery elements; the flap (clack) is held open by the flow. Under no-now conditions or on reversal of the pressure, the flap (clack) closes helped by its dead weight and, if necessary, slo wed down by an oil brake (dashpot).
2.10 Seals Seals are intended to stop leakage of fluids through the gaps of parts connected to each other (normally flanges; see H2.7. I). They must be easily deformable in orde r to compensate for the roughness of the sealing faces and must be sufficiently strong to withstand initial compression and internal pressures. Temperature and chemical stability must be conSidered, as well as the prevention of electrochemical decomposition of metallic seals or attack on the contact faces through electrochemical anodic action . A summary on seals, their function and deSignation is given in DIN 3750.
2.10.1 Static Contact Seals Figure 25 gives an overvie w o f the most inlpo nant types
•
of seal. They are distinguished by whether they are (a)
~ 5
~
ill
-
]
2
liJ! .:
I
~~H
~
6
8
7
~~~~
~~~-~ I
9
10
,.
I]·
~
~1 16
b Figure 14. a Butterfly valve ( Bopp & Reuther ). b Lens·shaped platt' with rubber sealing rings , body seal insert of stainless st eel.
_
Sealing elemel11
_
12
II
Direction of pressure grarneru
Figure lS. Stati c contaci seals [1';1.
15
2.10 Seals. 2.10.2 Dynamic Contact Seals
1:111
non-removable or conditionally removable (c.r.) and (b) removable. Interposed between them are (1) material contact joints with sealing materials or adhesives. Group (a) comprises (2) welded joint, (3) welded lip seal (cr.), (4) push-fit (c.r.), (5) rolled joint. Group (b) encompasses (6) flat seal (soft or hard), (~) face-to-face joint (without jointing material), (8) compound material
flat seal, (9) edge seal (plastic deformation), (10) fluid seal, (11) circular seal (O-ring of soft material or metal. resilient deformation), (12) seals of hard material (ring joint. resilient), (13) sdf-acting seals of soft material (compressed by internal pressure). (14) sdlCacting seals of hard material (ddta ring). (15) to (l7) seals of stuft1ng box type. Forms of seal implementation with scaling characteristics are as specified in DIN 2~0~. see Appendix H2, Table 1.
a 5
Flat Seals. These are discs, rings or gaskets that adapt to the jointing faces across the whole of their width. They consist either of a uniform material like ashestos board or paper (see DIN 3752: 0.1 to 10 mm thick: application up to 500°C) or It-plates (asbestos with inorganic fillers and an elastomer binder) in accordance with DIN 3754 for cold and hot water as well as oil, steam, saline solutions. etc., 0.5 to 4 mm thick. capable of loading up to 300 bar at 300°C, made of several materials sLlch as laminated metal foil (AI, CLI). or faced with sheet metal, or entirely of metal (see 112.7.1). For flat seals as llange seals see Fig. 26.
Profiled Seals (Fig. 25. 9 and
/0). These are discs or rings which, because of the form of their cross-section, do not make contact across the whole of their width. thereby achieving a higher bearing pressure. They consist of clastomerie materials. soft metals or material comhinations and are - depending on the material - suitable for high pressures (PN 4(0) and high temperatures (about ~OO 0e) (single use only).
Figure 27. a Douhle conical seal. b llhde-Bredtschneider seaL I Cover. 2 Wedge-section sealing ring. 3 Head of container. 4 Divided ring . .5 Fixing studs. 6 Pre-tensioning screws. 7 Retaining ring.
Toroidal Sealing Rings (O-Rings). These are rings of circular cross-section tnade of elastic materials or metals which, with low prestressing during installation, arc able
to form seals. aided by the operating pressure (Fig. 25, 11 and 13). Dimensions: see DIN 3770 (d l = 2 to 800 mm; d!, = 1.6 to 10 nrnI). AjJjJlications: oils; water; air; glycol mixtures from -SO to +200 °C and medium pressure (suitable fOf fepeated use).
High-Pressure Seals. Ca) Small DN Cpipe) (see Appendix H2, Table 1): grooved scal, ring-joint seal (frequent
opening), lenticular seal (lens ring); (b) large DN (apparatus flange) (see Fig. 25): delta ring (14), gap seal (17), or, as in Fig. 27a, double-cone seal self-acting with an intermediate layer of 0.3 to 1 mm aluminium foil, and lIhde-Bredtschneider seal (Fig.27b) - pressureaided. requires no bolts or expensive flanges.
2.10.2 Dynamic Contact Seals Stuffing Box Seals (Packing) Packings are sealing elements which scal off cylindrical faces in relative motion against fluids and gases. The stuffing box seal (Fig. 28) consists of C1) the fIxed part of the casing with the stuffing box space, (2) the sealing
3
5
%
'"
"0
:?
t5 Figure 26. Flat seals (gaskets) and flange sealing faces I:;]. a Smooth raised-face flange and gasket to DIN 2690 (PN 1 to 6, 10, 16. 2'), 40), b Tongue and groove flanges to DIN 2-512 and gasket to DIN 2691 (PI\; 10, 16,2-5.40, 6'l, 1(0). c ~alc and female spigot flanges to DIN 2")15 and gasket to DI]\; 2()92 (PN 10, 16.2'), 40.64. 100).
J
4 2
'"
tL
Figure 28. Stuffing hox seal (Goetze).
1:.,.1
Components of Thennal Apparatus. 2 Apparatus and Piping Components
170
~ 140
co
a
~110
"0
~
en co
~ 80 0.. 50
0
20
100 Internal diameter din mm
Figure 2,. Depth of packing space for laminated packing rings (Goetze).
b
Cf~ 12'
d,
k7
Figure 30. Packing rings (Goetze). a Hollow ring. 1 Lead or copper. 2 Graphite lubricant. 3 Radial holes. b Bevelled lip ring. I Bevelled washer. 2 Soft material infill. 3 Bevelled lip U-ring.
material (packing), (3) the collar plate (follower) screwed into the casing (flange or thread; providing adjustment), (4) the intennediate lantern ring (to distribute lubricating oil if necessary) as well as (5) the shaft or stem capable of rotary and axial movement. Packings can be used at relatively low sliding velocities (up to about 0.3 m/s), high temperatures (up to about 520°C), high pressures (up to about 300 bar) and shaft diameters of 10 to 200 mm; the external diameter of the packing may be from 18 to 245 mm (up to 800 mm for expansion compensators in gas lines). Sealing works on the principle that screwing down in the axial direction causes lateral defonnation and pressure against the cylindrical sealing faces. The width of packings of soft material is equal to '>I'd for small diameters d of the stem and = 2'>1'd for large ones.
Laminated Packing Rings (Fig. 29) These are made from corrugated metal intercalations such as soft lead, copper, nickel, or chrome-steel, embedded in layers in asbestos or cotton. The rings have an inclined split and can therefore be bent open and placed around the shaft.
Figure 31. Axial face seal (Burgmann). 1 Rotating mechanical seal. 2 Stationary counter ring. 3 Compression spring. 4 Washer. 5
If several rings are fitted, the splits must be staggered. In
form.
the case of gases, the seal should be improved by lubricating oil so that friction is reduced.
required. Such seals are suitable for very high pressures (in excess of 400 bar) in autoclaves, press and ultra-highpressure pumps.
Hollow Ring of Lead or Copper (Fig. ~Oa). These may be undivided or divided into two. The lead or copper casing is filled with graphite lubricant, which penetrates through small radial holes for self-lubrication; ground bearing-faces are required. A typical application is in hydraulic press pumps. Foil Packing Ring. Cotton wrapped with aluminium foil. Bevelled Lip Packing Ring (Fig. ~Ob). Axial stresses are transmitted to the running surface as a result of the wedge
Sealing ring. 6 Bearing ring.
Perfectly
functioning
external
lubrication
is
Rotating Mechanical Seals. Axial and radial rotating mechanical seals have increasingly displaced stuffmg box packings for rotating shafts. Figure ~ 1 shows the principles of construction of an axial rotating seal. This can be applied to shaft diameters of 5 to 500 mm, pressures of 10-' mbar to 450 bar, circumferential velocities of more than 100 mls and temperatures from -200 to +450 0c. For various configurations, leakage losses, axial face seal closure, friction losses and operational integrity, see [15, 16].
3.1 Tube-Bundle (Shell-and-Tube) Heat Exchangers
Types of Heat Exchanger 3.1 Tube·Bundle (Shell.and· Tube) Heat Exchangers The shell-and-tube heat exchanger is used in many branches of industry because of its versatile applicability to gaseous and liquid media within wide ranges of temperature and pressure. Figure 1 shows a DIN 28 183 heat exchanger featuring fIxed tube ends, one tube-side and one shell-side pass each. a shell compensator. and dished ends in the form of vessel domes. If the tubes and shell are of the same material and the temperature differences are not too great, thin tube ends can be used owing to the support action of the tubes. Easy mechanical cleaning of the tube space (inner space of tubes and dome) is possible. An exchanger with hairpin tubes (ll-tube exchanger, Fig.2a) requires thicker tube plates. This is constructed with two tube passes and, as counterllow types, with two shell passes (partitions must be watertight) and flat head closure. The floating-head version (Fig. 2b), in contrast to the fIxed-tube version, can deal with greater temperature differences between tubes and shell. There is easy cleaning of the shell and tube space. Construction is by means of four tube passes and one shell pass. The inner dome is sealed by means of a divided counter-flange. The pUllthrough exchanger (Fig.2c), in contrast to the exchanger shown in Fig. 2b, allows removal of the tube bundle without dismantling the floating head.
b
Disadvantages. These are large clearances between bundle and shell, bypass flow, and installation of internal slide rails for removal of the tube bundle (see Fig. ~). Figure 2. Various designs of tube-bundle (shell and tube) exchanger (Dupont). I, Shell-side medium. II, Tube-side medium. 1 Flat end (head closure). 2 Partition.
Important Standards DIN DIN DIN DIN DIN
28 28 28 28 28
008 080 180 182 191
Tolerances. Saddles. Steel tubes for heat exchangers. Tube sectious and tube connections. Flanged floating head.
omic concentration is shown in Fig. 4. After entty of the fresh solution, a thin liquid flIm is produced on the inner faces of the tubes by a suitable device. This flIm flows downwards under gravity, together with the vapour produced. In this way a given concentration of the solution is achieved during a single pass through the heat exchauger, if the exchanger has been suitably designed.
Tube bundle excbangers are also employed for phase changes: concentrators, evaporators with forced and natural circulation, condensers (see H4), waste-head boilers with steam generation. A falling film evaporator for econ-
~7;3
5
21 7 2
I
17
LrJ Typical fabrication in unalloyed steel
7
72 3
76
18
19
7
~
Typical fabrication In stainless steel Detail X
73
20 7 70
27 75
77
6
3
\
22 8
17
74
5 27 9
13
3
.~ Detail W
Figure 1. Tube bundle (shell and tube) heat exchanger with two fixed tube plates (1], as in DIN 28 \83. 1 Shell. 2 Tubes. 3 Segment baffle. 4 Shell moWIt. 5 Vent connection. 6 Drainage connection. 7 Tube end, tube plate. 8 Dome mount . .9 Dome end. 10 Dome flange. 11 Seal. 12 Compensator. 13 Tie-rod. 14 Spacer. 15 Vent boss. 16 Drainage boss. 17 Support. 18 Shell flange. 19 Flange stub. 20 Impingement plate. 21 Lifting eye. 22 Dome shell. 23 Shell/tubeplate stub.
Components of Thermal Apparatus • 3 Types of Heat Exchanger
Figure~. Wetded floating head [11 to DIN 28190. Two tube passes (passages), nominal diameter 350 nun, enveloping ring diameter 288 mm. J Made with flat plate. 2 Fabricated with domed end. 3 Slide rail 30 x 10 flat. 4 Internal
2
tube. 5 Tie-rod, 12 mm diameter.
1 Freshso'utioIl
Healilg
A
steam. ""'IHrIHHII"-.
Heater
DEgassing Condensate
W
1Vapours vapour
I
space
Alkaline sotutioll
Figure 4. Falling-film evaporator (Wiegand).
In Fig. S the walls of the heat exchanger are protected from unacceptably high temperatures by built-in tube spirals (steam generation). Cooling of the gases takes place in the built-in ftxed-tube exchanger by heat transfer to another process gas. Because of the high thermal stresses in thick-walled tube ends at high temperature differences, special designs are required [2]. A cooled tube end supported by a cradle is shown in Fig. 6. By suitable flow guidance, uniform cooling is achieved and deposits from the water (peak temperatures) are avoided. At very high temperatures (1000 to 1500 °C), additional sleeve tubes must be provided to protect the tubes and plates, or the tubes must be cooled individually at the gas entry end [2,3].
3.2 Other Types These include heat exchangers with fumed surfaces [4], which reduce thennal resistance in particular in the presence of gas flow (air coolers). Finned surfaces are also employed in evaporators (low fms) if large heat-transfer coefficients occur on the hot side, e.g. in the case of steam condensation.
Figure S. Heat exchanger in nitric acid installations with built-in residual gas heater (SteinmOIler). J Water-carrying spiral tube as wall protection. 2 Heat-exchanger packs (spiral tube for steam generation). 3 Residual gas heater.
Plate-rype heat exchangers [4] , which can be used in versatile ways to vary the flow of the media and can be cleaned easily (making them useful in the food industry), also need to be mentioned. They consist of a pack of proflied plates, separated by soft seals held together by means of a clamp. Large transfer areas can be accommodated in a small volume. Spiral heat exchangers (Fig. 7) represent a special form. They are produced by the winding of two or four plates, furnished with spacing bolts, around a stable core of up to 2 m diameter. The front faces are closed off by endplates and soft seals.
4.1 Principles of Condensation
Figure 6 (left). Waste-heat boiler with water-cooled and cradlesupported double bottom (Borsig). 1 Water inlet. 2 Distributor gap . .3 Baffle plates. 4 Support. 5 Load-bearing cooled bottom. 6 Supported membrane bottom_ 7 Water inlets. 8 Evaporation chamber. 9 Tube gap. 10 Brick lining.
)
I
Figure 7. Spiral heat exchanger in counterflow configuration (Kapp). I, 2 Inlet and outlet of cold medium. 3, 4 Inlet and outlet of warm medium.
Advantages. These are high flow-velocities, high heattransfer coefficients (1500 to 2300W/(m' K)), compact construction (20 to 70 m'/m'), dirt resistance and easy cleaning.
Section A-B (enlarged)
Disadvantages. These are low pressures and temperatures (up to 20 bar and 400°C) .
• • • • • • •;) Condensers and Reflux Coolers • • • • • • • • 4.1 Principles of Condensation When a vapour cools below its saturation temperature, or dew point, it changes to a liquid.
Areas of Application. In the case of condensers, this means the production of the highest possible vacuum (steam engines), the recovery of the condensate as a valuable liquid (distillation plant), and the precipitation of exhaust vapours damaging to the environment (vapours containing corrosive materials), as well as heating (water vapour as a heat-carrying medium). Cooling Agents. Water, air, refrigerating brine, and other substances which can take up heat, act as cooling agents. Types. These, and their working principles, arc as tallows:
Surface Condensers. Vapours are condensed by indirect contact with a cooling agent, usually via cooling surfaces
consisting of tubes (a form of construction known as 'closed').
Injection (Mixed) Condensers. Vapours are brought into direct contact with injected cooling water and precipitated.
Direct Air Cooling. In air-cooled condensers of open constmction, vapours are turned into liquid by heat transfer to the ambient air.
Indirect Air Cooling. Water is used as the cooling medium in surface or injection condensers and then transfers the heat to the air via cooling towers or watercourses. Surface and air-cooled condensers allow the recovery of pure condensates and the production of a higher vacuum than do ntixed condensers (air is dissolved in the injected water'); they are particularly useful for the precipitation of vapours of no commercial value. For heating and evap-
Components of Thermal Apparatus. 4 Condensers and Reflux Coolers
oration, the closed mode of construction of surface evaporators is required.
Non-Condensable Gases. These build up at the points of lowest pressure (lowest temperature) and there form a layer of increasing thermal resistance. Since the vapours must diffuse through these to gain the cooling surface, the vacuum becomes weaker. At constant total pressure, the partial pressure and the process-promoting temperature gradient between the vapour temperature and the coolant temperature are reduced. Condensers must therefore be vented at pressures above atmospheric and, when operating under vacuum, must be kept free from inen gas by pumping off.
4.2 Surface Condensers 4.2.1 Thermodynamic Design Heat To Be Removed
Q= mo(bo -
b K)
=mwCw(t2 -
t l )·
(I)
Cooling Surface of a Condenser A = mo(bo - bK)/k/!;.tM
•
(2)
In Eqs (I) and (2), mo, mw are vapour and coolant mass
flow respectively, b o and b K are the specific enthalpy of the vapour and condensate respectively, Cw is the thermal capacity of the coolant, tl and t2 are the inlet and outlet temperatures of the coolant respectively, k is the heattransfer coefficient and /!;.tM is the mean temperature difference (see HI.2.1).
Heat-Transfer Coefficient Ie (see CIO.2). This is usually detennined from the heat transfer on the coolant side, because the heat-transfer coefficients on the condensation side are large - panicularly in the case of water vapour; k increases with coolant velocity and reducing tube diameter. For steam condensation with cooling water flow in the tubes of between 1.5 to 2.5 mis, k "" 3000 to 4oo0W/(m2 K) (see HI.2.1). The cooling surface area A obtained from Eq. (2) is divided up into design units and k is recalculated on the basis of the geometrical data so obtained [1,2]. In this process, separate account must be taken of the effects both of layers of din and of inen gases [3]. Superheated Steam. In this case a film of condensate fonns on the wall when the wall temperature is equal to or lower than the saturation temperature of the steam; the heat-transfer coefficient for condensation itself (see CIO.4.2) changes only slightly during the process. The ranges for vapour cooling (dry wall) and condensate cooling require separate calculations. 4.2.2 Condensers in Steam Power Plant In steam power installations, the aim is the generation of the largest possible pressure and heat gradients. Owing to the large specific volume of steam under vacuum, large inlet cross-sections are required so that the pressure drops
do not exceed the gain in gradient. EconomicaUy attainable end pressures PI are 0.1 bar for piston engines, 0.025 bar for turbines (assuming low temperatures tl of the cooling water, which vary with locality and season). Values of tl and PI applying to central Europe: ground water 10 to 15°C and 0.03 bar, river water 0 to 25 °C and 0.04 bar, recooled water 15 to 30°C and 0.06 bar. The pressure PI is from 0.005 to 0.01 bar higher than the satu-
ration vapour pressure corresponding to the outlet temperature of the cooling water. The cooling water mass flow mw "" 70mo for steam turbines, mw "" 40m o for piston engines. If to is the temperature of the saturated vapour at the cooling waste outlet, then the relation to - t2 = 3 to 5 K applies. Supercooling of the condensate to - tK < 3 K, since otherwise inen gases will be dissolved and returned to the circuit. The pumping off of the inen gases must be effected at the coldest point (lowest total pressure), which must be screened against steam ingress (see H4.2.4).
4.2.' Condensers in the Chemical Industry Surface condensers for the recovery of valuable condensates downstream of columns and reactors are cooled either by water or by air (see H4.4). Products that need to be reheated or evaporated are also employed in increasing measure as cooling agents in order to save energy. Water, as the cooling agent, flows in the tubes (providing a better means of cleaning), pure condensate media on the sheU side of bundles (giving a greater cross-section and lower pressure drops). This demands particular consideration in the case of vacuum operation, which is employed for temperature-sensitive substances.
Heat·Transfer Coefficients. For condensing organic media, these are lower than those for steam. If it is a matter of reaching high values, condensation on horizontal tube bundles is more advantageous than on vertical ones. This applies above all at low boiling temperatures and shon tube lengths. Heat transfer is improved if the flow is lateral to the bundle. On the side of the cooling medium, conditions for good heat transfer must be created through high water velocities and the avoidance of dirt layers. This is best accomplished by automatic cleaning devices of brush or sphere type. If water conditions are unfavourable, coating of the tubes, to provide smooth surfaces, may help. Heating and Evaporation of Media. This is often carried out in condensers where steam from a power station is used as a source of heat. Since, under these circumstances, condensate heat-transfer coefficients are high, the conveying of phases and the geometrical layout on the evaporation side are detennined by the smaller values that occur in most cases. For this reason, vertical bundle arrangements with condensation at the columns on the sheU side of circulation condensers are commonly encountered, but horizontal bundle arrangements may also be found in reboilers of 'kettle' type, where condensation occurs on the tube side. For exact calculations concerning such evaporator condensers, the distribution of the heat-transfer coefficient for condensation and evaporation must be determined section by section. Muiti-Component Mixtures. The calculations require particular effon if such mixtures condense (panial condensers, reflux condensers), possibly with the aid of inen gases, and multi-component products are preheated on the side of the cooling media (part evaporators, flash evaporators). In this case, boiling or precipitation temperatures, which vary along the exchanger, must first be detennined from equilibrium calculations and then plotted in an enthalpy-temperature diagram. In exothermicaUy operating reactors, such q,eat exchangers serve for the heating up of reaction materials by the reactive product. Owing to the high temperatures (thermal stresses) and the danger of contamination, floating-head exchangers are preferred (see H3).
4.3 Injection (Direct-Contact) Condensers
4.2.4 Design Considerations Low-Pressure Saturated Vapour Condensers These are constructed mainly as horizontal tube-bundle configurations.
Low Pressure-Drops. For large steam volumes, a further steam chamber is located below the large entry connection tube and wedge-shaped steam passages in the upper tube bundle (Fig. 1). This leads to the danger of partial pressure minima occurring in the part bundles at which inert gases collect, which impede heat transfer if the gas is not pumped away. It would be more advantageous to narrow the condenser in the downward direction or to provide it with tube sections that taper downwards [2 J. However, this has not gained favour for reasons of design and cost. Inert Gas Removal. This is done exclusively from the coolest point (pressure minimum), with a minimum steam contribution. In accordance with [4], the best solution is to arrange for the pumping off to take place at the centres of the part-bundles through tubes running the length of the bundles with many suction orifices. Baffles screen against steam ingress; dead comers must be avoided. Prevention of Supercooling of the Condensate (Fig. 1). This is accomplished by deflectors (2), which keep the condensate away from the cooling tubes. Vapour traps or suction pumps continually drain the condensate. Design. Shells: > 500 mm dia. welded from sheet steel, length = 2 x dia. Tube ends: steel (or brass if water contains acid or salt) 20 to 30 mm thick. Tube dia.: 15 to 25 mm. Tube sections: (1.4 to 1.5) x external dia., tapered downwards. If there is condensation, no deflectors are required on the shell side. In order to avoid vibrations, supporting sheet-metal plates must be fitted at intervals of tube dia. x (50 to 70). For design in relation to vibrations see [1, 5 J. Thermal expansion must be taken up by expansion compensators or by prefonned S-shaped tubes (the
z
Figure 1. Single-flow Bakke-Durr condenser. Diameter 2.5 m. length 12 ffi. Tubes: number 4960, dimensions 19,05 mm x 0.89
mm, spacing 26mm. Vapour 44.2 kg/s, water 2.1 m-~/s, power 97 MW. 1 Vapour slots in tuhe-support wall. 2 Condensate deflector slats. 3 Condensate outlet. 4 Air suction tube connection.
S-bends should be located at the supporting plates). In double-flow versions, one half can be cleaned without shutdown of the installation. A safety exhaust valve must be provided at the vapour inlet.
Condensers in the Chemical Industry Vapour passages are not nonnally required (higher pressures, higher pressure-drops). Length: = dia. x (3 to 4), depending on the pressure drops and the layout of flow passes. An advantageous design consists of central vapour entry with division of the flow by a separating partition in the longitudinal direction of the bundle (i.e. split flow) [6J. Tubes dia.: 18 to 25 mm. Tube section: outside dia. x 1.3 to 1.5, to DIN 28 182. Floating-head condensers may be manufactured in the simplest pullthrough versions (see H3) using a strip gasket, because space is required for the distribution of the vapour and the collection of the condensate. If two tube plates are sufficient, the U-tube type is also suitable (tube passes cannot be cleaned). If condensation takes place in the tubes, the tube bundle is inclined for easier drainage of condensate. Apart from tube bundles, there are also spiral tube, double tube and surface irrigation coolers employed as condensers.
4.3 Injection (Direct·Contact, Condensers By injecting finely atomised cooling water into the vapour, greater heat-transfer coefficients can be obtained in comparison with surface condensers. These are measured in [7J for freely falling films, jets and drops as well as for pressure atomisation. In the latter case, values of k = 100000 W I(m' K) were measured for droplets of 0.6 mm diameter and velocity of 15 mls at a themaal flux density of 230 000 W m l The values reduce considerably with decreasing drop velocity or increasing dwell time, as well as with decreasing condenser pressure and increasing inert gas content (50% reduction at 1% gas content by mass). Since the area of the phase boundary per unit volume is also large, the dimensions of the direct contact condensers are smaller than those of surface condensers. Internal structures to increase the area of contact and the dwell time are relatively inexpensive. The specific cooling water requirement mw/mn is calculated from Eq. (1). Since t, = t K , mwlmo at 15 to 30 kglkg is smaller than in the case of surface condensers. For large power capacities and low pressures, the counterflow configuration (dry pumping-off of the inert gases at the head) is more economic than the parallel flow configuration (wet pumping-off). Discharging of the condensate and cooling water is usually carried out by means of a liquid receiving tank or via a water-jet pump, also, in the case of parallel flow configurations, via a jet condenser. Owing to the mixing of cooling water with the condensate, this efficient process can be applied only to vapours without commercial value. An exception is the Heller process [8], where vapour in an injection condenser is precipitated only by its own condensate, which has previously been cooled by air in a dry cooling tower. This indirect air cooling process (see H4.6) is applied only when there is a shortage of water. The injection condenser is only a third the size of a surface condenser of the same power. On the other hand, the investment costs of condensate cooling are considerable. By means of a threefold flow rate of air, the same vacuum is obtained as with a wet cooling tower [9J.
I: ••
Components of Thermal Apparatus. 4 Condensers and Reflux Coolers
4.4 Air·Cooled Condensers
4.5 Auxiliary Equipment
In the case of water shortage, apart from indirect methods, direct air cooling, requiring smaller surfaces, is applied in increasing measure. In most cases cooling takes place by blowing air against the fmned external faces using fans or, more rarely, by natural air flow. Owing to legal requirements, slow-running, low-noise fans with wide fan blades are in increasing use. Investment costs are higher than those of surface condensers. However, if air cooling is compared with surface condensers and the recooling plant is taken into consider-dtion, the investment costs are approximately equal, but the running costs for air cooling are lower as long as the product temperature lies above 60°C.
The condensate and the air which enters with the vapour, the cooling water and through leakage (piping systems, stuffmg boxes of machines) must be continually pumped away from the condensers. Wet-air pumps are used almost exclusively for direct-contact condensers in the counterflow configuration; they convey condensate and air Simultaneously at atmospheric pressure. In larger installations, dry-air pumps are employed with separate pumps for the condensate.
Installations for Power Stations. These are built with power capacities up to about 1100 tlh of condensate (400 MW). The tube bundles can be arranged vertically, horizontally or inclined (in the form of an A-frame or a Vframe) and, in order to save space, can be mounted above pipe bridges or on the tops of bUildings. The A-frame arrangement (Fig. 2), where steam is supplied from above, is in widespread use, producing parallel flows of steam and condensate. The decreasing condensing power of the tube rows that are situated in the already warm airflow is compensated for by the narrower fm spaces (1 in Fig. 2). Under frost conditions and during vacuum operation, liquid in the lower ends of the tubes is liable to freeze owing to the formation of dead zones through reflux streaming in the tubes with complete condensation and the occlusion and concentration of inert gas. In such cases, steam supplied from below (counterflow), giving poorer heat transfer, may be a solution. Alternatively, a combination of both configurations may be employed, which ensures that in the parallel-flow condenser, situated upstream, partial condensation occurs in all tubes, thereby
preventing supercooling of the condensate. If operating conditions vary, it is safer to provide each tube row with a separate collector.
Refineries and Chemical Works. Here too, aircooled condensers fmd increasing acceptance. In the form of head condensers for distillation columns, they are now being built for cooling rates of up to 40 G]lh.
4.5.1 Air Ejectors As far as turbines or steam piston engines are concerned, if no empirical values are available for design, average air quantities of about 0.1 to 0.25 of the maximum condensate mass percent may be assumed. Even with good cooling of the air to tL , the vapour flow mD pumped away with the air is greater than the air flow mL: (K)
10
2
4
6 1.2 to 1.5
12 to
5 to
2 to
13
6
3
Low values occur for total pressures of about 0.02 bar, high values for 0.1 bar. Cooling must be so designed that to - tL > 4 K, where to is the saturated vapour temperature at the condenser inlet. The most widely used pumps are water-jet and steamjet pumps. Apart from these, water-ring air pumps are employed, if the pressures are not too low, which can also be operated using low-vapour-pressure sealing liquids if necessary. Oil-filled rotary piston pumps are suitable for high-vacuum applications. The advantage of the jet pump over mechanical pumps is that it can be manufactured from special materials for dealing with corrosive media.
Water-Jet Air Pumps. These are built only as singlestage types. The air is compressed isothermally because the heat of compression is conveyed to the water. Theoretical vacuum (depending on the inlet temperature of the water) is achieved to 98%. Efficiency is low owing to the high water consumption (20 to 40 m' per kg of air), but greater than in the case of vapour jet ejectors. To obtain greater power, parallel operation should be employed. Water-jet pressure should be above 2 bar.
Advantages. The large pumping capacity means rapid attainment of the operational state of the condenser, while the simple mode of construction, with no moving parts, bestows high operational integrity, the more so because there is no sensitivity to contaminated water.
Disadvantages. Such devices are liable to power losses and to loss of condensate.
3
3
Vapour-Jet Air Ejectors. In these, a pressurised steam jet is brought to supersonic velocity by expansion in a Laval nozzle. Air is drawn in by subatmospheric pressure. The maximum single-stage compression ratio is 1 : 7, corresponding to a partial vacuum of 0.15 bar in conveying
Figure .1. Air-cooled condenser in A-frame configuration. 1 Finned tubes with varying spacing between fins. 2 Vapour supply. 3 Condensate drain. 4 Fan.
to atmosphere. In the case of low pressure, operation involves several stages where condensable components are precipitated in intermediate condensers in order to save energy and to reduce the size of the following stages. Turbine condensate (feedwater preheating) can be used for cooling. Figure 3 shows a two-stage version. Direct condensers as well as surface condensers can be employed. In many cases a water-jet air pump or a water-
4.6 Indirect Air Cooling and Cooling Towers. 4.6.2 Design Calculations
effect and where closed circuits are unavoidable, e.g. in nuclear power stations or in the Heller process using direct condensers (see H4.3). The planned combination of both processes [10, 11] allows change from the preferred dry operation in winter (where the characteristic depends on air temperature) to partial wet operation in summer (where the characteristic depends on wet-bulb temperature) .
a
4.6.1 Types
5
J
5
Figure 3. Vapour-jet air ejccwf (type: K6rting), t\-vo-stagc with intermediate condensation. a, b Vapour-jet ejector, nrst and second stage respectively. c. d Surface double condenser, first and second stage respectively. 1 Suction connection. 2 Operating stt·am. 3 Cooling water inlet. 4 Cooling water outlet.,) Condensate drain 6 Air vent.
ring air pump as the last stage (pressure range 0.2 to 1.0 bar) offers greater economy. Advantages. These are the same as for water-jet air pumps. In addition, suction pressures are lower and recov-
ery of the condensate is possible. Disadvantages. An additional condenser is required, and efficiency is lower.
4.5.2 Cooling.Water Pumps and Condensate Pumps Cooling-Water Pumps. For greater tlow rates (up to 15000 m'/h) and low conveying pressures (0.8 to 2 bar), these are usually of centrifugal type. [n power stations, duplicate drives are by steam turbine and electric motor for reasons of safety. Allocation is usually to several parallel sets of pumps; power matching occurs without change of speed by switching in Of switching out. Condensate Pumps. These are also designed as rotary pumps. They are over-sized in case of leakage in the cooling water pipes. A positive suction head is required. Such a pump is usually implemented in a multi-stage version, and is often arranged on a common shaft with the coolingwater pumps. The power requirement of the condensation plant is 0.4 to O. '1(.~) of the normal power of a prime mover.
4.6 Indirect Air Cooling and Cooling Towers If no fresh water is available or inlet flow temperatures
have been specified, indirect heat transfer to the air must be accomplished via an intermediate cooling agent (almost exclusively water). Depending on water now, the towers are termed 'dry' or 'wet'. In the case of dry cooling towers, the heat is transferred by convection over cooling surfaces (i.e., a finned tube bundle) which have a higher thermal resistance and a smaller enthalpy gradient than those obtained in the wet process. In the latter case, the heat is transferred mainly by evaporation of water. About I to 2% of the tlow to be cooled is lost and must be replaced. The environmentally friendly dry process requires greater ventilation power and higher investment costs. It comes increasingly into its own where the disadvantages of the wet process - damp mist, spray, formation of ice, extra water, encrustation and corrosion have a deleterious
Open cooling ponds (0.5 m'/h of water to be cooled per m' of ground area over a cooling interval of ]() K) or graduation works with 1 to 2 m'/(h m') have by now become a rarity. Nowadays the choice for large power capacities is mainly a dosed cooling tower built of concrete (5 to ]() m'/(h m'»; for small units, small coolers are made from plastics (up to 4()OO m'/h), while for medium power capacities, cellular cooling towers are built of prefabricated reinforced concrete sections. Types of Cooling Tower. A distinction is made between towers with forced and natural draught. At base load, natural-draught cooling towers are more economic for large power operation than are those with fans in spite of higher investment costs. Dimensions: These stnlctures measure approximately 1]() m in diameter and 150 m in height with concrete walls 14 em thick (power IO()O to 1200 MW for wet cooling towers), while greater dimensions, which would be required for dry towers of the same power (20() to 300 m diameter and height), can be realised cost-effectively by means of the catenoid form of constmction (Fig. 4). The paraboloid form with constriction is intended to prevent ingress of cold air from ahove [12, 13]. Forced·Draught Cooling Towers. These are economical only as discharge cooling towers (cooling of the discharge water tlow in summer) for peak-load operation of large power installations (up to 100 000 t/h). Normal application is for small to medium generator power capacities. The diameter of the round cooling towers is up to 70 m, that of the fans up to 26 m; operating power is 0.55 kW 1m' of droplet-receiving ground area. A pressuregenerating fan is used for the sake of low noise (with sliding sound screens at the air inlet); the diameter is 7 to H m. Tower height is under 50 m. The shell is easy to manufacture. Costs of construction are 10 to 15% less, the power requirement IO to 1';% higher, than in the case of an induction fan. Wet Cooling Towers. The water to be cooled is pumped to distributing troughs in the lower third of the tower and then tlows atomised by nozzles; or, distributed by overflow gutters to internal fittings, it flows in thin layers towards the air current. The water drop is 6 to 8 m. Air enters from below. Configurations with countertlow using drip plates are usual in Europe, with crosstlow the norm in the USA. 4.6.2 Design Calculations
Ileat transfer in dry cooling towers is calculated as in fmned-tube coolers. In towers with natural draught, the air tlow rate is calculated from the balance between buoyancy and pressure drop [I]. For wet cooling towers, the main equation [141 derived by Merkel applies approximately. Heat Flow. The heat extracted from the water by the air through heat transfer and evaporation is given by dQ
=rnwcw dtw ={3x(h'; -
hl)dA
(3)
1:.:1
Components of Thennal Apparatus. 4 Condensers and Reflux Coolers
Masthead
Ho 190m
Curb ring Oo92m H,146m Spoke wheel (radial lie-rods)
Oo92.7m
Ho112 ,3m
Spoke wheel (radial tie-rods)
0, 89.6 m H,68.7m
Figure 4. Catenoid dry·type cooling tower Schmehausen [12J, 1 Mast. Z Spoke wheel (radial stays) , 3 Spoke wheels (radial tie·rods), 4 Sheil, 5 lining, 6 Aircraft warning light, 7 Access ladder. 8 lift, 9 Catwalks, 10, II Internal and external access for inspection. 12 Heat exchanger, A-frame configuration. 13 Crane.
and is proportional to the boundary area of the phase cIA between water and air and the difference of the enthalphy
h;~
of saturated air at the temperature tw of the water
and the enthalpy h, of the air, f3x is the mass transition coefficient (mass flow per unit area),
Lewis Number. If a is the heat-transfer coefficient and CLm the mean thermal capaciry of the moist air, then Le = a/ (f3xCLm) ,
(4)
applies, Le = 1 for evaporating water,
Merkel Number. Neglecting changes in the quantity of water, Me = f3xA = jcw-dtw
mw
h~-hL'
(5)
follows from Eq, (3), The Merkel number is derived by an averaging process for h~ and h, [1] or graphically (Sherwood),
Number of Transfer Units. With mass flows air and mw for water. this is given by NTU = f3xANIc =
mw ' Me/m,..
m, for (6)
The loss of water is made up from f3xA plus the entrained droplets (0,3' f3xA) and must be replaced by additional treated water, The cooling effect increases with increasing water temperature, Below the limiting cooling temperature (wet-bulb temperature) , which is lower for unsaturated air than the dry air temperature. water cannot be cooled, For the usual mean values of 15 °C and 70% relative humidity measured for air. the limiting cooling temperature is about 12°C. The usual interval of the cooling region tW2 - tw , = 10 K, For the calculation and plotting of the changes of state in a Mollier h. x diagram, see C6.2,2, Values for f3xA or NTU may be found in the literature or may be determined empirically [1].
Components of Thermal Apparatus _ 5 Appendix H: Diagrams and Tables
I: ••
Appendix H: Diagrams and Tables Appendix H2 Table 1. Sealing data for gases and vapours in accordance with DIN 2505 [4}
Type of seal
Form of seal
Designaiton
Sealing data
Material
Imtial deformation ko mOl
Seals of soft materials
r-- Do ---j
Flat seals as
~
~
DIN 2690 to DIN 2692
r--d~-
Seals of metal and soft materials
Spiral asbestos seal Corrugated ring seal to DIN 2698
~11
U:
Sheel melal encased seal Metal seals
~~::
Flat metal seal
~
Diamond-edge metal seal
~.
Oval-profile metal seal
.-
Rubber PTFE It (asbestos with inorganic fillers and an elstometer binder Asbestos! steel
2 bD 25 bo
O,5bD 1,1 bD
bD~
b D(O,5+_5_)
50b D
1,3 bD
30 bD 35 bo 45 bu
O,6b D 0,7 bD 1 bD
50b o 60b" 70 bo
1,4bD 1,6 bD I,Sbo
I·b o
k,
M
bo +5
1,5
Ring-Joint seal Lenticular seal to DIN 2696
-----~ ~
mm
M
AI Cu, Ms Soft steel AI Cu, Ms Soft steel
Round-section metal seal
~
-
Operating condition
ko·K D N/mm
Z; number of crests
Crest profile seal to DIN 2697
0,5VZ
Appendix H2 Table 2, Standard velOCity values in m/s [6] Hot vapour (v; 0.025 m'/kg) 35 Hot vapour (v; 0.2 m'/kg) 50 Saturated vapour, also pipes in piston engines 15 Gas (long-distance pipelines) 5 Gas (domestic piping) 10 Air (STP) Compressed air Oil (long-distance pipelinesa ) Fuel lines in internal-combustion engines Lubricating oil linesa in internal-combustion engines Water: Suction side of pumpsh Delivery side of pumps Domestic piping Long-distance pipelines For water turbines
a
Attention to viSCOSity
h
Danger of cavitation.
to 45
to 60 to 2S to 10 to 20 to 40
to 10 to 2 approx. 20 0.5 0.5 1.5 1.5 1.5 2
to to to to to to
I
1 to 2 2to 4 2.5 35 4to 8
9+0,2·2
I : Mt.1
Components of Thermal Apparatus • 5 Appendix H: Diagrams and Tables
s
--
-
II
-i-
1
'-~~l~ r'\ -0
'f?
r'
10 2
<0
8
~":K
o !--
~\
10
X
V-
\
1\
X~~\ \~~
~
1\~kr; \( VI\
-'" V
2-
~ 1\
-'" ~\
-'" A
K
..
1\
~ -~
~ Y\ 1\
\ /'
:.c:::: -~
V~
--"-
--'" ,;,'9 1\/\\1.
\ ~~ V 1\ 1\
1~ l\~'(:,V
\ X\
1\
/'
-". --'" \
<:,
\ ~ K 1\
~~ y
f'lv<:'~
1'" ' t>( 1\\'(:,,,,:0+1.
-O~\ &~
~V\
o
---
""S'
oJ';
,,1~
1\ '(:,,,
/'
Y~
.':kV 1\ ~
--'" _1.1,1.,1'9
~
-"" VL
V\
r--r-t-
:>,'\1-1'91--
1\ ~Li1'(:,~1-1:O ~~ Vf\v\ t\ DI~I \ V ~ '" 1\ ~V [\ ~ 1\ 1\ 1\ --"PI:' \../" --'" r'\ \ K P 1\ \V X ~ ~ ~ V !"V \ V I\~ l\V ~\ V U l-'" ~ \; '\1\ I\~ V r\ ~
I\V
1\
V
r;
1\X
/'
./'\
./' -.~
P\~
f/
!"-.\,\V\
Y
\
\ / - 1-
I
1 I
1
6 810-'
6
B
1
6
, mbor/m
B 10
6
B 10'
Figure 1. Pressure drops in steel tubes DIN 2448 for cold water (+10 °C).
Nominal size in mm
5,0
= 40
Slraight-through valves Free flaw Type Boa
""c ;g 30 Q)
Q)
0
"
Q)
~
~
.'" ~
~ u::
DIN
2.0
32
40
50
65
80
100 125 150 200
1.5 2,1 4,0
1.4 2,2 4,2
13 10 2.3 2.3 4,4 4.5
10 2,4 4,7
1,0 2,5 4,8
1.3 2.4 4,8
1,3 2.3 4,5
13 1.6 2,1 2,0 4.1 35
1.5 2,8
1,5 10
1.7 3.3
1.9 3.5
2,0 3.7
2,0 3,9
19 3,8
1.7 3.3
2}
1.3 2,0
1,9
1,5
1,5
1,4
14
1.3
1.2
1.0
0,9
0,8
Angle valves
10
Type Boa
DIN 12
a
25
1.4 d/d2
16
18
Nan-return flap b (clack) valve
Figure 2. Flow resistance coefficient ,; a of shut-off slide valves with reducers, b of valves and flaps [17].
1.5
Components of Thermal Apparatus. 6 References
100
80
60
40
20
Temperature difference in K
10
12
Change in length in mm
14
Figure 3. Change in length of various materials as a function of temperature. PEh ~ polyethylene hard.
References HI Fundamentals. [I] VOI-WarmeatIas, 6th edn. VOl,
K. Beriihrungsdichtungen, 2nd edn. Springer, Berlin,
Diisse1dorf, 1991. - [2] Ahmad S, Linnhoff B, Smith R. Design of multipass heat exchangers: an alternative approach. Trans ASME/J Heat Transfers 1988; 110: 3049. - [3] Rummel K. Die Berechnung der Warmespeicher auf Grund der Warmedurchgangszahl. Stahl und Eisen 1928; 48: 1712-25. - [4] Hausen H. Warmeiibertragung im Gegenstrom, Gleichstrom und Kreuzstrom, 2nd edn. Springer, Berlin, 1976. - [5] Grassmann P. Physikalische Grundlagen der Verfahrenstechnik, 3rd edn. Salle, Frankfurt, 1982. - [6] Martin H. Warmeiibertrager. Thieme, Stuttgart, 1988. - [7] Glaser H. Der thermodynamische Wert und die verfahrenstechnische Wirkung von Warmeaustauschverlusten, Chem 1ng Techn 1952; 24: 13541. - [8] Gregorig, R. Warmeaustausch und Warmeaustauscher, 2nd edn. Sauerlander, Aarau, 1973.
1975. - [16] Mayer E. Axiale Gleitringdichtung, 7th edn. VOl, Diisseldorf, 1982.
H2 Apparatus and Piping Components. [I] AD-Merkb1atter: Richtlinien fur Werkstoff, Herstellung, Berechnung und Ausriistung von Druckbehaltern. Loseblatt-Sammlung. Heymann, Cologne. - [2] Klapp E. Festigkeit im Apparate- und Anlagenbau. Werner, Diisseldorf, 1970. - [3] Titze H. Elemente des Apparatebaues, 2nd edn. Springer, Berlin, 1967. - [4] Schwaigerer S. Festigkeitsberechnung im Dampfkessel, Behalter- und Rohrleitungsbau, 4th edn. Springer, Berlin, 1983. - [5] Tochtermann W, Bodenstein F. Konstruktionselemente des Maschinenbaues, pt 1, 9th edn. Springer, Berlin, 1979.[6] Richter, H. Rohrhydraulik, 5th edn. Springer, Berlin, 1971. - [7] Zoebl H, Kruschik J. Stromung durch Rohre und Ventile. Springer, Vienna, 1978. - [8] Grassmuck J, Houben KW, Zollinger RM. DIN-Normen in der Verfahrenstechnik. Teubner, Stuttgart, 1989. - [9] Richarts F. Berechnung von Festpunktbelastungen bei Fernwarmeleitungen. Heiz Luft Haustech 1955; 6: 220. - [10] Merkblatt 333. Halterungen und Dehnungsausgleicher fiir Rohrleitungen. Beratungsstelle fiir Stah1verwertung, Diisseldorf. - [11] Wagner W. Rohrleitungstechnik, 2nd edn. Vogel, Wiirzburg, 1983. - [12] Armaturen-Handbuch der Fa. KSB, Frankenthal. - [13] Friih KF. Berechnung des Durchflusses in Regelventilen mit Hilfe des k,-Koeffizienten. Regelungstechnik 1957; 5: 307. - [14] Ullmanns Encyklopadie der techno Chemie, vol. 4, 4th edn. Verlag Chemie, Weinheim, 1974, pp258-67. - [15] Trutnovsky
H3 Types of Heat Exchanger. [I] Grassmuck J, Houben KW, Zollinger RM. OIN-Normen in der Verfahrenstechnik.
Teubner, Stuttgart, 1989. - [2] Klapp E. Apparate- und Anlagentechnik. Springer, Berlin, 1980. - [3] Becker J. Ausfiihrungsbeispiele fiir Wiirmeaustauscher in Chemieanlagen. Verfahrenstechnik 1969; 3: 335-40. - [4] Shah RK. Classification of heat exchangers. In Heat Exchangers, Advanced Study Institute book. Hemisphere, Washington, 1981, pp 9-46.
H4 Condensers and Aftercoolers.
[I] VOI-WarmeatIas. Berechnungsblatter fiir den Warmeiibergang, 5th edn. VOl, Diisseldorf, 1988. - [2] Dornieden, M. Zur Berechnung ein- mehrgangiger Rohr-biindel-Kondensatoren. Chem Ing Techn 1972; 44: 618-22. - [3] Schrader H. Einfluss von lnengasen auf den Wacmetibergang bei
der Kondensation von Dampfen. Chem Ing Techn 1%6; 38: 1091-4. - [4] Grant !DR. Condenser performance the effect of different arrdngements for venting non condensing gases. Brit Chem Eng 1969; 14: 1709-11. - [5] Chen SS. Flow induced vibration of circular cylindrical structures. Hemisphere, Washington, 1987. - [6] TEMAStandards of Tubular Exchanger Manufacturers Association, 6th edn. New York, 1978. - [7] Kopp JH. Ober den Warme- und Stoffaustausch bei Mischkondensation. Diss. ETH. Juris-Verlag, ZUrich, 1965. - [8] Forgo L. Probleme der Mischkondensatorkonstruktion bei Luftkondensationsanlagen System Heller. Energietechn 1967: 17: 302-5. [9] Schroder K. Das neue Dampfkraftwerk. BrennstWarme-Kraft 1963: 15: 140-2. - [10] Berliner P. Kiihltiirme. Springer, Berlin, 1975. - [11] Vodicka V, Henning H. Oberlegungen zur optimalen Gestaltung eines Nass-/Trockenkiihlturms unter dem Gesichtspunkt der Minimierung des sichtbaren Schwadens. Brennst.-WarmeKraft 1976; 28: 387-92. - [12] Schlaich J, Mayr G, Weber P, Jasch E. Der Seilnetzkiihlturm Schmehausen. Bauing 1976; 51: 401-12. - [13] Ober den Windeinfluss bei natiirlich beliifteten Kiihltiirmen. Bakke-Diire: Die aktuelle Information no. 10-5/1976. - [14] Merkel F. Verdunstungskiihlung. VOI-Forschungsh. no. 275. VOl, Diisseldorf, 1925.
Machine Dynamics D. Faller, Frankfurt-on-Main; K. H. Kuttner, Berlin; R. Nordmann, Kaiserslautern
•••••LI Crank Operation, Forces and Moments of Inertia, Flywheel Calculations
K. H. Kuttner, Berlin
(3)
Forces and moments acting on the piston and other power unit components are used in calculations involving the power unit (see FlO.3), its smooth operation, crankshaft torsional vibration [11 (see J2), environmental mass effects, and resonance phenomena [21.
1.1 Graph of Torque Fluctuations in Multi·Cylinder Reciprocating Machines This is affected by type of construction, offset of cranks, the oscillating power-unit masses and cylinder pressure, as well as the firing sequence [3] in the case of engines. Pressure Graph. This is obtained [4] as p ~ I( 'P) from a cathode ray oscillogram or as p ~ I(x). The non-dimensional
quantity expresses the distance
X
travelled by the piston
in tenns of the crank angle
first three terms will be sufficient. Its steps A'P numbered k ~ 0 to i at the period 'PA ~ 360'aT for the working cycle are then given by A'P ~ 'PAlk. The number of strokes aT ~ 2 in the case of a four-stroke engine, otherwise aT ~ 1. For a compressor with 'PA ~ 360', A'P ~ 6° for k ~ 60. Its TDC (top dead centre) lies at t ~ 0, i.e. at k ~ 0 or 60, the BDC (bottom dead centre) at t ~ 2 and k ~ 30. For new designs without diagrams, the latter must be derived from the ideal process. In the case of compressors, this can be accomplished with the help of the simplified Seiliger process [;] by replacing compression and expansion by polytropes, and in the case of pumps by the suction and pressure characteristics. If the media are gases, the volumes are given by the p, V diagram [5]: Engine: Compressor:
V~
[1/(e - 1)
V~ [eo
+ t12]
+ t12] . Vh ·
. Vh ,
Total Torque. This applies to machines with z cylinders and equal piston and crankshaft offsets as in the case of engines [6], where the angle 'P is measured with respect to the angular position of the crank 1. For an in-line machine (Fig. Ia) this is (4)
The
total
moment
applies.
Torque. With the piston force FK ~ Fs - Fo and the difference between the gas and the oscillatory inertia forces (as FlO.2.1 to FlO.2.3), where in most cases two terms are suffiCient, the torque of a power unit is given by
repeats with
the periodicity
'Pp ~ 'PAlz, i.e. with the angle between two cranks. Here,
the torque fluctuations decrease with an increasing number of cylinders. For fan-type and radial machines and angles YF ~ 180° Iz and 'P, ~ 360° Iz respectively between the neighbouring cylinder centrelines, the following applies: M d ,,, ~
L Md ['P + (k -
l)y].
(5)
Equal firing intervals are only possible for radial engines with an odd number of cylinders. For the W compressor with YF ~ 60° (Fig. Ia), the following applies: M dw , ~ Md('P) + Md('P - 60°) + Md('P - 120°).
In the case of mUlti-stage compressors, the torque function of the individual stages differ. Owing to the coupling between the power unit and the driven unit, the torques of both must be taken into account (Fig. 1). In order to investigate the vibrations, the harmonic analysis of the diagram of Fig. Id must be applied.
Mean Torque, The following equation applies:
'p
(2)
In the case of double-acting cylinders and sytumetric diagrams, flO Eq. (19)
PKS ~ (2 - OPus
with period 'PA ~ 360' aT and the forces equal to zero in FlO.2.3. At increasing revolutions, the inertial forces at first reduce the gas forces only to exceed them later, a situation which also affects the torque fluctuations (see FlO Fig. 8 and Fig. Ie).
M dm
~ 2. JM 'Pp
dw ,
d'P;
(6)
o
it is detertnined by means of numerical integration. In the steady state it is equal to the mean value of the coupled machine and independent of the inertial forces.
Flywheel. This compensates for the maximum of individual energy fluctuations (Fig. 2):
Machine Dynamics. 1 Crank Operation, Forces and Moments of Inertia, Flywheel Calculations
0
150
80
100
I
/-'/" \
,
\~
120 I
I
80
I
',I
"-
,
E
~-100
:i!
40
,, I
\
E
z
c
\..1
a
-300 0,
-f
'"
-ZOO
60'
-80
360'
-240 E
-160
-18°0'
z
60 50 -5000~'----::'60':-:-'--:-:-:C:------:=---::-"-::---::::-;-------::
E
z
'"
b
"'~
40 30 10 10
60'
liD'
180' cp
140'
300'
360' 140'
M,
'fl,
I
180'
I
110'
I I
e:
I I I
60'
I
0' -60'
d Figure 1. Torque diagrams. a Single-stage W compressor, 'YF = 60°. b Four-stroke engine with single..stage V compressor with two banks of two cylinders each; MdmM = - MdmV , v. c Two-stroke engine idling. d Harmonic analysis of the torque of a two-stage compressor with spectrum of the instantaneous torque amplitudes and their phase angles Mk = ..JM~a + M~b or tan 'Pk = Mak/Mbk respectively.
w, =
7'
(Md - Mdm)dcp .
(7)
-k
Here the CPk or 'i'k+, occurs at the point where Md = M dm • Moment of Inertia. It follows from the energy conservation law, with Wsmax = ] (w~ax - w~jn)/2, the mean value Wm = (wm = + w min) 12 and the irregularity factor
8 = (wm~ -
w min)
1Wm that, in accordance with Table 1,
w,
W,
of which also includes control [5]. Reference values for four-stroke engines [6] are obtained from the intlicated power Pi and the constant k from Table 2, using -k Pi } - 8(nllOO)"
(9)
Thus at constant power, the moment of inertia decreases with the third power of the rate of revolution, the number of cylinders and the irregularity factor.
(8)
Design. The flywheel (Fig_ 3) consists of k discs of
This also takes into account the coupled machine and power unit, and is supplied by the flywheel, the function
width b k and outer and inner diameters Dk and d k respectively, and has a density p. Its mass and moment of inertia are therefore given respectively by
}= 8w;;, =
4'ITz
8n z '
1.1 Graph of Torque Fluctuations in Multi-Cylinder Reciprocating Machines
250r-----,-----,------,------,
E
l
= !:';'
i
-1000 \------y---j-----+--nrsT'hWT'J-----I
Figure 3-. Disc-type tly",rheei.
a Example. Torque diagram of a single-stage compressor with trunk piston. This is based on the simplified calculation of the pressure curve from the p, V diagram (Fig. 4).
I
w,
o£
Back Expansion 3-4. Using Eqs (1) and (2) it follows from the
polytropic equation
I
that
1111
b
Figure 2. Determination of available power: a torque, b energy curve.
It occurs between Pl and PI =P2/t/J, or between = 2e o (!)JUri ~ 1), respectively.
~=O
and
~ha
Table 1. Reference values of the irregularity factor Ship'S propeller Pumps and blowers Machine tools Piston compressors Vehicle engines Generators: - AC
- DC
Aircraft engines
Induction 4-1. This takes place along the isobar PI between
1;= I;h"," and 1;= 2. 1/30 1/30 to 1/';0 1/50 1/';0 to 1/100 1/1';0 to 1/300
Compression 1-2. Csing the polytropic equation pV" where VI == V, + Vh == (l + Eo) Vh, the following is obtained:
p-p ~
1/12'; to 1/300 1/100 to 1/200 1.\000
I
=Pi V7,
(.~)" eo + ~/2
This covers the region from PI to Pl == !)JP1, i.e. from
Exhaust pn:ssure
Table 2. Constant k in kg m 2 j (kW min:;) for four-stroke engines
2-3. This occurs between Pl.
~== ~nml
and
~=
0, at constant
Moment. 111e
force F~ == (P ~ Pa)Ak exerted by the medium is obtained from flO Eq. (18) with the pressure P=/('P) of the
Number of cylinders Diesel engine Spark-ignition engine
17.:; 7.2 4.3 0.92 1.63 O.~4 0.73 0.49 6.0 2.5 1.3 0.5 0.24 0.12
(10)
1
2.0
1.5
10
5-
~oxlr
where Dki-l = dko Accordingly, the outer ring has the greatest effect and accounts for about 90% of the moment of inenia for disctype flywheels and 9;% for those of spoke rype [7, 8]. In order to utilise material as effiCiently as possible, the outer diameter should be as large as the stresses due to centrifugal force allow. The limits lie at peripheral speeds of u = 50 mls for grey cast-iron flywheels and u = 75 mls for cast-steel flywheels.
0
1
l'
11
20
18
hp/15"
B
10
16
l'
11-0,1'
11
5--
Figure 4. Determination of the pressure curve of a piston compressor.
Machine Dynamics. 1 Crank Operation, Forces and Moments of Inertia, Flywheel Calculations
individual parts of the operating cycle. With the inertial forces as in FlO Eq. (22), Fo = morw' (cos cp + A cos 2cp), Le. with F, =F, - Fo
Using the non-dimensional constants
as in FlO Eq. (27), the moment is obtained from Eq. (3) (see Moo in Fig, 1).
(12) the resultant and its angular position are calculated from
Mere, = ;/M;' + M;,
1.2 Forces and Moments of Inertia In multi-cylinder machinery, resultants are found by vector addition of the forces of each power unit and the moments produced by them. These are the rotating forces F, = m,ror, or the first-and second-order forces F, = morw2 cos cp =P, cos cp and FII =AP, cos 2cp, respectively, obtained from FlO Eqs (21) and (22), as well as the moments M" M, and Mil' Their addition is carried out graphically [9] or analytically in accordance with the position of the cranks and the position of the centrelines. In the case of engines, the masses m, and mo of the power units are constant according to FlO Eqs (25) and (26), as are the distances a between the cylinders and the difference L\"k of the crank offset. Their centroidal axis SS lies in the middle of the crankshaft (Fig. S). Multi-stage compressors have different pistons [10] and hence masses mo. These forces and moments cause vibrations in the power unit and the machine [11]; in particular, torsional vibrations in the crankshaft [12].
1.2.1 Aaalytic:al Methods These are particularly suitable for the programming of computers [5].
In-Hoe Machines. Using the stroke number tl-r, the distance b k and the offset of a k crankshaft of z cylinders (Fig. Sa) is given by
"k
"k = (n -
1)360° aTlz,
(11)
b k = [0.5(z + 1) - k]a = v.a.
"k
Rotating Moments. Their vertical and horizontal components are given respectively by
M"
k 1
I
;/c;, + c;2 are
Moments of tbe mtb Order. With the force amplitudes Pmk = Fmk/cos (mcp), as in FlO Eq. (21), which act along the centreline of the cylinder,
The maximum is obtained from
dMm~Jdcp
=0
depending on the angle cp for its calculation and direction. If the pistons, i.e. the forces Pmb are identical, then using the constants
the instantaneous values and their maximum respectively are given by Cm = Mmres
;/r.., + r..2
= Pm a(cm1 cos mcp -
Cm2
sin mcp)
(17)
They occur at the crankshaft angle (18)
The index k = 1 to z designates the power units along the crankshaft starting at the coupling and the index n = 1 to z determines the angle and is counted in the direction of rotation.
M= = F, lbk sin( cp + "k)
tan X = M=IM" = c'2Ic,,; (13)
the rotating moments using c, =
and
Forces. For these b k = aVk = 1 applies in Eqs (11) and (16). This means that the constants are km' =
2: cos ma
k
and
km2 =
2: sin ma
k•
(19)
Most Advantageous Crank Sequence. The forces vanish if the mth order crank angular position diagrams with the
= F, lb, cos(cp + "k),
3 I
I
i
I
I
i : . .
i
l-~-1 I 1---3 -I3----i
Vk! a
S
F l
90'
b
Mrres
M5:1
Figure S. Seven-cylinder in-line engine: a crank schematic with first- and second-occter stars, b vectorial determination of the resulting rotating moments.
1.2 Forces and Moments of Inertia. 1.2. I Analytical Methods
6(~lli5 -- I-6
........
2
-
--
__ 4 \ J
3
--
3
----
W 6 __
I _-
r \
2
-
z+1
--
3
---) 7
Figure 6. Most favourable crank sequence for two-stroke engines with even and odd numbers of q'linders.
angles rn",. are symmetrical (Fig. Sa). Two-stroke engines (Fig. 6) have their smallest moments if their firstorder crank angular position diagram passes through the sequence 1, z, 2, z-l, n, n(z-n+ 1) [9,13]. In fourstroke engines the moments cancel each other if the angle "'. and the length of their lever arms hk are the same for any two cranks.
Engines. In the case of two-cylinder engines, the centrelines of the power units A and B displaced by the width of a pushrod fonn the angle y= 'PA + 'I'B (Fig. 7). The vertical and horizontal components of the first-order force are then Fix = (FIA - FIB) sin (y/2) FlY = (FIA
tan
0:,
Pia/PI
F'h/P,
FllafP"
FUb/Pli
1.867 1.707 1.50 1.259 1.0 0.5 0.0
0.134 0.293 0.50 0.741 1.0 1.50 2.0
1.673 1.307 0.866 0.411 0 0.5 0
0259 0.541 0.866 \.176 1.414 1.50 0
resent the semi-axes of the ellipses in accordance with Eq. (21) and are therefore given by F I, = 2PI cos'(y/2)
= Fix/FlY,
and
F'b = 2P, sin'(y!2). (22)
Their position is vertical or horizontal respectively and for y < 90° Flo is the maximum and F'b the minimum (see Table ~). For second-order forces with the component FilA = PIIA cos 2'PA and FIIB = P IIB cos 2'PB' the following apply: FUx = (FUA - FuB) sin (yf2)
and
F lly = (FilA + FUB ) cos (y/2).
+ FIB) cos (y/2)
and
30 45 60 75 90 120 180
0
and
respectively, because 'PA = y/2 - 'Pk and 'I'B = y/2 + 'Pk, with PIA = PIA COS 'P and FIB = P IB cos 'PBThis makes the resultant and its angular position FI = ...JFfx + Ffy respectively.
PI = morw and PH = API 'Y in
5-" 7
4
Extreme values of the inertial forces of V machines
Table~.
7
The resultant and its position angle are given by F" = (Xn = Fnxi Fny respectively. For pistons of equal mass,
;JPfIX + F~y and tan
Fux = 2Pu sine yf2) sin y sin 2'Pk,
(20)
Fuy = 2Pu cos(y!2) cos y cos 2'Pk.
Their extreme values, which occur at cos 2'Pk = 1 and 0 respectively, are
If the piston masses are equal then Fix = 2P, sin'(y/2) sin 'Pk
and
f;y = 2P, cos'( y/2) cos 'Pk.
(23)
(21)
Fu. = 2Pu cos (y/2) cos y FUb = 2Pu sin (y/2) sin 'P •.
At y=90°, Eqs (20) and (21) show that F,=P, and = 'P. The frrst-order forces can be balanced by counterweights at the countennass extensions. Their extreme values occur when cos 'P = 1 and 0 respectively. They rep-
and
(24)
"'I.
A/7~8 S
A1 81
AZ 8Z
A383
8
A
s a
a 1--
A
b Figure 7. V machine: a arrangement of the power units, b determination of the first-order force from its components, C vectorial determination of the second-order force.
8
b Figure 8. V in-line machines: a schematic construction and firstorder moments, b second-order crank angular position diagram with moments.
Machine Dynamics. 1 Crank Operation, Forces and Moments of Inertia, Flywheel Calculations
Table 4. Free forces and moments of inertia of various cylinder arrangements (compiled from [3, 6, 11, 12, 16])
8,
12,'
2 cylinders in line' : 2 cylinders In Iln' 3 '2 cylinders oppo~lldl 2 cylinders,
Designalion First,order crank angular, position diagram Schematic of crankshaft
I
Il
CD
(TI)
l~l
Free moments (no compensation) First-order
Seccld,order Higher order free forces H'gher order free moments
Counterwelghls ~fo~~I;~~eber
D·
Pj
oo "~I
o·
lIP •• P Fr·a.5fj
Torsional vibrations, critical Torsional vibration behaviour Genera! dynamic behaviour Assessment
05,15,1,15;" good serviceable serviceable
Designatron
] cylinders in line
large
I~~~l
].110' cranks Crankshaft constructron 140' -1,0' FlriQfl interval Free forces (no compensation! Firsl,order 0 0 Second,order l~ree rlomeots Ina ccmpensatlon) V3aP r Vl'G,P a Higher order free forces ]P. Higher order free moments JiJaP N
'
,
i
:!,
JeD, \.J) l
11-
° °0 4IP ·P•• ,,) . j------'
«F(,,'hP j )
0
! 'J '2: I
j
b,F a yzp., yzp~ b,yz(lP.; b,yznP. 1 lh(F; •.. P,I large see 110, 131
moderate moderate
4 cylinders, 2.180' v ,',Z,3' 4cyhnders opposed'Z,J
rS]
,
,;
2
I
4 cranks 180'-180'-180'-180'
1cranks 180' _180' ,180' lBO'
0 0
Yl aP, ~
o·pu
°
;P:P TI
0 4
1b F., 2b, F0
lb,P"C""l::.b-'c'P-'~~---l
IJzFr + • 'hP J moderate
small
(,----t--~,-----t--~~- ~-r--'
<<(Fr,'hP r )
moderate
~~o~ 45,
v,and h I,OP, v, OPa, h 1,41,Pa
v1,5 P",l 019] P, v, and h, a,B55Pa
l:a;~
I.P n
N
1 crank 450' - 170'
serviceable serviceable
1,1
//1
Jlv
I
"
I
~ b
i
b
I
moderate moderate
4 cylinders in line
:
,!,
i b·f
2.1 cranks 90' offset 4 cranks 180'_ lBO'-180' IBO'- I.L 90' - 90' - 90' _90'
I~~~~~~~~der
(
I']"
)08" 1
1,13
(TI) '2
bI, I D F, b,F a ,b F, Ol65P.,Ol55P. JiJP. Ol65,bP.; Ol65,DP. b,p.; b"ld3P~ 1 1 i ljdf;+ "P,) '1zIF,·",p,1 large large see 110, 1]1 see 110, 1]1
serviceable serviceable
lJ~1
,
Torsional vibrations, critical Torsional vibration behaviour
1
i
:I CD" 'I'
0
b·(Pl'1+ P\J .... )
4 cylinders in line
1,1,3.1
1
magnilude efforl
I
large 1,1,],,, good serv'ceable serviceable
Frrst'order crank angular position diagram Schematic of crankshaft
Coutllerwerghls' usual rumbe r
0 1
I
1 crank--- 1 crank 405' - ]15' 410' - ]00'
~1707P"h'0'19]P' vl]lPa, hOl4P,
DP, bP a
"2,3,4
12
$,J"t \I
__ 0
1
~~ge
(TI)
~l
lP, lP" 0 0
"2,3.4
2 cylinders, 60' V 2 cylinders, 90' V Il
~'I,~
~
D"
2 cranks I ' ]50'- ]50'
2 cranks ]50' - ]50'
0
,IIP.,P
o I
J!llrfL-
Ay' 2 cranks Crankshaft construclion lBO' - 540' Finnq interval Free forces (no compensation! Flrst,order 0 lP a Second,order
'2
~
45':';)'
f,.O.5P r large
4
1,4;5, moderate
I~O~d 8
moderate
medium
moderate
poor
«(Fr,O.SP j )
meolum gooo ~~miC behaviour __~~~----~~------_f~~~----~p~oo~r------~~go~o~d-------1 moderate ~~~~ medium
Designation
4cylinders 2~90' Ii
I·,U
4 cylinders 7.90· (J,J,,6 4 cylinders 2.90·
t.l·
4 ,5
I.
cylinders 500 v
Dod
1.3)
Frrst'order crank angular, position drag ram Schematic of crankshaft
5 cylinders rn fine
U,3
I
'@' J
2I I 11
I]', , "
I
I
~ ,," ,',' "
1cranks Crankshaft construction Firing interval 90' -180' - 170'- 180' Free forces (no compensation! First-order
0
Irs;eec8l0:;;nd;;,0;erOideSir;;C;:o;;e;;;at;C;;;tV,-"OPn, h lYZ PIT ~~~~~~;~~t~I", 0"' aF, "teWU'U'UCI
lbI a
1cranks 1cranks 90' offsel 1.110' cranks 50' offset 90' - 170' 90' -""1,,,70,-' _-+",180,-'_-9",0'_-
l1F, l'bF,
5.71' cranks
5.144'
____-+_____0____+"2V3 Pn 0
"I,vlF"b/z([F, 2oF,
a'P, b,P a
0,449,a,P, 4,9B,o,Pn
Torslonalvibrallons, critical 10.5,1,.5;15. [10,IJI~r:~~,5', ..[[IO,IJJ 105,1,'1,5,1,5" [IO,1l1 1;(,5, 1,1,5,1,5,3.5;4,,, TorSlo~al vibration behaViour Igocct' ~. _ I 9800 moderate moderate c,G"e"'ne"'ra"'I"'dy"n"'am"'ic"'bC'e'hCa-vC'io-ur-+m'-o'Cd-er"'/at-e- - - - - moderate ·--·--+m"'o"'lde-r"'lale---- ---l"m"'o""d""er""al"-e-----+'m"'od"'e"'ra"'te'------1 ---+mcc'o-';d'-er=Cale'- - - - -f;nod~--' +m:coCCde"'ra·-::le=----~.:,cm"'od7e:::ra:::te'-----'·Assessmenl moderate Continued
1.2 Forces and Moments of Inertia. 1.2.1 Analytical Methods
Table 4. Continued
Designation First-order crank angular, position diagram Schematic of crankshaft
6 cylinders in line
'(19'
]
•
6 cylinders in line
I.U,5
i .•
1
,@,
I 'J'
'
I
15~J'
5 ;6
.JI '"
I
I
:5 :5
5 'I
I
a
0
0
0 0
0
2n·a·F, 6P.
'/
6P.
Fr+'hP!
ll'l F.
0 6
lha·F IY ;
'~ ,
Crankshaft construction Firing interval Free forces (no compensation) First-order Second·order Free moments (no compensation First·order Second·order Higher order free forces H'gher order free moments Counterwerghts: usual number magnitude effort TorSional Vibrations, critical Torsional vibration behaviour General d namic behaviour
Assessment
Designation First-order crank angular, position diagram Schematic of crankshaft
I
.,sA
1*30
~
a
lllF.
ood
U.3
"lls: 6
8 :7 :
~ ~,
I
serviceable
serviceable
serviceable
8 cylinders 4.90' ;'.3.", 8 cylinders 4.180' V
'.,),1,
I'
' '
,
-1
GID
17 ]
I,.
7,8
!~ :'..IF I
large 4,8,12, . moderate serviceable
Assessment
serviceable
0 0
0 0
0 0 8P. (Fr+'hP[l
0 8
ViOa·F,
0
medium
4:8,12, ..
medium
good good
7 cylinders ,n line
'@'J i.
-
5
6 ,
5 :6 (
1
~ ~
,a:
I I
I'" '
I
8cyllnders,opposed
1),3
17~J6 'S
r-,-6
0 0
I
0
40· Fn
o·F"
t6
~
@7 ,8
J
I
]
( :8
_
,---
7
F:+lhP j
serviceable
I
4. ]0' cranKs, 90' offset 8·90' 0 0
0 0 0 4·b·P. , '·D·Pij, 8
moderate 4,8,12,. serviceable good good
,a
'~T
~(/\
0 0
I
I.D
'I~
I~ ~I 'J, --15 '
4 < IF,"11 P,I
small 2,4,6, .. medium moderate
.1,5
8cylinders 60' v
,I
4.180' cranKs, 90' offset 8.90'
I
5
~l:J~
~:,I' I B.90' cranks t.4\' offset 90'· 90'· 90'· 90' 45'· 90'· 90'·1l5' two·stroke 8·4\' 0 0 0448·a·P,
4cranKs.180'offset 4.180' two stroke
b~
1.1.3
8 cylinders In line
l~J
J
';
~
a
40·F" , 8
1},J
'I __ '~-8 J
c~
4VlF. 2'f).b·F"
I Fr .. Iii p)
!
IlJ 6
4.180' cranks,2.90'offset 4cranks, 90' offset 8·90' 8·90'
rankshatt. construction Firing interval Free forces (no compensation) First·order Second·order Free moments (no compensatlonl First-order Second·order Higher order free forces Higher order free moments Counterweights' usual number magnitude effort Torsional vibrations, critical Torsional vibration behaViour General dynamic behaViour
6 <(lhFr+lliP!)
18
1584
B
a
0 0 0 9,845·aP., 0,26J·a·P. 15·a·P. 7 8 Fr+ liz PI IF,.IIzP,1 large large 1:2.5,35,4.5,6,7,8, 2:2.5,3.5,4,4.5, ... moderate moderate serviceable serviceable serviceable serviceable
6
good 0.5, IS, 25,35,45, ... moderate
2669
~
small ]. 6, 9, .. good good good
0167·aP, 1006· a P
0 0
],6,9, . good
15
18
0 0
0 0
<(1hFr +l/tPr)
poor
moderate
27 IJ
1
3M3 a F", 31z(J ·b·Fij ]·bF ij 6 medium
IS, J, 45, .. good
J6Ei9'5
i2'(ia-F
'hFr+'hP j
'hFr+'hPI medium
0 J.b·P.
0 0
IF.
ij
0 0
] cranks 110' offset 7051.4]' cranks 120'·120'·60'·120'·120'·180' )·102.86'
1.5J'l·a·F , t5 Vla·F,
'{i·a·F, )'6·aF,
8cylinders ,n line
l~,
0 0
0
'iJ
b
1
>, "6 -~'~ :'.I, i , '1'·,'"
J cranks 120' offset J cranks 120' offset 150'·90'·150'·90'·150'·90' 6.120'
)'6·aI. , 'h'(i. b I 5
II5~J,~ ·'.l
,,0
I
-'-f,
0 0
serviceable
/.
.i I
'·i
"~ 'j '-., "!
I
lhFr+'hP 1
"@,, I ,2
25~:
J
f
~ ju~
medium J, 6, 9:.. serviceable serviceable
First-order crank angular, position diagram Schematic of crankshaft
J<' /5/
,/
6
Fr+ IIzP r
't
2
J/z·b·f VI
large l,5:9,. moderate moderate moderate '·23
.•
'@'
f
]\'IF.
3/z·b-F. 6
6 cylinders MO' ;,-,.3., 6 cylinders J. 120' y' . 6 cylinders J. 180' Y..
J
I
,
)(\~//
\2
3/z. o·f n
Designation
/!
'I
6 cylinders opposedl.1·3
]·IBO'cranks 110' offset 6·180' cranks 110' offset 6.120' 6.120'
0
lho.f n
«jFr+'hP!)
15, l, 4.5, 6, ... moderate moderate moderate
I~:'~ /' /1
,
]
1
0 0
smatt l, 6,9, good good good
medium
.'@'
,
6.60' cranks 6.120'
Q
0 6
15
'~~ J\" j,~ "I
I
6.60' cranks 6·110' cranks 110' -120'-IBO' _120' _120'_60' 6·110'
t
1~,2,3
6 cylinders 60' v I
J
, ~ ~ '
' ~.1.l
6 cylinders 60'V
16
I
,a,
Crankshaft construction Firing interval Free forces (no compensation) Firsl-order Second-order Free moments (no compensation) Flrst·order Second-order Higher order free forces Higher order free moments Counterweights: usuat number magnitude effort Torsionat vibrations, criticat Torsionat vibration behaviour General dynamiC behaviour Assessmen1
1.2,3
13.054,0.8181 a·P, 0 4J'lP. 2·b·P. 8 Fr+lhP J
I
medium 4,8,12:.. serviceable serviceable serviceable
Machine Dynamics. 1 Crank Operation, Forces and Moments of Inertia, Flywheel Calculations
Here, FIla is the maximum and FUb the minimum, when Y < 60° (see Table ~). The rotating forces are given by flO Eq. (26) as
respectively. This leads to the resultant and/or the maximum firstorder moment
(25)
From Eq. (14), the vector of the rotating moment with tan- 1 (0.2407/0.116) = 64.28° has the angular position
V In-Line Engines (Fig. 8). If the piston masses are equal the components of the first-order moments are, from Eqs (17) and (22), M,x
M,y
= 2P,a sin2 (y/2)(c" = 2P,a cos' (y/2)(c"
sin cp + C,2 cos cp),
(26)
cos cp - c" sin cp),
with CII = crl and Cl2 == erl Using Eq. (17) with m = II, the following apply for the second-order moments: M"x = 2P"a sin y sin (y/2)(c", sin 2cp
+ C"2 cos 2cp),
a,
M,
(27)
M"y = 2P"a cos cp cos (y!2)(C"2 cos 2cp - C"2 sin 2cp), The resultant and its angular position follow from Eq. (13). The extreme values of the first-order moments are given by M"
and
M'b
= 2P,ac, cos 2 (y/2) = 2P,ac, sin2 (y/2),
= 90° + 25.72° + 64.28' = 180°.
The maximum first-order moment occurs at the crank angles cp=-64.26° and 11S.75°, i.e. after rotation of crank 1 by 90°. The moment is zero at cp = 64.28 ° and 154.28°. For the second-order moment, Table S is recalculated for 2ok • From Eq. (16) this yields cm = 0.7862 and C IIol = 0.6270, i.e. CII = 1.006 and klil = kll.l = O. The maximum second-order moment is MIlre-;rnax./(A.Pla) = 1.006. It occurs at tan- 1 ( - cm/cm) = 38.57° when cp = (90 - 3857)° = 25.710, i.e. in the position shown in the drawing. In the graphical solution (Fig. Sb) as under J1.2.2, it follows from the moment diagr.tm that
= 2F,a(3 cos 64.28° + cos 38.57' =
2 cos 12.86')
0.2672 F,a.
Here the vector At,..", must still be rotated clockwise by 90°. No forces occur, because krl = kr2 = kill = kll2 = 0, Le. the crank stars are symmetrical
V In-Line Engines. If the vee-angle )' = 60 0 , the extreme values of the first-order moments are given by Eq. (28) as MIa/(P1a) = 2 . 0.2672 cos..! 30° = 0.4008
(28)
M,./(P,a)
and
= 2·0.2672 sin' 30° = 0.1336,
with c, = Vc7, + c72' For the second-order moments using clIl and CU2 as given by Eq. (19) and with c" = vei" + eil2 the following apply:
and the second-order moments in accordance with Eq. (29) as
2Pnacll cos l' cos "1/2 M"b = 2P"ac" sin y sin (y/2).
For the vee-angle 0=90°, the corresponding calculations yield
Mua:;;
and
(29)
=
0.8712.
M,,/(P,a)
The rotating moments are calculated using Eq. (14) and the mass m,y, as in the case of the in-line machine. Table 4 (see pp. J6 and ]7) shows the forces and moments of inertia of the most important engine types. Example. Forces and moments of inertia for an engine with crank sequence 1, 6, 3, 4, 5, 2, 7 of simple in-line and V in-line construction with 60° and 90° V angles.
= M,./(P,a) = 0.2612:
Mua/(AP1a) ::::; 0 MUb/(AP1a)
=--12
Fan-Type and Radial Engines. Here, the power units (Fig. 9) are distributed over a half or a complete circum-
ference. The angle between two centrelines is YF = 180° /z and y, = 360 0 /z respectively where they are offset by the
In-Line Engine. In accordance with Eq. (11) and Fig. S, the crank offs(:t and lever arm for z = 7 cylinders are Uk =
(n - 1)51.43°
and
Vk =
hk a = 4 - k.
The crank angle '1'= 51.43°/2 = 25.72°. From Table S, calculated by using these values and employing Eqs (12) and (19), it follows that Crl = 0.1160 and Cr2 = 0.2407 or Cr = 0.2672 and krl = kr2 = 0 1
J
5
1.1.11.1.
r=L Table S. Calculation of forces and moments of inertia of an in-line engine (see example)
n k
Ok
in °
1 0.0 2 6 51.43 102.86 3 4 154.29 205.72 5 6 257.15 7 308.58
cos
Ok
1.0 0.6235 -0.2225 -0.9010 -0.9010 -0.2225 0.6235 0 =
sin
Ok
0 0.7818 0.9750 0.4339 -0.4339 -0.9750 -0.7818 0
k-yl
=
k-Y2
11.
1\ cos
-2 1 0 -1 2 -3
3.0 -1.2470 -0.2225 0.0 0.9010 -0.4450 -1.8705
0.0 -1.5636 0.9750 0.0 0.4339 -1.9500 2.3454
0.1166
0.2407
=
Ok
C-yl
1\ sin
a
Ok
= C-y2
12
11L_
r=rrL 45 b Figure 9. a Fan-type machine. b Radial machine.
1.2 Forces and Moments of Inertia. 1.2.3 Compensation of Forces and Moments
Table 6. Fan-type and radial machines: constant first-order inertial forces rotating with the crank
z
Fan-type )' in
Radial yin 0
0
36 45 60 90
5 4 3
z
4
FI/P I
90 120
2.5 2.0 1.5 1.0
180
2.0
72
Table 7. Moments of inertia for opposed-cylinder machines
2
-
Crank angular position diagram
-
-
Mnmax
1
1
1
Vi.
Vi.
0
0
0
2
0
0
0
CD
-
1
~ Q9 ~ CD A
Mlmo'X
lje P,;e
1
2
4
Mr
--r;e
-
-
2
4
'
1
1.'
width of a pushrod. Here, up to five are located on one crankpin. They feature lightweight casings and the resultant of the first-order forces (see Table 6) rotates with the crank and can therefore be counterbalanced easily by weights at its countermass extensions. In the case of larger cylinder numbers, it is usual to employ a master connecting rod with pivoted-link connecting rods [14]. Opposed·Cylinder (Flat) Engines. Here, the power units 1 and 2, 3 and 4 (Fig. lOa) are allocated to the double cranks A, B, etc. If the masses are equal, the double cranks are free of forces, their moments are equal (Fig. lOb) and independent of the inter-cylinder distances. Equations (14), (17) and (18) then apply to the double cranks. This mode of construction with small moments due to mass (Table 7) is now mainly used for horizontal compressors. 1.2.2 Graphical Methods These show plainly the magnitude of the forces and moments [15, 16] and are particularly suitable for nonsymmetric machines. The polygons resulting from this method can also be evaluated by computer using vector programs. General Methods. The rotating forces are plotted in the direction of the cranks with which they rotate. In the case of oscillating forces, the auxiliary vectors PI = morw2 and P n = APl take the directions of their cranks or twice the latters' angle and are then projected onto the centreline of their power unit (Fig. 11).
4
A
: 1 ) --J-- ' "l] m'®.~:, r
a
T
2
3
1
i
i 4
i 6
Figure 10. Opposed 6-cylinder machines: a power unit arrangement with crank angular position diagram, b moment star with polygon.
4
- '®6 6~ 1
-
A
e4
2
6
5
3
1
ting of the vectors (Fig. lIb). In the case of the moments, these are rotated in a counter-clockwise direction into the cranks to simplify matters (Fig. 11c), and are marked with a cross-stroke. For negative lever arms as in Eq. (11), or if the cranks are located to the right of the centroidal plane, the vectors must be plotted in the opposite direction from that of the cranks. The rotated auxiliary vectors are then projected and added to the firstor second-order moments just as in the case of the forces. 1.2.3 Compensation of Forces and Moments Forces and moments can cause dangerous resonance phenomena. They must therefore be compensated within the machine or be avoided by tuning the foundations [17, 18].
1
2
--I0I210/2f--
In-Line Engines. Here, the first- and second-order crank angle pOSition diagrams with their crank angles and their double angles respectively serve to simplify the plot-
1
2
1.3
. I
i I I ~1112 )'112 a
S
~
,OC» c&TD~?>s~
2
l.Z"Y
"" b
~res
/'
Frres=f!r~s=O
Mrres =Fr a
"'lr=120' Mrresm(J'J.=P'O
fur" 90'
~"~ 1.1'"
Il
7511
~1~
fire5=ilI Mrres Furesmax=ZPrr M1re5 =0.B66fjo "'u=90' d
Figure 11. Graphical methods (two-cylinder in-line engine): a schematic of crankshaft and stars, b definition of moments; forces and moments, (: and d first- and second-order.
Machine Dynamics. 2 Vibrations
Figure 13. Crosshead power unit with complete mass compensation.
a
Figure l::1w Compensation of oscillating forces: a counter-rotating gears fOf first-order forces, b Lancaster drive for second-order forces.
Rotating Masses. Their forces and moments are compensated by counterweights at one or all cranks (see FlO Fig. S). If the forces vanish, it is sufficient to place counterweights at the crank countermass extensions to compensate for the moments; however, this leaves moments internal to the shaft in existence [6]. Osdllating Masses. These are compensated by weights moving in the opposite direction or rotating at twice the rate of revolution (Fig. IZa). Their components, which are normal to each other, compensate for the masses and the free centrifugal forces. They are driven by the crankshaft and lie in the centroidal plane below it, so that no additional moments are generated. In order to compensate for the moments, these weights are situated in front or behind the crankshaft. They are driven by a pinion from the shaft extension using an auxiliary shaft (Fig. 14). In the case of a Lancaster drive (Fig. IZb), this is accomplished by means of a toothed vee-belt. The oscillating forces of all orders of crosshead power units are compensated by the weight m = m K + m KS
Figure 14_ Compensation of mass effects by counterweights: 1 at
the crank countermass extensions for rotating moments, 2 at the shaft ends for first-order moments, 3 in the centroidal plane for second-order forces.
+ m Kr + m os, moving in a direction opposite from that of the piston (Fig. 13). Its two tooth racks are driven from the piston rod via two pinions. This arrangement is suitable only for lower rates of revolution and requires long power units. Example. Compensation of the forces and moments of a twostroke engine with two cylinders (Fig. 14:) whose masses mg of the counterweights are to be determined when the distances b and c are known. For the rotating moments. the relation mg = m,,(r/rg ) • (a/c) applies to each counterweight 1, since m g r g w2 b
= m"rbTa.
For the first-order moments of the two rotating weights 2, the relation mg := mo(r/rg) . (a/c) applies. For the second-order forces, the relation for weights 3 is mg = O.5Am,,(r/rg ), since m@rg(2w)2:= 2P" ; 2Am,.,.W'.
Vibrations R. Nordmann, Kaiserslautern
2.1 The Problem of Vibrations in Machines Machine dynamics in general deals with the interaction between forces and motions in machines. Here, apart from the desired dynamics that are necessary for the machine's function, there are also undesirable dynamic effects. These arise because machines and components possess elastic properties and mass and thus form systems capable of oscillations. If forces occur that change with time or movements of the base, these result in machine vibrations. Although compared with the required move-
ments, such movements are in general small, they can become dangerous under certain circumstances. So-called resonance phenomena are particularly hazardous and occur when an exciting frequency corresponds to the natural frequency of the machine structure thereby amplifYing the vibration amplitudes. Self-excited Vibrations, which are maintained by the presence of a source of energy, also represent continued dynamic loading. Machine vibrations always constitute a problem when excessive stressing of the materials takes place. If the permissible tensile values of materials are exceeded, they may be damaged. In order to guarantee the functional integrity of machines, deformation limits must also be observed. Thus rotor vibrations in turbines and electric motors must not become so large that they exceed the clearance between rotor and hOUSing. Vibrations are also inimical
2.2 Some Fundamental Concepts. 2.2.3 Model Parameters
to the environment. This applies not only to oscillatory movements often perceived as unpleasant but above all
to the noise caused by vibrations (e.g. body drumming). Finally, vibrations have a deleterious effect on the quality of manufactured articles in production processes. In the case of machine tools, therefore, attempts are made to
Mx(t)
ical and experimental aids are of special importance in
this respect.
= F(t).
(I)
M = the Quadratic N x N mass matrix. M contains the
inertial coefficients of the system and is symmetric. D
= the
K
= the Quadratic
keep the relative movement between tool and workpiece
as small as possible. A typical example of machine vibrations is found in vehicle engines, where the dynamics of the machine functions are of interest because of the crank drive. The problem involves the motions of the pistons and the crankshaft under the effect of the forces applied by the gas pressure (see FlO). The crankshaft itself is a system capable of oscillations that occur as flexural and torsional vibrations excited by the gas pressure and inertial forces transmitted via the connecting rod (see]l). This can be accompanied by resonance effects if the frequency of excitation is now the natural frequency of the crankshaft. In order to avoid dangerous vibration conditions, it is therefore important to know the amplitude and frequencies of the exciting forces as well as the dynamic properties of the crankshaft (natural frequencies, damping, eigenvectors). Engineers must be concerned with the problem of machine vibrations during the development and design of machines, as well as during the testing and later operation (machine supervision and diagnosis). Modem mathemat-
+ Dx(t) + Kx(t)
Here:
Quadratic N x N damping matrix. D contains the damping coefficients of the system. D may also be asymmetric (gyroscopic effects, sliding bearings and seal gap forces). N x N stiffness matrix. K contains the stiffness coefficients of the system. K may also be non-symmetric (rotating forces, sliding bearings and seal gap forces).
F(t) = the N x 1 vector of the time-dependent exciting
forces. Displacement or acceleration excitations
can always be transformed into force excitation. xU) = the N x 1 vector of time-dependent displacements or angles. X, are the associated velocities and
x
accelerations respectively. The equations of motion, Eq. (I), express the balance of forces and moments and take inertial forces into account. They are generally valid under the above conditions (linearity, time-invariant matrices), and can be applied to different types of machine as well as to different types of vibration (tlexural vibrations, torsional vibrations etc.). It is natural to use a graphical representation of the vibratory system. This can be done with a block diagram which relates input to output parameters (Fig. 1). Certain input parameters F(t) enter the system as exciting forces (e.g. out-of-balance forces, process forces, impact
2.2 Some Fundamental Concepts
It is advisable to base all investigations of the vibrating engine on certain model concepts. For this reason the real system is assigned a mechanical equivalent system
forces etc.) or as base excitations (floor disturbances). The system transforms the inputs in accordance with its transfer characteristics and responds with tbe output parameters x(t) or the velocities x(t) and the accelerations x(t) derived from these. The transfer characteristics are determined by the structure of the system, i.e. by the appropriate physical laws, and by the system parameters M, D and K. If M, D and K as well as the excitation vector F(t) are known (for instance periodic or pulse excitation of certain degrees of freedom), then
(vibration model), which permits the neglect or idealis-
first the natural oscillation parameters and then the
ation of certain features (see J2.6) and which is made up from simple mechanical elements (e.g. masses, dampers, springs, rods, beams etc.). It possesses a certain number N of degrees of mechanical freedom (displacements, angles).
response parameters x( t) can be determined by calculation (see J2.2 and J2.7).
By way of introduction, some important concepts in tbe field of machine vibrations will be explained. 2.2.1 Mechanical Equivalent System
2.2.2 Equations of Motion, System Matrices If the fundamental
mechanical equations (Newton, d'Alembert's principle of virtual work; see A3) are applied to the mechanical equivalent system, then the equations of motion that express the interdependence of a timedependent input parameter F(t) and the output parameters x(t) are obtained. These equations may be linear or non-linear. In the case of many practical tasks, in partiClllar, when dealing with small oscillations about an equilibrium point, linear models will suffice. We confine ourselves here to the description of linear time· invariant vibration systems with deterministic input parameters F(t). For a treatment of non-linear systems see A4.3 and [1, 2J. Under the above conditions it is always possible to obtain a system of linear, time-invariant, second-order equations of motion independent of the
number of degrees of freedom prevailing in each case (time-invariant means that M, D and K are not timc-
dependent):
2.2.3 Model Parameters: Natural Frequencies, Modal Damping, Eigenvectors
Natural Frequencies. Every linear oscillatory system has a particular natural frequency behaviour, which is determined by its natural frequencies, damping factors and eigenvectors. For instance, if a short disturbance in the form of a pulse Fk(t) is applied to the ventilator rotor shown in Fig. 2, the oscillatory system will subsequently undergo natural oscillations which are composed of several component vibrations (n = 1, 2, ''', N):
Oscillatory system x It) Structure: Linear Time-invariant / N degrees of freedom v Parameters: M, D, K Vector of the Vector of the N Input parameters N output parameters
Fill
"
Figure 1. Block diagram of an oscillatory system with physical pammeters.
Ia
•
Machine Dynamics • 2 Vibrations
II
approximately if the system is harmonically excited with a frequency that corresponds to one of the natural frequencies.
Eigenvalue Analysts. The modal parameters may be obtained mathematically by making p(t) = 0 on the righthand side of Eq. (1) (homogeneous equations), and using s(t)
=tpeAt,
*(t) = AtpeAt, ~(t)
(3)
= A2tpe'"
to define the eigenvalue problem b
(A2 M+ AD + K)tp= O.
(4)
If behaving in a purely oscillatory manner this has the sol-
ution An = an + iwn; A:; = an + iw. eigenvalues,
(5)
tpn = 'P::" + itp::";
(6)
II'! = 'P::" - icf>::" eigenvectors.
In many practical cases, it is difficult to construct a
damping matrix. In the case of weakly damped structures, which often occur in machine design (rotors in roller bearings with elastic, torsion and flexural properties, turbine blades, steel foundations) the assumption of "modal damping" suffices. The eigenvalue problem is initially solved for the undamped system (D = 0) in the purely real form (K-w'·M)tp=O
d FIpre lI. Natural oscillation parameters of a ventilator rotor: • time function of the force, b time function of the decrement of the amplitude of oscillation, c construction principles of the ventilator rotor (1 impact force generator, 2 vibration sensor), d time function of the eigenvectors.
s(t) =
~ An e"nt {'P::e cos(Wnt + "'n)
(2)
- '1'::" sin (Wnt + "'n)} . Each component vibration consists of an exponential function which describes the decrement or, in the case of unstable systems, the buildup of the component; and trigonometric functions that describe the vibration behaviour. The nth partial solution encompasses: w., the natural frequency (S-I); -an> the decrement factor (S-I); 'f/tf, ~ the real and imaginary part of the eigenvectorsII'; and An' "'n, constants; these are determined by the boundary conditions (impact). By measurement of the pulse response xl(t) or the acceleration xl(t) of degree of freedom I, the natural frequency parameters W n , a. can be determined by signal analysis and on processing further signals, the eigenvector components 'P::e, If'nbn as well. These are described as modal parameters. A knowledge of these quantities is extremely important, because they characterise the dynamic properties of a system capable of oscillations. With their help, it is possible to determine, Inter alia, at what frequency resonance effects can be expected and the magnitude of the resonance amplitudes (damped). The eigenvector indicates what deformation occurs when the system oscillates at its appropriate natural frequency. This deformation occurs
(7)
and the natural frequencies W. and the corresponding real eigenvectors 'Pn are then obtained. Any damping that does not enter this calculation is estimated or determined experimentally. A decrement factor -an or a modal damping value (degree of damping) Dn = - ani Wn is then assigned to each natural angular frequency w.. In practice, the follOwing parameters are most frequently used:
In = wn /27r
natural frequency (Hz),
Dn=-a.lwn
modal damping (-),
'Pn
real eigenvector.
(8) (9) (10)
Some numerical values for the modal damping D in %: Material/Components
Din%
Steel Cast iron Rubber (natural caoutchouc) Steel structures
0.1 1.8 to 2.0 1 to 8 0.2 to 1.5
Reinforced concrete structures
4
Steel foundations for tutbines without floor 0.5 to 1.5 damping Steelfoundations for turbines with floor damp- 1.5 to 3.0 ing
Knowledge of modal damping is particularly important in determining amplitudes of resonance oscillations caused by the excitation force P(t). Figure :z shows the two first eigenvectors tpl and 'P2 with their associated natural frequencies J. and 12 for the ventilator rotor with roller bearings. The first form of
2.2 Some Fundamental Concepts. 2.2.5 Frequency Response Functions
F(t)
""\
/
Oscillatory system Structure: Linear Time-invariant Ndegrees of !reedo. Parameters: "".On.OOn.cpn
xU)
J,.
v
Fi......, 3. Block diagram of a vibration system with modal parameters.
natural oscillation is visually similar to the static bending mode; the second mode of oscillation with a single node is deSCribed as the 5-mode. In contrast to complex eigen· vectors which occur when damping is taken into account, the ratio of the eigenvector components always indicates a constant deformation figure in the case of real eigenvec· tors. The indicated sintple procedure is not permiSSible in dealing with an oscillatory system capable of seif-exci· tation, as for instance in the case of rotating machinery featuring sliding bearings and seal gaps (pumps, turbines, compressors). In these cases it is necessary to solve the eigenvalue problem Eq. (4) and to assess the stability behaviour using the eigenvalues so obtained (see J2.7.4). 2.2.4 Modal Analysis In a way analogous to Fig. 1, the relation between input parameters F(t) and output parameters s( t) can also be
expressed by modal parameters (Fig. 3). if all natural frequencies 00.. eigenvectors 'Pn and damping factors (-a,,) or their modal damping factors Dn are known, they can be used to calculate the vibrations. In the case of a system capable of seif-excitation, the set of eigenvectors on the left is required as well [1, 2]. This calculation procedure is also described as "modal analysis" because the modes (eigenvectors) enter into the calculations. An advantage of this method is that the equations of motion Eq. (1), which were originally coupled, can be decoupled by making use of certain orthogonal propetties of the eig·
Ia
envectors. This sintplifies the calculation, and a physical intetpretation of the dynamic processes is made easier. The concept of modal analysis is now also employed in the determination of modal parameters from measurem· ents. The basis of the process is the representation of the system responses as a function of the modal parameters and the exciting frequency (Fla. 4). In matching analyti· cal system responses (frequency response of the model) to the measured system responses (measured frequency response), the modal parameters are changed until good matching between model and measurement is obtained. The result yields the required modal parameters. In a measurement process in general, test forces (pulse, sinusoidal, noise) are input into the system and the oscil· latory response is registered at the individual measuring points. The measured frequency responses are calculated from the time Signals after they have been converted to the frequency domain by Fourier analysis (seeJ2.4.2), and these are then used in the matching process that is required for the calculation of the modal parameters [13). 2.2.5 Frequency Response Functions of Mechanical Systems, Amplitude and Phase Characteristics Definition. If a linear vibration system which is described by the equations of motion Eq. (1) has its degree of freedom k excited by a harmonic exciting force
Fk = F. sin llt, (11) where ftk is the constant force amplitude and {l is the exciting frequency (all other forces being absent), then, after its transient state has passed, the system responds with movements that are also harmOnic (Fig. 4). All response parameters can be expressed by means of the vector x(t): X2(t)
x, x2
sin(flt + e'k) sin(flt + e2k)
x,(t)
XI
sine flt + elk)
xN(t)
xN
sine flt + eNk)
x,(t)
set) =
001. 01
Fipre 4. Harmonic exdtatioo of a linear vibration system.
(12)
Ia
Machine Dynamics. 2 Vibrations
The response for each degree of freedom is characterised by an amplitude and a phase angle in relation to the excitation force. For example, for the degree of freedom I, x,(t) = X, sin (!2t + e,k)'
(13)
X, as well as e'k (elk is negative) depends on the frequency of excitation. Therefore, the expression (14) is called the amplitude-frequency response (between I and k), and (15)
(22) and contain both the amplitude X, and the phase elk' If this calculation is repeated for other frequencies !2, further values of the frequency response function EIIk (!2) are obtained. For a system with N mechanical degrees of freedom (displacement and angles), there are N x N frequency responses (N degrees of freedom and N responses). The total matrix H(!2) of all frequency response functions EIIk (!2) (I = 1 ... N; k = 1 ... N) is obtained from the inversion of the complex (dynamic) stiffitess matrix K(!2) = K -!2'M + i!2D: (23)
is the phase-frequency response (between I and k). In practice, both functions can often be combined in the complex frequency response
EIIk = (.X,! F.) e'·" = IEIIkI e'·,k.
(16)
Since the quotient of the quantities x,I ftk represents a compliance parameter (displacement/force), EIIk (!2) is often described as a complex compliance frequency response. Figure 4, shows qualitatively the dependence of the amplitude IEIIkI = x,/Fk (amplitude response) and the phase elk (phase response) on the exciting frequency !2. The significance of the frequency response functions becomes particularly clear if the plot of the amplitude response is followed. If the exdting frequency !2 lies close to a natural frequency (W" W2 ••. ""') (resonance case) the response amplitude X, reaches a maximum whose magnitude in each case depends on the associated damping (ai, a" ... aN; D " D, ... DN respectively) (large damping, low amplitude increment). In the region of the resonance frequency the phase angle e'k (here defined as negative) changes relatively steeply. Calculation of Frequency Responses and H ..... monic and Periodic Systems Responses. If the equations of motion Eq. (1) are known, together with the matrices M, D, K, then the complex transfer function EI'k/ (!2) can be calculated by introducing the harmonic excitation function Fk/ (t) into the complex force function: Fk(t)
=ftk e'lU =F. (cos !2t + i sin !2t),
(17)
where in the case of excitation at a single point, only the kth component in the force vector has finite magnitude: F(t)
=Fe'{l,; F= {O,O,
... F.,O, ... OJ.
(18)
Putting Eq. (18) into Eq. (I) yields Mi& +
D* + lex = F e,{l,.
(19)
Using the complex formulation and its time derivatives, X=
i
e'lU,
it = i!l~ e iflt ,
(20)
s==-fl2it:!Dt,
The problem of harmonic. excitation and therefore of harmonic oscillations plays an important part in machine dynamics. Knowing the frequency response functions of a system, excitation frequencies at which particularly large response amplitudes occur can be assessed. An important application in rotating machinery is where harmonic excitation forces rotate with the angular frequency !2 (frequency of rotation) caused by out-of-balance masses. If the out-of-baIance force vector (out-of-balance forces are proportional to !2'; see J2.5) is substituted in Eq. (1), and the effect of the rate of revolution in the system matrices is included, special frequency response functions are obtained which describe the response amplitudes of flexural oscillations for the rotating shaft as a function of the exciting frequency. Since exciting frequency is equal to rotational speed, the term "critical angular frequency" is used when the angular frequency corresponds to a natural frequency of the system. If the exciting forces of a system contain several simultaneous excitation frequencies, as with periodic functions, then the amplitudes obtained from the frequency responses of the various exciting frequencies can be superposed with their correct phases to yield the total response. Periodic excitation forces or moments are found for instance in the case of gas pressure and inertial forces in the operation of the crankshaft of a vehicle engine (see Jl).
2.3 Basic Problems in Machine Dynamics In the treatment of vibration problems in machines many
questions arise. In the following survey it will be shown briefly that the problems occurring in various types of machine can be reduced to a few fundamental problems. In the explanations, an oscillatory system block diagram (Fig. I) and the associated equations of motion, Eq. (1), are used.
the complex system of equations follows: (K -!2' M+
iD)i=F;
(21)
from this it is possible to determine the vector .i- of the complex system response associated with any given excitation frequency !2 by solving the system of complex linear equations Eq. (21), provided that the M, D, K matrices and the force vector F are known. The components of.i- have the form
1I.~.1
Direct Problem
The direct problem is the task most frequently posed in practice and usually associated with the design phase of a new development. Here, the system to be investigated is usually given in the form of a design drawing (Fig. Sa). The basic task to be solved consists of calculating the time dependence of the system responses x(t) from the known critical time-dependence of the forces F( t) and the system
2.3 Basic Problems in Machine Dynamics. 2.3.5 Improving Machine Vibration
OSCillating system Parameters M, D, K
F(t) a
,
F(t)
OSCillating system Parameters: M D. K
,-----1, " L-----t/
I
x(t)
b
system response x( t) is given, and if the system properties M, D, K are also known then the question concerns the time dependence of the input parameters F(t) (Fig. Sb). A widely known application example of this problem is the balancing of rotors. Here the out-of-balance forces of a rotating shaft must be offset by mass compensation at the rotor, so that the bearings are not subject to forces at rotational frequencies. The vibration signals x( t) are measured before the balancing process.
r - - - - - - - - - - - - - - ---,
'
F(t)
~I I
Real Oscillating system Structure ? 0
:
~ xU) ~
L _ E9!1\r11.elel~0)_______ J
c
r------------------,
I
FU) ~
L--.V:
ReqUfrec Oscillating system Structure ? 0
I
~ x (II ~
L_.P1\rlf!1'!!Ilf§.':.L ______ J
d
Figure S. Fundamental prohlems of machine dynamic;: a direct problem, b input problem, c identification problem, d optimisation problem.
properties in the form of matrices M, D and K which are also known. The natural frequencies and modes are usually determined in a preliminary step. In accordance with [2] the following procedure is recommended for this important machine dynanlic analysis: 1.
Listing of All Load Cases (Excitation Forces). Load case of normal operation; loads arising from perturbations.
2.
~.
Idealisation of the Structure. Providing a mechanical equivalent system that represents the dynamic behaviour for the various load cases with sufficient accuracy. Decision regarding the type of modelling (multi-body systems, finite elements). Number of mechanical degrees of freedom. Derivation of Equations of Motion. In the case of discrete systems (multi-body systems, finiteelements) with linear system properties, the linear system of differential equations already given in Eq. (1) is obtained: Mx+Dx+Kx=F(t).
4.
S.
6.
Solution of Equations of Motion. The homogeneous solution of the linear equations of motion defines the natural frequency parameters and the stabiliry of the system. Then the particular solutions for the various load cases that describe the forced vibrations must be calculated. "'Where necessary. several load cases must be superposed. Graphical Presentation of Results. In order to present clearly the often enormous mass of data, the time dependence of displacements, the accelerations or shear loads and the frequency dependence of the amplitudes (frequency response) are shown by the computer in graphical form. Evaluation and Interpretation of Results. Several questions must be answered by the results, e.g.: Is the structure capable of standing up to the stresses occurring in all load cases? Is the system stable' Are resonance points closely approached' Where necessary, flaws must he removed, components changed or other material employed.
2.~.2
Input Problem
Here, the question that arises is the reverse of that of the direct problem, in that now the time dependence of the
2.~.~
Identifying the System Parameters
In the identification problem, the question is the determination of the equations (structure), including the system parameters, which describe the behaviour of the system from the measured input and output signals (Fig. Sc). Since criteria relating to the structure of the equations are frequently known (e.g. linearity, time-invariance, number of degrees of freedom) or assumptions regarding these are made, the task is reduced to one of so-called parameter identification. Here, test forces F(t) (pulse forces, step changes of force, harmonic or random excitation forces) are impressed on the vibration system in question and the resulting system response x(t) is registered. With the aid of the measured input quantities F(t) and output quantities xU) the required system parameters can be determined by a process of estimation, taking into account the known input-output relations (structure). Here, procedures in the time as well as in the frequency domain are applied. In more extended vibration systems it is particularly problematical to determine the system matrices M, D, K completely by identification of parameters. Since in general the parameters for simple mechanical elements (rods, beams, plates) can be obtained fairly well by calculation, experimental parameter determination is limited to certain system components with complex force displacement laws that in most cases possess only a few degrees of freedom. In the design of machines, such components are for instance sliding bearings, gap seals, couplings, etc., which often have a major influence on the vibration behaviour of the complete system and for which spring and damping coefficients are therefore required.
Modal analysis has become very significant in machine dynamics. In this identification procedure, the modal parameters of mechanical systems can be obtained from nleasured frequency responses. 2.~.4
Design Problem
The design problem is to construct a system so that certain desired output parameters x(t) are obtained from given excitation parameters F(t) (Fig. Sd). The task is to create an optimum dynamiC system. Frequently in these cases, the structure is given as well, so that parameter optimisation has to be carried out only within certain limits. 2.~.S
Improving Machine Vibration
This task often occurs in the practical operation of machines. Here some of the partial tasks already described have to be carried out. Machine vibrations are undesired phenomena that must not exceed certain limiting values. If the displacements x( t) are too large, the dynamic condition of the machine has to be improved and this can be accomplished in four steps. First the output signals xU) are measured and analysed in the time and frequency domain. Excessive
Machine Dynamics • 2 Vibrations
vibrations could be caused by excessive excitation FU) or poor system properties (wll' an. ipn). Therefore, as a second step, dynamic properties of the system are systematically investigated. Then, system propcmies can be identified (identification problem) by suitable test signals FU), and the corresponding measured output signals xU). A calculation model that represents the dynamic properties of the machine investigated with sufficient accuracy, can be matched to these results. The last step is now to fmd by simulation calculations those system modifications that are most effective in leading to a reduction of the vibrations. Here optimisation algorithms are applied that take account of the prevailing boundary conditions in each case (design problem).
nals, where the deterministic signals are here treated as the more important. They are subdivided once more into periodic and non-periodic oscillations. Harmonic sine and cosine functions are elementary signals belonging to the periodic category. General periodic signals are built up from sine and cosine components whose frequencies are multiples of a fundamental frequency The non-periodic signals are represented for instance by damped harmonic oscillations (natural oscillations), the pulse function and the step function. All the signals in Fig. 6 have in common that they are represented as time-dependent. While all the deterministic signals can be described by mathematical functions, random signals are not uniquely determined. It has turned out to be useful to characterise the various ways by which signals depend on time by mean values [I].
no.
2.4 Representation of Vibrations in the Time and Frequency Domains
Mean Values. The temporal linear mean value of xU)
is called the equivalent value,
2.4.1 Representation of Vibrations in the Time
fI T
Domain
xCt)
Machine vibrations manifest themselves by the time-varying displacements of individual points of the machine, which either repeat themselves or decrease or build up in one single process (natural oscillations of limited duration), or else proceed in an irregular (stochastic) manner. The time dependence of vibration processes is part of the study of kinematics (see A2). This deals particularly with the time dependence of individual components of xU). Since, however, the excitation forces FU) are also time-dependent we include them in the considerations. Thus, in accordance with the block diagram of Fig. 1, the study deals with the analysis of the signals that enter and leave the vibration system.
=
(24)
X(t)2 dt,
o
where T is the period of observation, the duration of the period in periodic signals. The mean-square value is
-ret) II T
=T
X(t)2 dt,
(25)
o
from which the root-mean-square (r.m.s.) value is derived:
fIrct) T
dt.
(26)
o
Classification. Figure 6 shows a classification of important vibration signals, where the "oscillating" quan-
For harmonic signals frequently encountered in practice,
tiry is here generally designated x( t) Cscalar). These can be roughly divided into deterministic and stochastic sig-
the mean value x(t) = 0 and the root-mean-square value is about 70% of the peak value: X df = -.J'i/2X.
I I I
Periodic signals
I Deterministic signals
Oscillatory Signals
j
Stochastic signals
I
Non-periodic signals
'PtAJ, '~ ~ xt= I
Harmonic sinusoidal
I
I
General periodic
I
I
t
Damped natural vibration
Step function
'L_ ,
t
Pulse function Figure 6. Classification of oscillatory Signals.
'~ Statistical random signal
I
2.4 Representation of Vibrations in the Time and Frequency Domains. 2.4.2 Frequency Domain
2.4.2 Representation of Vibrations in the Frequency Domain
80
In order to interpret input quantity F(t) and output quantity x(t) of an oscillating system more accurately, they can also be represented in the frequency domain as x( D) and F(D). Here fl = 2'ITJ is an angular frequency in S-I and J is the frequency in Hz. Representation in the frequency domain is often more informative because the frequency contributions of a vibration and their relationship to the dynamic properties of a system are clearer. By using Fourier analysis, it is possible to transform from the time domain to the frequency domain. The representation in both domains becomes clear by the simple example of a harmonic sinusoidal oscillation, as shown in Fig. 7. The sinusoidal oscillation x(t)
=x sin (flt + s)
(27)
is determined by the amplitude X, the angular frequency fl and the phase angle e relative to zero. In the frequency domain the value of x is plotted as a function of angular frequency fl in an amplitude diagram x(fl), and the value of e as a function of fl in a phase diagram.
Fourier Analysis of Periodic Oscillations. According to Fourier's theorem, every periodic function xU) of period T = 2'IT I flo can, under certain conditions, be expressed unequivocally as a sum of sine and cosine hmetions with the angular frequencies flo. 2flD , 3flo ... : x(t) = Xo
+
2: {so sin nflot + cn cos nflot} (28)
where 7
~ Jx(t) dt is the arithmetic mean, o T
So
=
Example. Figure 8 shows, as an example, a simple periodic function with two sinusoidal components in the time and frequency domains. This type of oscillatory signal can occur in rotating machines where the fundamental frequency ~) coincides with the angular frequency (out-of-balance oscillations) and with twice the angular frequen<'T 2Do caused, e.g., by the depanure from perfect circularity of a shaft cross-section (rotor of a genenltor, shaft with a crack). Numerical values: XI)::':: 0; XI ::= S, ::= 20 IJ..m; x2 ::= S2::= 10 J.Lm; c i ::=c'2=0.
Fourier Analysis of Non-periodic Processes. A transition from periodic to non-periodic processes is obtained when limiting values for infinitely long periods T are considered. In such a case the fundamental frequency has infmitely small values (flo ~ dfl) and the higher harmonics are closely spaced. This leads to a continuous spectrum. The time function can now be expressed by means of the Fourier integral x(t) =
Jx(!l)
ein,
d!l.
(29)
Here the complex spectral function x( fl) is the Fourier transform of the time signal xU): x(fl) =
f
x(t) e-il1, dt.
(30)
o
11=1
Xo =
= tan (co/so)
are the values of the Fourier phase spectrum.
~ JxU) sin n!lol dt
Example. Figure 9 shows qualitatively the contribution to the Fourier transfonn Ix(D)1 by three non-periodic signals. The first two are frequently used as test signals for the artificial excitation of oscillatory systems. The values of the spectral function Ix(D)1 of the pulse function (Fig. 9a) remain almost constant over an extended region. The position of the zero transit IxDI::= 0 depends on the pulse length (hard or soft pulse). In the case of the step function (Fig. 9b) the greater part of the energy is distributed over the lower frequencies. Hence systems with low natural frequencies are strongly excited. A very interesting result is found in the case of the third signal (Fig. 9c:). This is the pulse response function (weighting function) of an oscillator, i.e. the natural frequency of the system after a short pulse. If this function is transfonned into the frequency domain, the corresponding frequency distribution
o
,
Cn
=
~ Jx(t) cos nflot dl, D
are the Fourier coefficients
(n =
1,2, "',
x),
fl" = 2'ITIT
is the fundamental frequency (angular frequency),
x'LL'
are the values of the Fourier amplitude spectrum, and
xiI)
'll_
zo
o
')
.'-----!Q~I' Phase spectrum
Figure 7. Representation of a sinusoidal oscillation in the time and frequency domains: a time domain, b frequency domain.
2
Qo
l.Q o
l:~"~·'
Amplitude spectrum
b
X
10
Q
a
1
a
b
Qo
Phase spectrum
zno
Figure 8. Periodic function with two sinusoidal functions (x, = 20 j-Lrn; x2 = 10 j-Lm; 1':1 = 0; 1':2::= 0): a time domain, b frequency domain.
Machine Dynamics. 2 Vibrations
equations). The frequencies of oscillations are determined by the system properties (M, D, K), distinguishing between damped and undamped oscillatory systems. In the ideal case, free from damping, an exchange between kinetic and potential energy takes place (permanent oscillations). In the real case, the vibrations always decrease in the presence of genuine damping (see Figs 2,6 and 9).
2.5.2 Selt"-excited Vibrations
Frequency domain
Time domain
Figure 9. Spectral function ix([J) I for three non-periodic functions: a pulse function, b step function, c pulse response function. functions already defined in J2.2.5 are obtained. Figure 9 shows the frequency distribution for an oscillator with one degree of freedom.
These are a special case of natural vibration. As with free vibrations, no external excitation is present in the equation of motion (F(t) ~ 0). However, the oscillating system obtains energy from an energy source at the same frequency as the natural frequency. The energy input can lead to a buildup of (self-excited) oscillations if this is not prevented by countervailing damping forces. Strictly speaking, the self-excited vibrations have a non-linear character. However, linearised equations may be used in order to assess the stability at equilibrium. The tendency to self-excitation of an oscillatory system is recognised from the skew-symmetric contributions of the stiffness matrix K (circulatory forces) which is opposed by the damping factors (D matrix). In machine construction, examples of self-excited oscillations arc found, inter alia, in rotating shafts with sliding bearings and sealing gaps. Here the energy input comes from the rotation of the shaft.
2.5.3 Parameter·excited Vibrations
2.5 Origin of Machine Vibrations, Excitation Forces F(t} Vibrations in machines can have very different origins. In [5], a subdivision according to the originating mechanism is presented which differentiates between free, selfexcited, parameter-excited, and forced vibrations. The various cases are best explained by the equations of motion Eq. (I) using the block diagram (Fig. 1). These presuppose linear time-invariant oscillatory systems. The various origins for oscillatory movements x( t) are shown clearly in Fig. 10.
2.5.1 Free Vibrations Free vibrations occur only when a system is left to itself after receiving a pulse and is not exposed to any further external influences (see A4.1). Moreover, no further energy is supplied to the system. In the equations of motion, the right-hand sides representing the excitations are equal to zero (F(t) ~ 0, homogeneous system of
Single pulse Origin of natural vibrations
FIt)
Oscillatory system
External perturbations Origin of forced vibrations Energy supply controlled by the system Origin of self-excited vibrations
xlt)
Oscillations Systems parameters changing with time Origin of parameter-excited vibrations
Figure 10. Origin of machine vibrations.
Parametrically excited oscillations are characterised in that the osciilating system has time-dependent, mostly periodic, parameters. The presupposition of time-invariant equations of motion is then no longer satisfied, and the matrices are in general time-dependent: M(t), D(t), K(t). As a consequence, damped and undamped as well as stimulated oscillations can occur. For instance, rotors of electrical machines often have cross-sectional forms with strongly differing bending moments in two directions normal to each other (e.g. two-pole rotors of synchronous machines). When the shaft turns, the vertical stiffness of the shaft in a fixed coordinate system changes periodically with time. The stiffness matrix K of the rotor is therefore time-variant. Rotational speeds that produce a buildup of parameterexcited vibrations must therefore be avoided.
2.5.4 Forced Vibrations Forced vibrations (see A4.1), which probably occur most frequently in practice, are caused hy external perturbations, and their time behaviour is determined by them. These perturbations are represented as exciting forces (moments) by the vector F(t) on the right-hand side of the equations of motion. They depend only on the time t and not on the displacement x( t) of the oscillatory system itself. With regard to the exciting functions, the periodic functions and, as a special case, the harmonic functions are of particular interest in vibration practice. Next to these, pulse functions (perturbation through impact), step functions (switch-on processes) and random func-
tions are of great importance. Perturbations enter the system either as forces (moments) or as displacements or accelerations of the base. For instance, considerable exciting forces can occur in machines in the form of inertial forces through the translational or rotational movement of their masses. Other important excitations are caused by the coupling of mechanical systems with surrounding work media (gas,
2.5 Origin of Machine Vibrations, Excitation Forces F(t) • 2.5.4 Forced Vibrations
vapour) or with electrical systems (motors, generators) where strict coupling may be approximated by purely time-dependent perturbation functions. Perturbations in the environment of machines (ceilings of buildings, foundations of buildings) manifest themselves as excitations of the base of the oScillatory system. In regions subject to earth tremors, for instance, important machines and
machine aggregates (e .g. cooling pumps in nuclear power stations) must retain their functional integrity even under strong external influences. Some important cases of excitations are presented and discussed in the following paragraphs.
Excitation Through Harmonic Out-of-balance For· ces. In the construction of turbomachinery, flexural oscillations of rotating shafts are, in most cases, caused by out-of-balance forces. A clear explanation of excitation by out-of-balance forces is given by the example of a turbine rotor that has been idealised to the form of a disc (Fig. 11). Owing to manufacturing inaccuracies and non-uniform blade distribution, the centre of gravity S of the disc and the point at which the shaft passes through the disc W do not coincide. These two points have a fixed distance e from each other, described as the mass eccentricity, which represents a relatively small quantity compared with the turbine rotor diameter. During the operation of a machine, the mass eccentricity can increase because of deposits, by erosion, or through fracture of a blade. The product of the mass of the rotor m and the mass eccentricity e is known as the out-of-balance moment U = me. Rotation of the shaft generates the centrifugal force F= men'
(31)
which, as a result of tbe rotation of S about the centre of the shaft W, acts in the direction of the line connecting Wand S (centrifugal acceleration) and which rotates at the angular frequency n. The magnitude of the force increases with the square of n. An observer in a t1xed coordinate system sees the two components of the cen-
trifugal force as periodic or, more accurately, as harmonic fimctions: Fhor = men!. cos nt,
(32)
In the case of conlpiex rotors, the out-of-balance force
with eigenforms). Discrete out·of-balance force contributions must be attributed to the various degrees of freedom used in the calculations. The out-of-balance forces excite the shaft as well as the bearing blocks, the foundations and the housing, to harof the monic oscillations with the angular frequency shaft. In practice the intention will always be to keep the outof-balance exciting forces as small as possible. This is achieved hy the proc~ss of balancing in which suitable balancing weights are attached to the rotor. In balancing, it must be ascertained whether the rotor to be balanced is to be regarded as rigid or as elastic. Further details regarding the practice of balancing and the quality of balancing can be found in [6, 7 J .
n
Excitation Through Inertial and Gas·generated Forces in Piston Machinery. In the power units of piston machines (four-stroke engines, two-stroke engines, piston compressors), inertial forces (see ]1.3) caused by parts in longitudinal motion (pistons, parts of the connecting rod etc.), and by gas forces at the piston, occur in addition to the out·of-balance forces caused by rotating parts (crankshaft). This can lead to the excitation of considerable vibrations of individual components or of the entire engine [8, 9J (see FlO). In most cases, these forces vary periodically with engine speed (fundamental frequency = angular frequency), but the gas-generated forces in four-stroke engines have a period of two revol-
no
utions because a combustion stroke occurs in the cylinder of a four-stroke engine only once every other revolution. Of the various vibration phenomena in piston engines, crankshaft vibration needs special attention, so that its stresses do not lead to a fracture. For crankshaft vibration calculations, the time-varying exciting forces acting on the crankshaft which result from the above-mentioned inertial and gas forces must be known. The following details apply to the stationary condition (constant revolutions). The most important relations are best explained by means of a single-cylinder power unit (four-stroke engine). They can easily be applied to multi-cylinder engines. The resulting force rK(t) acting on a piston is composed of the force Fc;(t) due to gas pressure and the inertial force FM(t) (Fig. 12) (see FIO.2.3):
(33)
is distributed continuously along the shaft axis so that, apart from tbe force amplitudes, the relative angular positions have to be taken into account as well. Since the real distribution of the out-of-balance forces is never known accurately, certain sample distributions are assumed in the vibration calculations (e.g. distribution in accordance
Figure 11. Out--of-balance forces of a rotating disc: e mass eccentricity, fl angular frequency, m mass, 1 zero axis for the angle nt.
Figure 12. Forces occurring during crankshaft operation: !]I crank angle, r crank radius, f3 angle of excursion, I length of connecting rod.
Machine Dynamics • 2 Vibrations
The piston force can be resolved geometrically into the normal force FN(t) and the connecting-rod force Fs(t), where the connecting-rod force at the crankpin can once more be subdivided into the tangential component FT ( t) and the radial component FR(t) (see FlO.2). These are the exciting forces for the crankshaft which lead to torsional and flexural oscillations. These may again be subdivided into components due to gas-pressure forces and those due to inertial forces: FT(t) = FTG(t)
+ FrM(t),
= FR,/t)
+ FRM(t).
FR(t)
FT
l
From Eqs (37) and (38) and making use of (35), (36) and (39), the inertial tangential force and the inertial radial force can be calculated as
(34)
For each calculation, the force ratios FT/ FK and FR/ FK applying to the two types of force (gas forces, inertial forces) are needed. These are periodic functions which represent the geometry of the crank operation
F;,
e = 1, e2 = A + A'/4 + 15A'/128, e, = -A'/4 - 3A'/16 - ... , e6 = 9A'/128 + ....
TI
= A/4 + A'/16 + 15A'/512 + ...
T3 = -3A/4 - 9A3/32 - 81A'/512 - ... ,
+ B2 sin 2", + B4 sin 4", + ...
T,
(35)
= _A2/4 -
A'/8 - A6 /16 - ... ,
(40)
T, = 5A'/32 + 75A'/512 + ... . FRM =
= 1, B2 = A/2 + A'/8 + ... ,
rn OK
BI
r~ (Ro + i
""'I
R. cos k"').
with
B 4 = -A3/16 - 3A'/64 - ....
Ro = -1/2 - A2/4 - 3A4/16 - 5A6 /32 - ... ,
cos ("'+ (3) cos {3
RI
=Ao +AI cos ",+A 2 cos 2",+A 4 cos 4",+ ... , (36)
= -A/4 -
A3/16 - 15A'/512 - ... ,
R2 = -1/2 + A2/2 + 13A4/32 + 11A6/32 +,
R,
= -3A14 -
(41)
3A'/32 - 9:\'/512 - ... ,
R4 = _A2/4 - 5A4/16 - 5A6/16 - ... .
with An = -Al2 - 3A'/16 - ... , Al = 1, A2
A4
= Al2 + A3/4 + ... , = -A3/16 - ...
(if! = Do t crank angle, Do angular frequency of the crankshaft, (3 angle of excursion, A = r// connecting-rod ratio). The four individual parts of Eq. (34) can now be stated as follows:
=F,,(t) . (FT/ F K), =FM(t) . (FT/ F K) ,
}
(37)
=FG(t) . (FR/ F K) , FRM(t) =FM(t) . (FR/ F K) .
}
(38)
FTG(t) FTM (t) FRG(t)
However, the inertial force FM(t) as well as the gas-pressure force F,,(t) are also periodic functions under stationary operation. The inertial force FM(t) is given, for instance, by the product of the oscillating-mass mosc (mass of piston, mass contribution of the connecting rod), and the piston acceleration x.(t) and can be expressed by the following Fourier series:
+ e4 cos 4",+ with
,
T2 = -1/2 - A'/32 - A"/32 ... ,
with
FR
2: Tk sin kif!,
""'I
with
sin ('" + (3) cos {3 BI sin '"
rno,c rD~
FTM =
e6 cos 6",+ ... ),
(39)
Similar/y, the forces F TG and F RG resulting from gas-pressure forces on the piston may be obtained. If, for instance, discrete values of the force FG(t) are available over a period, then these are multiplied as in Eqs (37), (38) and thereafter a harmonic analysis of the force components FTG and F RG , so determined, is carried out. Here the different fundamental frequencies generated by the two-stroke engine (Do) and the four-stroke engine (Do!2) must be taken into account. Figure 1~ shows the results of the harmonic analyses of the radial force FRG(t) and the tangential force FTG(t) of a four-stroke engine. In all cases the values shown refer to the piston area A •. For multi-cylinder power units, it is generally assumed that all cylinders are equal and function in the same way, and that therefore the forces are the same for all cylinders. The forces for the various cylinders, however, suffer a time phase difference since the ignition points do not coincide. This phase difference yields different harmonic coefficients for the exciting forces for the different cylinders [8, 9], which can be derived from the stated values for the single-cylinder engine.
Excitation by Electrical Perturbation Moments. In electrical machines (motors, generators) considerable electrical perturbation moments may occur which can excite torsional oscillations along the entire length of the shaft. These will be introduced here in the form of perturbations in a power generator turbine set. In the stationary state, the torques exerted by the driving turbines and the braking generator are in equilibrium. This
2.6 Mechanical Equivalent Systems, Equations of Motion
~ li---~ ----11 I
M,IM o 120 0 - synchronisation error 2
:ftAftftAAr
30
I
20
"\
!
10
I
i
[\..)
~= xc~scp M,IMo
\
\
\J \.
\/ lOB
V
1\ 1\
1\
\
\
\.
V 200 inmst
W.866-0.866cosOH.SsinOt)
2-phase terminal short cirCUIt'
4rr inms
~ =1· xc~s9'
I
3.5
[sin 0 it-tol-O.Ssin 2.Q It-to))
--Me Figure 14. Air-gap moment Me(t) in a genemtor.
30 20 10
/
"
.0
/
h
120° synchronisation error:
'-'
. [0.866 - 0.866 cos nt + 1.5 sin nt}.
cot -10
,- :f II KrtkOI ' 0.5
1 1.5 2 2.5 Crank angle k
4rr
I
3.5
•
4
Figure 1:J. Harmonic analyses of the tangential force FT(, and the radial force FRG for a cylinder of a four-stroke engine.
equilibrium can be severely disrupted by electrical faults in the mains or at the generator, or by switching and synchronisation processes. The moment exerted by the generator then contains additional constant and oscillating components. Studies show that the greatest shaft loading occurs if the terminals are short-circuited or during faulty synchronisation with an angular error of 120°. For this reason, only these cases are used in most of the appropriate design standards and regulations. Figure 14 shows the timedependence of the air gap moment Me(t) of a generator in relation to the nominal moment Mo for an undamped two-pole terminal short circuit and for a 120° synchronisation error. The time dependence can be calculated from the following equations [10]:
Two-phase short circuit at the terminals: Me(t) = M"
+ ~ . _,,_1_ cos r.p Xd + XTR
. {sin nCt - to) - 0.5· sin 2nU - to)}.
(42)
where x~ = the subtransient reactance of the generator, X'I'R ;;;: the transfonner reactance, X N ;;;: the mains reactance, in each case normalised to the generator impedance, cos cp = the power factor, Mo = the nominal moment and fl = the mains frequency. With nominal torque Mo and the alternating contributions varying with frequency of rotation or twice that frequency respectively, the time-invariant contribution is clear. The given exciting moments must be inserted at the appropriate place into the excitation vector FC t) of the shaft equations of motion. The torques of drives with electrical speed control are increasingly important. Here, exciting torques that pulsate can occur as a consequence of energy supply via converters (voltage changes) because these cause harmonics in current and Voltage. In [11], exciting frequencies as a function of the rate of revolution are given for two of the most widely employed types of drive Cslip-ring motor with static converter cascade, current converter with synchronous motor drive).
2.6 Mechanical Equivalent Systems, Equations of Motion In order to obtain calculated solutions or to interpret the results of measurements, mechanical equivalent systems are needed that reproduce the true dynamic behaviour with sufficient accuracy. The way to proceed in formulating the model is shown in Fig. IS. The starting point is a consideration of the real system (design drawing) where it must be stated, inter alia, where the system limits are to be drawn. After formulating the task and its limits, the equivalent mechanical system can be
Machine Dynamics. 2 Vibrations
:r--~ Modelling as
El{zl.,u{z)
~
~~i______~~ equations Interpretation
Continuously distributed model b
~) IIIIIIIIIIIII~I x{z
Figure IS. Procedure for the fonnulation of a model.
set up. Here, simplifications based on assumptions and experiences or observations derived from similar systems are made. The mechanical model should be as simple as pOSSible, but should contain all essential features. Model and reality should conform closely in as much as the required information is concerned. The aim is a minimum model that gives relevant data regarding the dynamic behaviour of the system while making use of the smallest possible number of degrees of freedom. In addition to the equivalent mechanical system, it is also desirable to establish the corresponding mathematical model, which in the case of oscillatory systems often leads to a system of differential equations with constant coefficients (see Eq. (1». These equations can be solved subsequently, and the results interpreted and discussed. If necessary, and where doubts regarding the solutions remain or where strong discrepancies from the real behaviour exist, the mechanical model must be revised. In formulating a mechanical equivalent system, the system structure is defined, and then the appropriate system parameters may be determined (Fig. 1).
Z.6.1 Structure Definition Several questions arise in the definition of a structure. It is first necessary to choose between a continuous system with distributed masses and stiffness, and a discrete system. This leads to partial differential equations in the first case, and to ordinary differential equations in the latter case. It is also important to decide whether or not linear of non-linear relations apply. Further, there is the question of how many degrees of freedom are required, what elements (springs, masses, dampers, rods, beams, plates, etc.) the system is to consist of, and what boundary conditions apply. Figure 16 shows various possibilities of modelling using the example of a shaft and a turbine wheel (see Fig. Z). The continuous system, with its infinite number of degrees of freedom, is a structure that closely approximates to reality, because the continuous distribution of masses and stiffnesses is taken into account. However, in the case of complex systems the solution of the accompanying partial differential equations is possible only with a very considerable effort and is therefore not feasible for practical purposes. Good approximate solutions can be found using discrete systems. In the discrete process, which has already become classical engineering procedure, the continuously distributed masses are replaced by point masses or rigid bodies, and these are countered by massless springs and dampers (spring-mass-damper system, lumped mass models). As is shown in the examples, certain lumped
Finite-element model
c
Mathematical model: Hx.Ox.Kx-Flt)
Figure 16.. Possibilities of model formulation using the example of a shaft and turbine rotor: a real system; modelled as: b continuously distributed model; c finite-element model; d spring-damper-mass model; 1 flexural spring, 2 mass.
masses (rotors) suggest themselves for representation by point masses or rigid bodies to which neighbouring smaller masses (mass of the shaft) can be added in proportion. Springs, massless torsion rods and bending beams, among others, are employed as elastic connecting elements. Recently, finite-element analysis has become extremely Significant (see BS). The FE method is very versatile and allows the treatment of any arbitrary one-, two- or threedimensional oscillatory systems. Any boundary conditions and any distribution of mass, stiffness and damping are also permitted. Each element is treated on its own, and the dynamic behaviour is described in the form of a forcedisplacement relation using forces and moments with displacements and torsions, respectively, at the nodal points (Fig. U;). This is accomplished by so-called initial functions in which the mechanical degrees of freedom at the nodal points appear as free parameters. The properties of the elements are gathered together in mass, damping and stiffness matrices. This expresses clearly that the properties of inertia, damping and stiffness are all taken into consideration simultaneously in a fmite element. Finally, the elements are connected at the nodal points taking into account all boundary and transition conditions and formed into the total structure. The appropriate mathematical model takes the same form as the discrete engineering model, and in general leads to a set of ordinary differential equations already presented in Eq. (1), if the system behaves linearly.
2.6 Mechanical Equivalent Systems, Equations of Motion. 2.6.4 Finite-Element Models
2.6.2 Parameter Definition If the structure of the oscillator), system and hence the form of the mathematical equations are given, the next step is to determine the values of the system parameters. or the elements of the matrices M, D and K, respectively. In determining the parameters, jmportant infomlation is taken from the design drawings (dimensions, material parameters, masses) and the laws of mechanics (moments of inertia, tlexural rigidity, torsional rigidity) are applied. However, for several machine elements or mechanisms (sliding bearings, seals, couplings) properly developed theoretical models describing their dynamic behaviour do not yet exist. In such cases, empirical procedures are often unavoidable and attempts are mad,- to determine the unknown parameters of individual system components hy a (parameter) identification process [12, 13 J 2.6.~
Examples oC Mechanical Equivalent Systems. Spring-Mass-Damper Models
shown in Fig. 17 can be derived. They art used. for
[I
(eJ,
0 0
()
m
0
()
"W] [" o
o
.
+
2
."<,
x, x
@"
M
..J,..
x
()
kl2
()
()
kl2
0
0
l~
-k 21
()
+
k21
k"
0 0 0 0
0 Of)p
[-k"
0 0
0 0
+ 0
-k12
kZ2
K
0
-OfJp () ()
D
][~J x
F, [ F F, F,
=
H:]x (43)
F(t).
The elements of the stiffness matrix can be calculated from the data on unit deformations and the determination of the corresponding forces. The matrix D contains gyroscopic effects that are proportional to angular frequency and polar moment of inertia Bp. The inertial matrix contains the masses In and the equatorial moments of iner~ tia ®a in its diagonal _More detailed directions for the deri~ vation of the equations of motion are to be found, inter alia, in [15J.
n
2.6.4 Example.. oC Mechanical Equivalent Systems: Finite-Element Models Finite-Element Models oC a Turbogenerator_ Turbine sets for electrical power generation of up to 1200 MW are no longer a rarity. The shaft is about 35 m long, weighs approximately 220 t and rotates at 50 revolutions per second in order to generate electricity at mains frequency_ The worst torque stresses in the rotor are
a 8,
81
~B=
-
k, d]
El-~
-
x,
xiII + c
Model oC a Shaft with Rotor (Ventilator). The oscillatory system introduced in Fig. 2 can be used to represent a simple model for the calculation of low-fn:quency tlexural oscillations. To this end, the mass is considered to be concentrated in the rotor and the elasticity of the shaft and the bearings are taken as a single entity. In the case of the rotor, torsion as well as excursion must be included in order to take account of the effects of rotary inertia (Fig. 18). In writing down the equation of motion for this model, the gyroscopic terms for the shaft as well as the inertia and stiffness terms resulting from the law of angular mon1cntum must be taken into account. The complete equation of nlotion is therefore 0
Unrestrained Torsional Oscillators with Two Rotating Masses. The torsional oscillator), hehaviour of machine equipment can often be described in good approximation by a linear mechanical equivalent system with two rotating masses \\rith a torsional spring and a torsional damper between the two masses (Fig. 17). ("'l, and Ell are the moments of inertia of the two machines (e.g. electric motor-compressor) about the axis of rotation and kl and d, represent the torsional stiffness of the spring and the torsional damping constant respectively of the connecting shaft or a coupling with torsional elasticity between them. The moments of inertia of a given body about a fixed axis is given by @= f r' dm and the torsional stiffness of a cylindrical rod by k = GIrl I (G = the shear modulus, I, = the polar moment of inertia, I = the length of the rod J. In general, details regarding stiffness and damping properties of the coupling can be obtained from the manufacturer (non-linearities in the couplings must be taken into account). If XI and X, designate the two degrees of freedom of rotation and MI(t), M,(t) the exciting moments applied to the rotating masses, then the equations of motion
b
example, to calculate the torque maxima in the drive shaft (coupling) that result during startup with an asynchronous electric motor [14 J .
XI
o
xiII +
K
xII)
X"F, xz,Fz ----
Fill
Figure 17. Unrestrained rotary oscillations with two rotating masses: a machine set (J electric motor, 2 compressor), b equivalent system, c equation of motion
Figure 18. Excursion of a
sin~ly
equipped rotor disc.
Machine Dynamics. 2 Vibrations
Figure 1,. Representation of the real-system tuebogenerator as a finite-element model: a arrangement (construction) (1 generator, HP high pressure, MP medium pressure, IP low pressure), b mechanical model, c torsion element.
c:
caused by torsional oscillations brought about by electrical perturbations in the generator (see J2.5.4) or in the grid. During the design of the machine, the designer must cal· culate as accurately as possible the stresses at the shaft cross-sections resulting from these cases. Since the rotor system of a turbine generator unit represents a mechanical system with several shafts, a finely detailed model is required for accurate mechanical predictions. Since the shaft has to be subdivided into 200 to 300 elements for this purpose, the finite-element model suggests itself as a mechanical eqUivalent system [2, 10]. Figure 19 shows, apart from the real system of a turbogenerator with the turbines and the generator, also the appropriate FE model with N - 1 cylindrical torsional elements. The following constant entities belong to any given "finite" element e of constant cross-section: p'e = the rotating mass assigned to the element, GI!f = the torsional stiffness, and Ie = the length of the element. With the help of local initial functions which are substituted into the work integrals (principle of virtual work), there can be constructed for each element an element
stiffness matrix (44)
and an element mass matrix Me =
p.
e [<
[1/3 1/6] 1/6 1/3
(45)
which are of the second order owing to the two local degrees of freedom (one angle of rotation for each element node). The torsional oscillations of the total system are globally described by the angles of rotation Xi
which, in each case, are introduced at the nodal points (interfaces between two elements). A system of (N - 1) elements has N global degrees of freedom gathered together in the vector x. The total matrices M and K are built up from the super· position of the elementary matrices. In the present chainlike structure, simple overlapping results, leading to a form of band matrix that is advantageous as regards storage requirements and calculation time. Damping constants in these systems are generally defined as "modal" (see J2.2.3). In branched systems with sets of gears the structure is not quite so Simple, but does not present a serious problem when the well-tried FE method is employed.
FiAite-E1ement Model of a Multi·Stage Centrifugal Pump. The trend of development in centrifugal pumps, similar to that of other machines, tends to higher rates of revolution, lighter construction and greater efficiency. For that reason, dynamic behaviour is of ever greater importance, principally in terms of impeller flexural oscillation. Finite-element analysis is used in the majority of cases where, apart from the inertial and stiffness properties of the beam elements (shaft), other factors must be taken into account, namely the fluid forces acting on the rotor in sliding bearings, on seal gaps and on balancing pistons, as well as the interaction between the impeller and the guide vane ring (Fig. 210). In the case of the beam elements, four degrees of freedom per node are employed in order to take account of torsions as well as excursions. Further, shear deformation, gyroscopic effects and damping associated with the materials can also be taken into consideration. Pump impellers are, as a rule, considered to be rigid discs. Sealing gaps in centrifugal pumps make spaces with
Figure :10. Representation of a multi·stage cen· trifugal pump as a finite-element model: 1 axial bearing (journal), 2 balancing piston, 3 impeller, 4 seal gap (interactions: impeller-ring of guide vanes), 5 shaft, 6 sliding bearing, 7 coupling.
2.7 Application Examples for Machine Vibrations. 2.7.1 Torsional Oscillator with Two Rotating Masses
unequal pressures fluid-tight. Here, leakage loss through the gap, which is approximately 200 to 300 fLm wide, is tolerated because the advantages of low friction and low wear are more important. The sealing gap has, however, a considerable effect on the vibration behaviour. The surrounding fluid exerts forces on the moving rotor with its excursions (radial displacements
Xl, Xl
and associated
speeds Xlo x, and accelerations X" x2 respectively) which contribute in large measure to the out-of-balance oscillations and the stability behaviour of the machine. These forces can be described by inertial, damping and stiffness coefficients in the form of a linear force-displacement relation:
[ (')0'
0]. [X"X Z ] + B2
x
M
[k-k -k]. [X' k
XZ
+
K
]~
[M' ]
x
M2
F. (49)
If no external excitations are present, the vibrations of
the system are described by the homogeneous equation of motion (50)
M·x+K·x~O.
The solution is obtained using the relation x ~ 'P·eiw,. It consists of the natural frequencies Wn and eigenvectors 'Pn which result from the eigenvalue problem
['P'] -_0
-k ] k-w 2B2 . 'P2 (K -
(46) To determine the dynamic coefficients, any of a number of theories may be used that attempt to describe the flow in the gap by means of various initial formulations [16, 17]. All theories have in common that they describe a displacement outward from the central position. The matrices have a skew-symmetric structure which is confirmed by measurement. In the case of a hydrodynamic sliding bearing (see F5), the shaft is supported by a pressure field set up by the rotation of the shaft. This results in a strongly non-linear relation between the forces and the movement of the shaft relative to the housing. Linearisation is possible for small movements:
[~:: ~;:] [~: ] + [:;:
:::] [::]
~ [;:].
~
(51)
O.
w'(-k(('),
+ ('),) + w'('),('),)
~
o.
(52)
From this the natural frequencies are calculated as
(53) If these results are resubstituted into the eigenvalue problem, the corresponding eigenvectors are obtained: (54)
(47)
The dynamic coefficients d i, and ki' result from the solution of the Reynolds differential equation or from empirical studies. They are usually given in non-dimensional form as a function of the Sommerfeld number So [18] (see F5.2). Since the static bearing load Fl is a function of the bearing compliances, the weight forces and the hydraulic forces acting on the pump impeller, the problem of determining the stationary position of the shaft for each condition of operation is non-linear. lbe interactions berween the impeller and the ring of guide vanes can be described in a similar way. Taking into account all the above effects, an equation of motion for the centrifugal pump can be derived by superposition of the equations for the elements Mx+Dx+Kx~F.
w2M) . 'P
The characteristic equation is obtained in the form det {K - w'M} ~ 0:
Discussion of the results shows some interesting aspects. Since the system is not bound to any number of degrees of freedom, natural frequencies with value zero (w,., ~ 0) result. They are rigid-body displacements as shown by the corresponding eigenvectors. In this modal displacement the torsional composite shaft does not deform, and no internal strains appear. The other two solutions represent elastic modal movements. Their natural frequencies and modal forms depend on the two inertial rotary masses @l and B2 and the stiffness k. Figure 21 shows the modes of oscillation. In the special case of ('), ~ ('), the system is symmetrical and the natural frequency corresponds to that of a single mass oscillation with spring stiffness 2k. If one mass becomes
(48)
The matrices M, D and K have a band structure and are, in general, asymmetric. In addition, some of the matrix elements depend on speed.
2.7 Application Examples for Machine Vibrations The solutions of the equations of motion (natural vibrations, forced vibrations) can be discussed with the help of a few examples. This will reveal effects that occur frequently in machine dynamics. 2.7.1 Torsional Osdllator with Two Rotating Masses Natural Vibrations and Modal Entities. The equation of motion for the undamped torsion model with two rotating masses (Fig. 17) was given in matrix form:
X,
'1',.2 _.
1111111111111111111111111111111
el~ '1'].4-'
w1.Z~O W3'4~±V~+~
~;,-
Figure 21. Modes of oscillation for rotary oscillators with two degrees of freedom.
Machine Dynamics. 2 Vibrations
very large in comparison with the other, this remains at rest and the natural frequency corresponds to that due to rigid clamping at that point.
bine set is frequently subdivided into several hundred elements in order to reproduce the vibration behaviour with sufficient accuracy (Fig. 19).
Forced Vibrations. lf external forces (moments) act, the torsional oscillation with two masses is described by an inhomogeneous differential equation. In order to simplify matters, damping is not taken into account:
Natural Vibrations and Modal Entities. The natural vibrations of the system are described by the equation of motion without external exciting forces. The D matrix is omitted in the analysis of the natural vibrations because of the weak damping:
(55)
M·x+K·x=F.
If the system is excited by a force F(t) the solution consists of an homogeneous contribution X hom and a particular contribution xp,n' The forced contribution to the solution xparr results from the solution of the inhomogeneous equation of motion by an initial formulation in the manner of the right-hand side of the equation. In the case of a sinusoidal dependence of the force FU) = F sin(f.lt), xp,n = it sin(f.lt) is obtained, with -k
k-f.l 2 8, ,
X=
]-'
1
k
(57)
(58)
F
f.l2(f.l 28,82 - k( 8, + 8 2)) k-f.l 2 8,
Mx+Kx=O.
The mass matrix M is always definite positive, since all the matrix elements used represent masses. The stiffness matrix K is semi-definite positive, since a rigid body displacement of the "torsional train" is pOSSible. In accordance with the (N x N) order of the matrix, N natural frequencies and modal forms from the solution of the eigenvalue problem are obtained:
],
F.
(56)
For certain exciting frequencies, the excursions increase steeply. Such is the case at f.l ="'1,2 =0, the reason for which is the absence of damping of the oscillations and resonance, if the exciting frequency f.l agrees with the next natural frequencies f.l = "'3.4' Further, the excursions at the point of excitation can vanish if, for instance, the mass 8 1 is excited with the frequency f.l2 = k18" or vice versa. Figure 22 shows the plot of the torsional vibration amplitudes i\, X, as a function of the exciting freq uency f.l. Since any periodic function F(t) can be represented by a sum of harmonic functions, the oscillations due to such excitations can be given as the sum of several contributions of the above force. For non-periodic excitation, closed solutions are fre-
The solution itself can only be carried out economically with the help of numerical algorithms, distinguishing between direct and iterative procedures. The eigenvalues and eigenvectors are determined by iterative processes from initial values that are changed by iteration, until a predetermined termination condition is satisfied. Example. The modal entities of a 600-MW turbosystem are being considered, with the torsion train subdivided into 250 torsion elements. Since torsional vibrations are frequently only very weakly damped, it is sufficient to consider the undamped system. Figure 2~ shows the five lowest natural frequencies (fn = Wn/21T) and the nonnalised eigenvectors of the turbosystem. The rigid body form of the zero natural frequency is not shown. In the first natural mode, the HP (high-pressure), MP (medium-pressure) and LPI (lowpressure) turbine oscillate at 18.19 Hz with respect to the LP2 turbine and the generator. The natural mode has a zero transit in the region of the coupling (vibration node). A further node is added in each further natural mode. The low natural modes are distributed over the entire shaft train, while in the case of the higher frequencies only individual part rotors are in oscillation.
quently found. Complex force functions can be solved by numerical methods, or calculated using polygons that are closed piecewise.
2.7.2 Torsional Vibrations of a Turbosystem The case of the shaft assembly train of a turbosystem represents a rather more complex example and is of central importance in any power station. Apart from flexural vibrations, torsional vibrations in particular form a decisive criterion for plant reliability. Once again, calculations are carried out using finite-element analysis. Here the tur-
31.04 Hz
""'4lJlIllllll
1~
4111111111111111111111111111111111
36.94 Hz
0~'IijJjJJJJjjjjJ~
-1
;~
57.22 Hz
-1
1f
118.7ZHz
~
_~f-----~ Figure 22. Normalised amplitudes of torsional vibrations of a tOfsional oscillator with two degrees of freedom as a function of the normalised excitation frequency.
Figure 2~. Natural frequencies and natural vibration modes for the turbosystem.
2.7 Application Examples for Machine Vibrations. 2.7.2 Torsional Vibrations of a Turbosystem
Forced Vibrations. Because of the many degrees of freedom, the solution of the equation of motion for forced vibrations not caused by harmonic excitations is very timeconsuming and frequently numerically inaccurate. It is possible to decouple the equations by coordinate transformation when, as a rule, the number of equations is also considerably reduced (modal analysis; see J2.2.4). Once the decoupled equations have been solved, they can be retransformed, thus obtaining the required results. DecoupUng is accomplished with the aid of the so-called modal matrix <1>, which is made up from the calculated eigenvectors. This leads to generalised equations for simple inertial oscillators, which can be solved very effectively. It is also evident from the right-hand side of a modal
equation how intensively the vibration of this natural mode is excited. In the modal calculation of the forced vibrations, damping in modal form is also taken into consideration. Example. The short-circuit response of the above 600·MW turbosystem is being considered. The angle of torsion of each degree of freedom is composed of the superposition of the part solutions of the modal single mass oscillators. With the aid of the element matrices it is also possible to determine the shear moments from the calculated torsions that are Critical for the design of the shaft train. Figure 24 shows the contribution to these moments from the various natural vibration modes. Their sum results in a maximum loading of the coupling of the generator amounting to four times the nominal moment.
11,
HP
5000
E
12
MP
LP2
NO 1
Gen
SR
400 0
co
E
'"
E E
0
0
%i1Il
~
+-
-4
~ :H-~-~ .f
-4
Mode3
0
,
~<">"'""
100
+
I/, 7:
1000
:::~.
~ :,. 1134
~;
'. , ;~ < I::-
~ %
/
c~ " 1::-. V ~ /'I~ 1134
I/,
1::.
1134
2182 kNm Figure :14. Cross-sectional moments in the shaft of a turbosystem following a short circuit.
2 1 134 Mode
1-34
S
Machine Dynamics. 2 Vibrations
2. 7.~ Flexural Vibrations of' a Shaft with Rotor In calculating the flexural vibrations of rotors, the gyroscopic effect often plays an impottant pan. In the model for the simpler rotor introduced above, the equation of motion was characterised by a skew-symmetrical gyroscopic matrix D (see J2.6.3).
Natural Vibrations and Moclal Entities. A harmonic initial formulation of the homogeneous equation of motion leads to a characteristic equation, from which four natural angular frequencies w can be calculated: mBaw4 - mBpflw3 - (k Z2 m + k l1 0a)W2
+ klllJpflw + (kllk22 - kf2)
= o.
(59)
In each case the corresponding modal forms represent circular movements of the point at which the shaft passes through the rotor, where the torsions may be interpreted in a similar manner. Derivation of an explicit presentation of the results for eigenvalues w requires considerable effott. A qualitatively striking feature is that w does not depend only on the physical system parameters but also on the angular frequency fl. Ibis dependence is the more
150
\, -------~----
pronounced the more strongly the rotor tilts during a natural vibration and the greater the moment of inettia [15].
Forced Vibrations. Since the natural modes in this model lie along circular paths, it is impottant in considering forced vibrations to observe the sense of rotation along such circular paths. The out-of-balance excitation which, in the case of rotors, is frequently the dominant one, also moves along circular paths about the axis of rotation. This means that only those natural modes are excited that rotate in the same sense as the shaft. These are termed ganged natural frequencies, and, in the other case, counter-rotational natural frequencies [15]. 2.7.4 Flexural Vibrations of' a Multi-Stage Centrif'ugal Pump The flexural vibrations of a complex structure are considered via another example. The model of a multi-stage centrifugal pump is characterised by asymmetrical systems matrices M, D and K (see J2.6.4).
Natural Frequencies and Modal Entities. The solution of the homogeneous differential equation yields
'P,
Figure 2S. Eigenvalues as a function of the rate of revolution and modes for a multi-stage centrifugal pump.
2.7 Application Examples for Machine Vibrations. 2.7.4 Flexural Vibrations of a Multi-Stage Centrifugal Pump
Figure 26. Amplitudes of the vibrations of a centrifugal pump.
complex eigenvalues An and natural modes
= an + iwn:
= ~e + i~nm.
(60)
Modern numerical methods can exploit the special band structure of the matrices and can also, if necessary, calculate some or all eigenvalues and eigenvectors. The imaginary part Wn of an eigenvalue gives the natural angular frequency of the system. The corresponding real part an is a measure for the damping experienced by a free vibration. If an >0, the oscillatory movements xU) increase, i.e. the pump is unstable. The eigenvectors describe the natural modal vibrations. Since complex excursions cannot be visualised, the two occurring complex conjugate eigenvectors are taken together and transfonned into a real rep-
resentation. The result is elliptical paths at the various nodes, so that the mode of oscillation can change duriog a period. A rypical representation of eigenvalues as a function of the rate of revolution and of the complex eigenvectors is shown in Fig. 25. Possible resonance points occur at the
out-of-balance
point of intersection of the plots of the natural frequencies and the start·up line (w = D), and the intensiry of reson· ance amplification can be judged by the damping factor an at the appropriate point.
Forced Vibrations. The most important forced vibrations in such machines arise from out-of-balance excitations. In this case the vector F(t) is a harmonic function depending on the out-of·balance moment me and the rate of revolution (see J2.5.4). The steady·state condition x(t) results from the solution of a complex system of equations and describes elliptic paths for the various nodal pOints of the model similar to the natural modes (K - D'M + i!1D)
k F.
(61)
As a rule, the main interest is directed to the maximum
amplitudes that can occur at the various points. These are shown in Fig. 26 as a function of the rate of revolution. The strong excursions of the end of the shaft at the resonance points that were also noticeable in Fig. 25 can be noted.
Acoustics in Mechanical Engineering D. Fimer, Frankfurt-on·Main
Machine acoustics deals with the occurrence and reduction of the noise caused by machines. It is a branch of the science of acoustics applied to the requirements of mechanical engineering.
3.1 Basic Concepts An extensive collection of concepts from the entire field
of acoustics can be found in [1 J. Here, those that are most important for the acoustics of machines will be explained.
Noise, Acoustic Pressure, Acoustic Pressure Level Mechanical oscillations with frequency components in the audible range (16 to 16000 Hz) are described as noise. Oscillations in air and gases are called airborne noise, those in liquids (water, oil) fluid·borne noise, and those in solids (i.e. also in mechanical structures) structure· borne noise. Noise in air and other gases as well as in liquids only propagates as compression waves. The alternating pressure Pl(t) that is superimposed on the static pressure is called the acoustic pressure and is the most important
Machine Dynamics • 3 Acoustics in Mechanical Engineering
possible to match the sensitivity of a sound-level sensor to the natural sound sensitivity of the human ear. In practice it is usual to carty out the weighting in accordance with curve A, the acoustic pressure level so weighted being given in dB(A). The acoustic pressure level measured in the vicinity of a machine is not an entity specific to the machine but depends on the distance of the measuring point from the machine. In the case of machines with pronounced directional properties, it also depends on the location of the measuring point relative to the machine as well as on the size and nature of the room (wall reflection).
Acoustic Power Level
-20 L-....L---1_...l...----L_L-.-'----'_...l..-----:-~ 0.0160.03150.0630.125 0.25 0.5 16
In contrast to acoustic pressure level, acoustic power level Lp is an entity specific to the machine. It is given by
f in kHz
(2)
Figure 1. Normal curves of equal acoustic intensity (DIN 46630, Sheet 2).
measurable quantity for airborne and fluid-borne noise. Measuring probes are microphones or pressure sensors respectively. The acoustic pressure is given as an effective value P averaged over a certain time and within a certain frequency interval. The range for P within which the human ear can perceive sound lies between PHS = 2 . 10-4 fLbar = 2 . 10-5 N/ m 2 (audible threshold at 1000 Hz) and PSG = 2.102 fLbar = 2 . 101 N/m2 (pain threshold). Since P can range over six orders of magnitude, and in order to arrive at smaller numerical values, acoustic pressure is not stated in absolute microbars or N/m2, but in a relative measure of power known as the acoustic pressure level Lp in decibels (dB). This is defined as Lp = 10 19 (p2/Pf,) dB = 20 19 (PlPo) dB,
(1)
where Po = 2 . 10-4 fLbar = 2 . 10-5 N/m2, the internationally agreed reference value for the effective value (r.m.s. value) of the acoustic pressure. The dynamic range of the ear (at 1000 Hz) is then !:.Leo< = Lp • SG - Lp • H , = 120 dB. At the frequency of 1000 Hz the acoustic pressure level in dB is the same as the acoustic intensity level in phon. The dependence of the acoustic intensity on frequency and acoustic pressure level is reI>" resented by curves of equal intensity (Fig. 1). The weighting curves for acoustic pressure level sensors (Fig. Z) are the mirror-image-like approximations for selected curves of equal intensity. With their help it is
where it is usual to put Po = 10-12 W as the value to which the acoustic power P is referred. The determination of the acoustic power level Lp of a machine is always based on measurements of the acoustic pressure level L p , because P - p-' (DIN 45 635).
Structure-Borne Noise, Velocity, Velodty Level In solids, in contrast to gases and liquids, compressive
stresses occur in addition to tensile stresses. Therefore, in the case of structure-borne noise, transverse waves moving parallel to the shear strain appear in addition to the longitudinal or compression waves. Various other types of wave are composed of these two, of which flexural waves are the most important. These cause the largest displacement normal to the surface of a body, e.g. a plate, and in general make the strongest contribution to machine acoustic radiation, because only these normal oscillations can be transferred to the surrounding medium. They are excited not only by the transverse forces but also by bending moments. The most important measurable variable for structure-borne noise is the velocity of oscillation or the structure-borne noise velocity vet) normal to the radiating surface of a noise generator. Calculation of the velocity vet) from the acceleration aCt), obtained from measurements using piezoelectric sensors, is carried out using the relation
aCt) =
v(j) = ii(j)/(27rf)
CD "0
j-lO
/
.-{
. -/
ij;,
I
-'-20 I
I/o
II
/
c;;
~
/
I~c
I/~
""
"1\
4
6
810 2
(4)
applies for the r.m.s. values in a frequency band that is not too wide (e.g. third octave band) and whose centre frequency is f The velocity too is usually stated as a velocity level L." given by L.,
= 10 19 (iJ'/;ro) dB = 20 19 (v/vo) dB.
(5)
The reference value Vo has been chosen as 5 . 10-8 m/s.
V
I
(3)
so that
10
-30 10
d
dt v(t) ,
Radiation Coefficient 2
4
f in Hz
6
8103
4 6
810 4 2104
Figure 1. Weighting curves for acoustic level meters. The curves, with their difference relative to zero, state to what degree the acous-
tic pressure level at a given frequency is evaluated at a lower or higher value than would have been the case without weighting.
This relates the radiated power P to the mean-square velocity ij2 averaged over the total radiating surface S and is defined as a non-dimensional factor [2), P(j) u(j) = Pr.,cLSiJ'(j) ,
(6)
3.2 The Generation of Machine Noise
where PI. is the density of air and CL the velocity of sound in air. The mean-square velocity iJi is determined from N
iJ2(f) = (liS)
2: S;i!f(f) =
N
(lIN)
1=1
2: jj~(f),
(7)
1=1
where VI is the velocity of the tth of a total of N (equally large) area elements S" or the velocity at the Ith measuring point on S= ~ S,. Equation (6) is at the same time the definition of and the rule for measurement of the radiation coefficient.
Receptance, Impedance, Structure-Borne Noise Factor The (structure-borne noise) receptance is the ratio of generated velocity v to exciting force F and is, for instance, termed input admittance hE = ve(xo)1 F when Ve is the velocity at the point of excitation x o, or transmission admittance ho = v(x)IF, when v is measured at any other required point X. The inverse of the receptance h is called impedance Z = 11h =FIv. The mean square of the transmission receptance hl.m is important in radiation and is determined from the mean-square velocity iJi (Eq. (7)) of the entire surface from (8)
The above expression multiplied by the radiating area S, Sho.m is called the structural noise factor [8]. Investigation of structure-borne noise also requires the measurement of forces. In view of the range of acoustic frequencies, piezoelectric transducers are preferred for this purpose. The force level LF is defined as LF = 10 19 (PI F~) dB = 20 19 (F;Fo) dB,
(9)
where Fo = 1 N. Equation (8) may then be rewritten in a way that is simpler to use from a practical viewpoint: 10 19 (h~.m/hbo) dB =
T" - L
F•
Spectra The radiation coefficient and receptance are frequencydependent properties of a machine structure. Their determination as shown by Eqs (6) and (8) also requires that the force, velocity and acoustic power be ascertained as a function of frequency. TIlis frequency analysts is carried out by harmonic analysers which measure the r.m.S. value of the time-variant signals within a fixed or a variable frequency interval. The most frequently used types are third octave analysers and narrow-band sinusoidal tone or FFf analysers. They yield the amplitudes of third-octave or narrow-band spectra. The phase information lost in the case of analysers using filters normally plays no part in noise measurements. Sound generation processes in machines run periodically with speed, so that velocity and acoustic pressure are also periodic. Periodic processes possess a line spectrum. 11lis means that apart from the fundamental frequency fo = frequency of rotation in S-I, whole number multiples (harmonics) nfo also appear. The amplitudes of these spectral lines for f = nfo are given by the (continuous) pulse spectrum, which is determined by the time dependence (pulse form) of the corresponding measured quantity over one period. For the mathematical relation between these two spectra and the resulting measurement technique see [3]. It is frequently the case that several working processes with mut·
ual phase displacement take place during one revolution of the
shaft, e.g. in gear trains (meshing of teeth) or in hydraulic pumps. The fundamental frequency f~ is then given by frequency of rotation multiplied by number of teeth, or frequency of rotation by number of cylinders, and the line spectrum of such machines contains the corresponding hannonics nf~. It happens only too frequently that this main spectral line is found in measured spectra surrounded by a series of subsidiary lines, whose interval in each case is a/;;:: 10' These occur when the time dependencies of the sound-generating processes per working cycle are not exactly equal (amplitude modulation) and/or the frequency of rotation fluctuates (frequency modulation). Apart from this, measured spectra often have a back· ground of a continuous, more or less wide·band noise spectrum. This is caused by stochastic processes, e.g. frictional forces.
Shielding, Damping
Damping is the transformation of acoustic energy into heat. Shielding of airborne noise is achieved by means of porous or fibrous absorption materials with a high coefficient of absorption a. Damping of structure-borne noise is achieved by means of sound deadening layers or composite metal sheets with a high loss factor .,.,.
Shielding is the prevention of sound propagation. In the case of airborne noise, this is achieved by sound insulating walls (e.g. casings), in the case of structure-borne noise, by step changes in impedance (e.g. decoupling by rubber or spring elements). The transmission loss ALD is defined as the difference in level between locations in front of and behind a "shielding barrier", i.e. by
The insertion loss ALE by contrast describes the effect of the sound insulating method and is defined as the difference in level evaluated at a position on the far side of a barrier in relation to the point of sound generation with and without the sound insulating measure,
e.g. the level difference of a machine with and without casing. It is always true that 0 :5 ALE < ALn. In order to make ALE > 0, energy must be destroyed by damping on the sound generation side of the sound barrier. 11lis is required by the law of conservation of energy: no barrier effect without damping.
3.2 The Generation of Machine Noise Airborne Noise 111is is caused by machine processes, which in turn cause fluctuations in the air pressure and hence airborne noise. Such noise is generated [4] by the following:
Aeropulsive Sources. These displace the surrounding air in a pulsating manner (examples are induction and exhaust noise).
Aerodynamic Sources. These occur where sound is generated as a consequence of interrupted air flow. This manifests itself in a locally varying velocity profile (e .g. zones downstream of cascades, jet zones behind nozzles), which causes pressure fluctuations near objects moving at right angles to the flow, and these are propagated as airborne noise (e.g. fans, perforated disc sirens) or in an eddy zone which forms in the wake of an obstruction in the flow and directly generates sound (e.g. flow noise).
Thermodynamic Sources. Here the energy supply resulting from exothermic processes causes the expansion of numerous small-volume elements of gas with a
IB
Machine Dynamics. 3 Acoustics in Mechanical Engineering
corresponding buildup of pressure (e.g. noise from explosions, or electric discharges).
Structural Noise Generation In most machines, the greatest contribution is made by the initial excitation of noise in the elastic structure of the machine, which is then transferred from the external surfaces to the surrounding air and radiated as airborne noise. Structure-borne noise can be excited in two ways [4], as follows.
Force-generated noise is caused by changing motive forces which elastically deform the stressed parts, thereby exciting mechanical oscillations in the machine structure. This forced excitation does not occur exdusively at the point of action of the force, e.g. at the pressure face of a cylinder, but also encompasses other parts that are jnduded in the transmisston path of the force.
Velocity-generated noise, on the other hand, occurs in those parts of the machine structure that are not themselves stressed by the transmission of changing motive forces but are fixed to vibrating parts excited by these forces. They undergo base excitation at the points or edge of attachment where the vibration displacement or the vibration velocity is intposed upon them. This, as a rule, is the case with oil pans, protective panels and other mounted and relatively light components. Noise-Generating Forces These occur in numerous forms in mechanical engineering. The most intportant forms are: alternating pressure forces as in i.c. engines, compressors, hydraulic machines and piping systems; inertial forces such as out-of-ba1ance forces in rotating machine parts, or inertial forces in reciprocating components (coupling links in cam drives, slides in working machines); inertial and spring forces in cam gears; alternating forces caused by discontinuous force transfer such as the tooth-transmitted forces in gearing or the support forces in roller bearings; impulsive and intpact forces on intpact of machine parts as, for instance, when dearances are exceeded during roller bounce on cams, or of valves bouncing on valve seats; magnetic forces in electric motors, generators and transformers.
to develop the solutions of the (inhomogeneous) differential equations of vibration (e.g. the differential equation for flexural waves; see A4.2.4) in terms of (orthogonal) eigenfunctions of the oscillatory system (e.g. a plate) (theorem of series expansion). Complex structures, induding three-dintensional structures, can be dealt with by the finite-element method (FEM) , but the calculation of entire frequency plots requires a great deal of computer tinte. A first formulation for an estimate of the structural noise factor of plates, based on the theory of structural noise and treated in detail in [2], can be found in [3]. An improved estimate with regard to frequency dependence is investigated in [5]. According to this, the structural noise factor below the first natural frequency of a plate structure (f f1) it decreases on average at - 10 dB per decade (Fig. 3). The expected value of the structural noise factor (independent of the point of excitation x o) averaged over frequency, for f < J,., is given according to [8] by --II/, 1 , (Sh5 ) = - - - - - - - I/B" - lIh 6 .q 16" 1 + ..,2.fl m" ~m"B" '
and for f> f1 by
(Sh5e) = 1 ~ _ 1/(m" . 16"f..,m" m"If'
These working forces F(t) can be unequivocally resolved into an amplitude and a phase spectrum by Fourier analysis. The amplitude spectrum shows the exciting amplitudes as a function of the frequency f, and is called the excitation spectrum F(j). For the assessment of changing forces see [23].
The Behariour of Body Noise in Structures Structural noise amplitudes are proportional to the excitation amplitudes, but their ratio is not the same for all frequencies. For instance, amplified structure-borne vibrations occur in frequency ranges where the machine structure has natural frequencies. A measure of the frequency behavtour body noise of a structure when excited by forces is the structural noise factor Sbij,m(f) of Eq. (8). Closed numerical solutions for the structural noise factor are obtainable only in those cases where it is possible
lIh',
where m" = phis the mass, If' = Eh'/12 is the stiffness per unit area, p is the density, E is the modulus of elasticity, .., is the loss factor and h is the plate thickness. The application to ribbed plates and box-shaped machine casings is treated in [6].
Radiation of AhiMtrne Noise The structural vibrations normal to a radiating surface are not transformed into airborne vibrations equally efficiently at all frequencies. A measure for the radiation behaviour of a radiator is the radiation coefficient u(f) of Eq. (6). Even radiators whose surfaces vibrate entirely in phase, as for instance the zeroth-order spherical radiator, or the piston membrane, possess a frequency~pendent radiation coefficient. For the
above-mentioned spherical radiator this is given [7] by
Furthennore, all those forces give rise to noise which occurs during the processing of workpieces: separating forces in metal-remov-
ing production processes (milling, turning, drilling, grinding, sawing), as well as those occurring during cutting and stamping; deformation forces in material-funning manufacturing processes (fOrging, riveting, hanuuering, pressing).
~m"B") -
u. -
I'II~ = {f'Ifl" for 1<1.,
• -1'" f'1fl,
(10)
for I> /0,
1,
where I. = cd (21fR) is the cutoff frequency of the spherical radiator and R the radius of the sphere. The radiation coefficient U KN of the piston membrane is described quite well by the two asymptotic approximations of Eq. (10) if R and/. of the equivalent spherical
~ .5
120
~o
100
il~~~i
80
N-tZ::.'5
i5i
V '~
60
40 ~i;" 20
L.......:...J
--
- -
/ I
-
I
"".,... ~!
~
103
!f1
i~ 1
5
10 4
f in Hz
a f-'~
. i' i f3
'q Ws
1
Fipre 3. Structural noise factor of a circular plate excited at its centre (R = 60 mm, b = 10 mm, 11 = 10- 4 [5]): a exact plot of Sbfi.m(f), b smoothed plot (expeered value) (Sb6.m).
3.3 Methods for Reducing Machine Noise
radiator (equal volume flow) are determined. If SKM is the area of the piston membrane then R-= (5KM /2'IT)1/2 applies
According to [8]. the radiation coefticient Up of a plate which has been excited to flexural oscillation attains at most the radiation coefficient a KM of its equivalent piston membf'.me below the limiting frequency fg but may be reduced below U KM between the transmission frequency fo = f~/f. and the limiting frequency fg (short-circuit region) as a result of the interaction with regions of the plate vibrating out of phase (Fig. 4). The reason for this is that the propagation velocity c. of flexural vibrations is frequency dependent (dispersion): whereas (', = constant and hence ,1.= c,/f- 11f, c. --J] and thus A. = cBlf - 1/f] (Fig. 5). The frequency at which Al = AB is called the (flexural wave) limiting frequency. fg. As soon as A, > AB , i.e. f < f g , acoustic coupling occurs between the wave peaks and the wave troughs and this brings about a more or less pronounced hydrodynamic short circuit and correspondingly reduced radiation. For this case, the actual radiation coefficient for the plate is given according to [8] by
(II)
(lla) if",,, only depends on the size of the plate via fo, whereas the radiation coefficient (Tp in the short-circuit region also depends on the plate thickness, because
a~'~ 111~/2
plates has been calculated. The radiation coefficient of a box-shaped casing is then given by CTK
({T'o,
~, 1).
Basic Equations Cor Machine Acoustics The basic equation of machine acoustics can be derived using the definitions given in H.1. The radiated acoustic power in a given frequency range with centre frequency f, when excited by a force, is P(/)
= PIC! u(j)Sh[;.m
(/) p(/).
(12)
Here Pl_ is the density and Cl, the velOCity of noise in air, CT(f) the coefficient of radiation, 5h f '_m(f) the structural noise factor and F(f) the cm.s. value of the exciting force, which is known from the excitation spectrum. Written in level form, and taking account of weighting curve A which describes the frequency-dependent sensitivity of the ear, Eq. (12) becomes L,,(/) ~ L,(!) + [10 Ig (Sb[;m(j)/S" hi,.,,) (13)
where F,,::::;: I N, h('Il:::: '5, 1O--B mjs N, 50 -= 1 m 2 and dLA(j) is the level correction for the A weighting as in DIN 4S 633.
The structural noise factor and the coefficient of radiation are thus "weighting functions" and, through their filtering action, the airborne noise spectra of the machine noise are derived from the excitation spectra of the exciting forces. Finally, the total level LpA of the radiated acous· tic power is obtained if the levels LpA (f.) occurring at all the exciting frequencies!; as calculated by Eq. (13) are added logarithmically, i.e. with respect to power, to yield LpA = 10 19
(EI/lm")-'>/" -- h·'>!2.
The radiation coefficient of (box-shaped) maf-hine casings can be calculated from the radiation coefficients of the plates which form the pans of the casing [R]. It is given by the mean radiation coefficient;;: of the component surfaces, which are to be regarded as independent with respect to radiation as long as AI. is less than half the perimeter of the box (when the component plates are decoupled). When Al exceeds the perimeter of the hOUSing, the component surfaces are coupled. The radiation <:oefficient of a box can then be estimated in accordance with 181 as
-= min
(~
100"W!lOdB) dB.
(14)
The total level of several individual noise sources (independent of and decoupled from each other) is also calculated by this law of logarithmic level addition. In order to do so, the acoustics levels LpA.i of the i individual sources are put into the exponent of Eq. (14).
3.3 Methods for Reducing Machine Noise Survey
where lJ is the perimeter and J~ the limiting frequency of the individual component plates from which the mean value Uj.f'/~ of all the
10
h, ~ h2~ h3~ h4~
Methods for reducing machine noise are subdivided as follows.
4'10 2
1.5 mm 3 mm 6 mm 12 mm 1}
~
co
co b~
~~:
8 6
?
f---
,"'-
~~ ~"'- ~ ~
~
10' 8 6
--
.......
-- f--- -
--'-----"---'-6--'-6....1 10 4
Figure 4. Radiation coefficient CTp of plates of size 1000 X 700 mm 2 as a function of thickness b 18}, with piston radiation region 1< ft-· short-circuit region If) < Is, region of full radiation f > II!:
~,
r--..,. "'-~
~.
~
E'
. . .S 1
<"<~
-10
-3010",---"""",-L-..i6-'-a....I10""2----'--------.l4---'-6---'-a....110""3f in Hz
-
~~, 10 2 ~O 'IJ)
4
-
6
al03
fin Hz
f--- -
" " R. r......
Figure S. Flexural wavelength AB and wavelength in air AL as a function of frequency land plate thickness h, valid for steel and approximately valid for aluminium [2].
Machine Dynamics • 3 Acoustics in Mechanical Engineering
Primary Measures. These act ditectly on the generation of noise, that is to say, are applied to the noise source itself. In the case of airborne noise, this means an intervention in the generation of fluctuating pressures; in the case of structure-borne noise it means a reduction of the exciting forces (exciting spectrum) as well as alteration of the structural noise behaviour and the radiation (admittance and coefficient of radiation).
Secondary Measures. These are intended to reduce the airborne noise after generation and thus act either on the emission of noise from the machine (silencers, casing) or reduce the noise emission, e.g. at the operating point, by space-acoustic means. Secondary measures are meant to block the acoustic energy (barrier action) and then to destroy as large a part of it as possible by absorption (damping). Rules for low-noise design can be found in [11, 12], practical examples in [4, 13, 14J. Measures Applied to Sources of Ai1'borne Noise
Aeropulsive noise generation due to pressure equalisation processes is reduced, when pressure equalisation, e.g. via a valve, proceeds not in sudden bursts but slowly and steadily. Aerodynamic noise can be reduced by avoiding intemlptions to the air flow. At the inlet of flow machines, care must be taken to see that a constant velocity profile is maintained (no struts, gratings, guide vanes in front of fan rotors, inlet flow kept turbulence-free). If shadow zones cannot be avoided, then parts that move at right angles to the flow (e.g. a rotor), must not be located directly behind the obstruction, because at this point local variations in the velocity profile are at their greatest. For instance, the optimum distance from fan rotors according to [IJ is calculated as S" = 0.03(u/m S-I)' mm, where u is the peripheral velocity of the rotor.
period) [3J. Hence, if the slope dF/dt and the curvature d 2 p/dP of the time dependence of the exciting force can be reduced, noise levels from as Iowa frequency as a few hundred hertz can be decreased without affecting the working principle of the machine. A simple estimating procedure for the excitation spectrum that is developed in [3 J and shown graphically in [16-18J makes it possible to ascertain which properties of a force-time function must be altered in order to achieve a reduction in noise. Noise excitation by pulse forces is the smaller [3 J the smaller the transmitted pulse A = mv (i.e. the smaller the moving mass m and/or the smaller the relative velocity v between machine parts that impact each other). Reduction of the pulse mlv I to m 2 v, lowers impulsive excitation by
III
= [20 19
(m 2/m I ) + 20 19
(V 2/V I )J
db,
(15)
i.e. by - 6 dB when halving the mass m or the velocity v, by - 12 dB, when both together. In extending the duration of the pulse from 7',.1 to 7',.2 a reduction in the pulse excitation may also be achieved. From /'.1 = 1/7',.1 onwards, this amounts to (16) i.e. to - 6 dB on doubling 7',. Longer pulse durations are achieved by means of elastic intermediate layers or by increased compliance at the point of impact. 'The pulse is then damped by spring action and hence becomes "softet" . It is generally true [3 J that the longer the pulse duration 7' of an individual force application, the sooner the excitation spectrum flattens off. For this reason, simultaneous excitation (small 7') is avoided by applying the wedge principle (spiral cutter, oblique shearing, clipping punch with ridge cut, helical gearing).
Measures Applied to the Machine Structure The aim of these measures must be to keep the amplitude
Silencers
Dissipative silencers, like absorption and relaxation silencers, convert acoustic energy directly into heat. Both are relatively broad-band, generdting only a little back pressure. At low frequencies, however, they require relatively large cross-sectional dimensions of the acoustically effective lining and a rather extensive length if large insertion losses are to be achieved.
Impedance silencers, which include all types of resonance and interference silencers, primarily bring about a screening of the sound in that they introduce points with step changes in impedance into the flow channel (step changes in cross-section, coupled resonators, bypass lines) at which the sound waves are reflected. These reflections are associated with a correspondingly high back pressure; they occur, however, only at certain frequencies. Impedance silencers have to be tuned and are effective over relatively small bandwidths. Further information on silencer function can be found in [15J and on design in [4J.
Changing Excitation Spectra The excitation spectrum above the frequency f = 2/7' due to noise generating motive forces, is determined by the first and second and occasionally also the higher differential coefficients of the force-time functions (7' is the duration of the application of the individual force during one
of structural noise excitation force as low as possible at
radiating surfaces. Power flow due to the motive forces must therefore be confined to a narrow, massive and rigid region and not allowed to reach radiating outer surfaces.
The principle of functional separation [11 J should be applied. Forces must be absorbed in the interior of the machine, and walls with protective and sealing functions must be attached to force-transmitting structures by means that insulate them from structural noise. At the points where the forces are applied, input impedance should be increased by means of mass concentration [13, 19J. At the points of base excitation (velocityexcited structural noise) on the other hand, decoupling by means of rubber or spring elements (structure-borne sound insulation) must be applied. Increasing mass m" (density of materials, additional masses), stiffness B" (E modulus, ribbing) as well as the thickness h of the walls of the casing affected by flexural force (flexure B" - Eh') reduces the structural noise factor Sh6.m. At the same time, the radiation coefficient ITp in the short-circuit range is also reduced by m" but is increased by B" and h. The effect on noise generation can therefore be assessed only by considering Sh o.m and (]' simultaneously (Fig. 6). Since the effect evidently differs in magnitude depending on frequency range, the excitation spectrum P(f) must further be weighted by Sh,~.m and (J" if possible noise reductions are to be ascertained. If the exciting force leads primarily to tensile loading. the structural noise behaviour is mainly determined by the tensile stiffness (- Eh I ). In comparing different materials
3.3 Methods for Reducing Machine Noise
10
I
Rocketiaunch
~t'
Painful
/
/
~Ii "'2 ·4.5d8 .,. I- f""f ~
/V V-
Zfu.,
f,
I
I'--
fu"
90
I
Extremely
II
V 0 4 6 810'
-,
8dt . tli1 lk JVI ,
6
I
I
I
Ii .., I
, fin Hz
11
!
Ii
6
II
fdBrn
t{2-1-....
I
810'
sID 3
I
111 ,
6
conversation
L~ threshold ./ of auditory danger limi of aud' cry recovery .... Pelturbing effect on 10 nervous system
90
as
6lI 50
Quiet
Gentle radio muSIC Whispering
Upper limit for concentrated mental worle
<0
30
Rustling a leaves
!II
Silent
Relaxation, sleeo, rest
'0
Auditory threshold
Complete silence
al0'
dB1A1
(m" - p, B" - E), the effect of the dimensioning design rules that apply to the thickness h in order to achieve rigidity must also be considered. Relevant examples can be found in [4, 17, 18, 24, 25], and examples applying to product lines in [24, 26].
Damping of structural noise by sound-insulating layers
and composite metal sheets reduces the structural noise factor in the natural frequency range of the structure in accordance with the expression (17) 1)
o iceworle,
Moderalely
I
Figure 6. The effect of doubling of the thickness h on the coefficient of radiation Up and the structural noise factor (S~~_m) of a steel plate of size SOO X 350mm!, 'Tj = 0.1.
where
Driver's cab of loery Intense loud road traffic
Very
2f, fa
,..
PaJll threshold, acute auditory damage
at 100m
1
-40
Jet aircraft
Latge f()(ge Unbearable hammer Compressed air hammer Metal W()(king shop
~
/
SO' - 20
-30
J
h, ~ 3 m~ h, ~ 6mm
1mB
is the loss factor and acts as a measure of the
damping [2]. The reduction in noise through increasing
the loss factor YJI to 172 is thus on average
Figure 7. Noise conditions, perception and effect.
Shielding. The input attenuation 11LE as a measure of the effectiveness of an enclosure (see J3.I) depends on the (mean) sound attenuation factor R of the wall (transmission attenuation, property of the wall), on the (mean) absorption a as a measure of the (necessary) losses in the enclosure (absorbent lining, internal losses of the walls excited to structure-borne sound, absorption at high frequencies by the enclosed volume of air) and on the aperture ratio q = 5 0 /(5" + Sw), which is able to decrease conSiderably the possible effectiveness of the enclosure (keyhole effect). So is the entire unavoidable aperture area (shaft penetration, piping systems etc., transportation openings) whose transmission attenuation factor D can be matched to the wall attenuation factor R by means of silencers placed upstream. Sw is the remain-
ing wall area of the enclosure. [20] gives !lLE as i.e. - 3 dB if the damping is doubled. In this case, however, attention must be paid to the fact that the output damping of a structure is, as a rule, not determined by damping in the material used but by frictional damping in mounting and contact regions (screw assemblies, tolerances, bearings, etc.), which is often even greater than the damping in cast iron, and, further, that in relation to noise radiation an optimum loss factor exists in the shortcircuit region [4, 17, 18].
Enclosure of Machines High-powered machinery or numbers of smaller machines often generate noise at dangerous levels (Fig. 7). Noiseinduced hearing damage is therefore one of the main industrial injuries nowadays [22]. In view of legal requirements, the designer is therefore faced with the task of designing low-noise machines and manufacturing processes. However, primary noise reduction, in the sense of the design possibilitie.s already discussed, is not always sufficient to meet the required limits. In this context, enclosure, the function of which is based on shielding and damping, offers itself as an effective secondary measure.
!lLE = {- 10 19 [(I - q)'
+ q.
IO-f)/HldB]
IO-RIlOdB
+ 10 19
[a· (1 - q)
+ q]}
dB.
(18)
Since the wall attenuation factor R and the coefficient of absorption" (and in the case of silencers also D) are frequency-dependent, the input attenuation is frequencydependent: !lLE = !llE(f). For this reason, the machine structural power spectrum must always be weighted using the "filter curve" !lLE(f) of the enclosure in order to determine overall level reduction. The statement of a (frequency-)averaged attenuation value for R and also for !lLE is not a measure of the attainable attenuation noise reduction and is often too high. Below its limiting frequency fg = 12000 Hz/(hlmm) the wall attenuation factor R for enclosure walls made from sheet steel of thickness h and shortest edge length I lies [4, 20} between min(RlUin,Rmax) and R max , where Rm;n
= [151g ({/Hz) + 5lg (hlmm)
Rm=
= [20 19 ({/Hz) + 20 19 (hlmm)
+
IO 19 (lIm) - IS] dB,
- 27.5] dB.
(19)
(20)
Machine Dynamics • 3 Acoustics in Mechanical Engineering
With regard to cr, it should be noted that sheet metal enclosures without absorbent lining possess absorption coefficients Ct' = 10- 2 that arc practically independent of frequency and therefore, for b = 1 mm, have a considerable insertion loss 6.L£(f) > 10 dB (see [201 ). In the case of absorbent lining, the following applies [20];
" = {o.ss ~.95 VIII,
for I> Id for f
1
(21)
wherefd is given by /d = 5460 Hz em/d. If the largest possible value for a above a certain frequency f' is required, and with it the smallest possible loss of shielding action, then f' = /d should be substituted and the required thickness d of the absorption layer, and hence the specific flow impedance, calculated. This is given by SoPt
= (80 to 240 g S-l
cm--l)ld.
8
(22)
A suitable absorption material can then be chosen.
2
2
Soundproof Boxes. These are of the usual form of machine enclosure (Fig. 8).
Lining. This is provided by sheet metal, 1 to 1.5 mm thick, with an absorption layer 10 to 15 mm thick. Its sound attenuation is 6 to 12 dB from 125 to 8000 Hz. Material data are: panel weight 0.5 to 0.7 N/m2, temperature range - 25 to + 80°C and thermal conductivity 0.4 to 0.6 W/(m K). The surfaces must be resistant to aggressive media like petroleum, mineral oils, hydrochloric acid, caustic soda, and solvents.
Construction. 1be boxes should be as light as possible and as simple as possible to assemble. They should be watertight outside and inside and satisfy local fire regulations. They must be equipped with observation windows and be easily accessible at important points. To dissipate any heat generated by the machine, it is frequently necessary to provide an acoustically damped air-conditioning plant. Further, attention is required at the entry points of the supply lines for the operating media and control elements. They must be implemented in flexible form for machines with intense heat radiation or those mounted on resilient supports. They must be acoustically tight, since the acoustic level increases sharply in the interior of the enclosure. Directions for the construction of enclosures and attenuating feedthroughs, as well as working diagrams for their design, can be found in [4, 21 J.
a
/
____1/ K
,
-- --
100 90
~
//
1
.1,
99dB !AI
'"
1" ""
60
b
5063
115
250
~
~-:7B dB !AI
"
500 1000 1000 4000 BooO f in Hz
Figure 8. Single-casing movable enclosure for a high-speed wire bolt press: a construction (1 high-speed wire bolt press, 2 movable part of enclosure in tunnel form, 3 ftxed rear wall, 4 ftxed front wall, 5 guide frame, 6 bolt delivery opening (implemented as a sound-damping channel), 7 bolt-catching receptacle, 8 outlet or exhaust air duct for heat dissipation and oil disposal with forced ventilation), b acoustic octave level ['X:I as a function of frequency f full line, without enclosure; dashed line, with enclosure.
References ]1 Crank Operation, Forces and Moments of Inertia [1 J Haug K. Die Drehschwingungen in Kolbenmaschinen. Springer, Berlin, 1952. - [2J Kramer E. Maschinendynamik. Springer, Beriin, 1984. - [3J Maas H. Gestaltung und Hauptabmessungen der Verbrennungskraftmaschine. In List H. Die Verbrennungskraftmaschine, vol 1. Springer, Vienna, 1979. - [4J Woschni G. Thermodynamische Auswertung von Indikatordiagrammen electronisch gerechnet, MTZ 2517, 1964; 284-9. - [5J Kiittner KH. Kolbenmaschinen, 5th edn. Teubner, Stuttgart, 1984. - [6J Maas H, Klier H. Krafie, Momente und deren Ausgleich in der Verbrennungskraftmaschine. In List H. Die Verbrennungskraftmaschine, vol 2. Springer, Vienna, 1981. - [7J Hasselgruber H.
Massnahmen zur Verbesserung der Laufruhe von Verbrennungskraftmaschinen insbesondere von Schleppermotoren. Landtechnik 1965; 15: no. 1. - [8J Schmidt F. Schwungcader fur Grossdieselmotoren. VDI-Z 1930; 74: 230. - [9 J Sass F. Bau und Betrieb von Dieselmaschinen, vol 2. Springer, Beriin, 1957. - [lOJ Frohlich F. Kolbenverdichter. Springer, Berlin, 1961. - [11 J Haffner KE, Mass H. Theorie der Triebwerkschwingungen in der Verbrennungskraftmaschine, vol 3. Springer, Vienna, 1984. - [12J Haffner KE, Mass H. Torsionsschwingungen in der Verbrennungskraftmaschine. In List, H. Die Verbrennungskraftmaschine, vol 4. Springer, Vienna, 1985. - [13 J Kramer 0, Jungbluth G. Bau und Berechnung von Verbrennungsmotoren, 5th edn. Springer, Berlin, 1983. -
Machine Dynamics. 4 References
[14] Mickel E, Sommer P, Wiegand H. Berechnung und Gestaltung der Triebwerke schnellaufender Kolbenkraftmaschinen. Konstruksionsbucher, vol 6. Springer, Berlin, 1942. - [15] Kohler G, Rognitz H. Maschinenteile, pt 2, 7th edn. Teubner, Stuttgart, 1986. - [16] Schron H. Die Dynamik der Verbrennungskraftmaschine, 2nd edo. Springer, Vienna, 1947. - [17] Waas H. Fedemde Lagerung von Kolbenmaschinen. VOI-Z 1937; 26: June. - [18] Lang G. Zur elastischen Lagerung von Maschinen durch Gummifederelemente. MTZ 24/17. 1963.
J2 Vibrations. [1]
Kramer E. Maschinendynamik. Springer, Berlin, 1984. - [2] Gasch R, Knothe K. Strukturdynamik, vol!. Springer, Berlin, 1987. - [3] Holzweissig F, Dresig H. Lehrbuch der Maschinendynamik. Springer, Vienna, 1979. - [4] SchieWen W. Technische Dynamik. Teubner, Stuttgart, 1986. - [5] Magnus K. Schwingungen. Teubner, Stuttgart, 1976. - [6] Kellenberger W. Elastisches Wuchten. Springer, Berlin, 1987. - [7] Fedem K. Auswuchttechnik. Springer, Berlin, 1977. - [8] Maas H, Klier H. Die Verbrennungskraftmaschine. vol 2, Krafte, Momente und deren Ausgleich in der Verbrennungskraftmaschine. Springer, Vienna, 1981. - [9] Kuhlmann P. Schwingungen in Kolbenmaschinen. VOI-Bildungswerk, Schwingungen beim Betrieb von Maschinen BW 32.11.07, VOl-Gesellschaft Konstruktion und Entwick!ung, 1980. [10] Schwibinger P. Torsionsschwingungen von Turbogruppen und ihre Kopplung mit den Biegeschwingungen bei Getriebemaschinen. Fortschrittber. VD!, Dusseldorf, 1987. - [11] Grgic A. Torsionsschwingungsberechnungen fur Antriebe mit elektrisch drehzahlgeregelten Wechselstrom-Motoren. VOI-Ber 603, 1986. - [12] Natke HG. Einfuhrung in die Theorie und Praxis der Zeitreihen- und Modalanalyse. Vieweg, Brunswick, 1983. - [13] Ewins DJ. Modal testing: theory and practice. Research Studies Press, Taunton, 1984. - [14] Peeken H, Troeder C, Diekhans G. Beanspruchung elastischer Kupplungen in Antriehssystemen mit Asynchronmotoren. Antriebstechnik 1979; 18. - [15] Gasch R, Pfutzner H. Rotordynamik. Springer, Berlin, 1975. - [16] Diewald W. Das Biegeschwingungsverhalten von Kreiselpumpen unter Berucksichtigung der Koppelwirkungen mit dem Fluid. Fortschrittber. VOl, Dusseldorf, 1989. - [171 Dietzen F]. Bestimmung der dynamischen Koeffizienten von Dichtspalten mit Finite-Differenzen-Verfahren. Fortschrittber. VOl, Dusseldorf, 1988. [18] Glienicke]. Feder- und Dampfungskonstanten von Gleitlagem fur Turbomaschinen und deren Einfluss auf das Schwingungsverhalten eines einfachen Rotors. Diss., University of Karlsruhe, 1966.
J3 Machine Acoustics. [I] Schmidt H. Schalltechnisches Taschenbuch, 2nd edn. VOl, Dusseldorf, 1976. - [2] Cremer L, Heck! M. Korperschall - Physikalische Grundlagen und Technische Anwendungen. Springer, Berlin, 1967. - [3] Faller D. Untersuchung der Anregung von Korperschall in Maschinen und der Moglichkeiten fur eine primare Larmbekampfung. Diss., TH Darmstadt, 1972 or Forschungshefte Forschungskuratorium Maschinenbau eV, part 15. Maschinenbau, Frankfurt-on-Main, 1972. - [4] Faller D. et aJ. Gerauscharme Maschinenteile - Die Entstehung von Maschinengerauschen und konstruktive Massnahmen zu ihrer Verminderung. Forschungshefte Forschungskuratorium Maschinenbau eV, part 26. Maschinenbau, Frankfurt-on-Main, 1974. - [5] Kassing W. Untersu-
chungen zum Schwingungs- und Korperschallverhalten rotationssymmetrischer Maschinenstrukturen und ubertragung der Ergebnisse auf die Gerauschentwick!ung von Axialkolbeneinheiten. Diss., TH Darmstadt, 1975 or Forschungshefte Forschungskuratorium Maschinenbau eV, part 42. Maschinenbau, Frankfurt-on-Main, 1976. - [6] Welp EG. Untersuchung des Korperschallverhaltens von Platten- und Kastenstrukturen mit der Methode der Finiten Elemente. Diss., TH Darmstadt, 1977 or Forschungshefte Forschungskuratorium Maschinenbau eV, part 70. Maschinenbau, Frankfurt-on-Main, 1978. - [7] Skudrzyk E. Die Grundlagen der Akustik. Springer, Vienna, 1954. - [8] Faller D. Die GerauschabstraWung von Platten und kastenfOrmigen Maschinengehausen. Forschungshefte Forschungskuratorium Maschinenbau eV, part 78. Maschinenbau, Frankfurt-on-Main, 1979. - [9] Maidanik G. Response of ribbed panels to reverberant acoustic fields. J Acoust Soc Amer 1962; 34: 809-26. - [10] Sennheiser J. Ein Modell zur Bestimmung der SchallabstraWung von Platten unterhalb der Grenzfrequenz. Acustica 1975; 32: 244-54. - [11] Miiller HW, Faller D. Regeln fur Jarmarme Konstruktionen. Konstruktion 1976; 28: 333-9. - [12] VOl 3720. Sheet 1: Larmarm Konstruieren. Allgemeine Grundlagen, 1980; Sheet 2: Beispielsammlung, 1982. [13] Schmidt K-P. Larmarm Konstruieren - Beispiele fur die Praxis. Forschungsbericht no. 129 der Bundesanstalt fur Arbeitsschutz und Unfallforschung. Wirtschaftsverlag Nordwest, Wilhclmshaven, 1974. - [14] Heck! M. Larmarm Konstruieren Bestandsaufnahme bekannter Massnahmen. Forschungsbericht no. 135 der Bundesanstalt fur Arbeitsschutz und Unfallforschung. Wirtschaftsverlag Nordwest, Wilhelmshaven, 1975. - [15] VOl 2567: Schallschutz durch Schalldampfer, 1971. - [16] Faller D. Ein Verfahren zur quantitativen Beurteilung der Gerauschanregung in Maschinen. Tagungsbericht Akustik und Schwingungstechnik, Stuttgart, 1972, pp 418-21. VDE, Berlin, 1972. - [17] Faller D. Maschinenakustische ProbIerne in neuerer Sichl. Tagungsbericht DAGA '73, Aachen, pp 57-75. VOl, Dusseldorf, 1973. - [18] Faller D. Maschinenakustische Berechnungsgrundlagen fur den Konstrukteur. VOI-Ber 1975; 239: 55-65. - [19] Schroeder po]' Konstruktive Larmminderungsmassnahmen an einer Doppelstander-Exzenterschmiedepresse. VOl-Ber 1977; 278: 135-45. - [20] Fecher F. Abschatzung der Larmminderung mittels raumakustischer Massnahmen und Kapseln. [21] VOl 2711: Konstruktion 1976; 28: 341-6. Schallschutz durch Kapselung, 1978. - [22] Connert W. Liirmminderung - eine aktuelle Gemeinschaftsaufgabe. VOl-Ber 1977; 278. [23] VOl 3720, Sheet 7 (Developed): Larmarm Konstruieren - Beurteilung von Wechselkraften bei der Schallentstehung, 1989. - [24] Storm R. Untersuchung der Einflussgrossen auf das Akustische ubertragungsverhalten von Maschinenstrukturen. Diss., TH Darmstadt, 1980 or Forschungshefte Forschungskuratorium Maschinenbau eV, part 84. Maschinenbau, Frankfurt-on-Main, 1980. - [25] Storm R. Moglichkeiten zum gerauscharmen Konstruieren bei krafterregten Maschinenstrukturen in Leichtbauweise. Tagungsbericht DAGA '81, Berlin, pp 337-40. VDE, Berlin, 1981. - [26] Storm R. Zur Abschatzung des akustischen ubertragungsmasses von krafterregten Maschinenstrukturen in Baureihen mit geometrischer Ahn!ichkeit. Tagungsbericht DAGA '81, Berlin, pp 333-6. VOE, Berlin, 1981.
Manufacturing Processes K. Herfurth, Dusseldorf; L. Kiesewetter, Cottbus; J. Ladwig, Stuttgart; G. Mauer, Aachen; W. Reuter, Aachen; G. Seliger, Berlin; K. Siegert, Stuttgart; H. K. Tonshoff, Hanover; G. Spur, Berlin; H.-J. Warnecke, Stuttgart; M. Week, Aachen
_ _.. Survey of Manufacturing Processes H.K. Tonshoff, Hanover
1.1 Definition and Criteria Manufacturing is the production of workpieces of a geometrically defined shape (Kienzle). Unlike the other production technologies, i.e. process technology (chemical, thermal or mechanical process technology) or energy technology, manufacturing technology produces products distinguished by material and geometric characteristics. The selection of a manufacturing process is determined by four basic criteria:
Main Technology. This means the sizes and shapes that
omic efficiency and quality therefore not only should be applied to individual operations or manufacturing steps but also should aim at an overall optimum. To achieve this, opportunities for adaptation, substitution and/or integration (the A-S-l method) may be sought (Fig. 1) [3]. Adaptation is the favourable coordination of consecutive processes, e.g. production of unmachined parts by forging, followed by machining. The development of tools and machine tools may cause the substitution of one manufacturing process for another, e.g. replacement of grinding by hard turning. Integration of manufacturing steps shortens the sequence of operations and is often linked to direct cost savings, and certainly with shortened flow times and reduced control expense (indirect costs). Complete machining of components on multi-axis lathes or machining centres are current examples.
can be produced and the materials that can be worked by means of a manufacturing process.
Error Technology. Tbis means the errors in dimension, shape, position and surface that are caused by manufactur-
ing (error geometry). Besides the microgeometric form of a technical surface with its deviations from the matbematically geometric desired shape, manufacturing processes produce physical and chemical changes in peripheral zones [11. Quality of manufacturing means manufacturing within predetennined error tolerances.
Economic Efficiency. The quantities to be manufactured per unit of time (output), the cost of preparation (preparation costs) and of repeating the order (order repetition costs), the unit costs (directly assignable to an individual item) and the secondary costs (including storage costs) determine typical areas of use of competing manufacturing processes. In this context. the flexibility of a manufacturing process (flexibiliry of volume and flexibility of adaptation) is growing in importance, so that not only are the productivity and capacity utilisation of a manufacturing plant satisfactory but so too are the requirements with regard to the flow time of a product through the plant, the capital tied up in stock, and delivery reliability [2].
Adaptation of Work to People. Manufacturing processes and means should be designed so that people and the environment suffer the least possible nuisance or harm. Pollution limits (noise, vibrations, noxious substances) and safety standards should be complied with. Each of the four fundamental criteria should be observed to an equal extent.
Production engineering products, modules and components are manufactured in sequences of operations (manufacturing steps). Rationalisation to improve econ-
1.2 Classification According to Kienzle [4], today's and tomorrow's many and varied manufacturing processes can be classified into six main groups (Fig. 2) according to the criteria "changing of material cohesion" (creation, preservation, increase and reduction) and "changing of material properties". These main groups are: primary shaping, forming, cutting, joining, coating and changing of material properties. The main groups are subdivided into groups, e.g. cutting into severing, machining with geometrically defined cutting edges, machining with geometrically undefined cutting edges, metal removal, dismantling and cleaning. Within the groups, the manufacturing processes themselves are distinguished by subgroups. Index num-
l=J
Process C
Process B
• p • Process A
Adaptation
Substitution
Integrotion
Figure 1. AS! method of rationalisation.
Manufacturing Processes • 1 Survey of Manufacturing Processes
Create cohesion
Preserve cohesion
Reduce cohesion
1. Primary shaping
2. Metal 4. Joining 13. Cutting forming 6. Changing of material properties
Increase cohesion
Shape-changing
I
~earrangemenl Elimination f matenal particles
of matenal particles
5. Coating
IAddition
of material particles
Figure 2. Classification of manufacturing processes (DIN 8S80).
bers are assigned to this system according to the rules of decimal c1assification_
Standards and Codes DIN 3960: Definitions, parameters and equations for involute cylindrical gears and gear pairs. DIN 3971: Defmitions and parameters for bevel gears and bevel gear pairs. DIN 3975: Tenns and definitions for cylindrical wonn gears with shaft angle 90°. VDI code 3333: Hobbing of spur gears with involute profile. VDI-Verlag, Diisseldorf, 1977. DIN 8580: Classification of manufacturing methods. DIN 8593: Manufacturing process joining. Draft VDI code 2861 BI (9,80): Assembling and handling; characteristics of industrial robots; designation of coordinates. Draft VOl code 2861 B2 (5_82): Assembling and handling; characteristics of industrial robots; application-related characteristics.
Primary Shaping K. Herfurth, Diisseldorf
2.1 General According to DIN 8580, primary shaping is the manufacturing of a solid body from an amorphous material by creating cohesion. Thus primary shaping serves to give a component made from a material in amorphous condition an initial form. Amorphous materials are gases, liquids, powders, fibres, chips, granules, solutions, melts and the like. Primary shaping may be divided into two groups with regard to the form of the products and their further processing: 1.
2.
Products produced by primary shaping which will be further processed by fonning, severing, cutting and joining. The fmal product no longer resembles the original product of primary shaping in fonn and dimensions, Le. a further material change in form and dimensions is accomplished by means of other main groups of manufacturing processes. Products produced by primary shaping which essentially have the fonn and dimensions of fmished components (e.g. machine parts) or end-products, Le. their fonn essentially corresponds to the purpose of the product. The attainment of the desired fmal fonn and dimensions usually requires only operations that fall into the main process group "cutting" (machining).
The production of cast parts from metallic materials in the foundry industry (castings), from metallic materials in powder metallurgy (sintered parts) and from high-polymer materials in the plastics processing industry has major advantages for economic efficiency: The production of cast parts is the shortest route from the raw material to the finished part. It bypasses the process of forming and all the aSSOCiated expense. The final form of a finished component with a mass ranging from a few grams to several hundred tonnes is practically achieved in one direct operation. The production of cast parts by primary shaping from the liquid state allows the greatest freedom of design. This cannot be achieved by any other manufacturing process.
Primary shaping also enables processing of materials which cannot be achieved by means of other manufacturing methods. The direct route from the raw material to the pre-fonn or the end-product results in a favourable material and energy balance. The continual further development of primary shaping processes increasingly permits the production of components and end-products with enhanced practical characteristics, i.e. cast parts with lower wall thicknesses, lower machining allowances, narrower dimensional tolerances and improved surface qUality. In the following, primary shaping of metallic materials from the liquid state in foundry technology, of metallic materials from the solid state in powder metallurgy, and of high-polymer materials (plastics) from the plastidsed state or from solutions is discussed on a common basis with regard to the ftmdamental technological principles, the discussion being restricted to subjects relevant to mechanical engineering. For a better appreciation of the relevant principle of action, many detailed technological operations are omitted that, while they are vital to the specific manufacturing technology, are of minor importance. Furthermore, in discussing the specific primary shaping processes, only products with a simple fonn are referred to, as the diversity of the possible geometric fonns cannot be described here. Only the most important primary shaping processes are selected, as the large number of technological processes and process variants means that it is impossible to provide anything like a complete description_ The processes are selected first according to their technical importance and second according to the principle of action. Materials technology problems will only be mentioned briefly, although they are vital in order to understand the technological processes, their applicability and efficiency, and the changes in material properties brought about by the technological processes.
Process Principle in Primary Sbaping In the processes of primary shaping, the technological manufacturing process essentially comprises the following steps: Supply or production of the raw material as an amorphous substance.
2.2 Shaping of Metals by Casting. 2.2.1 Manufacturing of Semi-finished Products
Preparation of a material state ready for primary shaping. Filling of a primary shaping tool with the material in a state ready for primary shaping. Solidification of the material in the primary shaping tool. Removal of the product of primary shaping from the primary shaping tool. These individual steps are discussed in detail below.
Material State Ready for Primary Shaping. In primary shaping of metallic materials from the liquid state, the raw materials (pig-iron, scrap, ferroalloys and the like) are melted in a metallurgical melting furnace by means of thermal energy. The melting furnaces are usually physically separated from the primary shaping tool. The molten metal is carried by means of transfer vessels (ladles) to the primary shaping tools, termed moulds in the foundry industry, and cast there. In primary shaping of high-polymer materials from the plasticised state, bulk raw materials (granules, powder) are fed after proportioning into a preparation device which is usually integral with the primary shaping tool. There, thorough mixing, homogenising and plasticising of the material to be processed are accomplished under the action of heat and pressure. When solutions are used, these are produced in a mixing unit and then poured into the primary shaping tool. In primary shaping of metallic and also high-polymer materials from the solid state, the bulk raw materials (metal powder, plastic powder or plastic granules) are poured straight into the primary shaping tool, where they sinter, or first become plastic and then solidify under the action of pressure and thermal energy. Primary Shaping Tools. The primary shaping tool contains a hollow space which, with the allowance for contraction, usually corresponds to the form of the product (unmachined part) to be manufactured, but may be smaller or larger than the resulting unmachined part. Furthermore, primary shaping tools often contain systems of channels (runners) for feeding the material in the state ready for primary shaping. The allowance for contraction corresponds to the dimensional changes which occur in the material to be processed from the moment of solidification to its cooling to room temperature.
In the production of cast parts, a distinction is made between primary shaping tools for once-only and for repeated use. Primary shaping tools for once-only use are only used for primary shaping of metallic materials from the liquid state in foundry technology; they are termed expendable or "dead" moulds, Only one product (casting) can be manufactured, as the mould is subsequently destroyed. However, primary shaping tools for repeated use (permanent moulds) are also used in foundry technology. A larger quantiry of cast parts can be produced. The primary shaping technologies for processing of high-polymer materials and powder metallurgy use only primary shaping tools for repeated use. Primary shaping tools for repeated use are usually made of metallic, and more rarely of non-metallic, materials. Primary shaping tools for once-only use (dead moulds) are made with the aid of patterns.
FUIing the Primary Shaping Tools. Filling of the primary shaping tools with the material ready for primary shaping may be accomplished by means of the following principles of action: under the influence of gravity, elevated pressure or centrifugal Jorce and by displacement. The material to be processed can be put into the primary shaping tools in solid, pourable form (e.g. powder), as molten metal in the case of metallic materials, or in plasti-
cised condition, as a solution or as a paste in the case of high-polymer materials.
Change of State Ready for Primary Shaping into the Solid State of Aggregation. Liquid metallic materials (molten metals) change by crystallisation to the solid state of aggregation on cooling owing to the removal of heat. Thermoplastics are cooled in the primary shaping tool after forming. As a result of temperature reduction, which is accomplished either by heat removal in cooled tools or in downstream equipment (cooling baths), the plastic mass passes through the following states: plastic-rubberlike-elastic-solid. In setting by cooling, secondary valency bonds are restored. This process is repeatable; therefore thermoplastics can be restored to the plastic state by reheating. Thermosetting plastics or thermosets (cross-linkable plastics) are cured after forming by a hardening process. Primary valency bonds form, and the plasticised mass solidifies directly under the effect of heat and/or pressure. The curing is an irreversible chemical process: thermosets disintegrate on reheating without needing to pass through a plastic state. Fundamental chemical reactions during solidification are polymerisation, polycondensation and polyaddition. In primary shaping of high-polymer materials, if solutions are used then the transformation to the solid state may be accomplished by the physical process of solvent evaporation. In primary shaping by sintering, a process of shrinkage of the internal and external surface area of a body formed from powder by pressure takes place. Powder particles that are in contact are joined by the formation or reinforcement of bonds (material bridges) and/or by reducing the pore volume; at least one of the material constituents involved remains solid throughout the process. The bonding of the porous pressed body of powder takes place mainly through diffUSion mechanisms. In connection with the description of the technological aspects of primary shaping, further details of the processes that occur as a material changes from the state ready for primary shaping to the solid, dimensionally stable state are dispensed with (see D3.1.1 and 1)3.1.2).
2.2 Shaping of Metals by Casting 2.2.1 Manufacturing of Semi-finished Products This group of primary shaping processes involves the production of initial and intermediate products which are further processed by, for instance, metal forming (plastic deformation).
Ingot Casting Processes Here, ingots, slabs, wirebars, etc. are produced in permanent moulds made of metal (usually cast iron). These products are converted by metal forming (rolling, forging, pressing, wire drawing, etc.) into a semi-fmished product (sheet, plate, section, wire) that no longer resembles the original ingot in form and dimensions. In ingot casting a distinction is made between top pouring (downhill casting, Fig. la), where the mould is tilled by directly pouring the molten metal in from above, and bottom pouring (uphill casting, Fig. Ib), where one mould or several moulds simultaneously (group casting) are filled from below by means of a distribution system (pouring gate and runner bricks).
Manufacturing Processes. 2 Primary Shaping
I
~' T~--
5
b
a Figure~.
b
a
Figure 1. Ingot casting methods: a downhill casting, b uphill casting; 1 ingot mould, 2 baseplate, 3 casting pit, 4 molten metal supply, 5 molten metal, 6 pouring gate, 7 runner bricks.
Procedure. The prepared moulds are set up in the casting pit as illustrated. They are filled with the liquid metal, which solidifies in them. The moulds are stripped from the ingots, which are taken away. Contin.uous Casting Processes
In these processes, which are used to produce either intermediate products for metal fonning or semi-fmished products, the primary shaping tool (continuous mould, casting roller, casting belt, casting wheel) is always smaller than the product of primary shaping. With Contin.uous Mould. In this casting process, a bath of the molten metal is fed into a stationary continuous mould, where the solidification begins. Depending on the design, a distinction is made between batch or continuous vertical (Fig. 2a) and horizontal continuous casting systems (Fig. 2b). On leaving the continuous mould the resulting continuous casting (solid or hollow section) is cooled until it solidifies completely. The continuous casting is usually cut into defined lengths at intervals; like ingots from ingot casting, these are further processed by metal fonning. Travelling Primary Shaping Tools. In this continuous casting process, metal-fonning equipment for rolling or drawing is installed directly following the casting plant, thus dispensing with the manufactunng stages of metal forming. In this case there is usually no cutting of the continuous castings into sections. Strip and Wire Rod Casting Plants
Strip casting machine [41: a vertically uphill, b horizon-
tal: 1 pair of casting rollers, 2 molten metal supply, 3 solidified strip.
In horizontal casting (Fig. ~b), both the feeding of the molten metal and the discharge of the solidified continuous casting (strip) take place hOrizontally. In casting between a casting roller or a casting wheel having the profile of the desired strip or rod and an endless casting belt (Figs 4a and 4c), the molten metal solidifies between the casting roller/wheel and the casting belt and emerges into the open air. In casting in belt moulds (two endless, rotating casting belts), solidification is accomplished with the aid of further rotating equipment to restrict the product laterally between these casting belts (Fig. 4b); the solidified continuous casting then emerges into the open air as a strip. 2.2.2 Manuf"ac:turing of Cast Parts
Manufacturing of cast parts is accomplished with primary shaping processes by means of which a practically finished component, e.g. a machine part or an end-product, is produced without metal fonning. The product's shape and dimensions do not undergo any further significant change; however, primary shaping is followed by other manufacturing processes, e.g. cutting (turning, planing, milling, drilling), to obtain a component ready for fitting. The intention is to perfect and further develop the primary shaping techniques in order to, for instance, reduce the amount of machining work to a minimum. Table 1 gives a survey of the moulding and casting processes. Use of Expendable Primary Shaping Tools (Moulds)
This technique, which is only used in primary shaping of metallic materials from the liquid state in foundry technology, uses a pattern to produce the expendable primary shaping tool. Depending on the type of pattern used, a
In vertical uphill casting between two casting rollers (Fig. ~a) the molten metal is fed from below between two casting rollers. Solidification takes place between these two rollers, and the fmished continuous casting (a strip) emerges vertically upward from these rollers.
2
2
J
.
J
b
Figure 2:. Continuous casting machine [4J: a vertical, b horizontal; 1 open-ended mould, 2 molten metal supply, 3 molten metal, 4 solidified billet.
b
c
Figure 4. Casting machines [4]: a strip casting machine (rotary process), b strip casting machine (Hazelett process), c wire rod casting machine; 1 casting wheel, 2 casting belt, 3 guide rollers, 4 molten metal supply, 5 solidified strip or wire rod.
2.S to 5%
Tolerance range"
sao mm nominal
Single items, small production runs
Quantity range (approx. values)
afar
No limit, available transport facilities and melting capacity determine maximum weight
Single items, small to medium runs
0.3 to 0.8%
1 to 2%
Up to 1000 kg
Medium and large runs
Up to ISO kg
0.3 to 0.7%
Single items, small runs. Series production of suitable components ~
100000
~
80000
3 to S%
castings eu ~ lOOOO castings
AI
castings
'VIg
Series production. Life of mould: Zn ~ 'jOOOOO castings
No limit (maximum transportable weight); particularly suitable for heavy components
1 g up to several kg (up to 100 kg
in special cases)
All metals
All metals Light metals. special copper alloys, highgrade zinc. lamellar and nodular graphite cast iron
AI-, Mg-, Zn-. Cu-, Sn- or Pb-based die casting alloys (ironbased materials under development)
0.1 to 0.4%
Series production. Life of mould: Al ~ 100000 castings
IJ P to 'j000 kg
Lamellar and nodular graphite cast iron, cast steel, light metals, copper alloys
Centrifugal casting
0.3 to 0.6%
1%
Series production. Length of billet depends on machine Life of mould: 5000 to 100000 castings depending on ~ize of workpiece, casting material and type of mould
Al alloys: up to 45 kg Up to 100 kg (more Zn alloys: up to 20 kg in special cases) Mg alloys: up to 15 kg eu alloys: up to 5 kg (limited hy size of pressure die casting machine)
Chill casting
Pressure die casting
No pattern
Permanent moulds
size (approx. values), dependent on degree of accuracy, material, size of workpiece, shape. For material-specific tolerances see DIN 1680 and DIN 1683 to DIN 1688.
1.5 to 3%
Small to large production runs
Up to several tonnes, restricted by size of machines
All metals
All metals
Weight range (approx. values)
All metals
All metals
Materials
Full mould casting
Precision casting (lost-wax process)
Ceramic moulding
Hand moulding
Process
Shell moulding
Permanent patterns
Type of pattern
Mechanical moulding
Expendable patterns
Expendable moulds
Type of mould
Table 1.. Survey of moulding and casting processes (2J
0.8%
Up to several tonnes
Lamellar and nodular graphite cast iron, cast steel, copper and copper alloys, aluminium and aluminium alloys
Continuous casting
I
~
";!;
n
g,
&
~ 2 s· oo
N N
tv
•
J."
n
~
~
"~
g,
oo
!'"S·
N
N
Manufacturing Processes. 2 Primary Shaping
distinction is made between processes using a permanent pattern and those using an expendable pattern. A permanent pattern can be used to make many expendable moulds, but an expendable pattern can only be used for one expendable mould. Expendable patterns are also made in an appropriate primary shaping tool. The patterns are similar in shape to the case part to be manufactured, but are larger by the allowance for contraction of the material to be cast. They also incorporate the macbining allowances, which will subsequently be eliminated by machining of the casting with the aim of achieving accuracy in dimensions, shape and position, as well as tapers to enable the pattern to be removed from the mould. Most patterns have a pattern joint, i.e. they consist of at least two parts (pattern halves); in addition, for castings with hollow spaces the pattern has core marks for insertion of the cores in the mould. In the case of permanent patterns for making an expendable mould for casting, these patterns or their sections made from metals, high polymers or wood are used to make the moulds by the sand moulding, template moulding or shell moulding process.
Hand Mouldin.g (Fig. S). Mould. This is expendable (i.e. used only once). The medium used may be natural or artificial sand, with or without a syuthetic resin binder, CO, sand, cement sand, or a moulding compound. It is worked by hand.
Pattern. Patterns for repeated use, and patterns and core boxes, are made to DIN 2522, those in quality classes HIa and H I being mainly made of hardwood plywood, those in quality classes H2 and H3 of sawn timber and those in quality classes SI, S2 and S3 of foamed plastic. Process Characteristics. Hand moulding denotes the production of a sand mould without using a moulding machine. The mould consists of the external parts for the external profile and the internal parts for the internal proflle. Hollow spaces in the casting are fonned by cores
placed in the mould. The principle of moulding is illustmted in Fig. S. First of all the bottom half of the two-part pattern is moulded. After turning the moulding box over, the top half of the pattern and the pouring gate and risers are placed in position and the top mould is made. The top box is lifted off, the pattern halves are removed from the mould and the core is inserted. The halves of the mould are jOined and the casting is made.
Casting Materials. All metals and alloys that are castable with the current technology. Weight of Castings. The maximum tmnsportable weight and the melting capacity determine the maximum weight. Number of Castings. Single items, small production runs. Tolerances. From about 2.5 to 5%. Mechanical Mouldin.g (Fig. 6). Mould. Expendable (used only once). Natuml sand, artificial sand, sand with syuthetic resin binders, CO, sand. Prepamtion on moulding and core moulding machines. Used in semi-automatic and folly automatic production lines.
Pattern. Patterns and core boxes are made to DIN 1511, in quality class HI mainly of hardwood plywood, in quality classes MI and M of metal, in quality classes KI and K2 of plastic. Process Characteristics. Mechanical moulding is characterised by a semi-automatic or folly automatic manufacturing opemtion for efficient production of ready-to-cast sand moulds. The casting process is often incorporated into the production line. The main stages are: moulding station, core insertion section, casting section and cooling section. The emptying station releases the cast moulds. The moulding station may consist of one automatic moulding machine for complete moulds or of two or more for making separate top and bottom boxes. There are also boxless moulding units, where the moulds are made using only a frame, which is withdrawn after compacting the sand.
Casting Materials. All metals and alloys that are castable with the current technology. Weight of Castings. Limited by the size of the moulding machines: up to approximately 5000 kg. Number Of Castings. Owing to the mechanical preparation, mechanical moulding is suitable for series and
mass production of quantities of 1000 and multiples thereof.
2
Figure S. Hand moulding [3J; 1 bottom balf of box (drag), 2 top half of box (cope), 3 core, 4 casting, 5 plate with wooden pattern half, 6 risers, 7 pouring gate.
Figure 6. Mechanical moulding [3]; J plate with metal pattern, 2 compaction of the moulding sand in a frame, 3 boxless mould ready for casting, 4 casting.
2.2 Shaping of Metals by Casting. 2.2.2 Manufacturing of Cast Parts
Tolerances. From about 1. 5 to 3%. Suction Moulding (Fig. 7) Mould. Expendable (used once only). made from wet artificial casting sand.
Pattern. Wood, plastic. metal. Process Characteristics. The process is characterised by the formation of a vacuum by withdrawing air from the mould space and the incoming moulding sand. This accelerates the sand, which spreads over the wall of the pattern. The sand can be subsequently pressed against the pattern. Advantages of the process are optimum mould compaction around the pattern, no shadow effect with plane surfaces, decreasing hardness of compacted sand from the inside to the outside, high surface quality, dimensionally stable castings, reduced cleaning. This process should not be confused with vacuum mOUlding. Casting Materials. Iron, steel and aluminium. Weights of Castings. From about 0.1 to 120 kg. Number of Castings. Small, medium and large production runs. Tolerances. Conventional to DIN 1683; maximum offset of compacted sand 0.3 mm. Shell Moulding (Fig. 8). Mould. Expendable (used only once). Resin-coated sands or sand/resin mixtures.
Pattern. Patterns for repeated use, heatable metal patterns and metal core boxes.
Process Characteristics. These moulds are shell moulds with walls only a few millimetres thick. The mould material is poured onto the heated metal pattern. This cures the synthetic resins in the mould material, solidifying the mould. The result is a self-supporting, stable shell mould. Shell moulds are often moulded in one piece and then divided. After putting in the cores, the two halves of the mould are glued together. The shell moulding process is used in various stages of mechanisation and automation. This process is used not only for making moulds for shell casting, but also for producing hollow
~' ~ J .
. ,
. .
'
'.
. /'
2
.,.
p;A
~' :.' ~
~ -5 .
' ~ ,
"'~.
~6
.......w-IIIIt+++-""IIII
Figure 8. Shell moulding [31; 1 loose synthetic-resin--coated sand, 2 hardened synthetic-resin-coated sand, 3 heated metal pattern, 4 shell mould, 5 glued seam, 6 casting.
shell cores for sand and chill casting. These cores are produced on special core moulding machines. Shell casting offers high dimensional accuracy with excellent surface quality.
Casting Materials. All metals and alloys that are castable with the current technology. Weight of Castings. Up to = 1;0 kg. Number of Castings. Medium to large production runs. Tolerances. From about 1 to 2%. Ceramic Moulding (Fig. 9). Mould. Expendable (used only once), made of highly refractory ceramic similar in kind to mould materials for precision casting.
Patterns. Reusable, made of metal, plastic or specially varnished wood. Process Characteristics. A slip consisting of highly refractoty substances is poured around the pattern; these
J
7
Figure 7. Suction moulding [3]: I moulding sand, 2 pattern, 3 air connection, 4 vacuum, 5 mould space, 6 sand mouJd, 7 core.
Figure 9. Ceramic moulding l31; 1 pasty ceramic moulding compound, 2 plate with pattern half, 3 hardened split ceramic mOUld, 4 inserted core, 5 risers, 6 pouring gate, 7 casting.
Manufacturing Processes • 2 Primary Shaping
substances then hatden by chemical reaction. Often only one layer is poured, which is then back-filled with "normal" moulding sand. After removing the pattern, the ceramic is fired or skin-dried (Shaw process). To keep ceramic moulding, which is relatively expensive, to a minimum, it is usually only the parts of the mould that will be cast in finished or neat-finished form that are made from special ceramic. In the case of components for fluid flow machines, these are the thtee-dimensionally curved parts; in the case of tools, these are the contours that require no finishing by machining or which, after hardening, will be fmished only by electrical discharge machining or grinding. Castings from ceramic moulds have no casting skin in the conventional sense and are among the precision casting processes that, as the technology develops, are becoming more and more widely used owing to their efficiency.
a
b
Casting Materials. All metals and alloys that are castable with the materials.
current
technology,
especially
Weight of Castings. From about 0.1 depending on the production equipment.
to
iron-based 2500 kg,
c
'----_ ___ --'7-
Number of Castings. Single items, small and medium runs, also larger runs in the case of fluid flow machines.
Tolerances. Up to about 100 mm :!: 0.2%, over 100 mm :!: 0.3 to 0.8% of nominal dimensions. Vacuum Moulding (Fig. 10). Mould. Expendable (used only once); plastic sheet vacuum-moulded to the contours of the pattern, back-filled with fine-grained , binder-free quartz sand and sealed with a covering sheet. Dimensional stability is preserved by means of a vacuum of 0.3 to 0.6 bar.
Pattern. Permanent patterns, not subject to Significant wear. Quality classes 1 and 2, mainly made of wood. Core boxes according to the core manufacturing method.
Process Characteristics. The process is characterised by the use of a vacuum for both deep drawing the pattern sheet over a pattern with nozzle holes and maintaining the stability of the mould. A moulding box equipped with suction systems is connected by a pipe to the vacuum grid. The fine, binder-free sand with which the mould box is filled is compacted by vibration. After applying a cover sheet, the air is evacuated from the sand and the mould thus becomes rigid. The mould is constantly connected to the vacuum grid before, during and after casting. To empty the mould, the vacuum is switched off and the sand and cast parts drop out of the mould box without additional force. The advantages of the process are: high, reproducible dimensional accuracy with outstanding surface quality; the mould seam at the joins and core marks is very small; tapers can be entirely dispensed with in certain areas of the casting.
Casting Materials. All metals and alloys that are castable with the current technology.
Weight of Castings. Restricted by the equipment available, not by the process.
Tolerances. 0.3 to 0.6%. Casting Under (High) Vacuum. Mould. Expendable (used only once), shell moulds (for investment casting) and precision casting moulds made of special mould materials.
Pattern. Wax fo r investment casting, also of metal, plastic or the like depending on the type of mould.
d Figure 10. Vacuum moulding [3]; 1 heating, 2 plastiC sheeting, 3 pattern, 4 vacuum box. a The plastic sheeting is softened by means of a sheet-type heating element and drawn tightly against the pattern by vacuum through holes. b The mould box is placed on top, filled with sand, pre-compacted, and the top of the box is covered with plastic sheeting. (; A vacuum is applied to the mould box, compacting the sand. By switching off the vacuum, the mould box can be easily lifted off the pattern . d The top and bottom halves of [he box are joined. The vacuum is maintained during casting.
Process Characteristics. Titanium and zirconium are among the reactive metals that have high affinities with oxygen, nitrogen and hydrogen in the molten state. This is even the case when present as alloying constituents in appropriate percentages in, e.g., molten nickel. All these alloys must therefore be produced and cast under defined conditions, normally under high vacuum. The new mould ceramics, e.g. those made of yttrium and zirconium oxides, resist attack by reactive metals and melts. However, these special ceramics are not (yet) required for nickel-base alloys that are only alloyed with titanium, aluminium, etc . To optimise quality and structure, the castings are usually isostatically pressed at high temperature by means of the HIP process.
Casting Materials. Alloys
based on (in order of importance) nickel, titanium, cobalt, iron and zirconium.
Weight of Castings. Approx. 0.01 to 100 kg and more, depending on the manufacturing equipment. Number of Castings. Small series to fairly large production runs.
Tolerances. :!: 0.3 to :!: 0.8% of nominal size , depending on the mOUlding process. Precision Casting (Fig. 11). Mould. Expendable (used only once), made of highly refmctory ceramic; sin-
2.2 Shaping of Metals by Casting. 2.2.2 Manufacturing of Cast Parts Compact mould
Shell mould
Pallem making Back·filling
ForlJ1alion of shell by repealed imme~on and saJlding
¥IJ
terns are assembled into clusters by means of casting sys· tems, usually again employing injection moulding. The method of this assembly is crucial for the quality of the castings and for efficiency. Viscous ceramic coatings which cure by chemical reaction are then applied to these clusters. For aluminium, special plasters are also used. After melting out (lost wax process) or dissolving away the pattern material, the resulting one·piece moulds are fired. Casting takes place in moulds which are usually still hot from firing, so that narrow cross·sections and fine pro· files "tum out" cleanly. Precision casting, with its tight tolerances and high surface qualitty, is the casting tech· nology that offers the greatest freedom of design coupled with high quality.
Casting Materials (in order of importance). Steels and alloys based on iron, aluminium, nickel, cobalt, titanium, copper, magnesium or Zirconium, including aerospace materials, produced at atmospheric pressure or under vac· UUffi.
Casllng I/lemelal
Weight of Castings. 0.001 to 50 kg, also up to 150 kg and over depending on manufacturing equipment. Number of Castings. Small series to large production runs, depending on the complexity and/or machinability of the workpiece concerned.
Tolerances. :+: 0.4 top:+: 0.7% of nominal dimensions. Deslroying the mould 10 release lhe casting
~ U ~/._
Culling
~
~'t1r' ~
~
r L
Full Mould Casting (Fig. 12). Mould. Expendable (used only once), mould material usually self.curing.
Pattern. Expendable, foam materiaL Process Characteristics. One·piece pattern made of foam material (polystyrene). Shape and dimensions match the part to be cast (taking into account allowance for contraction). The pattern need not be removed from the mould after mould·making. The heat of the molten metal flowing into the full mould vaporises the pattern, which is continuously replaced by cast metal. Mould joints and cores are usually unnecessary. Bolts, sleeves, lubrication
Grinding
Figure 11. Schematic diagram of sequence of operations in precision casting 15 J.
gle or group pattern with runners, combined to form cast· ing units ("clusters" or "trees").
5
Pattern. Made by injection moulding from special waxes or the like, thermoplastics or mixtures thereof.
Process Characteristics. The distinguishing features are the expendable patterns, the one·piece moulds and the casting in hot moulds (= 900°C for steel). A casting skin in the conventional sense does not fornl. The patterns are injection·moulded in single or multiple tools made of alu· minium, steel or soft metal, for which an original pattern is required. The most suitable injection moulding tool in each particular case is chosen according to the planned total quantity, the form of the casting and the nature of the pattern material. The formation of certain undercut contours may require the use of water-soluble or ceramic cores. for which a supplementary tool is used. The pat·
Figure 12:. Full mould casting {3]; 1 pouring gate, 2 riser, .3 pipe to be integrtllly cast. 4 one-piece polystyrene foam pattern, 5 mould, 6 casting, 7 integrally cast pipe.
1:.,.,
Manufacturing Processes • 2 Primary Shaping
lines, etc. can be integrally cast in. The absence of mould tapers reduces the weight of the casting. The time and cost of manufacture are only a fraction of those encountered with a wooden pattern.
Materials Jor Casting. All metals and alloys that are castable with the current technology, especially those with high casting temperatures. Weight oJ Castings. From approximately 50 kg up to maximum transportable weight; especially suitable for large components. Numher oJ Castings. Single pieces, small production runs. Tolerances. From about 3 to 5%. Magnetic Moulding. Mould. once), iron granules.
Expendable (used only
Pattern. Expendable, foam material. Process Characteristics. Magnetic moulding is a type of full mould casting. The casting units prefabricated from foam material (patterns with pouring gates and runner) are coated with a refractory ceramic (similar to the shell moulds for preciSion casting). They are then back-filled with pourable iron granules in a mould box. By applying (or switching on) a d.c. magnetic field, the iron powder becomes rigid and thus supports the casting unit. After casting and solidification of the metal, the magnetic field is switched off, causing the iron granules to become pourable again. Then the casting is removed; the iron granules can be reused. Casting Materials. All metals and alloys that are castable with the current technology. As the thermal conductivity of the magnetisable mould material is higher than that of quartz sand, the cooling rate of the castings is higher and leads to a fmer metallographic structure. The properties in use are especially improved in the case of steel castings.
Number oJ Castings. Single items, small production runs. Tolerances. From about less than 3 to 5%. Use of Permanent Moulds Chill Casting (Fig.
1~).
Mould.
Permanent mould,
cast iron or steel, cores made of steel.
Pattern. None required. Process Characteristics. Casting takes place by gravity in permanent metal moulds. These moulds are made in two
2
or more parts for removal of the fmished casting. The higher thermal conductivity of the metal mould compared with moulding sand brings about faster cooling of the solidifying molten metal. The result is a relatively finegrained, dense structure with better mechanical properties than parts made by sand casting. High dimensional accuracy, excellent surface quality and good reproduction of contours characterise the chill casting. This process fully meets the specifications for gas- and liquid-tight valves, owing to the production of a dense structure. A rapid, efficient casting sequence, with machining generally being unnecessary or requiring only small machining allowances, are further features of this process.
Casting Material. Chill casting alloys, DIN 1709 copperzinc alloys, DIN 1714 copper-aluminium alloys, DIN 1725 aluminium alloys, DIN 1729 magnesium alloys, DIN 1743 high-grade zinc alloys, also copper, copper-chromium alloys, super-eutectic aluminium-silicon alloys, lamellar and nodular graphite cast iron. The standard chill casting alloys are denoted by the symbol GK. Weight oJ Castings. Non-ferrous metals and cast iron up to = 100 kg, more depending on equipment. Cast iron for certain purposes up to = 20 t (= 20000 kg). Number oJ Castings. From about 1000 and multiples thereof, depending on the material being cast (e.g. AI = 100000 castings). Tolerances. From about 0.3 to 0.6%. Low-Pressure Chill Casting (Fig. Mould. Permanent mould, cast iron or steel.
14).
Pattern. No pattern required. Process Characteristics. Casting is carried out under pressure (usually with compressed air) in permanent metal moulds. These moulds are made in two or more parts for removal of the finished casting. The higher thermal conductivity of the metal mould compared with
moulding sand brings about faster cooling of the solidifying molten metal. The result is a relatively fme-grained, dense structure with better mechanical properties than parts made by sand casting. The distinguishing feature is the application of pressure, which dispenses with the need for risers on the casting. High dimensional accuracy, excellent surface quality and good contour reproduction together with a rapid, efficient casting sequence and considerable machining economies are further features of this process. Gas- and liquid-tight valves can be efficiently manufactured owing to the dense structure of the casting.
3
1-
F=======~I "~~'L--___-
5 6 7
Figure 1.3. Chill casting (composite mould with metal and sand cores, the latter for undercuts) 13]; 1 riser, 2 metal core, 3 pouring gate, 4 sand corc.
Figure 14. Low-pressure chill casting (3]; 1 air or gas, 2 moving block, 3 fixed block, 4 ascending pipe for molten metal, 5 molten metal, 6 crucible, 7 heating.
2.2 Shaping of Metals by Casting. 2.2.2 Manufacturing of Cast Parts
Casting Materials. Light metal, especially aluminium alloys.
Weight of Castings. Up to = 70 kg. Numher of Castings. From about 1000 and multiples thereof.
Tolerances. From about 0.3 to 0.6%. Pressure Die Casting (Fig. IS). Mould. Permanent mould, usually high-tensile hot forming tool steel or special metals. Pattern. No pattern required.
Process Characteristics. The distinguishing feature of this process is that the molten metal is forced into the twopart permanent mould at high pressure and relatively high speed in pressure die casting machines. Two types of process are distinguished, namely: l.
2.
Hot-chamher process. Here the die casting machine and the holding furnace for the molten metal form a unit. The casting assembly is immersed in the molten metal. In each casting operation, a precisely predetermined volume of molten metal is forced into the mould. The hot-chamber die casting process is especially suitable for lead, magnesium, zinc and tin. The output of components manufactured by this process is considerable, but varies according to the size of the component and the casting material. Cold-chamber process. In this process the die casting machine and the holding furnace for the molten metal are separate. After being taken from the furnace. the
molten metal is poured into the cold pressure chamber and forced into the mould. The pressure chamber is mounted directly on the runner-side mould block. This process is chiefly suitable for aluminium- and copper-based alloys, as these would attack the steel casting assembly when molten if the hot-chamber process were employed. Cold-chamber die casting machines do not achieve the rates of output of hot-chamber machines, owing to the nature of the process. Pressure die casting is today one of the most efficient casting processes around. The machines are mostly semi-automatic or fully automatic. Pressure die-cast parts have smooth, clean surfaces and edges. They are extremely dimensionally accurate. Therefore only the fitting and bearing surfaces, at most, require machining. Vety low machining allowances result in short machining times.
Casting Materials. Materials suitable for pressure die casting are: DIN 1709 copper-zinc alloys, DIN 1714 copperaluminium alloys, DIN 1725 aluminium alloys, DIN 1729 magensium alloys, DIN 1741 lead alloys, DIN 1742 tin alloys, DIN 1743 high-grade zinc alloys. The standardised pressure die casting alloys are denoted by the symbol GO. For the hot-chamber process: lead, magnesium, zinc and tin alloys. For the cold-chamber process: particularly aluminium- and copper-based materials. Weight of Casting. Up to 45 kg for light alloys, up to 20 kg for other materials, depending on the material being cast and the working dimensions of the die casting machines. Number of Castings. Varies widely depending on the material being cast. Example: Zn alloys = 500000 castings. Tolerances. From about 0.1 to 0.4%. Centrifugal Casting (Fig. 16). Mould. water-cooled cast iron or steel mould.
Permanent,
Pattern. None required.
a
b
c
Process Characteristics. The centrifugal casting process is used to manufacture hollow products having a rotationally symmetrical hollow space and an axis coinciding with the
axis of rotation of the centrifugal casting machine. The external form of the casting is determined by the shape of the mould. The internal form is determined by the effect of the centrifugal force of the rotating mould. The wall thickness of the casting depends on the quantity of molten metal supplied. A variant of the process is centrifugal mould casting, which produces fmished hollow or even massive castings using rotating moulds. Composite centrifugal casting is also possible, as is centrifugal casting with a flange. The condition on delivety of centrifugalcast Fe, Ni and Co-based alloys is normally (at least) preturned.
3
Figure IS. Pressure die casting 131: a hot-chamber process, b to d cold-chamber process, b filling of casting chamber. c plunger forces molten metal into die, d ejection of casting; I die, 2 plunger,
3 crucible with molten metal, 4 casting vessel,S ejection, 6 moving block, 7 fixed block, 8 pressure chamber, 9 plunger, J () ca"iting, 11 slug.
Figure 16. Centrifugal casting [3]; J drive, 2 mOUld, 3 pouring ladle, 4 pouring basin.
':."4
Manufacturing Processes. 2 Primary Shaping
and light metals.
Weight of Castings. Up to about 5000 kg.
C
Number of Castings. From about 5000 to over 100000, depending on the mould and the casting material. In special cases, e.g. castings of stainless steel and the like, also single pieces and small production runs (from = 100 mm internal diameter upwards).
a
I
I I
----~
B
~U A
-- -l- -+s.Q![
Temperature
~
'"
E~
a:2
with the current technology.
Weight of Castings. Up to about 50 kg and over, depending on the manufacturing equipment.
I:>
I
Temperature
Pattern. None.
Casting Materials. All metals and alloys that are castable
I
IContraction i l l
posite casting.
Process Characteristics. These types of process are used: to cast structural parts from two or more different metallic materials which are firmly joined together. At least one material is poured in the molten state into a mould, which may also be part of a product to be manufactured; for composite casting of various metals and/or alloys in molkn or semi-solid condition, e.g. in centrifugal casting; and for casting in, casting round and lining solid components, which may be made not only of metal but also of ceramic. The bond may be fonned by shrinkage, positive force or both.
IShrinkage I ~ Iduring I ~ Isolidiflcatlon i -6
b
B I omractlonin
(or Integral) Casting (Fig. Metal mould, e.g. for centrifugal com-
I
----1 ;g
13
Shrinkaqe _: 'gl dUring i 1'l solidification 1.2
_ _ -tsQli
0 I I
Io§:
Tolerances. About 1%. Composite 17). Mould.
Contraction in liqUid state
0
Contraction in liquid state
Casting Materials. Especially cast iron, cast steel, heavy
8 E
Figure 18. Schematic diagrdm of contraction of metallic materials during cooling from the molten state: a for pure metals and eutectic alloys, b for non-eutectic alloys [6}.
Table 2. Contraction of various casting materials (approximate values) ~aterial
Lamellar graphite cast iron Nodular graphite cast iron Cast steel Malleable cast iron Copper alloys Aluminium alloys
liqUid (max. %)
Solid (max. %)
5 6 5.5 1.25
Number of Castings. Medium and large production runs. Tolerances. From about 0.1 to 0.6%, depending on the process. 2.2.~
Guidelines for Design
Forming by casting enables design ideas to be turned into reality to a particularly high degree, owing to the extensive freedom of design that it offers. A design appropriate for manufacturing, which contributes decisively to the efficient production of a casting, can generally be achieved only by close collaboration between the design engineer and the founder. Forming by casting differs from other forming processes in that the material only receives its shape, material structure and quality after cooling, with shrinkage - which may sometimes be considerable - in the liquid state and during solidification, and appreciable contraction in the solid state (Fig. 18, Table 2). The contraction in the solid state should be accounted for by means of a suitable allowance (allowance for contraction). The alloying specifications often suffer considerable deviations, owing to obstruction of contraction
by ribs, projections, more or less flexible cores and mould parts (Table 3). Provided that they fall within the acceptable dimensional variations or are compensated for by
machining allowances, as is usually the case with small castings, they do not present a problem. With large castings, though, empirical values for the deviations due to contraction have to be taken into account when making the pattem. If there is a one-sided obstruction of contrac-
tion, e.g. due to the mould or even the shape, especially in the case of longer castings (different cross-sections along their length and consequently different cooling rates and thermal stresses), the castings would distort unless the pattern is curved in the opposite direction. Large wheel centres, for instance, are not infrequently split, e.g. to prevent unacceptable out-of-roundness. Thermal stresses that are not reduced by plastic defonnation may, besides distortion, also result in undesirable "relief by cracking". Therefore if insufficient attention is paid to contraction of the material being cast at the design stage, taking into account the possibilities afforded by gating of moulds and risering, pipes (shrinkage cavities), shrinkage voids, sinks, hot (pipe) cracks, distortion and stress cracks may fonn.
Manufacture-Orientated Design Correct design of changes in wall thickness with a view to shrinkage during solidification and contraction in the solid state: Figure 17. Composite (or integral) casting [31; I composite material for casting, with expendable heads which are removed by cutting, 2 pouring basin, 3 clamped-on hub, 4 mould, 5 composite casting (two different materials).
Wall Thickness Graduations should pennit directional solidification.
Junctions fanned by the meeting of two or more walls form concentrations of material, i.e. hot zones. Therefore
2.2 Shaping of Metals by Casting. 2.2.4 Preparatory and Finishing Operations
JJ frfr Flexion
Table 3a Guide values for linear contraction and possible deviations [7[ casting material
Lamellar graphite cast iron Nodular graphite cast iron, unannealed Nodular graphite cast iron,
Guide value
Possible
%
deviation %
1.0 1.2
0.5 to 1.3 0.8 to 2.0
0.5
0.0 to 0.8
Cast steel Austenitic manganese steel White malleable cast iron Black malleable cast iron
2.0 2.3 1.6 0.5
Aluminium casting alloys
1.2
1.5 2.3 1.0 0.0 0.8 1.0 1.5 0.8
a
annealed
Magnesium casting alloys 1.2 casting copper (electrolytic) 1.9 eusn casting alloys (cast 1.5 bronzes) CuSn-Zn casting alloys 1.3 (gunmetal) CuZn casting alloys (cast brass) 1. 2 CuZn (Mn, Fe, AI) casting alloys 2.0 (special cast brasses) CuAl (Ni, Fe, Mn) casting alloys 2.1 (cast aluminium bronzes and cast multicomponent aluminium bronzes) Zinc casting alloys 1.3 Babbitt (Pb, Sn) 0.5
to to to to to to
2.5 2.8 2.0 1.5 1.5 1.5 to 2.1 to 2.0
b
1.1 to 1.5
Sudden Changes in Wall Thickness should be avoided, as they produce high thennal stresses due to different cooling rates. In addition, there is often increased obstruction of contraction by the mould. The risk of fonnation of hot cracks ("pipe cracks" between liquidus and solidus temperatures) and stress cracks (during further cooling in the solid state) is therefore high. Locations prone to cracking can be protected by ribs.
Sharp Comers additionally cause a heat buildup (sandedge effect) and, accordingly, not only hot cracks but also porosity due to contraction as well as drawholes. For a summary of design recommendations see Fig. 20.
Stress-orientated Design When designing castings, the main stresses occurring during manufacture have to be taken as a basis. Here, the freedom of design offers excellent adaptation to the technical requirements. Forming by casting pennits the efficient manufacture of parts of the most complicated kind with high strength in relation to shape. The stress condition of the design can often be made more favourable by suitable ribbing or slight modification (Figs 19 and 20). It is important to know the load-bearing capacity of the materials to be cast. For approximate values for lamellar graphite cast iron see Fig. 21. Examples. A casting made from GG-15 grey cast iron (top hOrizontal grey bar) with a wall thickness of 10 rom or a test bar diameter of 20 rom (vertical line) has a tensile strength of approx. 22 dN/mm2, a hardness of approximately 220 HB and an initial
Compression
1
2
~~ ,
F
Jf-=?DA! •
c
they should be separated as far as possible, or designed for efficient casting by narrowing the cross-section. Concentrations of material, especially at locations that are inaccessible to feeding, lead to piping.
Tensioo
F
0.8 to 1.8 1.8 to 2.3
0.4 to 0.6
Compression
~T_
n.8 to 1.6
1.9 to 2.3
1;.11
Tension
2
Figure 19. Examples of stress-orientated casting design for a material having higher compression strength than tensile strength according to [6]. a Pedestal; 1 tlexurally stressed - inadequate bearing surface, 2 compressively stressed - bearing surface widened. b Cylinder cover; 1 tensionally stressed - poor design, 2 compressively stressed - good design. I: Wall bracket ann; 1 poor cross-sectional arrangement, 2 stress-absorbing cross-sectional arrangement.
modulus of elasticity of almost 10000 dN/rom2. For a wall thickness of 4S mm, on the other hand, the tensile strength is approximately \0 dN/mm', !be hardness is approximately 130 HB and the modulus of elasticity almost 8000dN/mm 2 • If, however, the tensile strength of this 45-mm-thick wall is 22 dN/mm2 , a strength of approximately 180 HB and an Eo of approximately 11500 dN/mm2 should be expected. The material grade GG-30 should be selected. For a wall thickness of 10 mm, this cast iron has a tensile strength of approximately 3S dN/mm 2 , a hardness of approximately 260 HB and an Eo of approximately 13000 dN/mm'.
2.2.4 Preparatory and Finishing Operations Melting of Materials for Casting. For transforming the metal to be cast and the additives into the molten state, a wide variety of melting equipment - e.g. shaft (cupola), crucible and hearth-type furnaces - is available. These furnaces are heated with coke, gas, oil or electricity. The most important types of melting equipment are: for cast iron, including malleahle cast iron: cupola (shaft) furnaces, induction furnaces, rotary kilns (oilfired); cast steel: electric arc furnaces, induction furnaces; nonferrous metal castings: induction furnaces, electrically, gas- or oil-heated crucible furnaces. Cleaning of Castings. The moulds are emptied by means of emptying jiggers. The sand adhering to the casting is generally removed by means of abrasive blasting equipment employing, without exception, steel shot or steel grit made from wire.
':.1'
Manufacturing Processes. 2 Primary Shaping
/' TensJe stresses changed to / _ _ COOIpresson slresses by 'I': lOOd,f.ed shape Joinlf19 seJe
TenSile stresses present. wrong shape lor materlais With ~er compre5SlOll strength than te
_ ,_ No ooncentrallOO of maleroa!. dense Slructure
Wroog shape 101 bilttie mate
Stress·aosorbif1g lib shape Io! tensile stresses and broItfe matenals
Macho1ing dih~l, as no tool run-oot JlIOY1ded
' ____
DoYbI!!-sided machilYll9 rm«rt In this loon c'1g
-- MachlMg made easy by precast tool run-ool ( ~ ca$ng possible wilhou1 Cl)(8i
F'- - - II IS bener tomal
I
Meeting 01 several ribs leads 10 an undesirable ooocenIraliOfl 01 material
double-sided machinlll9 rm«rt by macwg
~~~~WS-- Stress-absorb"'!) pcSltIOO &
01 sliHening rib ' il IS now under compression SYess; beneficial with bnHIe mate
Wrong nb pcsnion lor matein and rLfKlUl ----not perpendiclklr 10 ~is of machiIrIg; tool gels off·line
-L
Slagge
Cross-nbbmg leads 10 concentratl()r\ 01 matenaJ ard oonsequeIItiy 10 loosening 01 tile Slruc1ure at tile [Unctions Unl1M)Jrabie stresses OOIIdilion _ IieruraJ stress
_
,
~
(1SSl~Ss::::::~:::s:~!?:::::~~~~::s:~~r-::,.; )
Com~ex mach ""'9. ----conce'1lration 01 material
Favourable stress OOIIdllion, co-npresSIOO SIJesses
-/-- - Easy to mad1ine.
savrlgs 00 matenal
Wrong
Right
General
General
$I1arp"ed<)ed changes In cross-scctlOO' Rrsk 01 cracks anc loosef1ing 01 stllltll1e. unfa'llUlahte stress OO1dltioo
~~C~~~KS
Figure 20. Illustration of important design guidelines [11].
Appro,. plale Ihickness in mm
7.5
10
15 I~
'--'.,---L..._l..--L_.L............L_..L...-l_
2.0
1.6
1.2
Surlaee·lt>-'-lliume ratoo rn em-'
-L_L-....J 0.6
O.S
0.4
0
Figure 21. Chart illustntting mechanical properties of lamellar graphite cast iron. Relationship between chemical composition, rate of cooling and mechanical properties (tensile strength, hardness, modulus of elasticity) in the casting (wall thickness) and the separately cast test bar 19, 10]. Each point on the diagram signifies a specific combination of mechanical properties for a specific material. It also detennines the material grade to be selected.
2.3 Forming of Plastics. 2.3.3 Calendering
':.....
Heat Treatment. Many materials only obtain the physical and teclUlological characteristics required in use from heat treatment. This treatment requires the use of electrically heated or gas- or oil-fired furnaces in continuous or batch operation. Their size is matched to the size and quantiry of the castings and their mode of operation to the wide variety of heat treatment processes (see D3). Inspection and Testing Methods. The diverse demands made on the casting, which become greater with advances in technology. and the trend towards lightweight construction and thus more efficient use of materials inevitably lead to stringent requirements with regard to casting quality, with particular emphasis on consistency. Inspections of the process and the castings begin with checking of the metallic and non-metallic feedstocks and end with the final inspection of the castings. Materials and workpieces are mainly tested by means of the nondestnlctive testing methods such as radiographic, ultrasonic, magnetic powder and liquid penetrant testing. Destnlctive tests, e.g. tensile, notched-bar impact and bending, are usually carried out with specimens cast either separately or as an appe ndage to the casting; in exceptional cases, specimens taken from the casting itself may be use d .
2.3 Forming of Plastics For the materials properties of plastics see D4.
Thermoplastics (in the form of injection moulding compounds) account for the greater part by far of production of moulded parts and semi-finished products compared with thermosetting plastics (compression moulding compounds and casting resins). Primary shaping may be accomplished by gravity (static) casting or centrifugal casting, but is more frequently done by compression moulding and, most of all, injection moulding and extrusion. The moulding technology is described in DIN 16700. Important factors for discontinuous (fixedcycle) production of moulded parts (compression moulding, injection mOUlding) and continuous production of sections, films, sheet. etc. (extrusion) from moulding compound (powdcr. granules, chips, etc.) are the "processing parameters", e.g. melting range, viscosity, melting index, flow behaviour, disintegration temperature range, stated by the manufacturers (cf. D4) . 2.~.1
Casting of Plastic Sheet
belt made of copper (belt casting, Fig. 22a). The sheet thus formed passes through a drying zone where the solvent evaporates, causing the sheet to solidify, and is removed from the rotating dnlm or belt with a stripper device. 2.~.2
Extrusion
The characteristic feature of extnlsion is that the material to be processed is continuously forced, in the form of moulding compound, in plasticised condition from a pressure cbamber through a suitably shaped extnlsion tool (die ) , emerging through a nozzle into the open air. The result is strip, rigid or flexible tubing, solid sections, fibres or sheet in a continuous stream. The purpose of the extruder (Fig. 2~) is to receive, compress and preheat the moulding compound (granules, powder) in the intake zone; to plasticise the moulding compound in the conversion zone; and to homogenise and compress the moulding compound and expel it from the extnlder at the correct te mperature in the expUlsion zone. To transform the processed material into its final solid state , the extnlded material has to be cooled with air or water after leaving the extrusion tool. Besides the continuously operating screw extnlders, there are also discontinuously operating ram extnlders, which deliver similar products in individual sections. Extruders are also used as plasticising equipment for injection moulding, calendering and blow moulding. The lastnamed process, however, begins with a defined form whicb simply undergoes a further change with the mate rial in n1bber-like elastic condition; it is therefore regarded as a fonning process. 2.~.~
In casting of plastic sbeet, the material to be processed flows by graviry from a storage tank with a controllable slot at the base onto a slowly rotating drum underneath the storage tank (drum casting, Fig. 22b) or an endless
•
zone Figure Z~. Extruder (4) ; 1 primary shaping tool (die), 2 cylinder, .3 screw, 4 heating elements, 5 COOling channels, 6 feeder head.
Calendering
Calendering is the term given to primary shaping of sheet or film from prebeated, prc-plasticised moulding compound between rotating rollers. Tbe pre-plasticised material coming from the preparation unit (e.g. an extruder, Fig. 2~) is fed by means of conveying equipment between the heated rollers of the calender (Fig. 24) , where it undergoes final homogenising and plasticis-
b
Figure 22. Plastic sheet casting machine (4): a helt process, b drum process: I marerial ( plastic <:ompound) supply, 2 sheet, 3 casting hell. 4 casring dnun.
Figure 24. Calendering unit (4) ; 1 extruder, 2 belt conveyor, 3 four-roller calender, 4 cooling rollers , 5 thickness meter, 6 reeling up.
':.GI
Manufacturing Processes. 2 Primary Shaping
ing and receives the desired thickness. After leaving the last calendering roller, the sheet or ftIm passes over cooling rollers to harden it. 2.~.4 ~ating
In laminating, webs of supporting materials (paper or fabriC) are soaked with high-polymer materials (resin carriers and resins; thennoplastic materials may also be used) and converted to laminates by pressing between heated plates as the primary shaping tool (mould). Depending on the desired thickness, several resin-soaked webs are placed on top of each other, covered with pressing plates on both sides and pressed into semi-fmished products in multiplaten presses. Between the heated pressing plates, the resin is plasticised, completely impregnates the webs and solidifies. In the production of tubes on this principle, the resin-coated webs are wound onto a mandrel and hardened by the action of heat and, usually, pressure as well. The most important laminates are resin-bonded paper and fabric as well as vulcanised fibre , which are supplied as boards, coiled, compression-moulded or non--compressionmoulded round tubes, solid bars and strip. 2.~.5
Injection Moulding
The characteristic feature of this process is that the plasticised material (injection moulding compound) is injected into either a cooled primary shaping tool (injection moulding tool) in the case of thermoplastics where the material solidifies due to cooling or into a heated tool in the case of thermosets at high pressure and solidifies there under the action of pressure. In Fig. 25 the material to be processed is fed to the heating cylinder of the extruder as pourable granules or powder. It is plasticised in the heating cylinder and injected in this state through a nozzle into a closed primary shaping tool at a pressure of 80 to 180 N/mm', where the material solidifies. The screw or piston is Withdrawn, the injection moulding tool is opened and the injection-moulded part is removed. 2.~.6
~ ; ~ -
c:
Transf"er Moulding
Transfer moulding is a primary shaping process in which the material to be processed (transfer moulding compound) is plasticised in a pressurised cylinder (loading chamber) and then transferred to a closed primary shaping tool (transfer moulding tool), where it solidifies. The loading chamber is filled with a measured, suitably pelletised quantity of the preheated transfer moulding compound, which has to correspond to the mass of the moulded part, the inlet and the distributor.
_.
:
...: ';,'
-
.
d
e
Injection mould ng COOIpound (granules)
f
Plaslicised injection mouldIng COOIpound Solidified injection moulding wnpound (i/1jeclion-moolded pan)
Compression Moulding
The characteristic feature of compression moulding is that the material to be processed (compression moulding compound) is softened in the primary shaping tool (compression moulding tool) under the action of heat and pressure, fiIls the hollow space with the tool closed and then solidifies. The heated primary shaping tool is filled with a quantity of compression moulding compound (powder, pellets, granules) corresponding to the mass of the moulded part, usually in preheated condition. At a pressure of 8 to 80 N/mm2 the compound fills the hollow space and begins to solidify. When sufficient cross-linkage of thennosets or sufficient cooling of thermoplastics has taken place, the mould is opened and the moulded part ejected. Compression moulding can be used to produce plastic components both with and without fillers. 2.~.7
b
Figure 2S. Injection moulding [4]. a Injection moulding tool is closed; b Nozzle is placed in position. c Moulding compound is injected and compressed. d Injection-moulded part solidifies. Moulding compound is metered in and plasticised. e Nozzle is withdrawn. f Injection moulding tool is opened and injection-moulded part is ejected.
2.~.8
Expanding
Production of parts from cellular materials plays a role in the case of high-polymer materials. The resulting parts contain only a fractional amount of the actual material, while a high proportion of their volume consists of hollow spaces (bubbles, voids). In the expanding of high-polymer materials, a distinction is made between three methods of working: I. First, a carrier foam is fonned by stirring air into a foaming agent (e.g. soap solution). Into this foam is poured the solution of a hardenable plastic, which spreads over the lamellae of the carrier foam and solidifies there
(churning process). 2. Two substances are mixed together that react with each other either immediately or only under the action of heat with liberation of gas, foam the material to be processed and then solidify (mixing process).
2.4 Forming of Metals and Ceramics by Powder Metallurgy. 2.4.3 Technology
3. To the plastic to be processed is added a special foaming agent, which is mixed with the molten material at atmospheric or elevated pressure. Cooling produces an expandable mixture. On reheating the foaming agent expands or decomposes, resulting in an expanded material whose structure is fixed by cooling.
2.4 Forming of Metals and Ceramics by Powder Metallurgy 2.4.1 General For terminology see DIN 30900. Powder metallurgy comprises the production of powders from metals, metal alloys and metal compounds (e.g. carbides, borides, silicides, nitrides, oxides and metals) and their conversion into semi-fmished and tlnishcd components. In this primary shaping process, powder with a particle size depending on the manufacturing process - of less than 0.5 mm (= 0.1 to 500 fern) is usually mechanically compacted (pressed) in moulds and generally converted into rigid finished parts by sintering under shielding gas at high temperature. Pressing takes place at room temperature or, in some cases, at elevated temperature (hot pressing) in moulds made of wear-resistant or high-temperature steel. The sintering temperature (to obtain cohesion of the particles by diffusion) is of the order of ~ to ! of the absolute melting point of the metal in the case of unary systenlS, but is often above the melting point of the lowest-melting component in the case of polynary systenls. If the powder mixture has a heterogeneous structure, it is entirely possible for a small quantity of the liquid phase to be present. It is essential to avoid extensive nldting. Bronzes and the like, for instance, are sintered at 600 to 800°C, ferroalloys at 1000 to BOO °c, hard metals at 1400 to 1600°C and the high-melting metals like molybdenum, tungsten and tantalum at = 2000 to 2900 DC. As the compression pressure (= 1 to 10 kbar), the sintering time and the sintering tenlpcrature increase and the particle size decreases, the density increases up to that of practically non-porous material. Consequently, technologically desirable porosities can be deliberately estahlished (Fig. 26).
Powder metallurgy is only economic for large production runs, hecause of the expensive pressing tools. and then chiefly for smaller components (less than one to a few thousand grams) of the Simplest possible shape, owing to the poor mould-filling ability compared with casting, the limitations with regard to suOicient and above all uniform
Up 10 50
·/.1
Up to 30 'I.
Up to 15 'I. I Up to 5
Examples 01 uses Den~ty
IFilters
~I====:;::::=======~I ~il-impregnated
Up to 20 '1.1
'/,1. I.
1. High-melting metals like molybdenum, tungsten, tantalum, niobium. With fusion metallurgy, in addition to the high temperature further problems are caused by (in some cases undesirable) strong reactions between the molten metals and the refractory crucible or the kiln lining, and by high gas solubility. 2. Hard metals as cutting materials. Production of a composite-metal-like structure from brittle hard metals like tungsten, molybdenum and tantalum carbides and a tough bonding metal like cobalt, which is liquid at the sintering temperature.
3. Composite products made from unalloyable or difflcult-to-alloy constituents, e.g. metal-bearing carbon bushes made from copper and graphite with the good conductivity of copper and the excellent sliding properties of graphite; materials for contacts with the high hardness of high-melting tungsten and molybdenum and the good conductivity of low-melting copper and silver; "diamond metals" made by homogeneous sintering of fine-grained hard materials such as diamond particles or corundum into a tough metallic matrix.
4. Filters and porous (oil-impregnated, self-lubricating and in some cases sealed-for-life) bearings with evenly distributed. interconnected pores; pore size and volume can be deliberately adjusted over a wide range. 5. Alloys of a metal with a high melting point and a metal whose boiling point is exceeded at this temperature, i.e. one with a high vapour pressure (c.g. iron,
cobalt, nickel on the one hand and zinc, cadmium, lead, etc. on the other).
2.4.2 Uses
ProportLOn 01 pore volume Pore vdume
compaction, and the low strength in unsintered condition. Disadvantages are the high capital expenditure requirements for presses, tools and kilns, the complex volume change conditions during pressing and sintering (for solid products, up to 20% linear contraction during sintering), the reiatively limited design possibilities and the gencrdlly lower strength and toughness compared with cast parts. Advantages are the low manpower requirements, the high yield, the high dimensional accuracy (after calibration) and surface quality and, in particular, the materials technology possibilities that are only feasible with powder metallurgy. Important technical applications that can only (or more easily) be achieved by sintering occur in the following t1elds:
. journal beanngs
1Componlll1lS
h'9hIFallty strength oonponents
I. High-strength compor1ents
Figure 2;6. Specifically variable porosities of sintered parts with regard to their usc.
6. Where very brittle materials that arc difficult or impossible to machine are to be converted (e.g. iron-aluminium-nickel-cobalt-copper-based permanent magnets or iron-aluminium-chromium-based brittle high-alloy steels), and machining would be comparatively time-consuming and expensive with other forming processes (e.g. mass-produced small components made of ferrous and non-ferrous metab), or a very high purity and a homogeneous composition are required (which cannot always be ensured with melting and the inevitable batchwise operation). 2.4.~
Technology
The sequence of manufacturing operations for sintcred components can be divided into four: powder production, forming, sintering and after-treatment.
Powder Production. This is carried out with particle sizes of = 1 fern to 0.5 mm by mechanical methods (crushing, grinding, granulating, atomising), physical methods (condensation) and chemical methods (reduction, electrochemical and electrolytic processes, decomposition) .
':.1:. f-7
Manufacturing Processes. 2 Primary Shaping
t--V
f--
V V V ,-/ / . / ""'-1
,..-
90
q:
I
.-!.'-V /
1
)- ./
V
./
/'
4 Compression pressure In kbar
./ :.--'
20 a
» __ J.-
/'
b
Formittg. F?rming into semi-finished and fmished components is mainly carried out by cold pressing in wear-resistant moulds, but also, for better compressibility, by hot pressing and pressure sintering, as well as by explosive compaction and, lastly, also by extrusion and powder rolling. Besides these, forming is also practised without compaction by simple gravity flow of powder or with suspensions of powder in liquids (slip casting), the result being a sintered component with high porosity even after sintering (e.g. metal ftlters). As the compression pressure is increased, the compaction ratio and consequently the density and space filling increase too (Fig. 27). As the rate of pressure transmission in powder mixtures is not uniform as with liquids, in order to achieve optimum unifOrmity of compaction, and thus a homogeneous material condition, the height of the compact or its height/diameter ratio is limited to "" 2 : 1 or, in favourable special cases, to 3: 1. Homogeneous compaction of complex components inevitably requires expensive tools with plungers at varying heights.
Cold Pressing Method (Fig. 28). Care should be taken when designing components to ensure that they can be manufactured at all by pressing and are as simple as possible to make. Guidelines for design (see [8]): adherence to dimensional limits and conditions: height/width < 2.5, wall thickness > 2 mm, holes > 2 mm. Insufficient tolerances should be avoided: holes 2" IT7, width > IT6, height 2" ITl2. Sharp edges, acute angles and tangential transitions should be avoided.
Figure 27. Compression behaviour of metal powders (12]. a Density after pressing versus compression pressure. b Volume ratio versus compression pressure. Harnetag iron powder < 0.3 mm: 1 unannealed without additives, 2 annealed without additives, 3 annealed with 0.5% lithium stearate, 4 theoretical density (non-porous material, 7.86 g/cm~).
Hot Pressing. This is chiefly used with powders made from brittle materials for better compaction. Pressure slntering, which is often used for composite materials, produces high compaction at a relatively low compression pressure. Better compaction also results from isostatic pressing, in which a specimen in a closed plastic or rubber envelope is subjected to very high pressure from all sides in a compression liquid by applying pressure with a plunger. Even higher pressures, required with difficult-tocompress powders, can be achieved by explosive compaction.
Powder Rolling. In this continuous compaction process for producing strip, pressures of several kilobars, as in conventional cold pressing, are applied in the roll gap. Lead bronzes and other composite materials that cannot be manufactured by fusion metallurgy are already being produced by this method on an industrial scale. Extrusion enables bulk or pre-
converted to various sections and to tube with a practically non-porous material.
Forming Without Compaction. This enables sintered products with high porosity and of simple shape to be manufactured, depending on the pouring method. In slip casting, fme powders are generally mixed with the smallest possible quantity of water or other liquid to form a castable slurry and poured into porous moulds which absorb the liquid, resulting in a practically dry porous moulded product which is then sintered. Among the materials suitable for this process are powders made from nickel, copper, bronze or ferroalloys. Another major advantage is that it can be used to convert difficult-tD-<:ompact compounds like oxides, nitrides and silicides, from which it may only be possible to manufacture components, including complex ones, by this method.
Sintering. The particles of powder are permanently a
b
bonded by diffusion in muffle, hood-type and often through-type furnaces under shielding gas (hydrogen, NH3 reforming gas and (to prevent carburisation) partly burnt methane, city gas or producer gas), and, occasionally, under vacuum. Heating takes place electrically or, with
c
d
Figure 218. Cold-pressing processes [131: a uni-directional pressing, b bi-directional pressing, c pressing with sprung: sheath, d pull-off method.
very high-melting metals, by direct current passage. Stages in this solidification process are the adhesion of the particles (even at room temperature, promoted by high compaction), bridge fortnation between the particles, and compaction of the moulded product, sometimes into a material with practically closed pores. The operations pressing (pouring), slntering and calibration (after-pressing to increase dimensional accuracy)
3.1 Classification and Introduction
Chromium
Do ~___________&_"_~_'____ SS_SSS_$(_S_~~H$ 11~
Platinum Rhodium
I
:WU
burising, nitriding, etc.), heat treatment and impregnation (increasing strength by ftlling the pores with low-melting metal, or with oil in the case of self-lubricating bearings).
r--= 2.5 Other Methods of Primary Shaping
Nickel Palladium
2.S.1 Electroformming
Cobatt
In electroforming, a metal deposit is produced on a
Iron
-
W=\§%i
Copper
~=----.---------------
Silver
- -..- ..- - - - - - - - - - - - - - - 1
Cadmium linc
~
r----------------------------~
Tin
Lead
o
=
=
=
100 200 300 400 500 600 700 800 900 1000 Hardness (Brinell or diamond-tipped pyramid scale)
Annealed, cast or machined Electrolytically deposited metal: max. known range Electrolytically deposited metal: normal practice
Figure :I!J. Hardness of various electrolytic.fir deposited metals.
may be performed in varying combinations to a certain extent, e.g. in:
Single pressing. Pressing-demoulding-sintering (also calibration where necessary).
Double pressing (high density, improved mechanical properties). Pressing-demoulding-sintering-pressingsintering (also calibration where necessary). Pressure sinterlng. Sintering in the mould (for difficultto-press powders; high density with comparatively low compression pressure)-demoulding-(calibration where necessary) .
Aftertreatment. The sintered mouldings may be given their final, ready-for-use condition by a variety of aftertreatment methods, depending on the application, e.g. by forming without cutting (calibration or pressing, rolling, drawing), machining, surface treatment to protect against corrosion or to increase wear resistance (chromising, car-
moulded cathode by e1ectrodeposition. This deposit is either removed as a solid product (electro) or, occasionally, remains bonded to the mould (core electro). Electroforming is suitable for both single items and for series production. The mould, a negative of the part to be formed, is produced by machining, casting or pressing. A positive original mould, on which the negative moulds are made by e1ectroforming, is used when it is not possible to make a negative mould. A positive mould is also the starting point if a sizeable quantity of negatives has to be manufactured for series production. Examples of materials deposited from aqueous and organic solutions and from salt melts for making shells are: iron, chromium, tungsten, niobium, copper, nickel, cobalt, tin, aluminium, silver, gold, and alloys of nickelcobalt, nickel-chromium, nickel-manganese, cobalt-tungsten, cobalt-tungsten-nickel-iron. Copper, nickel and, increasingly, nickel-cobalt alloys have in practice attained prime importance, in the last-named case not least because of the favourable combination of mechanical, wear and corrosion properties (Fig. 29). Advantages are the extremely high copying accuracy and, with the appropriate mould quality, the high dimensional accuracy. Peakto-valley heights as small as 0.5 ",m can still be copied exactly. The parts manufactured by electrodeposition are on the whole able to satisfy the specifications as to, e.g., high surface quality, uniform layer thickoess and required properties in use (structural parts, stencils for screen printing, spinnerets, household articles like pots, shakers, sieves and filters, including micro-filters).
2.S.2 A.utocataIytic PlatlAg Whereas in electroforming the metal ions are reduced by absorption of electrons at the cathode (external current source) and are deposited there as metal or metal alloy, in autocatalytic plating the reduction is accomplished by electron exchange with reactants which, in this non-electrical deposition process, inevitably change to a higher level of oxidation. This requires the use of catalysts.
Metal Forming K. Siegert, Stuttgart
3.1 Classification and Introduction Metal forming, according to DIN 8580, is the deliberate alteration of the sbape, surface and material properties of a workpiece while preserving its mass and cohesion. The workpiece is generally made of pure metal, a metal alloy manufactured by fusion or powder metallurgy, or a composite material.
Classification of Metal Formming Processes. There are various possibilities, as follows.
One possibility is to classify them according to the pre· dominantly active stresses (loads). Thus they are classified as follows: Forming under compressive conditions (DIN 8583) Forming under a combination of tensile and compressive conditions (DIN 8584) Forming under tensile conditions (DIN 8585) Forming under bending conditions (DIN 8586) Forming under shearing conditions (DIN 8587) Another possibility is classification into sheet/plate forming processes and massive forming processes. A very important question is whether or not forming
...
.:
Manufacturing Processes. 3 Metal Forming
produces a change in strength. Therefore a distinction is made between processes where fonning produces no change in strength, those where a temporary change in strength occurs during forming, and processes where forming leads to a permanent change in strength. Depending on whether the workpiece is heated before forming, the term cold forming or hot forming is used (DIN 8582), In cold forming processes, where the workpiece is introduced into the forming process at room temper-ature, with metallic materials whose recrystallisation temperature is significantly greater than room temperature, there is generally an increase in yield strength and ultimate tensile strength accompanied by a decrease in elongation after fracture as deformation increases. This is termed work-hardening. Furthermore, a distinction can be made according to the method of applying force, viz. forming processes with direct application offorce and those with indirect application of force. For instance, wire drawing, where the drawing force is applied to the forming zone by means of the already-drawn wire, is a process of indirect application of force. Forging, where the force is directly applied to the forming zone by means of the tool, is consequently a process of direct application of force.
The Forming Process. This is determined by several factors: the workpiece, the tool, the lubricant, the surrounding medium and the machine (including process control). Furthermore, mechanised or automatic transfer of the workpiece into, out of and between the tools should be taken into account. The tribological systems of the forming process are determined by the workpiece, the tool, the lubricant and the surrounding medium (Fig. 1; ct. 05). In describing the workpiece (e.g. metallographic structure, temperature, geometry, surface as well as technological data such as yield point, tensile strength, elongation after fracture and stress/strain curve), the following conditions should be referred to:
On delivery Immediately before forming During forming Immediately after forming After ageing at room temperature or after heat treatment For the initial parameters of the forming process, the condition of the workpiece immediately before forming is relevant.
It is important to view the forming process as a link in the "manufacturing chain" of a component. Thus the manufacture of the blank to be formed influences the forming process significantly. For example, forging of a cast blank can be optimised only if the alloying constituents, the structure resulting from casting and the heat treatment prior to forging are known. However, the method of further processing or treatment after forming, such as heat treatment, subsequent forming operations, machining, surface treatment, etc. should also be known in order to optimise the fOrming process, as the entire manufacturing chain determines the characteristics of a component. Optimisation of the forming process should therefore be carried out in the knowledge of and in coordination with the preceding and subsequent manufacturing processes.
3.2 Fundamentals of Metal Forming ~.2.1
Flow Stress
Flow of a material takes place when permanent deformation is caused by a specific stress condition. The flow stress k f (also known as the yield strength) is, in a uniaxial tensile test, the tensile force F in relation to the respective instantaneous cross-sectional area A at which the material flows, i.e. undergoes permanent deformation: kf
= F/A.
(1)
(Note: with IT = F/Ao, the force F is expressed in relation to the original cross-sectional area An.) ~.2.2
Characteristics of'Material Flow
The logarithmic deformation (degree of deformation) describes the magnitude of the deformation. In the Cartesian system of coordinates, the following applies: (2)
In the polar system of coordinates, the following is obtained: 'Pr = In
~ = CPt = ro
In
~. ro
(3)
If a body with the dimensions 10 , b o , ho is transformed by metal forming into a body with the dimensions 11, bl> hi, at constant volume
or
lc.,I?!.,1?!. = 1
( 4)
10 bo ho
Taking logarithms gives the following: (5)
Using Eq. (2), Eq. (5) can be written as 'PI
Figure 1. Tribological system a<.:cording to DIN 50320; 1 primary object, 2 opposing subject, .3 intervening substance, 4 surrounding medium.
+
'Ph
+
'Ph
=
O.
(6)
The total of the logarithmic deformations thus equals zero: l'P = O. The rate of deformation is the derivative of the logarithmic deformation in reaction to time: ip = d'P/ dt.
(7)
3.2 Fundamentals of Metal Forming. 3.2.4 Flow Curve
The acceleration of deformation is the derivative of the rate of defonnation in relation to time: ip = dcp/dt.
CPg, the rate ofprincipal deformation CPg and the temperature it of the material to be formed:
kf
(8)
=
f(cpg, CPg, it).
ation of principal deformation ipg.
The change from elastic defonnation to permanent plastic defonnation is described by flow criteria. In elementary metal forming theory, Tresca's shear stress hypothesis is generally applied (cf. B9.2.2). This says that flow occurs when the maximum possible shear stress T_ reaches the flow stress in shear k of a material: Trn~
= k.
(9)
In cold forming of metallic materials at forming temperatures well below the recrystallisation temperature, (20) For most materials (e.g. low-alloy steels, copper, brass, aluminium) the flow stress k f depends only on the principal defonnation CPg: (21)
It can be seen from Mohr's circle (cf. B1.1) that (10) where""",,, is the most positive and "min the most negative principal stress. For the uniaxial stress condition ("I 0,
*
"2 = "3 = 0,
k f = 2Trn~ = ("""", - "min)'
(11)
It should be noted, however, that with large deformations and rates of defonnation, even in cold forming (original temperature of material being fonned = room temperature) such high temperatures may occur in the area of the forming zone (e .g. cold extrusion of aluminium) so that the conditions of Eq. (20) no longer apply. If Eq. (20) applies, for most metallic materials the flow curve can be described by the approximation
This relationship is referred to as "Tresca's shear stress
hypothesis". According to this hypothesis, the principal deformation CPo is the numerically largest logarithmic deformation (12) A further hypothesis frequently used in metal forming is the maximum distortion energy theory fonnuiated by von Mises and Henky (cf. B1. 3.3). This says that flow occurs when the elastic shape modification energy "2, reaches a critical value. With the principal stresses the following applies:
"I>
"3
k f = ~! [("1-"2)2
+ ("2-"3)2 +
("3-"1)2]. (13)
If the mean stress is (14) Eq. (13) gives k f = ~(3/2) [(U I -Urn )2
+
(u2 -urn )'
+
(u,-urn )']· (15)
In pure shear stress, kf =
{3Tm~'
k f = acp",
~(2/3)(cp~ + cP~ + ~).
(16)
(17)
The principal defonnation CPg calculated according to the maximum distortion energy theory is also tenned the
comparative deformation CPV' Law of now. For isotropic materials, according to [1] the following relationship applies between the principal stresses "I, "2, "3 and the associated logarithmic deformations, taking Eq. (14) into account: CPI/CP2/CP, = ("I - "rn)/("2 - Urn)/(U3 - urn)· (18)
Therefore, if a principal stress equals the mean stress "m, the associated deformation is zero. ~.:I.4
now Curve
The flow stress k f of a material that is required to achieve and sustain flow depends on the principal deformation
(22)
where k f 2: RpO.2 or ReH (see D2.1). The exponent n is tenned the consolidation index, as it determines the rise of the flow curve. A high index n indicates that the material's hardness increases very rapidly with increasing deformation. As Eq. (22) is only an approximation, indication of the range CPgI os CPg os CPg2 to which an index n applies is recommended. If the flow curve is depicted on a double logarithmic scale, a straight line with the gradient n results for Eq. (22) (Fig. :I). For hot forming, it is generally the case that the flow stress falls as the temperature increases and rises as the rate of principal defonnation CPo increases. The influence of the principal deformation cP decreases with higher degrees of defonnation at elevated temperatures (Fig. ~). A plot of the flow curve is generally made at room temperature for a uniaxial tensile test in the zone of unifonn elongation [3] and for a uniaxial compression test [4]. At elevated temperatures and high degrees of deformation, flow curves are generally determined using a compression test and a torsion test. Other methods are given in [5] and [6].
According to the maximum distortion energy theory, the principal defonnation CPo is CPo =
(19)
In high-speed forming, k f also depends on the acceler-
now Criteria
~.:I.~
1:"...1
tan«~n
19 kf ~ n 19 'Pg+ 19«
Figure~.
Typical flow curve for it ~ {}recr.
Manufacturing Processes. 3 Metal Forming
C1
180 r - - - - - , - - - , - - - - , - - , - - - - - - r - - ,
n !--I '
i
160 I-----t-----t---t---t-l\lo" ~~--I
.
I
hs
I
Figure 4:. Earing as a result of planar anisotropy.
The vertical anisotropy is often denoted by the mean vertical anisotropy (25)
For denoting the suitability of a sheet metal for deepdrawing, however, this indication seems appropriate only to a limited extent. A better way is to denote the vertical anisotropy by the rmin value, The suitability of a sheet metal for drawing of rotationally symmetrical pots with as little trimming waste as possible is denoted by the planar anisotropy:
40bL--t-----t----t--
a
O,Z
zoo 1-20'C f-120'C
100
EEO 80 ;;;, .S
-><
60
r- 240'C
------~ ---
40 ~
20
-- ---- -----,.--
------
0,25
..........
J 53
b Flow curves of A199.5 [2): a flow stress in relation to principal deformation 'Pg at ip = 4 5- 1 , b flow stress in relation to the rate of principal deformation (Pg at qJg :=: 1.
Figure~.
(26) ~.2.6
Fonnability
Formability means the plastic deformation that a specific material is able to withstand in the fOrming zone up to the point of fracture at a specific stress condition, temperature and rate of deformation. Other parameters such as acceleration of deformation may also be relevant at extremely high rates of deformation. Formability, measured for instance as the deformation on fracture, depends greatly on the stress condition. The more negative the mean stress according to Eq. (14) or, in other words, the greater the mean compressive stress, the greater is the formability [7]. Here, however, the principal stress CT, is also relevant if CT] > CT, > CT3 applies. At the same mean stress
U ITJ
the fonnability is greatest when u 2 becomes
equal to CT,. It decreases as CT, becomes larger and is smallest when CT, = IT] [8]. Figure S illustrates the deformation on fracture as a measure of formability by means of the mean stress in relation to k,. Warning: In processes with indirect application of
~.2.S ~sotropy
Anisotropy exists if a material exhibits direction-specific properties. In sheet metal forming, vertical anisotropy r is defmed as the ratio of the logarithmic deformation of the width to the logarithmic deformation of the thickness in a uniaxial tensile test:
r = 'PhI'!',.
(23)
If r > 1, longitudinal flow of the material occurs more from the width of the sheet than the thickness. If r < 1, the material flows more from the thickness of the sheet. In sheet metal forming the highest possible values for r are aimed at. It should be noted, however, that r generally depends on the position of the sample in relation to the direction of rolling. In general, ro for a sample at 0° to the direction of rolling, r" for a sample at 45° to the direction of rolling and roo for a sample at 90° to the direction of rolling are determined, If ro oF r45 oF r90 , eaTing (Fig. 4) occurs in deep-drawing of rotationally symmetrical pots, i.e. the height of the pot is not constant over its circumference, Z
=
b -b (b B
g
T
+ b T )/2
.100%.
(24)
3>" Tensile test without clasping pressure (with necking)
-113
113
Figure S. Defonnation on fracture as a measure of formability over the mean stress Urn in relation to k f [8]; 1 tensile tests with clasping pressure, 2 torsion tests with axial tensile stress, 3 notched bar tensile tests, 4 compression tests with clasping pressure, 5 compression tests with radial tensile stress.
3.3 Theoretical Models
&&&
c±) c±) c±) c±) c±) c±)
J
c±) (j;) (B
1
do
c:=:>
':.41
1.0
eee
O.B
f)f)&,--i --../1
'" ~
0.6
III ;
0.4
...--
/
Figure 6. Defonnation analysis in sheet metal forming by grid measurement.
0.2 force, the failure case "fracture" generally occurs outside the forming zone. This is a question of the limits of deformation, which are generally process-specific. ~.2. 7
Forming Limit Curve (FLC)
In sheet metal forming, the deformations are often analysed by marking the sheet with a grid of circles (diameter of circles e.g. 4.5 mm) before forming and measuring the resulting ellipses after forming [9] (Fig. 6). The forming limit curve is obtained by plotting against one another the deformations 'PI and 'P2 in the plane of the sheet at which necking and fracture occur. The larger deformation is plotted above the smaller deformation (Fig. 7). This curve applies only if the path of deformation, until failure due to necking or fracture, occurs at a constant ratio of 'PI to 'P,. It should be noted that deformation of the thickness 'P, is calculated from Eq. (6) as (27)
Figure 8 shows a forming limit curve for 'P" 'P2 and 'P, = 'P, for the start of necking [10].
a
-
-02 I-- .-0.4
r-
'\
-0.6
\
"""
-O.B b -1.0
-04
0.2
-0.2
0.4
'1'1 Figure 8. a Forming limit curve. b Deformation of thickness according to Eq. (6).
3.3 Theoretical Models The elementary theory of plasticiry (cf. B9) originates from work by Siebel, Karmann, Sachs and Pomp [11-14]. This elementary theory was revised, generalised and expanded by lippmann and Mahrenholtz [15,16] (ct. also [17, 18]).
Three fundamental models arc taken as the basis (Fig. 9). The following assumptions are made for the discussion which follows: Homogeneous forming (pure elongations/shearing strains) in the individual strips, discs and tubes. The principal axes correspond to the axes of the bodies. The strips and discs remain flat during forming, the tubes remain
c Figure 9. Basic models of elementary fonning theory: a strip model, b disc model, c tube model; 1 strip, 2 disc, 3 tube.
Larger log deformation 1/>,
2
cylindrical. In this way of looking at the situation, the real conditions are deliberately ignored.
Homogeneous, Isotropic Material Friction According to Coulomb's Law of Fric· tion. Friction is constant over the area of contact between tool and workpiece T
Small log deformation q" Figure 7. FOrming limit curve (FLC); 1 fracture, 2 necking,
.3 strain path.
= P,Pn
(p,
= const).
(28)
Although forces of graviry and inertia can be allowed for in the model, they are usually negligible. They are not allowed for in this deSCription .
Manufacturing Processes. 3 Metal Forming
The flow stress kr is constant over the strip, disc or tube,
Strip Model This model was developed by Siebel and von Karman. According to Fig. 9, a strip of the material being formed is considered, which is formed from rwo parallel boundary surfaces a differentially small distance apart which are bounded at top and bottom by the forming tool (Fig. 10). As this strip width dx is required to be differentially small, the top and bottom boundaries can be described by straight lines to be regarded as tangents to the tool contour. The angle of these tangents to the horizontal is a function of x, as is the height of the strip:
"I
-(0', +dd,) = (p, +dp,)
Figure 11. Stresses acting on strip element.
= I(x); '" = I(x); h = I(x).
In the strip theory model the strip is assumed to be formed in such a way that the two cross-sectional surfaces bounding the strip remain flat and parallel to each other. Every time this theoretical model is applied to a specific metal forming process, these assumptions should be checked as to what extent they adequately describe what actually happens. If the forces of inertia acting on the material being formed are ignored, only stresses that act directly on the crosssectional surfaces and those that act directly on the boundary surfaces have to be taken into account. If there is friction berween the material being formed and the tool, edge shear stresses occur at the boundary surfaces. Shear stresses may also occur at the cross-sectional SUffaces if the tool contour forces the strip to undergo sudden deformation. Thus Fig. 10 shows that as the strip enters the forming zone, a jumps from 0 to a l or a 2 and jumps back to a = 0 as the strip leaves the forming zone. These shear stresses occurring at the cross-sectional sur-
h(x)
I
F
F+dF
dx
Figure 12. Forces acting on strip element.
faces are not considered for the time being. They are taken into account later as "losses due to shear strain"
(or, in the more recent literature [15], "tangential
adjustments") . According to Fig. 11, the stresses acting are therefore the compressive stresses on the cross-sectional surfaces Px and (Px + dpx), the compressive stresses on the boundary surfaces Pnl and Pn2> and the edge shear stresses 71 and 7 2 According to Coulomb's law of friction,
F
h(x)
F+dF
(29)
From the stresses, the forces acting directly and tangentially on these surfaces are obtained by multiplying by the respective surface areas (Fig. 12). By resolving the forces into horizontal and vertical forces, the following apply at the boundary surfaces (Fig. 1~):
dFvl Figure 13. Resolution of forces acting on strip element.
dFlIl
= Pn, dx elz(tana, +
IL,);
dFH ,
= Pn, dx elz(tana, +
IL,)·
dFvl
= Pnl dx elz(l -
ILl . tana l );
dFv ,
= Pn, dx elz(l -
IL2 . tan",).
(30)
(31)
If Py = dFv/(dx dz) is defined as the vertical compressive stress, with tan P = IL one obtains:
Figure 10. Strip model.
dFlIl
= Pyl dx elz tan(a , + P,);
dFH ,
= Py 2 dx elz tan( a 2 +
(32)
P2)'
Thus the horizontal and vertical forces at the boundary surfaces of the strip elements that are plotted in Fig. 1~ can be described by Eqs (30), (31) and (32).
3.4 Stresses and Forces in Selected Metal Forming Processes. 3.4.3 Wire Drawing
1~*"'1
- 'd, "p,
Disc Model The strip model is based on a strip of depth dz. For metal forming processes with an axially symmetrical forming zone (e.g. extrusion, wire drawing), it is advisable to imagine that in the forming zone the material being formed consists of "discs" (Fig. 9). These discs have a differentially small thickness and a defined outside diameter. A defined inside diameter is also used where applicable. For the disc model it is assumed that the discs are formed in such a way that the cross-sectional surfaces remain plane and parallel to each other. For solid crosssections the contour can be described by D
= j(x).
(33)
To avoid misunderstandings in differentiation and integration, the diameters are designated by U", Dl and D = I(x) (Fig. 14). The boundary surface of the disc is formed by the tangential surface at the contour of the forming zone at point x. The gradient of these tangential surfaces is obtained from
forming zones, 0' = const. Thus the following applies to the boundary surface ciA (x) of the disc element:
=
Solving this first-order differential equation, taking into account that a, = 0 at r = dl2 and p, = ~(}'" gives:
(34)
d(D(x»ldx.
If x is the angle of slope of the tangent and also a function of x according to Eq. (34), then 0' = I(x). For conical
ciA(x)
Figure IS. Stress conditions in tube model.
D(x)rr dx/cos a.
P, = kr,exp[2: «d/2)
P,
As in the strip model the boundary surface dA(x) = 0', Eq. (35) gives the following by analogy with 0'
+
1-'),
(36)
dF, = PnD(x)rr dx(1
~
I-' tan
(37)
0'),
and gives the following with the radial compressive stress
=
kf[1 + 2:
p).
(41)
= 0), (42)
The upsetting force F, is given by integration of Eq, (41) over the compressed area
( 43) as
,
F, = krAo
+
~ r)].
p, = k,.
p, by analogy with Eq. (32): dFH = PI dx D(x)rr tan(a
«d/2)
Under frictionless conditions (I-'
dz dxlcos Eq. (30):
dFH = PnD(x)rr dx(tan
(40)
By series expansion and truncation after the first term, this gives
(35)
and by analogy with Eq. (31),
~ r)].
(38)
[I + -J-td] , 3 1
~
h
(44)
At the end of the opet'Ation, i.e. when d l and hi are
attained,
3.4 Stresses and Forces in Selected Metal Forming Processes
(45) If the resistance to deformation is designated as
3.4.1 Upsetting of Cylindrical Parts Use is made of the tube model, Fig. 15. The equilibrium of forces gives, with <7, = au dr' dx = 0, sin(daI2) == dx/2, Tresca's hypothesis and assumption of Coulomb friction,
kw
= I
kf [1 !
1 J-tdl] + -:--, 3 hi
(46)
Eq, (45) can also be written as:
(39)
(47)
3.4.2 Upsetting of Square Parts Using the strip model, the following is obtained by analogy with Eq. (45): (48)
3.4.3 Wire Drawing
Figure 14. Geometrical conditions (example: wire drawing).
applying to
disc
model
In drawing the wire, the original diameter of the wire D" = DE is reduced to the diameter Dl = D A . The die, also known as the drawing hole, is the forming tool here, The drawing force acts on the emerging wire and is transmitted by this means to the forming zone. This process
1:W
Manufacturing Processes. 3 Metal Fonning
is therefore one with indirect action of force (Fig. 16). Characteristic geometrical values are A. =
1rm/4;
AA = 1rDi/4;
A(x) = 1rD(X)2/4.
(49)
l,,(D. - D A )/2 tan a; D(x) = DA
(51)
If the shear strain losses are ignored and frictionless conditions are assumed (p. = 0), the so-called ideal stress IT'd is obtained. It is given by Eq. (36) and the shear stress hypothesis as: IT'd(X) = lTx(x) = 2kfm ·In D./D(x).
(52)
Here, kfm is the arithmetical mean of the flow stress on entering the forming zone kfE and the flow stress in the cross-section under consideration k,(x): kfm = (kfE
+ k,(x))/2.
(53)
At the exit from the die, the following applies: IT'dmu
= IT'd(X = 0) = 2k, In (D./DA ),
(54)
where (55) k'A is the flow stress in the exit plane of the forming zone. Multiplied by the relevant cross-sectional area, the ideal drawing force is given with the aid of Eq. (54) as:
(56) Here the logarithmic total principal deformation is
(60)
fl.
If this portion is added to Eq. (58), the result is [19]
(50)
+ 2x tan a.
2
= 3" Ilf tan
U disc
lTx'm
=
k,_ j[
1
tan a
+ ---,;:-
][ (DJ 2P.] + 3" 1-
DA;;n.;
2
)
tan a . (61)
By series expansion and interruption after the first element, for the small angles a that occur in wire drawing the following relationship established by E. Siebel is obtained: lTx
lot
_ [1L2a] 1 + -; + - - .
= lTx(x = 0) = k,
3 'P
(62)
~ ~3 ~]. rp
(63)
a
Riot
glnt
The drawing force is given as
F"" = 1r '4Di k, rp..., [1 + a +
......
where lal is the angle a as a radian measure. The optimum angle lal op, is given by Eq. (63) and dF,o,/da = 0 as
=
~1.5p. rpg .
(64)
101
3.4.4 Extrusion In extrusion, the blank is pushed through a forming tool (die). According to DIN 8583, the extrusion processes consist of the forming processes tapering, extrusion and extrusion moulding.
(57)
Under conditions of friction (p. '" 9), the following applies to conical dies on the basis of Eq. (36): ux(x = 0) = k, (1
+
t~ a)[ 1 - (~y:.].
(58)
If the shear strain losses, i.e. the angular distortions of the "discs" on entering and leaving the fonning zone, are taken into account, an axial stress portion ITdbe is necessary for this. According to [19, 20] this is lTdl~ =
1
3" tan a(kre + k'A)'
(59)
Stresses in the Forming Zone The following discussion of the stresses in the forming zone applies to all three processes, assuming that a conical die is used (Fig. 1 '7). If the elementary theory is taken as a basis, the geometrical and kinematic conditions
in the forming zone are identical. The fonning zone is bounded by the wall of the die, the entry plane (x = lu) and the exit plane (x = 0). The difference compared with wire drawing lies in the action of the force. In similar fashion to wire drawing, for tapering the axial compressive stress in the inlet plane is P XE = Px(x = lu)
or, using Eq. (55), (65) For small angles a, tan a = lal. By series expansion and interruption after the first element, Eq. (65) gives
P XE
Fipre 16. Geometrical conditions and basic pattern of stresses in wire drawing with a conical die.
= k, rp...., [1
+
(p./a)
+~
(a/rp...) ].
(66)
FJ.aure 11. Geometrical conditions in extrusion - conical die.
1:*....1
3.4 Stresses and Forces in Selected Metal Fonning Processes. 3.4.5 Deep Drawing where 'P. = 2 In (DE/DA) and k, = (k fE = k'A)/2. In wirtdrawing and tapering, the die angles a divided by 2 are relatively small; thus, for
Pn = P,[l/(l
~
p.. tan
a))
it can be assumed that Pn = p,. This is unacceptable for extrusion. As " p 0, Pn '" p,
andp,
= k, + PX'
This gives, taking into account Eqs (55) and (22),
Px, = Px
(x
[(AE)C" = tu) = k,- C,C,~ 1.A:
1
~ 1 ,
(67)
where C, = (1 + p.. . c tan a)/(1 ~ p.. tan a). If Eq. (67) is expanded to include the axial stress portion |
-j
1+:3
Figure 19. Schematic diagram of deep drawing with punch applied: 1 punch, 2 blank holder, 3 drawing die, Do outside diameter of blank, do punch diameter, So thickness of blank, Fst force of punch, FN force of blank holder, rSt punch edge radius, TM die radius.
(73)
(68)
The compressive stress at the end of the ingot is obtained from this:
This is the axial compressive stress in the die entry plane.
If Coulomb friction applies, the following is obtained from Eq. (72), using Eq. (69) and taking Eq. (68) into
Px,
= k,
C,C,~ 1
[(AE)C" A:
~ 1
2 tan " ) .
PxCx = 10
Stresses in the Material Being Formed Outside the Forming Zone Consideration is given to the stress condition in the cylindrical, upset part of the ingot preceding the fonning zone X E ,,; x ,,; X E + to ~ s according to Fig. 18: for the edge shear stress acting on the surface of the ingot in this area, with Coulomb friction the following applies:
sheared off within a boundary layer: (70) The radial compressive stress p, is calculated from Tresca's hypothesis as
p, = Px
~
k,.
(71)
To determine the axial compressive stress P x in the
range 0 ,,; x ,,; to ~ s, the eqUilibrium of forces at a crosssectional disc is examined: (72) For shearing off, the result is as follows, using Eq. (70) and taking Eq. (68) into account:
= 2(kro/DE) (/"
~ s)
+ Px,
(74)
account:
Px
= PXE exp [4(p..o/D E ) x + kro (1
exp [4(Il-oIDE) xJl (75)
~
(cf. [23-25]). The resulting compressive stress at the end of the ingot is:
Px(x
= 10
~ s)
(69) An upper limit applies if the material being formed is
~ s)
= Px, exp[4(p..o/DE) + kro
(/0 ~ s))
(1 ~ exp[4(/Lo/DE) (/0 ~ s)] (76)
This relationship was detennined by Eisbein [23] and Sachs [24] (cf. [25]). The ram force P" is given by multiplying the compressive stress at the end of the ingot by the cross-sectional area of the ingot: PSI
= Px (x = 10
~ s) . ('ITDi,/4).
(77)
Equation (74) or (76) should be inserted in this equation, depending on the friction conditions.
3.4.5 Deep Drawing Deep drawing is a method of fonning a flat sheet-metal blank into a hollow part. Figure 19 illustrates the tool arrangement and the terminology for deep drawing of rotationally symmetrical parts. The forming zone is the area of sheet underneath the blank bolder up to the exit from the curvature of the drawing die. Figure :10 shows
IFst
FN
~
~ dol2
O./2---j
x Figure 18. Curves of axial compressive stress Px and radial compressive stress Pr over x, assuming Coulomb friction.
Figure 2:0. Schematic diagram of stress pattern for deep drawing with punch applied.
Manufacturing Processes. 3 Metal Fonning
the basic stress pattern for this process. It can be seen that the normal stress
Un
intersects the mean stress
Urn'
According to the law of flow as in Eq. (18), therefore, at this point no deformation takes place in the direction of thickness. It can also be deduced that an increase in sheet thickness towards the flange edge and a decrease in sheet thickness towards the radius of the drawing die entry take place. On average, however, the surface remains approximately constant in deep drawing: According to [26], the total drawing force F,m = F" is (78)
F'd is the ideal force required for no-loss fonning: (79)
with O"d = 10',1 = k'm In (D./do) and kfm = (kfi + kfu)/2. Here, kfi is the flow stress at the drawing die run-out for r = d o/2 and k,. is the flow stress at the outside flange diameter for r = D,/2. These are determined from 'Pg, = In
..j(IYo + d~
- DD/d~,
'Pg ,
= In Do/D,.(80)
FRN is the frictional force atising between the blank and the drawing die and between the blank and the blank holder. According to Paknin [26],
by the movement of the former (Fig. 22a). Owing to the friction between the former and the blank, uniform distribution of the elongations over the component is prevented. Failure occurs between the clamping jaws and the areas not in contact with the formers.
Tangential Stretch Forming (Fig. 22b). This enables uniform distribution of the elongations over the workpiece and a higher degree of forming in the central area. The sheet-metal blank is clamped in vertically and horizontally movable jaws and prestressed with them until the blank has undergone plastic deformation of 2 to 4%. If the formed component is to be embosssed, the stretch forming device can be incorporated in a single-acting press with countermould (Fig. 23). The Cyril-Bath process, in which the clamping jaws can be moved horizontally and vertically by eNC, enables the central area of large, drawn sheet-metal components to undergo a higher degree of fonning and thus enables a higher degree of work-hardening to be obtained [27, 28]. Sheet metal for stretch fonning should have the highest possible consolidation index n, so that the deformations are distributed as unifonnly as possible over the component and a premature local tear is avoided (cf. K3.2.4). The friction between the sheet and the stretch fonning tool should be as low as possible (I-' ~ 0). On materials see D3.1.4.
(81)
Here, FN = 'IT/4(~ - ~)Pn for the pressure of the blank holder Pn, which is set as constant throughout the drawing process. According to Siebel,
Pn
= (0.002
to 0.0025) [(/30 - 1)'
+
0.5(do/lOOso)] Rm,
where /30 = Do/do drawing ratio. F'b is the frictional force arising between the workpiece and the curvature of the drawing die: (82) F,b is the reverse bending force that is necessary after
leaving the drawing die: (83)
According to Fig. 21, the drawing force Fs, has a maximum which, according to [26], is h/hm~ = 0.4 for most metallic materials.
3.S.2 Deep Drawing A distinction is made between:
First-operation drawing (Fig. 19) Second-operation drawing (DIN 8584; Fig. 24) Whereas in stretch fonning the shape of the formed part is obtained by enlargement of the surface area at the expense of the thickness, because the sheet is clamped at the sides and cannot continue flowing, in deep drawing the sheet thickness over the drawn component is approximately constant in first-operation dC-awing, which means
that the surface area of the blank equals the surface area of the drawn product. A minimum holding-down force FN is required in order to suppress puckering underneath the blank holder (cf. K3.4.5). If possible, the sheet metal should have a dr of zero so as to avoid earing. r m'n should be as large as possible, as should the index n (cf. K3.2.5). To reduce friction at the interfaces of blank holder, sheet and drawing die and also in the area of the curvature of
3.5 Technology Original form of workpiece
3.S.1 Stretm Forming
Final form of workpiece
Stretch fOnning is used to produce large sheet-metal parts (VDI 3140). A distinction is made between:
Simple Stretch Forming. The
sheet-metal blank is clamped at two opposing sides. Forming is accomplished
'~ 0.4 Figure 21. Curve of punch force duct height.
Mh mcx PSt
over the specific drawn pro-
Figure 22. Stretch forming: a simple stretch forming, b tangential stretch forming; 1 workpiece, 2 clamping jaw, 3 former, 4 tool.
3.5 Technology. 3.5.3 Bending
Stretch forming
Deep drawing
Figure 25. Drawing of large noo-symmetrical parts by a combination of stretch forming and deep drawing; 1 shape of blank, 2 drawing frame.
drawing (Fig. 25). The flow of material underneath the blank holder is influenced by drawing channels, the shape of the blank, zones of higher and lower surface pressure and a specifically directed lubricant supply. In drawing with a single-acting press, a drawing device is placed in the bed of the press. This can be designed as a pneumatic or hydraulic drawing device [29, 30). Figure 26 shows a tool arr,mgement for drawing with a doubleacting press. Figure 23. Cyril-Bath process (Cyril-Bath Company). 1 to 3: sequence of operations.
.J.5 . .J Bending Bending is one of the most commonly used methods of sheet metal forming. Its uses range from mass production
Figure 24. Second-operation drawing [SJ: J punch. 2 blank holder, 3 support ring, 4 drawing die, 5 cup after first drawing, 6 cup being redrawn.
the drawing die, the coefficient of friction should be as small as possible (IL ~ 0). If the coefficient of friction between the sheet and the drawing die is relatively large (IL ~ 1), the drawing force applied can be increased by means of frictional forces and a larger limit drawing ratio obtained. On materials see D3.l.4. Drawing of non·symmetrical parts (e.g. car body panels) is by a combination of stretch forming and deep
Figure 26. Schematic diagram of drawing of large, irregular, sheetmetal panels with a double-acting press; 1 punch, 2 blank holder, 3 drawing tongue, 4 sheet, 5 drawing channel, 6 bed.
I:W1.,
Manufacturing Processes. 3 Metal Forming
of small parts to fabrication of single items in shipbuilding and industrial plant construction [31]. Apart from sheets, bending is performed above all on tubes, wite and bar of widely differing cross-sections. Cold forming is used in most cases; only in exceptional cases, i.e. for large crosssections or very small bending radii, is the material heated to reduce the forces requited for forming or to enable higher degrees of deformation to be achieved with a given material. Elementary bending theory [32] is discussed in B2.4. A phenomenon typical of bending is elastic springback. After removal of the load, the bending angle is smaller and the radius of the bent part larger than under load. In bending without radial stress, the spring-back on removal of the load is given by the spring-back ratio (Fig. 27):
The degree of spring-back depends on the material (modulus of elasticity, yield point, index n), the stress condition under which forming is carried out, and the initial deformation of the part to be bent. Depending on whether plastic deformation is successfully produced over the full cross-section of the workpiece to be bent, or whether elastic deformations predominate in the neutral fibre zone, the degree of spring-back is large or small. Spring-back can be reduced, prevented or compensated for by the following methods: Limiting the tolerances for the sheet parameters Rp02' index n and sheet thickness as a prerequisite for reproducible conditions. Over-bending. After-pressing in a forging die. Subsequent forming under tensile stress to obtain the fmal geometry, i.e. further forming under such high-tensile stresses that the whole cross-section undergoes plastic deformation. Superposition of tensile stresses during bending. In determining the blanks for bent products, empirical formulas have to be relied on at present, as precise determination of the geometry of the bent component is not yet possible [33]. If the bending radius is less than a
..!!
BendiAg Methods [32] Free BendiAg and FoldiAg. Technologically important methods are free bending with a three-point support or free bending of a sheet clamped at one side with a beam applied to the projecting part. If this beam turns through an angle, the method is called fOlding. BendiAg in a Vee-die (V·bendiAg). Here, two suboperations occur in succession [35]. Fitst, free bending takes place until the sides of the product being bent are in contact with the walls of the die (" = "G, Fig. 28a) or until r, < r" (r, = inside radius of component being bent, r" = radius of die). This is directly followed by further pressing in a die. The shape of the bent product is essentially altered to the shape of the tool. If r" is small, the workpiece is overbent until the bending legs lie against the die (Fig. 2Sc). If the load is removed in this position, the bending angle" may still be larger than the die angle "G' The inside radius steadily becomes smaller during further pressing. The same phenomena occur with regard to the bending angle applicable to further pressing both if the die radius rs. (or r,,/so) is large and if it is small. The precision of the bent components can be improved during further pressing, but this requites large forces (Fig. 29). Example.
Folding to produce a shape (Fig. :JO).
Channel or U·Bending. This denotes the sitnultaneous bending of two sides joined by a web, generally through 90°, in a die to produce a U-shaped component (Fig. ~l). A distinction is made between U-bending with and without a pressure pad. In U-bending without a pressure pad, the bellying of the web can be largely cured by further pressing. This leads to an increase in force at the end of the operation [34]. Influencing factors are the curvature of the die, its depth and the tool gap. In bending with a pressure pad (where the force of the pressure pad is approximately one-third of the bending force), the web remains flat during fOrming.
~:V:V
0.95
;
" 0.90
"".g
\V~:W-
~ 0.85 l;l
.0
.~ 0.80
-y
0.
UJ
0.15 0.10 0.65
material-specific minimum radius r, mim cracks occur in the outer fibre. Data on minitnum bending radii for steel sheet are given in DIN 6935 in relation to the material, sheet thickness, and bending axis position in relation to the roIling direction of the sheet.
g
1
1.6 2.5
16
25
40
63
100
Figure 7.7. Spring~back ratio K of various materials in relation to the bending radius [5); 1 Al99.5w, 2 St 1404, 3 51 1203, 4 CuZn33w.
Figure :a8. Bending in a 90° vee-
3.5 Technology. 3.5.4 Superplastic Forming of Sheet
Z.1
50
I
1.8
.~
u..
1.4 _.
1, 2-- r--'
g d:
c
~
_.
61'
0, 4t-
co
V-
~~,/ ' 1 ' : """,,/ / L /Sp ,)
I
\, / '
!
I
10
/' t-"'
o i1f' 0' 15'
0, 2 12 16 20 Travel of punch in mm Figure 29. Force-travel curve for bending in a 90° vee-die for a small and a large punch radius r s{ [5]; 1 r St = 2 mm, 2 r s{ = 15 mm (die width w = 42 mm, So = 2 mm, material St 14(4).
L/
resY
,
I
'g 20 Q)
.' M~ ~
I
.$2
2
--
I
30
C
0,8 O.
3i 0>
~ 1.0
.$2
So ~ 1,5mm b ~ 1000 mm Ws ~ 30,0 mm
40
1.6 - - ,--
3i
I
Material: Sf 1303
2.0
':.11
// /V
I
45' 60' 30' Angle of rotalion a,
,,- ,/'
V !
I 75'
90'
Figure 32. Bending force as a function of the angle of rotation as for different positions of the folding beam [361.
angle of rotation fr,. At small angles of rotation, the force required is low and increases only slowly, owing to a large effective lever arm. This increase is a consequence of the hardening of the material. ~.S.4
Figure '0. Steps in making a sheet-metal shape by folding.
Superplastic Forming of Sheet
Special metallic materials (chiefly eutectics and eutectoids) with an extremely fme-grained structure can withstand extremely large degrees of deformation (by over 1000%) under the following conditions: Tu> 0.5T, (T, = melting pOint, Tu = forming temperature in K) and low rates of deformation ip (usually < 10 - 2 1/ s). The material properties required for technically useful superplastic materials are: an extremely fme-grained structure (grain size < 10 fLm); high resistance to void formation by avoiding coarse inclusions, especially at the grain boundaries; low flow stress values k, at high rates of deformation ip (cf. K3.2.3). 5uperplastic forming is chiefly accomplished by means of blowing, The tool may take the form of either a positive mould (male mould method, Fig. ~~) or a negative mould (female mould method, Fig. ~4). The ratio of the drawing depth to the smallest plane dimension is limited to h/b < 0.4 for the female mould method and h/b < 0.6 for the male mould method [37]. The normal sheet thicknesses range from 0.5 to 3 nun. In the superplastic state, the flow stresses for e.g. aluminium and titanium alloys range from 4 to 20 N/nun', i.e. forming pressures ofO,S to 200 bar are required. The formed workpieces are practically free from intrinsic stresses and are thus free from spring-back. Approximate values for AI parts are given in [37], and for materials and parameters to be met in [38 to 42].
Figure 31. U-Bending [331: a without a pressure pad, b with a pressure pad.
Folding. Here, one side of the component being bent is clamped tightly and the other side is bent with a folding beam (Fig. ~2). If the smallest bending radius is larger than the radius of curvature of the clamping beam, free bending takes place. The pattern of the bending force exhibits two clearly defined zones, depending on the
a Figure ~3J. Male mould process. a Sheet metal blank; 1 in place. b Workpiece; 4 moulded; 2 moulding tool, 3 pressure space.
I :W.
Manufacturing Processes. 3 Metal Fonning
ling. For free upsetting (cold), this is s:5 2.3. Larger values require several stages. Figure 36 shows the manufacture of a cap bolt in several stages by tapering, initial upsetting and final upsetting. The force-displacement curve for upsetting rises steeply towards the end of the operation, which has a particularly large effect with smaller cap heights. Good filling of the mould and low burr formation are requirements for hot upsetting in a die. A special fonn of cold upsetting is embossing, which is perfonned as smooth embossing ( surface quality) or as sizing (thickness tolerance). Recourse is made to hot upsetUng only with difficult components, to keep the fonning forces small. Owing to the given force-displacement curve, presses with a fixed way of travel are particularly suitable for upsetting.
b
3.S.6 Forging c Figure 34. Female mould process. a Blank 1 in place. b Fonning the blank into a bubble by gas pressure 3. c Drawing over the mould 2. Showing 5 intermediate stages of fonning. d Final shape of work-
The basic forging processes belong to the manufacturing processes cutting, metal fonning, and joining. There are processes for changing cross-sections (straightening,
broadening, solid or cup extrusion, upsetUng, heading),
piece 4.
for changing direction (bending, folding, twisting), for producing hollow spaces (opening out, perforating,
3.S.S Upsetting
opening out with a hollow mandrel, indirect impact extrusion of hollow items), for cutting (parnng, deflasbing, piercing, preSSing-knife cutUng, chiselling, slitting) and for joining (shrinking, welding) if elements of com-
Upsetting is a basic method of forging and cold massive fonning, e.g. of fasteners (DIN 8583). It is important in theoretical studies as a model process. The limits to upsetting are:
plex forgings have to be united into the complete workpiece. Drop forging processes are heading, closed die
The degree of upsetUng (logarithmic defonnation) as the limit of fonnability: 'P = \n(ldlo) (Fig. 3S). In cold upsetting of steel, 'Pm= = 1.6 should not be exceeded, regardless of the number of upsetting stages. The upsetUng ratio s = lo/do as the limit to prevent buck5
·4.0 _ 2 slages limit: 91,' -1.6
,TJtL !
a
I
'15,-0,8 j
.~ -0.6
!
91,·-1.6 -1.4
JYL
:1:b
1.§ -0.4 "g
c55--0.2
-0.8 -0.6 -0.4
!
8
-0.2 1
Upsetting in one stage (singleaction process)
2
3 4 s'/o/do Upsetting in two stages (doubleaction process)
5
6
Upsetting in several stages (multipIeaction process)
911·-1.6 limit of formability of material e,=f{s) Limit to elevated tool loading in minor upsetting operations 5 = 2.3 Buckling point in single-action process 5= 4,5 Buckling point in double-action process Figure ~S. Process limits for cold upsetting of steel Cq35.
cI
e
Figure 36. Operations in malcing a cap bolt (transverse feed press): a feeding of material and shearing off with a shearing blade in separate stages, b feeding of blank up to die, C insertion into die and initial upsetting, d final upsetting, e ejection; 1 stop, 2 shearing die,3 shearing blade, 4 die (reducing die), 5 initial upsetting tool, 6 ejection pin (ram), 7 ejector (ram), 8 final upsetting ram, 9 ejection pin (die), 10 ejector (die) (VDI - Guideline 3171).
3.5 Technology. 3.5.6 Forging
I:.'
~ ~ a
~
c
b
Figure37. Process of drop forging in the narrower sense lSI : a heading, b closed die forging, (; open forging; 1 clamping jaw, 2 heading die, 3 workpiece, 4 ram , 5 container, 6" ejector, 7 upper
' 16
~
block, 8 lower block.
.
forging and open forging (Fig.
~7). In hot forging, the blanks are heated to a temperature above the recrystallisation temperature (850 to 1250 °C for steel), so that no permanent increase in the hardness of the material occurs. The production of blanks for forging comprises selection of the mill products, cutting the mill products into sections by shearing, breaking, sawing or parting, followed where necessary by setting (dosed die forging to produce flat parallel ends) and heating the blank to forging temperature.
250
,
c
b
Figure 39. Sequence of operations for forging in a multiple cavity die [44 1. a die; I stretching impression, 2 rolling impression, 3 impact surface, 4 final impression, 5 preliminary forging impreSSion, 6 bending impression. b Section through rolling impression . c Sequence of operations; I initial form, 2 stretched workpiece, .3 rolled workpiece, 4 bent workpiece, 5 workpiece after preliminary forging. 6 finished drop forging.
Hammer Forging. This process is generally used for one-off and small-scale production of parts with a mass ranging from I kg to 350 t. Typical operations are illustrated in Fig. ~8. Workpieces produced by hammer forging usually have to be fntished by machining. Drop Forging. Here, the blank is formed into the fntished workpiece via several intermediate shapes [43 J. The sequence of operations consists of mass distribution, preshaping of the cross-section (often by hammer forging) and closed die forging (Fig. ~9), which consists of the basic operations upsetting, spreading and rising (Fig. 40). The blank and the intermediate shapes should be matched to the tinished part in such a way that the best possible fibre pattem is achieved (Fig. 41). In preforming, the flash, which severely affects the process, is removed in the last stage of work by deflashing. Close-tol-
Figure 40. Basic types of operations for filling die impressions IS): a upsetting, b widening, c rising; I upsetting, 2 applying pressure,
3 filling.
erance forging enables the production of forgings of high
dimensional accuracy (lT9 to IT 11 , compared with !Tl2 to !TI6) and better surtace quality by perfonning at least one operation in a closed die and/or by means of forming at warm temperatures (600 to 900 °C for steel). Precision forging (e.g. under shielding gas with precise temperature control) produces ready-to-fit workpieces of even higher accuracy for selected machine parts (e.g. turbine blades, bevel wheels). r----~~=
I
11
)
~
~
c=Oi)
4
Figure 38. Application of basic processes for hammer forging of steel in sequence (51 .
c Figure 41. Selection of blank , process and pattern of fibres for forged parts {5]: a stretching, b stretching and upsetting, c upsetting.
Important types of forging die are, in the case of open dies, the full die as a single- and multiple-impression die, the insertion die as a single die and a die with several identical impressions, and the multiple-cavity die. Closed dies are dies without a flash gap with one or - in the case of multiple cavity dies - several parting lines. The high thermal and mechanical stresses (heating to 700 DC, stresses up to 1000 N/mm') mean that tool life is limited. Conventional steels for drop forgings are low-alloy hot
':.11
Manufacturing Processes. 3 Metal Forming
forming tool steels, e.g. 55NiCrMoV6, 56NiCrMoV7, 57NiCrMoV77, for full dies and high-alloy hot forming tool steels, e.g. X38CrMoV51, X37CrMoW51, X32CrMoV33, for die inserts. (ct. D3.1.4).
Cold Forging. This essentially comprises the processes of extrusion, also upsetting, embossing (q.v.) and - for smaller components made from steel and non-ferrous metals - also open forging and closed die forging. Here, forming takes place at room temperature without preheating. ~.s. 7
Extrusion
A survey of the extrusion processes is given in Fig. 42. A distinction is made between cold extrusion and hot extrusion. Cold extrusion denotes the pressing of ingots which are placed in the press without preheating. Hot extrusion, generally (as tWs is the norm) referred to simply as extrusion, denotes the pressing of ingots wWch are preheated before entering the press.
Direct Extrusion The ingot is first upset in the container until it assumes the diameter of the container bore. It is then forced through the die by the ram. Relative motion arises between the ingot and the container, so work has to be done to overcome friction.
With Lubricant (Fig. 42a). Friction is reduced by a film of lubricant between the ingot and the container, which is facilitated by conical dies. Uses are in hot extrusion of steel and cold extrusion of aluminium alloys.
Without Lubricant or Skull (Fig. 42b). This requires Wgher compression owing to greater friction between the ingot and the container and between the die and the ingot. Therefore tWs process is generally used for hot extrusion. The clearance between the dummy block and the container is such that no "skull" (the outer zone of ingot sheared off during extrusion) can form. Owing to the friction between the ingot and the wall of the container, depending on the amount of friction and the thermal conditions the outer zones of the ingot are obstructed to such an extent during movement of the ingot in the container that the core of the ingot flows ahead to a greater or lesser extent. Tills makes it possible to prevent the outer zones of the ingot from flowing into the forming zone preceding the die. This effect is exploited in extrusion of light metals. For instance, ingots with a continuous-cast surface are extruded up to certain ingot lengths in such away, depending on the cross- sectional shape and the ratio of the container cross-section to the product cross-section, that the outer zones of the ingot do not become part of the extruded product but remain in the discard. Without Lubrication and with Skull (Fig. 42c). If it is desired to ensure that no contaminated or oxidised outer zones of the ingot become incorporated in the extruded product, extrusion with a skull is carried out. In this method, the clearance between the container bore and the pressure disc is such that the outer zones of the ingot adhere to the container as a skull of metal. Thus only the inner part of the ingot is extruded. A disadvantage is the need to clear out the skull. Indirect Extrusion In indirect extrusion, too, the ingot is first upset in the container [45]. Here, a short closing ram closes the container on one side and the die, which is supported against a fixed hollow ram, enters the container from the other
side. The ingot and the container move together during extrusion, so no relative motion and thus no friction occurs between the ingot and the container. A disadvantage is the hollow ram, the internal bore of wWch limits the circumscribed circle of the extruded product.
With Lubricant (Fig. 42d). Friction between the die and the material being formed and between the die and the container is reduced, provided that the ingot is lubricated before processing and conical dies are used [46, 47].
d
3
9
8 10 7
~~Ii_~
9
3 10 8 7
Without Lubricant or Skull (Fig. 42e). In this process there is no friction between the ingot and the container in indirect extrusion. It is therefore suitable as a substitute for direct extrusion in cases where, in direct extrusion, one or both of the following occurrences would take place: 1.
/N~~rnt-4 ""'~'It;~~~'l-9
g Figure 41:. Schematic diagram of extrusion processes. a to c Direct extrusion: a with lubricant, b without lubricant or skull, c without lubricant, with skull. d to f Indirect extrusion: d with lubricant, e without lubricant or skull, f without lubricant, with skull, g hydrostatic pressing, h hydraftlm process. 1 Ram, 2 pressure disc,.3 container, 4 die, 5 hollow ram, 6 skull, 7 lubricant, 8 hydrostatic medium, 9 gasket, 10 ingot, 11 extrusion. Processes a, d., g and h are mainly for cold extrusion; processes h, c, e and fare mainly for hot extrusion.
2.
The force required to overcome the friction between ingot and container causes the total extrusion force to rise so rapidly that the extrusion force or the total extrusion force, in terms of the cross-sectional area of the container, restricts the use of tWs extrusion method too severely. The amount of heat resulting from the work done due to friction greatly lowers the rate of extrusion and/or the product quality.
In the skull-less process the clearance between the die and the container is set in such a way that the forces required to overcome the friction between die and container can be kept negligible compared with the forming force, and no skull (a tWn film of extruded material cover-
4.2 Machining with Geometrically Well-defmed Tool Edges. 4.2.1 Fundamentals
ing the walls of the container) can form between the die and the container. Because in this process the outer zones of the ingot become part of the extruded product, since they are not retained by friction on the ingot surface as in direct extrusion, either turned ingots should be used, or the ingots should have continuous-cast surfaces of adequate qUality. Without Lubricant and with Skull (Fig. 42f). The advantage of this process is that ingots with contaminated and oxidised outer zones can also be extruded, as the outer zones remain behind in the skull. The clearance between the die and the container is made so large that the outer zones of the ingot adhere to the container walls as a skull [48]. A disadvantage, again, is that the skull has to be cleared out.
Hydrostatic Extrusion In this process, the ingot is surrounded in the container by a compression medium (hydrostatic medium) (Fig. 42g). As the ram advances and the hydrostatic medium is compressed, the ram does not touch the ingot. Therefore the speed at which the ingot moves towards the die during extrusion is not equal to the speed of the ram, but is proportional to the displaced volume of hydrostatic medium [49]. Other features of the process are [4952]: low liquid friction at the surface of the ingot, sealing
1:-
between the ingot and the conical die by extrusion pressure, separate lubrication of the ingot is unnecessary if the hydrostatic medium has lubricating qualities, otherwise the ingot must be lubricated before use [53]. This process is used chiefly for cold extrusion. In hot extrusion, there is the problem of high thermal stress on all the components, and temperature control is also necessary. The Hydrafilm Process This process is also known as the "thick-film" process. The quantity of hydrostatic medium is kept so low that the ram can touch the ingot during extrusion (54, 55]. The die abuts against the container, as ingot and container are separated only by a film of liquid [52] (Fig. 42h). Furthermore, ingot speed and ram speed are practically the same, so extrusion can be interrupted at any time by stopping the motion of the ram. In this process, too, the ingot may be coated with a separate lubricant [56]. The process is mainly used for cold extrusion, but may also be used to a limited extent for hot extrusion, as additional equipment for the hydrostatic medium is unnecessary if use is made of hydrostatic media which can be applied to the ingot at normal pressure in the solid condition before placing the ingot in the extruder. The hydrostatic medium becomes viscous under extrusion pressure [56].
Cutting H.K. Tonshoff, Hanover
4.2 Machining with Geometrically Well· defined Tool Edges 4.2.1 Fundamentals
4.1 General Cutting is manufacturing by changing the shape of a solid object. The cohesion of the material is locally destroyed.
The final shape is contained in the original shape. Disassembly of assembled (joined) objects is considered to be part of cutting (according to DIN 8580). The main group "cutting" can be divided into seven groups:
Machining is manufacturing by cutting. Particles of material are mechanically removed as chips from a blank or workpiece by the cutting action of a tool. In machining with geometrically well-defmed tool edges, the number of cutting edges, the shape of the cutting tips and the position of the cutting edges in relation to the workpiece
are known and describable (in contrast to machining with geometrically undefined tool edges, e.g. grinding). Figure 1 illustrates important processes in this group. The processes are distinguished according to the cutting motion (cutting speed Vel, the feed motion (rate of feed v,) and the resulting effective motion (effective speed
Severing (DIN 8588) Machining with geometrically well-defmed tool edges (DIN 8589 Part 0) Machining with geometrically undefined tool edges (DIN 8589 Part 0) Chipless machining (DIN 8590) Disassembly (DIN 8591) Cleaning and Evacuation (DIN 8592)
The feed and cutting direction vectors fix the working plane. The angle between the two vectors is termed the feed direction angle 'P, while the angle between the effective direction and the cutting direction is termed the effective direction angle TJ. The relationship
Cutting by severing and machining is accomplished with the mechanical action of a tool on a workpiece. In chipless cutting, material particles are removed from a solid object by non-mechanical means. In cutting by cleaning, undesirable substances or particles are removed from the surface of a workpiece. Cutting by evacuation means the removal of gases from enclosed spaces; it is generally employed in connection with another manufacturing process such as electron beam welding or coating by ion plating.
applies to all processes. The mechanical operation of cutting particles off the workpiece, i.e. chip formation, can best be described with reference to the orthogonal process (two-dimensional deformation). The wedge is described by the rake angle y, the clearance angle a and the edge radius r H . The penetration of the wedge causes the material to undergo plastic deformation. Figure 2 shows the zones of plastic deformation by way of example during flowing chip formation. Five zones can be distinguished:
Vel·
sin'P
tanTJ
=
(vjv,) + cos'P
I:W"
Manufacturing Processes. 4 Cutting
Cutting with geometrically well-defined cutting edges
Milling
Turning
9~workPiece o
.
c-I~
TOOI~
~ece
Tool
Other processes • Planing, shaping • Broaching • Sawing
Figure 1. Processes of machining with geometrically weU-defmcd
tool edges.
The primary shearing zone comprises the actual area of chip formation by shearing. In the secondary shearing zones ahead of the face and at the flank, forces of friction act between the to.ol and the workpiece, causing plastic deformation of these layers of material. In the deformation advance zone, chip formation causes stresses to act which lead to plastic and elastic deformation of this zone. In the pressure and cutting zone, the material is deformed and cut under high compressive stresses. Because of these mechanisms, the stress thickness h in the undeformed condition changes to the chip thickness h', resulting in the chip compression Ah = h'lh. The plane of shear encloses the angle of shear 'P with the cutting speed vector. The angle of deformation X denotes the shearing of a particle which has passed through the plane of shear. Besides flowing chip formation, other kinds of chip may form: With flowing chips there is a steady flow of material from the workpiece. The deformation of the material is continuous_ Periodic changes may occur in the intensity of the deformation - usually at higher cutting speeds. Lamellae form in the chip, which may be pronounced up to cutting of the material and formation of chips [1). Shearing chiP formation occurs when the formability of the material is exceeded in the shearing zone and locally concentrated shearing takes place without complete cutting of the material. Chip formation is uneven. Tearing chip formation occurs in materials with low formability, e.g. lamellar graphite cast iron. The interface between the chip and the workpiece is irregular. Built-up edges may form when machining ductile, workhardening materials at low cutting speeds and sufficiently steady chip formation (flowing chip formation). They are parts of the material that have been severely deformed and work-hardened in the area of the compression zone and have become welded to the curvature of the cutting edge and the face under high pressure, thus becoming part of the cutting tip [2). In the chip formation zone the cutting energy supplied is completely converted. It is calculated as
(Fc' = cutting force, " = travel in direction of cut).
The cutting energy is made up of forming and shearing energy E, friction energy at the rake face E y , friction energy at the flank E", surface energy to form new surfaces ET , and kinetic energy due to chip deflection EM' The energy converted in machining one unit of volume is
(e, = specific energy, V~ = volume of metal machined)_ Like En the individual components of Ee can be expressed in relation to VW'
Workpiece
ton x=tan(-yi + 1!tan tan = cos y/(Ah-sin yi ).,h
=h'lh
Figure 2. Effective zones in chip fonnation and derived model of
deformation in the shear plane; 1 primary shearing zone, 2 secondary shearing zone at the face, 3 secondary shearing zone at the pressure and cutting zone, 4 secondary shearing zone at the flank 5 deformation advance zone.
Example. A numerical estimate shows that most of the cutting energy is converted into forming and friction energy. For ec = 2760 N mm -z, for a sample calculation where chip thickness h = 0.1 mm, cutting speed Vc = 60m min-I, rake angle 'Y = +10°, chip compression Ah = 3.9, the energy components are as follows: specific fanning and shearing energy e
From the specific energy ec introduced, the specific cut-
4.2 Machining with Geometrically Well-defined Tool Edges. 4.2.1 Fundamentals
':.*'
ting force kc can be derived as an index for calculating the cutting force:
where A = the undeformed chip cross-section A, b = the undeformed chip width and h = the undeformed chip thickness;
where Pc = the cutting performance, Qw = the chip volume over time, and Fc = the cutting force. Thus the specific cutting force kc can be seen as an energy-related variable. (The application and determination of kc are discussed in detail in K4.2.2.) The energy introduced into the chip formation zone is almost entirely converted into heat, with a small remnant being turned into intrinsic stress in the chip and the workpiece (spring energy). This produces high temperatures in the cutting tip, which is thus subjected to mechanical and thermal stress. The surface forces and the temperature distribution are shown in Fig. 3 [3]. From this, stresses can be calculated. In Fig. 4 only the direct tensile stresses which are especially critical for high-temperature ceramic cutting materials are plotted. Mechanical and thermal stress, assisted by chemical reactions, cause wear. The stress on the cutting tip depends on various influences. Besides control parameters like cutting speed, feed and depth of cut as well as environmental influences such as the cooling lubricant, the workpiece in particular influences tool wear.
Cutting edge diSlance y in !ml
a
0.90 0.90
Kinds of wear [4], Fig. 5 (cf. D54). Fractures and cracks; these occur in the area of the cutting edge due to excessive mechanical or thermal stress. Mechanical abrasion, mainly caused by hard inclusions such as carbides and oxides in the material. b
OJIting edge distance y in mm
0.90 0.90
Figure 4. Calculated principal tensile stress distributions in Al,p" ceramic indexable inserts, under a mechanical and b thermal stress [31; 1 flank, 2 face, 3 calculating plane.
Stress with
a 2
0,2 0.4
E E
0.6. so 1>1~
- 0.8
I>~" S~S
0.2
0.4
0,6
yin mm
0.8
Short-term effect
Constant mechanical stress
Alternating mechanical stress
Constant thermal stress
Alternating thermal stress
Internal chemical attack
Surface chemical attack
Types of wear Abrasion Adhesion Fracture Flaking Cracking Diffusion Oxidation
Figure S. Typical kinds of wear and their causes (d. D6).
1.0
I>l~
b
"
Long-term effect
S1~
1.0
1.2 1.2
Figure 31. a Assumed force and b temperature distribution in a cutting tip; 1 primary flank 2 face, 3 surface loads, 4 calculating plane.
Plastic deformation occurs if the cutting material has inadequate resistance to deformation but sufficient tough-
ness. Adhesion is the shearing off of pressure-welded points between the material and the chip, the point of shearing being located in the cutting material. Diffusion occurs at high cutting speeds and mutual solu-
';Wf:1
Manufacturing Processes. 4 Cutting
bility of the cutting and workpiece materials. The tool material is weakened by chemical reaction, becomes detached and is removed. Oxidation also occurs only at high cutting speeds. The cutting material oxidises in contact with atmospheric oxygen and the structure is weakened,
f
8
The macbinability of a workpiece is determined by the composition of the material, its structure in the machined area, the preceding metal forming/primary shaping operation, and the heat treatment it has undergone. Machinability is evaluated according to the following criteria: Tool wear Surface quality of the workpiece Machining forces Chip form
Feed dilection -
The nature of the machining job should be taken into account in weighting the criteria.
4.2.2 Turning According to DIN 8589E, turning is defined as machining with a continuous (usually circular) cutting motion and any desired feed motion in a plane at right angles to the cutting direction. The axis of rotation of the cutting motion maintains its position relative to the workpiece regardless of the feed motion. Figure 6 illustrates some important turning processes. In the following, longitudinal cylindrical turning is taken as an example of a turning process. Terminology, names and designations for describing the geometry at the cutting tip are laid down in DIN 6580 and ISO 3002/1. Figure 7 shows the surfaces and cutting edges defined for the cutting tip. The angles shown in Fig. 8 serve to determine the pos-
Figure 7. Terminology at cutting tip and directions of motion of tool (DIN 6580, ISO 3002/1); I indexable insert, 2 secondary flank, 3 secondary cutting edge, 4 nose, 5 primary flank, 6 primary cutting edge, 7 face, 8 tool holder, 9 workpiece, 10 working plane, v~ cutting speed, v" effective speed, Vr rate of feed, cp feed direction angle, 11 effective direction angle.
_
8
'"~~T~:~,'OOl·'~9 ~OT' ~~ .~, •
~ WP
or
p
e
.f: ~
rF
LW
w b
~w
'
'r :'
Feed direction
6
8
c
Figure 8. Turnmg tool angles (DIN 6581) a main view, b crosssection A-B (orthogonal plane of tool), C View Z (onto cuttlOg edge plane); 1 flank, 2 face, 3 tool cutting edge plane, 4 tool reference plane, 5 point on cutting edge under consideration, 6 assumed
working plane, 7 tool wedge measuring plane, 8 cutting edge plane of primary cutting edge, 9 cutting tip.
angle
c
K is the angle between the primary cutting edge and the working plane, The edge angle e is the angle between the primary and secondary cutting edges and is predetermined by the cutting edge geometry. The angle of inclination A is the angle between the cutting edge and the reference plane and is apparent when looking down onto the primary cutting edge. The clearance angle a, wedge angle {3 and rake angle l' are the angles measured in the wedge measuring plane and total 90°. The
d
WP
&¢~ t
e
I
T1
Feed directioo
values of the relevant tool angles are determined from
f
WP
f\
Direction of rotation
Figure 6. Turning processes (DIN 8589 Part 1): a facing, b parting, c tuming, d thread turning, e profile turning (workpiece contour is duplicated in tool), f fonn turning; WP workpiece, T tool.
approximate value tables in relation to the workpiece and cutting materials and the machining process. Table 1 shows some values for machining of steel. The entry angle K influences the shape of the undeformed chip cross-section to be removed and thus the power required for the machining process (Fig. 9). The chips flowing over the face of the tool have a different bulk volume depending on chip type and form. The
I:."
4.2 Machining with Geometrically Well-defmed Tool Edges. 4.2.2 TUrning
Table 1. Normal values for tool angles in machining of steel Cutting material
Cutting tip geometry Rake angle
Clearance angle
Angle of inclination
~
a
"
_.6 0 to +200 __ 6° to +15° -6° to 0°
6° to 8° 6° to 8° 6° to 8°
High-speed steel Cemented carbide Ceramic insert
Entering angle
Edge angle
10° to 100 0
Edge radius
0.4 O.S
45° to 100°
to 2mm to 2 nuu
~ I'I, IU ;090 b ~ Op /sin )/
•
~
parameter is the cbiP space coefficient RZ, the ratio of the time-related chip volume Qw to the bulk volume Q'. Here, =
~ III '" 25 e
Qw = ap/ve = apjD7rn.
Fc = kea p/ = kcbb.
,,50
.. so
b
c:
d
tJ'
f(JlII
Q'/Qw,
The chip space coefficient indicates the "bulkiness" of the chips. It is used to dimension machine tool working spaces, chip conveying equipment, and chip spaces of tools. The chip space coefficient RZ may have widely differing values depending on the shape of the chip (Fig. 10). The more brittle the material, the lower is this value. Brittleness can be influenced via the composition of the material. For steel, higher sulphur contents (mort than 0.04%, free-cutting steel with 0.2% S) have a ben eficial effect. However, this may impair the toughness a the material in the transverse direction, depending on th. form of the dispersed sulphides [5]. Chip-breaking step' sintered onto the face of the tool or attached chip break ers produce additional chip deformation, i.e. additiona stress on the chip material, and deflect the chip agains an obstacle in the direection of flow (Fig. 11). The chi! is bent by contact with the cut surface of the workpiec. or the flank of the tool and breaks (secondary chip break ing in contrast to tearing chip formation or lamellar chi! formation with material cutting (cf. K4.2.1 ), where th. chips leave the chip formation zone as small fragments) Favourable chip forms can also be achieved by selectinl the appropriate machine setting data such as rdte of feet and depth of cut (Fig. 12). Every material resists the penetration of the tool durinl chip removal. This has to be overcome by means of , force, the machining force F. This force is analysed b~ resolving it into its three components (Fig. 13):
.. 90
• •• •
h ~ f sin )/
Figure 9. Cut and chip variables in turning; 1 tool, 2 workpiece.
RZ
fu
",8 f
~
&:ir
~
t]}
11
..8
.. 3
g
b
Figure 10. Chip shapes (iron and steel testing sheet I 178-{i9): a ribbon chips; b snarl chips; c flat helical chips; d long, cylindrical helical chips; c helical chip fragments; f spiral chips; &" spiral chip fragments; h discontinuous chips.
-
Cultir>gspeed
a
Figure 11. Effect of chip shape stages: a contact with cut surface. b contact with flank.
.:*,.•
Manufacturing Processes • 4 Cutting
Material Culting speed Depth of cut Cutting material
E
~c ~
~'"
g'
:-a
2.5 2,0
20 MnCr 5SG 100 m/min
VC ;
op; 3 mm HM PI0
~
I '" I~ mc;0,24 "I 1,5 k e1l ; 15 . kN/mml
1 -t=-s.:1
'"
u
J:? 1.25 'u 2i
~,
if)
1,0 0.16
Depth of cut ap in mm Figure 12. Areas of favourable chip shape with tools having chipbreaking grooves (according to Konig).
v,
'!-----"fc F Figure 1~. Components of machining force (DIN 6584); 1 working plane.
0,63 1,0 1,6 0,25 0,4 Undeformed chip thickness h in mm
Figure 14. Specific cutting force as a function of undeformed chip thickness.
tal values m, may vary widely, It follows from me < 1 that for a given machining cross-section, the cutting force and power requirements increase at smaller undeformed chip thicknesses, The physical reason lies in the higher proportion of friction applicable to smaller undeformed chip thicknesses (cl". K4,2,)), ke depends on other variables apart from the material and the undeformed chip thickness, Additional influencing factors are therefore applied, The influencing factors for the cutting speed Kv, the rake angle K y, the cutting material K w " the cutting edge sharpness K~, the cooling lubricant K k , and the workpiece shape K, are also shown in Appendix K4, Table 1, The passive force l'~ as a further component of the machining force (Fig. 13) produces no work, as it is perpendicular to the working plane, and motion occurs only in this plane, However, it is important for the dimensional accuracy and accuracy of shape of the machine-workpiece-tool system, The third component is the feed force F,. The passive force Fp and the feed force F, can be combined into the resultant cutting force F D , For thin, undeformed chip cross-sections (b :P h), the resultant cutting force is perpendicular to the primary cutting edge, From this it follows that FdFp = tan
It is known from tests that the specific cutting force ke is also a function of the undeformed chip thickness h. It can be seen from the log-log representation (Fig. 14) that [6J
Here, ke!.l is the "primary value of the specific cutting force", i.e. ke at h = 1 mm (indices 1.1 due to kell =FJ1.1 at b= Imm and h= Imm). The exponent
m(:
indicates the increase and is the "incremen-
tal value of the specific cutting force". Kienzle's cutting force formula can also be written
kc 1.1 and 1 - me are listed for various ferrous materials in Appendix K4, Table 1. A direct comparison of the k, 1.1 values to indicate the machinability or the energy required for machining is unacceptable, as the incremen-
2,5
K.
For normal values of band h it can roughiy be assumed that FD = (0,65 - 0,75) Fc>
by means of which F, and Fp are to be determined, More exact determination is achieved by means of exponential functions corresponding to the cutting force formula, The exponents and principal values are shown in Appendix K4, Table 1, The surface fmish is determined by the proftle of the cutting edge which produces the workpiece surface and by the feed, From the shaping of the cutting edge comer radius r" the theoretical surface roughness R" 'h can be geometrically determined as R" 'h = f'l (8r0)' This value should be regarded as a minimum, which increases due to vibrations, especially at higher rotational and cutting speeds, on the formation of built-up edges (cf, K4,2,1) and with the progreSSive wear of the cutting edge,
4.2 Machining with Geometrically Well-defined Tool Edges. 4.2.3 Drilling
':.11
Unit costs K, KM, Kwz
J
2
Vc.opt
Cutting speed Vc
Figure 16. Manufacturing costs as a function of cutting speed vc; 1 unit costs K, 2 machine-specific unit costs K M , 3 tool-specific unit costs Kwz .
Cross-section IrB Figure IS. Forms of wear in turning (ISO 3685): 1 flank wear, 2 notching, 3 cratering; C, B, N zones, KB crater width, KM distance from crater centre to point of cutting tip, KT crater depth.
The tool is subject to mechanical stress due to the machining force. thermal stress due to heating and chemical attack due to the interaction of the cutting material, the workpiece material and the surrounding medium. This results in wear on the cutting tool (cf. K4.2.1). Typical forms of wear are illustrated in Fig. 15. In addition, cutting edge wear, rounding of the cutting edge and scoring may occur on the secondary cutting edge. The type of wear that determines the end of tool life (tool life criterion) is determined by the respective use. Weakening of the wedge due to crater wear, or an increase in the proportion of the machining force accounted for by friction due to flank wear, are critical in roughing. Cutting edge wear leads to changes in workpiece dimensions, and flank wear or scoring impair surface quality and determine the end of tool life in finishing. The end of tool life is often set at a width of wear land of 0.4 mm or a crater depth of 0.1 mm. The flank is divided into three zones for more precise identification of the wear. For a specific cutting-material-workpiece-material combination and a given tool life criterion, tool life depends
Here, To and Vc are reference values, To normally being set at To = 1 min. C is the cutting speed for a life of To = 1 min. The Taylor straight line is plotted on the basis of a wear/tool life turning test to ISO 3685. This lays down appropriate set values for high-speed steel, cemented carbides of all machining categories (cf. K4.2.6) and ceramic inserts. It is usually sufficient to determine the widtb of the wear land and/or the crater depth as well as the distance from the crater centre to the point of the cutting tip. Table 2 shows, for various materials, normal values of the gradient exponent k and the cutting speed C for a tool life of T = 1 min and a width of wear land of 0.4 mm. For metal cutting machines the optimum cutting speed has to be established according to commercial criteria (Fig. 16). The optimum cutting speed in relation to time is Vc
opt
=
C( -k-l) tU~·
Optimisation of the cutting speed to minimise unit costs takes into account not only the tool changing time twz but also the tool costs per cutting edge Kwz and the hourly machine rate K M:
chiefly on the cutting speed according to an exponential
function (Taylor's equation shown on a straight line on a log-log graph) [7]:
4.2.~
Drilling
Drilling is a metal cutting process with a rotary cutting motion (primary motion). The tool, the drill, performs a Table 2. Coefficients for determining Taylor straight lines Oxide ceramic (steel) Nitride ceramic (cast metal)
Taylor function Vc
=
c·
plk
Uncoated cemented carbide C
k
299 226 299 478 478 177 110 177
-3.85 -4.55 -3.85 -313 -3.13 -5.26 -7.69 -5.26
97
-6.25 -10.0
53
C
k
(m/min)
(m/min)
51 50-2 51 70-2 Ck 45N 16MnCrS 5 BG 20MnCr S BG 42CrMoS 4 V X155CrVMo 12 1G X40CrMo V , 1G GG-30 GG-40
Coated cemented carbide
385 306 385 588 588 234 163 234 184 102
C
k
(m/min)
-4.55 -5.26 -4.55 -3.57 -3.57 -6.25 -8.33 -6.25 -6.25 -10.0
1210 1040 1210 1780 1780 830 570 830 2120 1275
-2.27 -2.27 -2.27 -2.13 -2.13 -2.44 -2.63 --2.44 -2.50 -2.78
l;Wg
Manufacturing Processes a 4 Cutting
feed motion in the direction of the axis of rotation; Fig. 17 shows common drilling processes. In drilling into solid metal, either through holes or blind holes may be produced. The tool used is usually a twist drill. Enlarging a drilled hole is done with twist drills or with countersinks or counterbores having two or more cutting edges. Step drills produce stepped holes. They usually have multiple cutting edges; for manufacturing reasons, not every cutting edge has to support all parts of the contour (e.g. one cutting edge may break the edge of a step, while the adjacent one produces a flat surface). Centre drills are specialprofile drills with a thinner spigot and a short, stiff drill section in order to develop a good centring effect. Combination centre drills and countersinks cut a ring into the metal, a cylindrical centre hole being drilled at the same time. Taps are used to cut threads. Reaming is a holeenlarging process with a small undefonned chip thickness, for producing holes of precise size and shape with a high-quallty surface. For drilling holes with diameters of 1 to 20 mm, with drilling depths up to five times the diameter, the twist drill is the tool most commonly used (Fig. 18). The twist drill consists of the shank and the body. The shank is used to clamp the drill in the machine tool. It is straight or tapered. If high driving torques are to be transmitted, tangential flat surfaces transmit the force. The body has a complex geometry, which can be modified to adapt the drill to the respective machining duty. Essential parameters are the profile and the core thickness, the flute geometry and the helix angle, i.e. the pitch of the flutes, the land and the point angle. Of these, the land and the point angle can be influenced by the user. The profile of the twist drill is designed in such a way that the flutes provide the maximum possible space for chip removal while ensuring that the drill is able adequately to withstand torsional stress. These two main requirements may be accompanied by others, such as production of favourable chip shapes, which have led to a variety of special profiles and which enable the drilling process to be adapted to particular boundary conditions. Material also has to be removed ahead of the centre of the twist drill .
•
e
c
b
f'
6
I
~ '
7
3
'
8 3 2
5 4
13 17
18
Fiaure 18. Terminology and mode of operation of twist drill (DIN 6580,6581, 1412): n rotational speed, /) helix angle, d diameter, u cone angle, cp feed direction angle, 1J angle of effective direction of cut; 1 transverse cutting edge (angled part of primary cutting edge), 2 margin width b, 3 margin of secondary flank, 4 nose, 5 primary flank, 6 centre thickness K, 7 flute, 8 secondary flank, 9 land, 10 face, 11 secondary cutting edge, 12 primary cutting edge, 13 tool axis, 14 tool, 15 workpiece, 16 cutting motion, 17 effective motion, 18 feed motion.
This is achieved by the transverse cutting edge, which links the two primary cutting edges. Along the primary and secondary cutting edges, the rake angle ')', which is the most important variable influencing the drilling process, is not constant but decreases from the outside to the inside, even before the primary cutting edge is reached. In Fig. 19, the rake angles at three pOints on the cutting edge are shown by plotting the
•
Diagram not to scale
d
di ton (\ ) y. =orc ton ( - - - I D sin (0'/2)
g
Fiaure 17. Drilling processes (DIN 8589). a Drilling into solid metal; 1 twist drilL b Hole enlarging; 2 twist-type counterbore with
ton 6 =D1C/h
countersinking; 5 combined centre drill and countersink. fTapping;
Flaure 1!J. Rake angle at primary cutting edge of twist drill: b flute pitch, cr cone angle, /) helix angle, D drill diameter, d, diameter at relevant point on cutting edge t, YI rake angle at relevant point on
6 tap. g Reaming; 7 machine reamer.
cutting edge i.
three cutting edges. c Counterboring with multiple diameters;
3 step drill. d Centre drilling; 4 centre drill. e Centre drilling' and
4.2 Machining with Geometrically Well-defined Tool Edges. 4.2.4 Milling
pitch b of the flute over the development of the circles associated with the diameters [8). At the outside diameter it is identical to the helix angle 5 and decreases in direct proportion to the diameter. Negative rake angles may occur even before the primary cutting edge is reached. Preceding the transverse cutting edge, the rake angles are sharply negative. Here the workpiece material has to be displaced radially. Negative rake angles and the material displacement effect generate high pressures in the area of the transverse cutting edges. To alleviate this effect, twist drills are pointed. The centre of the drill is tapered by profile grinding in the direction of the flute and towards the drill tip on a conical or similar surface. In this way, the rake angle at the transverse cutting edge is increased and/or the transverse cutting edge is shortened. The most important type of wear on the twist drill is flank wear at the nose. This wear, which is mainly caused by abrasion, produces an increase in the torsional stress on the drill, as higher machining forces are present in the nose area. This torsional stress may cause the drill to break. Worn twist drills are therefore reground until the damaged area of the secondary cutting edge margin is removed.
Mlldtbdng Forces. The forces and moments present during drilling are calculated on the basis of Kienzle's approach [7, 9). Figure :10 illustrates the metal cutting geometry and the forces during drilling. The forces occurring at each cutting edge, which are assumed to act in the middle of the cutting edge, are resolved into their components Fe, Fp and Ft. The cutting forces Fel and Fc2 generate, via the lever arm D/4, the cutting moment
F,~Fcz P\
..
F. . cl
Fpz
012
Me = (Fel + Fe2)D/4,
I:• •
Fel = Fe2 = Fez,
Me = FezD/2. The feed forces Ffl and Fa are added together to give F,
F, = Ffl + Ff2 ,
F" = Ff2 = Fa,
F, = 2F,z·
The passive forces FpI and FP2 cancel each other out in the ideal case, i.e. with a symmetrical drill. If there are errors of symmetry, FpI and Fp2 generate interference forces which impair the quality of the hole. The cutting force per cutting edge works out as
Fez = bb"-mc) ke 1. I,
b =
f, sin K,
b = D/(2 sin K).
By analogy, the feed force is
Values are given in Appendix K4, Table :I. The feed forces are heavily dependent on the form of the transverse cutting edge. They can be greatly reduced by grinding the point. Wear causes them to rise to twice the original value or more. Surface quality in drilling with twist drills corresponds to roughing with R, = 10 to 20 f.Lm. The roughness can be reduced by reanting. Another possibility is to use drills made entirely of cemented carbide. When drilling solid metal, surface qualities, dimensional accuracy and accuracy of shape like those obtained with reanting are achieved.
Short-hole Drilling Short-hole drilling, with drilling depths of L < 2 . D, covers a large proportion of bolt hole drilling, through hole drilling and tapping. For this, short-hole drills with indexable inserts may be used for diameters from 16 to over 120 mm. Their advantage compared with twist drills is the absence of a transverse cutting edge and the increase in cutting speed and feed rate achieved with indexable cemented carbide or ceramic inserts. Owing to the asymmetrical machining forces, the use of short-hole drills requires rigid tool spindles such as are common on machining centres and milling machines. The higher rigidity of the tool enables pilot drilling of inclined or curved surfaces. Accuracies of m are achieved without further work [11).
4.:1.4 MilliAg •
012
Classification of MilliAg Processes In milling, the necessary relative motion between the tool and the workpiece is achieved by means of a circular cutting motion of the tool and a feed motion perpendicular to or at an angle to the axis of rotation of the tool. The cutting edge is not continually in engagement. It is therefore subject to alternating thermal and mechanical stresses. The complete machine-tool-workpiece system is dynamically stressed by the interrupted cutting action. Milling processes are classified according to DIN 8589 on the basis of the following:
b
d
FIpre ZO. Metal cutting geometry and machiniog forces in drill· ing: a forces, b drilling into solid metal, c hole enlarging; 1 tool, 2 workpiece.
The nature of the resulting workpiece surface. The kinematics of the metal cutting operation. The profile of the milling cutter. Milling can be used to produce a practically infinite variety of workpiece surfaces. A distinguishing feature of a process is the cutting edge (primary or secondary) that
':.11
Manufacturing Processes. 4 Cutting
cutter enters the workpiece at the maximum undeformed chip thickness, while in upcut milling the feed direction angle cp is < 90°, thus the cutting edge enters at the theoretical undeformed chip thickness h = O. This initially results in pinching and rubbing. A milling operation may include both upcut milling and downcut milling. The principal milling processes are summarised in Fig. 21~.
Plain Face Milling with End Milling Cutter The kinematics of metal cutting and the relationship of the metal cutting forces during milling will be discussed with reference to plain face milling with an end milling cutter. Further milling processes are described in [12].
Figure :1I1. Comparison of face milling and peripheral milling: a face milling - workpiece surface produced by secondary cutting edge, b peripheral milling - workpiece surface produced by primary cutting edge; I tool, 2 workpiece, 3 cutting edge.
produces the workpiece surface (Fig. 211): inface milling it is the secondary cutting edge located on the face of the milling cutter, while in peripheral milling it is the primary cutting edge located on the circumference of the
Kinematics of Metal Cutting. To describe the process, it is necessary to distinguish between the engagement variables and the metal cutting variables. The engagement variables, which are expressed in relation to the working plane, describe the interaction of the cutting edge and the workpiece. The working plane is described by the cutting speed vector Vc and the rate of feed vector vf • In milling the engagement variables are (Fig. 214): depth of cut a p , measured at right angles to the working plane; cutting engagement a c , measured in the working plane at right angles to the feed direction; and feed of the cutting edge J,., measured in the feed direction.
milling cutter. A distinction can be made on the basis of the feed direction angle cp, Fig. 2121: in downcut milling, the feed direction angle cp is > 90°, thus the cutting edge of the milling
a
b WP
a
Feed direction angle 90' < rp
~
c
d
e
f
180'
% 1234 N 5 X ... Z .. , F
b
Feed direclion angle 0' ~ rp < 90'
Effective direction angle 1/:
tan 11
sin rp
=-/-,--'-v, Vf +COsrp
Figure l:l:. Comparison of a downcut milling and b upcut milling (DIN 6580 E); I milling cutter, 2 working plane, 3 workpiece.
g -
Feed direction
C Direction of rotation
Figure :13. Milling processes (DIN 8589). Plain milling: a face milling, b peripheral milling, c combined face and peripheral milling, d thread milling, e hobbing, f profile milling, g form milling, WP workpiece, T tool.
4.2 Machining with Geometrically Well-defined Tool Edges. 4.2.4 Milling
The time-related chip volume is Q = aeapv,. The undefonned chip thickness is a function of the entry angle 'P and is thus not constant as in turning. The evaluation of the milling process is based on the mean undeformed chip thickness
-.,
hm
Jbeep) dip
=
(IN,·)
=
(IN,);; sin K(ens 'Pc - cos 'PA)'
Machining Force Components. The machining force required for chip formation has to be absorbed by the cutting edge and the workpiece. According to DIN 6584, the machining force F can be resolved into an active force F" which lies in the working plane, and a passive force F p , which is at right angles to the working plane. The direction of the active force F, changes with the entry angle 'P. The components of the active force can be expressed in relation to the following directions (Fig. 25): Direction of cutting speed Ve: the components cutting force Fe and perpendicular cutting force Fe~ relate to a co-rotating system of coordinates (tool-specific components of the active force). Direction of rate of feed v,: the components feed force F, and perpendicular feed force F'N relate to a fixed system of coordinates (workpiece-specific components of the
Delail Z Figure 24. Engagement variables in plain face milling with an end milling cutter; 1 exit plane, 2 entry plane. -3 tool cutting edge. 4 workpiece.
For a full description of the kinematics of metal cutting, the following data are required: milling cutter diameter D, number of teeth z, tool projection ii and the cutting edge geometry (side rake angle y" back rake angle yp' side clearance angle Ct: h back clear-ance angle Q'p' entry angle Kn angle of inclination As, cutting edge radius r, and land), Owing to the interrupted cut, the entry and exit con-
ditions of the cutting edge, and the rypes of contact, are especially important for the milling process. The rypes of contact describe the nature of the first and last contacts of the cutting edge with the workpiece. They can be determined from the entry and exit angles and the tool geometry. It is especially bad if the cutting edge tip is the first point of contact. From the engagement variables can be derived the metal cutting variables, which indicate the dimensions of the layer of metal to be removed from the workpiece. The metal cutting variables are not identical to the chip variables, which descrihe the dimensions of the resulting chips. The cutting edges describe cycloids in relation to the workpiece. As the cutting speed is significantly higher than the rate of feed, they can be approximated by circular paths. In this way of looking at the subject, the undefonned chip thickness is (Fig. 24): h( 'P) - ;;
sin K . sin 'P.
With the undefonned chip width b undeformed chip cross-section is
Tool-specific forces: F,;. FeN Workpiece-specific forces: F,. Fy
Angle of engagement
= ap/sin K,
the
~
Figure 2S. Components of machining force in plain face milling with an end milling cutter.
Manufacturing Processes. 4 Cutting
active force). For converting the active force from the ftxed system of coordinates into a co-rotating system, the
following apply:
+ FfN(cp) sin
FcCcp)
=
Ff(cp) cos cp
Fe,(cp)
=
Ff(cp) sin cp - FfN(cp) cos cp,
Fx(cp)
= F,(cp),
F,(cp)
=
cp,
FfN(cp).
This transformation is important if, for instance, the cutting force Fe is to be measured with a three-component force-measuring platform to which the workpiece is fixed. Figure 2S shows the pattern of the components of the active force in the tool- and workpiece-specific systems of coordinates for axial milling with an end milling cutter.
Relationship of Machining Forces. Kienzle's machining force equation [7] can also be applied to milling. For the machining force components cutting force Fc ' perpendicular cutting force FcN and passive force F p'
= Ak
Fi
i,
= c, cN, p,
where i
where A = the undeformed chip cross-section and k i the specific machining force. Owing to the wide range of undeformed chip thicknesses that is covered by milling (the undeformed chip thickness depends on cp), Kienzle's relationship applies only to certain areas. The undeformed chip thickness range of 0.001 mm < h < 1.0 mm is divided into three sections (Fig. 26) [13, 14]. For each section, a straight line can be detennined, which is established by the parameters: the main value of the specific machining force and the incremental value. For the specific machining force the following apply:
=k
i 1.0.01 •
k i = ki
h-mi 00>
1.0, I ' h-m,O.1
for 0.001 mm < h < 0.01 mm for 0.01 mm
0.1 mm
where i = c, cN, p. Thus the machining force for milling with an end milling cutter is
Ck 45 N HM P25 VC
= bk i
I.} •
hl -
mi ,
where i = c, eN, p.
The corresponding component of the machining force can be calculated for milling if the main value of the specific machining force component and the incremental value for the workpiecel cutting material pair and the cutting conditions are available. The machining indices for axial plane face milling with a milling head are given in Appendix K4, Table ~ for a number of workpiece materials and' cutting conditions [7, 13, 15]. Often, though, to estimate the machining force during milling, machining indices obtained from turning will have to be used. Milling machine capacity is designed on the basis of the average machining force
r;rn = bki
1. 1 •
h!"
m, Kvc/C!'VKMKw,"
where i = c, cN, p. In this equation, hrn is the mean undeformed chip thickness, K", = 1.2 to 1.4 is the correction factor for the manufacturing process (the factor takes into account the fact that the machining indices were obtained from turning tests), Ky is the correction factor for the rake angle (cf turning), Kv is the correction factor for the cutting speed (cf. turning), K~ is the correction factor for tool wear (cf turning) and Kw. is the correction factor for the cutting material (cf turning). Research into plain face milling shows that the influence of wear on the machining force components cannot be ignored [14].
Vibrations. Depending upon the elasticity frequency response of the complete milling machine-milling cutterworkpiece system, the metal cutting forces generate vibrations which may affect surface quality and tool life. According to the method of their generation, these vibrations are divided into separately excited and selfexcited vibrations [cf J2].
Separately Excited Vibrations. In the case of separate
for 0.1 mm < h < 1.0 mm
Workpiece material Cutting material Cutting conditions
Fi
=190 m/min
excitation, the complete system vibrates at the frequency
of the exciting forces. The intermittent cutting action of milling means that the cutting edges are not constantly in engagement. With a multiple-edged milling cutter, the number of cutting edges in engagement at anyone time should be taken into account. Depending on the ratio aciD, ziE cutting edges are in engagement, the following relationship being valid: Z'E
= (
where CPc/2 = aciD.
The mean cutting force acting on the milling cutter and thus on the spindle of the milling machine is
Fern =
E
.E z:
~
6
kc1, 0.01 = 75 N/mml
'\
~mco'Wlo.15o 1 !
kc1 01= 998 N/mml
~ 1~~cw04410 ~
r--
I 1 Oil< k kC1 .1 = 2100 N/mm
~~4
I I
Zone 3
-I
1
4
6
B10-1 1
Zone 2
1
4
6 B10- 1
Z'E
Fern"
where Fern, is the mean cutting force of a cutting edge. Superimposed on the mean cutting force is a dynamiC force element. The larger the value of Z'E, the smaller is the force amplitude; if zm is an integer, the cutting force amplitude is at a minimum. The dynamic force element leads to separately excited vibrations between the workpiece and the milling cutter.
IZonelt
1
4
6 B1
Undeformed chip thickness h in mm Figure 26. Specific cutting force in plain face milling with an end milling cutter [131.
Self-excited Vibrations. With self-excitation, the complete system vibrates at one or more eigenfrequencies, without
an external interference force affecting the system. SpeCial importance attaches to self-excited vibrations which arise because of the regenerative effect and are also referred to as "regenerative chatter". The chatter is caused by variations in cutting force due to changes in undeformed chip thickness [16]. Chatter can be influenced by varying the cutting speed, depth of cut, feed rate and cutting edge geometry.
4.2 Machining with Geometrically Well-defined Tool Edges. 4.2.5 Other Processes: Planing and Shaping
Wear. Owing to the intennittent cutting action in mill-
ing, the cutting material is subjected to alternating thermal and mechanical stresses. As a result, not only face and flank wear but cracking in the cutting tip may determine tool life. Figure 1.7 depicts the flank wear on the primary cutting edge and the crater depth for plain face milling with an end milling cutter. Approximate data for selecting the setpoint values are given in Appendix K, Table 4. With the development of cubic boron nitride, the fmishmachining of hardened materials by milling has been further developed [17-19]. Depending on the cutting conditions, surface roughnesses comparable to those obtained in grinding are achieved. In grinding, accuracy of shape is achieved by spark-out. As there has to be a minimum undeformed chip thickness in milling, shape defects occur which can be attributed to the following influencing variables [20]: environment, operating behaviour of the milling machine, inhomogeneities in material hardness, heating of the workpiece due to metal cutting, and a change in intrinsic stress in the edge zone of the workpiece.
Form Milling Hollow-mould tools such as deep-drawing tools are manufactured by both chip-forming and chipless machining processes. Milling plays a central role as a controlled shaping process. The essential characteristic in form milling is the number of actively controlled axes, a distinction accordingly being made between three-axis milling and five-axis milling (Fig. 1.8). In five-axis milling, not only the tip but the axial direction of milling cutter are continuously and Simultaneously controlled relative to the workpiece coordinate system. As a rule, three-axis milling is performed with a convex milling cutter, and five-axis milling is performed with an end milling cutter. The protile of the milled grooves determines the productivity and quality of the process (little finishing being required with a small profile depth). It is formed by machining a curved surface in parallel lines and depends on the milling cutter geometry, the workpiece geometry and the method of working. For a given groove depth t R , five-axis milling with an end milling cutter produces significantly larger
Figure 28. Form milling by a three-axis milling and b five-axis milling; 1 convex milling cutter, 2 end milling cutter, 3 tool axis, 4 surface normaL
groove widths b R than three-axis milling with a convex milling cutter [21].
4.1..5 Other Processes: Planing and Shaping, Broaching,
Workpiece materiat Ck 45N Cutting material HM PZ5 Depth of cut
§!;' 300 f----+-
Planing and Shaping
~:
~
! o ZOO r-----,IT-;!-"
c
.... Vc
Sa~ng
0 ZOO m/min,
fz
160 m/min,
160 m/min,
00.25 mm
200
0.Z5 mm 0.315 mm
I--HJ~-t-- ---+----t--~6+---- 100
'"~
* Q)
D
U
DIN 8589 Part 4 distinguishes between planing and shaping. Chip removal is accomplished during the working stroke with a single-point cutting tool. The following return stroke restores the tool to its original position. The feed is intermittent, usually at the end of a return stroke. In planing, the workpiece performs the cutting and return motion. Feed and engagement are accomplished by the tool (Fig. 1.9). In shaping, the tool performs the cutting and return motion, while feed and engagement are accomplished by the workpiece or the tool. The reciprocating motion of the workpiece (in planing) or the tool (in shaping) produces high inertia forces and limits the cutting speed. As a guideline for the cutting speed, for machining steels the ranges 11, = 60-80 m/min (for roughing) and 11, = 70-100 m/min (for finishing) are well established for cemented carbide tools. Frequently used special methods are hobbing by planing and shaping for manufacturing involute gears (cf. K5.2.1).
Miltlng distance i" In m Figure 27. Progress of wear in plain face milling with an end milling cutter.
Broaching In broaching (DIN 8589 Part 5), material is removed using a multiple-pointed tool, the teeth of which are one behind
.:.1:.
Manufacturing Processes. 4 Cutting
f Figure 29. Planing: a p depth of cut, VT return speed.
f
feed,
Vc
cutting speed,
the other anq are successively stepped by one layer of the metal to be removed. Thus no feed motion is required, as it is "built into " the tool, as it were. The cutting motion is translatory or, in certain circumstances, helical or circular. The advantages of the process are the high machining capacity and the possibility of finishing workpieces with a single tool. Furthermore, high surface qualities and dimensional accuracies with tolerances up to In can be achieved. Owing to the high cost of tools, the main areas of use are series and mass production; a new tool is required for each changed workpiece shape. A basic distinction is made between Internal broaching and external broaching (Fig. ~O). In internal broaching, the broaching tool (broach) is pushed or pulled through a hole, while in external broaching the tool is moved across the external surface. Broaches are divided into roughing, fmishing and sizing teeth sections. Normal undeformed chip thicknesses in flat broaching of steels are h, = 0.01 to 0.15 mm for roughing and h, = 0.003 to 0.023 mm for finishing.
d
2
-.---
Broaching of cast materials is carried out to a thickness of h, = 0.02 to 0.2 mm in the roughing section and h , = 0.01
to 0.04 mm in the fmishing section [22]. Cutting speeds are restricted by the hardness of the chosen cutting material at high temperature and by the efficiency of the machine. The cutting material most commonly used, high-speed steel (HSS), permits only low cutting speeds owing to the decrease in high-temperature hardness at approximately 600 °C; the capacity of the process can be increased by using TiN-coated HSS or cemented carbide. Cutting speeds of Vc = 1 to 30 m/min are used, with speeds of up to 60 m/min in exceptional cases. High cutting speeds require high drive power outputs to accelerate and brake the tool and the broach slide, causing a disproportionate increase in equipment costs [23]. Vibration problems also become greater, especially with thin internal broaches. In broaching, mainly mineral oils are used for lubricating and cooling in the contact zone area, but above all to prevent the formation of built-up edges and to carry away the chips. They usually contain EP (extreme pressure) additives, which are nowadays mostly chlorine-free [22].
f Figure 300. Broaching: a internal cylindrical broaching, b external surface broaching, c external cylindrical broaching, d internal profile broaching, e thread broaching, f external groove broaching; 1 workpiece, 2 tool, 3 original cross--section, 4 final cross--section.
Band sawing is sawing with a continuous, usually straight cutting motion of a rotating endless band. The motions and cutting parameters are shown in Fig. ~1. The normal cutting speeds with high-speed steels lie in the range Vc = 6 to 45 m/min with feed rates per tooth of J,. = 0.1 to 0.4 mm. If bands with inserted cementedcarbide teeth are used, the cutting speed can be increased to 200 m/min for steels and up to 2000 m/min for light metals.
Sawi1lg
Sawing is metal cutting with a multiple-pointed tool having a small width of cut for severing or slitting workpieces. The rotational or translatory principal motion is performed by the tool (DIN 8589 Part 6). The teeth of the tool are offset in alternate directions. By this means the kerf is widened in relation to the saw blade, reducing friction between tool and workpiece.
3 ~---------------- o. ----------------~
Figure ~1. Cutting variables in band sawing: I band saw, 2 workpiece, 3 working plane, Vc cutting speeed,/z tooth feed, Ve effective speed, a e engagement distance, Is cutting feed.
4.2 Machining with Geometrically Well-defined Tool Edges. 4.2.6 Cutting Materials
':.@'
In reciprocating sawing (hacksaw in g) , a tool of finite length clamped in a holding frame is used. The feed motion is carried out intermittently only as the tool advances or at a constant perpendicular force. Circular sawing is sawing with a continuous cutting motion using a circular saw blade. In terms of kinematics and metal cutting technique, circular sawing resembles peripheral milling.
4.2.6 Cutting Materials Tools for machining with geometrically well-defined tool edges consist of the cutting tip, the holder and the shank (Fig. ~2). Holders and shanks are designed according to constructional and organisational requirements, such as mating dimensions of the machine, the nature and extent of tool storage and tool changing, or the geometry of the workpiece. The cutting tip is responsible for chip removal. It is subject to mechanical and thermal stresses and chemical attack. The tip consequently wears (for types of wear cf. K4.2.2). A basic dualism applies to all cutting materials. They are either hard and wear-resistant - but afe then less tough and stable under periodically changing loads as well as under unstable cutting conditions and in intemtittent cutting. Or they are better able to bear variable mechanical or thermal stresses, but are then less resistant to wear. To overCOflle this restricting dualislll. various cutting materials are manufactured as composite materials. Coating with hard-wearing carbides or oxides produces a separation offunctions: the physically (PVD - physical vapour deposition) or chemically (CVD ~ chemical vapour deposition) vapour-deposited layers provide "rear protection, while the underlying, tougher substrate performs the supporting function, even under dynanlic load conditions. Cutting materials comprise: plain and alloy steels (still important for manually guided tools), high-speed steels, cemented carbides. ceramics and superhard cutting materials (diamond and boron nitride) (cf. D3.1.4):
High-Speed Steels. High-speed steels are used for drilling, milling, broaching, sawing and turning tools. Their high temperature hardness (up to approximately 600°C) is far superior to that of tool steels (Fig. 33). Their hard-
ness results from their basic martensitic structure and from interspersed carbides: tungsten carbides, tungstenmolybdenum carbides, chromium carbides, vanadium carbides. Accordingly, high-speed steels can be divided into four groups (high-speed steels are designated by S followed by W%-Mo%-V%-Co%):
Steels Steels Steels Steels
with with with with
18% W, 12% W, 6% W, 5% Mo 2% W, 9% Mo
e.g. e.g. e.g. e.g.
S HH-2-10 S 12-\-4-5 S ()'S-2 S 2-9-2-8
High-speed steels are standardised in "Stahl-Eisen-Werkstotlblatt" (iron and steel data sheet) 320. The through hardenabiliry of tools with large cross-sections is increased
J
2
Figure 32. Parts of a metal cutting tool: 1 cutting tip, 2 holder, 3 shank.
200
400 Temperature in "C
Figure .J3. High-temperature hardness of cutting materials.
with molybdenum and/or by alloying with chromium. Tungsten increases the wear resistance and tempered strength, vanadium the wear resistance (but is difficult to grind when hard) and cobalt the high-temperature hardness. High-speed steels are manufactured by casting. This determines their structure and segregations. These disadvantages can be overcome by means of powder metallurgy (sintered high-speed steels). PM steels have enhanced edge strength and cutting edge durability. They are used for thread-cutting and reanting tools. At high vanadium carbide contents they are easier to grind than cast highspeed steels. They have not yet gained wide acceptance because of their higher cost. High-speed steels are usually coated by PVD (reactive ion plating at low temperatures so as to remain below the tempering temperature). Simple shapes such as indexable inserts can be treated by CVD followed by rehardening. The coating material is titanium nitride (TiN, goldcoloured). The life of coated tools (drills, taps, hobs, form turning tools) is increased by two to eight times.
Cemented Carbides. Cemented carbides are two-phase or multi-phase alloys manufactured by powder metallurgy with a metallic binder. The materials used are tungsten carbide (We: (l'-phase), titanium carbide and tantalum carbide (TiC, Tae: 'Y-phase). The binder is cobalt (Co: f3phase) with a content of 5 to 15%. Nickel and molvbdenum binders (Ni, Mo) are also llsed in so--called "cermets" (also cemented carbides). Higher f3-phase contents increase toughness, higher (l'-phase contents increase wear resistance and higher y-phase contents enhance wear resistance at high temperature. Cermets have high edge strength and cutting edge durability. They are suitable for finishing under stable cutting conditions. The manufacture of cemented carbides by powder metallurgy permits considerable freedom in the choice of constituents (in contrast to casting). Cemented carbides retain their hardness up to over 1O0{) °c (Fig. ~~). They can therefore be used at higher speeds (three or more times as high) than high-speed steels. According to DIN 4990/ISO 513, cemented carbides are classified into the metal cutting application groups P (for long-chipping, ductile ferrous luaterials), K (for short-chipping ferrous metals and for non-ferrous metals) and M as a universal group (for ductile cast iron and for ferritic and austenitic steels). Each group is subdiv-
I;.,••
Manufacturing Processes • 4 Cutting
ided into toughness and wear resistant grades by adding a number; for example, 1'02 stands for very hard-wearing cemented carbide, P40 for tough cemented carbide. The metal cutting application groups do not indicate the material's composition. The classification is done by the manufacturer. Cemented carbides are coated with titanium carbide (TiC), titanium nitride (TiN), aluminium oxide (Al2 0,) or chemical or physical combinations of these substances. The coatings are usually applied by CVD. They are used to achieve longer tool lives or higher cutting speeds. They broaden the range of use of a grade. Coated cemented carbides should not be used for non-ferrous metals, highnickel ferrous materials or - owing to the edge rounding caused by manufacturing - in precision or ultra-precision machining (cermets are better for this purpose). Intermittent cutting and milling requires coatings of especially high bonding strength, which can be influenced by process control during coating.
Ceramic Inserts_ Ceramic inserts are single-phase or multi-phase sintered hard materials based on metal oxides, carbides or nitrides. They are distinguished from cemented carbides by the absence of metallic binders and exhibit high hardness even at temperatures above 1200°C. Ceramic inserts are therefore generally suitable for machining at high cutting speeds, usually exceeding SOOm/min. The use of aluminium oxide ceramic is restricted by its lower bending strength and fracture toughness compared with cemented carbide. In intermittent cutting and alternating mechanical and thermal stresses, microcracking, crack growth with peeling or total fracture can occur. This effect greatly depends on the nature and composition of the ceramic. The change from single-phase materials (Al 20,) to multiphase materials has improved toughness conSiderably: AlP" containing 10 to 15% zr0 2 (transformation reinforcement [24]) or AI 2 0, with TiC (dispersion ceramic). The main uses are turning of lamel-
lar graphite cast iron, under stable contlitions, at cutting speed> 500 m/min. Turning of steel is feasible. Additions of up to 40% TiC to the Al 2 0, ceramic (black mixed ceramic) increase toughness and edge strength. These are used for hard machining and the finishing of cast iron. Silicon nitride (Si,N,) exhibits ideal cutting material qualities (high strength, hardness, oxidation resistance, thermal conductivity and resistance to thermal shock, owing to strong covalent bonding of the elements). Here there is no limitation, because of a lack of fracture toughness. Si,N4 is used as a cutting material in three versions: sintered Si,N4 (p = 3.1 g/cm', Rm = 650 MPa), hotpressed Si,N4 (p = 3.2 g/cm', Rm = 700 MPa), and as the material system Y-Si-Al-O-N. The manufacture and use of Si,N4 is limited by the sintering auxiliaries (e.g. magnesium oxide, yttrium oxide) that are at present necessary. They determine the glassy phases in the cutting material. When machining steel or ductile cast iron, failure occurs due to severe wear. In contrast, Si,N, is suitable for turning and milling of grey cast iron, and also for highly intermittent cuttting actions and for the turning of highnickel materials.
Superhard Cutting Materials_ These are polycrystalline diamond (PCD) and boron nitride (PBN). The materials are synthesised at high pressure and temperature. PCD is supplied as an approximately O.S-mm layer on cemented carbide. It is used to machine aluminium and aluminium alloys, especially easily-earing AlSi alloys, fibre-reinforced plastics, graphite, and non-ferrous metals; it cannot be used for steel, owing to the high rate of
chemical wear. PBN, on the other hand, is chemically stable towards iron. It is used for hardened iron and steel and is supplied as a solid product or as an approximately O.S-mm layer on cemented carbide. Monocrystalline (natural) diamond is used for precision and ultra-predsion machining (turning, milling) of AI and Cu alloys with extremely sharp cutting edges (r ~ < I fLm).
4.3 Machining with Geometrically Nondefined Tool Edges 4.3.1 Fundamentals Machining with geometrically non-defined tool edges is cutting by the mechanical action of cutting edges on the material (DIN 8580, third group of the main group "cutting"). The cutting edges are formed by grains of hard material. They are shaped and arranged irregularly. The cutting edge geometry is therefore not described with reference to a single grain. The intlividual cutting edge is geometrically non-defined. The processes are broken down into the following subgroups: Grinding with a rotating tool Belt grinding Reciprocating grinding Honing Lapping Barrel polishing Machining by abrasive blasting (DIN 8200) The common factor in these processes is that the grains of hard material generally form several cutting edges. The important cutting edge angles for chip formation, the clearance angle a, the rake angle yand the wedge angle {3, are only indicated by means of statistical parameters such as means or distributions. On average, sharply negative face angles and large contact and friction zones are
formed between the grain and the workpiece. The cutting edges penetrate only a few micrometres into the material. The undeformed chip thickness distribution depends on the position of the cutting edges in the mixture of grains (microtopography of the cutting edge zone) and the geometry of the machined workpiece surface. Not only chip removal but also elastic and plastic deformations without chip removal take place. High normal forces result between tool and workpiece at the predominantly negative rake angles of the cutting edges. They lead to elastic deformations in the machine (stretching of the frame and deflection of the spindle), the tool and the workpiece. The deformations may markedly exceed the normal small feed motions. Therefore a distinction should be made between the theoretical and the actual feed motion (Fig. 34). Machining processes with geometrically non-defined tool edges are frequently used as fmal machining processes for workpieces subject to exacting quality requirements. Figure 35 shows a comparison of various precision machining processes with regard to operational results and efficiency. It can be seen that the grinding processes achieve high rates of metal removal, while honing and lapping are able to produce the best surface qualities. The cutting edges are formed by the contours of grains of the hard material. The materials used are hard-brittle materials such as zirconium-corundum (zr0 2 with Al 2 0,), corundum (Al 2 0,), silicon carbide (SiC), boron carbide (B4 C), boron nitride (BN) and diamond (C); their hardnesses are shown in Fig. 36. However, diamond is unsuitable for machining steel, as there is a high chemical
';-'
4.3 Machining with Geometrically Non-defmed Tool Edges. 4.3.1 Fundamentals Feed motion of grinding tool
6000
Stretch of Relaxation system _ _--il-__O_f_sy_s_te_m_ _--I
Diamond 4100
Cubic boron nitride Theoretical
Feed motion error during machining
mo
Boron C81bide
2\10
SiI.con carbide
2010
Effective feed motion
Corund\Jm
\olachining cycle
1000 2000 3000 4000 5000 6000 Knoop hardness HK 0.1
Figure 34.. Feed motion errors in precision machining due to elastic deformations in the machine-tool-workpiece system.
Fipre 36. Knoop hardness of various hard materials.
affinity between diamond and iron which leads to rapid tool wear. Sorting of the grains according to size is accomplished by screening (DIN 69 100). The basis of all standards (cf. Appendix )(4, Table 5) is the mesh of the screens through which the abrasive grains pass. The average grain size is determined by the shape (angularity) of the individual grains. Below a certain grain size, sorting can be carried out by settlement out of a slurried suspension of water and grains. The grains are either bonded together into a tool (grinding, honing) or are used loose (lapping, abrasive blasting). The bonding material (bond) is chosen according to the requirements of the machining process and of the grain material. Inorganic (ceramic, silicate, magnesite), organic (rubber, synthetic resin, glue) and metallic bonds (bronze, steel, cemented carbide) are used; the most popular bonds are those made from cetamic or synthetic resin. In manufacturing a tool, its structure can be influenced to a certain extent by varying the proportions of grains, bond and pore volume. The chip formation mechanism when using geometrically non-defined tool edges differs from that applicable to machining with geometrically well-defmed tool edges (Fig. ~7). A characteristic feature of this process is the often sharply negative rake angle of the individual grains. In Phase 1, this causes elastic deformation of the material.
,~
g>
~
'C
.!O
Positional accuracy
• -
Surtace
0
•
Favourable inHueoce on edge zones
0
•
• +
0
-
+
Metal removal performance
Mar>Jfacturilg costs
9.
¢-
_:::-='t-
~ --\ i !
1 7
I
Ly
-f'--::":.
.
)
I I
i
7
6/ --/
---' '\
,-_ ......
;>/....
..... .... __ ........
Figure ~7. Phases of chip formation: I elastic deformation, 2 elastic and plastic deformation ("ploughing"), 3 elastic and plastic deformation ("ploughing~) and deformation of chip by shear
(cutting deformation), 4 elastic and shear deformation of chip, 5 elastic deformation zone, 6 plastic deformation zone, 7 chip.
,~
Dimensional accuracy
0
f
'"'""')
!l
·• ....••
Accuracy of shape
.800
ZirconilJll\corundum
/ . \
• +
-
+ •
- Uttle influence + Great influence o Moderate influe!1ce • + Very great influence
l~"'I~
. , - C)WJdfiaI gMdif'g
......... pet-
/'
r--
)',
._eyIino:ticollJ'f"dog ~pellpIJo<~
/"I
""""JlI'Jdif'g
Loog-stroke honing Short-stroke honing Surtace lapping
~
.---V L -,i r--
/' o 0,1
r\10
1
100
Average sunace roughness R, in
)IlIl
Figure 3>S. Economic and technological comparison of various precision machining processes.
':-.1
Manufacturing Processes. 4 Cutting
In Phase 2 plastic defonnation occurs, while in Phase 3 the actual chip removal takes place. A large amount of friction occurs between the individual grains and the workpiece. The mechanical energy supplied is almost exclusively converted into heat. Figure ~8 shows a qualitative distribution of the heat flows around an individual grain. Most of the heat generated flows into the workpiece, while a small proportion fmds its way into the grain, the bond and the surroundings (cooling lubricant, atmosphere). Temperature increases in the workpiece may hann its edge zone. This is manifested in thennally induced inherent stresses, structure changes or cracks, which influence its subsequent behaviour in service. The use of readily heat-conducting granular materials (CBN, diamond) and bonds reduces the proportion of the heat that passes into the workpiece [25]. In machining with geometrically non-defined tool edges, the use of cooling lubricant is important for the end result. The cooling and lubricating effect can reduce tool wear. Furthermore, the temperature of the workpiece is reduced and the danger of damage to the edge zones is decreased. The lubricants used are non-water-miscible (oils) and water-miscible (emulsions, solutions) cooling lubricants (DIN 51 385), the effect of which can be further improved with additives (polar and EP additives to improve the lubricating effect, antifoaming agents, biocides and rust inhibitors). The cooling effect depends on physical parameters: specific heat capacity c in kJ/kg K, heat transmission coefficient ex in W 1m' K, thermal conductivity ,\ in W 1m K, heat of evaporation Id in kJ/kg and surface tension a in N/m. The lubricating effect is described by the tribological parameters of the cooling lubricant. 4_~_2
Grindin.g with Rotating Tool
Processes. Grinding is divided in DIN 8589 Part 11 into six processes according to the shape of the surfaces pro-
duced. Figure ~9a shows the classification and Figs ~9b to j show examples of various motion classifications and tool shapes.
Chip Formation. Material is removed by the penetration of abrasive grains into the material along a flat path. Owing to the generally unfavourable shape of the cutting edge, the actual chip formation is accompanied by friction and displacement processes. The process is evaluated' by calculating statistical averages. Figure 40 shows in simplified tenns how a comma-shaped chip is fonned by the successive action of two cutting edges. While grain I has travelled the path AB, the centre of the grinding wheel has moved from 0 to 01. The next grain 2 will travel the path CD. In the process, the thickness of an average chip increases from 0 up to hm~' A simple relationship for the average undefonned chip thickness h is obtained by applying the continuity relationship v"a.,ap
. - = -1a,d,q,
With I
V,p
---
rlwd,
= Ibh and d eq = d w ± d, (+ external
cylindrical grinding, - internal cylindrical grinding) or
h=
6
withr =-:c. h
Here, fj = average (undefonned) chip thickness, I = averaverage age (undefonned) chip length, b (undefonned) chip width, v" = rate offeed of workpiece, v, = cutting speed, a e = depth of cut, infeed, a p = width of engagement (width of grinding), d eq = equivalent grinding wheel diameter, d, = grinding wheel diameter,
= workpiece diameter ( - 00 in surface grinding), C = number of active cutting edges per unit of surface area of grinding wheel, r = ratio of average chip thickness to d'iN
average chip width. The maximum chip thickness h m . . is twice the average chip thickness fj thus determined. Owing to technical difficulties in measuring the number and distribution of grains, the equivalent chip thickness heq is often used as a parameter for evaluating the grinding process [26, 27].
Composition of Grindin.g Wheels. A grinding wheel consists of grains, bonding material (bond) and pores. The specification of a grinding wheel is standardised to DIN 69100. Grinding wheels made of diamond or cubic boron nitride (CBN) are not covered by this standard. They consist of a backing material to which the grinding layer is applied. Usual layer thicknesses are 2 to 5 mm. Grinding wheel wear may take place in the grains and the bond. Various kinds of wear and means of sharpening are shown in Fig. 41.
a
5
b Figure ~8. Energy conversion: a effects of energy conversion, b energy flows; 1 friction, 2 cutting, 3 shearing, 4 grain, 5 chip.
Limits to the Process. Restrictions on the process arise if the Original data, e.g. dimensional accuracy, accuracy of shape, surface quality or condition of workpiece edge zones, do not lie within the required limits. The interaction of the various influencing variables, such as workpiece, machine setting data, tool, cooling lubricant, etc., may be extremely diverse. Mechanical or thermal overstressing of the material in the grinding process may adversely affect the characteristics of a ground component [28]. Typical grinding
':..11
4.3 Machining with Geometrically Non-defined Tool Edges. 4.3.2 Grinding with Rotating Tool
a
b I
/
~
I~ /
('\.~d2
h
~
Feed motion, continuous Feed motion, intermittent Feed motion, reciprocating
Cutting motion Advancing metion, intermittent
Figure ~9. Grinding processes, schematic (DIN 8589): a classification, b longitudinal peripheral surface grinding, c transverse peripheral external cylindrical grinding, d longitudinal peripheral external cylindrical grinding, c transverse peripheral internal cylindrical grinding, f centreless throughfeed grinding, g longitudinal external thread grinding, h discontinuous external hobbing by grinding, i longitudinal external profile grinding, j form regrinding; I grinding wheel, 2 workpiece, ;3 regulating wheel, 4 work rest.
I !
'-'-'-'---1-'T--'-'//11 // . y,; /' I I /
/' . I I '.,
.I /
Ii ~ .'
a
12I.
....,.. .
A Figure 40. Engagement (explanations in text).
relationships
in
surface
grinding
defects attributable to poor process control are chatter marks, inherent tensile stresses [29]' grinding bum and cracks in the workpiece. Dre....ing. The purpose of dressing is to give the grinding wheel the required profile and concentricity (profiling) and to produce the necessary grinding wheel topography with sharp grains (sharpening). Both oper· ations are generdlly performed at the same time by passing a dressing tool over the face of the grinding wheel. A survey of common stationary and rotating dressing tools is contained in [30J. The main components of dressing tools are elements coated with diamond panicles; however, there are also diamond-free steel and ceramic elements or surfaces. Electrochemically bonded grinding tools coated
~--Sp,ntering
Bleaking off b
By fonning nell cuHing edges
grain groups
c
Figure 41. Types of wear and means of sharpening: a sharp grinding wheel, b types of wear, C means of sharpening.
':..1.
Manufacturing Processes. 4 Cutting
Figure 44. Schematic representation of process variants in surface belt grinding: Vo cutting speed, Vft workpiece speed, a e working engagement, bw grinding width, Pn perpendicular force; 1 grinding belt, 2 deflection roller, 3 contact roller, 4 workpiece.
Figure 4Z. Principle of grinding with continuous dressing (CD grinding).
with only one abrasive layer cannot be dressed. They have reached the end of their life when this abrasive layer is worn down. A special position is occupied by grinding with continuous dressing (CD grinding) (Fig. 42). Here the dressing tool, generally a diamond dressing roll, is in engagement during grinding and is continuously advanced radially to the grinding wheel. As a constant grinding wheel profile and a uniform grinding wheel topography with sharp cutting edges are permanently ensured, the time-related chip volume can be increased considerably [31]. With the aid of the machine control system, the dressing tool and the grinding wheel have to be advanced in relation to the workpiece in such a way that the decrease in the diameter of the grinding wheel is compensated for.
4.3.3 Belt Grinding Belt grinding is grinding with tools on a bed (DIN 8589). Belt grinding can be broken down as shown in Fig. 43 according to the surfaces that can be produced. Surface belt grinding predominates in industrial applications (Fig. 44). Belt grinding is normally performed at a constant perpendicular force Fn (pressing force). In this way, consistent surface qualities can be produced. The time-related chip volume is determined by the sharpness of the belt (of the active cutting edges). In belt grinding with constant working engagement a c , a constant timerelated chip volume is removed. Surface qualiry depends on the condition of the cutting edges. This process is especially suitable for removing large volumes of material (time-related chip volumes) [32]. The movement of the process variables during belt grinding stems from the changes to the cutting edges during working. In contrast to grinding wheels, grinding belts generally consist of a single layer of abrasive. Therefore the cutting edge zone changes over the period of use, owing to progressive abrasive and bond wear. Over the life of the grinding belt, three phases can be distinguished (Fig.4S). In the calibration phase, a rapidly changing grinding-in process takes place. The small number of grains that project funhest out of the cutting
Development Trends. Grinding has developed from a traditional precision machining process to improve dimensions, shape and surface qualiry into a very versatile and efficient manufacturing process. New grinding processes such as creep feed grinding, high-speed grinding and grinding with continuous dressing (CD grinding), the growing use of the superhard abrasives diamond and cubic boron nitride (CBN), together with CNC and sensor technology, have equally contributed towards increasing the performance of this manufacturing process.
I
I
I
Surlace be~ grinding
L 1
I
Belt grinding
,
I I~ror~1 I I grildJng
belt
,
Profile bell grinding
I
I
1
Form be~ grinding
t ~.~, • • 1
1
I
I
'; j
-
:iF
Longitudinal peripheral surlace belt grinding
~
periphetal surlace belt grindilg with c~cuIar feed rnooon
LongiludilaJ side surlace beK grinding
f
Transverse side surface belt grinding
Fipre 43. Survey of belt grinding processes with detailed classification of surface belt grinding (DIN 8589 Part 12).
.:..1.1
4.3 Machining with Geometrically Non-defmed Tool Edges. 4.3.4 Honing
Fn,t
fI, llr,
,...:....1
('v"'-i
3
•
2
f"-..i
,.. . . .J v
v
'-'
{",ox
Grinding belt lite t Fipre 4S. Process behaviour of a grinding belt during its service life: 1 perpendicular force Fn, 2 tangential force FI> 3 average surface roughness R~, 4 radial wear d.rs .
edge zone owing to the belt manufacturing process very quickly break off. As a result, radial wear initially rises rapidly and deeper lying grains engage in cutting. The increase in the number of active cutting edges causes a rise in the grinding forces and a decrease in the average surface roughness R,. The self-sharpening phase is characterised by continuous wear combined with formation of new cutting edges. The increase in the number of grains actively participating in the cutting process leads to a regular increase in the grinding forces and a decrease in the depth of roughness. The end of belt life is marked by rapidly advancing grain and bond wear, ultimately leading to failure of the grinding belt. Grinding belts consist of four elements: backing (paper, fabric), first-coat and top-coat bonding materials (phenolic resins), and abrasive grains (corundum, zirconiumcorundum, silicon carbide). Scattering of the grains on the ftrst coat of bonding material is carried out in an electrostatic fteld. 111is ensures that the abrasive grains are aligned perpendicularly. An even grain distribution and thus high reproducibiliry in belt manufacture compared with conventional gravity scattering can be achieved [33]. The grinding belt is supported in the working zone by contact elements. A contact disc is used in peripheral grinding, while a contact shoe or beam is used in side grinding. Hard contact rolls made of aluminium or steel are particularly suitable for roughing with coarse grinding belts (grade 36; 50) to transmit the relatively high grinding forces. Soft, rubber-sheathed contact rolls are used in ftnishing with fme grinding belts. They absorb the shocks caused by grinding belt wear [32]. Conventional applications of belt grinding are grinding of individual sheets and sheet coils, dellashing and deburring, and grinding down of excess metal in welded joints. A recent advance is the successful use of heavy-duty belt grinding in the motor industry and elsewhere as a substitute for turning and milling of components made of grey cast iron or aluminium alloys [32, 33]. Some principal grinding data are given in Appendix K4, Table 6.
/' b Figure 46. Geometry and kinematics in a short·stroke and b longstroke external cylindrical honing [341. Vr. axial rate offeed, lw workpiece length, lh length of honing stone.
groups: long-stroke honing and short-stroke hOning (Fig. 46) [34]. Long-stroke boning employs a large oscillation amplitude at a low frequency; in sbort-stroke boning the oscillating motion is performed at low amplitude and a correspondingly high frequency. The path curves in Fig. 4S depict the motion of a honing strip over a developed workpiece surface. Owing to the superimposed motion during honing, the workpiece surface exhibits intersecting tracks of the cutting grains, the two tracks enclosing an angle a (Fig. 47). The magnitude of the overlap angle a is determined by the selection of the ratio of the axial (v,) and tangential (v,) cutting speed components. For workpieces without longitudinal and transverse grooves, the angle a is generally 45°. The cutting speed Vc can be calculated by means of the aforementioned speed components
'-
/
'-
/
;~
b
4.3.4 BoniAg Honing is performed with a multiple-pOint cutting tool consisting of bonded abrasive grains using a cutting motion consisting of two components of which at least one is oscillating. The principal honing processes are external cylindrical honing, internal cylindrical honing, and surface honing. According to the oscillation amplitude, a further distinction can be made between two main
c Fipre~'" _ Working
operation in long-stroke honing: a working
principle, b honing movement of tool, c surface structure (a overlap angle).
,:-
Manufacturing Processes • 4 Cutting
according to v, = (v; + VDI!2, The cutting speed normally does not exceed v, = 1. 5 mls [35, 36]. During the cutting motion, the honing stones are pressed against the workpiece face to be machined at a perpendicular honing force Pm which may be generated by means of various feed systems (Fig. 48). In forcedependent feeding, a defined hydraulic pressure POi! is set on the machine. The resulting advancing force P, is transmitted to the honing stones via an advancing pin and cones. In path-dependent feeding, defined feed paths are generated, e.g. with a stepping motor [37], from which the perpendicular force Po at the honing stones results. Important variables influencing the result of honing are type of abrasive, grain size, type of bond, hardness and impregnation of the honing stones. The types of abrasive can be broken down into the conventional abrasive materials corundum and silicon carbide and the superhard abrasive materials diamond and cubic crystalline boron nitride (CBN). Grain size influences time-dependent chip volume and surface qualiry. The achievable surface roughnesses are R, = I fLm for long-stroke honing and R, = 0.1 fLm for short-stroke honing. Dimensional accuracies and accuracies of shape of the machined workpieces of I to 3 fLm are achieved. Unlike in grinding, the grains bonded in the honing stone are stressed on more than one axis owing to the oscillating motion. Honing tools are therefore self-sharpening. Cooling lubricants are used in honing as well as in grinding. Owing to the low cutting speed, however, heating is minimal, so the cooling effect plays a minor role. The surface contact between the honing stone and the workpiece requires instead a friction-reducing lubricating effect. Therefore pure oil, with additives if required, is generally used. The applications of honing can also be categorised according to whether long-stroke or short-stroke honing is used. Long-stroke honing is generally used for intemal-
4.~.S
Other Processes: Lapping, Inside Diameter Cut-off Grinding
Lapping Lapping is defined in DIN 8589 as metal cutting with loose grains of abrasive distributed in a paste or a liquid, the lapping mixture, which is carried on a generally shapeimparting counterpart (lap), the individual grains following cutting paths which are random. Lapping processes ate broken down into surface lapping, cylindrical lapping, hole lapping and ultrasonic machining (Fig. 49). Surface lapping or plane parallel lapping is carried out on single- or twin-disc lapping machines. The lapping discs act as the holder for the lapping abrasive. They ate chiefly made of perlitic cast materials or hardened steel alloys. The abrasive consists of the lapping powder and the fluid in the ratio 1 : 2 to 1 : 6. Particles of silicon carbide, corundum, boron catbide or diamond are used as lapping powder. The rype of grain for a patticular application is determined by the material to be machined. Grain sizes ranging from 5 to 40 fLm ate generally used. Besides highviscosiry oils or similar fluids, water with suitable additives has seen growing use as a vehicle for the abrasive in recent years. The purpose of lapping fluids, among other things, is to cool the workpiece and ensure chip removal from the effective zone. Lapping is a precision or ultra-precision machining process for producing functional surfaces of optimum surface quality. Surface roughnesses up to R, = 0.03 fLm, flatnesses < 0.3 fLm/m and plane parallelisms of up to 0.2 fLm are achieved. Typical applications of lapping are the machining of precision cemented catbide tools, calliper gauges or hydraulic rams. A special category of lapping is ultrasonic machining, which is particularly suitable for machining hard-brittle materials, e.g. fully sintered ceramic components [38, 39].
cylinder workpieces, e.g. piston contact surfaces in internal combustion engines. Short-stroke honing is mainly used for machining small cylindrical components, e.g. running surfaces of inner and outer rings and rollers of rolling-contact bearings [34].
2
•
b
Figure 48. Force- and patb-depcndent feed devices for honing [37J: a force-dependent, b path-dependent; I honing stone, 2 workpiece, 3 step motor.
Figure 49. Lapping processes (DIN 8589 Part 15): a plane-parallel lapping, b lapping of external cylinders, c ultrasonic machining; I lap, 2 workpiece, 3 abmsive, 4 lapping disc drive, 5 lapping cage, eccentric, mounted in bearings, 6 cage drive, 7 vibrating tool.
4.4 Chipless Machining. 4.4.2 Electro-discharge Machining (EDM)
Inside-Diameter Cutoff Grinding Inside-diameter (ID) cutoff grinding is a high-precision finish-machining process for hard-brittle materials. It is used to cut rod-shaped materials into thin slices (Fig. SO). Besides its applications to optical materials (glasses, glass ceramics), magnetic materials (samarium-cobalt, neodymium-iron-boron), ceramics and crystals for solidstate-type lasers, this process is used above all for semiconductor materials. Single-crystal silicon rods are cut into thin slices termed "wafers" (see K5.3). Compared with conventional abrasive cutting processes, the loss of material in the cutting gap can be reduced by approximately 80% by means of a narrow width of cut. This is a decisive advantage, especially for expensive, high-grade materials. To achieve the narrow widths of cut that are typical of this process, a comparatively unconventional tool is used for ID cutoff grinding. The basic body of the tool consists of a cold-rolled circle of high-tensile stainless steel with a thickness of 100 to 170 fLm. The inside edge of the cutting blade is coated hy electrodeposition with a diamond layer in a nickel-base bond; this layer forms the droplet-shaped cutting edge. Grain sizes commonly range from 45 to 130 fLm. Accordingly, the width of cut extends from 0.29 to 0.7 mm. Natural diamond is generally used as the cutting material. CBN may also be used for special applications. Workpieces with a diameter of up to 200 mm can be cut. To achieve the stiffness at the cutting edge that is necessary for cutting, the cutting blade, which can be compared to an eardrum, is clamped at its outside edge with a special clamping device. This widens the ID cutting blade radially until the tangential stresses at the inside edge reach around 18001\/mm'. In cutoff grinding, the workpiece undergoes a radial feed motion relative to the rotating tool [40, 41J
4.4 Chipless Machining 4.4.1 Survey Chip-producing machining processes operate by the mechanical effect of cutting edges on the workpiece.
~~ ~J Cross-seclion ~B
/
~--Z
. 5
\
( Vb?~) \
6
./
',---./
Figure so. Principle of 10 cul-off grinding; 1 clamping ring, 2 holding ring, 3 lD cutting blade, 4 Si crystal, 5 blade centre, 6 diamond-coated cutting edge, v<" cutting speed, lJ lr radial feed rate, Fm F(1 f~ process forces.
They therefore depend on workpiece properties such as strength, hardness, resistance to wear or toughness. Chipless machining processes utilise thermal, chemical or elec· trochemical processes for forming. They do not depend on the mechanical properties of the materials. They are well established for machining of materials that are diffi· cult or impossible to machine mechanically (highly tempered and hardened tool steels, nickel·base alloys or superhard materials such as diamond or cubic boron nitride). They are also used to machine complex, hard-toreach or very small (microtechnology) surfaces and contours. According to DIN 8590, chipless machining is manufacturing by removal of material particles from a solid object without mechanical action (ct. KS, Fig. 43). Thermal machining is characterised by the removal of material particles in solid, liquid or gaseous condition by the action of heat. The particles removed are carried away by mechanical and/or electromagnetic forces. This subgroup is further broken down according to the energy source that supplies the external heat required for cutting. The principle of chemical milling is based on chemical reaction of the material with an active medium to form a compound that is volatile or easily removable. The conversion of material takes place by a direct chemical reaction. Electrochemical machining is accomplished by the reaction of metallic materials with a dissociated, electroconductive active medium by the action of an electric current to form a compound that is soluble in the active medium or precipitates out. The current flow is initiated by an external power source. 4.4.2 Electro-discbarge Machining (EDM) Electro-discharge machining is used for machining of electroconductive materials in a dielectric. This is achieved by creating repeated discharges between an electrode and the workpiece in rapid succession [42J (Fig. SI), as follows. In the first phase, the dielectric is ionised at the point where the gap is narrowest (highest field strength) (t 1)' A discharge channel is formed cumulatively (ionisation by collision). The discharge current builds up, and the voltage falls to the physically dependent gap voltage of approximately 25 V (t,). In the third phase, the plasma is heated in the widening discharge channel. Tem· peratures of approximately 10 . 10' K arise, owing to constriction of the discharge. Small quantities of material are melted at the ends of the arc (electrode and workpiece) (t,). At the end of the pulse, the superheated molten metal evaporates explosively (t4)' The energy per pulse determines the crater size and the effect on the workpiece edge zone [43J. The dielectric has the following functions: insulation of workpiece and electrode, establishment of favourable ionisation characteristics, constriction of the discharge channel, carrying away of the particles removed, and cooling of electrode and workpiece. Hydrocarbons are used as dielectrics. The spark energy is produced by a generator (static pulse generators are used exclusively nowadays). The pulse is controlled with an electric switch. The current is restricted by the impedance Z; the pulse length is adjustable from I to 2000 fLs. The pulse-to-space ratio T = t;ltp is variable from 0.1 to 0.5, the no-load voltage U, can be varied between 60 and 300 V, and the pulse current 1, is adjustable within the range I to 300 A. Electro-discharge machining may he carried out in various alternative forms (Fig. S2). In cavity sinking by EDM, the tool is an electrode with the negative shape of
,..
Manufacturing Processes. 4 Cutting
8
9 Widening of discharge channel
lonisatioo of dielectric
10
Formation of discharge channel
Ffaure S3. Parts of an EDM cavity sinking machine [45]: 1 feed drive, 2 working head, 3 work container, 4 workpiece, 5 electrode, 6 return pipe, 7 dielectric supply unit, 8 axia1 drive control system, 9 generator, 10 power supply, 11 spark gap, 12 compound table, servomotor-
u(f)
the cavity to be produced. An EDM cavity sinking machine consists of the machine tool, generator, control unit for the axial drives and the dielectric unit ( .... S3). Drives in three spatial directions perform the positioning and advancing of the electrode. By monitoring the electrical data at the spark gap, its width is adjusted to the setpoint value ("" 10 to 80 ,..m) in a highly dynamic manner. The rate of feed is determined by the progress of the machin· ing process and cannot be preset [44]. Productivity and end result are determined by the elec· trode material, the dielectric and the electrical settings (current, pulse duration, pulse-to-space ratio and polarity). The machining process is divided into several
Time t
i(1)
14 Time t
I, Fipre SI. Phases of spark discharge [44].
I I
I I 1.
Drilling byEDM
I
I
I
I
Hole making by EDM
Cutting by EDM
I
I Cavity sinking byEDM
4. Cutting by
Cutting by EDMwith blade
EDMw~hwire
or strip
1
I
li
".-
Ffaure S:I.
5.
I
Cutting by EDMwith rotating ·wheet
1
-
-+.
I
Grinding by EDM
I
I
6.
Cylindrical grinding by EDM
I
/
-.
I
I
I
3.
2.
I
rnacn'mng (EDM) processes
Elect~iSl:narge
d·!"
./~ I ..
Classification of electro
I
7. Surface
grinding by EDM
I
4.4 Chipless Machining. 4.4.3 Laser Cutting
roughing and finishing processes. In roughing, removal rates of Qw = 600 mm'lmin at I , = 60 A and low relative wear (2 to 5%) are achieved. Finishing is carried out at low currents and with brief discharges. Surface qualities of R. = 0.3 fLm and dimensional deviations of less than 10 fLm can be achieved. The thermal machining process affects the workpiece edge zone to a depth of 5 to 50 fLm. An amorphous structure may form there. Intrinsic tensile stresses occur in the layer close to the surface, reducing the dynamic strength of the material. Electrodes are made from materials with high melting pOints andlor high thermal conductivity. Common materials are copper and graphite; sintered tungsten-copper materials are used in special cases [45] (A.ppendix K4, Table 7). Cavity sinking by EDM is used for making hollow moulds for primary shaping and metal forming tools. To the original EDM cavity sinking process with only one vertical feed motion have been added planetary EDM and path-controlled EDM (Fig. 54). In planetary EDM, a rotary motion of the electrode is superimposed on the downward vertical motion. This achieves better flushing, even distribution of electrode wear and a uniform undersize of the roughing and fmishing electrodes. Wider possibilities are offered by path-controlled EDM; electrodes of simple shape can be controlled to produce complex shapes. In cutting by EDM, a running wire electrode is moved along a curved path relative to the workpiece. The cutting gap is produced by electrical discharge machining. Cutouts of any desired contour are produced in flat components (Fig. 55). For diagonal prismatic cutouts the wire guides may be offset against one another. An EDM cutting machine comprises the actual machine tool with the wire feed, the generator, the control unit for the axial drives and the dielectric preparation unit. The end result depends essentially on the cutting wire. The usual wire diameter is 0.25 mm. The wire running speed is up to 300 mmls, the generator current is between 15 and 100 A, removal rates are up to 350 om,' lmin, dimensional accuracies and accuracies of shape are better than 0.01 mm, surface rouglmesses are R. = 0.3 fLm. Cutting by EDM is used in tool manufacture, e.g. for making punching, injection moulding and extrusion tools. 4.4.~
Laser Cutting
In laser cutting, light energy is generated in an optical resonator and transmitted to the material by absorption in
z
·1
.
.~
:r,~
. 0.
~.
FilJUl'C' 5'i. a Cavity sinking by EDM, b planetary EDM, c pathcontrolled EDM; x, y, Z, c: relative electrode motion.
I:..'
1r-- -- 6 FilJUl'C' 55. Principle of cutting by EDM [441; 1 wire feed, 2 invcrtcd-V guidance principle, 3 flushing nozzle, 4 flushing chamber, 5 power supply, 6 cutting wire.
the form of heat. (LASER = Light Amplification by Stimulated Emission of Radiation.) For cutting tools, only CO, lasers, Nd : YAG lasers and, recently, excimer high-capacity lasers are employed, owing to the extremely high beam powers required [46481. The beam characteristics of these lasers that are important for material machining are summarised in K5, Table 2. Laser cutting of metallic materials requires intensities of > 10" WI cm', which are achieved by focusing the laser beam with the aid of lenses or mirrors [49]. The thermal material removal process, which is directed into the depths of the material, produces a kerf in the material when feed motion is applied. The principle of laser cutting is iIlustrdted in Fig. 56. The material that is melted (laser fusion cutting), burnt (laser flame cutting) or vaporised (laser sublimation cutting) at the focal point of the laser beam, depending on the intenSity and the length of interaction, is expelled from the kerf by a stream of gas emitted from a nozzle coaxially to the optical axis. The cutting gas also serves to protect the sensitive focusing optics from spattered material. In laser flame cutting, oxygen or oxygen-rich gas is used as the cutting gas; at higher cutting speeds, however, this leads to oxidation of the cut surfaces owing to the introduction of additional exothermic energy. In contrast, in the other laser cutting processes mentioned above inert gases (e.g. argon, nitrogen) are used as cutting gases, which results in a slower cutting speed; however, they produce an oxide-free cut [50]. The relative motion between the laser beam and the workpiece that is required to produce a continuous kerf is achieved in practice in various ways. For laser cutting of small, easily handled components, the latter are generally
I:••'
Manufacturing Processes. 4 Cutting
Cutting dlrecOO1
WOIkpiece
Figure S6. Principle of laser clitting: v'-' cutting speed,
Zf
a
focus
position, J laser beam (wavelength A, laser power Pl' mode, pulse frequency J~ , pulse duration Ii), 2 focusing lens (focal length f) ,
Craie! width w,
3 cutting gas (gas pressure PJI.' type o f gas), 4 cutting tip (shape, diameter), 5 diameter of focal spot d r, 6 workpiece, 7 ex~lled material , Do diameter of the unfocused light beam.
c moved underneath the stationary laser beam, e.g. with the aid of an X, Y coordinate table. For laser machining of larger workpieces, the laser unit including the cutting tip is either moved across the stationary workpiece, or a movable system of mirrors is guided together with the cutting tip ("flying optics") berween the flXed laser unit and workpiece. For Nd: YAG lasers alone, flexible optical fibres may be used for guiding the beam [51 , 52]. The machining process is influenced by many different process parameters, the most important of which are given in Fig. 57 together with their defmitions. The maximum achievable cutting speed in relation to laser power and material thickness is shown in Fig. S8 by way
of example for structural steel using a CO 2 laser. These are average values calculated from data supplied by various users. In addition, the achievable cutting speeds of other metallic and non-metallic materials for a (C0 2 ) laser power of P L = 500 W are summarised in Table ~ . There is no current standard for the definitions for determining the qualiry of laser cuts and their measuring instructions; however, this is often done on the basis of a guideline issued by the CIRP-STC-" E" expert group [53]. This guideline is based on the definitions shown in Fig.
~
High-powered lasers of the type mentioned above generally fall into laser (protection) class 4, which has the highest hazard level (except for laser machining systems with closed working chambers, which are equipped with additional safety facilities such as interlock systems and r-.tdiation-absorbing protective windows). Allocation to this safery class means that even diffusely reflected laser radiation is a hazard to the skin and the human eye. Comprehensive instructions on radiation safety of lasers are laid down in DIN VDE 0837 and accident prevention regulation 46.0 (VBG 93).
--
d
depth h
Plana y-z
:y Actual profile I 'r~._._~ h !
Basicplofile
e
Figure S7. Definitions for evaluating quality of cut (53] : a groove deflection n, b unevenness u, c width of affected zone Wd (in Germ-
any: including matrix restoration and heat-affected zone) , d cratering, e groove depth b(z) ,
!
co
~
4
I,
/
"v~'1
/
/
/
Laser CO2 Material St 37
;/
V
'<>
// ~~
1 /
~
~: /
500
1000
1500
2000
2500
Laser power Pl in W
3000
4.4.4 Eledro<:hemicai Machining (ECM)
I'igure S8. Cutting material thicknesses.
The basic principle of electrochemical machining corresponds to an electrolytic cell. ElectrOlyte solution flows at high speed berween the workpiece (anode) and the tool (cathode) ; the gap berween the electrodes is 0.05 to 1 mm. Hydrogen ions are discharged at the cathode. Metal ions react at the anode with OH ions from the water, for-
ming metal hydroxide compOtmds which settle as a sludge. A widespread process is mould deflashing (Fig. 59). The tool electrode has to be matched to the workpiece. The flash is preferentially removed owing to the maximum current density being present at that point.
speed
versus
laser
power
for
various
4.5 Shearing and Blanking. 4.5.1 Classification
• •1
Machining parameters for laser cutting of various materials. Laser CO.d500 W, lens focal length J = 5 in.
Table~.
Material
Thickness (mm)
Cutting gas/ pressure (MPa)
Cutting speed (m/min)
PMMA (Plexi)
Air /0.06
3.5
Rubher
N, /0.3 Air /0.3 N, /0.15 Air /0.3 N, /0.5 0, /0.2 Air / 0.5 0, /0.45 O 2 /0.6
1.8 1.6
Asbestos Deal Eternit
4
AlTi ceramic
8 1.5
Aluminium Titanium CrNi steel Magnetic steel
3
4
2
0.35
55 0.8 0.07 0.4 1.9
Figure 59. Electrochemical
sheet Grey cast iron
N, /1
0.9
4.4.5 Chemical Machining Chemical machining is accomplished by a chemical reaction of the material with a liquid or gaseous medium. The product of reaction is gaseous or easily removable. An example of chemical machining is thermal deburring. This process consists of a thermal component (heating up the material) and a chemical one (burning of the material). In thermal deburring, metallic or non-metallic workpieces are pressed with a closing plate under a bell-shaped deburring chamber (Fig. 60). Oxygen and fuel gas (natural gas, methane or hydrogen) are fed into the chamber at a controlled rdte. The gas pressure and the mixing ratio determine the amount of material removed. While the mixture is being burnt, the temperature briefly reaches 2500 to 3000 dc. Parts of the workpiece with a large surface area and a small volume (low heat
mould deflashing (according to DIN 8590); 1 flash, 2 flow of electrolyte . .3 tool electrode (cathode),
4 workpiece (anode).
capacity) are burnt (oxidised). The burrs must be thinner than the thinnest parts of the workpiece. After deburring, the temperature of the workpieces is 100 to 160°C.
4.5 Shearing and Blanking K. Siegert and J. Ladwig, Stuttgart
4.5.1 Classification According to DIN 8588, the processes of severing - mechanical cutting of workpieces without formation of amorphous material - are divided into blanking, wedge-action cutting (single-blade cutting, cutting with two approaching blades, cleaving), tearing and breaking (Fig. 61). Particularly in sheet metal working, blanking predominates, often as a preparatory, finishing or inter-
1
Figure 60. Parts of a thermal deburring unit (according to Thilow); 1 spark plug, 2 mixing block, 3 fuel gas feed cylinder, 4 gas injection cylinder. 5 oxygen feed cylinder, 6 debuning chamber. 7 gasket, 8 workpiece holder, 9 closing pLate
~~(OOrP~S-----------r------------r-----~-----r-----------.r-----------,
pJRx~Jt
~
~J_' ~'JW'$~ Figure 61. Severing processes (DIN 8588); 1 tool, 2 workpiece, 3 rest.
I
:w
Manufacturing Processes. 4 Cutting
BlaoJ(!
•
Blank 2
b
c
b Fipft til. Blanking: workpiece and tool tenninology. a workpiece [55]; 1 cut edges, 2 cut surface. b Tool; 1 tool, 2 cutting gap, 3 cutting edge, 4 flank, 5 pressure surface, 6 wedge.
mediate operation in metal forming. There is a certain affinity with metal forming processes in that the cutting
processes involve plastic defonnation. The previously common designation punching is no longer contained in the standard [54). In principle, the tool-related tenns contain the word cutting (cutting edge, cutting surface), while the workpiece-related tenns have the word cut (cut edge, cut surface), Fig. 62. Blanking processes are divided according to the nature of the line of cut into processes with a closed line of cut and those with an open line of cut (Fig. 6~). Whereas closed cutting is perfonned on presses using punches and dies, open lines of cut are produced not only with these tools but also with longitudinal and circular blades on special machines (cf. K3). The processes with a closed line of cut include cutting out and piercing (Figs 64a, b). Cutting out produces the complete external shape in one operation. Piercing produces an internal shape in the workpiece. The processes with an open line of cut encompass cutting off, notching, slitting and trinttning (Figs 64c to ().
Cutting off is the severing of a part from the mill product (sheet or strip) or the semi-finished product. Notching is cutting out parts of the surface at internal and external edges. Slitting is partial severing of the workpiece without removal of material. It is generally used as preparation for metal forming. Trimming is used to sever material attached to the workpiece that is not required on the finished product. A special position is occupied by nibbling (Fig. 64g). In this process, the workpiece is gradually severed along a line of cut of any desired shape by means of a simple punch.
f
:~\
15
Detail (shown with tool)
Figure 64. a Cutting out; 1 scrap, 2 blank. b Piercing; 1 workpiece, 2 scrap. c Cutting off; 1 workpiece, 2 line of CUl, 3 metal strip. d Slitting. e Notching; 1 workpiece, 2 scrap, 3 finished part. f Trimming; 1 ejector, 2 finished part, 3 scrap, 4 edge cutter. I Nibbling; 1 workpiece, 2 cut edges, 3 tool (55].
Fs
4.S.2 Technology Application of Force. On the impact of the punch, the vertical punch force Fs and, with increasing depth of cut, the horizontal force FH are produced (Fig. 6S). The resolution of the punch force Fs and the reaction force F ~ leads to the punch-side forces Fv and FH on the one hand and F~ and F~ on the other, which act on the blanking
•
•
llL---B---'1 b
Fipft tilt. Blanking: a open, b closed.
Fi....... tiS. Action of forces in blanlting (punching). explanations in text (54].
die. Owing to the distance I between the pOints of application of the forces, a moment is generated which causes the workpiece to bend_ Sequence of Blanking (Punc:hing) and Formation of Cut Surfaces. These depend on the tool geometry -
4.5 Shearing and Blanking. 4.5.3 Forces and Energies
I:• •
J
Figure 68. Full use of material in blanking [551; 1 scrap-free shapes.
Figure 66. Cutting process in blanking [54]; J punch, 2 die, 3 blank, 4 metal sheet, u cutting gap, SA height of edge indentation, g width of edge indentation, h height of zone of cut, i height of zone of fracture, b(; height of burr. tE tear depth, s sheet thickness, Fs cutting force.
cutting gap u (Fig. 66), rounding or dulling of the cutting edge - and on the material and the characteristics of the mill product - sheet thickness s, mechanical properties, chemical composition and metallographic structure. The sequence of blanking is characterised by the following phases: Fig. 67 [55]. Owing to the influence of the vertical force, elastic defortnation occurs first of all. The sheet bulges under the punch and partly lifts off the face of the die. Then the sheet undergoes local plastic defortnation, producing permanent bulging of the sheet. The edge indentation is formed in the upper side of the sheet and in the blank. In the next cutting phase the material is sheared off, producing the smooth-cut part of the cut surface. In the residual cross-section the tensile stresses increase, leading to the fortnation of the first incipient cracks starting from the cutting edge of the die. Further incipient cracks then form
in the sheet at the edge of the punch. At the moment of
cracking, the maximum actual shear stress reaches the shear fracture point, leading to cracking [55]. When punching out parts from sheet metal, the aim is to use as much of the metal strip as possible (Fig. 68). The beginnings and ends of metal strips generally produce additional scrap; therefore the aim is to punch out the blanks directly from the coil. A range of CAD systems are available which enable computer-aided optimisation of the cutting of the blank (nesting drawings). 4.S.~
Forces and Energies
Among the most important parameters for the design or selection of presses is the maximum actual cutting force. The maximum cutting force is influenced by sheet thickness, punch geometry, tensile strength of the sheet metal, tool wear and the cutting gap u (Fig. 69). It should be noted that the burr height h" depends on the cutting gap. The maximum cutting force is determined according to an empirical equation in which k, = O.8Rrn and A, = l,s: F,rn.. =A,k,.
Here, I, is the length of cut, s the sheet thickness and Rm the tensile strength of the material.
Details of the factors influencing the maximum cutting force are shown in Table 4. The maximum cutting force can be reduced if the effective line of cut I, is shortened. The engagement of the punches may also be staggered over time (Fig. 70). As a result of the horizontal forces between the sheet and the punch, withdrawal forces,
Punch travel in mm Figure 67. Sequence of events in blanking (punching) [55]; 1
Figure 6,. Cutting force
punch, 2 sheet, 3 die.
gaps u [551.
F~
12
versus punch travel for various cutting
I;.'
Manufacturing Processes • 4 Cutting
Table 4. Variables influencing cutting force
4.5.4 Workpiece Properties
Influencing variable increases
Max. cutting force F, or specific cutting force k~
Cutting gap u Sheet thickness S Punch diameter d s\ Tensile strength Rm Tool wear
k, falls k, falls k, falls Rule of thumb: k, = 0.8 . Rm F~ rises to 1.6 times original value
F, = {"''',
Is length of line of cut
Cutting force Fs
_~_~~2,3,4
The blanks may exhibit a range of defects (Fig. 71): the shape defects edge indentation, tear depth, and burr height. With parts having small outside dimensions compared with the sheet thickness, deviations from flatness may occur. Edge indentation is constant only in the case of circular parts; in projections with small radii, it may be as much as 30% of the sheet thickness [55]. The tear depth depends on the cutting gap and the material. Burr formation on blanks is a consequence of cutting edge wear and the reSUlting change in the course of the crack. Dimensional defects arise in the event of dimensional inaccuracies of tools and/or, with multi-stage tools, as a consequence of feeding errors. The positional defects, which are usually deviation from parallel alignment, are caused by incorrect relative positioning of the tool elements. This may arise from inaccuracies in tool manufacture, tilting and shifting of the press slide, or feeding errors in the case of multi-stage tools. Angle defects of the cut surfaces result from angular spring, which is particularly severe in the case of C-frame presses. Owing to the plastic deformation at the beginning of the process, an increase in hardness occurs directly at the cut edges. The depth of the increase in hardness and the area over which this occurs depend on the material. Various investigations show that, with steel sheets, an increase in hardness of 2.0 to 2.2 times the original hardness may result over a distance of 30 to 50% of the sheet thickness from the cut surface. On materials, see D3.1.4.
a 4.5.5 Tools
u
b
Figure 70. a Cutting-foree-travel curve in relation to shape of cutting edge (b). c Staggering of punch action over time (progressive blanking) [54).
which are influenced by the cutting gap, the punch dimensions, and the thickness and mechanical properties of the sheet metal, arise when the punch is withdrawn (Table 5). The cutting energy is influenced by the tool geometry and the workpiece properties to a far greater extent than the maximum cutting force. It decreases as the width of the cutting gap increases and increases as the sheet thickness increases.
Types. Blanking tools are described according to the type of guidance of the cutting elements relative to each other as open, plate-guided and pillar-guided tools (Fig. 72). They are suitable in the order stated for small, medium and large quantities (Table 6). However, the guiding precision of the press greatly influences the qual-
ity of the cut. Depending on the requirements, the blank is cut out of a strip of sheet metal in one or more stages. Accordingly, a distinction is made between single-stage and multi-stage blanking tools. Tools for combinations of blanking and forming operations are termed compound multi-stage tools. In a single-stage tool, all the cut surfaces are produced in one operation. This is generally possible with simple blanks. A fmished blank is thus produced at each stroke of the press. The precision of the blank is determined by the precision of the tool. For difficult parts with narrow land areas, the workpiece is generally fabricated in several stages in a multistage tool. The blank remains connected to the metal strip during its passage through the stages and is cut out only in the fmal stage. With multi-stage tools, the precision of the blank is determined not only by the precision of the
Table S. Relationships of forces in blanking Side force/cutting force Withdrawal force/cutting force Ejector force/cutting force
0.02 to 0.2
0.01 to 0.4 0.00;
Figure 71. Shape defects on cut blanks. Height of edge indentation SA, burr height h G, tear depth tE [55].
4.5 Shearing and Blanking. 4.5.6 Special Blanking Processes
,;.,
Table 6. Types of punches
Guide
Free punch
Plate-guided punch
Pillar-guided punch
Press ram
Guide plate
Pillar guide frames
Press ram
(DIN 9812. 9814, 9816. 9819, 9822)
Advantage
Cheap due to simplicity of design
No positional defects; little risk of Very accurate workpieces, low wear; no buckling of thin punches; high pOSitional defects; simple and cheap to set tool life
up
Disadvantage
Difficult to set up in press; high wear if installation inaCCUi,ltC
Blanking tool element (punch) is used as guide element
Displacement forces and tilting moments arise from off-centrc clamping. expensive tools required
Use
Small quantities
Fairly large quantities
Large quantities
a Figure 74. Internal die face shapes. See text for explanations [55J.
"
Figure 77.. Types of blanking tools. a Free punch; 1 punch, 2 die, 3 bolster plate. b Plate-gUided punch; 1 punch guide plate (stripper), 2 guide strip. c Post-guided punch; 1 top, 2 guide bushing, 3 guide post, 4 stripper.
tool but also by the accuracy of the strip feed. This is ensured by means of pilot punches or pins [56].
Tool Position in the Press. Wherever possible, the tools should be positioned in such a way that the resultant
of the individual forces passes through the middle of the press. In this way, moments caused by eccentric loading and resulting defects in the precision of the workpiece as well as increased tool wear are avoided. In press design it is assumed that the resultant force acts at the centre of gravity of the lines of cut. The cutting gap, which influences the formation of the cut surfaces and the path of the cutting force, is established according to the requirements applicable to the cut surface - appearance, precision, further processing, and function [57]. For approximate values see A.ppendix K4, Table 8.
Blanking Elements. The punches are designed both to exert pressure and to prevent buckling (in piercing). Some punch designs are shown in Fig. 7~. The inside
"
Figure 730. Designs of piercing punches and punch guides.
faces (Fig. 74) of dies should be at less than 90° to the bearing surface if the blanked part has to be ejected in the direction opposite to the direction of cutting. Otherwise, clearance angles in the range 10:5 a :5 45° are usual, depending on sheet thickness. The height of the 90° inside face (Fig. 74b) is between 2 and 15 mm. At the design stage, a means of regrinding the blanking elements should be provided. For materials for blanking tools, see A.ppendix K4, Table 9.
4.5.6 Special Blanking Processes If flat parts with smooth, crack-free cut surfaces and high
dimensional accuracy are required, the blanks have to be either fmished or cut out by means of special processes.
Reblanking. This process forces the workpieces to be reblanked through a die whose inside dimensions are smaller than the workpiece by approximately twice the thickness of the layer of material to be peeled off (Fig. 75). Precision Blanking. The distinguishing feature of this process is that immediately before the workpiece is blanked, a knife-edged ring is pressed into the sheet a short distance away from the line of cut from one or both sides depending on the sheet thickness. During blanking, an ejector serving as a pressure pad prevents the bulging
.m .m
Figure 7S. Reblanking of cut parts according to [58]. a Punch smaUer than die. b Punch larger than die.
Manufacturing Processes a 4 Cutting
•
Figure 76. Sequence of events in precision blanking (58]. a Orig· ina! position. b Pressing of knife-edged ring. c Blanking with oppos· ing force exerted by pressure pad. d End of blanking. e Sheet stripped and blank ejected. FN Knife-edged ring force, FG pressure pad force ,
FSI
punch force.
81f1 a
Pa~ial
culling
b
d Fipre 78. Upset blanking 1551 . a Contact of punches; J sheet, 2 blanking punch, 3 upsetting punch, 4 die, 5 ejector. b Upset blanking. c End of upset blanking. d Cutting out; 1 scrap, 2 blank (workpiece).
Culting I rough
IlIlfJlfI b Parlial Culling
Cutting in opposi e direction
OJttmg tIIrough
c:
Fipre 77. Counterblanking 1561: a two-stage, b three·stage, c cut surface.
of the blanks that usuaUy occurs in normal blanking (Fig. 76) [59J. The effect of the knife-edged rings is imponant: they generate compressive stresses in the shearing zone at right angles to the direction of cut. In this way the proponion of the cut surface that undergoes plastic deformation by shearing is increased, improving the precision of the cut. A triple·acting press is required for precision blanking, owing to the knife·edged ring force and the pressure pad force that have to be exened in addition to the cutting force . Materials generally suitable for precision blanking are aluminium and its alloys, copper, brass with a copper con· tent '" 63%, plain steels containing :S 1% C, case·hardened steels, low-alloy hardened and tempered steels, and ferritic and austenitic stainless steels.
Counterblanking. Here, two or three blanking stages acting in opposing directions are employed (Fig. 77). In the fu-st stage, a panial cut is made 10 the point just before cracking begins. The second stage, performed in the opposite direction, also puts a stripped edge on the other
Fipre 79. High-pressure water jet cutting; / cutting nOlzle, p pump pressure, dnol. nozzle diameter, d depth of cut, a gap between nozzle and workpiece, s workpiece thickness, W kerf width, v rate of feed 155).
side of the workpiece. Sometimes the workpiece is again only panly cut in this second stage and not cut through until the third stage. The advantage of this process is that burr-free cut surfaces are formed on both the outer and the inner blank and both can be put to use. For thiS, either a single-stage or a multi-stage tool is generally required.
Upset Blanking. Upset blanking enables smooth, burrfree cut surfaces to be produced. First, the sheet is sheared off with a hollow outside punch as in Fig. 78. The remaining sheet thickness is then cut with the actual blanking punch and lastly the cut workpiece is expeUed with the ejector. The process is also suitable for cutting phenolic or epoxy resin laminates and glass-fibrereinforced plastics. let or Be.... Cutting. In these cutting processes, the material is removed along a designated workpiece contour by the effect of an active medium concentrated into a jet or an active energy source supplied in the form of a beam. The processes employing an active medium use a beam having mass as the "tool" (plasma cutting and water jet cutting, Fig. 79), whereas the processes employing active energy use a quasi-massless beam (laser cutting).
5.1 Thread Production. 5.1.3 Thread Cutting with Dies
':.'
Special Technologies 5.1 Thread Production G. Spur, Berlin
5.1.1 Single·Point Thread Turning Single-point thread turning is a spiral turning operation to produce a thread using a single-point cutting tool. Both external and internal threads can be produced. When using a universal lathe, the feed drive is accomplished via a lead screw, Figure 2. Shape of shank-type form-cutting tool for different thread pitches: a for small thread pitches; b for large thread pitches.
where nw = the rotational speed of workpiece, P w = the thread pitch of workpiece, nl. = the rotational speed of lead screw and PI. = the pitch of lead screw. Conventionally controlled automatic lathes are equipped with guide devices (feed curves, guide collets) to start the feed motion. On numerically controlled lathes the feed drive is separated from tbe main drive. The feed motion is computer-controlled and is accomplished with the aid of a recirculating ball screw. As the thread is produced in several stages, the tool has to be advanced repeatedly at the same point on the circumference of the workpiece. For this purpose the angular pOSition of the main spindle is recorded with mechanical devices on conventionally controlled machine tools and by means of digital transducers on numerically controlled machines. The thread turning tools correspond to the thread profile. There are Shank-type furm-cutting touls, round form-cutting tools and relieved round form-cutting tools (Fig. 1). The side rake angle of the tool is generally y, = 0°; the face is adjusted to the centre of the workpiece. To obtain the necessary side clearance angle of u, = 6 to go for a round form-cutting tool, the centre of the tool lies a defined distance h ahove the centre of the workpiece (Fig. Ib). At this point the round form-cutting tool has to have the desired thread profile. In the case of the relieved round form-cutting tool, the required side clearance angle is produced by the relief (Fig. le:). Regrinding is carried out on the face, which is positioned radially relative to the tool axis. The effective side clearance angle is arc = 3 to 5° (Fig. 2). It depends on the thread pitch. A symmetrical ground surface is adequate for smaller pitches (Fig. 2a). For larger pitches the shank-type form-cutting tool should be ground at differing angles (Fig. 2b). This produces different effective side wedge angles 13k , and 13k " which may lead to unfavourable cutting conditions. To avoid this, the shank-rype form-cutting tool is positioned at an angle (Fig. 2b). Profile distortions arise. however, which have to be compensated for by contouring the shank-type form-cutting tool accordingly. Precision threads are pro, duced with two cutting tools which each machine one flank.
Figure~. Thread chaser: lA bevelled length, vj' feed rate, bs chaser width, P pitch, 1 workpiece, 2 chaser.
5.1.2 Thread Chasing Thread chasing is turning to produce a thread with a tool having multiple cutting edges in the feed direction. The chaser is usually bevelled at the leading side (Fig. ~). It may be engaged either radially or tangentially. Chasing is frequently employed on turret and automatic lathes with the use of a guide device. The chaser is guided by a guide collet or a chasing curve. Single- and multiple-start internal and external threads can be chased. Several thread pitches may be turned with the same chasing curve by means of a change gear train [1]. S.l.~
Thread Cutting with Dies
This is spiral turning to produce a thread using a tool with multiple chasers. Solid (non-opening) and opening dies are generally used for thread cutting by hand and for threads with low precision requirements. Solid dies (Fig. 4) may
~.~ a b e :
Figure 1. SingJe-poim thread-turning tools: a shank-type form-cutting tool; b round form-cutting tool; c relieved round form-cut-
ting tool
Figure 4. Threading die in a stock: 1 threading die, 2 stock.
.;.:1
Manufacturing Processes • 5 Special Technologies
Figure S. Shape of cutting edges of a tap: O'p back dearance angle (flank relief angle), O'pi back clearance angle on tapered end, /3p back wedge angle, YP back rake angle, h relief, hi relief on tapered end, d_, diameter of tapered end, 14 length of tapered end. K centre diameter.
Figure 7" Long-thread milling: 1 tool, 2 workpiece.
S.l.S Thread Milling be slotted or closed. Opening dies usually have four radially or tangentially positioned chasers. These are adjustable and interchangeable for different thread diameters and pitches and can be opened after cutting, so that the unscrewing necessary with solid dies is dispensed with. For series production, self-opening dies with radially or tangentially adjustable chasers are used. If the die hits a stop which marks the end of the thread length, the chasers open automatically.
S.1.4 Tapping Tapping is spiral drilling with a tap to produce an internal thread. The geometrical cutting edge shape of a tap is shown in Fig. S. Taps are relief-ground to reduce the work done by friction. The back clearance angle (relief angle) is "p = 1 to 5°; the back rake angle Yp is 0 to 3° for grey cast iron, 3 to 15° for steel and 12 to 25° for aluminium alloys. The chamfered end of the tap performs most of the cutting, while the rest of the tool is mainly for guidance and is slightly tapered (1 : 1000). Hand and machine taps are used. Hand taps consist of a set of several taps, selected according to the material to
be cut. Sets of three taps are normally used. The cutting work is split between the individual taps as follows: taper tap 50%, plug tap 30% and bottoming tap 20%. Machine taps are normally used as single-cut taps. Poor chip removal and the resulting risk of tool breakage necessitate low cutting speeds in machine tapping. Figure 6 shows common designs of machine taps. Shortchipping materials are cut mainly with straight-flute taps (Fig. 6a). Curling taps (Fig. 6b) achieve better chip removal when machining through holes. Tapping in sheet metal is carried out with taps having short flutes (Fig. 6c). Taps with sharply helical flutes (Fig. 6d) promote good chip removal in bottoming operations with small fUnouts [2J.
Figure 6. Tap shapes (see text for explanations).
Long Thread Milling. Here, the length of thread that can be produced is independent of the tool (Fig. 7). Disc-shaped, relieved proftie milling cutters are used, the profile of which has to be adjusted in the case of large pitches. The axis of the milling cutter is angled in relation to the workpiece axis according to the thread pitch. If the partial thread angle is less than 10°, profile distortions arise from the lateral free-cutting of the milling cutter. Either upcut or downcut milling may be used. Long thread milling is used with longer threaded spindles. The cutting speeds of relieved profile milling cutters made of highspeed steel are v, = 4 to 20 m/min for steel, depending on its tensile strength [2J. Thread whirling is another method of long thread milling. The principle of thread whirling - also known as thread peeling or fly milling - is illustrated in Fig. 8. One to four cutting tools with their tips pointing towards the centre of rotation of the holder rotate around the workpiece on an eccentric orbit. The whirling head is inclined
in relation to the workpiece at the pitch angle. The tools may be positioned radially or tangentially. Such thread whirling devices can also be used on centre lathes. In thread whirling with cemented carbide, the cutting speed for steel, depending on its tensile strength, is v, = 100 to 125 m/min l3J. The peripheral speed of the workpiece is between 0.5 and 4 m/min. Thread whirling may be performed in either the upcut or the down cut direction. For internal threads the required orbit should be 2 to 5 mm smaller than the minor diameter. Internal threads and external threads with small thread depths are cut using only one tool.
Figure 8. Principle of thread whirling.
5.1 Thread Production. 5.1.8 Thread Rolling
,:.'
several thread proftJes next to one another on the grinding wheel according to the thread pitch. These proftJes are stepped on the chamfer of the grinding wheel. The feed distance depends on the length of the thread and the width of the grinding wheel. Threads with a collar cannot be produced. The pitch range is P = 0.8 to 4 mm.
Figure 9. Short-thread milling. 1 tool, 2 workpiece.
Short Thread MilliAg. This is done with roll-shaped thread milling cutters (Fig. 9). These have adjacent thread proftJes with no pitch, the spacing of which corresponds to the thread pitch. One tool can only produce threads of the same pitch, but to various diameters. During about 1/6 of a workpiece revolution, the milling cutter is radially advanced to the required thread depth and displaced axially during a further workpiece revolution. Large-proftJe multiple start threads, e.g. wOrtnS, may be economically manufactured by hobbing. The tool moves on rolling contact along a line parallel to the axis of rotation at the periphery of the workpiece. 5.1.6 Thread Grinding The three most important thread grinding processes are illustrated in Table 1.
Longitudinal Spiral Grinding. The thread starts are ground in succession using a single-profile grinding wheel. The grinding wheel is inclined in relation to the workpiece axis according to the thread pitch. Every possible pitch can be ground. The small machining force components promote the high degree of precision achievable, but the grinding time is relatively long. In multiple-profile longitudinal spiral grinding, there are
Transverse Spiral Grinding. In multiple-proftJe transverse spiral grinding, the grinding wheel is fed inward to the full thread depth during a one-quarter revolution of the workpiece. The thread is finished during a further revolution with simultaneous axial displacement by the amount of the pitch. The machining force components are relatively large. Only thread lengths up to about 40 mm can be ground. The pitch range is P = 0.8 to 4 mm. Multiple-profile transverse spiral grinding results in the shortest grinding times. Centreless thread grinding may be perfortned by the throughfeed method or the transverse grinding method. The proftJed grinding wheel is pivoted according to the helix angle of the pitch diameter. The grinding wheels have to be dressed to produce the desired thread proftJe at the circumference. Single-proftJe grinding wheels are dressed with diamond dreSSing tools with a single layer of grains, while multiple-proftJe grinding wheels are dressed with shaped rollers made of hardened tool steel or with rotating shaped diamond rollers. 5.1.7 Electro-discharge Machining of Threads Electro-discharge machining of threads is used for difficultto-cut materials, usually for producing internal threads. The tool electrode, made of brass, copper or steel, bears the thread profile and screws itself into the workpiece, which has usually already been centre-drilled.
5.1.8 Thread Rolling In thread rolling with flat dies (Fig. lOa), the pair of dies carries the opposite proftJe to the thread with the helix angle of the thread. One die is fixed and the other is movable. The workpiece is rolled between the two dies under the effect of friction forces, fonning the thread over its entire circumference. The rolling dies have a chamfered lead-in and lead-out and a straight grooving section.
Table 1. Working methods and accuracies in thread grinding [4]
Transverse spiral grinding with multiple-profile grinding wheel
~ Feed distance If Workpiece revolutions iw Grinding wheel width bs Accuracies: Pitch diameter Partial thread angle
,z 11m ,5'
Finishing , 4IC 5 11m Rough-working ,10 to 15 f1m ,5'10 10'
Pitch over a length of 25 mm
,z te 3 f1m
,5 f1m
Pitch over a length of 300 mm
,511m
,10 11m
,lOw 20 f1m
,10' ,51lm
[Z5 mm wheel Width 1
':.,'
Manufacturing Processes. 5 Special Technologies
cooling lubricant used [6J. Lubricating channels may be cut in the tools to improve cooling lubricant delivery.
S.1.10 Thread Pressing Thread pressing (Fig. 12) is mostly used to produce cylindrical threads in fairly thin sheets. The thread is pressed into the workpiece with two shaped rollers.
5.2 Gear Cutting M. Week, G. Mauer and W. Reuter, Aachen S.2.1 Cutting of Cylindrical Gears Figure 10.. Thread explanations).
rolling
processes
[5]
(see
text
for
Thread rolling with roller-type dies (Figs lOb-d) may be carried out with infeed or through-feed of the blank. Eitber short or long threads may be rolled, even right up to the collar. For infeed rolling, the rollers have the required thread profile with the same helix angle but the opposite direction of twist. The workpiece is either held with a guide (centreless) or clamped between centres. The workpiece remains axially at rest during rolling, apart from slight compensating movements. As the roller-diameter-workpiece-diameter ratio and the number of turns have to be precisely matched, a pair of rollers can be used only for a specific thread. For through-feed rolling, the rollers have adjacent pitchless thread profiles. They are pivoted about their horizontal longitudinal axis through the required helix angle. In this way the axial feeding of the workpiece by the depth of pitch is accomplished in one revolution. The rollers may also be used to a limited extent for various workpiece diameters. This method, though, results in a lower pitch accuracy than does infeed rolling. Self-opening thread rolling heads usually have three rollers. They are advanced axially towards the workpiece and are automatically pulled onto tbe workpiece by the inclined position of the rollers. When the required thread length is reached, the rolling head opens automatically and can be withdrawn.
S.I.9 Thread Forming Thread forming is the indentation of a thread into a workpiece using a tool witb a helical effective surface. The process (Fig. 11) resembles tapping in terms of kinematics, however the tool has no flutes and its cross-section is shaped like a rounded polygon with three or more forming webs. The torques to be exerted are considerably higher than in tapping and depend heavily on, among other things, the diameter of the forming hole and the
Fundamentals Figure 13 gives a survey of gear manufacturing processes. Initial gear cutting is mainly carried out by hobbing, generating by shaping, and broaching. Generating by planing may be used to cut large gears. Finishing before heat treatment is performed by shaving. Finishing processes after heat treatment are generating by grinding or form grinding. The processes are divided into form-cutting processes and generating processes (Fig. 14). Form-Cutting Processes. The tool (disc or end milling cutter, shaping tool, broach, grinding wheel) has the profile of the tootb space. The spaces are produced individually. To machine the next tooth space, the gear blank is turned through the circular pitch (individual indexing process). For each gear blank design (number of teeth, modulus, meshing angle, helix angle, addendum modification, and tooth adjustment) an appropriate tool profile is required (cf. F8. 1). Tool profiling for helical gear cutting is complicated, as the line of contact between tbe gear blank and the tool is a three-dimensional curve that cannot be derived from the transverse profile of the gear blank in a simple manner (it also depends on the tool
diameter). The grinding wheel profile has to be calculated with a computer.
Generating Processes. A rolling motion is produced between the gear blank and tbe tool during machining by kinematic linkage (self-contained gear train, electronic control loop). The flank shape (involute) is formed as the envelope of the straight-sided flank tool edge (Fig. IS; cf. FS.1.7). The involute shape is generated by rolling coupling of linear motion (translatory component of generating) with the rotation of the workpiece (rotational component of generating). The tools used are: hobs, rackshaped cutters, disc and taper grinding wheels, grinding worms. With the involute profile, the tools have a more universal range of uses than with the form-cutting process (independent of the number of teeth, the helix angle or the addendum modification). Continuous generating is possible witb a worm-shaped tool (hob, grinding worm) or a gearwheel-shaped tool (cutting wheel, peeling wheel, shaving wheel). Rack-shaped cutters and disc, flat or taper grinding wheels are used to machine one or more
tooth spaces (generating with indexing). When the mesh-
Figure 11. Thread forming: 1 thread forming tool, 2 workpiece.
Figure 121 .. Thread pressing: 1 pressing rolier, 2 workpiece.
5.2 Gear Cutting. 5.2.1 Cutting of Cylindrical Gears
l
, ••
)
y 'Heat treatment
Figure 131. Gear manufacturing processes.
Figure 14. Cylindrical processes.
ing zone has been machined, the workpiece is turned through one or more circular pitches and the generating operation is repeated (reversing).
Form Milling
Uses. Large-pitch or large-diameter gear blanks, gear blanks with non-generatable profiles, gear blanks with large gear cutting tolerances and for initial machining; also in the manufacture of single units (spur teeth can be produced on conventional universal milling machines with an indexing attachment).
gear
manuf3cturing
Machine. The tool motor directly drives the form milling cutter. Precise indexing equipment is required. For helical teeth, the rotary motion of the gear blank is derived from the feed motion of the tool (generation of spiral motion in the gear blank·specific coordinate system). The spiral motion depends on the helix angle of the gear blank.
Tool. End or stde-and-face milling cutters (also fitted with cemented carbide cutting tips). In profiling the tool, the grinding wheel is guided by a template, a cam mechanism or numerical control. The cutting capacity is high, because the tool cuts along the entire length of the profile.
':.'.1
Manufacturing Processes. 5 Special Technologies
m'
a
I
I
-~-----
f---
Spur gears
~ \'
r
I
~-----,
\I
-------;
Skew,tooth rack if there is limited space beside the teeth
Helical gears Figure 16. PrinCiple of generating by planing; 1 cutting motion, 2 feed motion, 3 reversing motion. Figure IS. Production of the involute-shaped tooth flank. a Theoretical production principle. b Production principle in machine; 1 tooL cutting edge, 2 tooth flank, 3 dedendum circle radius, 4 gear blank axis, 5 rolling feed, 6 base circle radius, 7 pressure angle, S line of action, 9 pitch circle radius, 10 rotational component of rolJing.
Generating by Planing Principles. An oscillating rack-shaped cutter rolls with the gear blank (Fig. 16). The cutting motion is accomplished by the reciprocating motion of the rack. The feed motion is accomplished by radial plunging or tangential feeding of the rack and the rolling motion of the blank. During the return stroke, the rack is lifted off the blank. The rolling feed takes place cyclically at the top of the return stroke of the tool. Only one group of teeth can be machined at anyone time. The rack is next drawn back and the gear blank then indexed by the corresponding number of teeth (group generating process with indexing).
Gear Shaping Principles. The rotation of the gear-shaped tool (cutting wheel) is matched to the rotation of the gear blank by a kinematic linkage so that both elements revolve like a pair of cylindrical gears (Fig. 17). The cutting motion is accomplished by the stroke of the cutter. The feed motion is accomplished by radial feeding up to the depth of cut and by a roling motion (rolling speed in relation to stroke frequency).
Uses. Initial cutting and fmishing of large-size spur and helical gears of high-tensile material and large tooth width. Machine. The rack-shaped cutter is attached to a slide which can be pivoted in the helix direction of the gear teeth. The rolling motion is accomplished by means of a cradle and a turntable. The rolling feed is transmitted to the cradle and the turntable by the ram stroke motor via the rolling feed crank and the rolling gears. With smaller machines, the tool travel is accomplished by means of a crank mechanism. Tool. Involute gears are cut with a relieved, straightsided flank involute rack; the rake angle is adjusted by hollow grinding of the face. The tool is cheap and is easy to manufacture with a surface grinding machine. The wear rate is low and tool changing is simple.
TM
Figure 17. Principle of gear shaping. Cutting wheel in engagement with straight-toothed gear blank, H tool travel, RT tool rotation, WPR workpiece rotation, TM helical tool motion in cutting of helical teeth. a PrincipJe. b Spur gears. c: Helical gears.
5.2 Gear Cutting. 5.2.1 Cutting of Cylindrical Gears
On the return stroke of the cutting wheel, a lift-off motion has to take place to prevent collision between the continually revolving cutting wheel and the gear blank (the flank of the gear blank is not yet fully profiled and therefore cannot roll, continuous radial feeding takes place during the stroke). The direction of lift may be determined with reference to tbe gear blank flank by tangential offsetting of the cutter and gear blank axes. For cutting helical gears, a spiral motion corresponding to the helix angle has to be superimposed on the stroke.
Uses. These are the initial cutting and finishing of internal gear teeth, teeth with too small an axial idle travel distance of the tool for hobbing (collar following teeth, step wheels), and short teeth. Machine (Fig. 18) Mechanical Drive. The rolling drive drives the cutting
wheel and the gear blank. The kinematic coordination of the rolling feed of the cutting wheel with that of the blank is accomplished by the table change gears. A separate reciprocating drive generates an oscillating ram motioo. The lift-off motion is controlled via the reciprocating drive and the lifting cam. NC Drive. Each axis has a separate drive. The rolling link-
age is accomplished electronically in the control system. The lifting cam is also linked to the stroke. The spiral motion is produced mechanically via an inclined guidance sleeve.
Tool. The cutting wheel is made of high-speed steel (HSS) or coated TIN. The helix angle of the cutting wheel depends on the gear hlank helix angle. For cutting helical
I;.'
tool (hob) so that both elements rotate like a wornl and wheel. Through the additional superimposition of a feed motion (axially, radially, radially-axially, tangentially or axially-tangentially to the gear hlank cylinder) the hob cuts material from the tooth spaces. Figure 19 illustrates the engagement of the hob with the blank. A rolling motion is produced by superimposition of an imaginary tangential motion of the hob teeth on the rotational motion of the blank (during a revolution of the hob, tangentially offset cutter teeth of the worm start to engage in succession). The gear blank profile is composed, in polygonal fashion, of enveloping cuts. Axial Method. The hob feed direction is axial to the gear
blank cylinder. This is the most common method of cutting cylindrical gear wheels on a hobbing machine. Diagonal Method. The hob feed direction is both axial and tangential to the gear blank cylinder. It is used to cut
cylindrical wheels. The tangential motion of the cutter mUSl be balanced by additional rotation of the gear by an equal amount (analogous to the rack rolling with the gear blank). Tangential Method. The hob feed direction is tangential to the gear blank cylinder. It is used for cutting worm wheels. The kinematics are as in the diagonal method (no axial feed). Radial lfJethod. The hob feed direction is radial to the
gear blank cylinder. It is used for cutting the teeth of worm wheels and very thin wheels.
teeth, the face is often ground in such a way that it is perpendicular to the helix direction (step grinding). The tlanks of the cutting wheel teeth are relieved in such a way that the cutter can generate the desired gear profile in every transverse position (i.e. after every regrinding of the face).
Gear Robbing Principles. The rotation of the gear hlank is matched by kinematic linkage to the rotation of the worm-shaped
2
Section A-B Figure 19. Terminology for hob-workpiece pair. Gear: d 2 gear Figure 18. Diagram of mechanism of a gear shaping machine: I reciprocating drive, 2 rolling drive, 3 turntable change gears.
4 radial feed, 5 lift-
7
reciprocating
diameter, z-'- number of teeth,
j3~
helix angle, b gear width. Hob·
hob diameter, Zo number of starts, Yo pitch angle, e axial pitch, i number of teeth. Machining: Tj pivoting angle (YJ = f32 == 'Yo), /" axial feed, T depth of cut. d~o
'*il
Manufacturing Processes. 5 Special Technologies
To produce helical gear teeth, the axial feed must be linked to the rotation of the workpiece and the hob. To achieve this, the feed motion is returned to the table gear train via the differential gear and the differential. The outer cage of the differential, which is stationary during cutting of spur teeth, moves in line with the translations in the differential gear train and superimposes an additional motion on the table gear train. Depending on the pitch direction of the gear blank teeth, this results in an increased or reduced table speed to produce helical teeth on the gear blank.
Figure 1:0. Gear train of a hobbing machine; J hob. 2 main motor, 3 gearbox, 4 differential, 5 indexing change gears, 6 differential change gears, 7 feed spindle, 8 feed gearbox, 9 indexing worm gear.
Uses. Initial cutting of automotive gears in series production, initial cutting and finishing of soft, quenched and tempered and hardened (with a cemented carbide peeling hob) large gear teeth up to approximately 4000 mm outside diameter, also initial cutting and finishing of worm wheels and special teeth (axial compressor rotors, serrated or splined shaft teeth, chain wheels). Machine (Figs 20 and 21)
Conventional Drive. All the motions are derived from the main motor. The rotational speed of the hob (cutting speed) may be varied by means of a steplessly adjustable drive. The rotation of the hob and the gear blank are linked by means of the indexing change gear train. The relationship of the gear blank rotation to the cutter rotation is the same as the relationship of the number of starts of the hob (number of teeth) to the number of teeth of the workpiece. This adjustment is made by means of the indexing change gears. The axial feed is taken from the worm shaft of the table drive mechanism. This feed is steplessly adjustable.
NC Drive. Figure 22 shows a diagram of the drive system of an NC gear hobbing machine. All the axes are driven by separate motors. The kinematic linkage is accomplished by means of the control computer. From the information supplied by the hob angle measuring system and, for helical gear teeth, by the axial drive measuring system, the setpoint value for the table rotation is calculated in the computer. Comparison with the information on the workpiece motion gives a difference which is compensated for via the table drive. Tool. For cutting involute gear teeth, the enveloping surface of the hob is an involute screw with straight-sided flanks which is broken by flutes at right angles to the involutions. The teeth are relieved so as to form flanks at the tips and sides of the teeth which permit regrinding of the face at a constant rake angle and tooth profile (radial regrinding). Tooth profiles are standardised as reference profiles (i.e. the normal section of the rack) in DIN 3972.
Solid Hobs. These are made in one piece from high-speed steel (HSS) with or without TIN coating. For tilting-tooth hobs, the cutting teeth are ground in ancillary devices as involute screws and then tilted into the working position in the tool body, producing clearance angles at the tip and sides. The teeth are made of HSS or cemented carbide, while the body is made of tool steel. Cutter Plate Hobs. The teeth are relieved in the body as with solid hobs. The hob tooth length is extensively regrindable owing to back supports. The teeth are made of HSS or cemented carbide, while the body is made of tool steel. The stress on the cutting edge varies along the zone of engagement between cutter and gear blank, so there is no uniform distribution of wear in the longitudinal direction of the hob. This is remedied by gradual tangential shifting of the working range of the hob when it is out of engagement after the permitted maximum width of wear land has been reached, or by continuous tangential shifting by the diagonal method. Gear Finishing
IndexWl9 gear Irain AxIal differeolial gear lIain Figure 2:1. Generating module of a conventional hobbing machine; I axial cradle, 2 flywheel, 3 hob, 4 workpiece table , 5 axial feed change gears, 6 differential change gears, 7 indexing change gears, 8 differential, 9 main motor.
Gear fmishing is carried out by sbaving if the teeth are in a soft condition (before heat treatment) and by grinding, bob peeling or peeling gear shaping. The main purpose of fme machining is to remove the geometrical deviations on the gear blanks such as envelope cutting and feed markings (Fig. 23). The making of tooth flank corrections is growing in importance. As Fig. 24 shows, relieving the tip zone or making corrections in the direction of tooth width can improve the running and stress behaviour of the gears. Topological corrections can be used to make specific adjustments in the dynamic running characteristics.
Shaving with a Shaving Cutter
Principles. The initially cut gear blank rolls with a gearshaped tool (shaving cutter) with intersecting axes and
5.2 Gear Cutting. 5.2.1 Cutting of Cylindrical Gears
Figure 22. Electronic generating module of an NC hobbing machine: 1 angle measuring system Z, 2 axial drive, 3 angle measuring system B, 4 hob drive, 5 angle measuring system C, 6 table drive, 7 input data, 8 booster, 9 regulator, 10 subtraction, K machine constant, f3 helix angle, mIl nonna! modulus.
b
Uncorrected, tooth flanks
Correct transmission of motion
Correction of tooth width
Improvement of load-bearing characteristics, compensation of· directional errors
Figure 2~. Deviations from the enveloping cut in initial gear cutting: a hobbed tooth flanks, b tooth flanks generated by shaping; 1 deviations from the enveloping cut, 2 axial feed
without kinematic linkage. The shaving cutter drives the gear blank. Intersection of the axes (conditions corresponding to a spiral toothed gear train) gives rise to sliding of the shaving wheel tooth flank on the gear blank tooth flank in vertical and longitudinal directions of the tooth. On the flank of the shaving cutter are cutting teeth. If the flank of the cutting wheel runs over the flank of the gear blank with application of force . chip removal takes place. The feed motion is axial (parallel shaving), tangential (transverse shaving), diagonal (diagonal shaving) or radial (plunge shaving) relative to the gear blank cylinder. Figure 2; illustrates the shaving cutter engaging with the gear blank. Uses. Fine machining of soft spur and helical gear teeth after initial cutting, series manufacture of autonlotive gears, improvement of surface roughness and tooth-cutting errors, and reduction of distortion due to hardening by initial correction of the gear blank flanks.
Relief of base and tip
Reduction of meshing jerk
Topologically corrected tooth flank
Improvement of dynamic running characteristics
Figure 24. Correction of tooth flank geometry.
':.41
Manufacturing Processes. 5 Special Technologies
'i;
Form-cutting method
RA I
I
Conditions of contact
Principle
Method
~ Line ~
(
I
-
~
I
~.
Generaling met1hod ;' wilh indexing . .-
\
':f"
."' ~
IA Figure 2:S. Shaving wheel SW meshing with helical-tooth gear blank GB (after Buschhoff K. Thesis, RWTH Aachen, 1975). RGB rotary motion of gear blank, RT rotary motion of tool, AX axial direction relative to gear blank, TA tangential direction relative to gear blank, RA radial direction relative to gear blank, OJ diagonal direction relative to gear blank, CT cutting tooth.
Fonn GriAding. The conditions arc similar to those of form milling. The grinding wheel is dressed to the gear blank profile according to the desired line of contact. This is easy with spur gear teeth but complicated with helical gear teeth. Figure 27 iIlustrates the possible grinding wheel positions. The advantages of the process are a high metal removal capacity due to line contact over the full width of the flank and minimal profile errors (no kinematic rolling errors during machining). It is suitable for internal gear teeth. Generating by Grinding with Indexing. The grinding wheel flank embodies the flank of an imaginary rack which rolls with the gear blank flank to be generated (conditions analogous to generating by planing). Figure 28 shows a machine for generating by grinding with indexing with a double-conical grinding wheel. From the kinematic point of view, this machine may be compared with a machine for generating by planing. The rolling feed takes place continually during grinding. It is a general-purpose machine, used mainly in small- and medium-scale series production and for grinding of large gear teeth.
Besides the already mentioned generation of the rolling motion by means of a self-contained rolling gear train, in gear tooth grinding the rolling motion may be generated by means of pitch blocks, rolling belts and cradles.
Continuous Generating by GriAding. The contlitions are similar to those of hob bing, but the hob is replaced by a grinding wheel of larger diameter, the outer
w"
GeneraUng method with indexing
Gear Tooth GriAding Principles. Finishing of mainly hardened or quenched and tempered gear teeth (improvement of surface roughness and tooth cutting errors, elimination of distortion due to hardening, and corrective grinding). By analogy with the initial gear cutting methods, the grinding methods are classitied into form-cutting methods, continuous methods and generating methods with indexing (Fig. 26). The motions correspond in principle to those of the initial cutting processes. Owing to the lower cutting forces and higher cutting speeds, machine designs have to be adapted to these circumstances. Corundum and CBN grinding wheels are used as tools.
LD +'
Two-
One-point contact
.
Generating method With IndeXing
contact
~ ~_ I
'. \.
.
...
.
__
...,
Point and line contact
~ n-Point contact
Continuous generating process
Figure 26. Types of gear tooth grinding.
moo
GB~ SG
SG
GB Figure 2:7. Position of form grinding wheel relative to gear teeth (after Dudley DW, Winter H. Zahnriider. Springer, Berlin, 1961): DG double-flank grinding wheel, SG single-flank grinding wheel. GB gear blank.
shell of which is dressed in the form of an involute screw. Figure 29 shows the machine construction with six NC axes. The tools are corundum or CBN grintling wheels with a flange-mounted polishing worm. By means of the diagonal method (simultaneous axial and tangential
5.2 Gear Cutting • 5.2.3 Cutting of Worm Gears
':."
Flank Shape K The axial profile of the disc-shaped, tap-
ered rotary tool lies in the normal plane. Owing to the three-dimensional line of contact between the tool and worm flanks, the axial profile of the tool is not reproduced in the normal plane of the worm; in the case of a tool profile with straight-sided flanks, therefore, the normal profile of the worm is convex. Flank Shape I This corresponds to a helical-toothed cyl-
indrical wheel, with involute profile in the transverse plane. The flanks of the worm are produced by form cutting or generating.
Form Milling and Form Grinding with Discshaped Tool
Figure 28. Machine for generating by grinding with indexing, with double-conical grinding wheel: 1 helix angle adjustment, 2 stand, 3 infeed, 4 change gear box, 5 machine bed, 6 workpiece slide, 7 table feed, 8 workpiece table, 9 table rotation, 10 grinding slide, 11 tool travel, 12 grinding slide support.
The contlitions are the same as in the cutting of helicaltoothed cylindrical wheels. If a tapered tool profile with straight-sided flanks is used, only flank shape K is possible (Fig. ~l). The other flank shapes are possible if the tool profile takes into account the conditions of contact with the worm flank. With a flat tool, flank shape I is possible if the tool axis is pivoted in the normal plane of the worm and tilted through the generating angle.
Form Turning To produce flank shape A or N, the trapeZOidal turning tool cutting tips are guided in the axial or normal plane of the worm with an axial motion linked to the rotation of the worm (generation of a spiral motion in the coordinate system of the workpiece) (Fig. ~2). Flank shape I is produced if trapezoidal turning tool cutting tips lie in a plane which is tangential to the base cylinder of the involute worm. Flank shape N can also be approximated with a srnall-diameter end milling cutter or side-and-face milling cutter with straight-sided flanks.
Hobbiog and Hob Peeling The contlitions are the same as in hobbing of helicaltoothed cylindrical gears. The flank shape is I. The worm blank, instead of the hob, and the tool (peeling wheel), instead of the gear blank, are clamped into the generating machine. The cylindrical wonn performs a tangential motion. Worm rotation, tangential motion and peeling Figure 29. Axes of motion of a continuous-type machine for generating by grinding: X radial feed, Z axial feed, V tangential feed, A pivoting angle of tool, B tool rotation, C workpiece rotation.
feeding), tooth flank corrections can be made in the direction of tooth height and width with the aid of a profile which is variable over the length of the involution. Continuous generating by grinding is used in large-scale series manufacturing.
wheel rotation are kinematically linked. To produce an enveloping worm, the peeling wheel is fed in the radial direction of the hobbing machine. Flank shapes A or I are possible. The enveloping worm is matched to the curvature of the circumference of the worm gear. During manufacture, the tool cutting edge must rotate about the centre of the worm gear in kinematic linkage with the rotary motion of the worm gear. Figure ~~ illustrates the motion relationships. S.2.~
Cutting of Worm Gears
S.2.2 cutting of Worms
Principles
Principles
The flank of the worm gear is a helical surface; the basic body is globoid in shape. The flanks are produced by generating the enveloping body of the tool corresponding to the worm with which the worm gear is to be paired.
Four blank shapes of cylindrical worms are standardised according to DIN 3975 (Fig. ~O): Flank Shape A. Trapezoidal tool cutting edges lie in the
axial plane. The axial profile of the worm is trapezoidal and has straight-sided flanks. Flank Shape N. Trapezoidal tool cutting edges lie in the
normal plane. The normal profile of the worm is trapezoidal and has straight-sided flanks.
Radial Method A radially fed cylindrical hob plunges into the worm gear until the centre-to-centre distance between the worm and the worm gear is reached. The effective hob length must cover the profile formation zone of the worm gear. This
I:.~:I Designation
Manufacturing Processes. 5 Special Technologies
Tools and tool incidence for worm cutting
Illustration Axial section A-A
Straight-
~ sided flanks
Flank shape A (ZA-type worm)
• A
Turning tool with trapezoidal profile Axial incidence parallel 10 axis
A _
_
_
Turnrng tool
Normal section N-N Straight-
~ sided flanks\\
Flank shape N (ZN-Iype worm)
4B" -'
.
\\ . .
..
-
Norma~ Convex Flank shape I (ZI-Iype worm)
• -
Normal section N-N
\\
.
Turning tool
Hob or grinding wheel with straight-sided ftank profile Incidence as for manufacturing of involutetoothed helical gears
C
~ onvex
Flank shape K (ZK-type worm)
.
Turning tool with trapezoidal profile Normal incidence at right angles to direction of tooth space
Disc cutter or grinding wheels with trapezoidal profile Normal incidence at right angles to direction of tooth space
method is suitable only for wonn gears with an 8° lead angle. At larger lead angles, the hob cuts away parts of the flank which belong to the helical surface at the full depth of cut before the final centre-to-centre distance is reached (Fig. 34a).
Tangential Method A hob with a chamfer (Le. a tapered part of the hob; its purpose is to spread the load on the cutting edge by increasing the addendum of the hob, and to shorten the tangential approach distance) is rolled tangentially past the dedendum circle cylinder of the wonn gear with the same centre-to-centre distance as that of the wonn and the wonn gear. Only one hob tooth need have the full profile. It is therefore possible to manufacture wonn gears by the tangential method using a fly cutter (Figs 34b, c:).
Radial-Tangential Method This method combines the advantages of the radial method (short feeding distance) and the tangential method (precise flank fonnation). It comprises radial plunge cutting until the centre-to-centre distance between the wonn and the wonn gear is reached, followed by tangential feeding. Straight hobs shorter than the wonn gear profile formation zone are possible (Fig. 34d).
Figure ~_ Flank shapes of cylindrical worms, standardised according to DIN 3975.
S.2.4 Bevel Gear Cutting Principles Bevel gears are used to transmit motion between axes which intersect or cross. The basic bodies are cones in the case of gear pairs without axial displacement and hyperboloids in the case of axially displaced gears. Any shaft angle is possible; in practice, however, the shaft angle is usually 90°. During manufacture (by generating), both wheels of a bevel gear pair roll with the imaginary generating wheel (crown gear); the tool embodies one tooth flank, one tooth or several teeth of the generating wheel. The teeth are described by the profile in the vertical tooth direction and by flank lines in the longitudinal tooth direction. The gear profile depends on the tool profile and the relative motion between the tool and the gear to be manufactured. Flank lines result from the kinematics of the generating process (straight, arc-shaped, epicycloidshaped or involute-shaped teeth). The cutting motion is performed in the longitudinal tooth direction.
Shaping Methods The tool (milling head, planing tool, side-and-face milling cutter, end milling cutter, or grinding wheel) has the profile of the tooth space. The tooth spaces are produced
5.2 Gear Cutting. 5.2.4 Bevel Gear Cutting
':.0
TP FCT
WRD
0·
9
RGB
TM Figure 3:1. Trapezoidal, straight-sided-flank form-cutting tools (TIl turning tool, EM end milling cutter, SI side-and-face milling cutter) in engagement with cylindrical worm CW (DIN 3975): "generating angle, NE normal plane of worm, OP cross-sectional tool profile, TM cutting motion of tool, RGB rotary motion of gear blank, AFaxial feed motion of worm.
CW Figure '1. Disc-shaped, tapered form-cutting tool FCT in engagement with cylindrical worm CW (DIN 3975): " generating angle; 'Ym centre pitch angle of worm; TP straight-sided-flank, trapezoidal tool profile, NP slightly concave normal profile of worm, TM cutting motion of tool, RGB rotary motion of gcar blank, AFaxial feed motion of worm.
GSN
RGB
\
8----
individually or continuously. During manufacture, an imaginary tooth of the generating wheel machines the material of the tooth space by means of a straight or curved cutting motion 4 (Fig. ~S).
PW
Generating Methods
The gear proille is formed as the envelope of the tool cutting edge. The motion 2 of the generating wheel is matched to the rotary motion 1 of the gear blank by a kinematic linkage as if a gear and a mating gear were revolving in a bevel gear train. A tool proille with straightsided flanks is usually used. Continuous generating is possible with a milling head or a bevel hob (tapered worm). Generating with indexing is carried out using planing
tools, side-and-face milling cutters, and disc or cup grinding wheels. Principle of an NC Bevel Gear Generating Machine
The cutting motion of the tool D (Fig. ~6) and the rotation of the cradle A supply speed sensor pulse
Figure 33. Disc-shaped, tapered form-cutting tool CT and peeling wheel PW in engagement with enveloping worm GSN (after Thomas AK. Zahnradherstellung. Hanser, Munich, 1965). RGB rotary motion of gear blank, RT rotary motion of tool, CR centre of rotation = centre of worm, TM cutting motion of tool, RF radial feed motion.
sequences for the continuous control of the gear blank rotation B, to achieve the kinematic linkage predetermined by the indexing and rolling transmissions_ During the infeed phase there is no rolling, only plunge feeding X. The machine is set up by means of electronic positioning axes: workpiece positioning Y, workpiece pivotTF
TF
c Figure 314. Worm gear cutting methods: a radial method, b, c tangential methods, d radial-tangential method; RW rotary motion of worm gear, cr cutting motion of tool, A centre-to-centre distance between worm and wonn gear, SH straight hob, TH tapered hob, FC fly cutter, TF tangential feed motion, RF radial feed motion, EWG enveloping worm gear.
':.:1.•
Manufacturing Processes _ 5 Special Technologies
._._._., I
I
Cradle
Gear tiank axis
MachirlE!
axis
Figure ~S. Basic system configur.ttion for profile milling and hob-
bing of bevel gears: a degrees of freedom of machine adjustment and movement, b gear-blank-generating wheel-tool configuration; 1 rotary motion of gear blank, 2 rotary motion of generating wheel (cradle), 3 adjustment of eccentricity of multitooth cutter, 4 cutting motion of cutter, 5 adjustment of angle of inclination of cutter, 6 adjustment of machine axis angle, 7 axial adjustment of gear blank, 8 feed motion, 9 adjustment of axial displacement.
ing axis C, cutter distance adjustment V and cutter positioning axis E. There is a manual axial displacement adjustment Z.
Bevel Gear Cutting Processes
Figure ~6. Design of an NC bevel gear hobbing machine (according to Messrs Klingelnberg, Hiickeswagen): B continuous NC gear blank rotation, D cutter rotation, A cradle rotation (alternating with feed motion X). NC machine adjustment axes: Y workpiece positioning axis, C workpiece pivoting axis, V cutter eccentricity adjustment, E cutter positioning axis, Z manual axial displacement adjustment.
edges. The rolling motion and indexing are the same as in generating by planing (generating method with indexing).
Cyclo-Palloid Process (Klingelnberg). This process is used for manufacturing spiral-tooth bevel gears. The flank lines of the generating wheel are epicycloids (Fig. ~7). The ratio of the cutting motion 4 of the cutter to the continuous indexing motion of the gear blank (Fig. ~S) is the same as the ratio of the number of teeth to be cut on the gear blank to the number of starts of the cutter. The rotary motion of the cradle 2 and the additional rotation of the gear blank are performed in the same ratio as the number of teeth to be cut in the gear blank and the number of teeth on the generating wheel. The crowning is generated
by an indexed multitooth cutter with various orbit radii.
Spiroflex Process (Oeriikon). This process is used for manufacturing helical-toothed bevel gears. The flank lines and the process of generation are analogous to the cydopalloid process, but with no indexed multitooth cutter; instead, the crowning is generated by the inclination of the cutter spindle 5 (Fig. ~S). Generating with a Bevel Helical Milling Cutter
Pal/aid Process (Klingelnberg). This process is used for manufacturing helical-toothed bevel gears. The bevel heli-
Ml
Generating by Planing. This process is used for manufacturing straight-tooth or helical-tooth bevel gears. One or two planing tools (a straight-sided flank cutting edge may be used) perform a reciprocating cutting motion. The planing slide is mounted in the cradle in place of the cutter (Fig. ~S). The gear train for linking the cutting motion of the tool with the motion of the table is dispensed with. The rolling motion is produced by rotation of the cradle and balancing rotation of the gear blank. When a tooth space has been completed, indexing is carried out and the cradle returns to its initial position
(generating method with indexing).
Generating with a Multitooth Rotating Cutter
Cutting of Curved Teeth (Gleason). This process is used for manufacturing curved-tooth bevel gears. The multitooth cutter has straight-sided flank or spherical cutting
Figure ~7. Formation of epicycloid flank line on generating wheel (continuous generating process) from the cutting tip paths by rolling of the rolling circle RR on the base circle RB; cutting tip groups MI, M2, M3, consisting of outer and inner cutting tips, cut successive tooth spaces 21, Z2, 23; 1 epk.ycloid flank line of generating wheel, 2 triple-start multitooth cutter, 3 outer cutting tip, 4 inner cutting tip, 5 generating crown gear.
5.3 Manufacturing in Precision Engineering and Microtechnology • 5.3.1 Introduction
cal milling cutter replaces the multi tooth cutter on the cradle. The flank lines of the generating wheel are involutes (Fig. ~8) which arise from the kinematic linkage between the cutting motion of the cutter and the rotation of the gear blank. The indexing motion of the gear blank is continuous. The generating feed is accomplished in snch a way that starting from position 1, the cutter plunges into the gear blank as far as position 2 and clears to position 3. Grinding of Bevel Gears
This is done to improve the surface linish and to remove the distortion due to hardening and the tooth-cutting errors. Principle: in grinding of straight- and helicaltoothed bevel gears, the proftJe of the grinding wheel (disc wheel) embodies the proftJe of the imaginary generating tooth. For bevel gears with curved flank lines, the body of the grinding wheel (cup wheel) corresponds to the envelope of the multitooth cutter. The motion relationships are essentially the same as in the bevel gear cutting processes.
5.3 Manufacturing in Precision Engineering and Microtechnology L. Kiesewetter, Cottbus S.~.l
Introduction
As miniaturisation progresses, precision engineering has to solve special problems with regard to both design and manufacturing technology. Precision engineering is not a kind of "mechanical engineering in miniature", but has a technical character of its own which derives from the
Figure 38. Bevel worm hob WH in engagement with hypoid bevel gear blank GB (palloid process, according to Messrs Klingelnberg): GW generating wheel, TM cutting motion of tool, RGB rotary motion of gear blank, RGW imaginary rotary motion of generating wheel, FMT rolling feed motion of toot, U point of origin of involute.
':.:11
smallness of the components, the high absolute precision, the signal-orientated mode of operation and the mass production that is typically encountered here. These characteristics necessitate the use of specific functional elements, production processes and highly relined materials. Accordingly) there is often a close interrelationship with other diSciplines, especially physics, optics and electronics (Fig. ~9). Design and manufacturing technology in these lields is concerned with small objects such as instruments, measuring and control system components, data processing devices, clocks and watches, balances, small drives, and even toys. The trend within precision engineering is in the direction of microtechnology, meaning components and systems that are manufactured by the production methods of semiconductor technology but which functionally take far greater account of stmcturing in the direction of the third dimension. The product range of precision engineering therefore ranges from geometric bodies with very tight tolerances and high-quality surface tlnishes to mass-produced appliance technology articles and further to microtechnology, which offers maximum preciSion but works with extremely small dimensions. The associated manufacturing technology has to encompass the areas that lead on the one hand to extremely accurate one-off pieces and on the other to high-precision mass-produced articles. The former culminates in all the processes of fine machining [10,111. By means of high-precision machines and tools, conven-
tional machining techniques lead to extremely precise surfaces and tight tolerances. In addition, small-size designs have long been used as machines for precision engineers. For the manufacturing process of turning, table-top machines and bench lathes without integral stands are available. For large quantities, automatic lathes are used which process bar stock and use coils of wire for extremely large quantities. An analysis of the motions shows that machines of a completely reversed design are feasible (Fig. 40), where the workpiece performs translatory motion analogous to feeding and the tools perform the radial and rotational motions in a rigid tool plane. Starting material in coil form is up to 30% cheaper than bar stock. Modem manufacturing processes of precision engineering, however, are often based on the application of new
Figure ~9. Schematic diagram of the sectors of application of important diSciplines within precision engineering.
'M:.
Manufacturing Processes • 5 Special Technologies
2"n
5.3.2 Laser Beam Processing
~4
Physical Principles
----------------------------
v
a
Figure 40. a Longitudinal turning and b turning from the coil: tool, 2 angular velocity of cutting, 3 feed step = component length, 4 tensioning step, 5 angular velocity for straightening.
physical effects [12]. To understand these manufacturing processes, the physical effects on which they are based must be known. One of the primary tasks in mass production by precision engineering is to optimally adapt the products to economic production. As up to 70% of the manufacturing costs of a precision engineering product are accounted for by assembly and the associated quality assurance, in the field of product design special emphasis must be placed on design for ease of handling. Comparisons of the cost of alternative solutions must not end at the production of the components, but extend up to the condition in which the component is operational in its fmal position. For this purpose the positional probabilities of the components have to be calculated, the supply functions and supply eqUipment selected [13] and the level of automation precisely matched to the manufacturing task. Technological progress is particularly expected in the field of microminiaturisation. It is necessary to create a synthesis of innovations in design science and materials technology in conjunction with newly developed and not yet developed unconventional manufacturing methods [14].
In 1960 T. H. Maiman (USA) succeeded in achieving an inversion of the population numbers of discrete energy levels with lingering periods of a few milliseconds. The first LASER (acronym for light amplification by stimulated emission of radiation) had been invented. In contrast to thermal radiation sources, a laser emits amplified and correspondingly intensive, highly monochromatic light of high coherence in space and time. The almost parallel beam of light has the properties of sharp directionality, high coherence length, good focusability down to almost one wavelength and extremely high power densities. These densities are achievable only up to values of 10" W / cmz if the stored energy of the laser is withdrawn in pulses and focused on small focal spots. 106 to 107 W/cm2 is the power density at which most materials evaporate. Thus manufacturing technology appears as the preferred field of application of the laser [12, 15-17].
A laser is a device for prodUCing inversion of population numbers in different energy levels with lingering periods of a few milliseconds in the metastable band. Such systems are pumped with continuous or pulsed light or with a d.c. or a.c. voltage to produce a gas discharge. To emit laser light, precisely defined absorption and emission bands are passed through, which on broad excitation leads to the poor efficiency encountered here and on emission leads to very narrow-frequency spectra and thus to the defined wavelength of the light beam.
Applications. The types of laser most suitable for manufacturing technology applications are solid-state lasers and gas lasers (Table 2) [18]. Of the solids, the most important are ruby (AlzO, as host material doped with Cr'+ ions, A = 0.69 fLm), glass and garnet (Y,Al,Oz, YAG for short, as host material doped with active Nd ions, A = 1.06 fLm), while the COz laser occupies the primary position among the gas lasers (CO z mixed with N pumping atoms, A = 10.6 fLm). Both may be operated continuously or in pulses. Recently, so-called excimer lasers have come to be used more and more in manufacturing. Exci· mers are diatomic excited molecules in high-pressure gas which consist of a noble gas and a halogen atom. On disintegration they emit light with a particularly short wavelength of 193 to 248 nm, i.e. in the u.v. range. Therefore
Table 2. Types of laser and their applications in manufacturing technology Type of laser
Power range in W
Mode of operation
Application
Excimer laser (0.193 !Lm/ 0.248 !Lm)
5· 106 to 3· 10-
pulsed (15 to 300')
Chip less machining, notching, photochemiStry, spec· troscopy
He-Ne laser (0.632 !Lm)
< 100
continuous
Measuring systems
Ruby laser (0.693 !Lm)
1 . 104 to 4· 10' 1 . 102 to 2· 10 2
pulsed (I to 10 ms) continuous
Chipless machining
Nd-YAG (1.06 !Lm)
10(' 1.5·10'
pulsed (I to 10 os) continuous
Chipless machining joining
CO2 laser (10.6 !Lm)
<5.106 2 to 2.5· 10'
pulsed (I to 1 . 10' !L') continuous
Cutting, joining, chipless machining, surface treatment
5.3 Manufacturing in Precision Engineering and Microtechnology • 5.3.3 Electron Beam Processing
they are ideally suited to the machining of even more precise dimensions. For manufacturing technology, the laser beam is often focused with the wavelengths of adapted lens systems and guided to the point of action by means of beam deflection systems or optical fibres. Being a "tool" that is not subject to wear, lasers are
suitable for welding, notching, engraving, sheet cutting, drilling and for altering the properties of various materials such as metal, glass, silicon, diamond. ceramic. plastic, paper and textiles. At the often low energy levels. high power densities are achievable only with small working ranges. Alongside this, the trend is towards CO, lasers with 25-kW beam capacities for applications in machine manufacturing. These have cooled mirrors and aerodynamic output windows to prevent heat losses in the area partitioned off against the low pressure.
Welding. The work is performed in air or under shielding gas. mainly with Nd or CO, lasers. The welding eqUipment is perfected, especially for microwelding; seam and spot welding are performed in the millisecond range [19J. It is especially important to coordinate the geometry, choice of material and manufacturing technology in order to minimise the high reflection losses in the surfaces at the start of machining. In addition, the laser beam must be prevented from reflecting back into itself. laser beams can also be used through transparent walls, e.g. behind glass. Microsoldering is regarded as a process requiring particularly sensitive execution in making microcontacts with high packing densiry in microtechnology. In this fairly complex process, the increased absorption of laser light on the melting of the solder is sensed from the measurement of the thermal radiation by means of infrared sensors and the laser beam is switched in fractions of 0.1 s 1201. Drilling. Practically all materials can be drilled. even quite hard materials like glass, corundum and diamond, at power densities of 10' to 108 W /cm' and machining times of 10- 4 to 1O- 6 s 121J. Owing to the beam acoustics, cylindrical holes can be drilled only under limited aspect conditions. The material must vaporise, but the formation of
plasma must not screen the laser beam. Because of the small photon mass, the beam penetrates the surface only to a depth of a fraction of a micrometre; thus the material is removed layer by layer.
Sheet Cutting. The CO, laser is ideally suited to sheet cutting. Most industrial materials can be cut, often under gases such as inert gas or oxygen due to the high continuous wave power; sheets up to 5 mm thick can be cut at power densities of 108 W/cm' and at speeds of 6 m/min. Not only steels and metal alloys but also organic materials and ceramics can be cut. The two last-named materials in particular can be effectively notched with the laser by introducing stresses into the workpieces by cutting rows of holes in such a way that on subsequent bending the folds run in the desired directions. Chipless Machining. The laser beam is used as an accurate material removing tool for trimming, e.g. for tuning of tuning forks and quartz crystals or for adjusting hybrid-connected resistors and capacitors to their setpoints. For resistors, initial tolerances of - 10% can be trimmed to 1%, whereas for quartz tuning forks, accuracies of 106 are achieved by evaporation of extremely thin gold layer zones. Extremely precise chipless machining of plastics is accomplished with, e.g., pulsed excimer lasers up to 250 W. Polymer laminates are photochem-
':':JCI
ically removed without imposing thermal stresses on the base materials, even in narrow edge zones. Chip less machining in the sense of evaporation of four-component sintered compacts rotating in a vacuum chamber may also be carried out with excimer lasers having a frequency of 30 Hz, a pulse duration of 40 ns and A = 248 nm. With a 5-min cycle time for evacuating, heating to high temperature, vapour deposition and dismantling, substrates of Sr, Ti, 0, for computer chips can be coated with superconducting polycrystalline films consisting of yttrium-barium-copper oxide (YBa,Cu 2 0 7 ) which withstand current densities up to 1.5 kA/cm'.
Coating. Ceramic substrates can be given a structured metal coating by means of the laser chemical vapour deposition (LCVD) technique [22J. This is achieved in a chamber at low pressure with u.v. lasers which release the metal atoms from gaseous metallo-organic compounds by pyrolysis or photolysis in the surface zone by direct programmed inscribing or by means of templates. Metals such as Au, Rn, Pd or Os are used. A further area of emphasis for lasers is remelt coating. Here the surfaces of workpieces are completely or locally treated, chiefly with CO 2 lasers at 500 W, in such a way that coating materials previously formed by thermal coating or application of powders, pastes or solids are alloyed with the surface or diffused into it. Stereolithography. perhaps better termed multilayer laser polymerisation in manufacturing technology, appears to be becoming an efficient method of manufacturing three-dimensional prototype shaped parts from polymer materials. Using HeCd lasers, only the top 0.05 to 0.15-mm thick liquid monomer layers are inscribed with laser light and approximately 70% cross-linked. Then the component is lowered in the bath of monomer by the thickness of the layer and the next layer is structured with defined geometrical data. The three-dimensional component is ultimately obtained by post-curing in an oven under u.v. light. s.~.~
Electron Beam Processing
Physical Principles In machining of workpieces with electron beam devices,
a concentrated beam of highly accelerated electrons is directed at the point of action in a vacuum. It is perhaps surprising that the effect often resembles that of a laser beam, although the rest mass of electrons alone is 3 . 10' times greater than that of a photon and the same energy level is achieved at acceleration voltages of only approximately 2 V. With acceleration voltages of 200 kV in electron beam generators, the deep welding effect and the dependence of the manufacturing process on the density of the material to be machined can thus be explained. The beam generator consists of a hot cathode, an anode and the control electrode (Wehnelt cylinder) [23, 24J. The latter focuses and switches the beam intensity up to power densities of 109 W / cm' in the focal spot. The beam may be shaped on the way to the point of action and guided and deflected without inertia by electrostatic or electromagnetic deflection equipment. The smallest focal spot diameters are less than 1 fLm.
Applications. Three types of electron beam machines are distinguished: high-vacuum, semi-vacuum and atmospheriC machines. Chamber-type machines, phased machines with turntables and batch-operated throughtype machines are used for series production. The controllability of capacity, focusing area and beam direction in conjunction with automatic fmding of the point of action
.;.:11
Manufacturing Processes. 5 Special Technologies
by measuring the intensity of revertively controlled electrons pennits a broad spectrum of applications, particularly in precision engineering. Extremely precise spot and seam welding can be carried out under vacuum, as can the removal of material by cutting, drilling, pierdng, engraving and fusion notching [25, 26]. The materials used are metals, alloys, ceramic, precious stones and the like. Only a small amount of thermal stress occurs around the point of action and the processing in vacuum maintains a high material purity. The duration of beam positioning and the effective duration of e.g. spot joining or metal removal is a few milliseconds. One of the main applications of electron beams is vapour deposition in a vacuum (Fig.41) [12]. Thin films for opto-electronic components, in semiconductor technology, for film capacitors, large-area coating of window glass and the like are produced by this means. Vapour deposition by means of electron beams shows in this case the advantages of minimal crucible contamination and the possibility of varying the properties of the film by using different deposition materials in several crucibles which have to be individually controlled by the electron beam with regard to intensity and time. Evaporation rates ranging from 1 g/h to 100 kg/h pennit high working speeds, even when producing thicker layers of over 10 f1lD [27, 28]. A non-thermal application of electron beam processing is found in electron beam lithography, where the beam is used to inscribe mask structures in photosensitive layers. This is the most important technique for producing the master masks in IC technology and micromechanics.
can achieve reproducibility, 100% cleaning efficiency and machining times ranging from seconds to minutes. This is accomplished using austenitic stainless steel 18/8 troughs in the 20 to 40 kHz range to which are attached transducers leading to a sound field of maximum possible homogeneity. The latter'S main effect is cavitation, which at 20 W /1 chiefly occurs at contaminated material and induces pressures of over 1000 bar [29-31]. Machining in the defined effective range of the workpieces is accomplished by means of the ultrasonic equipment illustrated in Fig. 4.:1. An acoustic head mounted at the vibration node transmits its sympathetic vibration to a booster and, to increase the amplitude, into the effective tool range of the machine by means of a sonotrode. Here amplitudes of 5 to 35 11m are produced at 20 to 40 kHz. In welding of metals, the shearing effect of the surfaces to be joined is exploited, while in joining of plastics the compression and tension phases within the thermoplastic materials are made use of. The preferred metal is aluminium, the oxide skin of which performs considerable frictional work until it is fully destroyed. Thus ICs with 27-l1m thick aluminium wires are contacted by "wedgebonding" on a large scale [32]. The phenomenon of ultrasound offers the following advantages: short welding and post-clamping times of
S.:t.4 Ultrasonic: Processing
•
Ph)'Sic:al PriodpJes Ultrasound is an elasto-mechanical vibration above the hearing threshold. It ranges from 20 kHz to beyond the megahertz range and is generated from electrical energy by means of piezoelectric or magnetostrictive trans-
ducers. AppHc:ations. In manufacturing technology, ultrasound is used for cleaning and for joining processes such as welding, riveting and embedding, in instrument tech-
c:
nology and in medicine. There is no other process that
~--......-
/.
d
""
'
~ ./ -""
e
/~"
f
~ (II . :0/ ~-"---
./
g
.
Figure 41:. • Schematic construction of an ultrasonic device and the most important applications: 1 of the HF generator, 2 trans-
Fipre 41. Principle of electron beam vapour deposition: 1 electron gun, 2 magnetic field B, 3 evaporated material, 4 substrate
heating, 5 substrate, 6 vacuwn pwnp, 7 recipient, 8 vapour stream, 9 evaporator diaphragm, 10 crucible (cooled).
ducer, 3 booster, 4 sonotrode, 5 force, 6 anvil. b Principle of ultrasonic metal welding. c Principle of ultrasonic plastic welding. d Principle of ultrasonic wedge bonding. e PrinCiple of ultrasonic spot welding. f Principle of ultrasonic riveting. If Principle of ultrasonic far-field welding.
5.3 Manufacturing in Precision Engineering and Microtechnology • 5.3.5 Electnxlischarge Machining
about 1 s, the possibility of joining parts with widely differing wall thicknesses, high strengths. no pretreatment, and no structural changes in the material. With plastics. joining is performed in the near-field range of 6 mm and in the far-field range [33]. Ultrasonic cavity sinking, also termed ultrasonic drill-
to dry processes such as ion-beam etching, these processes take place in liquid active media {34-37].
Electro-discharge Machining, Electrochemical Machining, Metal Etching
Electro-discharge Machining. As electric arc machining produces inexact copies, under the heading of electrical machining, electro-discharge machining is often used, which takes the form of removal of material or migration of material between electrically conductive contacts. According to the definition in VOl Code 3402: "Electrical machining comprises the removal of electroconductive materials caused by electrical discharges between electrodes under a working medium for machining purposes." The electrodes are the shaping tool and the workpiece to be machined. Electrical machining therefore represents the electrical alternative to ultrasonic processing. The polarities of the workpiece and the tool have to be borne in mind deliberately to achieve low relative tool wear. The sparks across a gap are temporary local discharges, the effect of which on the workpiece surface is characterised by the pinch and skin ~ffects. The machines, which are operated with pulse or relaxation generators, can carry out the processes of cavity sinking, wire EDM, grinding and sawing [38].
In the survey of manufacturing processes according to DIN 8580 (cf. Kl), the chipless machining processes (DIN 8590) can be classified according to Fig. 43. The most interesting processes in precision engineering are marked with an asterisk; ultrasonic processing is a purely mechanical process, while the effective mechanism of the beam-based processes often lies in thermal effects. For shaping small components. the processes of electrodischarge machining, electrochemical machining and metal etching which are listed in VOl Codes 3400 and 340 I are often also used. What they have in common is that electric current, sometimes with "localised voltaic cell formation", is responsible for their effect. In contrast
Electrochemical Machining. This is an electrochemical process in which metal atoms of the anode pass into solution under the influence of a d.c. voltage of about 20 V in aqueous solutions of salts or acids as electrolytes. It is the reverse of electroplating, in which a material migration of one gram-equivalent is caused by 96487 C. To determine the geometry, the electrolyte is fed through an insulated nozzle at a velocity of up to 30 mls and achieves very high removal rates at current densities of 250 A/cm2 . With regard to specific machines and applications, the processes can be divided into electrochemical etching, surface removal at up to 40 cm'lmin and, by analogy and in conjunction with chip-forming processes,
ing, is based on the machining of hard, brittle materials
with an abrasive suspension acting in the effective zone between the workpiece and the end of the sonotrode, which is the tool. Machining can be carried out with a relative tool wear of 1%, and material removal rates of 1200 mm'/min, chiefly in hard, non-conductive materials unsuited to electro-discharge machining. which is normally the alternative manufacturing process [12]. Glass, diamond and materials used in the gem-cutting and semiconductor industries are machined. The feed force in the direction of oscillation must permit the tool to lift off during the decompression phase to create space for the material removed to be washed away and fresh abrasive suspension, such as oxides and carbides, to be supplied. S.~.S
Joining - - Surlace coating - - - Modification of material properties
DIN 8580: Primary shaping - - Metal forming - -
Severing
-~~
Dismantling - - - Cleaning - - - Evacuation
Machining by metal cutting
Machining by etching
EC form machining'
Thermal machining by heating
Thermal-chemical deburring
EC surlace machining (vat ECM),
with solid bodies with liquids with gases
Chemical-thermal machining
Thermal machining by friction
I
-I
I
I
I
EC etching (metal etching)'
with sparks' } with arcs VDI 3400 and 3402 with plasma beams
I
Thermal machining by high-energy radiation with an electron beam' with a beam of photons'
Note: EC etching is based on anodic removal of metals (in contrast to EC processes, usually without an extemal current source).
with ultrasound' Figure 43. Classification of chipkss manufacturing processes according to DIN 8580 and 8590, VDI Codes 3400, 3401 and 3402.
':.:0
Manufacturing Processes • 5 Special Technologies
shaping
by electrocbemical macbining (e.g. EC grinding). Thus they are at the same time processes for achieving component geometries and surfaces with roughnesses down to R, = 0.5 f.Lm with no burr [39). Metal Etching. This is carried out with an external current source applied to the component, which serves as the anode. However, voltaic cell formation also takes place locally in the electrolyte, e.g. in the HCI or FeCI, bath in the case of copper components. Other etching solutions include ammonium persulphate, sulphuric acid, nitric acid, hydrofluoric acid, copper chloride and caustic soda. The machining is then carried out by direct reaction of the etching medium with the component material, often with liberation of hydrogen or an oxygen reaction. This process is often used to manufacture complex shapes in the form of foils or sheet metal components or for structuring the strip conductors of printed circuits (foil etching, shape etching). Etching is carried out by immersion or with centrifugal or spray etching devices; etching rates of up to 50 f.Lm are achieved. Defined structures can be produced if a masking layer (etching resist) is applied to the surfaces that are not to be etched. Isotropically acting etchants cause undercutting of the resist, which leads to narrower components or lead geometries. More accurate sheet metal components are therefore obtained with double-sided coating and etching; attention needs to be paid to the overlay. This process is especially efficient with layer thicknesses up to 0.2 mm and with exacting requirements with regard to geometry and freedom from burr or in small production runs. S.~.6
Coating Processes
Coating is normally used for decoration and protection of surfaces [12, 40-42). In precision engineering, however, the layer - particularly if it is structured - often becomes the performer of the function, while the coated material becomes the substrate. Depending on the requirements, electroconductive, semiconducting, insulating, superconducting, soft and hard magnetic, hard-wearing and selflubricating surfaces are required. According to DIN 8580, coating is the application of an adherent layer of an amorphous substance to a workpiece. In view of the possibilities offered by manufacturing technology particularly for making coatings, the implantation or burial of coating material must be included here. With regard to function and manufacture, a distinction is made between thin coatings ranging from 0.01 nm to 1 f.Lm and thick ones of greater depth. The coating material may be present in any state of aggregation: in the gas phase, the liquid phase (electroplating [43]) and the solid particulate material [44J. An interesting variant of wide-area coating with a monomolecular film thickness is the Langmuir-Blodgeu process, in which the fmely distributed coating material is floating on a liquid and fully and directionally wets the substrate when lifted out of the liquid [45]. All processes, especially for coating of thin-walled substrates, depend on minor internal stresses, which may result from disorder and/or the bringing in of foreign atoms and different coefficients of expansion. Moreover, an imponant evaluation criterion is the bonding strength of the coating, which is produced by the bonding forces between the coating material and the substrate. In the case of glass or ceramic substrates, the bonding forces to the desired metal coating can be deliberately increased by means of reactive intermediate layers of metal in the nature of layers of bonding agent consisting of Ti or Cr. Thin coatings are produced by means of the PVD and
Figure 44. Schematic diagram of a diode sputtering unit: 1 insulation, 2 cathode with magnetic field, 3 target, 4 argon, 5 substrates, 6 holding device, 7 cooling, 8 valve, 9 pump system.
CVD processes. PhYSical deposition from the gas phase (PVD) comprises vacuum metallising, sputtering and ion implantation, together with their reactive variants. In vacuum meta/Using, coating material is deposited in straight-line propagation from the vaporiser to the substrate in a vacuum chamber according to a cosine distribution law. Spuuering is a purely mechanical process in which gas ions strike the coating material - the target - in the vacuum chamber and "lever out" the atoms (with maximum effect at an angle of incidence of 45° to 60°), which are accelerated towards the substrate in an a.e. or d.c. field. Figure 44 illustrates a magnetron system in which free electrons are kept away from the substrate by means of directional magnetic fields. Thus sputtering is accomplished with the substrate at a fairly low temperature. In ion implantation, ions in an electric field are accelerated to such a degree that they penetrate deep into the surface of the substrate and thus alter the properties of the material; it is clear that if the technical conditions are changed slightly, one ean very quickly jump between the main categories of DIN 8580 within the processes. Chemical vapour deposition (CVD) denotes the deposition of coating material from the vapour phase by means of the activation energies in the nature of thermal CVD, plasma CVD, photon CVD and laser-induced CVD. For example, silicon layers may be produced according to the reaction equation SiH, ~ Si + 2H, at a rate of 0.5 f.Lm/min. Often it is required to coat metals or insulating substrates with plastics. Table ~ gives a summary of the most
Table 3. Processes for coating metals with plastic Starting material PLastic paint Painting
Plastic powder Powder coating
Plastic film Roller coating
Brushing
Fluidised bed sintering
Calendering
Spraying
Electrostatic fluidised bed sintering
Roller melting process
Dipping
Flame spraying
Extrusion coating
Curtain coating Centrifuging
Electrostatic powder coating (flock coating)
Film laminating
5.3 Manufacturing in Precision Engineering and Microtechnology • 5.3.7 Plane Surface Structures
common manufacturing methods. Of these. an interesting process is "resist centrifuging", where, for instance, for the lithography technique, an excess quantity of photosensitive resist is placed with a dispenser in the centre of the substrate to be coated and then centrifuged to produce defined dry film thicknesses in a few seconds, depending on speed and duration. TIle fields of application are IC technology, microtechnology and liquid crystal technology [46].
':M:tl
Substrate p-SI
Substrate Si
~
~~~""""
Oxidation (SiO,)
Oxidation(SiO,), Coating (AI)
'~~~'~"~~~ Application of positive photosensitive resist
Application of negative photosensitive resist
S.3.7 Production of Plane Surface Structures For the products of precision engineering, the feasibility of producing plane surface structures is a determining factor for the high packing density to be achieved [12, 471. A characteristic feature is always the job of structuring a surface or its coating into surface elements or paths in such a way that its properties are fundamentally and clearly different from those of the surroundings. This binary statement may relate to any chemical and physical properties; a simple example is the printed circuit board [48, 49]. It should always be assumed, however, that plane surface structures exist on or within a substrate and the lateral extents of the overall layout are considerably larger than its measurements in the third dimension, height. Regardless of this, there are practical examples where, within a cross-sectional area, the depth dimensions are greater than those of the widths (vertical structure). If DIN 8580 is followed, plane surface structures are produced by coating, chipless machining and changing substance properties. The oldest examples are found in the production of written and printed matter by means of letterpress, gravure,flatbed and screen printing. while new methods are encountered in all lithograPhy and moulding processes for producing video discs, printed circuit boards, thick-film and thin-film circuitry, solid-state circuitry and especially for producing masks for these manufacturing methods. In screen printing, a squeegee is drawn over a gauze stencil, forcing ink or electroconductive paste through the parts of the surface where the mesh is open [50, 51]. By this means, bonding frames 2 to 3 I-'m thick and 200 fCm wide in LC manufacture can be made just as accurately as the printed gold strip conductors on cerdmic substrates
for manufacturing multilayer circuits on ceramic substrates with alternating strip conductor levels and ceramic insulating layers [52]. In accurately fitting multilayer techniques, special attention must be paid to the problems of the overlay. In impact printing processes, the contrastproducing ink is applied to the paper by mechanical pressure. Thermal printing processes (so-called non-impact processes) apply the structuring material by the action of heat at low mechanical pressure. Complete freedom in the design of the structure, with dependence on the software alone, is achieved in printing with laser printers, which produce electrostatic charge images and thus come under the heading of flatbed printing. With the use of light, however, one comes closer to entirely new techniques and the feasibility of producing considerably fmer structures still. The techniques of photolithography use extremely precise masks with the scale copy of the structural data in optical beam paths in lithographic coatings of the substrate in the contact method or with an extremely small mask gap (proxinlity). The principle of photolithography (photoresist technique), as used to produce printed circuit boards up to production of structures in the submicrometre range of silicon technology, is shown in Fig. 4S. Method b can also be carried out as a reciprocal variant, in which the metallised photosensitive coatings are lifted off.
-PlaCing 01 mask, exposure
Placing 01 mask, exposure
-'all Developing
Developing
~~' ,';'~','.;';~""
Etching
Etching
~~ Removal of resist and diffusion p-~~
a
Removal of resist
~~ ...~
n-Si
b
Figure 45. Principle of photOlithography: a ~elective etching and diffusion with negative resist, b selective metal coating with positive resist.
The limits to the photolithographic processes lie in the achievable edge sharpness and the desired fmeness of the structures; both these are limited or only achievable at all [53] by the wavelength of the light, any diffraction that occurs, and interference [54]. It appears possible to increase the gradation by means of special photosensitive dyes which are applied to the layer of photosenSitive resist to a thickness of 0.3 fCm. The satisfaction of the desire for ever finer structures (I-megabit chips) with strip widths in fractions of a micro metre is a challenge and a stimulus to manufacturing technology, for the "packing density" on a substrate is inversely proportional to the square of the structural width. Technical circumstances such as copying errors, thermal expansion and intrinsic stresses due to "retention" of the substrates pennit sufficiently precise matching of the mask and the substrate only within limited areas of the surface. Large complete fields can consequently be achieved only by "tacking together" small subfields by the step-and-repeat method. Laser systems can achieve positioning accuracies of 0.1 fCm. Xray lithography uses a "wavelength window" of 0.2 to 4 nm for exposing special photosensitive resists to produce structure widths of less than 0.5 fCm. This process is described in K5.3.8 as a specimen application. In the picture of the wave-particle dualism, the electron beam has an even shorter wavelength (A < 0.1 nm) [55]. Electron beam methods are of fundamental importance to lithography for another reason. Whereas in electron beam cathode projection on an image scale of 1 : 1 the electrons are liberated directly from the mask by u,v. light, in electron beam projection the image scale is reduced by means of further optoelectronic systems.
1;.:1:.
Manufacturing Processes. 5 Special Technologies
The most interesting variant of this application is the electron beam inscriber. It offers the possibility of inscribing directly with the electron beams, with control by software and with an adjustable focal spot size, to produce the extremely precise masks required elsewhere and for "direct inscribing" of structures and substrates for producing prototypes or small runs for product trials.
5.3.8 Manufacturing of Microstructures In [56] a process for manufacturing quartz tuning forks for watches is described, which can be chemically etched out of an SiOz wafer 125 11m thick at 85° in the nature of "wafer batch processing" [57]. The bath of hydrofluoric acid and ammonium fluoride etches the quartz in a highly anisotropic manner, depending on the directions of the crystal axes. This means that with an etching rate of approximately 4 11m/min in the z-direction, an extremely small amount of undercutting can be achieved in the xand y-directions. Thus, by "leading" in defined axial directions, not only the lateral dimensions of etched components but also the consequences in the third dimension can be predetermined and used for the geometric designs of the microstructures. Microelements and modules composed of them are products of microtechnology, which are themselves made up of elements of microelectronics, micromechanics and microoptics. Microtechnology products therefore include semiconductor circuits, integrated optical and optoelectronic systems, sensors made of silicon, microjets, microactors, subminiaturised mechanical, electrical and optical connections, and switches. All the components are distinguished by extremely small dimensions in the sub-micrometre range, an integrated structure and a system concept which is vitally necessary here. Electronic system components have been produced as integrated circuits for many years, so it was logical, in the spirit of monolithic or hybrid integration, to use the materials and technologies for micromechanical parts as well [14, 58]. In production technology, microtechnology comprises special coating, lithography and etChing techniques. Silicon Technology A material of supreme importance is the monocrystalline material Silicon, which is extremely attractive because of its mechanical properties such as minimal attenuation, absence of fatigue, maximum crystal purity, an electrical conductivity that can be determined by doping, etchability and coatability. The most important thing about this material, however, is the possibility of spatially exploiting the third dimenSion, that of depth, micromechanically with special selective and anisotropic etchants in relation to the crystal orientation. This anisotropic and isotropic etching, for the production of extremely fme structures and masks not only with lasers but also with electron beams and X-rays and the erosive methods, and electrocoating processes, justifies the great efforts made in the field of microtechnology [59]. The starting material for silicon processing is a wafer, from which many identical elements are produced by batch processing. The designing and structuring of the wafer is accomplished by means of a wide variety of manufacturing processes and eqUipment which, while mostly familiar from Ie technology, have to be specially adapted to the thicknesses in question [60]. The more finely the lateral structures are to be resolved, the more exacting are the demands made on the lithography processes; here, use is made of every type of radiation, such as light, X-rays and particle radiation. Besides processes employing a focused beam such as
an electron beam, even fmer resolutions in the submicro metre range are achieved with X-rays, which are formed tangentially to the acceleration sections of electron synchrotrons and are conducted into vacuum tubes. They are emitted as a wide beam from a window covered with plastic sheet with Gaussian intensity distribution at a distance of approximately 10 mm. Thus lithographic exposures can be achieved through beryllium masks with absorber zones by the proximity method, either directly or, at a greater distance, oscillating together in the vertical direction. Structuring in the direction of the coating thickness, i.e. to determine the geometry of the components within the silicon wafer, is achieved by means of additive and subtractive techniques. The former pennit the building up of insulating layers, e.g. SiOz or Si 3 N., or doped semiconductor layers, as well as coating with metals such as AI, AI/Si, AI/Si/eu or organic material and glasses. The technologies for this are epitaxial processes for monocrystalline silicon, chemical deposition from the gaseous phase and condensation of the products of decomposition (eVD), thermal oxidation, or vacuum deposition and sputtering. Specific material removal from silicon and coating structures is achieved by means of the wet chemical etching processes, which are isotropiC and thus produce identical etching rates in all spatial directions, as well as extensive undercutting. Important influences on the etching arise from the nature and freedom from defects of the etched material, the maskings and the orientation in relation to the crystal axes, the etchants with regard to temperature and age, and the external influences such as cleanliness. The advantages of the dry etching processes are the often higher material removal rate, the excellent structure resolution and an often-observed anisotropy or directional dependence. Material removal can also be accomplished by ion bombardment in low-pressure chambers. One of the main representatives of these equipment categories is a sputter-etching device, which is familiar in the inverse mode of operation as a coating device. Here a negative potential is applied to the component to be etched in a low-pressure plasma consisting of a chemically inert gas such as argon. The positive argon ions dislodge molecules and atoms from the substrate, which is partially protected in other places with photosensitive resist. In ion beam milling, a beam of argon ions is formed in a chamber and accelerated to 0.5 to 1 keY, striking the substrate under conditions of high vacuum. By adding a chemically reactive constituent, this process becomes reactive ion beam etching with initialising by the ion bombardment. Plastics can chiefly be structured by adding 0" e.g. webs 1. 5 11m wide and 30 11m thick. So far, Si, SiO and AI can be easily processed with etching rates of 0.1 to 111m/min and adjustable profiles or gradients of the sidewalls and aspect ratios of 10 : 1. The aspect ratios are always understood as the depth in the direction of processing in relation to the channel or web width of the structure.
Anisotropic Silicon Etching. Silicon has a lattice structure like that of diamond, as shown in Fig. 46 [14]. With Miller indexing of the orientated single crystals, there are preferred planes which are removed at widely varying rates with anisotropically acting etching solutions such as the alkalies KOH and NaOH or ethylenediamine with catechol and water or hydrazine and water. This anisotropic behaviour is caused by the lattice structure of the crystal and the associated varying bonding forces. Because the energy required to detach a silicon atom is greatest in the direction of the III plane, this direction is preferably
5.3 Manufacturing in Precision Engineering and Microtechnology _ 5.3.8 Manufacturing of Microstructures
stresses - compressive stresses in the case of boron. However, methods of compensating for these internal stresses by doping with, e.g., Ge atoms have become known [61]. As in general the "crystal detertnines" what is feasible with regard to manufacturing technology, it is extremely important that even within the wafer plane, the mask structures are aligned with the crystal direction. One solution so far has been to determine the direction of the crystal axes precisely on the "flat" of the wafer in anisotropic etching tests. For clarification, it should be imagined in this context that with prolonged etching, a window opening of any shape in the wafer always leads to an etching geometry whose shape lies through the tangents to the window opening in the direction of the crystal axes. Internal comers form sharp edges, whereas convex structures are undercut (Fig. 48). Etched walls pointing vertically into the thickness of the wafer are achieved if they are formed by the 111 plane of the crystal. Silicon microelements have to fit into a geometrical concept for the component periphery. To join multilayer systems made of, e.g., silicon and glass, the anodic bonding method has emerged, in which Pyrex glass is chemically joined to silicon at approximately 300°C at low pressure and voltage through becoming conductive in the joint gap. By anisotropic etching, three-dimensional microelements can be made from a solid silicon wafer 0.5 to 0.8 mrn thick by chipless machining.
•
b
':.:1'
5035'
c Figure 46. Lattice structure of silicon. a 100 plane; b, 111 plane; c, 110 plane. 1
maintained. Here the material removal rates are over a hundred times lower than in the other crystal directions 100 and 110. The production engineer is now faced with the task of placing the wafer surface in relation to the crystal axes in such a way that the etching produces the required geometries. A characteristic angle of 54.735° will very often determine the geometry of the components. This is the angle between the 111 and 100 planes with their widely differing etching rates. For if the surface of the wafer forms the 100 plane, four areas are formed in an etching window which, depending on etching depth and wafer thickness, may run out to a point. Double-sided etching with a precise overlay then leads to double-conical penetrations (Fig. 47). At etching rates of 5 to 150 fLm/h in the desired etching direction, the processes take a long time. Therefore the masking layers are made of SiO, or Si 2 N,. Boron-doped silicon with 1020 boron atoms per cubic centimetre is also resistant to such etching solutions. If, however, self-supporting tongues and the like are to be produced, it must be borne in mind that the lattice constant and atomic radius of boron differ from those of silicon, and this doping process leads to internal
Production of extremely accurate metal and plastic components of approximately identical thickness by means of a "buildup technique" is accomplished with the "LIGA" process, derived from the manufacturing steps lithography, electroforming and moulding [62]. Figure 49 illustrates the process sequence, in which a resist structure which can be easily modified by radiation physical means is first irradiated with high-intensity parallel X-rays for several hours via a mask at approximately 40 fLm distance. Depending on the resist material, the irradiated and unirradiated areas are selectively removed by developing, leaving structures with extremely fme
r&;] /.' [1;;]
Basic structures
after etching time T I
I- •
,
a
11001
:
-~:-
after 2T
LI010)
after 3T Figure 47. Isotropic and anisotropic etching of silicon.
Figure 48. Undercutting at convex comers (Si technology).
':WI,'
Manufacturing Processes. 5 Special Technologies
a Masl<
Aesist - - -_ _m~~Il8888llggg88E:83 Baseplate
While silicon technology can be mainly recognised as a direct method for manufacturing workpieces by wafer technology, the LIGA process is more likely to develop into a method of producing extremely accurate moulding tools.
Irradiation
~
Reslsl ~ructure
Developing
5.4 Surface Coating
Melal structure
Electroforming
H. K. Tonshoff, Hannover
Moofd insen
Sprue plate Moulding compound'
' . .,'
-=-~ . ~ -~
:~~re --~
Moufdmaking
Moolding
Demoulding
Figure 4,. UGA process: a manufacturing steps, b plastic honeycomb structure. The wall thickness is 4 j..lffi, the height ofthe structure 350 /Lm (Photo; KlK).
resolution owing to the short wavelength of the X-rays. At lateral dimensions of a few micrometres film thicknesses of several hundred micrometres can be produced. The gaps or clearances, now precisely represented, can be filled with metals such as nickeL Identical height is achieved by mechanical fmishing. After removing the resist, the result is a metal mould which can be used as a spray-coating mould for an infinite quantity of plastic components. The plastic elements can of course be used in their turn as tools for further electroforming.
Surface coating is the application of an adherent layer of amorphous material to a workpiece (DIN 8580). The coating and the substrate (base) form a sandwich of different materials. This enables a separation of functions: the coating assumes contact functions like protection against chemical or corrosive attack of tribological stress, influences the friction behaviour or serves optical or decorative purposes; the substrate performs support functions, for which its properties can be adapted to the specific stress regardless of the contact behaviour. This degree of freedom, which is achieved by combining the properties of the coating and the substrate, is the reason for the growing interest in surface coating technology. Multilayer coatings can be used to achieve further property-specific advantages, e.g. reduction of the coefficient of friction with the top contact layer, followed by diffuSion-blocking layers and layers to increase the bonding strength with the substrate. In ptinciple, three zones are distinguished: the coating zone, the bonding zone joining the coating to the base and the substrate as the shape-giving, supporting body. Coatings are applied to metals, ceramics, single crystals, glasses and plastics. Depending on the material composition and the coating process, the coating and bonding zones take an almost infinite variety of forms (Table 4). According to the state of aggregation of the amorphous substance to be applied, a distinction is made between coating from the gaseous or vapour state, that from the liquid or powder (or solid) state, and that from the ionised state with film thicknesses from less than I 11m to more than 100 11m. Coating from the gaseous or vapour state may be achieved by physical processes (physical vapour deposition, PVD) or chemical processes (chemical vapour deposition, CVD).
PVD Processes. These consist of three phases [631: evaporation of the coating material, transfer from the source to the substrate, and condensation on the substrate. The gaseous state is achieved by heating - evaporation (emission energy of particles low, < 0.5 eV; transfer vacuum high, !O -4 Pal or by particle bombardment - atomis-
Tabl., i. Examples of coatings Process
PVD, ion plating
Application
Coating
Material
tiN
Thickness
Hardness
(/Lm)
(HV)
3 to 8
2300
Drills, milling cutters Cutting tools Metal forming tools
CVD
TiC
7
3500
Indexable inserts
CVD
TiC
4
3500
Antifriction bearings/nuclear engineer· ing
Plasma spraying
Cemented carbide
50 to 300
1600
Currentless deposition
Ni dispersion
10 to 100
550
Cylinder liners
Electroplating
Cr
10 to 50
900
Piston rods
Nuclear components
, ••
5.4 Surface Coating
(Fla. SOd). These are based on chemical reactions of gases. The process temperatures range from above 700 to 1500 DC. Here too the trend is towards lower temperatures. The reaction takes place between the metal joining gas (e.g. Tia.) and the reactive gas (e.g. CH.); the substrate (e.g. cemented carbide) may act as a catalyst. A third, inert or reducihg, gas perfortnS the transfer of the reaction gases. (In the example, TiC is precipitated [64).) In CVD coating, the energy is supplied by heating of the substrate (by radiation) and, lately, also by plasma discbarge or by means of lasers. A controlled laser beam is able to produce coating patterns, permitting local vari· ations in properties. Coating from the liquid state includes the application of organic coatings by brushing or spray painting, cold enamelling, buildup welding and laser coating. Explosion cladding, roll·bonding and powder spray coating are examples of coaring from the solid state. Powder coattng is used for corrosion protection or optical surface treat· ment. Thermosetting plasticS (based on epoxy polyester and acrylic resin) are applied to workpieces in an electrostatic field and the powder is baked at 150 to 220°C. In jlutdtsed bed stntering, heated workpieces are inunersed in fluidised powder (based on polyamide, polyvinyl chlor· ide and polyethylene). The powder melts to form a protective coating, the thickness of which is determined by the inunersion time. In electroplattng, coating takes place from the ionised state. The coating materials are Cr, Ni, Sn, Zn, Cd, etc. Pure metals or alloys are electrolytically deposited from an aqueous solution (an exception is aluminium, which is deposited from a non-aqueous solution). Metal ions are discbarged and deposited at the cathode; at the anode they pass into solution (where the anode is soluble). Deposition takes place according to Faraday's law, m = kIt, where m = the deposited mass, I = the current, t = time, k = a material constant. The rate of deposition is 0.2 to 1 fLm/min. CVD Processes
\tl
3
1"1:!i=A--r2
2 0.5 to SkV
b
a 0.1 to 1kV
Reaction gas =(>
Exhaust
J
L=(>
-~kr d
n ....... so. Coating from the vapour phase: a evaporation (PVD), b sputtering (PVD), c ion plating (PVD), d chemical vapour deposition (CVD); 1 substrate, 2 coatiog matetial, 3 cathode, 4 plasma. atton (sputtering) - (emission energy high, < 40 eV,
transfer vacuum lower, 1 to 10- 3 Pa) (1'iJ. SO). With vapour deposition, condensation takes pllice without a major change in the temperature of the substrate, whereas in sputtering a large change in temperature occurs owing to the high kinetic energy of the particles. Ion plattng combines advantages of vapour deposition and sputtering (Fla. SOC). The substrate carties a negative potential; the plasma is formed by glow discharge at a vacuum of 1 to 10- 1 Pa and a particle energy of 10 to 100 eV. The high impact energy removes foreign coating matter at the same time. In all PVD processes, the process temperature is < 500 DC, with a trend towards lower process tempera· tures so as not to affect the base material.
Assembly G. SeDger, Berlin
I Main groups
Groups
I
1
2
Primary shaping
4.1
1
I
4.2
I
2uAssembling
Filling
DIN 8593 Part 1
DIN 8593 Part 2
4.3
I
I
I
3
Metal forming DIN 8582
4
Cutting
4.4
I
Manufacturing processes
I
4.5
T
I Joining DIN 8593 Part 0
I 4.6
I
I
5
6
Changing of material properties
Surface coating
4.7
I
1
4.8
I
Pressing againsl Pressing into
Joining by primary shaping
Joining by metal forming
JOining by welding
Joining by soldering
Gluing
DIN 8593 Part 3
DIN 8593 Part 4
DIN 8593 Part 5
DIN 8593 PartS
DIN 8593 Part 7
DIN 8593 Part 8
Fipre 1. Classification and subdivision of the manufacturing process "joining" according to DIN 8593.
';.9
Manufacturing Processes • 6 Assembly
6.1 Definitions Assembling. Generic term for all operations directed towards the joining together of geometrically defined objects. Amorphous material may be additionally used for this purpose [1-3]. The manufacturing process joining, which accomplishes the actual process of creating a connection between two or more parts, should be regarded as the main function of assembly.
Elementary functions
Joining. Joining should not be equated with assembling. Although assembling is always carried out by means of joining processes, it includes the subsidiary functions of materials handling, adjusting, inspection and special operations. As main group 4 in the overall system of manufacturing processes according to DIN 8580, joining is classified into nine groups (Fig. 1) [4]. Materials HancUing. According to VDI Code 2860, Sheet 1 (draft), materials handling is the establishment,
divide combine
turn shift
retain
divide off
pivot orientate position arrange guide transfer
clamp
-------- -----Controlled storage Partly controlled storage
Composite functions
allocate branch off bring together sort
test
unfasten
unclamp
test presence test identity test shape test size test colour test weight test position test orientation measure
count measure orientation measure position
---
~ - - - - - - - - - ~- - - - - - - - - - - - - - - - - -
Supplementary functions
Loose storage'
a
(convey)' 'Not actually a handling function, but necessary for fully describing functional secuences
Classification criterion: Main function
Examples:
belt
separating device
pallet
feeder
store
switch
[=
grab
testing device
arranging device
receiver
measuring device
industrial robot
turnbuckle
sensor
b Figure 2:. Classification of materials handling according to VDI Code 2860, Sheet 1 (draft): a subfunctions; b breakdown of handling equipment into groups according to main functions.
6.3 Realisation of Assembly
defined changing or temporary preservation of a predetermined physical location of geometrically defined objects in a system of reference coordinates. The physical location of an object in the system of reference coordinates is defined by its orientation and position. The orientation of an object is the angular relationship between the axes of the object's own system of coordinates. The position of an object is the place that a specific point on the object occupies in the system of reference coordinates [5]. Materials handling is divided into the following functions (Fig. 2): Storage (keeping of quantities). Changing of quantities. Moving (establishing and changing a defined physical location). Securing (preserving a defined physical location). Inspection (measuring and examining completed handling operations) [5].
Adjusting. Generic term for all activities that are routinely necessary during or after product assembly in order to compensate for unavoidable deviations for manufacturing technology reasons with the aim of achieving specified functions, functional accuracies or product characteristics within preset limits [1]. Inspection. Inspection is divided into measuring and testing. Testing is ascertaining whether specifk characteristics or conditions are met. The result is binary in nature, e.g. of the rype good/bad or yes/no. Measuring is the term used when characteristics or conditions are ascertained by means of a predetermined reference value. Inspection occurs as a subfunction in all manufacturing sequences and stages [5].
Special Operations. These comprise activities that cannot be assigned directly to one of the above-named functions, but are nevertheless considered essential constituents of assembly. Examples are application of fluxes or securing of nuts with varnish [I, 3].
6.2 Tasks of Assembly At the interface with development and marketing, assembly as the final stage of the manufacturing process becomes a logistical orientation point of works management. In assembly, a technology- and procedure-related coordination of the productive factors takes place. Technologically, the ability of the products to function is demonstrated in assembly. Organisationally, the elasticity of
production in the face of demand fluctuations in the market is demonstrated in assembly. Product design and production resource planning for ease of assembly harbours a great potential for rationalisation. Figure 3 illustrates the siting of assembly between market, development, design and manufacturing [6]. Assembly in production is necessary for various reasons, e.g.:
Achievement of function-related mobility. Combination of various material properties. Simplification of manufacturing. Replaceability of wearing parts. Reduction of manufacturing costs. Testability. Increasing the variety of different models. Weight reduction [7].
6.3 Realisation of Assembly The Assembly Process This process is accomplished by the interaction of product-specific, production-resource-specific and cycle-specific variables. The product is described by parts lists and by the geometrical and technological characteristics of the components and modules to be assembled. The cycle is technologically determined by the individual assembly operations and their interrelationships. These can be illustrated graphically with the aid of the priority graph. A priority graph is a critical-path-like depiction of sub-operations of assembly and their sequential relationships (Fig. 4). Organisationally, the cycle structure is determined by the production programme and the control of assemhly. Control of assembly refers to the coordination and control of the cycle in order to complete the endproducts on schedule and in the specified quantity and quality. The production resources comprise all the function-performers in their interaction in performing the tasks of assembly.
Assembly Planning The goal of systematic assembly planning is to assist the planner in the individual plannipg phases from analysis through drafting and desllW up to the introduction of assembly systems. Information technology tools may be used to model assembly processes in order to increase reliability of planning and productivity.
Development and design • Design for ease of (dis-)assembly Product and process innovation
D M a n u f a
c
t u r
• Flow of information • Flow of matenals <:=> • Methods • Materials
Potential fields of assembly Products Production resources Organisation Quantifying
<=>
• Quality Range of different models • Delivery service • Environment
M a r k e t Figure 3. Siting of assembly between the market, development, design and manufacturing.
, :_,
Manufacturing Processes. 7 Production and Works Management
~
Loudspeaker cover
~\\~l~~~~\~\
Top shell of handset
LCD module
Loudspeaker Loudspeaker Sw~ching
rocker Housing
Bottom shell of handset Circuit board
•
Base tray
No. Action
Loudspeaker cover
10 11
Engage switching rocker F~ loudspeakers F~ foil Fit circuit board F~ LCD module Insert ends of foil Engage base tray Fasten base tray wtth screws Lock housing into unit Contact loudspeakers Engage loudspeaker cover
c
FJpre 4. Sample case of an assembly task - communication terminal (telephone): a exploded view, b parts list of structure, c operations for assembling the housing, d priority graph for housing assembly.
6.3 Realisation of Assembly
I :W.
Organisation forms of assembly . -_ _ __ _ Movement of product _ _ _ __ _-, being assembled
Place of
Time dependence of materials flow
Type of conveyor motion
Motional state of product during assembly
Flpre S. Organisation fonns of assembly [II.
Number of different products being assembled
Organisadonal Forms of Assembly Assembly systems may be classified according to the movement of the product being assembled into locally concentrated systems and systems distributed over several stations (Fig. S) [I]. A distinction is made between quantity.fJased and type·based division. Quantity·based division is the parallel performance of the same assembly
Flpre 6. Fields of application of different assembly equipment {9]: a automatic assembly machine , b flexibly automated assembly line, c flexibly automated assembly station, d mechanised individual workstation, e manual individual workstation.
operations, while type-based division is the sequential performance of different assembly operations at the respective capacity points.
Assembly Systems The variety of the components, their joining behaviour and different tasks of assembly lead to a diverse range of assembly systems [7].
':.1.
Manufacturing Processes. 7 Production and Works Management
assembly, joining behaviour, joining and handling kinematics, methods of supply, components, modules and joining sequences can be largely standardised for the product range to be assembled.
Automated Assembly Automation of assembly is intended to increase efficiency and productivity. Reduction of employee stress and increasing of product quality are also imponant. Automatic assembly devices are technical equipment items by means of which assembly operations can be automated, either fully or with manual assistance [3 J. Automated assembly systems consist of assembly stations, the links between them and the peripherals [2J. The characteristics of automated assembly systems are: The nature of their structure. The flexibility achieved with the assembly system. The extent of the automated areas [1 J. Figure 7 . Integrated mechanical and automated assembly (Bosch GmbH).
Depending on the quantity to be produced and the design of the product, the entire task of assembly is broken down on a quantity or type basis. Flexibility and the level of automation need to be adjusted according to economic criteria (Fig. 6). For the assembly of different products on an assembly system, a low flexibility requirement is desirable. Through product design for ease of
For efficient integration of manual and automatic assembly stations, standardisation of the materials flow is essential. In the case of physically separated manual and automatic assembly, standard transfer containers are necessary for direct transfer of components without intermediate handling (Fig_ 7). Scanning elements for position recognition and the use of coding with mobile data storage media or bar codes permit automated transfer. The use of standard transfer systems can achieve an integrated flow of materials. Product-specific devices facilitate automated positioning and orientation of the workpieces [8].
Production and Works Management H.-J. Warnecke, Stuttgan
subdivided into production planning and production control.
7.1.1 Production Planning
7.1 Job Planning Job planning comprises the sum total of all measures, including the preparation of all the necessary documents and the provision of production resources, which minimise the cost of manufacturing products by means of planning, control and supervision (definition according to Germany's AWF - Ausschuss fur wirtschaflliche Fertigung: Committee for Efficient Manufacturing). Job planning is
Production planning comprises all the one-off measures. These measures relate to the design of the product, the production engineering, the planning and the provision of the production resources and fmish with clearance for production (definition according to the AWF).
Tasks of Production Planning (Table 1) The main task is the preparation of the job schedule. Alongside the drawing and the pans list, the job schedule
Table 1. Scope of production planoing
Activities
Advice on manufacturing techoology
Methods planning
Materials planning
Operations and time scheduling
Planning of production resources
Cost scheduling
Advising of design dept. on designing workpieces for ease of manufacture and assembly Checking of drawings
Planning of new methods and procedures Preparatio n of planning documents Experimental procedure comparisons
Specification of material (unmachined dimensions and shape) Optimising of waste Material storage planning
Preparation of job schedule Product classification Work instructions Design of: working methods workplaces Allowed time system
Production facility Machines Jigs and fixtures Tools Measuring and testing equipment Conveying equipment and storage facilities
Cost of materiaJs Cost of working equipment Labour COStS Initial cost accounting Follow-up COSt accounting
7.1 Job Planning. 7.1.1 Production Planning
I I
I '*'
Quanli~ ordered Uril pes Batch size Dale 01 issue 1Ij(Jo pes. 500 6. 1.
Processing lime
Product name
Drawing 00.
/f~
':TO -:rIO
"'if Material quantity
,,11
Cos! loc.
Workstation
1.9.> r.rMtolM
Unit
Ie•.
Material Lid blank GG-22
OperaOperation lion no ~ 1.<>#Is I
RAAIot'l
1:J.1.'·~s
",,"~./l'>Fila.eJ
,l.
$-<'1/., d~bw>'
IS~no
.. 30 Cost type account
&30-3'0.11
.r-/oPL
~
Type.
=,
lOOJigJ Ilxture
tk..
Job schedule no '0 2.1 I?
;r,l.
~zeJmodeI "'.
lid UNfK r;t; - 2.')..
..
.:
~ -.....-. :JJr;f(,~ji1. ~,it . /3 ""If
.r/~£'.jiw I<';~I lilt.. . /3
Wage group
"
Ir
I,
~
.f~
,..
ff'
'"
S
72- 7.0
C'_;"'" q;A..';1.~
.2~
,.,-...
~Aif·
:s
ffjjJ.,;~
jIf;d'~''''IIlCf!!i
Figure 1_ Job schedule for a manufacturing operation with chip removal, according to Sonnenberg.
C"~elr4.~
is a further basic document in the technical organisation of the firm. The job scbedule data comprise drawing, parts list and contract data. The infonnation which a job schedule should contain is determined by the tasks to be accomplished in the various divisions of the firm.
ties, wage group, preparation time, time per unit, and notes where necessary. The time data are usually based on the breakdown of the allowed time according to REFA (Fig. ::I). Allowed times are target times for operations performed by people and production resources. In the case of mainly manual operations, systems of predetermined times are often used. These are methods by which times can be determined, with the aid of time tables, for the performance of those operational elements which can be fully influenced by people (e.g. manual assembly). The best-known methods are MTM (methods time measurement) and
Job Schedule. This generally contains the following infonnation (Fig. 1).
Title Block Data. Part name, part number, material of construction or raw material, dimensions, batch size range, name of person in chatge, date, clearance indication or validity.
work factor. Application. The job schedule is intended first and fore-
Operation-Describing Data. Operation number, cost location, designation of operation, machine number, machine name, necessary tools, fixtures and testing facili-
most as working instructions for manufacturing. However, the job schedule data are also the basis for the following:
Allowed time I
I
for the production resources
,for people
Basic prepa- Preparation ration time recovery time trer Irg
Basic time
I"
Ig
Recovery time ler
Allowance
Production resources basic preparation time
Production resources preparation allowance
Iv
IrgB
I"B
Production resources basic time 19B
Production resources allowance IVB
Figure 2:. Breakdown of allowed time according to REFA (REFA·Verband fur Arbeitsstudien und Betriebsorganisation e.V., Darmstadt); m = number of units.
'Mf:'
Manufacturing Processes. 7 Production and Works Management
Time scheduling of operations, determining the capacity requirements of machinery and personnel, materials planning, planning of production resources and procurement. Preparation of contract documents, routing cards, wage slips, production resources supply lists, materials supply lists. Initial, interim and follow-up cost accounting, valuation of repeat work and rejects. Long-term planning tasks, organisation of data management if using EDP.
technological input data by which the working cycle is influenced.
Replanning PrinCiple, This is based on a universally valid analysis of the manufacturing process. Based on a description of the blanks and the fmished parts, the job schedule data are calculated by means of an inter-plant planning logic. Alternative solutions may be optimised according to predetermined target criteria (minimum cost, minimum time, or optimum alternative strategy in the event of capacity bottlenecks).
7_1_2 Production Control Computer-Aided Production Planning With regard to the use of computers, a distinction has to be made between job schedule management and job
schedule Issuing. Job Schedule Management_ This expression is used when the job schedule data are ascertained by the planner in the conventional way and recorded on a form. The job schedule can then be input into the computer and stored in a job schedule master me. The stored job schedules may be output at any time, with the addition of up-todate customer contract data if necessary. The advantage of computerised job schedule management lies in the fact that the stored job schedule data are available as input data for further computer programs, e.g. for time scheduling and expediting as well as materials and time management.
Job Schedule Issuing_ In this case, the computer performs some of the activities of the job scheduler (Fig_ ~). Based on a description of the production task, the job schedule data are calculated by machine and the job schedules prepared by means of a programmed planning logic and the corresponding meso The program systems so far developed for computer-aided job schedule issuing can be traced back to two principal planning methods: the "variant principle" and the "replanning principle" [1, 2].
Variant Principle. Here, for workpieces of the same kind a standard solution is developed in the form of a basic type with an associated job schedule, which embodies the respective individual solutions by means of variations of the basic type within preset limits. An up-to-date job schedule can be prepared by varying geometrical and/or
MarUacturillg lask
- ~'-
o oec=.IST.Cion","s---, < = 8 =>rAid,,,., ·,,..S- - - - ,
""i
• Sections of
00
• Coding system
::::. JjJ "::''' EDP program Files
W 0 ~Od
Figure 3-. Sequence of computer-assisted job schedule issuing.
Production control comprises the measures required to perform a contract in the sense of production planning (definition according to the AWF). It plans and supervises the flow of the contracts, particularly in the area of production. Its special responSibility lies in efficient capacity utilisation, ftxing of key dates and contract fulfilment. Production control, with its two functions materials management and time management, is an integral part of the operational or techno""rganisational information systems [3]. Figure 4 illustrates the main functions of such a system, which is particularly suitable for contract flow control. The production process and the associated planning, control and supervisory system form a unit in the form of a control loop. The large quantities of data to be processed and the necessary high-speed transmission of the control information are increasingly leading to the use of EDP systems in production control. Furthermore, such computerised systems enable complex planning models and methods to be applied which may signiftcantly improve the profttability of the operational processes and thus the ftrm's trading result [4].
Materials Management The task of materials management is to plan, control and superVise the materials in the form of modules, individual parts, raw materials and consumables. It follows from this that the most important goal of materials management is to ensure high availability of the materials required for component production and for assembly. Under the aspect of cost minimisation, further objectives are low capital tie-up by virtue of low warehouse and short-term stocks, low planning and procurement costs, and high machine capacity utilisation due to coordinated provision of materials. The stated objectives are partly conflicting. The desired direction must be established as part of the stock-keeping policy of the ftrm. Figure S shows the various subtasks of materials management, on which the following comments may be made [5].
Stock-Keeping Polley_ Establishment of guidelines for the level of availability of the material and in-process stocks, the maximum level of capital tie-up and frequency of ordering; planning of minimum and safety stocks for every stock item; and coordination of procurement and production. Demand CalaJIation
Gross Demand Calculation. Calculation of the gross demand for modules, components, raw materials (secondary demand) and consumables (tertiary demand) on a quantity and due date basis by derivation from the customer orders received and/ or from the current production programme (primary demand), or by extrapolating or estimating the demand trend based on past deinand.
7.1 Job Planning. 7.1.2 Production Control
':.@'
Planning a/ld COI1trol system
WOO
Production plOcess
~c=> ~
I Di~t~ I
L::::::::j
Supervision system
Figure 4. Structure of operational information systems [5].
Materials management
Availability control
Figure S. Scope of materials management.
Net Demand Calculation. Calculation of the net demand by offsetting the gross demand against the corresponding stock level, and possibly against workshop. order and reserved stocks.
batch sizes), taking into account preparation and storage costs, the production rate and similar factors.
Order Calculation
mum production run quantity and the forecast future
Batch Size Determination. Calculation of economic production run quantities (optimum order and production
Order Quantity Calculation. Establishment of the quantities to be produced on the basis of the calculated optidemand situation.
Calculation of Order Dates. Calculation of the order dates based on the dates when the respective goods are
.:.,,1,.
Manufacturing Processes. 7 Production and Works Management
times and the calculated optimum production run quantities.
Production Sequence Planning. Establishment of the chronological and physical production sequence for all production orders.
Stock Management
Survey of Capacity Utilisation. Periodic totalling of the
required, taking into account processing and replacement
Stock Recording and Updating. Recording and accounting of additions to and withdrawals from stock in terms of quantity and value, separately for different types of stock.
Stock Statistics. Records and statistical analyses of stocks and consumption, e.g. according to types of material and components as well as products. Stocktaking. Recording of the actual stock level and comparison with the stock accounts, making corrections if necessary. Stock Rationalisation. Checking of stock levels in respect of items with an above-average storage time ("slow movers"), possibly initiating their scrapping.
Availability Control. Checking whether the available stock levels satisfy the requirements for the planned production and assembly jobs. Organisation of Parts Lists. Management of the product structure data and preparation of parts lists and component use records of various kinds. Time Management The field of time management encompasses the planning, control and supervision of all the firm's manufacturing operations. It consists essentially of the chronological aSSignment of production orders to machines or workstations. This task is characterised by the goal of on-schedule completion of products at the lowest possible cost. From this the following subgoals of time management, some of which coincide with those of materials management, can be derived: high utilisation of the available capacity (production resources and manpower), short processing times for production orders and low capital tie-up in the firm's current assets. The goals of "high capacity utilisation" and "short processing time" conflict ("operations planning dilemma"). Figure 6 shows the subtasks of time management [6].
Job Scheduling Strategy. Establishment of priority rules according to the goals to be achieved as a matter of priority.
Scheduling of Processing. Calculation of the starting, interim and completion dates based on the operation and transfer times, without taking the available capacity into account.
utilisation values per production capacity (capacity group or individual capacity), which are derived from the scheduling of processing; graphic representation of the utilisation situation per planning period and/or production capacity.
Job Distribution. Allocation of contracts and job documents to the individual capacities; establishment of the definitive starting date for the work and the definitive order of the individual operations.
Capacity Balancing. Coordination of the orders with processing schedules (capacity demand) with the capacity actually available (capacity supply) by deferring entire orders or individual operations (physical capacity balancing allocates alternative capacity to the order or operation, while in chronological balancing new production deadlines are established). Production Supervision. Supervision of the measures taken by means of place-specific and time-specific expediting.
Sequence Planning. Establishment of the sequence in which orders waiting for production capacity are to be processed (assignment of job and operation priorities).
Job Schedule Organisation. Management of the job schedules by amendment, deletion and addition of job schedule data.
7.2 Manufacturing Systems 7.2.1 The System "Manufacturing"
The performance of a manufacturing task requires various coordinated operations which are carried out by subsystems of manufacturing. Figure 7 shows the functional structure of manufacturing and the linking of the dynamic subsystems by the flow of materials, energy and information. The subsystems, which are installed singly or in multiple, perform the follOwing subfunctions (by analogy with [7]):
Work System. Changing of geometrical and/or material characteristics of the workpieces in line with the manufacturing task (e.g. with the aid of a machine tool). Control System. Processing, transmission and storage of technical and/or organisational information.
Figure 6. Scope of time management.
7.2 Manufacturing Systems. 7.2.2 Automation of Materials Handling Functions
';WI,II
Blanks/semifinished products Finished products
Waste products
~
l
Consum/. abies
~
LLLL/.~< /.LLL£'(~'(LLLL//.LLL.-'j
/.LLL£/LLLL'(
Energy
Energy Supply System. Conversion, transmission and storage of the energy required in all subsystems. Workpiece Handling System. Storage, supply, positioning, clamping and transfer of the workpieces. Tool Handling System. Storage, supply, damping and changing of the tools. Measuring and Testing System. Comparison of the actual values with predetermined target values. Consumables Supply System. Supply of the consumables required for the manufacturing process in the work system (e.g. coolant). Waste and Consumables Disposal System. Removal of the unconsumed consumables an the waste generated during the manufacturing process (e.g. chips). Because of technological progress, the demands made on people in the system "manufacturing" have changed as fotlows: Relief of physical stress on human beings through mechanisation of the energy supply system and the work system, partial relief from control of the work system, and mechanisation of transport tasks, and full relief from manuat activity and control function in mass production through the use of production resources such as transfer lines; the direct link between people and the work system has been abandoned. In series manufacturing, this state still has to be achieved (e.g. by using "flexible manufacturing systems"). The trend is in the direction of the "automatic factory", which enables maximum productivity and quality while minimising the direct ties between people and the manufacturing process.
7.2.2 Automation of Materials Handling Futtctions For an analytical description of handling processes, these processes are broken down into individual handling functions (in accordance with VOl Code 3239: Supply Functions). Each function can be represented by a symbol and an associated code number. The functions that are important for the use of handling eqUipment are described by the characteristic functions, the performance of which may require several installed functions (ct. E4.3.1O). The automation of materials handling processes means
Figure 7. Functional structure of the system ~manufacturing" .
that account needs to be taken of the handling characteristics of the workpieces (object,; to be handled), the conditions of the respective manufacturing equipment and the technical possibilities of handling equipment, as well as their interdependence [8]. Owing to the wide range of these influences, handling devices are usually tailor-made individual solutions. The associated high cost of development often means that automation is possibly only for frequently recurring handling tasks (large-scale series production or mass production) if standardised handling equipment cannot be used, the further development of which is favoured by the progress achieved in control technology and information processing. Feeding, delivery, transfer and similar handling functions are performed with insertion devices, programmable handling devices ("industrial robots") and telemanipulators.
Teleoperators. These are remote-controlled manipulators [9] without program control. Control is performed by a human controller, who takes the necessary decisions and initiates the motions. Teleoperators are strength, per-
formance and range boosters for the human handling characteristics. If an appropriate communication system is available, the teleoperator may be set up and work at any desired distance from the human controller. In industry, heavy load manipulators are used where it is desired to relieve people of heavy physical work, but control of the motion sequences must remain in the hands of human operators (e.g. in nuclear engineering, marine engineer-
ing, space engineering).
Insertion Derices. These are mechanical handling devices, usually equipped with grabs, which perform predetermined motion sequences according to a fixed program [10]_ They work at presses ("iron hands"), on assembly lines, in the packaging industry, etc., i.e. wherever the same handling task is to be performed over a long period. Industrial Robots. In contrast, these are automatic handling devices equipped with grabs or tools which are designed for industrial use and which are programmable in several axes of motion (Fig.8) [11]. The difference between them and insertion devices lies in their programmability and in their usually more sophisticated kinematics.
Materials handling tasks can normally be automated only with the aid of robots if some of these tasks
I:W'••
Manufacturing Processes. 7 Production and Works Management
Assembly. Joiniog, gluing, screwing, pressing in, joining by diffusion, riveting, soldering, mounting electronic components on printed circuit boards. Workpiece HancWng
Axis N Axis 'l
Figure 8. Programmable handling unit - industrial robot (Volkswagenwerk AG). The working space (hatched areas) results from the rotary motion of axis I and the pivoting motion of axes II and III; the working space can be extended by means of the hand
axes IV (dot-and-dashed line) and V (rotary motion).
(especially arranging in component manufacture, arrangiog and positioniog in assembly) are taken over by otber equipment. For arranging, there are two possibilities. The reqUired arrangement state is produced by placing every workpiece io a predetermioed location and position.
The arrangement is recognised, and the location and position of every workpiece are determioed. For this, sensors record specific features of the workpieces, a simple control system processes these data witb tbe aid of a predetermioed "iotemal model" , i.e. a program, and derives from this signals for controlliog the handling device. In the process, the volume of data is reduced to the degree acceptable for solviog the problem at hand, tbus achieviog simple, rapid processiog. The tasks that are performed today by iodustrial robots can be divided into workpiece handling and tool handliog [12].
Tool HancWng Coating. Paiotiog, enamelliog and spray application of adhesive. Spot welding. Weldiog of car bodies. Continuous welding. Guiding of a welding torch (in gas-shielded weldiog, often with the aid of sensors which find the beginning of the seam and guide the torch in the middle of the weldiog groove during welding).
Handling at presses, forging machines, pressure die casting and injection moulding machines. Loading and unloading of machine tools or test systems. Palletising, commiSSioning, interlinking of two or more machines. 7.2.~
Transfer Lines and Automated Production Lines
In automated production, the Taylor principle of division of labour is particularly marked in transfer lines. A transfer line is a production line in which workpieces are passed from machining station to machining station in a cycle (automatic workpiece processing). Transfer lines are designed as special machines for machining workpieces, usually in large quantities, which are very similar in their manufacture. Their typical field of application is the motor vehicle industry. Although transfer lines are single-purpose systems, they are largely assembled from packaged units on the modular construction principle (cf. L5.12). These packaged units are self-contained modules, each of which embodies one or more subfunctions of a machine tool. A distinction is made between basic units, main units and supplementary units [13]. To ensure that units from different manufacturers are interchangeable, the principal main and mating dimensions of the various equipment sizes are standardised in DIN 69 512 et seq.
Rigid Interlinking. The stations of a transfer line (Fig. 9) are rigidly interlinked. The characteristics of this are: common control of the interlinking device and tbe machining stations; processing of workpieces on an even cycle prescribed by the slowest work cycle; a fault in one machining station causiog the entire production system to shut down. Loose Interlinking. In comparison, the characteristics of loose Interlinking are: iodependent work cycles of tbe machiniog stations (no common cycle); greater freedom in installing and positioning the machining stations; a workpiece reservoir as a breakdown buffer with physical and chronological unlocking of the workpieces between the machining stations; tbe ability of the preceding and following machining stations to remaio in operation io the event of a breakdown; and consecutive downtimes not being cumulative provided the capacity of the breakdown buffers is sufficient. Series machines are often connected by loose interlinking; it also enables manual and automatic machining stations to be uncoupled in production lines. Combinations of loose and rigid interlinking are found especially in automatic production lines with many stations, machining stations with the same technological operations and productivity being rigidly interlinked. In this way tbe advantages of tbe direct, short workpiece passage on tbe one hand and the section-by-section bridging of downtimes by means of buffers are partially combined.
7.2.4 ftexible Manufacturing Systems Workpiece and Tool HancWng Machining. Deburring, grioding, polishing, water jet cutting, cleaning of castings, laser or plasma cutting.
A flexible manufacturing system consists of several individual machines that are generally numerically controlled and loosely interlinked and is able, owing to the linkage with regard to material flow and infonnation technology,
7.3 Quality Engineering. 7.3.1 Scope of Quality Assurance
'Mlli.
followingjimcUons or equipment items usually form part of a flexible manufacturing system:
Workpiece changing (by means of handling devices in the case of rotationally symmetrical workpieces, by means of workpiece cartiers (pallets) and pallet-changing devices in the case of prismatic workpieces). Tool changing (changing of individual tools, multiplespindle drill heads or tool magazines). Consumable supply and disposal, chip removal system, automatic control of subsystems (NC, CNC), adaptive control. NC data distribution with a DNC computer (direct numerical control data distribution computer). Buffering and storage equipment (depending on the workpiece spectrum, machine set-up and type of interlinking), automatic washing and cleaning of workpieces, automatic measurement of workpieces. Computer control of the entire system including transfer control. Computer-aided capacity and scheduling calculation (organisational control) by a higher.ranking production control computer. Online operational data logging. Automatic fault diagnosis (monitoring).
Fipre 9. Transfer line for machining steering knuckles; outline and machining sequence (Mauser Schaerer GmbH). The jig carriages are returned to the loading and unloading station (station 1) by means of inclined hoists; cycle time 0.96 min.
to machine workpieces automatically in medium-sized and small batches down to a minimum batch size of one. Various workpieces undergo machining at the same time, passing through the system on various paths. The machining stations within the system may substitute for or complement each other with regard to the installed production functions. Substituting stations have the advantage of high utilisation of the system as regards time, optional work· piece passage and high flexibility of the overall system, as they can each alternately perform the same machining tasks and have the same technological functions and working geometty. Complementaty stations only perform machining steps that cannot be performed on other sta· tions because of differing technical functions and working space geometty. Systems with complementaty stations have a high technical utilisation, realise the line principle and have high productivity. The interltnking of the machining stations in flexible manufacturing systems may take the following form [14): lnterlinkage by means of a single transfer vehicle (e.g. a high·bay storage and retrieval unit, or a mobile industrial robot). lnterlinkage by means of several transfer vehicles (e.g. an inductively controlled industriaI truck). lnterlinkage by means of a fixed conveyor line (e.g. a frietion-driven roller conveyor). Depending on the level of development of a system, the
Figure 10 illustrates the fundamental structure of a flexible manufacturing system. The machinery and equipment control systems perform the processing of NC programs and transfer data and the logging of the operational data. A higher-ranking process control computer performs the tasks of distributing and managing the NC programs (DNC cf. U), control of the flow of materials and logging of further operational data. The process control computer is linked to a central in-house computer which carties out the tasks of production planning, production control and the management information system (MIS).
7.3 Quality Engineering 7.3.1 Scope of Quality Assurance Quality engineering is responsible for the lion's share of the quality assurance tasks. These tasks comprise all the organisational and technical activities for ensuring high product quality while taking profitability into account. They can be divided into the following sub-activities in accordance with DIN 55350 Part 11 and [15): Quality Management. Performs the overall management task to establish and implement the quality policy. Qualtty Planning. Selects quality features, classifies and weights them, and puts all the individual requirements as to the condition of a product into concrete form, taking into account the feasibility of their realisation. Quality TesUng. Ascertains the extent to which a unit meets the quality requirement. The quality requirement corresponds to specified and assumed requirements imposed on a unit. It may be documented in bills of quan· tities, specifications, drawings and the like. Quality tesring particularly involves the planning and performance of tests.
1;.1.11
Manufacturing Processes. 7 Production and Works Management
12
00000 Figure to. Flexible manufacturing system for non·rotating parts (Burkhardt und Weber Group): 1 pallet store, 2 loading and unloading station, 3 pallet transfer vehicle, 4 high-capacity tool store, 5 tool transfer device, 6 horizontal centre (working distances machining 1250 X 1000 X 800 mm), 7 CNC control, 8IC
9
adaptor, 9 hydraulic unit, 10 thyristor unit with power pack, 11 cooling unit, 12 washing station, 13 pallet transfer station.
IJ
Quality Control. Performs preventive, supervisory and corrective tasks in order to meet the quality requirement. It generally analyses the quality testing and corrects processes.
7.3.2 Quality Systems Quality systems establish the structural and procedural organisation for performing the quality assurance tasks. It describes the responsibilities, methods, processes and means for implementing the quality management. The individual contributions of activities or processes to quality and planning, implementation and utilisation phases may be represented as a quality circle (Fig. 11). Quality systems are normally documented in quality assurance manuals. The requirements for proof applying to such systems are standardised in DIN ISO 9001 to 9003.
7.3.3 Methods and Procedures To perform the quality assurance tasks, numerous methods have been developed and applied in industry in recent years. Some of the most important are:
Assembly and operation Purchaser (user)
Sales and distribution
Packing and storage
Marketing and market research Product definition and developmenl
Manufacturer
FMEA (Failure Mode and Effect Analysts). FMEA [16J is a method of estimating, at the planning stage, the risk of failure of a product component, a process stage or a system. Failure modes and their causes and effects are mostly determined, systematically recorded and valued in teamwork. If the valuation index, or the risk priority number, exceeds an acceptable maximum, then measures are drawn up to prevent or detect the potential failure or its cause.
Test Planning [15 J. For planning the quality tests, testing characteristics are selected and the testing systems, the frequency, and the method and location for their supervision are established. Planning of the testing frequency is carried out by statistical methods. As part of this, random sampling systems are used to evaluate a batch (cf. DIN 40080). CAQ (Computer-Aided Quality Control). A high proportion of the quality assurance tasks for planning and performing tests can be assisted with CAQ systems. Standard fuoctions of such systems are the preparation of testing schedules, testing order management, test data logging and analysis of, e.g. measurements, failures, quality costs. A CAQ system, as a subsystem in the entire spectrum of operational systems, should at least be combined with CAD (drawing data) and PPS (contract data) systems. SPC (Statistical Process Control) [17J. Mathematical statistical methods are an essential element of quality assurance in manufacturing technology (cf. test planning). Direct control of the quality of the manufacturing process is achieved by means of quality control cards for variable and attributive characteristics. Observation of statistical variables such as mean, span or standard deviation enables the capabilities of machines and processes to be assessed. The computer systems used for this purpose also permit automatic data logging and control of measuring operations.
Quality tests and inspections
Production
Process planning and development
Figure 11. Quality circle.
7.3.4 Testing Systems In mechanical engineering, it is mainly testing systems for geometrical length testing that are used. A distinction has
':W".
7.4 Operational Costing. 7.4.2 Types of Cost
Cost Location Accounting. This is carried out with the aid of a manufacturing cost sheet (MCS); it distributes the costs that are not directly attributable to the product (overheads) among the cost locations.
to be made between representations of measurements (gauge blocks, gauges, yardsticks) and measuring means (measuring tools, measuring instruments and measuring devices). As the level of automation in manufacturing technology increases, computerised measuring instruments are being used. Flexible·use CNC-controlled coordinate measuring instruments [18] with mechanical or optical data recorders are especially widespread. The use of a computer enables the control programs to be written away from the tnachine. This increases the main utilisation time of the capital-intensive appliances. The facility to gen· erate the control program straight from CAD data rationalises programming and avoids errors in transferring data caused by the manual input of data taken from the drawing.
Cost Unit Accounting. In the fonn of time-based cost unit accounting (operating statement), this calculates the profit as the profit per period, while item-based cost unit accounting (cost accounting) determines the cost per product. The cost type account and the cost location account are accounts for specific periods. The time·based cost unit account also relates to a specific period. The item·based cost unit account, on the other hand, is an account for a specific item. Costing is essentially perfonned in the following stages [20]: (1) recording of the costs by type of cost, (2) allocation of the costs to cost locations or cost units, and (3) use of the costs to measure operational activity for monitoring operational behaviour and/or for planning purposes. Costs can generally be broken down according to two aspects: (1) cost breakdown into direct costs and overheads according to their attributability to a cost unit; (2) cost breakdown into fixed and variable costs according to their response to changes in employment.
7.4 Operational Costing 7.4.1 Fundamentals of Operational Costing The operational accounting organisation has the task of recording and monitoring all the operations of procurement, production, sales and fmancing in terms of quantity and value. It is institutionally broken down into financial accounting, costing, statistics and budgeting. Costs are the consumption in value tenns of goods and services for the production and sale of operational outputs and for maintaining the necessary operational readiness [19]. The purpose of costing is to keep a check on the profitability of the output production process by reo cording, distributing and allocating the costs incurred in performing the finn's objectives. In detail, costing is the basis for [20] cost accounting (bid price, price limit), manufacturing control (comparison of costs and profits, comparison of budgeted and actual costs), operational planning and oper-
7.4.2 Types of Cost Accounting by cost type covers all the costs incurred in a finn in the procurement, storage, production and sale of operational outputs during a work period. In addition, it delimits the costs vis-a·vis the expenses of the finn as a whole. The importance of accounting by cost type lies in its subdivision of the overall costs into individual types of cost and the resulting possibility of attributing the individual costs to cost locations and cost units according to their origin (cf. E2.5.3). According to the most important operational functions, a distinction is made between procurement costs, storage costs, production costs, administrative costs and marketing costs. According to their source, five types of cost can be dis-
ational policy. Costing as a whole is divided into three areas (Fig. 12) [19]:
Accounting by Cost Type. This is used to record the costs in full detail by type.
Cost type accounting Class 2 Delimi· tation of financial account· ing in relation to produc· tion accounting
Class 4 Delimi· tation of extra· ordinary expendi· ture Calcula· tion of costs for cost account· ing purposes
Cost un~ Jccounting
Cost location accounting
}" "-
"-
"-
"-
"-
~
"-'-
---
Types of cost
Overheads
-
Cost locations
IT
}
Ext
d'
,--
Manufacturing cost sheet (MCS)
"-
I I I
.Cost unit I
Direct costs
t
__ ~~~~~s
HKS
I T
r-I--I I I I I I I I I I I ~-
HKS
HKS
HKS
I I
I I
I I
I I I I I I
I
I
I
I
r-I- - -
I
I I
L.L.i_-1-_1 __ I
IL _ _ _ _ _ _ _ _ _ _ _ _ _ _ ..l.. I __ _
HKS =main cost locations Figure 12. Costing system according to Schonfeld.
Cost unit IT
Cost unit
m
I.M
Manufacturing Processes. 7 Production and Works Management The tasks of the MCS are:
tinguished [20]: (I) labour costs (wages, salaries, subsidiary wage costs, entrepreneuriaJ income), (2) material costs (cost of raw materials and supplies), (3) cost of capital (interest, depreciation, capital risks), (4) cost of outwork (cost of repairs, transport services) and (5) the costs Of human soctety (taxes in the nature of costs, fees, contributions).
Distribution of the primary overheads among the cost locations according to origin. Allocation of the costs of the general cost locations to subsidiary cost locations. Allocation of the costs of the ancillary cost locations to the main cost locations. Calculation of the overhead surcharge rates for every cost location by comparing direct costs and overheads. Rechecking of the costs charged, i.e. calculation of the difference between the envisaged costs charged and the actual costs incurred. Control of the efficiency of the cost centres by calculating indices [22].
7.4.3 Cost Location Ac:couatina and the Manufacturina Cost Sheet Cost location accounting lies between accounting by cost type and cost unit accounting (cost accounting). In firms with a diverse production programme, it enables the overheads to be allocated to the cost units in a way which best corresponds to their origin. While the direct product costs can be attributed directly to the cost unit even without cost location accounting, in the case of the overheads the lack of a cost location account would make for a very imprecise distribution of costs. By the establishment of cost locations (accounting sectors) within a firm, the overheads can be recorded location by location and allocated to the products with the aid of a special distribution code according to the demands made on the location by the product (Fig. 1:J). As individual cost locations (e.g. energy generation) pass on in-house outputs to other cost locations (e.g. manufacturing), a compensation of the in-house outputs must be performed as pan of cost location accounting. On a formal basis, cost location accounting is carried out with the aid of a manufacturing cost sheet (MCS), which lists types of cost and cost locations in tabular form as rows and columns. Table :J shows an example of the design of a manufacturing cost sheet, in which the compensation of the in-house services has been dispensed with in order to simplify matters.
7.4.4 Calculation of Machine-Hour Bate The calculation of the machine-hour rate represents the furthest-reaching breakdown of the cost locations in cost location accounting. Here, the cost locations are individual machines. The total costs of a machine are termed machine costs. The purpose of such in-depth cost location accounting in the form of the machine-hour rate is to achieve increased precision in charging the overheads. The machine-hour rate is calculated according to VOl Code 3258 - costing with machine-hour rates - by applying the calculated machine costs to the planned or average customary annual period of use TN in h/yr: KMH
=
KA
+ K z + K. + KE + K, TN
.
Here, KA is the depreciation for costing purposes in moneyIyr. It is calculated from the replacement value (including installation and stanup costs) according to
Table:l. Example of the structure of • simple manufacturing cost sheet according to [21] Types of cost
Figures in accounts
Cost locations Production division
Production cost location [ Salaries Ancillary pay Social welfare expenditure Consumables Office supplies Outside repairs Energy consumption DepreCiation Taxes Postage Advertising expensees Misc. expenses
2600 1800
Total overheads Direct I.hour costs
8000 4500 5300
Direct material Production costs
Overhead surcharge rates
Cost type accounting
900 500 400 400 350 250 100 150 350 200
300 800 300 100
50 40
Production cost location II 400 200 150 100
100
60
Production cost location III
200 150
300 50 50
10
40
50
1600 2000
1050 1000
800 1500
Materials division
Administration division
200 300 50
1200 100 180
100
200
20 20
80 40 100 50
Marketing division
500 200 70
300 100 100 50 40 100 350
10
30
60
700
1980
1870
14000
14000
14.1%
13.4%
5350
80.0%
105.0%
53.3%
13.1%
Cost location accounting
7.5 Basic Ergonomics
.:.11*1
fering repair-pronenesses of various types of machine should be taken into account.
Production materials. Material overheads
7.4.5 Cost Accounting
Cost of materials
C_':!:J
Direct labour costs
+ r---,
+
L!~~J
Production overheads
L--[lIJ +
Production costs Special direct costs of production Manufacturing costs
[5EKJ
-~ +
C}~~J
Development and design costs Administrative overheads
[f~KJ
Marketing overheads
L __ J
rV~K'
+
Administrative and marketing overheads
The purpose of cost accounting is to allocate the costs incurred in producing the operational outputs and in selling these outputs commercially and in-house. Cost accounting may form the basis for: price calculation (initial cost accounting), price review (follow-up cost accounting), profit calculation, performance of comparative calculations and output valuation. Wherever several products with various material and manufacturing labour costs are produced by various manufacturing processes, surcharge cost accounting is used. This cost accounting method is based on separate allocation of the direct costs and overheads to the cost units. The direct costs are directly charged to the cost units with individual vouchers (e.g. material requisition slip), while the overheads are charged indirectly by means of overhead surcharges (cf. MCS). The procedure for calculating the cost price with the aid of surcharge cost accounting can be demonstrated by means of the diagram in Fig. 1~.
Cost
7.5 Basic Ergonomics Figure 13. Procedure for surcharge cost accounting.
business principles and applied to the expected useful life of the machine. K z is the interest for costing purposes in money/yr. It is applied at the normal interest rates for long-term debt. To simplify the calculation and for ease of comparison, the interest is calculated on half the replacement value. KR is the space costs in money/yr. They are usually applied to the floor area taken up by the machine including the necessary secondary areas. They include depreciation and interest on buildings and plant, building maintenance expenses, and the cost of light, heating, insurance and cleaning. KE is the energy costs in money/yr. They are calculated for electricity. gas, water, etc. on the basis of the actual average annual figures. K, is the maintenance costs in money/yr. They are to be cal-
culated for regular servicing and for uncapitalised repairs as average annual values over extended periods. The dif-
The subject of ergonomics is human work. Work in this sense is regular human activity directed at the creation of a permanent result, using one's physical, mental and psychological powers. Accordingly, ergonomics is concerned with the expressions of the characteristics of human work (loads) and their physical, mental and psychological effects on people (stresses). The results of ergonomic studies are used to create or change working conditions (workplaces, working procedures, environmental influences) in such a way that they can be termed humane in the broadest sense of the word [231. The adaptation of the working environment to people by methods engineering contrasts with the adaptation of people to the requirements of the work. This process can be assisted by instruction and training measures (cf. E4. 3.7). The starting data for designing workplaces are the dimensions of the human body. For this purpose, mean
Percentiles (dimensions in cm) Male
5%
5
6
95%
50%
95 %
61.6 13.8
69.0 28,5
76.1 35)
187.0
200.0
5%
Forward reach
66.1
72.1
2 Depth of body
23.3
27.6
78.7 31.B
J Upward reach
191.0
205.1
Z21.0
174.8
4 Body height
162.9
173.3
184.1
151.0
161.9
172.5
5 Height of eyes
150.9
161.3
172.1
140.2
150.2
159.6
6 Height of shoulders
134.9
144.5
Height of etbow above standing surtace
102.1
109.6
154.2 117.9
123.4 95)
133.9 103,0
143.6 110,0
8 Height of hand above standing surtace 9 Width of hips, standing
72.8
76,7
B2.8
66,4
73.8
80,3
31.0
34.4
36.8
31,4
35,8
40.5
70 Width of shoutders
36.7
39.8
41.8
31.3
35.5
38.8
(both arms)
J
Female
50 %
Figure 14. Body measurements of Gennan adults - standing - according to DIN 33402.
1;.111:.
Manufacturing Processes. 7 Production and Works Management
values and distributions of body measurements have been calculated in serial studies with representative samples (cf. DIN 33403). Some of these body measurements are shown in Fig. 14 for standing adults and Fig. 15 for seated adults. In physical work, it is usually the transmission of forces from the worker to the workpiece by means of tools and equipments that is to the fore. These tools and equipments must he designed so as to pennit the largest possible forces to be transmitted with little load on the worker. On machines, levers, handwheels, pushbuttons, etc. should be positioned in such a way that their operation approximates to a natural movement [24]. Informational work can be broken down into information receiving, information processing and information issuing. Information is received via the sensory organs, mostly by sight and hearing and to a lesser extent by touch, smell or taste. The receiving of information via the eye can take place only if the associated signals are supplied in the field of vision. The field of vision is a circle whose diameter d G (in metres) increases linearly with the distance from the eye a (in metres) according to the numerical value equation d G = 1.64a. The working area requires illumination, the strength of which depends on the type of work to be performed (Table 3). Further information can be found in DIN 5034 and DIN 5035. For visual means of information (indicating instruments) there are design possibilities of varying worth. For instance, round instruments are preferable to rectangular displays. Figure 16 compares common types of display [25]. Issuing of information to technical systems is generally accomplished by means of operating elements. Speech input is a possible alternative. Human efficiency is also affected by the environmental conditions (climate, nOise, dust) in which the work is done. The climate in working premises is described by the air temperature, the relative humidity and the air motion
velocity (Table 4). For an air space of 15 m' per person, the supply of fresh air should be at least 30 m' per person an hour, even in the case of very light work [26].
Table 3. Illumination required for specific visual tasks Level
Nominal illumination
Classification of visual tasks
(Ix)
I; 30
Orientation; temporary stay only
60 120
; 6
250 500
8
750 1000
9 10
1500 2000
It
3000 5000
12
Easy visual tasks; Jarge details with high contrast Normal visual tasks; medium-size details with medium contrast Difficult visual tasks; small details with low contrast Very difficult visual tasks; very small details with very low contrast Special cases, operating field
c.g.
illumination
of
Almost all work operations generate noise in some form. The noise is measured with instruments standardised to DIN lEe 651 by methods laid down in DIN 45635. The statutory order on health and safety at work (Arbeitsstattenverordnung) specifies 55 dB(A) as the maximum in premises where the work is mainly intellectual in nature, 70 dB(A) for simple office work and 85 dB(A) for industrial workplaces. Gases, dust and vapours are subject to MAC (maximum allowable concentration) or TLV (threshold limit value) values. They indicate the concentrations of noxious substances which, at an effect duration of 8 hours a day, are not harmful to health even over a prolonged period [27].
Percentiles (dimensions in cm) Male 5%
50 %
95 %
::emale 5%
50 %
95 %
71 Body height, seated
84.9
90.7
96.2
80.5
85.7
91.4
12 Height of eyes, seated
73. 9
79.0
84.4
58.0
73.5
78,5
13 Height of elbow above seat II, Length of lower leg with foct (seat height)
19.3
23.0
28.0
19.1
23.3
27.8
39.9
44.2
48.0
35.1
39.5
43.4
32)
35.2
38.9
29)
32.2
35.4
16 Sitting depth
45.2
50.0
55.2
42.5
48.4
53,2
17 Length from bottom to knee
55.4
59.9
64.5
53.0
58}
63.1
18 Length from bottom to sole of foct96.4
103.5
112.5
95.5
104.4
112.6
15 Distance from elbow to gripping axis
19 Height of upper leg
11.7
13.5
15.7
11.8
14.4
17.3
20 Width across elbows
39.9
45.1
51.2
37.0
45.5
54.4
21 Width of hips, seated
32.5
35.2
39.1
34.0
38)
45.1
Figure IS. Body measurements of German adults - seated - according to DIN 33402.
I
7.5 Basic Ergonomics
Analog display
Digital display
Designation
Round scale display
Sector scale display
Vertical scale display
Window scale display
Luminous bar display
Active element
POinter
Pointer
Painter
Scale
Bar
40
40 50 60
Z~~~~'~'~t/~'~O 1Ooi~
30
CD
10
~5~
10
0-
a
Reliable read-off Qualitative read-off
:! '"
Recogni- { quick changes tion of slow changes
c
~ Quantitative read-off <1l Comparison of displays en
o unsuitable
•a
a a a
~
~
a
a
()
()
a
~
•
523 * 567810 4711
10
0
•
very suitable
~
0
0
()
()
• •
a
•a •
•
~
~
0 0
a
Characters
~" BEl8 a • a
()
()
Numerals
30
~
a a
Screen (alphanumeric characters with dot matrix)
10
()
()
Electronic numerical display (line grid)
a
0
~
() suitable
~ limited suitability
a
()
a a
a a
Adjustment of values Control
:WI.g
•a •
()
~
~
extremely suitable
Figure 16. Comparison of common analog and digital displays according to [25].
Table 4. Climatic data for specific activities Air temperature
Activity
eel
Office work
Ught manual work, seated Light manual work, standing Heavy work
Very heavy work
ReI. humidity
Air motion
(%)
velocity (m/s)
min.
opt.
max.
min.
opt.
max.
max.
18 18 17
21 20 18
24 24 22
50 50
17
21
14
16
20
70 70 70 70 70
0.1 0.1 0.2
15
30 30 30 30 30
50 50
50
0.4 0.5
• • • • • •OAppendix K: Diagrams and Tables • • • • • • Correction values:
Cutting speed correction factor 2.023
Kv = V~.153 for
Vc
< 100 m/min
KM
= 1.05
Kw ,
= 1.0 (cemented carbide)
KM
= 0.9 to 0.95 (cutting ceramic)
Tool wear correction factor K~
for
Vc
= 20
(HSS)
to 600
K~
= 1.3 to 1.5 = 1. 0 (work-sharp cutting edge)
Cooling lubricant correction factor Kh
Rake angle correction factor K,
= 1.09 to 0.015 . 1:
K,
= 1.03 to 0.015·1: 0 (castings)
Cutting material correction factor
= 1 (dry)
K k , = 0.85 (non-water-miscible cooling lubricant) 0
(steel)
K k,
= 0.9 (cooling lubricant emulsion)
Workpiece shape correction factor K,
= I (external turning)
K, = 1. 2 (internal turning)
I
:W'I..
Manufacturing Processes. 8 Appendix K: Diagrams and Tables
AppeacUx K4 Table 1. k, •.• and 1 - m, values for ferrous materia1s Cutting conditions
Material
St 50-2 St 70-2 0:45 N 0: 45 V Ck60N Ck60V 40 Mn 4V 37 MnSi 5V 18 CrN!8BG 30 CrNIMoBV 34 CrNIM06V 34Cr4V 41 Cr 4 V 16 MnCr 5N 16 MnCr 5BG 20 MnCr 5N 20 MnCr 5BG 34 CrMo 4V 42 CrMo 4V 50 Crv 4V Ck 35 V Ck 55 N 55 NiCrMoV6V 100 Cr 6 GG30
v = 100 n min-I =3 mm Cemented carbide PIO
Cutting speed Depth of cut Cutting materia1s
Qp
ex
'Y
Steel castings
5° 5°
6° 2'
Rm (N/rom')
Specific machining forces k, •.• (N/rom')
559 824 657 765 775 873 755 892 618 971 1010 902 961 500 500 588 588 1000 1138 1050 622 661 1141 624 HB= 206
0° 0'
8
K
r.
90° 90'
70° 70'
0.8 rom 0.8 rom
ket .l
1-m,
k f1 . 1
I-mr
kpl.l
I-mp
1499 1595 1659 1584 1686 1662 1691 1656 1511 1704 1686 1536 1596 1411 1575 1464 1523 1632 1773 1698 1527 1396 1595 1726 899
0.71 0.68 0.79 0.74 0.78 0.77 0.78 0.79 0.80 0.82 0.82 0.78 0.77 0.70 0.81 0.74 0.76 0.80 0.83 0.78 0.72 0.65 0.71 0.72 0.59
351 228 521 364 285 337 350 239 318 337 291 327 291 406 391 356 356 276 354 295
0.30 -0.07 0.51 0.27 0.28 0.29 0.31 0.31 0.27 0.46 0.37 0.36 0.27 0.37 0.30 0.24 0.33 0.34 0.43 0.28 0.25 0.16 0.21 0.14 0.09
274 152 309 282 259 249 244 249 242 371 284 222 215 312 324 300 271 172 252 195 291 255 198 362 164
0.51 0.10 0.60 0.57 0.59 0.53 0.55 0.67 0.46 0.88 0.72 0.59 0.52 0.50 0.54 0.58 0.52 0.48 0.49 0.44 0.46 0.42 0.34 0.47 0.30
344
316 269 318 170
AppeacUx K4 Table 3. Machining forces for drilling [9. IOJ Material
18 CrNi 8 42 CrMo 4 100 Cr 46 MnSi 4 0:60 St 50 16 MnCr 5 34 CrMo 4
Rm (N·mm-')
I-me
600 1080 710 650 850
0.82 ± 0.86 ± 0.76 ± 0.85 ± 0.87 ± 0.82 ± 0.83 ± 0.80 ±
560 560 610
I-m f
ket .t
kfl.l
(N·mm-') 0.04 0.06 0.03 0.04 0.03 0.03 0.03 0.03
2690 ± 2720 ± 2780 ± 2390 ± 2200 ± 1960 ± 2020 ± 1840 ±
230 420 220 250 200 160 200 150
(N· rom-') 0.55 ± 0.71 ± 0.56 ± 0.62 ± 0.57 ± 0.71 ± 0.64 ± 0.64 ±
0.06 0.04 0.07 0.02 0.03 0.02 0.03 0.03
1240 ± 160 2370 ± 230 1630 ± 300 1360 ± 100 1170 ± 100 1250± 70 1220 ± 120 1460 ± 140
Grey cast iron
Up to G-22 OVer G-22
0.51 0.48
504 535
0.56 0.53
356 381
Manufacturing Processes. 8 Appendix K: Diagrams and Tables
'M"I
Appenclix &4 Table 3. Main and incremental values for axia1 plain face milling Cutting
Material
Cutting speed Vc
material
Cutting edge Main and incremental values for spec. machining force geometry in axial face milling
(m min-I)
5t 52·3N
HMP25
120
negative positive
1831 1469
0.29 0.25
809 447
0.54 0.57
705 174
0.41 0.56
Ck 45N
HM P25
190
negative
1506
0.45
708
0.62
653
0.52
X22CrMoV121
HMP40
120
positive
1533
0.29
497
0.70
164
0.77
Cutting edge geometry
'Yr
'Yp
fx,
fXp
A,
K,
negative positive
-4° 0°
-7° 8°
6° 9°
23° 29°
-6° 8°
75° 75°
90° 90°
K,
Uutd (mm)
60°/30%° 45'/0'
1.4/0.8/1.4 0.8/1.4
0.1 to 0.8 70 to 205
0.1 to 0.8
I,
c,
I,
I,
;;
Iz
Iz
P25
P40
MIS
KOI
KlO
K20
v,
v,
v,
v,
v,
v,
;;
PlO
HB
HB
35 to 130
0.1 to 0.4 80 to 250
0.1 to 0.8 70 to 180
0.1 to 0.8 35 to 90
0.1 to 0.8 55 to 140
0.1 to 0.4 90 to 165
110 to 200 200 to 265 265 to 450
HB
Unalloyed steel and steel castings
;; mmfU Vr.; m/min
Cutting material
Appendix K4 Table 4a. Negative wedge geometry; y = - 6°
HB
0.1 to 0.8 45 to 105
0.1 to 0.8 70 to 165
0.1 to 0.8 45 to 100
0.1 to 0.8 65 to 150
0.1 to 0.4 0.1 to 0.2 115 to 190 135 to 170
125 to 200 200 to 265
HB
0.1 to 0.8 25 to 70
40 to 105
0.1 to 0.8
0.1 to 0.2 95 to 120
280 to 345
HB
Alloyed steel and steel castings
HB
HB
5B
0.1 to 0.8 55 to 130
0.1 to 0.8 75 to 210
0.1 to 0.2 90 to 180 0.1 to 0.8 85 to 155
0.1 to 0.4 80 to 150
0.1 to 0.8 50 to 120
0.1 to 0.8 30 to 90
0.1 to 0.2 0.1 to 0.2 100 to 170 100 to 170
0.1 to 0.2 0.1 to 0.2 120 to 180 100 to 160
0.1 to 0.5 90 to 200
110 to 265 265 to 340 130 to 200 200 to 280
HB
Stainless steel and steel Grey cast iron castings
150 to 180
125 to 230
0.1 to 0.8 50 to 95
0.1 to 0.2 170 to 195
HB
HB
0.1 to 0.3 60 to 100
0.1 to 0.15 100 to 150
0.1 to 0.3 70 to 130
White heart malleable cast iron
Nodular cast iron
0.1 to 0.5 70 to 170
0.1 to 0.2 120 to 220
0.1 to 0.5 90 to 200
130 to 180
HB
Black heart malleable cast iron
Appendix K4 Table 4. Guide values for milling of ferrous materials. The guide values apply to pre-machined workpieces and stable machining conditions. In the case of unstable conditions or in the presence of mill scale, forging scale or casting skin, the stated cutting conditions should be reduced accordingly. f, feed/cutting edge; v, cutting speed
&1
~
[
l'
~ tI
~
rJ:l
• ~
~
il
~
""
s·
§,
~ 2
I
K20
KIO
K01
M15
P40
P2<;
j;
PIO
Vc
J,
VC
J,
VC
J,
VC
J,
l'L'
J,
VC
f.
VC
f.mm/U vern/min
Cutting material
0.2 to 1.5 35 to 135
0.1 to 1.2 120 to 200
0.1 to 0.4 80 to 160 0.2 to 0.6 30 to 70
0.2 to 0.8 40 to 80
0.1 to 0.6 70 to 130
0.1 to 0.2 80 to 120
0.1 to 0.2 50 to 90 0.1 to 0.4 30 to 90
200 to 265
265 to 450
0.1 to 0.4 65 to 110
280 to 345
0.1 to 0.6 40 to 80
0.1 to 0.4 40 to 90
0.1 to 0.4 95 to 130
26<; to 340
11 0 to 265
0.1 to 0.2 80 to 250
HB
HB
110 to 200
HB
HB
HB
200 to 265
HB
HB
Stainless steel and steel castings
Alloyed steel and steel castings
Unalloyed steel and steel castings
Appenclix K.4 Table 4b. Positive wedge geometry; 'Y = +6°
0.1 to 0.4 80 to 150
0.1 to 0.4 80 to 135
0.1 to 0.2 120 to 180
0.1 to 1.2 90 to 200
130 to 200
HB
0.1 to 0.4 55 to 110
0.1 to 0.4 60 to 110
0.1 to 0.2 100 to 160
0.1 to 1.0 80 to 150
200 to 280
HB
Grey cast iron
0.1 to 0.2 100 to 150
0.1 to 0.6 70 to 130
150 to 180
HB
0.1 to 0.2 120 to 220
0.1 to 1.2 90 to 200
130 to 180
HB
White heart malleable Black heart malleable cast iron cast iron
I
f
8-
~
i
?:i
C/J
•
I I
g. ""
o~
~
I
;W'"
Manufacturing Processes. 8 Appendix K: Diagrams and Tables
Appendix K4 Table S. Standards for grain size of abrasives (according to DIN 69100, VDI Code 3394) Average spacing of two screening machines
US Standard ASTM Ell
FEPA grain size designation
Designation according to DIN 848
Range
Range
Range
Average grain diameter (corundum SiC)
ILffi
ILffi
Narrow
Broad
Narrow
Broad
Narrow
Broad
1-300
420
0426
0350
0427
40/50
40/60 0356
1-150 297 0280 0301
50/60
0250 1-125
250
0251
60170
0220
0252
60/80
210
70/80
1-105
0213 Dl80
1-90
177
80/100
0181
Dl50
149
1-75 0140 100/120
0151 62
125 120/140
0126
0110 DlOO
105
I - 4S
1----140/170
0107
090 1-37
88
170/200
070
091
I- 31
065
74 200/230
076 1-27
63 230/270
064
270/325
054
055 1-22
S3
325/400 37 32
050 045
44 046
035 300/500
1-18
Manufacturing Processes. 8 Appendix K: Diagrams and Tables
Appendix K4 Table 6. Principal data for belt grinding Material
Rougbing
Finishing
Manipulated variables
Result
Specific rate of metal removal
Cutting speed
Q;,
(m/s)
Measured
surface roughness R1(I'm)
(mm 3 jmm· s)
Manipulated variables
Result
Specific rate of metal removal
Cutting speed
Q;,
(m/s)
(mm 3 /mm· s)
Measured surface roughness R z
40 to 60
>40
2 to 8
30 to 40
5 to 8
Up to 200
30 to 50
>50
5 to 10
20 to 30
8 to 10
Up to 100
20 to 40
>60
I to 5
30 to 40
\0 to 12
Steel for anti- Up to 150 friction bearings (62
HRe) Grey cast iron AlSi alloys
AppencUx K4 Table 7. Material pairs in eiectro-discharge machining Electrode material
Workpiece material
Machining operation
Electrode polarity
Quality of machining
Electrode wear
Notes
Copper-tungsten alloy
Copper-tungsten alloy
Rougbing Finishing
negative negative
good good
reasonable reasonable
Low metal removal capacity
Copper-tungsten
Steel
Rougbing Finishing
positive positive
good good
low low
Used for small, high-precision press tools
Copper-tungsten
Tungsten carbide
Rougbing Finishing
negative negative
good good
moderate moderate
Used for small, high-precision press tools
Copper alloy
Steel
Rougbing Finishing
positive positive
good good
low low
High cost of electrodes, ideal for small steel workpieces
Copper alloy
Tungsten carbide
Rougbing Finishing
negative negative
good good
moderate moderate
Ideal for tungsten carbide workpieces
Graphite
Cast iron
Roughing Finishing
positive negative
good good
low moderate
Higher metal removal capacity with negative polarity, but electrode wear is higher
Graphite
Copper
Roughing Finishing
negative negative
moderate moderate
reasonable reasonable
Low metal removal capacity
Graphite
Nimonic
Rougbing Finishing
positive negative
good good
low low
Good metal removal rate Reasonable metal removal rate
Graphite
Higb-speed steel
Rougbing Finishing
negative negative
moderate moderate
reasonable reasonable
Moderate metal removal rate Reasonable metal removal rate
Graphite
Stainless steel
Rougbing Finishing
positive negative
moderate moderate
moderate moderate
Good metal removal rate
Graphite
Steel
Rougbing Finishing
positive negative
good good
low low
Good metal removal rate High metal removal rate
Graphite
Stellite
Roughing Finishing
negative negative
moderate moderate
reasonable reasonable
Moderate metal removal rate
Graphite
Tungsten carbide
Rougbing Finishing
negative negative
moderate moderate
higb higb
Moderate metal removal rate; sparking is a problem if ti is too long
Steel
Steel
Rougbing Finishing
positive positive
poor poor
reasonable reasonable
Low metal removal rate Use for special purposes only (continued)
I:W'II
Manufacturing Processes. 8 Appendix K: Diagrams and Tables
Appendix K4 Table 7. Continued Electrode material
Tungsten carbide
Workpiece material
Steel
Machining operation
Electrode polatity
Quality of machining
Roughing Finishing
positive positive
low low
Low metal removal rate
good
moderate
Electrode
Notcs
wear
Tungsten carbide
Nimonic
Roughing Finishing
positive positive
moderate good
low low
Reasonable metal removal rate For machining small openings
Aluminium
Steel
Roughing Finishing
positive positive
good poor
low high
Stability doubtful for some grades of AI
Aluminium
Tungsten carbide
Roughing Finishing
negative negative
poor very poor
high very high
Not universally recommended
Brass
Copper
Roughing Finishing
negative negative
good good
reasonable
Use for special purposes only
reasonable
Brass
Steel
Roughing Finishing
negative negative
good good
high high
For narrow openings Reasonable metal removal rate
Brass
Ste1lite
Roughing Finishing
negative negative
poor moderate
high high
Unstable Reasonable metal removal rate
Brass
Titanium
Roughing Finishing
negative negative
good good
high high
Reasonable metal removal rate
Brass
Tungsten carbide
Roughing Finishing
negative negative
moderate moderate
high high
Not universally recommended
Copper
Aluminium
Roughing Finishing
positive positive
good good
low low
Low metal removal rate
Copper
Brass
Roughing Finishing
positive positive
good good
reasonable reasonable
Use for special purposes only
Copper
Cast iron
Roughing Finishing
positive positive
good good
low low
Reasonable metal removal rate
Copper
Copper
Roughing Finishing
positive positive
poor poor
Copper
Graphite
Roughing Finishing
positive positive
moderate moderate
reasonable reasonable
Use for special purposes only
Copper
Nimonic
Roughing Finishing
positive positive
good good
low reasonable
Reasonable metal removal rate Good metal removal rate
Copper
Stainless steel
Roughing Finishing
positive positive
moderate moderate
reasonable reasonable
Stability doubtful for some grades of steel
Copper
Steel
Roughing Finishing
positive pOSitive
good good
low low
Good metal removal rate. Never use negative polarity
Copper
Stellite
Roughing Finishing
positive positive
good good
reasonable reasonable
Reasonable metal removal rate
Copper
Tungsten carbide
Roughing Finishing
negative negative
good good
high high
Reasonable metal removal rate Stability becomes problematic if t; is too long
Not recommended
Manufacturing Processes. 8 Appendix K: Diagrams and Tables
':.111
Appendix K4 Table 8. Guide values for the relationship die clearance/sheet or plate thickness Sheet/plate thickness
Tensile strength of material
(mm)
(N/mm')
<250
250 to 400
400 to 600
>600
Independent of sheet/plate thickness
0.03
0.04
0.05
0.06
0.025
0.025
0.03
0.035
lto2
0.03
0.03
0.035
0.04
2 to 3
0.035
0.035
0.04
0.045
3 to 5
0.04
0.04
0.045
0.05
5 to 7
0.045
0.045
0.05
0.055
0.05
0.05
0.055
0.06
7
to 10
Appendix K4 Table 9. Conventional materials for cutting tools and field of application Tool material
Approx. service hardness HRC, HV
Sheet! plate thickness
Characteristics
(mm)
1. Cold work steel X 155CrVMol2 1 X 165CrMoVI2 X 21OCrWI2
62 up to 65 HRC
up to 4 mm
Materials with low toughness and high wear resistance for shearing hard, thin sheets
4 up to
Low-distortion materials of average toughness and wear resistance
X 210Crl2 X 210CrCoWI2
S 6-5-2
90MnV8 lOSWCr6 45WCrV7 60WCrV7 X 45NiCrM04 X SOCrMoW9 11 X 63CrMoV5 I
60 up to 64 HRC
6mm
56
over
up to
6mm
63 HRC
2. Cemented carbides GT 15' GT 20' GT 30' GT 40' lHR-F'
1450HV 1300HV 12
3. Hard material alloys Ferro-Titanit-C-Specialb Ferro-Titanit-WFNh S 6.5.3 (ASP 23Y CPM lOY"~ CPM Rex M 4<..1
68 68 61 61 61
to 71 HRC to 71 HRC to 6S HRC to 64 HRC to 65 HRC
Tough materials for absorbing high stress peaks in shearing of thick plates; lower wear resistance to abrasive wear mechanisms
I mm
Brittle materials for shearing thin plates; extremely high wear resistance to predominantly ab1J.sive wear mechanisms
up to 8 mm
Hard-wearing materials with high ductility due homogeneous structure.
up
to
to
a
'M":'
Manufacturing Processes. 9 References
References Kl Survey of Manufacturing Processes. [1] Tonshoff HK.
Randzonenbeeinflussung durcb Spanen und Abtragen. Ann CIRP 1974; 23: 187-8. - [2] Wiendahl H-P. Belastungsorientierte Fertigungssteuerung. Hanser, Munich, 1987. - [3] Tonshoff HK. Processing alternatives for cost reduction. Ann CIRP 1987; 36: 445-7. - [4] Kienzle O. Begriffe und Benennungen der Fertigungsverfahren. Werkstattstechnik 1%6; 56: 169-73. K2 Primary Shaping. [1] Hilgenfeldt W, Herfurth K. Tabellenbuch Gusswerkstoffe. VEB Deutscher Verlag fur
Grundstoffindustrie, Leipzig, 1983. - [2] ZGV. Giessen heute. Ed. Zentrale fur Gussverwendung, DUsseldorf, 1974. - [3] Guss Produkte '89. Hoppenstedt, Dannstadt. [4] Herfurth K. Einfiihrung in die Fertigungstechnik. Kapitel Urformen. VEB Verlag Technik, Berlin, 1975. - [5] Feinguss fur alle Industriebereiche. Ed. Zentrale fur Gussverwendung, Dusseldorf, 1984. - [6] Leitfaden rur Gusskonstruktionen. Ed. Zentrale fur Gussverwendung. Giesserei-Verlag, DUsseldorf, 1966. - [7] Verein Deutscher Giessereifachleute (VDG). Giesserei-Kalender 1977. Giesserei-Verlag, DUsseldorf, 1976. - [8] PahI G, Beitz W. Konstruktionslehre - Handbuch fur Studium und Praxis, 2nd edn. Springer, Berlin, 1986. - [9] Patterson W, Dopp R. Betriebsnomogramm fur Grauguss. Giesserei 1960; 47: 175-80. - [10] Colland A. Giesserei, techn.-wiss. Beib. 1954; 14: 709-26, and 1955; 15: 767-99. - [11] ZGV-Mitteilungen. DUsseldorf, 1976. - [12] Eisenkolb F. Einfiihrung in die Werkstoffkunde, vol. V: Pulvermetallurgie, 2nd edn. VEB Verlag Technik, Berlin, 1%7. - [13] Technikum fur berufliche Bildung des Ministeriums fur Erzbergbau, Metallurgie und Kali (Technical Centre for Vocational Training of the Ministry of Ore Mining, Metallurgy and Potash). Lehrbuch Metallurgie. VEB Deutscher Verlag fur Grundstoffindustrie, Leipzig, 1971. Standards and Codes. DIN 1680 Part 1: Rough castings;
general tolerances and machining allowances; general. Part 2: Rough castings; general tolerance system. DIN 1683 Part 1: Steel raw castings; general tolerances; machining allowances. DIN 1684 Part 1: Malleable iron raw castings; general tolerances, machining allowances. DIN 1685 Part 1: Raw castings made from nodular graphite cast iron; general tolerances, machining allowances. DIN 1686 Part 1: Rough castings of grey iron with flake graphite; general tolerances, machining allowances. DIN 1687 Part 1: Heavy metal alloy raw castings, sand castings; general tolerances, machining allowances. Part 3: Rough castings of heavy metal alloys, gravity die castings; general tolerances, machining allowances. Part 4: Heavy metal alloy raw castings, pressure die castings; general tolerances. DIN 1688 Part 1: Light metal alloy raw castings, sand castings; general tolerances, machining allowances. Part 3: Light metal alloy raw castings, gravity die castings; general tolerances, machining allowances. Part 4: Light metal alloy raw castings, pressure die castings; general tolerances. DIN 1690 Part 1: Technical delivery conditions for castings made from metallic materials; general conditions. Ferrous cast materials: DIN 1691: Grey iron with flake graphite. DIN 1693: Cast iron with nodular graphite. DIN 1694: Austenitic cast iron. DIN 1695: Abrasion-resisting alloy cast iron. DIN 1692: Malleable cast iron; concepts, properties, final inspection.
DIN 1681: Cast steels for general engineering purposes. DIN 17245: Ferritic steel castings creep-resistant at elevated temperatures. DIN 17445: Stainless steel castings. DIN 17465: Heat-resisting steel castings. SEW 410: Stainless steel castings (Iron and Steel Material Data Sheets). SEW 685: Steel castings tough at sub-zero temperatures. SEW 510: Heat-treatable steel castings with wall thicknesses up to 100 mm. SEW 515: Heat-treatable steel castings with wall thicknesses over 100 mm. SEW 595: Cast steel for crude oil and natural gas plants. SEW 471: Heat-resisting cast steel. SEW 390: Non-magnetisable cast steel. SEW 835: Cast steel for flame and induction hardening. Light Metal Casting Materials. DIN 1725 Sheet 2: Alu-
minium casting alloys. DIN 1729 Sheet 2: Magnesium casting alloys. Heavy metal casting materials. DIN 1705: Copper-tin and copper-tin-zinc casting alloys (cast tin bronze and gunmetal), castings. DIN 1709: Copper-zinc alloys castings (brass and special brass castings). DIN 1714: Copper-aluminium casting alloys (cast aluminium bronze), castings. DIN 1716: Copper-lead-tin casting alloys (cast tin-lead bronze), castings. DIN 17655: Unalloyed and low-alloy copper materials for casting; castings. DIN 17658: Copper-nickel casting alloys. DIN 1743: High purity zinc casting alloys (Part 1: Ingot metals; Part 2: Castings). DIN 1741: Lead alloys for pressure die castings. DIN 1742: Tin alloys for pressure die castings. DIN 17730: Nickel and nickel-copper casting alloys. K3 Metal Forming. General reference: Lange K. Hand-
book of Metals Forming. McGraw-Hill, New York, 1985. [1] Henky H. Z angew Math Mech 1924; 4: 323-34. - [2] BUhler H, Hopfner HG, LOwen]. Die Formiinderungsfestigkeit von Aluminium und einigen AluntiniumJegierungen. BBR 1970; 11: 645-9. - [3] Krause K. Formiinderungsfestigkeit der Werkstoffe beim Kaltumformen. In: Grund!agen der bildsamen Formgebung. VDEh, DUsseldorf, 99145. - [4] Kienzle 0, BUhler H. Das Plastometer, eine PrUfmaschine fur Staucheigenschaften von Metallen. Z MetalIkd 1%4; 55: 668-73. - [5] Lange K. Lehrbuch der Umformtechnik, vol. 1, 2nd edn, 1984; vol. 2, 1988; vol. 3, 1990; vol. 4, 1993. Springer, Berlin. - [6] MillIer G. Formiinderungsfestigkeit beim Umformen in der Warme. In: Grundlagen der bildsamen Formgebung. VDEh, DUsseldorf, pp. 146-61. - [7] SiebelE. Grenzen der Verformbarkeit. Mitt fur die Mitglieder der Forschungsgesellschaft. Blechverarbeitung 1952; 16: 177-84. - [8] Stenger H. Uber die Abhlingigkeit des Formiinderungsvermilgens metallischer Werkstoffe vom Spannungszustand. Thesis, RWTH Aachen, 1%5. - [9] Hasek V. Untersuchung und theoretische Beschreibung wichtiger EinflussgrOssen auf das Grenzformiinderungsdiagramm. Blech-Rohr-Prollie 1978; 25: 213-20, 285-92, 493-9, 620-7. - [10] Siegert K. Grenzen des Ziehens von Karosserieteilen. Werkst Betrieb 1985; 118: 709-13. - [11] Siebel E. Kriifte und Materialfluss bei der bildsamen Formiinderung. Stahl Eisen 1925; 45: 139-41. - [12] Siebel E. Die Formgebung im bildsamen Zustand. Stahleisen, DUsseldorf, 1932. - [13]
Manufacturing Processes • 9 References
Sachs G. Zur Theorie des Ziehvorgangs. Z angew Math 1927; 235-6. - [14] Siebel E., Pomp A. Zur Weiterenrwicklung des Druckversuches. Mitt K-Wilh-Inst Eisenf 1928; 10: 55-62. - [15] Lippmann H, Mahrenboltz O. Plastomechanik der Umformung metallischer Werkstoffe, vol. 1. Springer, Berlin, 1%7 - [16] Ismar H, Mahrenboltz O. Technische Plastomechanik Vieweg, Brunswick, 1979. [17] Lippmann H. Die elementare Plastizitatstheorie der Umformtechnik. Bander Bleche Rohre 1962; 374-83. [18] Spur G, Stoferle T. Handbuch der Fertigungstechnik, vol. 2. Hanser, Munich, 1983. - [19] Korper F, Eichinger A. Die Grundlagen der bildsamen Formgebung. Mitt KWilh-Inst Eisenf 1940; 22: 57-80. - [20] Pawelski O. Grundlagen des Ziehens und Einstossens. In: Grundlagen der bildsamen Formgebung. VDEh, Dusseldorf, pp. 384433. - [21] Sachs G. Zur Theorie des Ziehvorgangs. Z angew Math Mech 1927; 7: 235-36. - [22] Lippmann H. Theorie der Einstoss- und Strangpressvorgange. Bander Bleche Rohre 1%3; 223-5. - [23] Eisbein W. Kraftbedarf und Fliessvorgange beim Strangpressen. Thesis, TH Berlin, 1931. - [24] Sachs G. Spanlose Formgebung der Metalle. In: Handbuch der Metallphysik, vol. 3, Lief. 1, 1937. - [25] Rathjen C. Untersuchungen tiber die Grosse der Stempelkraft und des Innendruckes im Aufnehmer beim Strangpressen von Metallen. Thesis, RWTH Aachen, 1966. - [26] Panknin W. Die Grundlagen des Tiefziehens im Anschlag unter besonderer Beriicksichtigung der Tiefziehpriifung. Bander Bleche Rohre 1961; 133-43, 201-11, 264-71. [27] Siegert K. Ziehen von f1achen Karosserieteilen, Verfahren-Maschinen-Werkzeuge. VDJ-Z 1989; 131: no. 4. [28] Cyril Bath Company. Streckziehen von Karosserieteilen. Werkstatt und Betrieb 1%5; issue 3. - [29] Neuere Enrwicklungen in der Blechuruformung, ed. K. Siegert. DGM-Informationsgesellschaft mbH, Oberursel, 1990. [30] Siegert K. Zieheinrichtungen im Pressentisch einfach wirkender Pressen. In [29]. - [31] Ziinkler B. Biegeuruformen. In: Spur G, Stdferle T. Handbuch der Fertigungstechnik, vols 2 and 3. Hanser, Munich, 1985 - [32] Ludwik P. Technologische Studie tiber Blechbiegung. Techn Blatter 1903; pp. 133-59. - [33] Oehler G. Biegen. Hanser, Munich, 1963. - [34] Ziinkler B. Rechnerische Erfassung der Vorgange beim Biegen im V-Gesenk Ind Anz 1%6; 88: 1601-5. - [35] Kienzle O. Untersuchungen iiber das Biegen. Mitt DFBO 1952; 57-65. - [36] Fait]. Grundlagenuntersuchungen zur Ermittiung von Kenngrossen fur das CNC-Schwenkbiegen. Ind Anz 1987; 109: 45-6. - [37] Eichner A]. Superplastisches Fertigen komplexer Formstucke. Werkstatt und Betrieb 1981; 114: 715-18. - [38] Winkler po], Keinath W. Superplastische Umformung, ein werkstoffsparendes und kostengiinstiges Fertigungsverfahren fur die Luft- und Raumfahrt. Metall 1980; 34: 51925. - [39] Pischel H. Superplastisches Blechuruformen. Werkstatt und Betrieb 1989; 122: 165-9. - [40] Bunk W, Kellerer H. Neue Fertigungsverfahren zur Verbesserung der Wirtschaftlichkeit. Aluminium 1985; 61: 247-51. [41] Richards JH. Einsatz superplastisch umgeformter Blechbauteile im Bauwesen. Aluminium 1987; 63: 3607. - [42] Hojas M, Kiilein W, Siegert K, Werle T. Herstellung von superplastischen Aluminiumblechen und deren Verarbeitung mit numerisch gesteuerten Pressen. In [29]. - [43] Lange K, Meyer-Nolkemper H. Gesenkschmieden, 2nd edn. Springer, Berlin, 1977. - [44] Bruchanow AW, Rebelski AV. Gesenkschmieden und Warmpressen. Verlag Technik, Berlin, 1955. - [45] Rathjen C. Die historische Enrwicklung des Strangpressverfahrens. Ind Anz 89. 1967; 47: 17/2. - [46] Ziegler W, Siegert K. Spezielle Anwendungsmoglichkeiten der indirekten Strangpressmethode. Metall 1977; 31: 845-51. - [47] Ruppin D, Muller
I
:W'O
K. Kalt-Strangpressen von Aluminium-Werkstoffen mit
Druckftlmschmierung. Aluminium 1980; 56: 263-8, 32931, 403-6. - [48] Ziegler W, Siegert K. Indirektes Strangpressen von Leichtmetall. Metallkunde 1973; 64: 224-9. [49] Pugh H Li D. The mechanical behaviour of materials under pressure. Applied Science Publishers, London, 1971. - [50] Hommark N, Ermel D. Kupferurnhiilltes Aluminium, ein neuer Werkstoff fur die industrielle Fertigung von Kompoundleitem. Draht-Welt 1970; 56: 424-6. - [51] Fiorentino R], Richardson BD, Sabrow AM, Boulger FW. New Developments in Hydrostatic Extrusion. Proc Int Conf Manuf Techn 1967; 25/28: 941-54. - [52] Fiorentino R], Sabrow AM, Boulger FW. Advances in hydrostatic extrusion. Tool Mfng Engr 1973; 00: 77-83. - [53] Pugh H Li D, Donaldson GHH. Hydrostatic extrusion - a review. Ann CIRP 1972; 21/2: 000-0. - [54] Fiorentino R], Meyer GE, Byrer TG. Some practical considerations for hydrostatic extrusion. Metallurgia and Metal Forming 1974; 00: 210-13,296-9. - [55] Fiorentino R], Meyer GE, Byrer TG. Technical and economic potential of hydrostatic extrusion over conventional extrusion. Preliminary reports for the symposium "Neue Verfahren fur die Halbzeugherstellung". Deutsche Gesellschaft f Metallkunde, 1973. - [56] Fiorentino R], Meyer GE, Byrer TG. The thick-film hydrostatic extrusion process. Metallurgia and Metal Forming 1972; 00: 200-3.
K 4 Cutting. [1] Patzke M. Einfluss der Randzone auf die Zerspanbarkeit von Schmiedeteilen. Thesis, University of Hannover, 1987. - [2] Warnecke G. Spanbildung bei metallischen Werkstoffen. Fertigungstechnische Ber, vol. 2. Resch, Griifelf'mg, 1974. - [3] Bartsch S. Verschleissverhalten von Aluminiumoxidschneidstoffen unter stationarer Belastung. Thesis, University of Hannover, 1988. - [4] Tonshoff HK. Schneidstoffe fur die spanende Fertigung. wt-Z Ind Fert 1982; 72: 201-8. - [5] Knorr W. Bedeutung des Schwefels fur die Zerspanbarkeit der Stahle unter Beriicksichtigung ihrer Gebrauchseigenschaften. Stahl und Eisen 1977; 97: 414-23. - [6] Kienzle 0, Victor H. Die Bestimmung von Kriiften und Leistungen an spanenden Werkzeugmaschinen. VDJ-Z 1952; 94: 299-305. - [7] Taylor FW. On the art of cutting metals. Trans Am Soc Mech Engrs 1907; 28: 30-351. - [8] Gawehn H. Das Spanwinkelproblem des Spiralbohrers. Maschinenbau und Betrieb 1931; 00: 440-6. - [9] Spur G. Beitrag zur Schnittkraftmessung beim Bohren mit Spiralbohrem unter Beriicksichtigung der Radialkriifte. Thesis, TU Brunswick, 1%1. - [10] Hutte, Taschenbuch fur Betriebsingenieure, vol. 1: Fertigung, 5th edn. Ernst, Berlin, 1957. - [11] Tuffentsammer K. Kurzlochbohren mit unterschiedlichsten Werkzeugen moglich. Ind Anz 1980; 102: 100, 38-41. [12] Victor H, Muller M, Opferkuch R. Zerspantechnik, vols I-III. Springer, Berlin, 1985. - [13] Kamm H. Beitrag zur Optimierung des Messerkopffriisens. Thesis, University of Karlsruhe, 1977. - [14] Victor HR. Zerspankennwerte. Ind Anz 1976; 98: 1825-30. - [15] Muller M. Zerspankraft, Werkzeugbeanspruchung und Verschleiss beim Frasen mit Hartmetall. Thesis, University of Karslrnbe, 1982. - [16] Roese H. Untersuchung der dynamischen Stabilitat beim Frasen. Thesis, RWTH Aachen, 1967. - [17] Borys WE. Vergleichsuntersuchungen zum Einsatz hochharter polykristalliner Schneidstoffe beim Frasen. Thesis, University of Hannover, 1984. - [18] Chryssolouris G. Einsatz hochharter polykristalliner Schneidstoffe beim Drehen und Frasen. Thesis, University of Hannover, 1984. - [19] TOllner K. Frasen von hochharten Eisenstoffen. wt-Z Ind Fert 1982; 72: 493-6. - [20] Tonshoff HK, Bussmann W. Forrnfehler bei der Hartbearbeitung: Frasen gehiirteter Fiihrungsflkhen. Ind Anz 1988; 110: 29,
I
:w....
Manufacturing Processes. 9 References
35-6. - [U] Tonshoff HK, HernJindes-Camacho J. HFF-Ber 10. Hannover, March 1987, pp.I-20. - [22] Laufer H}. Einsatz von Prozessmodellen zur rechnerunterstutzten Auslegung von Riiumwerkzeugen. Thesis, University of Karlsruhe, 1988. - [23] Opferkuch R. Die Werkzeugbeanspruchung beim Riiumen. Thesis, University of Karlsruhe, 1981. - [24] Dworak U. Mechanical strengthening of alumina and zirkonia ceramics through the introduction of secondary phases. Sci Ceramics 1987; 9: 543. - [25] Choi H-Z. Beitrag zur Ursachenanalyse der Randzonenbeeinflussung beim Schleifen. Thesis, University of Hannover, 1986. [26] Kurrein M. Die Messung der Schleifkraft. Werkstattstechnik 1927; 20: 585-94. - [27] Snoeys R. The mean undeformed chip thickness as a basic parameter in grinding. Ann CIRP 1971; 20: 183-6. - [28] Tonshoff HK, Brinksmeier E, Choi HZ. Messung und Berechnung mechanischer und thennischer Werkstoffbeanspruchungen beim Schleifen. Jahrbuch Schleifen, Honen, Lappen und Polieren, 53rd edn. Vulkan, Essen, 1985, pp. 31-47. - [29] Brinksmeier E. Randzonenanalyse gesch1iffener Werkstucke. Thesis, University of Hannover, 1982. - [30] Salje E. Abrichtverfahren mit unbewegten und rotierenden Abrichtwerkzeugen. Jahrbuch Schleifen, Honen, Lappen und Polieren, 50th edn. Vulkan, Essen, 1981, pp.28498. - [31] Salje E. Abrichten wilhrend des Schleifens Grundlagen, Leistungssteigerungen, Wirtschaftlichkeit. Jahrbuch Schleifen, Honen, Lappen und Polieren, 53rd edn. Vulkan, Essen, 1985, pp. 1-30. - [32] Tonshoff HK, Dennis P. Hochleistungsbandschleifen - ein massgebendes Verfahren. Werkstattstechnik 1988; 78: 665-9. [33] Konig W, TonshoffHK, FromlowitzJ, Dennis P. Belt grinding. Ann CIRP 1986; 35: 487-94. - [34] Mushardt H. Modellbetrachtungen und Grundlagen zum Innenrundhonen. Thesis, TU Brunswick, 1986. - [35] Tonshoff T. Formgenauigkeit, Obert1iichenrauheit und Werkstoffabtrag beim Langhubhonen. Thesis, University of Karlsruhe, 1970. - [36] Salje E, Mohlen H, See vM. Vergleichende Betrachtungen zum Schleifen und Honen. VOI-Z 1987; 129, (1): 66-69. - [37] See vM. Prozessoptimierung beim Honen. Jahrbuch Sch1eifen, Honen, Lappen und Polieren, 55th edn. Vulkan, Essen, 1988, pp.401-14. - [38] Spur G, Simpfendorfer D. Neue Erkenntnisse und Entwicklungstendenzen beim Planlappen. Jahrbuch Schleifen, Honen, Lappen und Polieren, 55th edn. Vulkan, Essen, 1988, pp. 469-80. - [39] Nolke H-H. Spanende Bearbeitung von Siliziunmitrid-Werkstoffen durch Ultrascha11-Schwinglappen. Thesis, University of Hannover, 1980. - [40] Tonshoff HK, Brinksmeier E, Schmieden VM. Grundlagen und Theorie des lnnenlochtrennens. Jahrbuch Schleifen, Honen, Lappen und Polieren, 55th edn. Vulkan, Essen, 1988, pp. 481-93. - [41] Brinksmeier E, Schmieden vW. Werkzeugaufspannung und Prozessverlauf beim 10Trennschleifen. Ind Diamanten Rundsch 1988; 00: 21419. - [42] Weckerle D. Prozessstorungen bei der funkenerosiven Metallbearbeitung. Tech. Mitt. F Deckel AG, Munich, 1985. - [43] Schmohl H-P. Ennittlung funkenerosiver Bearbeitungseigenspannungen in Werkzeugstiihlen. Thesis, TU Hannover, 1973. - [44] WijersJLe. Numerically controlled diesinking. EDM-Digest 1984; 00: 9/10. - [45] Schumacher B, Weckerle D. Funkenerosion - Richtig verstehen und anwenden. Technischer l'achverlag, Velbert, 1988. - [46] TonshoffHK, Seemrau H. Laser beam machining in new fields of application. Proceedings ASME Symposium, Chicago, December 1988. - [47] Tonshoff HK, Butje R. Excimer laser in material processing. Ann CIRP 1988; 37: 000-0. - [48] Dickmann K, Enunelmann C, Hohensee V, Schrnatjko KJ. Excimer-Hochleistungslaser in der Materialbearbeitung. Laser Magazin, pt 1, 1987; issue
3: 26-9, and pt 2, 1987; issue 4: 34-44. - [49] Bimberg D. Laser in Industrie und Technik, vol. 13. Expert, Grafenau, 1985. - [50] Semrau H. Erzeugen von oxidfreien Schnittfliichen durch Laserstrahlschneiden. Thesis, University of Hannover, 1989. - [51] Beske EU, Meyer e. Schweissen mit kW-Festkorperiasem. Laser Magazin 1989; (3): 42-6. - [52] Beske EV. Handhabung einer lichtleitfaser zum Fiihren eines Nd-Yag-Laserstrahls. Laser und Optoelektronik 1989; 21(3): 60-1. - [53] Konig W, Schmitz-Justen C, Trasser Fr:J, Willerscheid H. Provisional list of tenninology for laser beam cutting. Ann CIRP 1988; 37: 675-80. - [54] Spur G. Handbuch der Fertigungstechnik, vol. 2/3: StOferle, T. Hanser, Munich, 1985. - [55] Lange K. Umformtechnik: Handbuch fur lndustrie und Wissenschaft, vol. 3. Springer, Berlin, 1990. - [56] VOl Code 3368. Die clearance, punch and die plate size for blanking punches. VOl-Verlag, DUsseldorf, 1965. - [57] Tschatsch H. Taschenbuch Umformtechnik: Verfahren, Maschinen, Werkzeuge. Hanser, Munich, 1977. - [58] Guidi A. Nachschneiden und Feinschneiden. Hanser, Munich, 1965. - [59] VOl Code 3345. Fine blanking (with further literature). VOl-Verlag, DUsseldorf, 1980.
Standards and Codes. DIN 2310: Themtal cutting. DIN 4990: Groups of application of carbides for machining by chip removal. DIN 6580: Movements and geometry of the chip removing process. DIN 6581: Reference systems and angles on the cutting part of the tool. DIN 8200: Blasting techniques. DIN 8580: Manufacturing methods. DIN 8589: Manufacturing processes cutting. DIN 8590: Manufacturing processes removal operations. DIN 69 100: Bonded abrasive products. DIN 51 384: Cooling lubricants.
ISO (International Organisation for Standardisation). ISO 513: Application of carbides for machining by chip removal. ISO 3002: Basic quantities in cutting and grinding. Part 1: Geometry of the active part of cutting tools. Part 3: Geometric and kinematic quantities cutting. ISO 3685: Tool life testing with single point turning tools.
VDI Codes. VOl Code 3332: Chip breakers at carbidetipped lathe tools. VOl Code 3335: Groups of application in the field of chip removal and operating angles for turning on a lathe with carbide metal tools. VOl Code 3400: Electrical discharge machining - concepts, methods, application. VOl Code 3401, Sheet 2: Electrochemical machining bath electrolytic machining. Stahl-Eisen-Priifblatt 1160: Chip removing tests, general basic concepts. K5.1 Thread Production. [I] Spur G. MehrspindelDrehautomaten. Hanser, Munich, 1970. - [2] Stock-Taschenbuch. R. Stock AG, Berlin, 1979. - [3] Stender W. ScMlen von Gewindespindeln. Brochure from Messrs Waldrich, Coburg. - [4] Druminski R. Analytische und experimentelle Untersuchungen des Gewindeschleifprozesses beim Uings- und Einstechschleifen. Thesis, TU Berlin, 1977. - [5]lickteig E. Schraubenherstellung. Verlag Stahleisen, DUsseldorf, 1966. - [6] Siebert H. Werkstattblatt 501: Gewindefurchen. Hanser, Munich, 1970.
K5.3 Manufacturing in Prectston Engineering and Mlcrotecbn%gy. [10] Degner W, Bottger He. Handbuch Feinbearbeitung. Hanser, Munich, 1979. - [11] Griinwald.
Manufacturing Processes • 9 References
Fertigungsverfahren in der Geriitetechnik. VEB Verlag Technik, Berlin, 1980. - [12] Schweizer W, Kiesewetter L. Moderne Fertigungsverfahren der Feinwerktechnik, ein Uberblick. Springer, Berlin, 1981. - [13] Lotter B. Wirtschaftliche Montage. Handbuch fur Elektrogeriitebau und Feinwerktechnik. VDI-Verlag, Dusseldorf, 1986. - [14] Heuberger A. Mikromechanik. Mikrofertigung mit Methoden der Halbleitertechnologie. Springer, Berlin, 1989. - [IS] Weber H, Herziger G. Laser. Grundlagen und Anwendungen. Physik-Verlag, Weinheim, 1972. - [16] Bimberg D. Laser in Industrie und Technik, 2nd edn. Expert, Sindelfingen, 1985. - [17] Kiesewetter L. Laser ein Werkzeug der Feinwerkteclntik. Wissenschaftsmagazin TU-Berlin, issue 9, 1986, 33-7. - [18] Herziger G. Werkstoffbearbeitung mit Laserstrahlung. Feinwerktechnik + Messtechnik 1983; 91: 156-63. - [19] Seiler P. Festkorper-Impulslaser ZlUn Fugen und Abtragen im Mikrobereich. Company publication of Carl Haas, Schramberg. [20] Moller W. Laser-Mikroloten mit Temperatur- und Zeitsteuerung. Optoelektronik Mag 1988; 4: 684-9. - [21] Benninghoff H. Werkstoffbearbeitung mit dem Laser. Tech Rundsch 1989; 6: 26- 31. - [22] Brochure issued by Kammerer GmbH, Pforzheim-Huckenfeld, 1989. - [23] Schiller S, Heisig U, Panzer S. Elektronenstrahltechnologie. Stuttgart, Wissenschaftliche Verlagsanstalt, 1977. - [24] Dobeneck YD. Die Elektronenstrahltechnik - ein vielseitiges Fertigungsverfahren. Feinwerktechnik + micronic 1973; 77: 98-106. - [251 Behnisch H. Einsatz des Elektronen- und Laserstrahls in der Schweiss- und Schneidtechnik. Technica (CH) 1976; 25: 1341-7. - [26] Schulz H. Schweissen von Sondertnetallen. Deutscher Verlag fur Schweisstechnik, Dusseldorf, 1971. - [27] Schiller S, Panzer S. Thertnische Oberfliichenmodifikation metallischer Bauteile mit Elektronenstrahlen. Metall 1985; 39: 22732. - [28] Schiller S, Heiseg U, Frach P. Elektronenstrahlbedampfen. In: Sudarshan TS, Surfacing technologies handbook. Marcel Dekker, New York, 1987. - [29] Lehfeldt W. Ultraschall. Vogel, Wurzburg, 1973. - [30] Matauschek J. Einfuhrung in die Ultraschalltechnik. VEB Verlag Technik, Berlin, 1962. - [31] Millner R. Ultraschalltechnik. Grundlagen und Anwendungen. Physik-Verlag, Weinheim, 1987. - [32] Dorn L. Schweissen in der Elektround Feinwerktechnik. Expert, Grafenau/Wiirtt., 1984. [33] Abel F. Ultraschall in der Kunststoff-Fugetechnik. Herfurth GmbH, Hamburg, 1979. - [34] Berger A. Elektrisch abtragende Fertigungsverfahren. VDI-Verlag, Dusseldorf, 1977. - [35] Degner W, Bottger He. Handbuch Feinbearbeitung. Hanser, Munich, 1979. - [36] Grunwald F. Fertigungsverfahren in der Geriitetechnik. YEB Verlag Technik, Berlin, 1980. - [37] Degner W. Elektrochemische Metallbearbeitung. VEB Verlag Technik, Berlin, 1984. - [38] Janicke ]. Anwendungstechniken der Funkenerosion. Tech Rundsch 1975; 31: 10-11. - [39] Schadach P. Elektroerosive und elektrochemische Metallbearbeitungsverfahren. VDI-Z 1975; 117: PT32-PT37. - [40] Haefer RA. Oberlliichen- und Diinnschicht-Technologie. Pt 1: Beschichtungen von Oberfliichen. Springer, Berlin, 1987. - [41] Simon H, Thoma M. Angewandte Oberfliichentechnik fur metallische Werkstoffe. Hanser, Munich, 1985. - [42] Czichos H. Konstruktionselement Oberlliiche. Konstruktion 1985; 37: 219-27. - [43] Paatsch W. Technologische Eigenschaften galvanisch abgeschiedener Schichten. Galvanotechnik 1985; 75: 1234-41. - [44] Frey H. Diinnschichttechnologie. VOl-Verlag, Dusseldorf, 1987. - [45] Ikeno H. E1ectrooptic histability of a ferroelectric liquid crystal device prepared using polyimide Langmuir-Blodgett orientation ftIms. Jap J Appl Phys 1988; 27: L475. - [46] Kiesewetter L, GIeske G. Baufortn und
l;Wt...1
Fertigungsverfahren fur Flussigkristall-Anzeigen. BerlinTronics 10. Verlag fur technische Publikationen, Berlin, 1988,4-8. - [47] Hanke H:J, Fabian H. Technologie elektronischer Baugruppen. VEB Verlag Technik, Berlin, 1977. - [48] Joachim F-W. Kupferplattiertes Invar als Metallkern in Leiterplatten mit einsteUbarem Wiirtneausdehnungskoeffizienten. Feinwerktechnik und Messtechnik 1986; 94: 507-9. - [49] Huber B. Leiterplatten- und Hybridtechnologien im Vergleich. Feinwerktechnik und Messtechnik 1986; 94: 215-20. - [SO] Duppen vJ. Handbuch fur den Siebdruck. Verlag der Siebdruck, Lubeck, 1981. - [51] Scheer HG. Siebdruck und Elektronik-Druckforrnherstellung in der Elektronik. IS + L 1983/4 (August). - [52] SteinbergJJ, Horowitz SJ, Bacher RJ. Herstellen von Mehrlagenschaltungen mit niedrig sinterden griinen Keramikfolien. EPP Hybridtechnik, October 1986, pp.43-7. - [53] Lehmann HW, Gale T. Submikrongitter. Tech Rundsch 1989; 00: 46-53. - [54] 0.4 fLm-Strukturen mit norrnaler Optik. Elektronik 1984; 17: 22. - [55] Jagt JC, Whipps PW. Elektronenempfmdliche Negativlacke fur VLSI. Philips Tech Rundsch 1981; 39: 368-75. - [56] Staudte H]. Proceedings 27th Annual Symposium on Frequency Control, 1973, 50-4. - [57] Zwingg W. Miniaturquerschwinger und -Quarzsensoren. Jahrbuch der Deutschen Gesellschaft fur Chronometrie e.V., vol. 36. Stuttgart, 1985. - [58] Johansson S. Micromechanical properties of silicon. Acta Universitatis Upsaliensis, Faculty of Science, Uppsala, 1988. - [59] Petersen KE. Silicon as a mechanical material. Proc IEEE 70: 1982; 420-57. [60] Hohm D. Mikromechanik erOffnet neue Wege zu elektroakustischen Wandlern. Spektrum der Wissenschaft 1988; 00: 38-50. - [61] Herzog H-J, Csepregi L. X-ray investigation of boron- and germanium-doped silicon epitaxial layers. I. Electrochem Soc 1984; 131: 000-0. - [62] Becker EW, Ehrfeld W. Das LlGA-Verfahren. Phys BI 1988; 44: 166-70.
K5.4 Surface Coating. [63J Pulker HK. Verschleissschutzschichten unter Anwendung der CVD/PVD-Verfahren. Expert, Sindelfingen, 1985. - [64] Gunther Ke. Advanced coating by vapour phase processes. Ann CIRP 1989; 38: 645-55.
K6Assembly. [11 Spur G, St6ferle T (ed.). Handbuch der Fertigungstechnik, vol. 5: Fugen, Handhaben, Montieren. Hanser, Munich, 1986. - [2] Wranecke H-J, Schraft RD (ed.). Handbuch - Handhabungs-, Montage- und Industrierobotertechnik, vol. 3: Montagetechnik. Verlag Modeme Industrie, Munich, 1984. - [3] Lotter B. Wirtschaftliche Montage. Ein Handbuch fur Elektrogeriitebau und Feinwerktechnik. VDI-Verlag, Dusseldorf, 1986. - [4] DIN 8593: Manufacturing production processes joining. Classification, subdivision, concepts. Beuth Verlag, Berlin, 1985. - [5] VDI Code 2860, Sheet I, Draft: Assembly and handling units. Handling functions, handling units, tenninology, definitions and symbols. VDI-Verlag, Dusseldorf, 1982. - [6] Seliger G (ed.). Montagetechnik. gfmt, Munich, 1989. - [7] Andreasen, Kiihler, Lund. Montagegerechtes Konstruieren. Springer, Berlin, 1985. - [8] Deutschliinder A. Integrierte rechnerunterstiitzte Montageplanung. Series: Produktionstechnik Berlin, vol. 72. Hanser, Munich, 1989. - [9] Severin F. Flexibel automatisierte Montageeinrichtungen - Innovationspotential der achtziger Jahre (pt I). ZwF 1982; 77: 529-40.
K7 Production and Works Management. [I] Olbrich W. Arbeitsplanerstellung unter Einsatz elektronischer Datenverarbeitungsanlagen. Thesis, RWTH Aachen, 1970. - [2] Warnecke HJ, Hirschbach 0, Metzger H. Rechnerunterstutzte MontagearbeitsplanersteUung. wt-Z ind Fertig
I
."4
Manufacturing Processes. 9 References
1975; 65: 147-52. - [3] Warnecke m, GrafH, Kunerth W. Stand und Entwicklungstendenzen technisch organisatorischer lnformationssysteme. In: Hansen HR. lnformationssysteme im Produktionsbereich. Oldenbourg, Munich, 1975. - [4] Hahn R, Kunerth W, Roschmarui. K. Fertigungssteuerung mit elektronischer Datenverarbeitung. Beuth, Berlin, 1973. - [5] Graf H. MethodenauswahI fiir die Materialbewirtschaftung in Maschinenbau-Betrieben. Thesis, University of Stuttgart, 1977. - [6] Rabus G, Nakonzer K. Analyse der Fertigungssteuerungsaufgaben im Hinblick auf den EDV-Einsatz. Unpublished research repon of the lnstitut fiir Produktionstechnik und Automatisierung (lPA) , 1976. - [7] Scharf P. Strukturen flexibler Fertigungssysteme. Gestaltung und Bewenung. Krausskopf, Mainz, 1976. - [8] Frank E. Handhabungseinrichtungen. Krausskopf, Mainz, 1975. - [9] Droge KH. Telemarupulatoren - Stand der Technik. Paper for the 5th workshop of the Institut fiir Produktionstechnik und Automatisierung (IPA): "Erfahrungsaustausch Industrieroboter". Stuttgart, 1975. - [10] Warnecke m, Schraft R-D. Einlegegeriite zur automatischen Werkstilckhandhabung. Krausskopf, Mainz, 1973. - [11] Warnecke lV, Schraft RD. Industrieroboter. Krausskopf, Mainz, 1989. - [I 2] Schweizer M. Robotenechnik. Bibliothek der Technik, vol. I. Verlag moderne industrie, 1987. - [13] Gerlach B. Spanende Sonderwerkzeugmaschinen. Techn. Verlag Grossmann, Stuttgart, 1977. - [I4] Warnecke m, Gericke E, Vettin G. Auslegung der Verkettungseinrichtungen flexibler Fertigungssysteme mit Hilfe der Simulation. Proceedings of the CIRP Seminars on Manufacturing Systems 1976; 5: 155-64. - [I5] Masing W. Handbuch der Qualitiitssicherung, 2nd edn. Hanser, Munich/Vienna, 1988. - [16) VOA: Qualitiitskontrolle in der Automobilindustrie - Sicherung der Qualitiit vor Serieneinsatz. Verband der Automobilindustrie e.V., Frankfurt, 1977. - [I7) Ford Motor Company publication: Statistische Prozessregelung. l.eitfaden
Eu880b, April 1986. - [I8] Warnecke HJ, Melchior KW, Ahlers R:J, KringJ. Handhuch Qualitiitstechnik: Methoden und Geriite zur effizienten Qualitiitssicherung. moderne industrie,l.andsberg/Lech, 1987. - [19] Warnecke lV, Bullinger H-J, Hichen R. Kostenrechnung fiir Ingenieure. Hanser, Munich, 1979. - [20] Mellerowicz K. Kosten und Kostenrechnung, vol. I, 5th edn. De Gruyter, Berlin, 1973. - [21] Bussmann KF.Industrielles Rechnungswesen. Poeschel, Stuttgart, 1963. - [22) Warnecke HJ, Bullinger H:T, Hichen R. Wirschaftlichkeitsrechnung fiir Ingenieure. Hanser, Munich, 1980. - [23) lnstitut fiir angewandte Arbeitswissenschaft e.V. (ed.). Arbeitsgestaltung in Produktion und Verwaltung: Taschenbuch fiir den Praktiker. Bachem, Cologne, 1989. - [24) Bullinger H:T, Solf JJ. Ergonomische Arbeitsmittelgestaltung I: Systematik/Forschungsbericht no. 196, Bundesanstalt fiir Arbeltsschutz, Dottmund. Bremerhaven, Wirtschaftsveriag NW, 1979. - [25) Neudtirfer A. Anzeiger und Bedienteile: Gesetzmilssigkeiten und systematische Ltisungssammlungen. VOl Verlag, DOsseldorf, 1981. - [26) Lange W. Kleine Ergonomische Datensammlung, 4th edn, ed. Bundesanstalt fiir Arbeitsschutz TOv Rheinland, Cologne, 1985. - [27) Schmidtke H (ed.). l.ehrbuch der Ergonomie, 2nd edn. Hanser, Munich, 1981. and Codes. DIN 5034: Daylighting of interiors (principles). DIN 5035: Artificial lighting of interiors. DIN 5036: Radiometric and photometric properties of materials. DIN 33402: Body dimensions of adults. DIN 40080: Sampling procedures and tables for inspection by attributes. DIN 45635: Measurement of airborne noise emitted by machines. DIN 69 513-69643: Machine tools (various subheadings). DIN IEC 651: Second level meters. Standards
Manufacturing Systems B. Behr, Aachen; E. Dannenmann, Stuttgart; I. Dorn, Berlin; G. Pritshow, Stuttgart; K. Siegert, Stuttgart; G. Spur, Berlin; M. Weck, Aachen; T. Werle, Stuttgart.
Machine Tool Components M. Week and B. Behr, Aachen
1.1 Fundamentals 1.1.1 Function Structure System Structure Manufacturing systems are classified according to DIN 8590. Subsystems are machine tools which are defined according to DIN 69651 as "mechanised and partially or fully automated manufacturing systems which can generate a prescribed fortn or change in the workpiece as a result of the relative motion hetween the tool and workpiece". Individual and multiple machine tool systems consist of one or more basic machine tool systems and other operational and auxiliary systems. The assemblies (i.e. drives, frame components, tool carriers and workpiece carriers) required for the execution of the basic operation fortn the basic macbine tool system. The designs of the various tool and workpiece carriers range from rigid tables to a number of combinations of linear and rotary guides and/or bearings depending on the machine tool stmcture. Tools and workpieces are held or clamped onto the respective carriers. The enlbodiment design of the mechanical interfaces between the equipment components and the machine is determined by the interchangeabiliry and flexibiliry of the machine tools in a variety of different machining tasks. The complete machine tool system consists of various components of machine tool and machine tool flow systems, depending on the degree of automation, and there are a number of similarities between the components for the execution of the handling, transport and storage operations and the components of the basic machine tool system. Handling systems are linked to the basic system at the relevant clamping points (Fig. 1).
Action Pair, Effective Motion A workpiece with a specific basic form is transfortned into a given fortn as a result of the relative motion between tool, workpiece and process-related power transmission, i.e. separating and re-forming. The technical qualiry of a workpiece is determined by the dimensional accuracy and surface fmish. Technological advances in machine tool components are constantly producing improvements in machining accuracy (Fig. 2). Effective motion is composed of cutting motion, infeed motion and feed motion. These take the form of linear, rotary, continuous, or intermittent motion. This prod-
uces a three-dimensional working space in proportion to the size of the feed or infeed axis and the work length, where applicable, i.e. in planing, shaping and slotting, and metal-forming machines. This space is cylindrical in turning machines and circular grinding machines; in milling, drilling, shaping and slotting machines it is nortnally cuboid. The working space in metal-forming machines is determined by the maximum travel and maximum tool surface area at right angles to the direction of travel. Rotating motion normally occurs in the fortn of cutting motion in cutting machine tools, e.g. turning, drilling and milling. The required speed range is limited by the maximum and minimum cutting speed required and the largest and smallest workpiece and/or tool diameter. For each machining task an optimum rotational speed may be specified at which the most economic cutting speed is achieved. Higher cutting speeds are being achieved all the time with the improved efficiency of cutting materials. These high speeds make great demands on the constmctions of spindle-bearing systems (Fig. 3). For example, a rotational speed of n = 12 500 min-' is required for a cutting speed of 2000 m/min at a milling rate of d = 50 mm, which represents a critical load for a roller bearing of diameter 100 mm or over. To allocate effective motion purely in tertns of the fortn of the workpiece would be misleading, however. The execution of the required motion with the workpiece carrier and tool carrier may be effected in a wide variety of ways with kinematic reversal, where the components of effective motion may he interchanged. This results in various types of machines which make an extremely wide range of demands on the components of linear and rotary motion, namely the guides. Rational arrangements may be derived from the individual machining task, as may the specific requirements for automatic workpiece and tool change facilities. These types range from machines where motion is confmed to the tool carrier, through the various intertnediate stages with their respective combinations, to those where motion is effected by the workpiece carriers. Motion is nortnally actuated by separate main and feed drives and, less commonly, the feedrate is derived from the main spindle. Gears alter rotational speeds and speedtorque characteristics. Transmission components, such as threaded spindles and toothed belts, transmit the motion to the tool and/or workpiece carrier, nortnally consisting of slides with linear motion. The forces generated by the production process at the active point, including friction force and force due to weight, are taken up by guides and bearings and conducted to assemblies such as slides, spindle drive boxes and tailstocks. The flow of force is completed through the
Manufacturing Systems. 1 Machine Tool Components
Tool flow system Handling Transport
"
~~
SlOrage
..
"'\.....r
A( V
lIleaSIJflng ana testing system
_"0-
f
)
-
Workpiece llow system ~
Handling Transport Storage
be
[ A r\ Interlaces in flow V V of force of rnacIWle
and clamp>ng system ) Fj~turing
1
Imerlaces fOI supply and discharge
Figure 1. System setup for machine tools and equipment accessories.
Coins
10
_-l ____ tnventions and t>s1onca' eve'1S
.'"'" ~
~~
,.,
~ ~-
<»1"" 5 e
"5,
~ 'i£ ~
~~!~I~~~~ ... u-,!:... & os~
.•
co
~
E
i~ .~ ~
... ~
a>
10-' 1150
~ ~ .~ ~ ~~ E-I~i i:l!""":8i! .~
e?
~ ~
"',.
1800
b
0.0
'" '" '" ~ ~
0
:l!c758<.:>~~oS' =!ri g .,; ~:Il~!1! ~ '
~
M
1850
:g e?
~ Year
~!~ 1900
!
~
-
§
...
~
M
§!
N ..... a>
M
!B ~ 1950
Figure 2. Chart showing historical progression of machining accuracy achieved in machine tools.
frame components, such as the columns and bed, which
1.1.2 Mechanical Characteristics
also provide the connection to the foundations, Static, dynamic and thennal loads produce elastic defonnation in individual components which may result in surface imperfections on the workpiece or may affect economic
The static, dynamic and thennoeIastic characteristics of a machine tool, regarding either the whole assembly or an individual part, may have a substantial effect on the process capability and quality of manufacture which is achievable with that tool.
efficiency.
1.1 Fundamentals. 1.1.2 Mechanical Characteristics
il
4.10 3
c
i
Fa
10 3 8
5
~f ------
--
~
~
-g 8 .~
~
6.
I
~--~
:
'""""'------'-.--~
Xo
Deformation x
10 2 8
5
k=
(i)\ :!i x fo xa
.'!l
UJ
I
"",,,_-,,;--_.~I Xo Deformation x
k*=(Qi) =~, dx f xa-x k*= ton lX D b
a
10 0 1850
Figure 4. Definition of rigidity rate, a with secants, b with tan-
1870
1930 1950 1970 1990 Year
10'---
gents.
rigidity rates k, of the components involved, calculated from the sum of their respective compliance rates 1/k"", as reciprocal values of their rigidity rates, thus: 1/k"" = ~ Ilk,. The total machine is, thus, always "less stiff" than its most compliant part within the flow of force. The rigidity normally occurring at the cutting point in cutting machine tools is between 20 and 500 N/f.lm and between 10' and 10' N/f.lm in metal-forming machines measured between the ram and machine tool table.
8
5
Criteria for Dynamic: Load
b
Action
10'
n (min-I)
Figure~. Trends a in cutting speed~ and b in rotational speeds in machine tools in steel machining
Criteria for Static: Load The static characteristics of a machine tool are represented by elastic deformation, which occurs under a continuously rated load over time, i.e. process forces and forces due to weight. This consequently produces the static rigidity k as the most important factor. It is a measure for the resistance to changes in form and is shown as the ratio of load F to displacement x of the part in the direction of the application of force: k = dF/dx. The dependence of the det()ftnation x on the t()rce brought under load F is represented in the form of characteristic curves (Fig. 4; see A4.1 and F2.!). The theoretical relation is linear: k = Fix (spring rate) In practice, however, a progressive relation occurs as a result of a number of contact surfaces between the parts. There are two definitions for rigidity at a tool centre point. The secant is taken from the origin to the point under observation 1'0, Xo in the first diagram (Fig.4a) and in the second diagram (Fig. 4b) the slope of the tangent to the characteristic curve at the point under observation 1'0, x" is introduced. Depending on the type of load, this is known as the tensile, compressive, bending or torsional rigidity; the last (k,) is shown as the ratio of torque M to the angle of rotation '1', k, = dMldip. The resulting rigidity k,o,"1 at the point of application of force is always obtained by carrying over the individual
The dynamic characteristics of a machine tool are determined primarily by the static rigidity, spatial distribution and size of the component masses, as well as by system damping. Specific spatial natural oscillation forms are produced for the structure of each machine andlor part at certain individual frequencies, depending on these characteristics. It is essential for the description of the dynamic characteristics of complex machine tool structures of these types that the natural oscillation forms are known. This enables the individual components which predominantly cause the natural oscillations to be determined (weak point analysis). Figure 5 shows the natural oscillation fornl of a bed-type milling machine for an indi-
vidual supply frequency of 105 Hz. A bending oscillation can be seen in the vertical section of the column, and
Figure S. Natural oscillation forms of a bcd-type milling machine.
Manufacturing Systems. I Machine Tool Components
slight torsion occurring in the horizontal section of the column. In order to visualise the dynamic characteristics, one should imagine the machine tool divided up into mass elements elastically coupled to one another. Equilibrium conditions for exciter forces F(t), displacement-related spring energy, rate-controlling damping forces and proportional gain Inertial forces may be described by a system of differential equations. Dynamic characteristics of existing machines and frames may be obtained experimentally by exciting with different frequencies f [60]. The compliance frequency response Ilkdyn = xdynlFdyn is obtained from the ratio of the dynamic displacement x dyn and exciter force Fd,n at the point of application of force and the phase displacement cp between the force sensor and displacement transducer. It may be shown separately as amplitude response and phase response or as a vector diagram (circle diagram). Figure 6 shows measured frequency response curves and circle diagrams for an assembly with two resonant frequencies. Static compliance may be determined at f = O. Resonant dynamic compliance is approximately 2 to 5 times higher than static compliance depending on system damping. Natural frequencies of a minimum of a factor of approximately 1.2 to 1.4 should be located outside the exciter frequency range generated by cutting forces or feed drives for example, to prevent resonant oscillation via external exciters. There is a danger of regenerative chatter [61] and destruction of tools and workpieces in dynamically weak machines. High individual frequencies are achieved by setting high static rigidity rates and minimising mass. These
1
'1
1
11k
1\
z 3
,
/
\
---)-----
" ""-
a
are to be distributed in such a way that high-mass structures such as gears and motors are located at rigid points, i.e. the bed or the column underbody, where possible. Damping is to be as high as possible. The design of the joints and guides is critical, with particular attention to aspects such as oil fllnl. Damping may also be affected by the selection of materials, e.g. polymer concrete has a higher material damping effect than cast iron and this, in tum, has a higher value than steel. Both sand intill, where the casting core is not removed, and concrete infill are useful for increasing damping. In welding constructions the joints within the welded joint act as suppressors.
Criteria for Thennal Loads The thermal characteristics of machine tools may be described in terms of the thermoelastic relative displacement at the active point between workpiece and tool as a result of the effects of heat sources. These displacements are determined by all parts in the thermal chain of effect and their themtal deformation properties. Variable temperature distributions over time (isotherm lines) and, consequently, time-dependent deformations occur in the parts as a result of internal heat sources present inside machine tools, such as bearings, motors, gears, process heat, etc., and external heat sources acting on machine tools, e.g. temperature of surrounding components, sun's rays, day-night temperature fluctuations, etc. Figure 7 shows the potential heat sources and attributive effects of these thermal influences, using a knee-type milling machine as an example. The typical temperature distributions occurring in the parts due to the heat sources are determined by the specific thermal material properties, such as heat capacity and heat conductivity, and the conditions for heat exchange from the surroundings or adjacent parts. The connections of the individual parts in relation to the machining position, the relative position of the individual parts and the
interactive effect of the deformation of the parts may be cumulative, or may also have a balancing effect; nevertheless, they all have an effect on deformation resulting from the temperature distribution at the cutting point in conjunction with the heat expansion coefficient. Interactive compensation of thermal-related displacement in relation to the machining point may be exploited intentionally by way of a targeted design in terms of the heat sources (thermosymmetrical construction).
1.2 Drives f in Hz
lmt
Re
c: Figure 6. Compliance frequency response of a revolving turning macWne with two resonant frequencies, measured on excitation of the plunger by Fdyn. a Amplitude response. b Phase response. c: Circle diagram. d Oscillation form.
Drives are required mainly in machine tools for cutting and feed motion [1-9]. These are normally employed as separate drives for each individual motion, particularly with numerically controlled machines and less frequently as a common drive with power takeoff gears. Stepless drives are becoming more popular because of their greater compatibility with machining specifications. A distinction is made between electrical, hydraulic and pneumatic drives, depending on the type of drive circuitry and power supply (DIN 24 300) and hybrids such as electro hydraulic drives. The term drive includes assemblies such as motors, power transfomters, gears and transmission components.
1.2.1 Motors Electrical Three·Phase SUpring Motors These are traditionally employed in machine tools as asynchronous motors in conjunction with gear cone trans-
1.2 Drives. 1.2.1 Motors
frequency f, according to n ~ 2f!p. Pole-changing motors may be designed for all speeds with constant torque or with constant power. Modem asynchronous motor applications cater for speed-controlled operations. Such drives are known as servodrives. Speed control is determined by instantaneous position and magnetic field strength. The magnetic field and torque are generated to the desired level independently of one another by the controlled input of stator currents. The combined effect of the two factors produced on the stator results in a constant rotor speed. The regulated asynchronous motor is based on so-called fieldvector control (Fig. 8): see also 1.2 [10-121.
a
Example.
Figure 8 shows the relation between the field and
stator winding coordinates. The field-vector control takes the established relationships as the basis, according to which the torque is controlled via the torque-producing current components and the speed is controlled by the flow-producing current components. The effect of the temperature on the rotor time constant and the magnetic saturation on the motor parameters represent the limits of the concept. Control of these influencing factors may further improve the quality of the regulated asynchronous motor (13].
b
/ Figure 7. Examples of thermal-related deformation in a milling machine. a Principal heat sources: 1 bearings, 2 gear and hydraulic oil, 3 gears, clutches, 4 pumps, motors, 5 guides, 6 cutting point and swarf, 7 external heat transfer, b Deformation due to internal heat sources. c Defonnation due to external heat transfer.
Figure 9 shows an asynchronous motor of the squirre\cage rotor type designed as a servomotor. To compensate for the relatively complex control procedure associated with the servo amplifier, there are advantages, such as the fact that it requires no maintenance and has a broad speed range under field control. The latter characteristic enables the speed to be adjusted over a broad range with a constant power output, see Fig. 10. This is why the regulated asynchronous motor is becoming increasingly popular as a main spindle motor. Up to 80 kW of power and speeds of up to 8000 min- 1 are adequate for main spindle drives. The speed may reach 14000 min- 1 for feed ranges. Slipring Motors. These are installed in machine tools with larger drive outputs and those with flywheel drives, because the energy stored in the flywheel may be released during the working stroke when the motor speed is reduced. Synchronous Motors. These have developed from the permanently excited d.c. motor, where the role of the stator and rotor are interchanged. In synchronous motors the electrically generated exciter field in the stator rotates
mission units (see LJ .2.2). The controlled asynchronous motor is being selected increasingly for use as the main spindle drive of the machine, whereas the synchronous motor, even in its controlled form, is now being employed in specialised feed tasks. Both types of motor have a broad speed range (10' to 104 ), so that changeover gears are not normally required [2].
Squirrel-Cage Motors These are the most common type of construction, as they require little maintenance and offer stable operation under rated conditions. However, they have a high starting current with a low starting torque. Thanks to the variety of squirrel-cage rotor designs, the motor may be adapted to suit the characteristics of the machine too\. High-torque cage motors (deep-bar squirrel-cage motors) have an extremely high breakaway torque with a low starting current and are therefore suited to direct-on-line starting. High-resistance sqUirrel-cage motors have the highest specified breakaway torque (as high as the breakdown torque) and a low starting current. The rotational speed n may be changed in three-phase slipring motors by changing the number of poles p, as the speed is dependent on the number of poles p and the
in relation to speed. Permanent magnets are installed in the rotor. and the primary windings are attached to the stator to generate the field of rotation. The separation of the current supplying the stator windings is effected in relation to the rotor position angle which has to be measured for this to be achieved. Sensors for rotor position and speed measurement are typically non-contact transmitters, so that electrical torque transmission is not
required via collectors or brushes from the stator to the rotor or vice versa. The attributes of the design of synchronous motors, also known as brushless d.c. motors, therefore include lack of any need for maintenance and low heat buildup. At the same time, a more expensive electronic control system is needed than with d.c. motors, but the performance characteristics are broadly similar. Example. Figure 11 shows the principle of a six-pole, permanently-excited synchronous motor (the supply frequenl.'Y is three times as high as the torque frequency of the motor). In rotor position J the current flows into winding phase a and out of winding phase c, whereas in position 2 the current flows into winding phase b and out of winding phase c. The polarities set with a time switch in this way in the stator windings produce a torque in the same direction on the magnetised rotor, which sets the rotor in motion in a clockwise direction. Current supply of the winding phases is carried out alternately synchronised at 600 to the rotor position. The current supply of each phase winding occurs periodically with
Manufacturing Systems. 1 Machine Tool Components Stator winding coordinates (i,},,) and i", =-(i" +i,,)
Field coordinates (iSdlisq)
Sd [ I
Isq
.=T['Sl.]
where T=[
's2
cos\"+t sin\" .
1
sin\" -2 cos\"
Speed controller
_ .____ --[gJisq,Se~;"'____I Field table
T
FiekJ controller
L-r'.........-lL
Field coordinates -~_- Stator coordinates - -
Figure 8. Asynchronous motor control based on the principle of field orientation. W rotational speed, i current, isd flow-producing current components, isq torque-producing current components, 'P field coordinate angle, 'Y rotor position angle, s slip, Tr rotor time constant, T transformation matrix.
Figure 9. Structure of an asynchronous motor of the squirrel·cage rotor type (ABB): 1 holding brake, 2 motor and brake connections, .3 stator winding, 4 rotor winding, 5 measuring system, 6 thermistor.
200,-------,,----,-----,-----,----,20
4000
8000
n in min-1
12000
Figure 10. Typical characteristic curve of an asynchronous motor (AMK). P output, M torque, no rated speed, n max maximum speed, IjJ magnetic field strength, I constant torque, 2 constant power.
a period of 40° supply and 20° rest, where these figures refer to the rotor revolution. In electrical terms, this produces angle deviations of 120 0 and 60". The responses of the induced voltage in the individual phase windings and at the tenninals are also shown in Fig. 11. The response of the induced voltage within the current-free rest phase is, in principle, insignificant for constant power output. The trapezoidal response occurs as a result of skewing in the magnet plates on the rotor which is carried out to avoid problems with slotting [14, 151. A basic distinction is made in synchronous motors between supply with sinusoidal and square-wave currents. The advantage of square-wave current supply lies in the simplified signal forming and the use of a simple sensor for detecting the position of the rotor. Two different rotor position sensors may be employed for sinusoidal current supply, depending on the degree of accuracy required. The sensor with only three sensor elements detects the initial points for the V, V and W supply in each case, whereas the cyclical sensor operating with absolute values also provides additional information on the absolute rotor position in addition to the three sensor elements for detecting the initial points for more accurate sinusoidal supply. In general, supply with sinusoidal currents may cause attenuation in the harmonic waves and thereby result in high synchronisation in the drive [15,16].
1.2 Drives. 1.2.1 Motors
characterised by the fact that the servo characteristics have been much improved. The boundary circle frequency extends from 500 to 100 S-l, with speeds of up to 3 m/s and maximum acceleration even up to 10 g [181. lt is already clear that the use of servodriven linear motors of the asynchronous type is set to increase dramatically in the future.
EI2
Electrical d.c. Motors
ElZ
Electrical Shunt Motors. These are characterised by their high speed constant under load, and are employed where possible in main and feed drives for flywheel drives, with additional series field windings due to their stepless speed control characteristics. The high starting current in the ar-mature circuit is limited by preliminary resistors and/or by thyristor power units. The direction of rotation may be changed by interchanging the armature
E
or field connections.
The speed may be increased by increasing the ar-mature voltage with constant torque and/or field control with constant power and reduced torque. In the Ward-Leonard system the rotational speed which may be achieved amounts to up to B = 40, with a thyristor power unit and speed control in the main drives. Armature control range of B A > 50 is standard, and in feed drives with servomotors the range is even wider. The heat dissipation is lower at low speeds and external ventilation is therefore required.
E
Figure 11. Function diagram of a permanently-excited synchronous motor (Bosch). (,.hoc winding phase current, lla,h_c winding phase voltage,
Uah.lx,ca
terminal voltage, U, V, W, X, Y, Z motor ter-
minals, m internal torque, E intermediate circuit voltage. j
Electronic circuitry may be produced with the synchronous motor control as shown in Fig. 11 (see also Fig. 12). In this control concept, current is supplied in quasi-sinusoidal or (more precisely) trapezoidal form. The resolution required in the rotor position sensor for genuine sinusoidal supply would have to be far higher than that produced by three sensor elements offset to 60'. This is the only way constant-phase displacement or phase coincidence can be guaranteed between the induced voltage and phase winding current. There is no commutation limit for synchronous motors which is comparable to that in d.c. motors. The power tends to be restricted by the servoamplifier instead. A typical characteristic curve field of a synchronous motor is shown in Fig. 1~ for the purpose of comparison with the d.c. motor. The standard speed range is up to 3000 min- l , with a maximum output of up to 10 kW. Asynchronous Linear Motors. These have low power at low speeds with less power efficiency and generate high levels of heat [17]. They are suitable for use in forging machines as pile hammer drives and in flexible manufacturing processes as transport drives for work-
piece pallets. New developments in asynchronous linear motors arc
Permanently Excited d.c. Motors. These are employed exclusively in feed drives with speed control [19]. They may be employed as shunt-characteristic motors with permanent field excitation. The speed may be adjusted via the armature voltage, while the power supply is by way of thyristor or transistor electrical inverter units from the three-phase supply with smoothing reactors in series. Speed control is effected via a reverse tachometer (coupled directly to the main shaft), which enables the speed to be reduced to virtually nil while maintaining high rotational regularity, thereby producing a broader control range of B = 10' to 10 4 , suitable for continuous-path control operations. Special-purpose constructions (Fig. 14) reveal improvements in dynamic characteristics with the minimal mass moment of inertia compared with conventional constructions. Rotor wind-
ings have high specific loading and high phase spacing. High rates of increase of current due to low armature
inductance, highest possible starting torques (3 to 10 times rated torques, depending on the current limiter of the inverter and switching frequency), with short-term high current overload capacity, enable extremely high startup times of 5 to 50 ms. Figure 15 shows the structure of a d.c. motor control system with a transistor ampli-
fier. One special problem associated with d.c. motors is limiting the transmittable current. The reason for this is the type of current transmission which takes place via brushes and collectors. This provides a natural boundary for the maximum transmittable current, which may still be transmitted without damaging the contact elements. This characteristic is demonstrated in Fig. 16. The commutation limit is dependent on speed and decreases rapidly at increasing speeds. A speed-dependent current limiter is usually installed in the servoamplifier to compensate for this. The ratio between the maximum available torque and the rated torque is thereby effectively reduced. The characteristics for the motor govern its design, as the
motor is normally designed for rated operation. One long-term handicap with d.c. motor applications
..
Manufacturing Systems. 1 Machine Tool Components
Peal< current reauction
Rotor position-related directional current controller
Controller enable
Electronically commutated synchronous motor
.I\--,
l. Intermediate circuit voltage
,10V
Speed setpoint
~--~--~~fL--~----~1
I I
I
'~+---~I
I
I I
I
_J
Ble 1
Ble 2
Ble 3
From supply module
ii
Synchronous generator with phase-sensijive power rectification
~ -~ady for operation
Speed listing value Figure 12. Structure of a synchronised motor control system (lndramat).
has been the relatively short lifetime of the commutator, which incurs high maintenance costs owing to the frequent replacement of worn parts. This has led to a prefer-
ence in favour of the three-phase slipring motor over the past decades. Commutator technotogy has advanced in recent years, however, and tifetimes of over 30 000 hours are now being achieved. Direct-current technology is therefore rising in popularity again [20]. Bar-Type Rotors. These have a thin, unslotted rotor with a uniform winding and high winding density, rotational speeds of up to 3000 min-' (in certain con-
.-
r,
!\
215 [, 2 186 ----------------: \-~ I
E ;;;.
v
: 1)-- 3 I '~
Short-time operation max. 300 ms
10.0 ~10'5. __
Q)
129f-r....~~~~~~~~_.,Irrmnl
Short-time operation 65% of 15 min.
e-
.=t
• Extremely low-mass inertial torque • Broad speed control range a • Great positioning accuracy
• Robust motor
• Low mass inertial
~ 2.
e
torque
b • High speeds
r.==::J
-=
,:?
Continuous operation
5
6 1000 Speed (min-')
2540 3000 2870 e
Universal motor for standard applications
Figure 130. Characteristic curve field of a synchronous motor (Indr-dmat). 1 maximum torque, 2 voltage Limiter at rated current, 3 voltage limiter at 15% supply undervoltage, 4 continuous torque, 5 maximum speed, 6 buckling speed.
Figure 14. Constructions of d.c. motors: a disc-type rotors, b bartype rotors, c hollow-type rorors, d slow-type rotors, e conventional construction.
1.2 Drives. 1.2.1 Motors
iActuai
External current limiter
Figure IS. Structure of d.c. motor drive circuitry with transistor amplifier (Siemens). n speed of rotation, i current, G tacho-generator, M motor.
M,I
such, may act both as motors and as measuring devices. Highest speeds are approximately 3600 min-I. The startstop frequency is limited in relation to external inertial torque, with no loss of increment. Owing to their low torques, they fmd application only as feed motors in smaller machine tools and as controlled auxiliary drives. Hydraulic torque amplification is required for larger machine tools. Linear stepper motors were developed for sequence switching in five-stator rotating stepper motors. They are used in small, double-axis, point-<:ontrolled machine tools (e.g. boring and drilling machines fitted with Fujitsu motors and can achieve increments of 0.1 mm).
Mmax.!mox
nmox
n
Figure 16. Progression of speed-dependent current limiter (ABB): 1 commutation limit, 2 charJ.cteristic curve of motor, 3 current limiter, e.g. /0 := 10 A, II = 30 A, [max ::= 100 A, n'nax = 2500 min -], Mo ;::: 8 N m, M]
:=
21 N m, Mmax = 69 N m.
ditions up to 14 000 min -I) and require backlash-free gearing connected in series (rapid-type rotor).
Disc-Type Rotors. These have a lightweight, fibreglassreinforced, non-ferrous, synthetic resin disc-type rotor with bonded-on current conductors running between permanent magnets. Rated speeds are 2100 to 4500 milC I, maximum speeds up to 6000 min-I. Hollow-Type Rotors. These have a bell-shaped winding assembly which is surrounded both inside and outside by the field system. Slow-Type Rotors. These have large numbers of poles (torque motors), and typically an annular, slotted, largediameter rotor. The speed range of n < 1 min I to 1200 min-I enables the provision of a direct link to feed spindles without intermediate gearing (freedom from backlash) for high torques of up to 4000 N m. Electrical Stepper Motors. These have three, five or more stators and may execute angular and position increments with the appropriate field control and, as
Hydromotors Rotational Hydromotors. These (G2.2) are mainly found in machine tool applications in feed drives, acting as direct main drives only on special-purpose machines. The most common constructions (including pump operation) are gear-type machines, vane-type and radial machines, and axial and rotating piston machines. Typical applications are in pump motor systems with stepless speed adjustment or as electrohydraulic motors. Figure 17 shows the structure of an electrohydraulic feed drive based on the positive displacement principle, and a system under constant pressure. Index 1 shows the supply side under constant system pressure. On the delivery side, the adjustable hydromotor supplied direct from the mains supply produces linear motion in the machine slide with the aid of a threaded spindle. The hydromotor setting is actuated via the adjusting piston, which is controlled in tum by the output of the positional controller, the feedback on the regulating distance y and the spindle speed n 2 . The oil flow irQ through the valve regulates a double-sided acting cylinder piston, which alters the intake volume of the hydraulic motor according to the rotational speed and/or setpoint position of the slide to be controlled. In practice, the displacement controller is energy-saving, as the adjustable pump which is driven by an electrical control signal generates hydraulic power only as and when the drive (delivery) requires it. One disadvantage, however, is that the timing is somewhat slow, as
Manufacturing Systems. 1 Machine Tool Components
Constant supply of system pressure Non-return valve
1.5
limiter
I
zp/zpm" o 1.0
1.0
servomotor
1O}5
Tank line
0.5
0.5 Supply side (Index 11
0.25
Delivery side (Index_2_l~~~~~_
a
.~E "-
.",-~
-OS
I
-0.25
i
-OS
-O}5 -1.0
-1.0 -15 -1.0 b
-0.5
0.5
Figure 17. E1ectro~hydraulic feed drive based on displacement control principle [21,22], a Circuit arrangement: p pressure, V volumetric flow, n speed, x path, Uf. control VOltage, 1ml reduced mass moment of inertia, FL force under load, T, P, A, B valve connections, y regulating distance. b Characteristics of servo-control pump: ~PL pressure change under load, Zp pump setting. VI. volumetric flow under load.
fairly large masses have to be moved over long distances (e.g. 10 to 100 kg mass over a distance of some 10 to 100 mm). This is why this control principle is mainly of
interest in high-power applications [21- 24].
Hydraulic Linear Motors (Hydrocylinders). These are employed in machine tools. as the main drives of planing. shaping and slotting and broaching machines, in presses, in the feed drives of grinding machines, in crossover turning machines and machining centres, and also for auxiliary drives, e.g. for automatic tool change in machining centres or on workpiece transport systems in transfer lines. Electrohydraulic Motors. The electrohydraulic servomotor consists of a hydromotor or hydrocylinder which is driven via a flanged electrohydraulic servovalve. It is suitable for the continuous-path direct drive of feed spindles thanks to its excellent dynamic characteristics, which consist of low mass inertial torque, high torque, short startup times and speed ranges of B = 10' to 104 (with tachometer feedback). Figure 18 shows the structure of an electrohydraulic feed drive based on the prinCiple of rheostatic control under constant system pressure. The proportional control valve and the servomotor form the drive which moves the slide via a threaded spindle. The slide position XaClUal and the motor speed nactual are obtained and fed back to the positional controller and/or speed controller, The system deviation regulates the volumetric flow Vl to the motor via the valve and the speed is adjusted accordingly. Rheostatic control is characterised by its extremely good dynamic behaviour, but also by its low level of efficiency due to high energy loss as a result of the inductance effect. This high dynamic response may be attributed
the movement of low mass over very short distances (e.g. 0.1 kg mass over a distance of approximately 0.1 to 1 mm) in the valves. Rheostatic control is normally employed in applications at ratings ranging up to 10 kW. Displacement control may be substituted for even larger power ratings [21-24]. to
Example. Figure 19 shows the combined roller vane motor with a two-stage servovalve. The input quantity of the valve is the low control current i, while the output quantity is the proportional oil flow qA and/or qp, which may be translated into proportional speed in the motor. Power amplification is between 10.3 and 10 5 . Control current i causes displacement of the flapper (nozzle-baffleplate system, Stage I) via control coils and armature. As a result, different pressures are produced on the left and right side of the piston valve, causing this to be displaced (Stage II). Depending on the setting of the piston, the oil under pressure flows to A or B. Constant regulators and nozzles (adjustable-gap inductors) fonn the basic bridge connections. Extremely fine-grade filtering of the oil is required for the sensitive inductor system. The effect of the inductor causes pressure loss and extreme heat generation in the valve, and coolers are therefore reqUired in most case~. The roller vane motor converts the direct tangentiaL force of the internal supply of oil under pressure into continuous torque with a positively driven rotor and rotary slide valves. Two diagonally arranged rotary slide disc valves each revolve in relation to the internal diameter of the rotor to seal the rotary pressure chambers. The rotary slide disc valves are thereby driven by the motor shaft via the planetary gears at a ratio of 2 : 1, and lens-shaped depressions are provided for the passage of the rotor. This arrangement is compact and fuUy symmetrical, thereby providing uniform operation even at low speeds.
Electrohydraulic Stepping Motors. These are employed where the torque of electrical stepping motors is not adequate for feed drives. Hydraulic torque amplifiers are then coupled to the stepping motor shaft
1.0
1.2 Drives. 1.2.2 Transmission Units
...
1.S 1.0
0.5 B ,S"
Figure 19. Electrohydraulic servomotor. a Ekctrohydraulic servovalve (Moog) with equivalent cross-.section. b Roller vane motor (Hartmann) in longitudinal section and cross-section (actual size ratio = 1 : 5). I control coils, 2 flapper, 3 hydraulic amplifier, 4 filter, .5 constant inductor, 6 piston valve , 7 consumer's terminals. 8 rotor, 9 rotary disc valve, 10 planetary gears, 11 oil under pressure , 12 oil return , 13 leakage oil, 14 roller vanes; Po (= 140 bar) pressure of supply unit , qu available oil volume , PR ( = 0) oil return.
E
.....
-0,25
,~
-0.5
-0}5
-to
..
-1.0
-1.5 -I,D
- 0.5
0
0.5
1.0
6Pl/PO
Figure 18. Electrohydraulic feed drive based on the principle of rheostatic control [21. 22]. a Structure: I proportional control valve , 2 servomotor, .3 slide, 4 spindle, 5 bearing, 6 controller, Pu constant system pressure, V volumetric flow, nA<-"\ual spindle speed, X slide position. b Characteristics: y regulating distance, Po system pressure,
/j.PL
pressure change under load, VI. volumetric flow
Figure 20. Electrohydraulic stepper motor (Fujitsu). 1 electrical stepper motor, 2 servovalve,.3 hydromotor, 4 bolt, 5 nut , 6 connection for oil under pressure. 7 oil return line.
under load.
(Fig. 20), these consisting of a servovalve and hydrometer. These two form the internal hydro mechanical servocontrol circuit. The piston valve of the servovalve and the hydromotor shaft are mechanicaUy coupled via a bolt and nut connection. The system deviation is produced at this point. The rotary action of the piston valve produced by the stepping motor or angle deviation of the hydromotor shaft causes axial displacement of the piston valve, the oil flow released driving the motor impeller until the piston valve is screwed back to the midpoint setting. 1.2.2 Transmission Units Mechanical Gears In machine tool construction, these gears primarily reduce the rypicaUy high speeds of the motors to achieve the operating speeds and to generate defined feed motion in the tool supports [28]. Distinctions are made between gears for uniform and non-uniform transmission (see F9) .
Gear Cone Transmission Units [25] . The smallest functional group in the gear train consists of a single pair of gears, where the gear and mating gear are located on
different shafts. Transmission ratio i is given by i = ndriving/ndriven::;: driving speed/driven speed. The gear ratio u is given by u = IIi = z,lz, = number of driving gears/number of driven gears (see F8.1).
Types. There are a number of gear-switching mechanisms, e.g. pickoff gears, change gears [26, 27], sliding gears, and mechanically and electrically operated clutches. The smallest switching mechanism is the two-speed basic gearbox with two driven speeds, while the next largest unit is the three-speed gearbox with three driven speeds (Fig_ 21) The arrangement of the sliding gears for sliding-gear transmission units is possible on both the drive and the output shafts. The lighter gears are displaced in such a way that less mass is to be moved and shorter gearshift forks are required owing to the smaller diameter, Switching takes place via clutches in power Shift gears. It is therefore possible to carry out switching under loads and in a rotating condition. Gears with a number of driven speeds are produced by switching the basic gearboxes in sequence. Two twospeed basic gears (Fig. 22) switched in sequence produce a four-speed gear with four driven speeds, for example.
Manufacturing Systems. 1 Machine Tool Components
always switched via a clutch. The flow of force either passes directly from shaft I to shaft III or first to shaft II and from there to shaft III. In the first case, the gears are engaged without effect, so that driving and driven speeds are equal. In the second case, a large overall transmission ratio is achieved by switching two pairs of gears in sequence. A small overall volume is produced as a result of the design structure (feedback of flow of force to the coaxial shaft III). Transmission gearing is normally attached to the driven shaft to provide operation with the gearing for as long as possible at high speeds, i.e. at small torques.
T
b
Design. There are graphics aids for the design of speed
c Figure 21. TIrree·speed basic gearbox. Sliding-gear transmission: a standard arrangement; b expanded arrangement, Zb width of tooth; c power shift gear, 1 clutch.
.1,.J LJ
LJ
rates for stepped and continuously variable transmission units which simplify the task considerably (Fig. 24) [28J. The modular network (Fig. 24a) forms the basis of the preliminary gearing design. It shows the various possible different divisions for the progressive ratios within the transmission unit and the individual switching blocks. The modular network is always drawn symmetrically. Each connecting line corresponds to a transmission speed. The size of the individual transmissions is not determined at this point. In addition to the modular network, there is also the speed flow diagram (Fig. 24b) showing the speed of each shaft and size of the transmission. The output speeds are shown in the speed flow diagram in geometric progressions (cp = constant) at equal intervals using a logarithmic scale. This interval may be viewed both as the ratio between two consecutive speeds and the power of cpo The transmission is characterised by the gradients of the connecting lines of the speeds of two consecutive shafts. In Fig. 24b these are, for example i1 =
=
i.-'J
= cpO = 1,
i4 = cp2 = n 4/n Z = n;"Inl>
Other aids employed in gearing design are the gearwheel arrangement (Fig. 2oic) and the flow of force diagram (Fig. 24<1). The gearwheel arrangement shows
1 1 ---r
b
>4lb
c
LJ
LJ
OJ
OJ
J I
>5lb
Figure 2:2:. Four-speed three shaft gear: a basic gearbox; b single mesh gearing; c double mesh gearing, Zb width of tooth.
Meshed gears are employed to produce smaller overall lengths and save on gearwheels. These are composed of one or more wheels of different driving gears. The meshed gearwheels are marked by shading areas in Fig. 22. As the meshed gearwheels are engaged with two gears, the same module has to be employed for all three wheels. The size of the module is determined by the driving gear with the largest torque, as a result of which larger shaft-centre distances may be produced under certain conditions. A shorter overall length in axial terms therefore means an increase in the gear in radial terms. One construction commonly employed is transmission gearing (Fig. 2~), which consists of three shafts and is
a
:or
1 J
.:lmJ .:~
tB]~ml
tHE~m
Figure 2:~. Transmission gearing: a gearwheel arrangement; b modular network, I clutch; c speed flow diagram.
1.2 Drives. 1.2.2 Transmission Units
2
f--+l-t-f--+-- Drive
a k
f - - - t - - nJ
d
'---'--- n4
b
B > 9'sl
c k>l
Speed range
b Figure 24. Basic requirements for gearing design. a Modular net· work. b Speed flow diagram, B speed range. c Gearwheel arrangement. d Flow of force diagram.
the arrangement and number of shafts, gearwheels and any clutches which may be employed. Symbols are employed to illustrate the construction. The flow of force diagram shows which wheels transmit the flow of force in the individual switching positions. The flow of force diagram may also be employed to establish how the individual blocks are to be switched to produce a specific output speed. Further developments in the control of electronic drives have enabled an increasing variety of combinations of step less electrical drives to be employed with stepped transmissions as the main drives for machine tools (Fig. 25). The operating speed Bo of the stepless drive may be extended by a stepped transmission connected in series. A speed coverage of k > 1 is desirable, so that all
speeds within the operating speed range Bo may be achieved. Thus: B tot = BoBso
k=Blrps<
where Bo = the operating speed range of the d.c. motor, Bs< = the operating speed range of the stepped transmission, 'PSt = the progressive ratio of stepped transmission, k = the speed overlap. Example. Figure 26 shows the structural design of a twelvespeed c1utch-controlled gear in the spindle box of a numerically controlled turning machine with the respective speed flow diagram. The drive consists of a pole-changing three-phase stipring motor with a V-belt and V-belt pulley at shaft I. The speed switching is carried out via slipring-free electromagnetic multiple-disc clutches KI to K". The figures in the speed flow diagram and at the gearwheel speeds show the transmission ratios. They should amount to between 0.5 (speed-increasing) and 4 (speed reduction) depending on the perfonnance and the overall size. The progressive ratio amounts to 'P = 1.6 in the example selected; smaller progressive ratios are calculated in the main operating range of 'Po." = 1.25.
flexible Drive and Friction Drive Belt Drives (see F6). These are suitable for the trans-
mission of rotaty motion between the motor and gears or directly to the main spindle in machine tool construction.
IIIIIlIIlIIIIl Range 1 mmII!!!ll Range 2 Range 3
~
Figure lS. Combination of controllable electrical motor with range transmission. a Block diagram: I controllable d.c. motor, 2 range transmission. b Negative overlap. c No overlap. d Positive overlap.
Their advantages include shock suppression and protection against over-ranging. Rapid traverse speed in spindles commonly uses smooth-running belt transmission. The tension force is taken up by separate bearings at slow speeds via toothed wheels. Flat belt transmission is employed for the highest speeds and lowest torques, e.g. for grinding spindles, otherwise V-belt drives are the norm, with toothed belts
and nylon toothed belts less common, the latter also usable in conjunction with oil. High-torques applications may feature a number of belts switched in series. Stepped speed pulleys are installed in high-speed spindles and with lower power ratings, e.g. in small-scale drilling and boring machines and small, rapid-turning machines. These have significant downtime for belt adjustment. Even belt tension is achieved at all speeds, where the shaft -centre distance is given by a 2': IO( d m " - d min ), where dm" is the maximum belt pulley diameter and d min is the minimum. The total belt pulley diameter must remain constant by comparison. An idler pulley must be provided for smaller shaft-centre intervals. Chain Transmission (see F6). This is normally used only
in conjunction with roller chains in machine tool construction for auxiliaty and transport motion. Low-noise, inverted-tooth chains are also employed in feed and main spindle drives of small-scale automatically controlled machines. Stepless chain-driven gears are employed primarily in main drives of up to 40 kW. With positive, multiple-disc chain drive the speed range B is up to 6, with roller chain construction up to 10. Non-positively driven roller chain is also suitable for higher speeds. Wheel gears are frequently connected in series to extend the speed range (see the
Manufacturing Systems. 1 Machine Tool Components
Crank Mechanisms (see F9 and FlO). These are installed in machine tools for straight-line back-and-forth motion, where non-uniform speeds are admissible or even desirable [29J.
Slider-Crank Mechanisms. These have equal back-andforth times, i.e. 50% non-productive time. Consequently they are rarely employed in meta!-cutting machine tools, although they are frequently in use in metal{ormtng machine tools. In this case the adjustable crank journal is extended to form an eccentric cam. The connecting rod may be strained to buckling and is consequently designed to be short and compact. The rotating joint (connecting rod journal) consists of ball bearings in a ball cup. A flywheel is provided to ensure even operation of the machine tool. Crank-Rocker Mechanisms. These are employed in shaping and slotting machines. In Fig. 28 the shaping and slotting spindle is driven by a rocker arm with a toothed wheel section. The plunger stroke may be adjusted on the crankwheel (eccentric cam).
I -------------------~~~------
b
n,
Figure 26. a Twelve-speed clutch-controUed gear of a numerically controlled turning machine (Gildemeistcr, Bielefeld), b Speed flow
Slider-Crank Mecharlisms. These are used in horizontal shaping and slotting machines as oscillating or totating sliders to achieve more rapid return times (see F9). It is normally necessary to calculate the dynamic forces only for reversal points; static forces may be determined from diagrams. Slider-crank mechanisms are employed in infeed areas of shaping and slotting machines for step-by-step infeed motion via stops or a locking wheel. The crank or wrist pin may be adjusted at this point.
Electric Transmission
diagram, K J to K.;, slipring-free electromagnetic multiplc-disc clutches. c Drive via pole-changing three-phase slipring motor and V-belt. Figures arc transmission ratios.
The Ward-Leonard speed control system traditionally employed in many d.c. motor control applications has become outdated and is being superseded by power elec-
planetary gearing in Fig. 27) . Specifically compact construction is availabJe via torque division.
now be employed in high-precision feed drives, e.g. in gear-cutting machines for toothed wheel production with control electronics (see K5, Fig. 22).
tronics, i.e. thyristor drives. Stepless electronic drives may
Frictional Gears (see F7). These are infmitely variable, and are employed in main and feed drives in medium to small-scale boring, drilling and turning machines, where a limited speed range of B < 5 is adequate at high speeds.
Figure 7.7. Stcples.o;; (prv = peak inverse variable) chain transmission 1 with planetary gearing cOlUlected in series 2 (arrows mark the flow of force at the torque division), 3 drive. 4 output , 5 internal
gear, 6 sun wheel, 7 planetary gear wheels, 8 web.
Hydraulic Gears
These use pressurised liquid, normally oil (see G3), for power transmission. The hydraulic gears employed in machine tools are virtually all hydrostatic ones. These remain virtually unaffected by the kinetic energy of the liquid flow, in contrast to hydrodynamic gears. The liquid serves purely for the transmission of the compressive forces. With hydraulic transmission step less adjustment of
Figure 2:8. Drive of the shaping and slotting spindle 1 of a gearshaping machine (Lorenz, Ettlingen) with crank-rocker mechanism 2; 3 eccentric disc (drive), 4 adjustable crank journal, 5 connecting rod, 6 <.:yliodrical toothed rack , 7 helical guide bushing, 8 cutting wheel.
1.2 Drives _ 1.2.2 Transmission Units
the output speed may be carried out within broad limits. Constant operating speeds are achieved, also bump less changeovers, and the oil pressure may be exploited for clamping and control motion and for braking. The hydropumps and motors employed are circulating gear-type or vane-type pumps with constant or variable
----"".~
VA
delivery volumes or straight-stroke piston pumps. Pumps and motors may he of similar or different constructions. Rotary drive and output motion Of rotary drive with linear reciprocating output motion may be obtained, depending on the composition [30-32, 34J.
Hydraulic Gears with Rotary Drive and Output. These are used in broaching, planing and surface grinding machine tool applications. Both pump and motor are located in a single housing and may normally be adjusted independently of one another. The power characteristics of a hydraulic gear are similar in principle to those in electrical transmission. The selection and design of the oil circuit are important (see G.1, Fig. 1). The pump removes the entire delivery stream from the tank in an open circuit (Fig. 29a), whereas the return oil from the motor, minus the leakage oil, is fed back to the pump in a closed circuit. In a closed circuit (Fig. 29b) the motor is "hydraulically clamped", so its torsional rigidity is higher than in an open circuit. The dosed circuit is therefore suitable for braking, rapid change in the direction of rotation and feed drives where the machine tool table has a tendency to stick or slip. Precautions have to he taken to ensure that the heated oil in the circuit is continuously replaced by oil from the tank or cooled with additional aggregates to achieve the required heat dissipation.
Hydraulic Gears with Rotary Input and Linear Output. These may be used for the main motion in planing, slotting, broaching and flat grinding machines and presses, for feed drives in constnlctiunal units and automatically controlled machinery, and for auxiliary and tension motion in devices. The oil supply to the cylinder is via a constant pump in the reactance circuit or a servopump with a rapid-traverse switch. Reactance Circllit (Fig. 30). This is eqUipped with a
constant delivery pump (see G3.3.2). Fine tuning of the oil flow is via an inductor, with free flow on the return of piston provided via a directional valve. Where possible the inductor should not be installed in forward motion, since back pressure then builds up via forces within the table and there is a danger of chatter occurring with tluctuating forces. Where the inductor is located in the return motion, back pressure builds up which stabilises the piston (back pressure control). This back pressure is even more effectively controlled where a variable-displacement
Figure ~O. Inductor circuit in back-pressure operation with four/three-way directional con· trol valve.
Figure 31. Rapid-traverse switch with four/three-way directional control valve and additional three/two-way relay valve.
pump is installed as a gauge box in place of the inductor. Although the reactance circuit is cheaper and simple, and extremely reliable in operation, it is less effective, as the difference between the continuous tlow of oil and that consumed in the cylinder is discharged via the pressurereducing valve causing heat dissipation.
Rapid-Traverse Gears. In cylinders with single-sided piston rods (Fig. ~O), larger piston surface areas are normally employed to produce greater power and slower operating speeds 1'." with higher rapid-traverse speeds vI' in smaller ring surface areas with less power. Where the rapid traverse also applies to the direction of operation, an additional relay valve is employed (Fig. ~1). The cylinder chambers are interconnected in position 1, where the oil is exchanged and the total volume of oil conveyed by the pump responds to the smaller differential surface area (corresponding to the piston rod cross-section). A smaller volume of oil is required as a result, so fixed-displacement pumps are frequently employed instead of variable-displacement ones, with a considerable saving in costs. Pneumatic Gears
These are normally employed in machine tools as cylinders for automatic clamping, auxiliary and transport motion (see G5 and 135]). Advantages are simple installation, extremely reliable operation and high operating speeds of up to 3 m/s. Disadvantages include low rigidity of air cylinder, non-unifonn motion with fluctuations in forces under load and frictional forces (though these may be overcome with hydraulic regulator), with intermediate positions difficult to canto!. With larger air cylinders consumption costs are high. Noise is generated by air being discharged, but this can be avoided with the aid of a silencer. Standard supply pressure p is 4 to 6 bar, with a maximum of lObar. Piston force F is given by the formula F = 1)jJAw, where Aw is the effective cross-section. Efficiency 1) is 0.8 to 0.95, depending on the pressure and the size of the cylinder.
Single-acting Cylinde,-. This is employed for clamping, lifting, extracting, etc. There is a standard production size with a stroke of up to 100 mm. Return action is brought about by a spring mechanism or dead weight. Figure 29. Oil circuits. a Open circuit with servocontrol pump and four/three-way directional control valve. b Closed Circuit without directional control valve with reversing servopump.
DOUble-acting Cylinder. This type may also be titted with a penetrating piston rod. Where uniform operating feed is required, pneumatic action must be combined with
...
Manufacturing Systems. 1 Machine Tool Components
2
hydraulic action, either in separate cylinders (Fig. ~:la) or in a common cylinder (Fig. ~:lb). The latter construction is more compact [33]. 1.:I.~
Mec:banical Feed Drive Components
All parts and machine components located in the flow of force between the motor and tool or workpiece constitute mechanical feed drive components. The following feed drive components are significant: gearing for translating rotary motion into linear motion; gearing for speed/torque alignment; couplings; bearings; joints. The design of these mechanical feed drive components contributes considerably to the performance and precision of a numerically controlled machine tool. The main criteria for their design are as follows:
-
High geometric and kinematic precision High rigidity and freedom from backlash High initial resonant frequency Low mass moment of intertia and mass of moving mechanical parts.
FIpre 33. BaIlscrew with backlash elimination. I fir5t bail nut, Z second ball nut, 3 ball return, 4 preload spacers, 5 ballscrew.
The ballscrew mechanism is ideally suited to the requirements of the transmission characteristics of feed drive components. It embodies the following principal
The low system damping is a disadvantage, however. As the ball bearings roll along the retaining grooves of the spindle and the nut, they move in a tangential and axial direction. The ball bearings, therefore, require a recirculating mechanism (see F4). The ballscrew and nut assembly cannot, however, be manufactured to provide the full elimination of backlash. To achieve this, i.e. minimum backlash on reversal and high overall rigidity, the ballscrew and nut assembly must be preloaded. Double or single nuts are used for this. Where double nuts are used, the preload is achieved by pushing the two halves of the two nuts together or apart using preload spacers (Fig. ~~). Single nuts are preloaded in an axial arrangement where the relevant ball bearing tums are offset by a distance tJ. The rigidity of the system is directly dependent on the preload force generated and on the number of load bearing ways. A minimum feed must be maintained in addition to any external loading in order to produce the required rigidity in the sys-
advantages:
tem.
There are also requirements relating to adequate damping and low friction and linear transmission characteristics of the components.
Feedscrew and Nut Drive This is the machine tool component most commonly employed for translating rotary motion into linear motion in feed drives of machine tools. Acme tbread spindles (see F1.6.3) with bronze nuts are used for simple systems, and ballscrew mecbantsms are employed in modem, high-precision, numerically controlled machine tools (Fig.~~).
- Extremely high mechanical efficiency (11 = 0.95 to 0.99) owing to low roller friction (IL = 0.01 to 0.02) - No stick/slip effect (slide back) - Low wear and hence long service life - Low heat buildup - High positioning accuracy and repeatability as a result of adequate rigidity and the elimination of backlash - High traversing speed
FIpre 3:1. Pnewnatic-hydraulic feed units. a separate cylinder: b air cylinder, c oil brake cylinder. b common cylinder: b operating cylinder, f air-oil actuator. Working cycles with .: .low feed: air aI-bI-web b, oil cI-d-e2; rapid retum: air aZ-bZ-b, oil cZ-e-cI. Working cycles with b: rapid forward travel: air al-bl, oil bZ-cIf2; feed rate: air aI-bI, oil bZ-d-J2; return: air aZ-fJ, oiIf2-e-bZ.
The spindle bearing should also be mentioned as another important component of the ballscrew mechanism. Its function is to guide the spindle in a radial direction and take up the feed force in the axial direction simultaneously, while the spindle deformation and dislocation must be kept within admissible levels at the same time. This is why the main requirements of ballscrew bearings include high axial load-bearing capacity, high rigidity, low axial backlash, low bearing friction, high rotational speeds and a high degree of freedom from vibration. These various individual criteria take on a greater or lesser significance depending on the individual application. Whereas the rigidity of the bearing plays a major role in large-scale milling machines with high cutting forces, friction in preloaded bearings forms the main criterion for low-load grinding machines. A low-friction bearing is also essential at high speeds in this type of application. Thrust angular baU bearings or roller and needle bearings are normally employed in threaded spindle bearing mechanisms (see F4). Thrust angular ball bearings have a wide angle of compression (60°) and are therefore able to withstand high axial loads. They are placed against a second bearing, which acts as the return guide because of their single-sided action. Thrust angular ball bearings should preferably be employed in pairs or in groups in X, o or tandem arrangements (Figs 34 and 38b). In order to prevent misalignment or compensate for this more easily, it is recommended that matched bearings are installed in the X arrangement because of the small area of support. Roller and needle bearings are employed as complete needle-thrust cylindrical roller bearing units. The special
1.2 Drives. 1.2.3 Mechanical Feed Drive Components
Preload loree F,
=c> in !he spirxle _
in the casil1g_
Figure ~4. Examples of bearings for ball screws (SKF , Schweinfurt). a Feed spindle bearing for low loads, single-sided clamping. b With high rigidity, double-sided clamping.
collar in the thrust bearing simultaneously acts as the outer ring of the needle bearing. Thc width of the inner ring is thus adjusted to fit the outer ring with the respective axial cylinder roller collars, in such a way that a specific axial preload may be achieved by tightening the groove nut (Figs 38 and 40). Both types of bearing require grease or oil lubrication . Bearings are constructed in different ways, depending on the loading requirements. A threaded spindle permanently fIXed on one side in an axial direction with one free end is the standard solution for small loads (Fig. 34a). A rigid spindle guide is required in feed drives with high rigidity requirements, and double-sided clamping is preferable (Fig. 34b). Thrust angular ball bearings are used to achieve high rigidity at both ends as back-to-back bearing seats in a tandem a!T".lngement. The spindle is distended as a result of this and the preload of the ball screw mechanism is consequently increased. Spindle feed with rigidly clamped spindles should be designed so that it is not off-
2
set by the operating forces or the spindle expansion
occurring as a result of frictional heat. These various different types of bearing require axial rigidity chamcteristics for the ball screw drive which are dependent on the tmverse path of the feed slide . The rigidity of the a=ngement in traditional types of bearings with a fIXed bearing and a moving bearing is reduced hyperbolically according to the distance of the slide from the fIXed bearing. Where the second bearing scat is also designed as a fIXed bearing, a rigidity curve which is latemlly reversed may be superimposed, so that a symmetrical curve is produced (Fig. 3S). The ovemll rigidity is thereby considerably greater where there are two thrust roller bearings and this is virtually constant over a large area at the centre of the spindle. The following constructional rules should be observed in the design of rigid spindle bearings in general: - Needle bearings and roller bearings should be used in preference to ball bearings where possible owing to their line contact and hence greater rigidity. - Thrust bearings should always be preloaded. - Rigid connections should link separable surfaces (rigid jointed bolts). - Bearing and spacer rings are to be a voided where possible, so as to obtain the smallest possible number of contact surfaces which reduce rigidity .
kM ~ kL km;n ~E x02/4 L -+----t kMlkm;n d
0.25
0.50
0.75
100
Slide position Figure ~S. Rigidity rate characteristics c of a spindle drive with a sing1e-sided flXed thrust bearing and b double-sided flXed thrust bearing; I single thrust bearing , 2 two thrust bearings.
- Fitting surfaces and spacer surfaces should be ground to provide a high contact area ratio and high rigidity. The structural design of the ballscrew mechanism should be carried out in conjunction with the preset pammeters relating to loading, traverse path, traversing speed and positioning accuracy according to the criteria of rigidity, tensile strength, bending, critical operating speed) mass moment of inertia and service life. This normally concerns the spindle diameter where there ultimately has to be a compromise between the rigidity requirements and the mass moment of inertia.
Manufacturing Systems. 1 Machine Tool Components
Example. The required torque of the spindle Msp:: F~"h/(2'ITTJ) with thread pitch band efficienc..'Y TJ with a given axial force FaxFor ballscrew spindles 1/ = 0.8 to 0.95, for Acme thread spindles 1) =: 0.2 to 0.55, corresponding to a pitch angle of between 2 and 16°. The relation between linear speed v and rotational speed nsp of the spindle is given by nsp ::= lIlb.
Rack-and-Pinion Drive Long feed spindles would be severely deformed by the axial load and dead weight where there are long traverse paths, e.g. in long turning automatic lathes, longitudinal milling machines and apron-type horizontal boring machines. They have a tendency to buckle. There is also the danger that the spindle rotational frequency may fall within the natural bending frequency range of the spindle. It is therefore recommended that rack-and-pinion drives are employed where the traverse path is greater than 4 m. Any length of feed path may be achieved with jointed toothed rack sections. The overall rigidity of the rack-andpinion drives is always independent of the traverse path in such a case, and is primarily determined by the proportions of torsional rigidity from the pinion shaft and the pinion-toothed-rack couple. The power transmission at the pinion is characterised by the extremely low rotational speed and high torques, and additional gear speeds are therefore required. The construction of the entire drivetrain should be torsionally rigid and free from backlash. Backlash elimination is achieved by preloading two equal tt'dins of gears A and B (Fig_ .J6), where their pinions are engaged in a single toothed rack at the high speed. Backlash is eliminated in the last three gear speeds by reciprocal preloading of the divisions. The last three speeds of the two trains of gears are preloaded, i.e. free from backlash, as a result of the axial shifting of a helical toothed pinion shaft with the two gearings via spring washers or hydraulic pistons facing in the opposite direction of skew. Gearing faults may thereby be offset. The slide is driven via the train of gears A,
depending on the direction of travel of the slide, whereby the train of gears B is tightened free from backlash, or vice versa.
Worm-Rack Gearing Worm-rack gearings arc frequently employed instead of the rack-and-pinion drive where mUltiple step gearing is to be avoided with long traverse paths. Work-rack systems are constructed with a hydrostatic lubrication system
Figure 36. Backlash-free feed drive with rack-and-pinion system. A, B gear divisions, I drive, 2 bracing mechanism, 3, 4 with rack,
6 braced pinion (output), 5 positive and negative bevel gearing.
to ll1lO1ll1lSe friction (Fig_ .J7). The worm is equipped with oil-pressure pads which are only fed with oil under pressure in the area affected on the tooth surfaces in the rack internally via a steady distributor (control plate). The rack tooth surfaces are plastic-faced. The high-precision construction is achieved by way of a second casting process from an impression of a master worm before the plastic applied is left to harden. The oil-pressure pads on the worm flanks are produced in the milling process. In more modem constructions these pads may be positioned directly in the rack tooth surfaces, so that these may be manufactured economically directly during the casting process by way of wax-coated material bonded to the rack tooth surfaces. The supply of oil under pressure is still provided via the worm.
Feedgear Mechanisms Additional feedgear mechanisms are installed in feed drives between the motor and ballscrew or rack-andpinion shaft ;0 cut the high motor speeds to match suitable spindle or pinion speeds with higher torques and to further reduce the slide-side mass moment of inertia in terms of the engine shaft. The gear mechanisms should be torsionally rigid, with low inertia and no rotational backlash. Toothed gears should therefore have wheels of small diall1eter, because this increases to the fourth power in terms of the mass moment of inertia. One 'way to eliminate backlash in practice is with tangential preloading of the meshing wheels. A toothed wheel is divided for this purpose. The two halves are twisted against one another until the desired backlash elimination is set with the counter wheel from the width of the two toothed wheels (Fig_ .J8a). Otherwise, it is possible to position the toothed wheels or wheel shafts in adjustable eccentric bushes. The axle distance may be varied by turning the eccentric bush until the tooth surface backlash is elimin· ated. Nowadays, synchronous belt drives are being used in many applications instead of toothed-wheel drives, in conjunction with threaded spindle-and-nut systems (see F6), where additional speed is essential on constnlction grounds (Fig_ .J8b). The synchronous belt drive satisfies the requirements for feedgear mechanisms employed in numerically controlled machine tools as far as rigidity, power transmission and precision are concerned, thereby providing a particularly economical solution. Tensioners made of fibreglass or stranded steel provide a high level of tensile strength, bending strength and minimal strain.
Figure 37. Hydrostatic worm rack gearing (Waldrich, Coburg). 1 worm, 2 worm rack, 3 drive wheel, 4 oil pressure pockets, 5 oil distributor. 6 supply of oil under pressure for front and/or rear tooth surface.
1.2 Drives. 1.2.3 Mechanical Feed Drive Components
flat-rype construction (a2). Both consist in principle of a rigid cylindrical ring with internal gear teeth (circular spline 1) which is pennanently jointed to the housing. Inside this ring there is an elastic steel bushing with external gear teeth (flexspline 2) which are pressed into the internal gear teeth of the circular splines and caused to rotate, thereby displacing an elliptical cam connected with the drive via a tightened piston bearing (wave generator 3), generating rotary motion between two opposite points in the large elliptical axis. A relative tum is produced between the circular spline and the flexspline owing to a difference of two in the number of gear teeth between these two, which is directly transmitted to the output in the can-type construction via the flexspline and in the flat construction via the flexspline and the dynamic spline (4). The drive and output move in opposite directions. The transmission ratio i is obtained from the number of gear teeth z in the circular spline and the flexspline,
a
al
a2
Figure ~8. a Backlash-free transmission gears for feed spindles (Scharmann, M()nchengiadbach): 1 feed spindle shaft, 2 direct current actuator motor. b Feed drive of bed-type milling machine with integrated toothed-belt gear mechanism (Maho AG).
The toothed belt is preloaded to increase rigidity and prevent backlash. The toothed-belt pulleys are made of aluminium to improve their dynamic response. The high material damping of the toothed belt material ensures lowoscillation transmission to the motor actuator motion. Moreover, the toothed-belt drive also provides for substantially more favourable structural design options, owing to the larger shaft -centre distance. This has led to feed drive concepts with smaller footprints and thereby smaller machine designs as a whole. The toothed-belt drive is ultimately also the most favourable manufacturing solution in tenns of cost owing to the small number of parts involved. The spindle bearing should be designed to provide exceptionally high axial rigidity in this drive concept, and the same applies for the radial and tilting rigidity. The special-purpose harmonic drive and Cyclo feedgear mechanisms satisfy the requirement of providing the highest possibk speed transmission with a compact
construction, high rigidity and coaxial input and output drive. The harmonic drive gear mechanism (Fig. ~9a) may be constructed as a can-type construction (a1) or a
Figure 39. Types of high-transmission gear mechanisms. a Harmonic drive (Harmonic Drive System GmbH, Limburg); a1 camtype construction; a2 flat-type construction; b Cydo gear mechanism (Cyetu Getriebebau Lorenz Rraren GmbH. Markt Indersdorf); explanations in text.
Manufacturing Systems. 1 Machine Tool Components thus j = ZFI/(ZFI- z<;;). Transmission ratios of i = 50 to 320 and output torques of M = 1·5 to 4000 N m may be achieved. The gear mechanism possesses extreme torsional rigidity and also freedom from backlash, owing to the large meshing area which amounts to 15% of the total number of gear teeth. In the cyclo gear mechanism (Fig. ~9b), a cam plate 1 is driven via an eccentric cam 2 (drive shaft 6) and revolves around a rigidly mounted ring. Each point on the plate thereby describes a cycloid curve. Rotary motion is generated at a far lower speed in the opposite direction on the plate and is dependent on the ring-plate diameter ratio.
In order to prevent any sliding when rolling, the plate in the cyclo gear mechanism is fitted with an enclosed cycloidal train, thereby providing an external form, and the ring is replaced by bolts arranged in a circle. Each cam plate thereby has one cam curve segment fewer than there are bolts in the ring of bolts. The curve tracks on the cam plate engage positively in the rollers 5 of the rigid outer ring and move on rolling contact with this. The reduced rotary motion of the cam plate is transmitted to the output shaft 3 via the bolts 4 which engage in drill holes in the same plate. The transmission ratio is determined by the number of cam curve segments in the cam plate. Rollers are mounted onto the bolts of the ring of bolts 5 and output shaft 3 which generate power transmission via their pure rolling action between the cam plate and ring of bolts 5 and cam plate and retaining bolts 4 of the output shaft. Friction losses, noise generation and wear are thereby reduced to a minimum. The gear mechanism is fitted with two cam plates offset by 1800 driven via a double eccentric cam, thereby providing a mechanical balance. These high-transmission compact gear mechanisms are employed as intermediate gearing in feed drives or for
Where this value is exceeded, the flow of force is interrupted to ensure the reliable protection of the endangered parts from any damage. Slip couplings are constructed as spring-loaded conical friction clutches and dog clutches. They are frequently integrated into the spindle-side toothed-belt pulley in feedscrew-and-nut drives within an established synchronous belt drive. The coupling is mounted on the shaft journal of the ballscrew and connected to this via a conical clamp 12 which is frictionally engaged with this and free from backlash (Fig. 40). In normal operation the torque transmission in the dog clutch is carried from the toothed belt plate 1 and the flanged ring bolted to it 2 via the ball bearings 3 to the driving collar 4. The ball bearings are thereby fed to the narrow-tolerance penetration drill holes in the driving collar and pressed into the conical collars of the flanged ring head-on by the spring washer 5 and index ring 6. The spring washers 5 are preloaded with the aid of the setting screw. The torque to be transmitted may thereby be adapted to suit the respective operating conditions. The driving collar twists towards the flanged ring where there is excess loading, whereby the ball bearings are pressed out of the conical collars of the flanged ring against the force of the spring washer. The four-point contact bearing 8 takes on the bearing function between the moving belt pulley and rigid driving collar. The flow of force is interrupted as a result. The clutch continues to slip until the torque falls back below the prescribed critical limit value. It then re-engages automatically. The excess loading is detected immediately via the proxintity switch 9 so that the drive may be switched off.
Collision Force Calculations and Estimates. The arrangement of the slip coupling in the flow of force depends on the one hand on the position of the parts to
driving rotating tables, tool magaZines and tool changers.
Compact gearing mechanisms also constitute one of the main components of articulated joint drives in robot applications. The criteria of high transmission within the smallest available construction, concentricity, high dynamic response, low backlash, high torsional rigidity and high excess load-bearing capacity provide highly dynamic drives with extremely low backlash and high positioning accuracy and repeatability.
7
2 a
Couplings Special-purpose flexible couplings may be employed to connect two shaft ends, in particular those of the motor shaft and ballscrew in feed drives, which still have a high level of rigidity in a perpendicular direction. The rotary motion in the perpendicular direction is thereby transmitted with great accuracy. Radial and axial displacement of the shaft ends and angular displacement are tolerated to a limited degree in this coupling (see F3). Friction-locked clutches (e.g. bellows and diaphragm couplings) are normally employed in high-precision feed drives. They satisfy most aptly the high requirements of torsional rigidity, freedom from backlash and low mass moment of interia. Their structural design is determined by the torque to be transmitted, the shaft diameter and the torsional rigidity. Slip couplings are employed as effective protection for numerically controlled machine tools against damage from excess loads and impacts following tool breakages and programming and operating errors, where the effective torque in a drive train is limited to a maximum limit value.
It Figure 40. Integrated slip coupling for feed drives with synchronous belt drive Oakob, Kleinwallstadt). a General arrangement: 1
spindle bearing, 2 ballscrew, 3 synchronous belt drive, 4 slip coupling,5 servomotor, 6 machine tool table. b Slip coupling: 10 spindle bearing, 11 spindle (further details in text).
1.3 Frames. 1.3.1 Requirements, Types
be protected and on the other on the position of the machine components that cause the high collision forces. These forces are brought about primarily by two mechanisms. First, mass forces are released in a case of collision with the sudden delay of impact on a machine axis; they are determined by the kinetic energy of the moving machine parts, e.g. slide, workpiece, spindle, motor, etc.
Second, in a case of collision, motor torque increases rapitlly by three to ten times the rated torque depending on the type of motor. The mass forces and peak torque of a motor are added together to provide a total force on impact which may cause elastic deformation, and in the worst case even permanent deformation or breakdown in the machine components located in the flow of force. An estimate of the total force on impact may be derived on the basis of a simulated model.
a
1.3 Frames 1.~.1
Requirements, Types
Frames and frame components are load-bearing and supporting foundations of machine tools [361. They carry and guide the parts required for relative motion between the workpiece and tool, e.g. supports, gears, motors and control systems. The basic shape and dimensions of these parts are determined by the working space, the level of process forces and the required degree of accuracy (rigidity). The necessary access to the machine for maintenance, service and assembly purposes must also be avoided. Because of production and assembly considerations the frames themselves are in many cases manufactured from a number of individual parts which are then bolted together at the joints and even cemented together in certain cases. Frames consist of bed-type, column-type, tabletype, knee-type and crossbar designs. (See L4.2.2 and LS for examples of machine tool frames.) The structure of the bed of the machine and pOSition of the operating spindle are important structural considerations for turning machines (Fig. 41).
which are vertical in relation to one another, and the positioning of the cross-table on the wide guideways of the bed produces high static and dynamiC rigidity. "Cross-bed" construction is where two vertical directions of feed are provided on the bed, with one assigned to the tool-carrying group (normally column-type frames) and the other to the workpiece-carrying group (normally table-types). The H{rame construction (also known as the double column frame with crossbar) provides a particularly stable
Flat-bed construction 1 is used mainly in large-scale
and suitable construction for higher cutting power with
turning machines (rolling and turning machines). The angled-bed construction 2 permits the hot swarf and the coolant lubrication to be discharged or removed from the working space, so reducing any danger of clogging or thermal overloading of the machine bed. Front-bed turning machines 3 are particularly suitable for chucking part processing with automatic tool change. Vertical turning machines based on the column constnlction 4 have the advantage of being particularly suitable for accommodating and processing large parts without any bending stress being sustained hy the spindle. The spindle (workpiece) may be positioned both parallel 5 and vertical 6 to the base. Figure 42 shows the major horizontal and vertical drilling and milling machine constructions, structured according to their frame design (knee-type, bed-type and H-frame type) and the number of axes in the tool carrier or workpiece carrier. The knee-type design is employed only in small-scale machine tools owing to the mass to be moved in a perpendicular direction. Bed-type milling machines are employed for processing heavier workpieces. In contrast to the knee-type, in this design the table is mounted on a rigid machine bed. A distinction is made in bed-type designs between cross-table and crossbed constructions. In cross-table constructions the workpiece carrier, i.e. the table, has two directions of motion
bulky workpiece items. Of the two designs available, the longitudinal table-type design (Fig. 42b) is fitted with a table that may be moved in one direction only and the bed is twice as long as the table. All coordinated motion vertical to the feed movements of the bench is carried out by the machine tool. In contrast, the gantry design employs a permanently fixed clamping plate and a moving gantry, with the advantage that the machine has only to be as long as the longest workpiece to be machined and/or the clamping plate, whereas machines with moving tables require twice the length. Both the constructions just described have to be capable of high accuracy in fmal machining and possess a high cutting capacity for premachining requirements. A distinction is also made between crank presses and eccentric presses in terms of the frame construction (see L3 and L4). There are also open, discharging C{rames and closed O-frames in twocolumn designs. The C-frames have the drawback that they open up as a result of the forming force (Fig. 4~a), which may result in misalignment in the two halves of the tool, but the advantage that access is provided to the working space from three sides on the machine. Closed-frame constructions (Fig. 4~b) are employed primarily for medium-scale and larger constructions and where the machine tool requires a particularly rigid and accurate
b Figure 41. Classification oftuming machines based on their frame constructions. a Bed-type designs. b Relative pOSition between tool and workpiece (explanations in text).
I:!IfII
Manufacturing Systems. 1 Machine Tool Components
Types Bed type
Knee type
r----; ~r
I -!.- ~ :!l
0
Y'
lQ,
Qz' + X'
Cross-table type
Knee-type column construction .~
~
~
'"
~ ~
'0
.2l §
Et· X'
Cross-table type
Y~ L H J I I ~~r
~.
X'
c
Knee-type bed construction
z
I
II
Cross-bed type construction
~~f1 Y~~I
Traveliing-column construction
a
.,
<'5
machine construction
JJ.~ Y'
Cross-table type
Knee-type column construction .~ ~
~
.,'" ~
.. 0
!;j l'0
.2l §
1!l
f=.
b
~m ~m
Machine ta e construction
z
Travening-coiumn construction
guideway to contain the forces generated during the fOrming process. 1.3.2 Materials for Frames
Steel, cast steel and cast iron may all be employed as materials for frames and frame parts. More recently. polymer concrete has been proving increasingly popular in
Gantry construction
Figure oil_ Designs of drilling and milling machines: a horizontal, b vertical design.
smaller machine frames. Table 1 shows the most important physical characteristics of the above materials for frames.
Advantages of Steel Construction. The coefficient of elasticity is about twice that of cast iron, resulting in material savings and lower weights. There are no modelling costs, making steel especially suitable for products made to
1.3 Frames. 1.3.3 Embodiment Design of Structural Components (Frames)
5
5 a'-_ _ _ _ _
~
6
Figure 43. a Single column and b double column frames: 1 flow of force and 2 displacement chardcteristics, .3 workpiece, 4 tool, Fax axial components of machining forcc.
J order. Steel constructions may be of either plate type or cell type. The plane method relates to cast constructions: plates or moulded units made of thick sheet metal are welded together with ribs to frames. This method is frequently found in presses. cutters and similar machines where cast iron is not rigid enough. Frames made on the cell-principle consist of a wide range of individual cells constructed from thin welded sheets, which gives a considerable saving in weight while maintaining a high level of rigidity. The heat-generating capacity is correspondingly smaller, owing to the minimum of material employed, which results in a greater danger of thermoelastic deformation (see Ll.1.2).
8 Figure 44. Polymer concrete machine frame: 1 polymer concrete machine columns, 2 polymer concrete machine bed, 3 polyurethane foam core (suds tank), 4 bolted trim (cemented in subsequently), 5 gUideways (bolted on), 6 threaded bush, 7 bearing flange (cemented in subsequently), 8 conduit for retaining core (suds).
and hence rapid production of a second cast following constructional modifications. Parts to be positioned accurately (e.g. guideways) are subsequently cemented in place in premachined grooves or furrows using mortar [37J. An example is shown in Fig. 44.
Advantages of Cast Iron. These are its high damping capacity, good sliding characteristics in guideways, good machining properties and good dimensional stability. These are further enhanced by employing special formulas to achieve good casting properties and different wall thicknesses and a high degree of stability (Rm = 400 N/mm' and over), using nodular cast iron with a high coefficient of elasticity.
1.~.~
Embodiment Design of Structural Components (Frames)
The embodiment design of structural components (frdmes) is determined on the basis of the requirement for static and dynamic rigidity and the lowest possible material consumption. These are, therefore, constructed to be rigid and lightweight, by increasing the moment of inertia with the appropriate design of the cross-section. Open cross-sections and penetrations should be avoided hecause these reduce rigidity conSiderably. In addition, the construction should be as sturdy as possible, as the width between supports and projection will have a substantial effect on all bending stress. The bending and torsional rigidity of frame parts may be increased with the appropriate ribbing. Figure 4S shows the ribbing types most frequently employed in column-type parts. In cases A to D, longitudinal ribbing is provided. The columns shown under E to H are fitted with cross-rails. The relative bending and torsional rigidity of the different column ribbing shown in Fig. 46 are calculated using the fmite-element method (see BB).
Advantages of Polymer Concrete. This material is frequently employed in the manufacture of small and medium-sized « 5 m) frame components, specifically machine heds. It possesses an even higher damping capacity than cast iron, and therefore higher dynamic stability. It has a lower heat conductance and higher heat-generating capacity
than other materials, and so is insensitive to short-term fluctuations in temperature. luere is a wide variety of applications. Embedded parts, e.g. clamping surfaces for bolting on cover plates, motors, spindle boxes, etc. may be positioned in such a way that no subsequent reworking is required. Conduits, cahle and ducting channels for the power supply may be incorporated directly into the casting. Modular systems provide a simple rdnge of options, while simple and rapid modification of fomls is possible
Table 1. Physical characteristics for frame materials in machine tools [37} Material
Steel Nodular cast iron Cast iron Polymer concrete
Coefficient of elasticity
Specific weight
Heat expansion coefficient
E (N/mm')
')' (N/dm')
" (I/K)
78.5 74.0 72.0 23.0
11.1 . 10"" 9.5.10 6 9.0·10·(' 10 to 20.10-6
2.1·10' 1.7·10';
0.5 to 1.1·10" 0.4·10'
SpeCific heatgenerating capacity C Q/(g K»
Heat conductance
Rigidity area
A (W/(m K»
II
0.45
14 to 52 29 54 0.9 to 1.1
400 to 1300 400 to 700 100 to 300 IOta 15
0.63 0.54 0.9 to 1.1
(N/mm')
Manufacturing Systems. 1 Machine Tool Components
vide considerable stiffening in the walls as protection against local denting and warping and thereby contribute towards inhibiting local deformation at the points of application of force.
OR
B
A
D
[
Example. Figure 47 shows a right-angled cross-section of a column-type frame, where the walls are ribbed in accordance with the cell-type design. As the interior of the column remains free to take a counterweight, discontinuous cross-rails are employed to increase the torsional rigidity. longitudinal ribs increase the bending rigidity. The guideways are not positioned in the vicinity of the side walls,
J
F
E
H
6
Figure 45. Types of ribbing in columns.
The longitudinal ribs improve the bending rigiclity of the part by increasing the axial geometric moment of inertia. Vertical ribs running parallel to the outer walls do not provide any notable improvement in torsional rigidity. The torsional stress occurring in frame components normally results from a couple acting on the guideways, which can cause extreme cross-sectional distortion. Diagonal ribs in the longitudinal rib design are particularly suitable for preventing this cross-sectional distortion. Horizontal ribs (cross-rails and end-plates) also effectively help to prevent cross-sectional distortion with torsionalloading via a couple. Horizontal ribs have virtually no effect on bending rigiclity. Nevertheless, they may pro-
FI2--
o
s~
4mm
5
D B B EB EEl
Figure 47. Frame of a horizontal drilling and milling macWne with ribbing: 1 column, 2 bed slide, 3 end plate or hood, 4 gUideways for spindle box.
[0
ill
158
FI2--E;tl1ll
2
100'1.
With coup Ie
100'1.
Without Couple
100'1.
Without couple
~94'/.
120'1.
101'1.
111%
~12.1'!.
120'1.
101'1.
112'1.
[Z]
118'1.
[2J
118%
IV1
With coup Ie
_182'1. 115'1. 104%
14 2'/.
~
~
119'1.
14 2'1.
o
B o
§
50
100·/.
Model
100'1.
Model
100
Relative bending rigidity
117%
104'1.
J 150
50
I
100
Model
109'1.
Relative torsional rigidity
Model
I
150
Figure 46. Relating bending and torsional rigidity in various types of ribbing (finite element method calculations).
1. 4 linear and Rotary Guides and Bearings
but are supported by the cross-rails. Circular cross-sections are particularly torsionally stiff and do not cause distortion under load conditions. The manufacture of these forms is complex, however.
Examplea Figure 48 shows the cast-iron construction of the travelling column of a roller milling machine. The transverse ribbing and the diagonal longitudinal ribs provide sufficient bending and torsional rigidity for the column. The connection of the diagonal longitudinal ribs to the front wall in the vicinity of the guideways provides an even load distribution from the guideways over the
entire column and prevents excessive local deformation at the points of application of force. 1.~.4
Calc:ulation and Optimisation
Computer technology is extremely useful for making effective predictions about the characteristics of a machine tool in the construction and development phase. A high-performance user software package is a basic requirement for this type of computer application. The software nonnally employed for calculating the mechanical characteristics of frame components, such as static, dynamic, thermal and stress analysis, consists of programs which are based on the finite-element method (FEM) (see BS). Static and dynamic displacement and/or compliance and rigidity, natural frequencies and temperature distri-
butions may be calculated for given heat sources and thermo-elastic deformation. The execution of a structural analysis with FEM involves the data acquisition phase (preprocessing), calculation and evaluation Of results (postprocessing). Optimisation strategies based on the FEM method are becoming increasingly popular in structural analysis. Such systems programs serve to optimise the weight and rigidity of mechanical structures and to minimise peak loads at the margins of the curve. They are capable of automatically varying the geometric component parameters, such as wall thickness within specific limits, so that the optimisation target (optimum) is achieved. A complete FEM analysis must be performed at each iteration stage. One of the chief targets with cutting machine tools is to minimise deformation occurring at the cutting point during machining. This means optimising frame components in cutting machine tools with respect to maximum rigidity for a given total weight. Example. Figure 49 shows the travelling columns of a drilling and milling machine. The target of this optimisation was to minimise
deformation at the structural point P in the vicinity of the rigbthand guideway. The loads shown affect the columns depending on the operating forces, causing Slight bending and extreme torsion in the column. The external and ribbed wall thicknesses of the column are determined as the optimisation parameters. Since the columns
Al.
+ +
+
s1.
+
I
+
I
t
I ~
t t t
are designed as a welded structure, the optimisation parameters may only assume eight discrete wall thicknesses of between 8 and 40 mm with a number of restrictions. The wall thickness distribution is shown in Fi&. 4:~ before and after the optimisation calcu-
.iA
lation, with a given material consumption (weight) in the form of a bar chart. The effect of the redistribution of the volume of material on the deformation of the guideways in the column is shown in
Fig. 4911. Where a double optimisation algorithm was employed (University of Lo.ttich, Fleury, Braibant), reductions in deformations
of up to 17% were established compared with the original structure.
.is
In the case of metal-forming machine tools, e.g. presses, where the forces within the component are vital, as is a sufficient degree of rigidity, local excess loads which tend to occur as a result of fatigue in discontinuous cross-sectional junctions (drill holes) should be noted specifically. These may result in the breakdown of an entire machine in a substantial number of cases. Example.
FilJllft SO shows the optimisation of the curve of a
C-frame press to minimise peak loads occurring at the margins of
the curve. It is clear that the peak: loads at 235 0 may be reduced
by around 30%.
1.4 Linear and Rotary Guides and aearings
Section A-A
Linear and rotary guides and bearings in machine tools have the function of providing a precise, linear trajectory for the execution of the cutting and feed motion of specific components, such as slides, spindle boxes, rams of presses, spindle sleeves, etc. The weights of the guided components and workpieces also have to be carried and process forces have to be absorbed with the minimum of deformation [3B]. Important requirements for the linear and rotary guides and bearings of machine tools are high machining accuracy and high performance in the long term, with low manufacturing and operating costs [39]. In order to satisfy these requirements, the linear and rotary guides and bearings must have the following characteristics:
Section B-B
- Low friction and freedom from stick/ slip as a requirement for accurate positioning at low feed power. - Low wear and protection against seizure, so that longterm accuracy may be maintained.
Section C-C
Figure 48. Machine columns with diagonal ribbing and cross-rails.
Manufacturing Systems _ I Machine Tool Components
o Inilial soIutiofl
•
x
a
25 0
30 25
0
c:
..\..-\
30
30
20
X2
i
Xl
8
X5
X6
I
Xl
iO' Hr--;-L:-; eft:-;: .h7:a/ld ::;----,-r.;;;;--,. ,--;;;::;:;-;=----,.~Ml
~
guideway
O.S·Hf-':"'-'I-'-+--/-T----i
b
II
mmll Xl
X5
XI
40
25
20
0
f-
'-..I--
30 25
\
M
40
Solution aftel third ileralion
Ol-H
Xl
X6
Xl
- High rigidity and low clearance of linear and rotaty guides and bearings and/or freedom from backlash to minimise changes in position of guided components. - Good damping in load-bearing direction and direction of motion to prevent excessive oscillation in feed drives and tendency of machine tools to chatter. Other criteria also affect the machining accuracy and the performance of the machine tools and must therefore be taken into consideration; these include power loss and thermal characteristics depending on heat conductance, protection against swarf, contaminant or coolant penetration, and obstruction in the linear and rotaty gUides and bearings. Manufacturing and operating costs are determined primarily by the choice of the guide and bearing prinCiple. Figure 51 shows linear and rotaty guides and bearings classified according to the physical principle on which they are based and/or the type of lubrication and lubrication system together with the friction characteristics. The manufacturing costs (Fig. 52) may be reduced by rationalising the manufacturing process, using preprocessed and/or standard components and selecting suitable materials, such as plastic side linings which may be preformed. Another important consideration is the ease of
d
Percentage delormalion
Figure 49. Minimisation of defo rmation in machine tool columns
due to variations in wall thickness with a constant total weight. Optimisation: minimum defonnation at structural point P. Restrictions: same weight, use of sheet thicknesses of 8, 10, 12, 15, 20, 25, 30 and 40 mm. a Schematic diagram, b deformed structures, c optimisation parameters Xl to X7, d deformation of gUideways (in space).
assembly of a complete linear and rotary guide and bearing system [39). Operating safety and fault tolerance with the capacity to withstand excess loads also affect the operating costs of linear and rotaty guides and bearings. The maintenance requirement and resistance to contamination of the various basic guide and bearing principles and methods are also criteria which affect operating costs and must therefore be taken into consideration in their selection. 1.4.1 Linear Guides
nat Slideways. These are the most common design employed in machine tools irrespective of the design principle. They are designed to transmit high weight, mass and cutting forces mainly vertical to the slideway (Fig. 53). Holding strips are provided to prevent the slide from lifting or skewing. Backlash elimination is provided for horizontal slideways by adjustable gib-strips. Aspect ratios vaty between 1 : 40 and 1: 100. See Fig. 53e, f for adjust· ment controls. Dovetail Slideways (Fig. 53b). These prevent the slide from lifting by setting the side surface area at an angle of 55°. They may be adjusted via gib-strips arranged
1.4 Linear and Rotary Guides and Bearings. 1.4.1 Linear Guides
3
Frictional force
I
.~ Sliding speed
a
b
G
600 400
c
3
E
~k-==
hl I
ZOO §.
1-
00 .~
>-
-ZOO
d
Figure 51. Principles of guides and friction characteristics. a Hydrodynamic slide: 1 bed, 2 slide, 3 oil sump. b Hydrostatic slide. c
'!
Aerostatic slide. dRoller slideway.
-400 1--j,L,4+W:,LX~----I- .--1-----1 -600 '---_---L_ _--'-_---"_ _--'-__----' -400 -ZOO 400 -600 ZOO Xaxis (mm)
100
~ ~ I I
80
I
I,
:
2
I
-i--11--li+l-f-----J
1
.
I
60
n
Milling. forming Guideway length and/or table size
40~-----+-~~--~----~----~--
~
~
ZOII----+-H---+------I---~~-1---___I
O'----~----'--~--~--
50'
100'
150'
Angle
ZOO'
Z50'
300'
~
d Figure 50. Optimisation of curve in C·frame press. a Block diagram. b Finite element network (line loads each 800 kN respectively). c Curves: I prior to optimisation, 2 following optimisation, 3 within the admissible V'Miation r.mge. d Stress characteristics at the margins of the curve: I prior to optimisation, 2 following optimisation, where the boundaries along the angle '" are to be plotted as evoluted (anticlockwise).
Figure S2:. Trends of manufacturing costs of various guide principles. I Hydrostatics, one pump per pad, cast iron/cast iron. milling/milling. 2 Slideway cast iron/cast iron, grinding/grinding. 3 Roller slideway, fittings for recirculating rollers/tempered steel holding strips.
at an angle. Advantages compared with the flat guides are low height and good damping qualities. Designs may also be employed with a slope on one side and flat guides on the other (Fig. S;k). Use dovetail guides in short planing, slotting and shaping and small milling machines, otherwise only on slides for auxiliary and feed motion. They are mainly employed as stideways.
Vee and Flat SUdeways (Fig. S3d) (V- and flat shape). These take up force in two directions. A flat fonn is used as a stideway for the main support in small and
Manufacturing Systems. 1 Machine Tool Components
40
30
S~. Types of linear and rotary guides and bearings. a Flat slideway with holding strip and adjustable gib-strip. b Dovetail slideway with gib-strip. c Flat sLideway and dovetail combination. d Vee
Figure
and flat (cambering) slidew3Y combination with adjustable holding
strip. Gib-strips may be adjusted: e externally, finternally via hexagonal socket screw.
Material mixes ~ C
I!I!lI Total medium-sized turning machines, and also in combination with a flat stideway. Protection against lifting is via holding strips, which may be adjusted in pitch for clearance.
Figure S4. Material mixes employed in slideways and combined slideways/roller slideways.
CyHndric:al Ways. These are installed as directional guides (e.g. drilling spindle sleeves) or slideways with shaft sleeves (Spieth sleeves) with adjustable clearance, or as roller slideways (see F4). They have the advantages of being simple to manufacture and possessing high guid" accur-Ky, but are complex in assembly (shaft-centre distance) and are only suitable for limited lateral loads. Hydrodynamic Lubrication Slides. These are most commonly employed in the machine tool construction sector. The grounds for this are the high damping factor and a high degree of accuracy and rigidiry in conjunction with relatively low construction and manufacturing costs (38). The relatively high frictional forces may adversely affect the feed drives. Material Mix. Figure S4 shows the material mixes employed in slideways and combined roller slideways/slideways r40). In this case , 30% cast-iron-cast-iron material mixes and 28% cast-iron-plastic material mixes are employed respectively, whereas the remaining mixes are far less widely employed. Cast-iron and epoxy resin and Teflon-based plastics (PTFE) are normally used for the moving part of the guideway (slide). The fixed part (bed) is normally of cast iron, or in a few cases steel (Ck 45, 16MnCr5 or 90MnV8). Manufacturing and Processing. The manufacture of plastic-coated guideways is carried out by bonding on plastic sheeting or using moulding technology. In the moulding process, the plastic compound is applied to the roughly prepared sliding surface and then pressed in before it hardens with the opposing component which has been prefinished and sprayed with a parting compound (application technique). In order to achieve the correct shape for the gUideway and an even layer of compound, pOSitioning and spacing strips between the two parts are adjusted before embedding. The excess compound is squeezed out of the gaps by the force of the weight and additional loading. In the injection moulding technique, the lining is formed by pressing the plastic compound into the space between preset and adjustable components (Fig. SS). Good adhesion is achieved between the plastic
Detail
Depth Mining wi1Il single cutting edge
Fipre SSe Injection moulding techniques for plastic-coated slideways (SKC-Gleitbelagtechnik): I slide, 2 bed, 3 plastic sliding surface, 4 press--fit drill holes, 5 bolted guide rail and spline.
and slide by planing with a diamond-point tool or milling with a single cutting edge. The majoriry (some 60%) of the slideways which are primed or sprayed with compound are abraded following hardening to form the oil pads. A smaller proportion (some 25%) are put to use without any further processing [40]. Four fmishing processes, i.e. abrading, circutnferential grinding, end milling and fme milling, are undertaken in the most popular guideway material, cast iron,
1.4 Linear and Rotary Guides and Bearings _ L 4.1 Linear Guides
whereas steel is nonnally processed only with circumferential grinding and end milling. Load-bearing guideways should be tempered to counteract seizure or wear. Cast·iron tempering is performed with flame hardening or induction hardening or casting permanent mould tempering (Brinell hardness, HB = 4.5 to 6 kN/mm2). Surface tempering steel guides (Rockwell Hardness C, HRC = 58 to 63) are available in cylinders, block strips, plate or spring band steel.
Tribological Cbaracteristics. In the tribological survey (sec 06) of friction and wear, the collective load must be taken into account in all cases [41 J. The collective load includes the type of motion (sliding, rolling, etc.), the time-related sequence of motion (continuous, oscillating, etc.), and the loading parameters (normal force FN , speed v, temperature and length of load tB ). The characteristics of the parent and offspring substances with their materials and surface structures and the type, viscosity and volume of the precursor are particularly important. The frictional characteristics of the different guide principles and of slideways with their various materials and surface structures are shown in Fig. S6 [39J. The hydrostatic slide has the lowest friction coefficient. The friction coefficients in hydrodynamic slides arc significantly higher than those in hydrostatic slides and roller
Ao50,Z50mm 2 p o 50N/cm 2
Surlace section
,.-_..L\..:--=----"_.L-_y Under-surlace
section
Lubricating oil = 170 mPa; v = 3mm3 Lubrication frequency Lubrication distance
0
0
15 min 60 km
slideways. The surface structures have a substantial effect on the frictional characteristic curve (Stribeck curve) in this type of guide. The use of the machining characteristic for circumferential grinding on the fixed under-surface section (bed) and the moving surface section (slide) reveals a sharp fall in the friction coefficient at increasing speeds (characteristic curve 1). This encourages the undesirable stick/slip characteristic (slideback) at low feed rates. A part of the slideway. preferably the slide, should have tool marks crossing the slideway to alleviate this steep fall [42 J. This may also be achieved with end grinding or preferably end milling (characteristic curves 2, 3 and 4). In this case, the overall level of the friction coefficient in the bottom section of the sliding speed range is significantly lower, and the stick/slip tendency is alleviated as a result. A favourable frictional characteristic curve, and an appropriately lower stick/slip tendency, is achieved with resin and PTFE (Teflon) inftlls using bronze (characteristic curves 5 and 6). Teflon also pennits dry operation, but has low compressive strength (edge tear resistance) . Table 2 shows the test results reiating to the wear characteristics of different stideways [43 J. The wear of lubricated, untempered cast-iron slideways is around 1 to 3 j.Lm per sliding couple for a 60-km slide path with a load of 50 N/cm2 , which corresponds to an operating life of around five years in single-shift operation. Tempering metal guides do not cause any substantial reduction in wear in lubricated sliding load operation. The moulded plastics employed nowadays adversely affect the gap height change (i.e. the gap becomes smaller) by around 3 j.Lm because of their swelling properties. Because coolant emulsions may also inadvertently find their way onto the guideway during the manufacturing process in addition to the slideway oil required, higher swelling values may be anticipated in general in plastics. Extremely soft guide materials, such as pure PTFE, suffer excessively high wear under the normal loads in machine tool construction of 50 N/cm2 Lower wear values may be achieved, while favourable frictional characteristics are retained, by employing suitable additives, e.g. pulverised bronze. Lubrication of hydrodynamic slideways is an important aspect with respect to wear. Most machine tools (up to 80%) are fitted with pulsed lubrication systems. Continuous dropper oil lubrication and manual lubrication are Table 2. Unear wear rate in
~m
for slide path of 60 Ian
6
7 8 O~·==·d~~·-£·~·~~~~~ 1
m
Surtace section/processing 1Cast iron 25/ circumferential grinding 2Cast iron 25/end milling HM hard metal 3Cast iron 25/end grinding 4Cast iron 25/end milling with cutling ceramics 5Moulded reSin/moulding 6PTFE with bronze/ clrcumierential grinding
m2
Sliding speed v (mmlmin) Under-surtace section/processing Cast iron 25/circumferential grinding Cast iron 25/clrcumferential milling Cast Iron 25/circumferential grinding Cast iron 24/clrcumferential grinding Cast iron 25/circumferential grinding Cast iron 25/end milling HM hard metal
Operating
processa Surface
section/ undersurface
Material mix in slideways (surface section/undersurface section)
Cast iron/
Cast
Resin"/
cast iron
iron/
cast iron
cast iron<-
cast ironb
section
2.7/1.1 (2.0/0.6)' 1.8/1.7 2.3/1.7 2.5/0.6
1.5/0.6 -2.8/0.8 (-1.2/0.6)' (-1.7/0.5)'
a
PTFE compound/
3.5/0.3
Test setup and machining process in accordance with Fig. Sh. Tempered. C With bronze. d With slide path of 20 km. to Filled.
Figure S6. Frictional characteristics of different guides. 1-6 slide-
b
ways, 7 roller slideways, 8 Hydrostatic slides
, With slide path of 5 km.
Manufacturing Systems. 1 Machine Tool Components
rigidity and excess load capacity (Fig. ;9, curve 1). At low power loss, as in the second case, the manufacturing costs are lower at half the operating rigidity (Fig. ;9, curve 2).
Advantages of Hydrostatic Slides. These are: wear-free operation, hence no startup friction and only slight friction without backslip (stick/slip effect) in the range of feed rates; very good damping characteristics due to an even oil film across the guideway; high rigidity in smaU footprints, achievable within generous limits.
SHdes with Aerostatic Lubrication. Gas-lubricated
Figure S7. Lubrication groove fonns. a Forms. b Lubrication groove cross-sections; cross-section 2 occurs more frequently than I.
employed only in a very few cases. Slideway oils with viscosities of n 50 = 30 x 10-3 to 80 x 10-3 N s/m2 are used in lubrication. Figure ;7 shows lubrication groove fonns and cross-sections of slideways. To prevent dirt from penetrating them, slideways have wipers (nonnally plastic constructions) mounted on them.
Hydrostatic Lubrication SHdes. The slides of the machine components fed in this guideway construction are separated from one another by an oil film which is under pressure and is maintained via an external oil supply system [38]. The compressed oil is fed over the supply drill holes onto hydrostatic pads and flows away into the parallel gap between the sliding surfaces, thereby losing pressure. The oil supply is provided via either a separate pump for each pad (Fig. ;Sa) or a common pump at constant pressure with hydraulic cutoff to the pads via preliminary regulators, and normaUy takes the fonn of capillary tubes (Fig. ;Sb). The first case provides higher
bearings operate on the same functional principle as those with fluid lubrication. The differences between them mainly concern the characteristics of their lubricants. The advantages include very low friction, low heat generation, very good repeatability and low construction costs due to the omission of seals and a lubrication return system. The disadvantages are larger construction sizes, lower damping, inadequate emergency running properties and increased costs involved in manufacture and air treatment. Self-excited pneumatic instabilities ("air hammer") may occur owing to the compressibility of the lubricant, but may be corrected via structural measures, and by restricting the supply pressure to 4 to 10 x 105 Pa. The extremely narrow bearing gap of around 10 fJ.m requires very high machining accuracy and low static, dynamic and particularly thermal-related displacement. The calculations for aerostatic bearings, based on the assumption of viscous gap flows, are carried out with the aid of the NavierStokes equations. Figure 60 shows an aerostatic slide bearing with a rotational table. In order to reduce the manufacturing costs, the slide is preloaded with spring-loaded support roUers which have an extremely low level of rigidity compared with the aerostatic bearings.
RoUer SUdeways. linear guides in roller slideways are used for a wide range of applications, as are standard slideways. They have the advantages over slideways of smooth operation due to roller friction, low startup resistance, absence of stick/slip and freedom from maintenance. Their disadvantages compared with hydrostatic and hydrodynamic guides is their low damping normal to the direction of motion [38]. Combined roUer-standard slideways are frequently employed for this reason.
lOO
}
0.75 2j
0.50
Y
0.25 a
L!J
L!.J
~
"
-0.25 -0.50
I
f-- ~
./
---
b Fi......, SS. Oil supply to hydrostatic compressed oil pads lover multiple pumps a, common pumps b and capiIlaty regulators 2.
{ -4
-3
-2
-1
.-
I--
/
t
-0.75 -tOO -5
~
Y "......
~
2 0 F/Fo
Fip.re 5'. Displacement-load characteristic curves for hydrostatic slides as shown in Fig. S: bo gap height in initial condition, Fo internal preload force of holding strip. Technical specifications: surface area ratio q; = 0.6, initial gap height ratio A = 1.0, regulator
ratios
~,
= 1.67 and
~,
= 0.6.
1.4 Linear and Rotary Guides and Bearings. 1.4.2 Rotary Guides, Bearings
Figure 60. Aerostatic slide 1 in table base with rotating table 2 (Wotan, Diisseldorf), 3 hed, 4 cemented in tempered steel plates,
5 cemented in steel strip, 6 cemented in plastic plates, 7 springloaded support rollers, 8 air supply, ') inlet stream aperture with jets as regulator.
Figure 63. Slide conveyor of a milling machine: 1, 2 fittings for recirculating rollers, 3, 4 gUideway, 5 damping strip (INA Lineartechnik).
Table 3. Comparison of types of bearings w
Speed limit
ti II II ~e·rlelj ~o
~ ~
Servioe lile
~
~
~
~
~g>
~.8
Figure 61. Recirculating ball bearing assembly.
Roller slideway components are available in a variety of sizes and precision classes. Ball bearings, cylindrical rollers and needles may be used as rollers (see F4). Rotating components containing a return for the rollers with one, tWo or four rows which are especially suited for long traverse paths may be incorporated to achieve greater rigidiry where preloading is employed as well (Fig. 61). The ball-bearing-shaped rollers with two or four-point contact (pointed arc) may also be fitted between the fitting for the recirculating rollers and the guide rail (Fig. 62). Example. Figure 63 shows four roller slideway components (fIttings for recirculating rollers) positioned on top of one another in the slide guide which take up the main load of the horizontal slide. The damping strips are located adjacent to these components The remaining oil in the capillary gap between the strip and the guideway acts as an oscillation damper.
1.4.2 Rotary Guides, Bearings The suitability of various types of bearings for standard application criteria in machine tool construction is shown in Table 3 [51 J. Table 4 provides a summary of the resultant areas of application of the various bearing systems.
a
b
Figure 62:. Ball bearing guides: a with 2-point and b with 4-point contact of ball bearings (Deutsche Star, Schweinfurt).
Freedom Irom vibrllUon
~
~
Damping
0
~ ~
Rigidity
I~
LuIJr~lion
(cost)
Price lacquisitioo, maintenance) 0
() I ~
e () ~ e o IX e e 0 0
() Oc () ~ ()
Dependenl on lubricatin g system and type 01 roller beari ng in
ope
~ Unlim~ed
e
b
~ O< ~
Power loss
C
e ble eO e e e
lau~-free
Exl/emely high
~ High
() Average
0
• Low
Main Bearings. These serve to guide and take up the force of the rotating components which generate part of the cutting or shaping motion. The highest requirement is freedom from vibration for spindle bearings employed as boring, milling, turning and grinding spindles. This is why the dimensions of the various components in the spindle bearing system, i.e. the bearing, the spindle, the casing, etc., are given such narrow tolerances [53 J. In addition to the rotational speed limit, the required speed range and characteristics also affect the selection of the bearing. Where the roller bearing is employed, the lubrication principle to be employed (i.e. minimum oil and grease volumes or oil coolant lubrication) is to be selected to correspond to the speed of the application, system loading and admissible power loss [54-56J. Roller and bottom bracket bearings normally have to transmit the most power in small constmctions. They are therefore frequently constmcted as sliding bearings [52 J.
Feed Spindle Bearings. These make great demands on the thmst bearings at extremely high operating speeds in terms of accuracy and loading, and therefore always take the form of rolling bearings which are preloaded. Gear Bearings. Shafts, gear hubs, etc. operate in these as components of wheel gears. They normally transmit
Manufacturing Systems. I Machine Tool Components
Table 4. Areas of application of various bearing systems in machine tools
I.~ II If §j II £-.8
.'.'.'
c.'• • •• •• •• v ' •• •• •• •• • • • ~
Slandard m~ing machines
~i
~ .~
~ .~
1=
-= f~ eE
cylirIOOcaI
!=C')Imncal Turning Boring
:'~~bOO~nes Roler and ball bearings
Feed spindles
Gear sllafts
0 0
()
H~muling
~
() () () ()
()'
() ' () ' () '
0
0 0 0 0
0 0 0 0 0 0
() () ()
, Where surtace roughness 01 less tIIan 02 ""' is req\.rired , Lin~ed suitability willi greasing
,
~
•
2 3
> 10' rnrrVrnin
o Unsuitable
() UmHed suHability
•
Extremely suitable
relatively high speeds over a small to medium speed range. Applications are found in standard rolling bearings with small relative speeds and in small constructions; sliding bearings made of bronze or cast iron are also used.
Fonn
5 b
F,-4 Carrying 100ce I;, Narrowesl poinl 01 gap e FOIm 01 eccenlricit)o
Figure 64. a Grinding spindle bearing with hydrodynamic multiple sliding valves: J bearing bushes, 2 thrust bearing, 3 spindle, 4 spindle box. b Cross-section of multiple sliding valve: I spindle, 2 sliding surface, 3 pressure envelope, 4 bearing sheU, 5 oil groove.
Sliding Bearings with Hydrodynamic Lubrication [38,45-47.49] (see F5). These are employed in machine tools as main bearings where high accuracy and good damping characteristics are required at high, virtually con-
stant operating speeds, i.e. wear-free operation in viscous friction area, or where high power is to be transmitted in small constructions to mixed friction areas in some cases. Circular sectional bearings are used in heavy machine tool constructions, such as in rolling machines, large-scale turning machines and eccentric presses. Multiple sliding valves (see F5.6) are used as spindle bearings for low-power, high-speed grinding, fine boring and fmish cutting machines. These bearings should be avoided in multi-directional or frequent startup applications with fluctuating power. Sliding bearing bushes are normally designed with a conical form for controllability and are press-fitted or in some cases cemented in. To prevent strain on the edges, these must be installed accurately and have minimal shaft bending deflection. Slide materials [48] include tin or lead plating or bronze. The surface of the shaft must be tempered, ground and superfinished. Surface roughness and circularity errors are between I to 2 fLm. Diameter tolerance is h7 to h8, bearing clearance is 0.4 to 3%, of diameter and length/diameter ratio is 0.5 to 1. Example (see Fie. (4) . The grinding spindle bearing 1501 shown is centred by two multiple sliding valves with ftxed sliding surfaces with the clamping effect of four pressure envelopes.
Sliding Bearings with Hydrostatic Lubrications [38, 44] (see F5.1O). These are employed as the main bearings in grinding, fmish-turning, and boring and milling machines where high loads have to be taken up and high rotational speeds are required. However, almost any operating characteristics may be achieved at will with the
appropriate selection of structural parameters. These advantages are countered by the high costs of an oil sup-
ply system and safety precautions required in case of breakdowns. Low oil viscosity should be used to maintain lower levels of friction loss and heat buildup at high sliding speeds (15 m/s and above) and with a small oil gap (around 30 /-Lm). Careful installation and consideration of the shaft bending resistance are required, because dry friction may occur as a result of tilting. Example (see Fig. 65).
RoUer Bearings (see F4). These have a broad area of application in machine tool constructions, owing to their adaptability to extremely high requirements, such as high accuracy in continuous operation, high load-bearing capacity and rigidity, and high operating range at high speeds with low heat buildup. These requirements are satisfied by a combination of the advantageous roller, retainer and bearing surface design an.d arrangement, bearing clearance and/or preload, lubrication and choice of quality classification. Roller bearings are standardised and therefore less costly, and are of a simple design. Roller bearings are employed in spindle bearings up to the highest precision classification defined in DIN 620. A small to medium preload should be applied to the bearing to provide the highest possible rigidity and concentricity levels and the lowest possible wear [57,58]. Lubrication of roller bearings is essential because otherwise they would break down after a short period of operation, and anyway the bearing temperature would be too high. Grease is used to lubricate small to medium-sized bearings. Where the bearings are employed in high-speed applications, with operating speed characteristic values
1.4 Linear and Rotary Guides and Bearings. 1.4.2 Rotary Guides, Bearings
i-' I
I I
i
•
v,
';1
Figure 67. Milling machine spindle (SKF, Schweinfurt): 1 taper roller bearing.
ity and damping, particularly in two-row design. Clearance adjustment is via a taper roller seat on a spindle.
Figure 65. Hydrostatic spindle in bearing (FAG, Schweinfurt): I oil supply, 2 oil drain, 3 hydrostatic pad, 4 gap, 5 pressure envel· ope, V oil volume. a Cross-section of pres.'iure envelope (resulting force of pressure). b Longitudinal section.
ndm of 0.5 to 1 X 106 mm/min (dm = mean bearing
diameter) , oil injection lubrication or minimum oil lubrication is recommended in most cases owing to the increased levels of operational safery these offer [54,55]. In injection lubrication, a substantial volume of oil is guided around a cooled oil circuit and therefore also serves to cool the bearing. Where precision angular-contact ball bearings are employed, as is normally the case in high-speed spindle bearings, this greasing method may be employed with operating speed characteristic values of up to ndm = 0.8 X 106 mm/min. Special-purpose synthetic greasing is then required, with accurately regulated meter-
ing according to the roller bearing. In addition, an accurate input process is essential in this case, together with gradual increases in speed and intermittent operation [54,55]. Cylindrical roller bearings are frequently employed as radial bearings (Fig. (6), the rollers producing high rigid-
Figure 66. Milling spindle (SKF, Schweinfurt): I cylindrical roUer bearing, 2 thrust angular ball bearing, 3 oil supply, 4 labyrinth seal, 5 oil drain.
Taper roller bearings provide a means for adjustment by way of the axial advance of a bearing ring (Fig. (7). They possess good damping charact"ristics, but rotational speed has an upper limit imposed by the rim friction of the rollers. The O-shaped arrangement of the taper roller bearings helps to compensate for the temperature expansion. Thrust angular ball bearings cater for operating speed characteristics of up to ndm = 5 x 10' mm/min with low preloading (Fig. (6). Thrust cylindrical roller bearings are used where there are substantial axial forces and the rotational speeds are not too high (ndm:S 0.4 x 10' mm/min), e.g. for table bearings in large-scale turning machines or feed spindle bearings. In the case of the latter, the spindle should be supported axially at both ends to improve its overall rigidity. Axial grooved ball bearings are used to transmit axial
power, and are therefore little used as spindle bearings. In order to be able to take up high axial spindle loads, thrust angular ball bearings are employed in preference. Angular-contact ball bearings operate at high speeds.
The low rigidity of these bearings, particularly in an axial direction, may be increased by a series of bearings (up to four) arranged back-to-back in a tandem arrangement
Figure 68. Main spindle of turning machine (FAG, Schweinfurt): 1 L'Ylindrical roller bearing, 2 angular-contact ball bearing, 3 grease chambers, 4 labyrinth seal [591.
Manufacturing Systems. 2 Control Systems
which are preloaded with up to two single thrust bearings (Fig. 68). Angular-contact ball bearings are frequently employed in combination with one or more rows of cylindrical roller bearings. Where the spindles to be supported
are to be operated at extremely high speeds (operating speed characteristics ndm = 1 to 2 x 10" mm/min), angular-contact ball bearings should be used in preference to all other types.
Control Systems • • • G. Pritschow, Stuttgart
2.1 Fundamentals of Control 2.1.1 Definitions of Control
DIN 19226 and IEC SO/chapter 351 define control as a process within a system whereby one or more variables as inputs affect outputs, depending on the individual system environment. The tenn control is also used to define equipment where a combination of mechanical and electronic engineering and infonnation technology, in jargon technotronics, nowadays fonns the basis of automation in machine tool construction. The control system is an essential part of a machine tool, allowing an operating process to be carried out independently according to a specified program. DIN 19237 and IEC 117-15 part 15 provide a specification for the term [21], defining it in tenns of types of infonnation layout and signal processing. 2.1.2 Information Layout
A distinction is made in infonnation layout between analog, e.g. cam-throttle control, cam-type control, copying control, and digital, e.g. NC, systems. The latter operate with digital (quantised) signals which are normally represented in hinary fonn (bivalent). A further categoty of control classification in the information layout is the distinction between digital and binary control systems. Binary control systems operate primarily with binaty signals, which do not normally form part of data displayed in numerical fonn. 2.1.~
Program Control and Function Control
Machine functions such as motion and switching functions which may be accessed manually are known as manual control systems; where these are accessed through the various steps of a program stored in memoty they are known as program control [1]. Digital program control systems have a processor which interprets the user program step by step. Program control is used to process source statements into individual function requests and to coordinate the function process automatically. Where the control status is determined by a time factor, for example with the guide of a tool cutter over a cam plate, where the angle of rotation is a function of time, then the term time control (e.g. cam-throttle control) applies. All other program control systems are process-control-based, i.e. the conditions enabling advance to the next step of the program are dependent on specific process variables being satisfied, such as path, temperature, force, etc. Positioning control systems are frequently employed in the control of machine tools, the commonest alternative being numerical control [3, 4]. Conversion of machine functions requested manually or
via a program is carried out via function control (Fig. 1). This analyses the functions requested according to a predetermined series of machining sequences and initiates their execution. Function control may also include program control, depending on the complexity of the tasks set. This tenn also covers a wider, more general context, as program control is ultimately used to convert logical functions (Fig. 2). In this case, function control consists of measuring elements and actuators. Actuators are elements which have a direct influence on the system or process as the output of the control or monitoring device. Elements which are actuated include e.g. hydro and electrical motors, hydraulic and pneumatic adjusting rotors, couples and transmission units. Where the operating program is based on process control, measuring elements are attached to the machine, e.g. position monitoring systems, which signal the status of the process to the control system, i.e. the position of the tool. This enables machining sequences to be initiated or halted in relation to the path travelled or to specific positions. 2.1.4 Input and Output of Signals
A signal at the input of a functional component is described as an input or incoming signal; in the same way, signals at the output are called output or outgoing signals. Signals are normally controlled by input and ontput elements before and after processing. The following func-
tions apply for these elements:
-
Input
element: resetting, re-fortning, directing, isolating potential, conditioning, converting (analog/digital, digital/analog). Output element: amplifying, converting, storing, decoupling.
Input and output elements may be omitted where the processor technolOgy is modified to suit the signal
Individual requests
Input
l
D
Program oontrol (processing) C>
<= .~ <>
~.
i2
"*
D
,
Program sequence , 'Process oontroO'ime controij
~
D
~unctlons
_~
Jl1!Qt~.!~~e!l:,~~~~QJl~~~_ Interlocking devices
0Cl
Control Operating program
Function control (execution)
U
I;rocess control e.g. ,II continuous path oontrol
H'
Measuring elements
"'-
Actuators Process, system
Figure 1. Control structure.
Jl
ITime oontrol e.g. I oontrol shaft
,
2.1 Fundamentals of Control
• 2.1. 7 Control Programs
=
T
Function control level n: F,
I
nk
""' Fn_1
I
T
~
Program control
""" T~ Fn-1 Program contra
ft
I
Program control
J
X
-
I
r=7I
Y777>
T
Program control .,J,. T~
T I I'
~
environment within the controller (system-compatible signals).
Figure Z. Function and program control.
Remole action condition
2.1.S Sipal Formiag Input and output signals in a control system are signals from a signal-forming generator. A distinction is made between the following, depending on the signal:
-
process (acoustic or visual signaJ1ing as specified in DIN 19235 and IEC 73); acknowledgement - signal which has a direct effect on a command.
2.1.6 Signal Processing Each control function may be classified as either signal input, signal processing or signal output, irrespective of the scope and level of control. Signal processing takes the form of either logic control or sequence control.
Logic Control. This means that output signals are assigned to specific input signals in the context of logic operation. Signal processing is carried out via basic logic elements. The following are examples of basic logic elements: -
-
logic elements: AND, OR, NOT; monoflops for pulse contracting, signal delay or signal stretching; flip-flops, such as RS flip-flops (R = reset, S = set), D bistable flip-flops and JK flip-flops.
Sequence Control. Control systems with positive stepby-step sequences are known as sequence control systems. A distinction is made here between control with time control or process control step-enabling conditions. The control problem may be described in the form of a sequence cascade (Fig. 3) . The following are important attributes of sequence control systems based on process control: -
Remote action condition
reporting - signal to inform operators about status or
Only one sequential control element is set. The step enabling condition is only dependent on conditions following the current step. The safety interlocking function operates independently of the sequence cascade. Extensive control tasks frequently require a number of
Figure'. Description of sequence cascade.
sequence cascades, which may be derived from the structure as shown in Fig. 4. There is another distinction based on time control of
signal processing:
Clocked Control. With this, signal processing occurs only in the individual control elements at specific times which are synchronised by a clock pulse. This process is particularly useful where different signal propagation times in various parts of the control system and their spread would mean that any control result which may be produced could be ambiguous. It is employed chiefly in electronic control. Asynchronous Control. Asynchronous signal processing is based on demand and operating time, and not linked to a fixed time. The type of control ensures that no operating time-dependent errors occur in signals which may influence one another. This normally takes the form of a prescribed series of signals, where data processing is only enabled after specific signals for release have been given and a subsequent operation may only be initiated in conjunction with an "operation completed" message relating to the previous operation. 2.1.7 Control Programs Charac:teristics. Control programs include all the commands and directives required for signal processing, as a result of which a system (process) is influenced in accord-
Manufacturing Systems. 2 Control Systems
Parameters
I I
I II II
Step-enabling conditions
II II II
Safety intertock
connections or pcb connections, or programmable, i.e. variable, by way of plug-in line connections, diode matrices, plug boards, variable crossbar distributor panels or interchangeable modules. In the case of PLCs, a distinction is always made between freely programmable control systems and programmable controllers with interchangeable memory. The progranunable memory in freely programmable electronic control systems is a read/write memory or RAM (random-access memory), where the entire contents may be changed at will without any mechanical intervention and to any degree however small.
Programmable controllers with Interchangeable memory on the other hand have a ROM (read-only memory),
where the contents, once programmed, may be only changed by mechanical intervention in the control system after they have been programmed. A distinction is made here between controllers with read-only memory that are programmed following manufacture at the factory and may be changed subsequently (RPROM = reprogrammable ROM) and those that may be programmed only once during or following manufacture and are subsequently invariable (PROM = programmable ROM).
2.1.8 Organisation of Control a1
Operating field
an a2 Control commands
Figure 4. Structure of process control system.
ance with the task-related conditions. This may occur in various forms. Rigid systems operate with fIXed programs, where a selection may be made between a number of programs. Where programs are changed frequently, it is more practical to employ interchangeable, freely programmable memory. This may apply in the case of e.g. cam plates, cams, stops or notched bars in mechanical control systems, while in electrical control systems applications may
include program cylinders, crossbar distribution panels, punched tape or electronic data carriers. Exchangeable programs produced by the users of the processes to be controlled are known as user programs. Electronic control systems require additional internal system programs to enable them to interpret and process these user programs.
Program-Dependent Hardware_ Classifications are provided in DIN 19237 [21] and IEC 117-15 part 15 for hard-wired programmed controllers and programmable logic controllers (PLC) (Fig. S). Hard-wired programmed controllers may either be permanently programmed, i.e. invariable, e.g. owing to their hard-wired
Great importance is attached to hierarchical process control system organisation in industrial applications. The various control systems assigned to the different hierarchical levels are as follows:
IndiriduaI Control. A process within a control system is normally influenced via actuator intervention at the level of the individual controller. This is the smallest control element which monitors the drive components and may be operated either manually or from a superior level. The appropriate operating mode is determined via specific input in each individual case. Where operation is selected with a volatile command, the individual controller is assigned a control signal memory to detennine the time
of issue of the command. The operation (command input) is only activated where the safery interlock conditions are satisfied and a release is triggered, where this is applicable. The individual controller normally includes a monitoring and reporting sector for reporting the operating status of an actuator (steady state, transition, not ready, fault). All the individual control systems (the drive controllers) together make up the individual control level (drive control level).
Cell Control. The functional unit required to control a subprocess is known as the "cell control". It is superior to the individual control system forming part of the subprocess (the drive controller). A number of cell controllers may be arranged hierarchically in order of importance where this is required for the process enabling conditions. The cell controller group constitutes the cell controller level. Primary Control. Primary control is superior to cell control and is used to control the complete process. This division into individual, cell and primary control is structured according to functional units, where the nearest superior level is the control level of the inferior level in each case.
Control Levels in Manufacturing Systems Figure ,. Control classification in accordance with DIN 19237 (functional as for program implementation).
Classification of control tasks in terms of levels leads to decentralisation of the data processing tasks and thereby
2.2 Means of Control. 2.2.1 Mechanical Memories and Control Systems
to less complex system parts with their own separate data management and configurable standardised interfaces and modular software. The advantages of the autonomous system parts are wider accessibility over the entire system and simplified conditions for commissioning or modifications. The controller tasks are hierarchically structured in production engineering in tenns of primary, cell and machine controller levels. This classification also represents a structure based on functional units (Fig. 6). The above designation may not be adopted for all applications. The cell control level may be linked to the primary control level, depending on the size of the production system, and may also undertake machine control tasks where the system hardware requirements are appropriate. The control tasks in a chained production system may not, therefore, necessarily be pennanently assigned to the above-mentioned levels; nevertheless, they nonnally folIowa structure which is similar to the one illustrated in the following example:
Primary Control Level -
Control data generation for workpiece and tool flow (internal planning). NC program management. System mapping. Production data acquisition (PDA)/machine data acquisition (MDA) for composing and editing for display, documentation and controlling.
Cell Control Level -
-
Tool data management. NC program organisation. Acquisition and evaluation of PDA/MDA data. Equipment synchronisation at machine controller level. Evaluation of measured data and control where applicable.
Machine Control Level -
-
Manual operation/initialisation. Program debugging. Processing tool offsets. Generation of axis motion. Processing logic functions. Monitoring and diagnostic functions. Measuring processes. PDA/MDA data acquisition.
2.1.9 Databases and Link Structures in Manufacturing The requirements for manufacture are as follows: to be able to satisfy market demand flexibly, and to frequently supply up to lot size 1. This means that the response time between order and delivery must be minimised via the deceptively simple fonnula: turnaround time/processing time -
EDP equipment linking via networks with standardised network protocols; uniform data fonnats for communication between sectors; coordination of flow of information.
2.1.10 Safety Standards No fault or error occurring in automation systems may result in any danger to human life, and those who design, install and operate such systems have a responsibility to provide safety functions in addition to executing control and monitoring tasks. Current engineering practice is to
be observed for industrial control technology in the form of:
-
i
-
MaChine CO=tntrOllevel Machine controllers
. - '=1'TT
. '\ TT--T Individual controllers
Process, system Figure 6. Controller hierarchy.
1
with the maximum degree of flexible automation of the means of production. Only if planning and implementation at all production levels, including the relevant databases and data sinks, are computer-based and if the appropriate databases and computer systems assigned to them are integrated into a computer network via a commnnications system may this be achieved. The short response times which may be achieved between the areas involved, plus data integrity and the power of the decentralised yet interlinked process engineering closed-loop control circuits are of considerable help in achieving this ambitious target. Figure 7 shows which areas are principally affected by this in tenns of the areas in a plant that generate data that may be combined within a CIM system (CIM = computer-integrated manufacturing). The data provision requirement for integrated data processing can only be attained with the aid of individual database systems which are networked and coordinated. In addition to the actual transmission technique employed, there are also the questions of data management and data content which have yet to be addressed. The following bastc functions are required for a CIM solution:
The law relating to the safety of operating equipment (Geriitesicherheitsgesetz GSG, Bundesgesetzblatt (Gennan Civil Code) Pt I, p. 717. Accident prevention regulations in accordance with the Commercial Trade Association (VBG) Register (VBG regulations). Safety regulations: Rules and principles relating to ZHI Register (Workers' cooperative guidelines on work safety). Technical standards: DIN 31 OOO/VDE 1000 General principles for safety structures for technical products; DIN 57 113/VDE 0113, lEC 68, lEC 529, lEe 536, lEC 742 and lEC 801: Regulations for electrical equipment in machining and fmishing machines with rated currents of up to 1000 V; VDI/VDE 3541, Sheets 1-3: Control systems with interlocking functions .
2.2 Means of Control 2.2.1 Mechanical Memories and Control Systems Cam·Throttie Control. Cam gears are frequently employed for generating path and feedrate variation, i.e.
--.0 -.:I11I:II
Manufacturing Systems • 2 ControISystems
Market
Customer
~Iity assurance (OA) (Computer-aided
Planning level
1--: 0--:-,-::.---:-I------~--------:T:-ech--;---:-nical--;----j ~~~_n~~~~~a _ _ _ _ _ _ _ _ _ _ _ _ _ _ _~
q~ity
assurance - CAO) - Products - Pricing - Customers - Ordering - Settlement - Complaints - Orders - Stocks - Capacity - Production volume monnoring - Cycietime monnoring
- Distribution and settlement
- Tools -Carriers - Machines - Material
Order data Staff data Material data Machine data Production
il
Corrections! Revision
.!
~
Application process (CAP =computeraided planning) - Work sequences - Programming - Calculations - Simulation
characteristics - Work schedules
- NC programs - Calculations
Setpointiaclual -
(CAll)
-Testi Product definition
Production planning and control (PPC) - Order planning - Order execution - Progress raports - Recalculation - Parts logistics
as
Development and construction (CAD =computeraided design) -Draft - Calculations
- Part classifications - Alternatives - Standards - Parts lists
SetpoinUaclual
8
Operating levels (CAM - Computer-aided manufacture) Primary level (order-based prodJction)
D
Cell level Individual control
Data Functional level
Fipre 7. Databases in works organisation.
curves represent mechanical reference systems for path and feedrate variation. The required motion displacement
is transmitted via the sensor head of the transmission element to the pan to be moved, e.g. to the machine tool slide during one revolution, depending on the path and feedrate. These curves may be either three-dimensional (drum-type cams) or two-dimensional (ptate cams) (Fig. 8). One imponant area of application for cam-throttle control is in automatically controlled turning machines or presses. Process control is carried out automaticalty via plate cams and cams which are positioned on control shafts and normally rotate at constant speeds (time-program controt). The curves form program references for the paths and feedrates, where the relations are as follows:
Figure 8. Principles of caJDMthrottle control: a pushing - drumtype cam (fonn-fit), b pushing - plate cam (adhesion), c turningdrum-type cam (fann-fit), d turning - plate cam (adhesion).
Path reference. Stroke fls = f( a); Feedrate reference. v
= w(ds/dOl);
fls_ = r_ - rmID' w = 27f/T, 01=
wi.
Here a = the angular position of the curve, T = the cam plate radius, w = the angular Velocity, T = the period of rotation. These transmit the feedrate required to the actuator and the moments and/or power required for acceleration. The transmission mechanism consists of mechanical components such as rollers, levers, ball bearings, guideways and springs.
CIUIl Control. Cams move a ram by travelling over it, thereby triggering a mechanical, hydraulic or pneumatic logic function. Electrical spring-operated limit switches trip logic functions accurately. Examples of logic functions include main spindle speed changes, coolant supply circuitry, switches for feed motion, etc. The cams are normally attached to cam strips or cam rollers which generally have several cam tracks and may be clamped at any point at will (Fig. ,). The cam strip attached to the slide or tool bed serves as a reference p0sition for the cam program controller, whereas the cam roller rotating with the control shaft represents time-program control. Cam tracks are constructed in the form of a rectangular, T-shaped or dovetailed groove. Tracer Control. Tracing (copying) denotes a machining process where the tool motion is controlled by a control curve or surface area (model, template), so that the profile of the sample is transferred to the workpiece. Trac-
2.2 Means of Control. 2.2.2 Fluidics
9 Equidistant sections
~-"
6 Figure 9. Continuous path c..:am-type control 1 to 5 adjustable cams, 6 cam limit switch, 7 bed, 8 message, Y slide.
ing is employed for manufacturing workpieces which are complex to create (e.g. shaped tools). specifically in small-scale series production, but is being superseded increasingly by NC technology. A distinction is made in tracing between singie-, doubleand triple-axis kinds. In single-axis tracing the feed motion of the tracing slide is only controlled in one axis, whereas it passes through the machine in the other direction of axis with a constant sliding feed. Similarly. two or three axes of motion are controlled in double- or tripleaxis tracing, where both processes permit three-dimensional tracing. The slide motion in double-axis control then runs along the third axis with the sliding feed (= 21axis transport), e.g. for traverse milling and multi-layered rotational milling (Fig. 10). Transmission from flat curves in samples to the workpiece is effected via eq uidistances (parallel curves). The curved radii of the sample curve, the probe radius rand the tool radius R have to be viewed as critical limit values and in relation to profile distortion. Thus, r "" R leads to distortions in transmission. This means for the case R < r (Fig. 11) that the profile radius of the workpiece for concave profiles is greater (smaller in the case of convex profiles) than that of the template. A probe radius of the template of Ps < r will result in coarser workpiece pro-
Figure 11. Touch trigger probe transmission from sample curve to workpiece via equidistant sections with different probe and tool mdius. R tool radius, r probe radius, P paint, p profile radius, suffix S template, suffix W workpiece.
motion is normally signalled via directional control valves (Fig. 12), where the speed of the motion is set via flow valves. The directional control valves are activated either mechanically, directly from' the system, or electromagOn
Wiring diagram
1-1
W====r-" I
1.3
1.1
1.1
Oft
On
files.
2.2.2 Fluidics
Fluidics (see G) operate with compressed air or hydraulic oil [5-7 J. These are applied where fluid drives are used because their special characteristics and the control tasks are simple. This saves on conversion from one energy form to another. The introduction and end of
With advance in Z-axis
With probecontrolled feed in Z-axis
Control system flowchart (path-time diagram) No.
Designation
1.0 Cylinder
tion
lOft On
a I ~/two-way direc.l 1 -I tional control valve I o I 1.3IThree/;;w~Y~ ,tional control valve I o I - tional control valve
I
1.5
b
Figure 10. Profile milling methods: a traverse milling; b contour milling; F sensor, z line feed.
Two/two-way directional control valve
1 6 Four/two-way direc. Itional control valve
I
t
'+,
I
I
I
~~-I
~ (
I
-H
1
I
I I
0
I
I
I
1 i Ofi 1
~-
.-
I
+
11
(
I---
-1+-
lWrf I ,
c- t---~ 0
a
- --
1 1 Three/two-way direc- 1 . ,tional control valve I
~-T;olhvO:;ay direcl XZ,(Y,Zl f [X,,Yo ,lrl
Work process I 1 2 3 4 5 6 7 8 9 10
I~OSi-
I'
i '
II I
I
i
t-l~Mtion
ri
'
I
I
I H---
Figure 12. Switching examples for hydraulic drill feed equipment.
I!IDII
Manufacturing Systems. 2 Control Systems
netically, on the basis of electrical signals. Occasionally, pressure-related switching is also used. The combination of electrical signal processing and hydraulic oil mechanical advantage is known as electrobydraultcs. Simple logic operation and ease of handling of electrical signals are combined in this case with the high mechanical advantage and good time response of hydraulic oil drives. The combined effect of directional control valves and flow valves may be produced with servovalves. These consist of an electric motor with a single- or multiple-step piston cylinder system, enabling the restrictors to be adjusted continuously depending on the excitation of the motor. Continuously variable speed drives may be set up with servovalves, where their characteristics may be improved specifically with the employment of a speed control mechanism (servohydraulics). With servohydraullcs, speed adjustment takes place via an analog adjustment to the restrictors which affects the oil flow. The same aim, i.e. an adjustment to the oil flow, may be achieved with digital hydraulics by opening and closing the oil flow via a rapid switching valve at high frequencies and the time intervals after which this may be opened or closed are variable. Fluid drives are available as rotary or linear motors. Transmission units are not always required, in contrast to electric motors, on account of the high pressure of the fluid media and the associated high torques. The ratio of the moment of intertia to the torque which corresponds to the startup time of the motor, is lower in rotary bydraullc motors than in the majority of rotary electromagnetic motors. AI; the favourable ratio is also due to the low moment of inertia, the additional moment of inertia of the machinery to be driven also needs to be taken into account in the final calculation. Linear hydraultc motors may be manufactured extremely competitively on the piston-and-cylinder principle. They also save on the mechanical transmission of rotary motion with their linear motion, which is more fre-
quently required. Hydraulic otl linear drives provide for high loads at average speeds (up to I m/s) with adequate rigidity. Their rigidity is directly proportional to the total enclosed oil volume, and may be improved by increasing the operating pressure. Higher operating pressures at the same output produce lower manufacturing tolerances (due to the associated losses), however, and therefore more expensive components. Thus, the dependence' of the operating characteristics on the temperature is increased. PneumaUc linear drives are cheap and extremely fast. They are designed for positioning rates of between I and 10 m/s in borderline cases. The forces to be generated are restricted by the dimensions and limited operating pressures, nortnally <6 x 10' Pa «6 bar). Their rigidity is low, and the same applies to the potential speed adjustment. They are therefore used almost entirely for positiOning motion. The use of hydraulic 011 drives in a machine requires a hydraulic generator which converts electrical energy into hydraulic energy. For maintenance purposes it is advisable to install this generator with the control valves in a hydraulic control cabinet, where possible. The control circuits may therefore be composed of individual elements or a number of valves may be connected to form a common control block (hydraulic block). Where there are high requirements as regards time response, the switching elements should be arranged directly adjacent to the drives to reduce the volume of oil to be switched. Hydraulic oil systems are described in hydraulic draw-
ings and parts lists. The symbols to be employed for this are listed and explained in DIN 24 300 and/or ISO 1219.
2.2.3 Eledrica1 Coatrol Electrical control systems are designed as either contactcontrolled or electronically controlled.
Contact Coatrol. Large outputs may be switched easily and rapidly via contacts. These are also suitable for binaty circuits (DIN 19237), where a change in the system status is carried out via a change in a binary-coded signal. The combination of contacts and an electromagnetic drive is known as a contactor or relay. AI; both the switching of three-phase motors and the operation of actuators are frequently effected via contactors, linking may also be carried out with additional contacts to these elements. The load factor and logic level are amalgamated within the system. Where the control systems are not too elaborate in the function control sector, contact controls are a suitable solution. However, it should be noted that the number of switches for contactors is limited mechanically to around 10" to 107 switches and the electrical Switching capacity of the contact itself has to be considered. Furthermore, contactors require Switching times of 10 to 200 ms which have to be taken into account in high-speed processes, e.g. with cutoff at high speeds. The switching times are dependent on the type and output capacity of the equipment and are affected by the magnetic flux. Contact controls are described in circuit diagrams and parts lists and are structured with progranuned connections (Fig. 13). The opportunities for rationalising the manufacturing control system are therefore limited. Symbol representation and the standards to be applied in connection with their use are contained in DIN 40 703, DIN 40713 and IEC 117-3. The VDE regulations are also to be observed in association with their practical design, as these define state-of-the-art technology. Specifically, these include VDE Olooff and VDE 0113, which cover systems for processing and finishing machines. The switching devices to be used are to correspond to specific switching classifications. The control voltage in contact controls should amount to 220 V. It is not nortnally advisable to switch voltages of under 24 V using contacts owing to switching jitter. Direct currents should only be switched with the contacts specifically designed for the purpose, and protective equipment should be used to prevent arcing. Electronic: Coatrol. Where the data processing extends beyond simple logic tasks, electronically operated control is nortnally employed. This is used both for binary (bit) and digital (word) signal processing. The processing of simple functions such as an AND or OR logic with simple functions within a semiconductor component is carried out in the same way as the operation of a counter or a digital/analog converter. Electronic control systems are unlimited in terms of the number of switches or their useful service life, they provide extremely rapid Switching (nanoseconds or microseconds) and this is achieved with a low load factor. PLCs (programmable logic controllers) are employed as the hardware solution for bit and word processing of process and logic-orientated control problems. In addition, digital control is used for special applications using computers (and microcomputers). In both cases, the control algorithm is achieved via a program. In certain cases, bard-wired program controllers are still employed. These are only economical in large volumes with the same control systems due to the development, pre-production and inspection costs associated with pcb-
2.3 Programmable Logic Controller (PLC) .2.3.1 Structure
R, 5, T,
eli
'OF 54 12 f-+y--
It-
r-
50
.1 1 I
K29
.1 .2 .1 .2 5 41 51 13 I I I ~
\--t-l--t-t6 42 52 14 2 4
Po K2
K1
55
rVI
Y1[:::~:J-~
N
R1
.2 .2 .2 .3 .1 1 3 5 41 51
r- t-t-t--\-I
I ~
I
2
4 6 42 52 14
a1
I
.1 13
I
K1 Main contact clockwise rotation K2 Main contact anticlocl
~-V-r I
V2 V2 R2 V3 V3 R3 Q1 S5
Binary, digHal and analog signals Switch
E
type data carriers. Electronic control systems may be subject to mechanical or electromagnetic interference from peak currents owing to their rapid switching and the status conditions that may be stored. Measures to counteract this include careful proof testing of the mains supply equipment, sufficient printed conductors and screening of the equipment itself. Inputs and outputs should be kept free of interference via a low-pass filter and be decoupled mechanically if necessary. Care should also be taken to ensure the reference potential is well-defined with adequate earthing. AI; the control equipment is only heated slightly, owing to the low power factor in signal processing electronics, it is easy to provide protection for the electronic components against dust and humidity. The output signals may be set to the level required for the actuators by Switching amplifiers or infinitely variable power amplifiers. Actuators are thereby frequently driven via interconnected contact controllers. Continuously variable power amplifiers are required for the relevant variable-speed d.c. drives. These may also be constructed as thyristors or transistor regulators . Programmable logic controllers (PLC) and numerical control (NC) are dealt with in greater detail below in view of their significance in the control of manufacturing systems.
2.3 Programmable Logic Controller (PLC)
According to VDE Guideline 2880, the term "programmable logic controller (PLC)" is defined as follows: a programmable-memory automation system with user-orienClutch speed 1 tated programming language which is employed primarily Rectifier to V2 for control purposes [14]. Resistor to V2 This type of controller normal1y consists of a bit or Clutch speed 2 word based processor with RAM, ROM and PROM Rectifier to V3 (Fig. 14), where specific software enables control probResistor to V3 lems to be described in a user-orientated programming lanSwitch speed I, 2 guage (circuit diagram, logic diagram, Boolean algebra, Switch release spindle brake status graphs, sequence cascade). (Ex1ract)
Fipre 1~. Example of contact control: spindle drive of turning machine.
0
Y2~{/R2ZQ Y3_ck~R3~
List of Control and Drive elements
Ml Main motor SI Main OFF switch
C
K2]
Klit-
c?
.1 3 I
h
53E--~ Kl)
K2 It Kl
B
,--Inchlng operaijonLl
5H-
52
A
III
Ml 3~
Ro
3
1
2
1
...
:1.3.1 structure
Simple PLC controllers are almost all programmable as far as logic operations are concerned. AI; such operations in the control sector are based on bit data, the processors preferably operate with word lengths of a single bit. Each logic operation to be executed by the controller is arranged in a fixed order in the memory and may be
Binary, digital and analog signals LCD lamp
Contact switch
Relay
Decade switch
DigHai readout
Light barrier
Solenoid valve
Measured values
Setpoint values
Fipre 14_ structure of programmable controller with bit and word processor.
Manufacturing Systems. 2 Control Systems
r-------iJ
: r , "" '" I
~l
I
I I
I I
I
3.
H
l
Process image inputs
1st command 2nd command
e
C>
~
!l? 0
!
5.
I I
E
ec-
6.
-~---
--1-
7.
-+-----
f-------t---
I
Whereas programming types 1 to 4 form part of logic control, types 5 and 6 are classified as process control.
I
r--
I Last command Set outputs
L I IL_ _ _ _ _ _ _ --1 III
I
H Process image outputs l
Figure IS. Program organisation with process I/O image.
accessed cyclically and processed accordingly (Fig. IS). The overall PLC program consists of a system program and user programs. The system program encompasses all the commands and directives of internal device restart ftmctions and forms a permanent part (EPROM) of the PLe. This program may not be modified by the user. The user program is processed cyclically by the PLe. When the program has run through to the end once, the cycle begins again. Inconsistencies, which may be generated when an input is accessed which changes state during the course of a cycle, may also be avoided by way of a program organisation with a process I/O image. All outputs are updated simultaneously at the end of a cycle and their values are stored as a process image in an integral memory corresponding to the simultaneous access of all inputs and stor-
age of these values in a process image.
2.3.2 Programming PLC programming is similar to the standard descriptions of those of conventional controllers which are more or
less standardised. Programs are classified as follows: I.
2.
4.
Logic diagram in accordance with DIN 19239 Symbols [22] (graphics, logic diagram). Boolean equations in accordance with DIN 19239 (mathematical description). Flow chart programming in accordance with DIN 40719, Sheet 6, IEC 113, IEC 750, DIN IEC 65A (SEC) 67 and IEC 1131. Flow diagram in accordance with DIN 66 001, ISO 1028-1, ISO 2636-1 and ISO 580H. Use of higher programming language in accordance with DIN IEC 65A (SEC) 67 and IEC 1131.
Ladder programming in accordance with DIN IEC 65A (SEC) 67 [23] and IEC 1131 (circuit diagram, graphics). Mnemonic designation in the form of a command listing in accordance with DIN 19239 (alphanumeric).
Program type 7 is employed for complex control tasks. The types described under 1 to 4 produce the same type of processing within the PLe. Only their representation is different (Fig. 16). The latest PLC programming systems provide direct graphics programming of sequence cascade representation (Fig. 17). The French standard guideline known as GRAFCET has been adopted by many manufacturers and much of the new standard guidelines forming DIN IEC 65 A (Draft) and IEC 1131 are based on this. The most important elements in the language are the steps which specify actions or conditions in accordance with DIN 40719, Sheet 6 IEC 113, IEC 750 and stepenabling conditions (transitions T) which designate the point of deactivation of a step and the transfer to the next step (activation). In addition to a straight sequence in a chain, it is also possible to represent branches and synchronisation of branched chains. Branches may run both forwards (parallel sequences, jumps) and backwards (loops). Two time variables are allocated to each step: the wait time, which describes the minimum lapse in the step irrespective of the action time of the step, and the check time, which specifies the maximum length of time a step may take. At the end of the check time the program reverts to
hold mode and generates an error message. This type of representation provides a description of all machine construction control problems and has been introduced by several PLC manufacturers.
2.4 Numerical Control (NC) 2.4.1 Definition The concept of numerical control was developed at the Massachusetts Institute of Technology (MIT) in 1951 for process engineering tasks. "Numerical" means that the input of the control data takes the form of numbers. These are represented in binary code and may be processed
Ladder programming
Ust of instructions
Logic diagram
Boolean equation
Programming with graphics symbols as in circuit diagram
Programming with mnemonic abbreviations for function designations.
Programming with graphics symbols
Programming with mathematical description
Corresponding to DIN 19239
I~~n'~
Corresponding to DIN 19239
,----
List of instructions U E1 UN E 2 U E3 ON E4 0 E5 = A1
Corresponding to IEC 117-15 DIN 40700 DIN 40719 Din 19239 Logic diagram
£1~u E2
E3
Corresponding to DIN 19239
! E1 &N E1
&
E4 E5
& E3
"1
A1
IN E4 I ES = A1
Figure 16. Types of coding for PLC programming.
2.4 Numerical Control (NC) • 2.4.3 Data Interfaces
form is considered to he numerical control, irrespective of the input device or data storage system [8-10].
2.4.2 Programming
Workpiece programming describes the generation of workpiece-related control data for numerical control (Fig. 18). Path data and alter statements are to be input to a data storage device in a predetermined order. One data storage device which used to be frequently employed was punched-hole tape, because it is robust and the sequence of operations is invariable; a large volume of data could be stored cheaply and read stage by stage as the process was being carried out. The use of inexpensive electronic memory has enabled large volumes of data to be stored far more simply nowadays, so that the input of entire programs is normally carried out either manually, e.g. via magnetic data storage media, or via data links from a main host computer. NC programs may be either generated on-line, i.e. via the operator directly on the machine (shop-floor programming), or offline, as part of the operational planning stage. Workpiece deSCriptions in the form of construction drawings (Fig. 19) or CAD data serve as output data for program generation where one of the more complex process engineering programming languages such as EXAPT is employed. The source program is translated into CLDATA (cutter location data) code with the aid of a processor (conversion program). The processor thereby processes the geometric data and supplements the technological processing requirements using materials and workpiece data meso A post processor converts the computer-independent CLDATA code for use with a specific NC machine.
a Start
Good - software switen Next workpiece
2.4.3 Data Interfaces Poor - software sWitch Next workpiece
As can be seen from Fig. 20, a further data interface is relevant to numerical control systems in addition to the NC program interface as given in DIN 66025, ISO DP
Next workpiece inserted Preloading
Drilling
Withdraw
Tool and chuck list or drawing
Release workpiece b Figure 17 . a Flow diagram in accordance with DIN 40 719 Part 6 for automatic drilling station. b Flow diagram from GRAFCET designed to address the same problem.
directly from the controller. Numhers have to he input to describe the workpiece geometry (path data) and technological specifications relating to the tools and operating speeds (alter statements), in numerical form in each case. This figure designation is characterised by an address letter prefix (DIN 66 025, ISO DP 6983/3 and ISO 646). Any control in which the path data are entered in numerical
t::3Figure 18.. Flow
of
(conventional method).
data
from
Workpiece drawing
to
machining
I!IIIII Drilling pattern
Manufacturing Systems. 2 Control Systems
z
c.t>20
M10
6983/3 and ISO 646 - namely, the CLDATA (cutter location data) interface for computer-independent programming of machine tools, e.g. with the complex progranuning language EXAPT (cf. L2.4.2). CLDATA may only be run on a control system when it has been assigned control-related parameters in contrast to DIN 66 025, ISO DP 6983/3 and ISO 646. This either takes place via the post processor referred to above, which compiles the CLDATA in a form corresponding to the specifications in DIN 66 025, ISO DP 6983/3 and ISO 646 (Fig. 21), or the NC controller has an interpreter which executes the translation and modification line by line as part of the program control.
x
x
2.4.4 Control Data Processing
PARTNO 1PLA HE, 06 - 70 MACHIN IPPl TRANSI200,160,0 ZSURF 130 Pl • PDrNTl40,50 PZ .PDrNT 1100,50 Geometric definitions
Technological definitions
PART IMATERL,l CLDrSTI2 BOHR. ORILL/DrAMET, 10,OEPTH,30 SENK ·SINK/SO,DrAMET,ZO,OEPTH,lO GEWI • TAP/DrAMET, 10, DEPTH, 20,TAT,1,BllNO,1 Executive commands Worklweight WorkidrilVsink Finish Figure 19. EXAPT parts program.
The control data programmed and stored in memory is processed in the numerical controller to produce position setpoints for the individual axes or output in the form of switching commands. Continuous motion in a number of axes takes the form of continuous computerised stage-bystage output of separate position setpoints which are synchronised with the process. The position setpoints of each axis are compared with the respective actual positions. A pOSitioning control speed is obtained from the position tolerances by multiplying them by a factor which is the same in all axes (vector amplification Ky (S-I). Different position setpoints in the individual axes lead to different position tolerances, known as/ollowing errors, and consequently to different speeds which are necessary for motion over different track angles (Fig. 22). Impulses are generated for stepper motors from position setpoints. The calculation of the position setpoints from the programmed control data is carried out according to fIxed methods of calculation and is known as interpolation. The aim of interpolation is to reduce the volume of control data to a level which is sufficient for generating a combination of workpiece contours at will from simple straight lines, circles or segments of a parabola. Reducing
the volume of control data frees control data storage media and input devices. Linear and rotary interpolation suffices for the majority of applications, i.e. the control data input is then used for the interpolation points,
I
CAD ope!ating processor
Technolog'y pr«:esso! I
Tedv101ogy processor II
~ntegraI
(integral to sjSlemJ
to system)
1i,tAOJ.Ve-processo1
Interlace module
integral to systefl1
I
l
I
,IPo& pr«:esso! I (
Parts program
I
I
rl~ech~,1 processor IV
I
I
Po& processor II
1 ,1 ~ 1I 1
I
I
DIN 66 025
Language description e.g. EXAPT
e.g. EXAPT
I I l CLDATA (CUlling location datal)
I
/
CAD post processorJ
Technology pr«:esso! III
I
~
Workpiece description: construction drawing or CAD data
C) Data interlace Figure ,)0. Data interfaces in control technology.
Technological data: tool data, materials data, GLDATA code (DIN 66 2151 cutting values, operating sequences ,---_ _ _..L.._ _ _--,
) NC program (DIN 66 0251 Figure :U. Computer-based NC programming.
2.4 Numerical Control (NC) • 2.4.5 Numerical Basic Functions
and editing, and dissemination, technological data processing and geometrical data processing (GEO). These basic ftmctions are described briefly below.
Operation and Control Data Input/ Output. Man/machine communication is becoming more and more essential in the numerical control sector. New developments are taking place in terms of the operator interface with menu-driven technology, graphics screens, windows functions, etc. Operation and programming facilities are becoming more and more complex. This part of the system software already accounts for over half of the overall system in modem NC systems. The following main ftmctions all form part of the operator environment of an NC system nowadays:
Figure 1:1. Continuous path NC control for linear interpolation and directional change in path: ~s following error, kv acceleration, Vx and Vy speeds.
between which intermediate values are calculated on these curves as pOSition setpoints, so that a position setpoint is output every 5 to 10 ms to each axis. In the less common method used for three-dimensional curves, i.e. parabola interpolation, a compensating parabola is set using three points in each case. Tolerance calculations are normally to be carried out prior to interpolation (i.e. coordinate transformation, tool length offsets and radius compensation, etc.). Position adjustment may be effected via a separate external module, or in some cases it is carried out from the NC control within the control software. In the latter case, scanner systems provide computerised solutions, where discrete position control is executed via computer programs and the position control circuit within the control computer is closed. Alter statements are normally only stored in numerical control and output in sequence via a defined interface (VDI 3422) to the subordinate ftmction control units. Where programmable memory control systems are employed in ftmction controllers, the interface may be omitted as defined in VDE 3422, where the PLC is integrated directly within the NC.
-
NC program memory storage (including any associated management operations). Operation and operator guidance. Plausibility checks for input data. Editing ftmctions.
NC Data Management and Data Composing and Editing. The main tasks fOrming part of this ftmctional unit include: Preparation of NC records for decoding and display. Decoding NC records (ASCII character conversion into internal control image). Resolution of operating cycles and subprograms, parameter calculation. Execution of tolerance calculations (tool length offset, tool radius compensation).
Technological Data Processing. Technological data
2.4.5 Numerical Basic Functions
processing is responsible for the execution of the alter statements (= technological instructions) which activate the switching of main spindle speeds, feedrates, tool change systems, coolant supply, etc., for example via the individual control level.
Where the ftmctions of an NC system are divided up into ftmction-related parts, there are four basic tasks as shown
Geometrical Data Processing (GEO). Geometrical data processing encompasses all the basic ftmctions
in Fig. 23 which represent the minimum scope of func-
required for patb generation. A patb is generated via the primary control motion of the individual axes. The ftmctional units controlling setpoint generation
tions within the system. These include operating and control data input/output, NC data management, composing
Central control system
rro~ration ~nd con;rP
!Id;ta inpuVoutput
0-'IC\.
- Operation and . _ operator guidance - Editor function - NC program memory storage r
=---'''''-----,
i=l
,
Geometrical data processing (GEO)
I
:::
;::N"'C--cda-'ta-p-roc-es-'sing-,---''-, composing and editing NC ~
I O :: -~::am:: - Subprograms and cycles - Equidistancing
I'
[J (managed by operationll~nd , I j
Position se1point generation
~ -~:::!I
V
'I
-
/)
~n~.~.!::=J V
Memory intertace Exchange of data (the interface between the central control system and the function blocks is not shown in the drawing)
and monitO!ing - Startup and braking ramj) (slope) - Reference poinllravel
T
f\-..
'V
ecltnological data processing (PLC)
fl -
U
Execution of
ru:;ableIOQiC
fl
U
- Machine monitoring
Figure :U. Examples of function blocks in numerical control.
Manufacturing Systems • 2 Control Systems
(interpolation), setpoint adjustment (slope) and position control are required when adjusting the position of an axis. The next section (L2.4.6) goes into these functions
in greater detail.
2.4.6 Position Adjustment Position Setpoint Generation Position setpoints are obtained for the individual axes of the controlled system from geometric input data in numerical control. Distinctions are made in relation to the kinematic sequences between three types of control system, namely point-to-point control, line motion control and continuous path control. In pOint-ta-point control the point defined by the setpoint may be approached via any path at will, as the tool is not operative during startup (Fig. 24a). The shortest path is nonnally selected to save time; the geometric shape of a tool only affects the traverse path in exceptional cases. This type of control constitutes the simplest fonn of numerical control and is nonnally employed in applications with drilling and boring machines, spot welding machines and tooling machines for electronic components. Line motion control is related to point-to-point control, the difference being that the tool may be operative during installation. The sequence of motions is thereby carried out parallel to the axes of motion of the machine and the operating speed may be predetennined (Fig. 24b). In special cases, parallel operation of two or three axes at the same speed is possible (motion of less than 45°). Where processing of arbitrary two- or three-dimensional curves is required, as is normally the case in milling machines, turning machines and flame-cutting machines, for example, continuous path control is employed. Its characteristics are comparable to those of tracer control. Where a tool traverses from point A to point B, for example (Fig. ~u), the tool follows the function shown, where y ~ f(x), and is operative. The relative motion between workpiece and tool is thus continuously variable in tenns of size and direction. The slide motion is therefore to be controlled at a minimum of two coordinates during processing. Position setpoint generation (interpolation) for continuous-path control is explained in greater detail below. The input data required for the desired traverse path are already present in digital fonn in NC machine tools with continuous path control, as has already been mentioned. The position setpoints are generated from the geometrical and motion data as position reference variables in the fonn of a finely stepped function of path/time by
the interpolator, which is a computer device. This function is converted by the position control system, which adjusts the individual machine slides in accordance with the position reference variables (Fig. 25). The path produced by the position reference variables is primarily dependent on the interpolation process (single-stage, two-stage), the interpolation increment and the interpolation calculation process. These three influenCing factors are detailed briefly below.
Interpolation Process. In Single-stage interpolation the interpolation points are calculated directly as reference variables for the position control. There is a drawback in this as regards the calculation input required for non-linear interpolation, e.g. circular interpolation. This may be reduced significantly by twa-stage interpolation. A rough interpolation calculation is carried out initially when using this method, where interpolation points are generated at larger intervals. This is subsequently followed by a detailed interpolation calculation in the fonn of a simple linear interpolation where the intervals are cut by half (Fig. 26). A polygon that is preselected in this way is smoothed by the low-pass characteristics of the drives and does not, therefore, produce a continuous curve. Interpolation Increment. Interpolation may be carried out in the fonn of:
- fixed time Increments: the path to be travelled is predetermined for each interpolation cycle;
- fixed path increments: the interpolator outputs individual path elements in the fonn of the smallest unit of travel in accordance with the prescribed speed.
Interpolation Calculation Process. The following processes may be employed: step search process, digital differential analyser (DDA), direct function calculation and recursive function calculation. Both the step search and DDA processes are inter-
polation processes with fixed-path increments and require standard measuring systems, nonnally consisting of special hardware interpolators for increment lengths in the resolution area, and are now outdated. Direct or recursive function calculations are both interpolation processes with fixed time increments and may be implemented relatively simply using microprocessors and are, therefore, in popular use. The latter process provides for relatively simple calculations, although
y
Figure 24. Types of NC control: a point-to-point control, b line motion control, c continuous path control.
Figure 2:5. Signal flow diagram for generating relative motion between workpiece and tool.
2.4 Numerical Control (NC) • 2.4.6 Position Adjustment Path segments
Rough
I
Thus, the parameter Llr = (v 8 /s)LlT applies, where LlT is the interpolation cycle time and I' is the overall travel time.
I
Conversion of Spatial Coordinates into Axis Coordinates
interpolati~
Path intermediate points
I
Geometrical programming is normally carried out in the
~tailed interpolation
x, y, z coordinates of the Cartesian system. Where the axis
I I I
Position reference variables
Y
x Example: Rough interpolation - circular and linear. Detailed interpolation -linear Path at x-y level • Path segments x Path intermediate points o Path interpolation peints lcorrespending to discrete time position reference variables)
coordinates are not identical to the main coordinates of the Cartesian system, the relevant conversion from a spatial to an axis system of coordinates is carried out by the control computer. This normally takes the form of a matrix operation. The conversion should be carried out within the interpolation cycle and may involve substantial calculation for multiple-axiS machines, such as five-axis milling or six-axis robot guidance. This is why the principle of rough interpolation is frequently employed with the three-dimensional system of coordinates and detailed interpolation is used in axis coordinate systems. The flow of data from interpolation in spatial coordinates (Cartesian coordinates) to the reference variable of the individual axes may be seen in Fig. 28. Where conversion is omitted altogether owing to the amount of calculation involved, and interpolation is only carried out in the axis coordinates, substantial discrep-
Initial value
Figure 26. Principles of two-stage: intt:rpulation.
Target value
Intermediate points in interpolation
Spatial coordinates i
I I
-II
the calculation has to be carried out with a greater degree of accuracy because of error propagation. The calculation of interpolation intermediate points using the two-function calculation processes is illustrated in Fig. 27, based on the example of two-dimensional linear interpolation (also known as straight-line interpolation). If a parameter T is increased in the interpolation cycle I' by one increment Llr in each case, this
Llr/l = MIT.
VU
= siT
I~
I~
I", IIII
-j-
I Axis coordinates
produces a constant path speed of VB in a Cartesian-based working space. The size of the increment 6.7 is proportional to the programmed path speed VB and inversely proportional to the spatial traverse path s:
s = -fa; + bi,
Iill
Conversion
1
Initial pesi- I tion setpoinl~
Interpelalion in spatial coordinates
I
Initial value
I
I
Spatial coordinates
I
I I I
1 I
I
i
Axis coordinates
x:xo+a,r y=yQ +b o~r~1
"
NC record
1
Conversion
Xa
Target value
Intermediate peints in interpolation
- - - - 1 - - - - - - - - - 1I-
Unear interpolation
Y
Direct calculation:
Position setpcint
x,
x
Recursive calculation: xn",r::xn+o, z1 r Yn.'= Yn+b, Ll r
Figure 27. Function calculation of linear interpolation.
1
1
1
____ 1- ________ 1_ "1 peSI-. I1 11 1mtla 1
tion setpcin~ts.L1.ll-.ll-J..L.l.LLi-U-"--'
Target position setpeints
Interpelation in axis coordinates [] €I Converted coordinates o 0 Given coordirl9tes Figure 28. Flow of data on interpolation to spatial coordinates and to axis coordinates.
Manufacturing Systems • 2 Control Systems
Machine components
Position control Setpoint value generation
-
'II
~ansmission units, spindle, ~lide Direct
Actual position Controller
Feed drive
Integral contact
potential acceleration Ky. The Ky factor in tum determines the following error, e.g. via the relation Asx = Xl Ky 1 for the x-axis with ~s, in relation to the velocity Xl = constant. Where this following error As, is compensated by suitable circuitry, the additional following error for constant acceleration Xi = constant, where dsx = xjK;} TAl may be calculated. As can be seen, the drive constant only has a direct effect on the following error in acceleration procedures. These relations enable the effects of a follow· ing error to be derived under a variety of influencing factors for linear travel in a Cartesian configuration. Contour distortions in linear travel may then be avoided, if both the acceleration factor Kv and the drive time constant TA are equal in both axes.
2.5 Equipment for Position Measurement at He Machines Equivalent to :
where ~nd
Figure :19. Simplified model of closed·loop position control axis.
Position-measuring systems in NC machines register a given line motion as an analog geometrical figure and produce this in the form of a digital position value. These are essential components of the position control circuit and their accuracy also helps determine the manufacturing quality of a machine tool. Their structure consists of a material measure, e.g. in the form of a scale, readout and analysing unit (cf. terms employed in instrumentation [24, 25 D, The following terms are considered to be particularly significant: resolution capacity, accuracy, sensitivity towards external influences and expansion capacity. 2.5.1 Types of Position Data Registration
ancies are likely to occur compared with the desired three-dimensional path (linear, circular).
These measuring systems vary depending on their particular characteristics (Fig. ~O).
Position Control
2.5.2 Measuring Spot and Data Sensoring
The relative motion between the tool or measuring instru· ment and workpiece is effected by the primary control motion of a minimum of two axes in continuous-path-controlled NC machine tools [11]. Figure 29 shows the structure of a c1osed·loop position control axis using a simplified control structure chart, where the drive is shown as a system of the frrst order and the position controller is typically designed as a P-controller with the acceleration Ky. In order to prevent excessive oscillation, damping of DI. = 0.7 is recommended for the position control circuit described, i.e, the drive time constant TA dictates the
The distinction between direct and indirect measurement depends on the position of the measuring spot [18]. The measuring system is located directly on the machine slide in direct measurement. In indirect measurement, intermediate contacts transmit the change in position or line motion to a measuring system which is normally of a rotational design. Intermediate contacts may constitute the main spindle with a measuring gear and rack-and-pinion gearing, where necessary. Indirect measurement is often carried out on constructional or cost-effective grounds. Poor machining accuracy in the intermediate contacts and changes in size due to
Characteristics
Position and path measurement
Data logging Measuring process Measuring spotdata testing Measuring system
Figure~.
Types
of position and path
measurement: D direct, I indirect, R rotary, T linear.
2.5 Equipment for Position Measurement at NC Machines. 2.5.3 Digital Data Logging
temperature influencing factors have a direct effect on results of measurements, however. This is why direct measurement is more accurate than indirect measurement. Options for installation in-line motion or position measuring systems and error-influencing factors are shown in Fig. ~l. Error-influencing factors are basically always smaller, the better the measuring apparatus satisfies the conditions of the Abbe coefficient (after Abbe, 1890). According to this, the sample and comparative line motion should be arranged in alignment with the direction of measurement. The alignment angle error is then only of the second order in the measuring result. :!.5.~
Digital Data Logging
In digital data logging, the dimension to be measured is in a quantitative form. A distinction should be made between two different function principles, namely incremental and absolute measuring systems. The functional principles of incremental measuring systems are discussed first below.
Digital Incremental Measuring Systems The digital incremental measuring system is based on the subdivision of the path into parts of equal size (increments) and the principle of incremental dimensioning. The measurement of a line (of an angle) is carried out by adding the increments on an incremental grating or incremental plate detected by a scanner device in relation to direction. To distinguish between two adjacent increments, the individual increments are alternately
Indirect data logging
Direct data logging Measuring system operates in linear direction and is directly coupled wnh the longrtudinal travel of machine slide.
assigned different physical characteristics. The scanner device then provides the so-called count pulse from the relative motion in terms of the increment which is added up by the counter. The majority of incremental measuring systems use photoelectric pulse transmitters. In addition, pulse transmitters with magnetic sensors are also employed (e.g. gear wheel and magnetic pulse transmitter) where the requirements for resolution are not high. There are a number of different constructions of photoelectric pulse transmitters, namely transmitted-light operation, front-lighting operation, polygon reflecting operation, and multi-prismatic operation. The main functional units of these measuring systems include incremental scale, grating, lighting unit, counter device and direction discriminator. The incremental scale is characterised by a regular series of markings. Depending on the design, these relate to light-permeable and non-light-permeable, reflecting and non-reflecting fringes. The pitch (grating constant) may be extremely finely stepped (up to 1 f,lm). The grating forms part of the scanner and determines the form of the signal to be evaluated. There are three types of grating (Fig.~:!). In the first form, the grating pitch is the same as the scale pitch and is parallel to this. When in motion the entire grating area alternates between light and dark. In the second form, where the grating area is positioned at an angle, Moire fringes are generated. In the third form, the grating pitch is not the same. The light and dark zones occurring run lengthwise to the direction of scale. The same aim is achieved with all these embodiments,
Drive element and transformer separate Conversion from longitudinal travel to rotary motion via gear rack and pinion
Pinion
Drive element and transformer identical Longitudinal travel logging via rotary motion of operating spindle
measuri"9 spindle
Gear rack
Error-influencing factors
Error-influencing factors
Error-InfluenCing faclors
Temperature Pitch errors Distance and angle errors Errors at intersection points of scales
Pitch errors of gear rack and pinion Eccentricity of pinion Errors in any transmission unITS employed Temperature Pickup error Sudden load change in gear rack
Elasllc deformation of spindle Angle errors Play Obstruction of the spindle Pickup error Temperature
Figure "1. Error-influencing factors in direct and indirect data logging [21.
Manufacturing Systems • 2 Control Systems
_: III/
O''lITlIIO'l71I'
ically with any motion in the scale in relation to the scanning unit. These signals are displaced by 90° (= t pitch period or grating constant) in relation to one another, in order to be able to carry out directional recognition. After preamplification, the two analog measuring signals (between 1 : 4 and 1 : 64) may also be subdivided via a voltage-multiplier connection. Two right-angled pulse sequences displaced electrically by 90° are generated from the analog signals in a transducer switched downstream. Directional recognition and pulse counting are carried out in a counter circuit with directional logic for display or processing purposes. The counter level provides a measure for the path travelled in relation to the point of reference after completion of the reference point travel. The maximum traversing speed is limited by the maximum counting frequency. In addition to the incremental measuring system described for linear motion, there is one for rotary motion which operates on the same operating principle. For standard applications the revolution counters have between 1000 and 3000 lines per revolution and for precision applications up to 3600 lines are possible, providing an angular resolution of up to 0.5".
Digital Absolute Measuring Systems
nb
~ ~~~~~~ ~~~~~
~l
b
a
Figure 32. Designs of gratings in incremental measuring systems: a construction; b output signal; 1 scale, 2 gratings, 3 light/dark, 4 Moire fringe, 5 vernier fringe.
namely that the fmely graded light/dark line structure of the scale is converted into large areas of light/dark zones. This is needed for scanning with the light-sensitive elements.
Transmitted-Light Optical Diffraction Systems The method of evaluation is explained in detail here and applies for all transmitted-light systems. The light travels from the light source via an optical element which generates parallel light through the grating and the grating plate on the photoelectric elements (Fig, 33). The photoelectric elements generate virtually sinusoidal signals period-
2
3 Figure 33. Operating principle of transmitted-light optical diffraction process: 1 light source, 2 condenser, 3 scale grating, 4 reference mark, 5 scanner grating, 6 scanner disc, 7 silicon photoelectric diodes.
In digital absolute measuring systems, a single measured value based on a fIxed datum is assigned to each line motion component. These data are marked from the datum onwards by a uniquely recognisable code word. The datum is set mechanically. A datum offset may be carried out mechanically or by adding a sum to the position measured value. A linear encoder serves as the material measure in linear position measuring systems and a rotary encoder is used in rotational measuring systems. The physical possibilities of measuring the codes and the scanning process correspond to those employed to generate increments in incremental systems. Photoelectric scanning is used in preference in this area as well. However, there are problems with ambiguity in scanning the increment in this instance, owing to the simultaneous change of state in a number of tracks on transition of data from one measured value to another. There are three ways of preventing this: 1. 2. 3.
introduction of an additional clock track; use of single-step code; use of double scanning or V-scanning.
Regarding I, the additional pulsed track ensures that the measurement reading is only enabled when the transition has defInitely been completed in all tracks. Regarding 2, the Gray code is the best single-step code. It has the advantage that only one signal change takes place on transition from one increment to the next one in line. Regarding 3, double scanning is employed in preference in dual or binary-decimal encoded increments, in order to render pitch errors in the increments and the scanning arrangement ineffective. Two scanners are installed in each code track, apart from the fmest-stepped, one of which is employed as the reference track. The advantage of the dual system is exploited for this purpose, where on generation of the signal L in a track the right datum line is always critical in the subsequent track. Where 0 is generated in a track, the left datum line is always critical in the subsequent track. Two scanners are, therefore, always arranged in the subsequent track and are displaced by half the pitch width
2.5 Equipment for Position Measurement at NC Machines. 2.5.4 Analog Data Logging
of the preceding track (Fig. ~). This generates a field of tolerance which is large enough to allow for the transition from one value to another in the tracks. The scanner used in each case is selected from the preceding track. This V-scanner is more costly to employ than the Gray code, yet the tolerances of the tracks may be greater where the values are higher. This advantage is extremely
important for rotational measuring systems with rotary encoders switched in sequence owing to the precision of the intermediate gearing.
2.S.4 Analog Data Logging Analog data logging is characterised in that a measuring signal value may be assigned to each measured value of the data volume continuously. In the Simplest case, the change in resistance is employed in relation to the length of the electrical conductor to generate an electrical measuring signal. In practice, potentiometers are switched as voltage dividers. Measuring devices of these types (linear or rotational) are only employed in special cases, e.g. for rough positioning as they are not wear-free and their resolution capacity is limited. Their linearity is not normally higher than 1% either. Inductively operating measuring systems have become more popular for analog pOSition and line motion measurement , i.e. in the form of synchro-resolvers (or rotary transducers) and Inductosyn scales.
Synchro-resolvers These are rotational measuring systems where angles are registered inductively without contact being made. They normally consist of a rotating transformer with a rotor (impeller) and stator (column) (Fig. ~S). The arrangement shown in Fig. ~Sa has a single-phase winding stator and rotor in each case. This has no practical significance, although the input and output voltage proportions are the most clearly represented in this casco Where the alternating voltage is II, = U, sin wt at the stator winding, the magnetic flux generated in the rotor winding induces an amplitude-modulated voltage of the same frequency, where Il, = (U, sin wt) cos a = If, sin wt. As Fig. 3Sa shows, the time -related amplitude modification is modulated by the angle-dependent change in cos a. The envelope reflects the respective angular position and the modulated voltage causes a phase shift of 180~, at the voltage zero in this envelope. This produces a clear relation between amplitude and angle.
Figure ~Sb shows a synchro-resolver with a singlephase rotor winding and two stator windings physically displaced by 90° as an example of the many possible winding arrangements and types of circuitry available in prac-
wi
Figure 3S. Synchro-resolver (principle); a with stator coil, b with two stator coils.
where a cos and a sin voltage are applied to the stator windings:
u 2 = IU, cos
+
[U, cos
(a +~)
cos wI
= [U, cos a] sinwt - [U, sin a] coswt = U, sin (wI - a)
The voltage at the secondary winding changes constantly with the spatial angle in the phase position, compared with the voltage at one of the primary windings. A phase discriminator produces a signal proportional to a .
Inductosyn Scale. This is an inductively operating scale based on the synchro-resolver principle. Squarewave divisions are llsed on the scale on the printed circuit
board with Inductosyn systems (Fig. ~6). A contactless slide with two square-wave windings spatially displaced by 90' back-to-back moves above this. This device operates in exactly the same way as a two-phase
tice.
The voltage induced in the rotor winding is as follows,
A
ro
1011 12 131415 16171 819 m l nIRIDIKI~I~IV ~u I I I I I I I I I LI 1'd. lPl I t I I I
2' ~
3' 4' C;== ~X,
0"0 D "l Figure~.
--
'0...2
~3 \'
Q4
X,
--1
X
Double scanning for binary-coded absolute encoder.
Figure ~6 . Principle of Inductosyn scale system. a linear scale !J, rotor in synchro-resolver. b C\.lrsor !J, stator in synchro-resolver.
Manufacturing Systems • 3 Shearing and Blanking Machines
synchro-resolver, as the two coils which are both displaced by 90° within the pitch also generate a specific field vector. Inductosyn scales all have pitches of to" or 2 mm. Resolution is thereby provided in the lower micrometre range. Absolute measurements can only be carried out from within the pitch.
2.5.5 Laser Interferometer The laser interferometer is an extremely high-precision measuring system for scanning and ranging machine tools, i.e. scanning, ranging and calibrating scales, scanning and ranging the accuracy of roller ball spindles and rack gears, etc. It is also employed in high-precision and large-scale machine tools and as a position measuring system [20]. In a Michaelson interferometer a light source emits a coherent monochromatic beam of light of the same frequency and phase length which is split into two parts by a semitransparent mirror. The two beams are reflected onto two mirrors with total reflection, where one mirror
is fixed and serves as a reference line and the other is movable and is located on the device under test. Where the reflected parts of the beam meet they interfere with one another. This causes a reciprocal attenuation or amplification according to the respective mutual phase position in the photoelectric receiver. Pulses are formed from the signal received, where their number is proportional to the path. The precision of the laser interference process depends considerably on the stability of the light source length, which itself is affected by ambient conditions, such as air pressure, temperature, humidity, CO 2 content in the air and the operating condition of the laser. There are various ways of increasing the stability of the waves, which are particularly well suited to the dual-frequency laser process. A measuring error of approximately I fLm/m is achieved with such stable laser interferometers, and linear speeds of 18 m/min may be recorded. A resolution of 5 X 10-3 fLm may be achieved via frequency multiplication.
Shearing and Blanking Machines K. Siegert and T. Werle, Stuttgart
3.1 Shearing Machines Plate Shears. These are used for shearing strips or making linear blanking cuts in sheet metal. A blankholder with appropriately shaped cutting blades top and bottom is used to drive one or both cutting edges to generate cutting surfaces which are free of flash where possible and which run at right angles to the sheet metal (Fig. 1).
Figure 2:. Functional diagram showing principle of strip shears.
Where the top cutter runs parallel to the bottom one, the
cut edge produced is slightly angled. Angled or oscillating cutters improve the right-angled shape of the cut surface. Crank mechanisms or toggle drives and variations on these constructions are employed as gearing in these machine tools. Hydraulic drives are also used. Nowadays, CNC-controlled machine tools are available for all types of drives where the cutting angle, blade clearance and maximum cutting power may be pre-programmed.
Strip Shears. These cut strips in continuous operation (Fig. 2). They may be used in applications with sheet
thicknesses of up to a maximum of 6.5 mm and minimum widths of 40 mm per strip.
Circular Shears and Curved Shears. These are designed for cutting along curved lines. The diameter D of the cutting tools (Fig. ") must not exceed a specific threshold limit value (D < 120so), which represents the sheet thickness, because of the flexibility required for extreme curves.
Billet Shears and Billet Parten. These are used for the production of unrnachined parts, e.g. for drop forging. In this case, the emphasis is placed on the production of a constant volume of cut, unmachined parts. A complex measuring and control device is required for setting the feed length stop owing to the variations in the thickness of the billet. Shears for Unmacltined Parts. Used specifically for processing unmachined parts in cold forging, these oper-
.] Figure 1_ Plate shears. a Parallel top cutting edge. b Angled top cutting edge. c Oscillating top cutting edge. dAngle 00 shearing
blade.
Figure 3. Functional diagram showing prinCiple of curved shears.
3.3 Nibbling Machines
2 3
a
b
Figure 4. Shears for unmachined parts: a open cutting edge, b closed cutting edge; 1 flank, 2 support, 3 cutter, 4 wedge angle, 5 blade clearance.
ate at a relatively high number of strokes per minute. They shear off either the unmachined parts using rolled wire, or rolled bars with an open cutting edge (Fig. 4), or cropping bars with a closed cutting edge. Crank mechanisms are employed as the standard drive.
3.2 Blanking Machines The typical force/path curve (Fig. S) illustrates the requirement for machines with high-rated cutting forces even for relatively small machining capacities, although the energy capability has to be greater for blanking sheet metal with higher ductile yields, owing to the longer cutting distances involved, than with sheet metal with lower ductile yields and the same stability. High-speed crank presses are employed for blanking and even small-stroke, high-speed, toggle-joint presses and hydraulic presses with stroke arrestors are used in special-purpose applications. Cutting impact damping is carried out as an additional process in the interests of noise abatement in blanking machines, where the fall-through of the slide is prevented at the end of the process. Stroke rates of up to 800 min-I are normal for mechanical automated blanking machines. Automatic slotting machines with toggle mechanisms which oscillate around their line position achieve stroke rates of 1300 min-I. These therefore require correspondingly high-performance and accurate feed mechanisms (feedrate accuracy ± 0.01 mm, average continuous speed
of up to 120 m/min) in conveyor operation or for precision part mechanisms for cutting slots in sheet metal
a Figure 6. Alternative slideways in high-speed presses. a Schematic diagram of slideway above conveyor level. b Schematic diagram of slideway at conveyor level.
adjustment of the insertion depth of the cut. Insertion depth of cuts, automatic slide stops at the upper dead centre and feed lengths are monitored by computer, whereas band thickness, bandwidths and tool parameters such as the band feed height, feedrate accuracy of the roller feed mechanism and cutting power are monitored by the machine tools (see Fig.8) [1]. The latter also provides information on tool wear by comparing actual and setpoint forces. Presses should have as rigid a construction as possible in order to minimise wear, and the slideway guide should be optimised. High-speed presses are particularly suitable for use in flexible punching centres. Table 1 shows industrial punching techniques classified according to characteristics such as accuracy, areas of application and stroke rates [2].
3.3 Nibbling Machines These are available in all sizes, to incorporate all the various intermediate stages of automation, from manual nibblers to numerically controlled sheet metal working centres. In the case of stationary machines the workpiece
stators Of rotors of electric motors. Modern precision auto-
matic punching machines can cater for stroke rates of up to 1800 min-I with a rated cutting force of 200 kN thanks to the use of CNC controllers. Such stroke rates are only possible with optimised slideways (where the slide is guided at conveyor level to prevent tipping - Fig. 6), mechanical balance (compensation for dynamic forces in high-speed presses - Fig. 7), and complex sensory analysis (monitoring all essential influencing parameters) and
Figure s. Force~path curve during blanking: 1 precision blanking, 2 sheet metal blanking with high ductile yields, 3 sheet metal blanking with low ductile yields.
Figure 7. Function of mechanically balanced automatic punching machine (Bruderer), with four-point suppon of slide and punching force reduction on eccentric cam. TIH = tool installation height.
Manufacturing Systems. 4 Presses and Hammers for Metal Forging
Table 1. Classification of industrial punching techniques [2] Punching techniques
Characteristics
High-performan<:e punching Punched parts from strips of 0.1 to techniques 3 mm thick, high accuracy, high number of strokes (up to 1800 min-I), extremely high piece numbers (e.g. plug-in contacts, electronic system data carriers, laminated rotors and stators)
Figure 8. Diagram of eccentric press with feed mechanism; setring adjustment is via numerically controlled actuators. F compression force, H stroke, e eccentricity, 1 roller feed, 2 slide, 3 eccentric bush, 4 slide strike adjustment, 5 eccentric cam.
is moved in relation to the tool [3]. Sheet metal working centres have a tool data store from which a variery of standardised tools may be selected at will for automatic operation. The numerical control system generates control commands for moving and positioning the workpiece via the positioning table, switching commands for initiating and repositioning strokes (individual continuous strokes) and commands for tool changes. The body of the machine consists of a protruding throat-rype frame, or occasionally a straight-sided press (see LI.3). The slide is normally driven by a crank mechanism.
3.4 Beam Cutting Machines These generate individually programmed geometrical blanks or unmachined part cuts (see K4.5.6) using a beam
Conventional punching techniques
Punched parts from strips of between 1 and 6 mm thick, average accuracy (e.g. machine components, ftxtures and fittings), average number of strokes (up to 400 min-I)
Precision blanking & punching techniques
Punched parts with smooth cutting surface (R, ~ 0.8 to 2.4 fLm) of between I and 15 nun thick, high degree of accuracy and wide variety of shapes, including threedimensional (e.g. coupling halves, trip levers, gearwheel sections)
Nibbling and laser blanking techniques
Punched parts from 0.5 to 3 mm (and thicker) sheet metal or blanks for lot sizes of "one" or more, predominantly large-scale, bulky parts. Table motion CNC-controlled (e.g. blanked sheet metal casings)
Large-scale punching techniques
Punched parts of 0.3- to 2-mm-thick sheet metal blanks in large sizes (e.g. coach parts)
of bundled active media or active energy. They frequently form elements of a manufacturing station, and are specifically suited to small and medium-scale blanking operations owing to their high degree of flexibiliry. Water torches can produce tolerances of 2: 0.1 mm, while laser cutting tool equipment may have tolerances of up to 2: 0.005 mm.
Presses and Hammers for Metal Forging ]. Siegert and E. Dannenmann, Stuttgart These are designed to activate forming tools, carry out reciprocal guiding of tool parts and provide the deformation forces, moments and energy input requirements for the process. Figure 1 shows the forming machine classifications. Presses are the most frequently employed machine tools for unit loads and are here dealt with first of all.
4.1 Characteristics of Presses and Hammers The characteristics describe the attributes of a forming machine. Compared with the requirements of the fonning process, these enable the most suitable machine tool to be selected for the relevant process. Three groups of characteristics are relevant for presses: energy and force
characteristics, time characteristics and preciSion characteristics (Table 1). In addition to these characteristics and their numerical values (characteristic values), there are also mechanical data which are essential for press applications, such as the height of lift of the slide or pile hammer, the dimensions and design of the installation area inside the machine tool, the connection ratings, the space requirement and the weight. Construction sizes have been standardised for a wide range of presses (DIN 55 170, DIN 55 181, ISO/DIS 9188-1989, DIN 55184, ISO 6898/1984, DIN 55185, ISO/DIS 9189-1989, DIN 55222). The main force and energy characteristics associated with forming machines are the slide force f~, and energy capabiliry EM' These values always have to correspond as a minimum requirement to the forming force F and fonning action W required for the process, in order that the process may be carried out on the machine. In addition to
4.2 Mechanical Presses
Figure 1. Classifications of forming machines (D continuous process, E plunge process).
fonning forces and action, additional force and action may also be required for the operation of additional aggregates, such as drawing machines, blankholders, ejectors, etc. Spring action stored within the machine-tool system should also be taken into account, depending on the process. Presses are distinguished according to the way in which the machine generates the characteristic deformation forces and energy: continuous-path control, adaptive control and closed-loop machining (Fig_ :1) .
Time characteristics describe machine-dependent process times and speeds, such as impact and stroke sequence time, pressure contact time and tool speed. Precision characteristics provide a reference for the workpiece speeds which may be achieved with a forming machine. There are characteristics for no-load machines (accuracy of manufacture) and machines under load. These guidelines for accuracy of manufacture relate to the geometric dimensions of the installation area inside the machine tool and accuracy of motion of the slide and are specified in DIN 8650, ISO 6899-1984 and DIN 8651 for continuous-path control presses according to the type and size of the relevant machine tool construction. Accuracy characteristics of machines under load are defined in DIN 55 189 for mechanical (continuous-path control) and hydraulic presses, and describe the offsets in the tookarrying surface areas under load compared with no-load conditions. An offset is only generated in the direction of operation (v,o") under average load in presses with symmetrically structured frame constructions (O-frames) and driving
mechanisms (Fig. ~). This is composed of the initial offset v'z, generated by clearance and load-related elastic deformation l'dZ in the
individual press components (Fig. 4). Rigidity Cz is one of the accuracy characteristics and is obtained from the linear section of the offset curve, where Cz = /li'z/ c;'VCIZ ' Eccentric loading (Fig. S) leads to tilting between the table and slide vertical to the direction of operation and a shift in the centres of the table and slide (offset), irrespective of the design of the frame. The overall tilt k"" is composed of k, (mechanical balance of guide clearance) and elastic tilt k" (frame, slide and driving mechanism deformation) (Fig_ 6). Breakdown rigidities C kA and C kB about the x- or y-axis are obtained as accuracy characteristics from the linear section of the tilt curve at CkA = /li'z . MY/ tJ.kelA , or C kB = /li'z . c;'lx/ MelB (Mx, tJ.ly eccentricity of load application in X, Y direction, set at 10% of the productive slide depth or width). The characteristic for the offset vertical to the direction of operation (shift) is the distance of the central vertical line of the slide in relation to the central vertical line of the table, measured as half the distance between the table clamping area and the slide area (Fig. S). This is obtained as the overall offset V lOtX == VaX + VelX, or V tOtY = vaY = VelY at a force under load (Fz ) of 50% of the rated force (FN ) and an eccentricity (C;.lx(Y)) of 10% of the productive slide width or depth (Fig. 7).
4.2 Mechanical Presses In continuous-path control presses (Fig. :Za) the machine slide runs through a path prescribed by the kinematics of the main gearing. The latter is driven by an electric motor via a flywheel and clutch. Intermediate gears may be arranged between the flywheel and main transmission unit. The force emitted by the slide F" is dependent on the slide setting h. The dominant characteristics thereby
Manufacturing Systems • 4 Presses and Hammers for Metal Forging
Table 1. Characteristics of presses Characteristics:
Force and energy
,J! F~
w,
Slide force, energy produced by machine at any point in process Rated force, force determining design of machine Force on impact, compression force at highest speed on impaCl excluding effective energy losses (closed-loop machining presses) Maximum admissible force (in continuous operation) (spindle presses) Energy capability Rated energy capability, maximum amount of energy available for operating cycle Spring action, potential energy stored in machine and tool in operating process
i
Q)
~~--------~
:!2
en
LL
i
g'
j
Top Forming path 5
a
,J! 2l
Bottom
-
Slide path h
Rated force
FN~pA
.Ee-----L------------, Q)
:!2
en LL
~
.E g'
§
~
Time to impact, stroke sequence time, duration of pile hammer or slide stroke before machine is ready for next stroke No. of impacts, no. of strokes, reciprocal value of stroke sequence time, in presses with crank mechanisms equal to crankshaft speed n K Tool speed at any point of process (normally equal to slide speed lJSt )
Time
Accuracy
No load machines
Machines under load
Flatness and paraUelism of tool clamping surfaces, right-angled slide motion in relation to table surface C z Spring rate in operating direction CkA , C klJ Stable rigidity tJ tot, tJ totV Offset vertical to direction of operation (shift)
b
Forming path 5 _ Slide path h W~3·EN
1st cycle
2nd cycle
3rd cycle
Forming path 5 Figure 2. Principles of presses: a continuous-path b adaptive control, (: closed-loop machining.
21 L-. __ _
FZ
consist of the curve of the slide force in relation to the slide path, i.e. Fs, = F,,(h), and the relevant maximum admissible value, the rated force F N , which the parts affected by the flow of force are designed for. The energy requirement of an operating cycle is almost exclusively covered by the energy released by the flywheel. The energy capability is a further significant characteristic and is determined by the design of the flywheel and operating mode, i.e. the maximum available rated energy capability EN in continuous-stroke operation. In single-stroke operation a higher energy capability EE = 2EN may be employed owing to the relatively brief machining cycle and, consequently, the lower thermal loading of the drive motor.
~~~~~~~~~
control,
-1\
~otZ
f77:7777;~'77777/l-~/ i 1
Figure 3. Offsets in symmetrically structured press frames and centre loading (DIN 55 189): I table, 2 slide, Fz force under load, V totz overall shift between table and slide.
4.2.1 Types
Presses are known as crank or cam mechanisms (Fig. 8) depending on the type and structure of the main gearing. Cam mechanisms are limited to a small rated force F N ; nevertheless, they provide virtually unlimited possibilities for the sequence of movements. Crank mechanisms are
Figure 4. Offset V z as a function of the force under load Fz (offset curve, DIN 55 189).
4,2 Mechanical Presses. 4,2,2 Assemblies
Type of gearing
Simple
Structure
Slider-crank mechanism
Slide-crack toggle mechanism
Figure S. Tilt and offset vertical to the direction of operation in eccentric loading (DIN 55 189): 1 operational side, 2 slide, 3 table,
Complex
+k
k08 (kaA ) +---+---------------~----------
Swivel mechanism
ka8(k aA )
kel8 (k,IA)
-k Figure 6. Tilt k as a function of the force under load Fz with a given eccentricity A/x(y) (tilt cunres, DIN 55 189); kaA' kaB =
Cam mechanism
initial tilt.
Figure 8. Structures of main gearing.
_ _ _¥C,/'-_ _
L~!~~~~~g!l!kea~) I
Vax(VaY)
+---4---------------0-5-~~----------F~z
-J,oadaPplication lett (frOnt)1
~::-
I
I I -Vi~xlY)
Figure 7. Offset VtotX(Yl vertical to direction of operation as a function of the force under load Fz with a given eccentricity 1l1x (y) (DIN 55 189); v..x, vaY = initial offset.
divided into the simple and the complex, The popular such presses are those with slider-crank mechanisms, either with a fIxed overall stroke (crank presses) or a variable overall stroke (eccentric presses), Complex crank mechanisms are employed where a large F" is required with a small stroke (toggle mechanisms) or where reduced machining speeds are desirable in the working space (swivel mechanisms),
4.2.2 Assemblies Frames (Fig. 9)-
Clrames come in single- and double-column design, and may be vertical, inclinable, horizontal or in some cases
Figure 9. Shapes and constructions of frames for continuous-path control presses. a C-type frame, single-column construction. b ctype frame, doub1e-column construction. cO-type frame, two-column construction. dO-type frame, pillar-type construction.
have tension rods, These are normally used for presses with small-to-medium-scale rated forces,
O-frames may be of two-, three- or four-column design with penetrations in the side columns for tool changes and workpiece supply and discharge, and are occaSionally found with pillar design; these are medium-sized presses, normally of a unitary construction, For large-scale presses, multi-part two-column frames are employed, with a table, side columns and cross-beams connected via tensioning rods, Frames may be of cast-iron, cast-steel or, increasingly nowadays, of welded steel plate construction, Drive. The flywheel is normally driven via three-phase asynchronous slipring motors, The stroke rate adjustment for presses with small to average rated forces is via gears
Manufacturing Systems • 4 Presses and Hammers for Metal Forging
between the drive motor and flywheel, while in presses with high rated forces it is via variable-speed drive motors.
- ro-, 1-"', '-"", =
UN"75'
..
Clutc:h, Brake. Nowadays, clutches are mainly of the friction-locked type (friction clutches), and of single- or multiple-disc construction (F3). The forces brought under load are normally applied via a pressure medium (air, or occasionally oil). Brakes are basically similarly in design to clutches. The forces brought under load are generated from spring mechanisms, for reasons of safety. Positive clutches (rotating wedge-and-pin couplings) are today becoming less popular because of current safety requirements. 4.2.~
\
\
"-
1""'-
i'-..
03.c
V
0.2 ;}
o
:,
hN/H"032
\ \ I'\. "'Z..J.,.-'-.~
\
\
'-. 0.15
V
1 008
V
\\
\ '-
f'...
""'-
F" has a minimum value (F"mJn) where Ci = 90° (h = H/2, H = 2r overall stroke) and at the end position (Ci = 0°, Ci
~.O
/
V
I II III
./
0.04 002 0.008
./
/
II
~ 15
1.0 09 0,8
'"
~
'" .~
0;
a:
0.7 0.6 0.5 0.4 0]
0.2 0.1
09 0.8 0.7 0.6 0.5 0.4 0] 0.2 0.1 00 Relative slide path hlH
Figure 10. Critical values of slide forces in relation to slide path for various designs of crank press. a Design in accordance with rated angle of force aN' b Design in accordance with rated force path b N (A = 0.1).
H" 0,5 Hmox
1\
\ \
1985 for eccentric presses with throat-type frames, the drive is to be constructed such that the rated force FN at CiN = 30° (corresponding to hN = 0.073 Hm~) is available
adjustment (Fig. 11), as does the slide speed Vso although the energy capability remains unaffected. In slider-crank toggle mechanisms (Fig. 8) with tension or pressure applied to the connecting rod, the slide motion is delayed near bottom dead centre. As a result, the slide forces in the range h > hN are lower than with slider-crank mechanisms of the same design as regards the rated force path hN and overall stroke H (Fig. 12).
---
10'
a
F" = Md(r sin Ci) where MK = the crank torque, r =the crank diameter and Ci =the crank angle.
and magnitude of PSt changes as a result of the stroke
7
1.0 0.9 0.8 0.7 0.5 05 0.4
._-f-- f-- 01
Slide path h and slide speed v" are dependent on the crank angle Ci in slider crank mechanisms (Fig. 8; see FIO.I). This can be simplified, so that the following applies for the slide force F,,:
for the maximum stroke Hm~ (standard design). Further standard designs depend on the area of application include drop forging Ci N = 10°, blanking CiN = 20°, extruding Ci N = 45°, deep-drawing up to Ci N = 75°. Critical limits for slide forces for Ci > CiN or h > hN are provided by F" =FN (sin CiN/Sin Ci) (Fig. lOa). The design specification in more recent standards relating to construction sizes of continuous-path control presses is based on the rated force path h N • For presses with throat-type frames (DIN 55 184), rated force paths are specified in the region of 2 mm < hN < 9 mm. For presses with straight-sided frames (DIN 55 181 and ISOIDIS 9188-1989), requirements for rated force paths of hN = 3, 5, 7, 12.5 and 25 mm are specified. In the latter case, critical limits for slide forces for the various designs are shown in Fig. lOb. The rated energy capability in continuous stroke operation (EN) is normally EN = FN X h N. The overall stroke in eccentric presses is normally variable within the range of Hm~/HmJn = 10. The direction
20'
I
I I
3~' ./'
........
Dynamic:s and Kinematic:s
= 180°) tends towards infinity. For the components in the flow of force, the slide force has to be limited to a finite value, the rated force F N , either over a specific crank angle (rated angle of force Ci N ), or over a specific slide path (rated force path hN)' before bottom dead centre. The force is limited via excess load protection (shear plate, hydraulic die, force sensor element reacting in response to machine control). The magnitude of rated force angle Ci N or rated force path hN depends on the structure and the area of application. According to DIN 55 170 and the standards valid up to
/
45' . /
-......
\ \
II
/
\
-
1\ \
,JAW 1\06'" r--- ./ VIA
0/'-~ 09\ 0.8"-.....
"
/, (Jj 'i,
V
10 Hmox
1.0 0.9 08..c
.~
0.7 '"
0.6~
'"
05~
O.4.~
OJ
03 &' 0.2 0.1
10 0.9 0.8 O} 0.6 0.5 0.4 03 0.2 0.1 Relative slide path h/Hm" Figure 11. Critical values of slide forces in relation to slide path with stroke adjustment (design: UN = 30° for H = Hma:x).
Swivel Mechanisms (Figs 8 and 18a). These provide
low and virtually constant slide speed v s• in the working space. No-load paths are traversed rapidly (Fig. 1~). Higher slide forces are available for h > h~ compared with slider-crank mechanisms with a rated force path hN of the same size (Fig. 14).
4.2.4 Applic:ations Continuous-path control presses constitute the majority of forming machines employed in unit load manufacturing, with a variety of construction requirements designed to suit the relevant individual application. The high rigidity required for massive forming (drop forging, extruding), in view of the machining accuracy
4.2 Mechanical Presses. 4.2.4 Applications
1.0
\
,1
II
1/
\ \ 2
.......
k-'
0.9 O.BoC 07~ 06 ~ 0.5 ~
/
Ul
04 ~ 0.3~ 0,2 0.1
1.0 0.9 08 0.7 0.6 0.5 0.4 0.3 0,2 Q1 Relative sfide palh IiIH Figure 12.. Critical values of slide forces in relation to slide path for slider-crank toggle mechanisms. Tension is applied to connecting rod (1) and simple slider-crank mechanism (2) with the same rated force path h .... and the same overall stroke H (bNI H = 0.073).
Operatill9 range Deep-drawing
: ;/1 .0
).
2\.... t--
-
1
3>- V
.../
,
,
/
I I
I
~I
:;:; , ';i l
250 200 150 100
0
I
50
o
0
Relative slkle palh (mm) IUTi
and pressure contact time, is achieved by way of the shape of the frame (i.e. an O-frame with small column width) in conjunction with the design of the driving mechanism. The main drive shaft in drop forging presses is in the form of a rigid eccentric shaft with a short, wide connecting rod. A wedge press with a slider-crank toggle mechanism is used in extrusion machines. The slide drive in the wedge press (Fig. 1;) is moved by the connecting rod via an interposed wedge (wedge angle 30°). Thus, only approximately half the slide force is applied to the connecting rod. The wedge prevents the slide from tilting as a result of the guide clearance. Wedge presses designed with rated forces of up to 12S MN are employed in the manufacture of long, precision-forged items. In addition to the most popular construction with vertical feed motion, there are also designs with horizontal feed motion for processing work pieces, with long shafts (horizontal upsetting presses) or die-shaped parts (e.g. machines for tube extrusion) providing advantages in workpiece handling. Presses with throat·type and O-type frames are employed in sheet metal forming. In throat-type frame presses (Fig. 16) the working space is freely accessible from three sides. This design is normally constructed with stroke adjustment, as it may be adapted easily to suit a variety of applications as a result (universal presses). Presses with O·frames are normally used for sheet metal fonning owing to their relatively large column widths with mUlti-point slide drive; nowadays the majority are used in cross·shaft drives (Fig. 17). Deep drawing with a blankholder requires a device on the press for activating the blankholder in the tool. The tool blankholder is activated in two-way-action presses via blankholder slides (Fig. 18b) separated from the die slide with step-motion linkage (see Fig. 19 for the sequence of motion of die and blankholder slides), in simple-action presses normally via pneumatic-powered drawing mechanisms. More recently, hydraulically powered drawing mechanisms
Figure l~. Slide speed Figure 14. Critical values of slide forin relation to slide path ces in relation to slide path (Schuler): (Schuler): 1 swivel mech- I swivel mechanism, 2 slider- crank anism, 2 slider-crank mechanism. Rated force path hN =
mechanism.
12.5 mm.
FillW'e IS. Wedge press (ElJMlJCO)
Figure 16. Eccentric press with Ctype frame (MOllerWeingarten); FN = 1600 kN, HmVl. = 160 mm, H m'n = 20 mm, n K = 50 min- I.
Manufacturing Systems. 4 Presses and Hammers for Metal Forging
Fipre 19. Sequence of motion of die slide (1) and blankholder slide (2) of press as shown in Fig. 18 [2[. C, standstill phase of blankholder slide.
Figure 17. Two-column press with two-point transverse shaft drive (Schuler); FN = 1600 kN, H = 300 mm, n K = 32 min-'.
common of the two types, and operate on the principle of hydrostatiCS (see G 1.1), The high-pressure energy of the medium under pressure (oil, water) is converted into mechanical action in the cylinders, The pressure p and the volumetric flow V are the most important characteristics of hydraulic drives. (See G2 for the design.) The slide force F" is determined by the pressure p and the piston surface area A; thus, F" = pA. It is not, therefore, dependent on the slide setting (Fig. 2b). The maximum value of Fso i.e. rated force FN , may not be exceeded, FN constitutes the most important force characteristic; the energy capability plays a secondary role in direct pump drives, as the energy required for the cycle is provided in sufficient quantities by the drive motor. In stored energy mechanisms the energy capability EN is specified by the quantity of energy stored and thereby constitutes another important characteristic. Hydraulic presses are easily adapted to the requirements of the process as regards force and machining requirements, speed and forming path. They are normally employed in processes with large force and/or machining requirements and long action paths with small to medium stroke rates, depending on the stroke size.
a
Figure 18. Basic structure of gearing in two-way action press [2}. a Swivel mechanism for die slide. b Stop-motion linkage for blank holder slide.
have been employed more frequently because the blankholder force may be more accurately reproduced and there is the option of making targeted adjustments in relation to the die path [3].
4.3 Hydraulic Presses Adaptive control presses (Fig. 2b) consist of hydraulic and pneumatic presses, Hydraulic presses are the most
b
1~
FiJ1ll"e :10. Basic layout of hydrauliC circuit of press: a with volumetric flow source (direct pump drive), b with pressure source (stored energy drive); 1 vessel, 2 pump and motor, 3 speed governor, 4 pressure-reducing valve, 5 non-return valve, 6 hydraulic accumulator, 7 four/three-way valve, 8 throttle valve, 9 hydraulic cylinder of press.
4.4 Hammers and Screw Presses. 4.4.1 Hammers
Figure 21. Hydraulic circuit of press with direct pump drive and safety monitoring devices to prevent the slide from sinking and accidental pressure buildup (SMG).
4.3.1 Types There are various different types of drives:
Hydraulic Presses with Conveyor System (Direct Pump Drive). See Fig. 20a for the layout of the hydraulic circuit. For the basic design with a safety monitoring device (to prevent the slide from sinking, and unforeseeable pressure loss) see Fig. 21. Characteristics: the pump and the drive motor are designed for the maximum instantaneous output requirement of the pump. Oil is used as the medium under pressure. The slide speed is normally infmitely variable by means of adjustment to the delivery of the high-pressure pump.
Hydraulic Presses with Pressure System (Stored Energy Drives) (Fig.20b). These are characterised by pumps and drive motors designed for average output. Oil or water is used as the medium under pressure. There is an increased tendency towards direct pump drives owing to reduced stroke sequence times as a result of mechanisation and automatic tool handling.
4.3.2 Assemblies In hydraulic presses, in addition to one- and two-column
frames, pillar-type frames with 2 and 4 pillars are also frequently employed. The latter are normally used in presses with high-rated forces for hammer forging and extrusion. Multiple-piston pumps (axial, radial and serial piston pumps) with a small stroke and piston diameter are used as high-pressure oil pumps. The various constructions include those achieving both constant and infinitely variable delivery rates. Flow and pressure may be adapted to suit the operating cycle via control devices (output rating, pressure and zero strike regulators). In addition, gear pumps may also be employed for constant delivery rates. Standard pressure levels p are between 2 x 107 and 3.15 x 107 Pa (200 and 315 bar), in exceptional cases even more. Hydraulic accumulators directly pressurised with compressed air are used in stored energy drives, nitrogen
batch storage, piston energy storage, or with water as the medium under pressure.
4.3.3 Applications Hydraulic drives are frequently employed in machines for mass working and sheet-metal forming owing to the high level of controllability of the slide force and speed. Hydraulic presses for serial manufacture of sheet metal parts (with precision blanking and drawing (see Fig. 22) and press baking) and for cold forging (extruding, hob bing, embossing) almost eXclUSively with direct pump drive. Forging presses (hammer forging, drop forging of lightweight metals) with rated forces of up to approx. 30 MN and slide speeds of less than 80 mm/s also employ direct pump drives. Stored energy drives are preferred at higher rated forces with high slide speeds of up to 250 mm/s. Hammer forging presses often use pillar-type columns which provide easy access to the working space. There are special advantages in this respect for under-floor drives. Extruders (Figs 23 and 24) are almost exclusively reserved for horizontal constructions with pillar-type frames. Pneumatic drives in small-scale presses are limited to drawing, blanking, bending and riveting applications.
4.4 Hammers and Screw Presses Closed-loop machining presses (Fig. 2«:) consist of hammers and flywheel spindle presses. The main characteristics include energy capability E which is transmitted in full for each operating cycle, except in clutch spindle presses. Otherwise, the rated force FN , the maximum (continuous) permissible force Fmn penn and the impact force Fimp are also important considerations in spindle presses.
4.4.1 Hammers These are the cheapest forming machines for generating substantial forces and transmitting high-energy capability.
Manufacturing Systems. 4 Presses and Hammers for Metal Forging
Of simple construction, they are impossible to overload, as hammer frame and drive are not located within the flow of force in the operating cycle. The forming process in hammers follows the law of impact. The potential energy E is translated into productive energy WN and no-load operation Wy (pile hammer return and anvil bed losses). One chardcteristic of energy transmission is the efficiency of impact 1), = WN/ E. In theory the following therefore applies for the anvil hammer: 1)5
= (l -
k' )/(l + m./m 5 ) ·
Here k is the number of impacts: in an upsetting operation k = 0.1 to 0.3, in drop forging k = 0.6 to 0.8. The ratio between the anvil mass m5 and pile hammer m. affects the loads applied to the foundations and the return acceleration of the anvil (the jump of the item being forged). Minimum values are m 5 /m B = 10 to 20 with a permanently fIXed anvil, m5/m. = 3 to 5 with a moving anvil.
f - - - 3300 - ---I
Structures There are anvil hammers, which include drop hammers, doub/e-acting hammers and counterblow hammers (Fig. 2S). Anvil hammers (Fig.26a) have a permanently fIXed anvil, counterblow hammers (Fig. 26b) have two pile hammers moving in opposition to one another.
Applications Figure 2:J. Hydraulic oil press Weingarten); FN ~ 6000 kN.
with
die
cushion
(MOller-
The main areas of application are in hammer forging and drop forging and in special-purpose applications in embossing, hot extrusion and sheet metal forming. For the areas of application of the various constructions see Fig. 27.
Figure :13, Direct press extruder (SMS Hasenclever): 1 counter bar, 2 tool slide or tool turret , 3 shears, 4 block pickup, 5 running bar,
6 stamp, 9 cylinder bar, 10 oil container with drive and controls.
Figure :14. Indirect
press extruder (SMS Hasenclever): 1 counter bar, 2 tool slide, 3 shears, 4 block pickup, 5 running bar, 6 matrix stamp, 7 seal, 8 cylinder bar, 9 oil container with drive and controls.
4.4 Hammers and Screw Presses. 4.4.1 Hammers
Pile hammer stroke: mechanical pneumatic hydraulic
Pile hammer drive: (mechanical) pneumatic hydraulic
Pile hammer drive: pneumatic hydraulic Pile hammer couple: mechanical hydraulic
Pile hammer drive: (mechanical) pneumatic Pile hammer couple: hydraulic
Figu.rc 2:5. Classification of hammers.
Here g = the acceleration due to gravity, H = the drop height. The stroke is limited to H = 1 to 1.6 m, in order to provide numbers of impacts n H of 50 to 60 min-I. Pile hammer impact speed is between 4 and 6.5 m/s. Technology has developed from strap-lift drop hammers and gravity drop hammers to pneumatic or hydraulic drop hammers, which have the advantages of low levels of wear on the stroke components and simple control and energy metering. Double-Acting Hammers. These possess extra energy
storage capacity in addition to the pile hammer in the form of compressed air, steam (0.6 to 0.7 MPa - 6 to 7 bar) or hydraulic oil (2 to 20 MPa or 20 to 200 bar). Hydraulically driven double-acting hammers (Fig. 28) are employed increasingly today owing to their favourable energy consumption figures. Their energy capability EN is given by where Pm. = the average index operating pressure and A = the piston area. Double-acting hammers provide shorter strokes of H = 0.4 to 0.7 m at the same pile hammer impact speeds as drop hammers, and thus substantially higher numbers of impacts (n H = 55 to 250 (450) min-I, depending on the type of construction and type of drive). Counterblow Hammers. These have only a third of the
Figure 2:6. Basic principles of hammers: a anvil hammer, b counterblow hammer; I frame, 2 (top) hammer, 3 (bottom) hammer, 4 anvil or baseplate, .5 backing plate, 6 base.
1.0
I
mass of double-acting hammers, with the same energy capability. Correspondingly smaller foundations are therefore possible. Design is the same in both vertical (more common) and horizontal feed motion. Both pile hammers are mechanically (conveyor) or hydraulically coupled in their movement (Fig. 29). In addition to traditional constructions where the mass of the top and bottom pile hammer are more or less equal, there are also more modem developments where the mass of the bottom pile hammer is significantly greater than that of the top pile hammer (Fig. '}O). As a result, the stroke of the bottom pile hammer is much smaller than that of the top pile hammer, producing advantages in loading. The number of impacts depending on the type of drive may be between 30 and 120 min-I.
I I
160 ,00 1000 10000 100000 100 250 630 Pile hammer mass m" (kg) (c:oonterlliow hammer has 1 pile hamRlel) Figure 1.7. Areas of application for anvil hammers (drop and double-acting hammers) and counterblow hammers for drop forging.
Drop Hammers. The energy capability (excluding fric-
tion and windage losses) is given by Figure 2:8. Hydraulically driven double-acting hammer (Lasco).
Manufacturing Systems. 4 Presses and Hammers for Metal Forging
Where the tool makes an impact on the workpiece, kinetic energy is translated in full from the flywheel, spindle and slide into productive energy and no-load operation (lengthwise and torsional spring losses in spindle and frame and frictional losses on guide and spindle). Energy transmission is characterised by the degree of effect of impact 1)s. One significant characteristic is energy capability E, given by
E where
] We
(jw'ij/2) + (m B d,/2), the moment of inertia of flywheel and spindle, = the angle speed of flywheel or spindle on impact with workpiece, = the slide mass, = the speed on impact of slide on workpiece (nortnaJIy between 0.5 and 1 m/s).
Figure Z9. Counterblow hammer, hydraulic drive and pile hammer couple (Beche & Grohs).
Otherwise, the rated force FN , maximum admissible force (in continuous operation) Fm~ pe= and impact force Fimp are essential factors, as these compressive forces are applied to the spindle and the frame (depending on the type of construction). The impact force Fimp occurs where the total energy capability is translated into spring energy without deducting productive energy. Fimp may be estimated from the rated energy capability EN and rigidity of the press in the direction of operation cz, where FimP = .J2czEN • The following nortnaJIy applies for FN , Fimp and Fmax perm:
Screw presses with a high energy capability (for hot metal forming) cannot be designed to be free of impact
Figure 30. Drive system (schematic diagram) of counterblow hammer with unequal pile hammer mass (Lasco): 1 top pile hammer with mass m I, 2 bottom pile hammer with mass m2, 3 oil, 4 air, Ht top pile hammer stroke, Hz bottom pile hammer stroke, VI speed of top pile hammer, V l speed of bottom pile hammer, m,/m, = 1/4, H,/H, = 4/1, v,/v, = 4/1.
4.4.2 Screw Presses In the traditional construction of the (flywheel) screw press, the spindle and flywheel are either continuously positively connected or else friction-locked. Rotary motion of the flywheel and spindle over a treble or quadruple re-entrant winding of the coarse pitch (angle of pitch 12 to 17°) is translated into linear slide motion.
Figure ~1. Direct drive screw press with hydraulic excess load protection (MOller-Weingarten).
4.5 Safety
effect on economic grounds. The forces occurring in the machine may be limited using hydraulic cushions between the tension rod nuts and the frdme (Fig. ~ 1) , or a friction clutch between the flywheel and spindle, or via energy metering. Stroke rdtes of between 12 and approximately 65 min- I may be achieved in screw presses, depending on the type of construction and size.
Types of Construction The friction disc drive is the classic form of drive, with two or three continuously rotating flat lateral discs with a spindle with longitudinal motion, or two tapered lateral discs with a permanently ftxed spindle (Vincent friction screw press). However, the high load application and the associated high wear to the friction linings are disadvantageous of such drives, and recent use has tended instead towards reversible electrical motors which drive the flywheel via friction rollers or pinion gears, or indeed are directly coupled (Fig. ~ 1). Large-scale screw presses (the largest with energy capabilities of 4.5 MN m and impact forces of 315 MN) are driven via a number of electrical reversible motors or hydromotors (Fig. ~2) positioned along the flywheel perimeter via pinion gears on the toothed flywheel rim. More modem developments have included the clutch screw press (Fig. ~~) with a continuously rotating flywheel that is connected to the spindle via a switchable friction clutch to initiate the working stroke. When a preselected force is reached, the clutch disengages the flywheel from the spindle, and the return stroke of the slide is then carried out via a pullback cylinder. The acceleration times of the slide are Figure 3~. Screw press with continuously rotating flywheel and clutch (Siempelkamp): 1 clutch cylinder, 2 clutch piston, 3 flywheel, 4 thrust bearing, 5 puB-back cylinder.
consequently shorter, because the masses to be accelerated during the working stroke are small. The compressive forces may be limited via the frictional torque of the frictional dutch.
Application Screw presses are found both in forging operations (drop forging of non-ferrous metals, production of precision forging parts) and in cold forging (cutlery manufacture, mintage and sizing, and coining) and sheet-metal forming (production of flat die parts from thick sheet metal).
4.5 Safety
Figure 3:1. Large-scale screw press with hydromotor drive (SMS Hasenclever) .
Work on presses is subject to safety regulations. These are designed to prevent accessibility to the hazardous area (tool installation area) while the tools are making contact (e.g. via hand guards), to eliminate any accidental contact movement by the tools (press safety devices) and to limit noise emissions. The rules and regulations and standards summarised in Fig. ~4 , from the relevant standard engineering practice according to §3 of the German law on working materials, apply for work safety on presses in general. Accordingly, hand guards may take the form of secured tools, permanently fixed screening of hazard areas, two-handed switching and no-contact protective equipment (photoelectric barriers) . Press protective eqUipment includes safety clutches, safety brakes, safe control , follow-up monitoring, removable screening and devices for automatic pass dis-
I:!DIiII
Manufacturing Systems. 5 Metal-Cutting Machine Tools
From 1.1.1993
Until 31.12.1992
I
Technical working tool regulation Tool safety regulation 1968
®
GS-Symbol optional
~ I
1987 1987 1961 1974
1989\
Controls for powerdriven metal working 1978 presses Two-handed gearchange for power-driven metal working presses 1978 Non-contact protective equipment for powerdriven metal working 1980 presses Flexible shielding for power-driven eccentric and related metal 1975 working presses Bending processes on power-driven die bending metal working presses (bevel presses)1981 Figure
[IT]
IA
Technical regulations General
Safety regulations
1987
EC-Guideline for machines CE-Symbol compulsory
B
B
Accident prevention regulations Power-driven working tools Eccentric and related presses Hydraulic presses Screw press Noise
I
\
DIN-standards (e.g. 31001) VOE-regulations (e.g. 0113) VDI-guidelines
Harmonised standards in preparation: EN-presses EN-two-handed gear changes EN-non-contact protective equipment EN-controls
*_ safety regulations for working on presses.
abling. Figure 16 shows the structural design of the latter protective measure. The operating brake is separated into two equal units operating independently of one another. In the German accident prevention regulation on noise (VBG 121: German accident prevention regulations of the Employers liability Insurance Association), the acceptable
noise level specified is 90 dB (A), with levels in excess classified as "areas of excessive noise". In many areas of operation presses have as yet been unable to adhere to
this acceptable noise level. In these cases, secondary noise reduction measures (such as partial or full encapsulation in enclosures are required), along with acoustic protection for each individual employee.
Metal·Cutting Machine Tools G. Spur, Berlin
5.1 Lathes S.I.I GeAeral Parts generated by rotation are manufactured on lathes. The workpiece performs a circular motion about an axis of rotation, while the tool carries out the feed motion in a plane perpendicular to the direction of cut. In specialpurpose constructions, the tool may also be rotated. Power-driven tools are also used for light metal drilling and milling operations as well and are thus also suitable for end-face machining, slotting and eccentric milling, or drilling diagonal to the axis of the workpiece on the lathe.
Clusification. Classification in terms of universal lathes, automatic lathes, frontoperated lathes, vertical boring and turning mtlls and special-purpose lathes has come from the lathe construction industry itself. A system-
atic classification may also be made according to the position of the main axis in vertical and horizontal machines, the number of spindles in single or multiple spindle machines and the type of control, i.e. manually operated machines, automatic mechanical PLC machines and numerically controlled machines.
Construction. The lathe processing system may be divided into subsystems, including workpiece systems, tool systems, kinematics systems, energy systems, information systems and auxiliaty systems.
Workpiece Systems. These include the workpiece and workpiece clamping and suppott mechanisms. The most popular form of clamping is the three-jaw chuck (Fig. I). It is employed in conventional machines as a manual chuck and in NC machines as a power-operated chuck in hydraulic, electrical or pneumatic operation. Lathe chucks with two, four or six jaws are also used in special cases. Other clamping devices consist of collets, faceplates and lathe mandrels. Long workpieces are held
5.1 Lathes. 5.1.2 Universal Lathes
5
6
numerical control has made this form more popular in small and medium-sized serial production.
Auxiliary Systems. These include functional systems such as coolant lubrication, swarf-clearing installations, and central lubrication. The frame forms a deftnitive part of the basic shape of the machine as the carrier of the other assemblies, carrying the guides of the tool slideways in the form of the bed, and the main spindle and main drive in the form of the headstock. Cast, welded, and concrete constructions are employed. The horizontal bed-type machine, the standard manually operated machine, is being superseded predominantly by the inclined-bed construction in numerically-controlled machine tools because of the improved chip clearance. Lathes with a vertical main axis are designed as column-type machines. 5.1.2 Universal Lathes
Figure 1. Power-opcrated wedge-type collet chuck with mechanical centrifugaiforce balance (Forkardt GmbH, Dusseldorf); 1 chuck
body, 2 chuck adapter for universal spindle assembly, 3 threadcd ring for connection to air line, 4 faceplate jaw, .5 centrifugal force, 6 base jaw, 7 protective bushing, 8 bar capacity, 9 standard jaw attachment.
between points, driven by drivers and supported by a steady rest (backrest) where applicable.
Tool Systems. These include the tool, tool clamp and tool carrier. In addition to rapid-change tool holders, turrets are in common use in program-controlled machines as tool carriers for between 4 and 16 tools each. The tool holders are carried in standard straight-shank carriers, Tbolts or V-guides. However, more and more tool change systems are being adopted where either only the cutting head or the entire tool including shaft and holder ftxture are changed. The latter are also speciftcally designed for changing power-driven tools.
Single spindle lathes are dominant in small and mediumsized serial production and are available in a variety of sizes. They are classifted in terms of drive ratings and operating ranges. The operating range is determined by the maximum turning diameter and the longest distance between centres. The most common sizes have a
maximum pitch of between 100 and 500 mm and a maximum distance between centres of between 250 and 1250 mm. Then come the small lathes and above this the large-scale lathes with pitches of up to approximately 2500 mm and distances between centres of up to approximately 10 m. The manually operated sliding and screw-cutting lathe has the basic form of a universal lathe (Fig. 2). The main spindle is driven by a multiple-stage, change-speed gear drive, in order to be able to operate over a broad speed range at a constant output. The feed drive is taken from the main drive via the feedgear mechanism, feed rod and bed slide drive. The main spindle serves to provide the kinematic connection between the main drive and longitudinal feed in screw cutting operations. With the use of turrets as tool carriers for the tools required for processing the workpiece, turret lathes were
Energy Systems and Kinematics Systems. The kinematics
developed. A distinction is made between the drum tur-
system is divided into one system for generating the cut-
ret, turnstile turret, disc-type turret and flat-bed turret, depending on the orientation of the tool axis and the operating axis (Fig. 3). Special-purpose constructions include the block turret, cross-type turret and crown-type turret (Fig. 3). The manually operated turret lathe has been superseded to a large extent by numerically controlled machines. Mechan-
ting motion and another for generating the feed motion
that is transmitted directly from the energy system. Singlestage or reversible-pole three-phase motors are combined with mUltiple-stage mechanical switching mechanisms to cater for a broad speed range in main drives. D.C. motors with electrical speed adjustment with two- to four-stage electrically switched transmission units are normally coupled in numerically controlled machines. The main spindle is usually seated in a roller bearing. High requirements relating to thermal characteristics and high speeds may require special lubrication systems. Electrical motors and occaSionally hydraulic motors are employed in feed drives not derived from the main drive. The slideways normally run in V-guides, although enclosed flat-bed guides, circular guides and combinations thereof are also employed.
Information Systems. These are employed for the control of the interaction of functions between the subsystems. They incorporate external information such as machining programs or parts sequences into the manufacturing system and submit status reports to the primary control sys-
tem. Control cams, stroke rates and trip cams are employed as mechanical data stores in automatic lathes for mass production. The greater flexibility offered by
ical program-controlled models are used in mass pro-
duction environments as a single spindle lathe.
Copying Lathes. These usc mechanical, electrical or hydraulic systems to scan a two- or three-dimensional standard shape which is subsequently retrieved to control the feed motion of the lathe cutting tool. Devices for out-oftrue copying by lathe are also employed in addition to longitudinal copying lathes for shaft processing. Mechanical systems operate with direct power transmission and a lead cam or swivel guide bar (taper turning). Powerassisted systems operate with sensitive scanning devices and electrically or hydraulically controlled reproductive motion by the tool. Because NC continuous-path control systems carry out the same tasks, and the description of the workpiece contour in the NC program is less complex than the manufacture of a lead cam, lathe copying has also been mainly superseded by the use of NC machines. In numerically controlled universal lathes the main
Manufacturing Systems. 5 Metal-Cutting Machine Tools
Figure Z_ Engine lathe and bar lathe (Gebr. Boehringer, Goppingen): 1 flange motor drive,
J
5
6
11IJ I~ 15
789 10 11
2 feed drive, 3 headstock with main gearing,
16
4 speed switch, 5 switch cabinet, 6 operator's panel, 7 main spindle, 8 faceplate, 9 longitudinal feed stop, 10 saddle (longitudinal slide), IJ toolholder, 12 cros~slide , 13 top support, i 4 transverse stop, J5 dead centre, 16 taHstock, J 7 swarf guard, 18 lever for tailstock clamp, 19 tailstock clamping lever, 20 handwheel, 2 J gear rack , 22 guide screw, 23 feed rod, 24 operating shaft, 25 remote-control lever, 26 base of bcd, 27 swage box, 28 intermediate switch lever for direction of feed, 29 split nut, 30 handwheel for cross-slide, 31 handwheel for longitudinal slide, 32 lock box, 33 bed, 34 feed and thread-rolling drum, 35 feed box, 36 switchover mm-inches, 37 engine-lathe-bar-Iathe changeover lever.
1
I
J5
,
JJ Jl JI Xl 19 28
b
d
..J
Figure~. Constructions of turret heads. a Drum turret (Pinier, Langen). b Turnstile turret (Pittler, Langen). c Flatbed turret (Pittler. Langen). d Disc-type turret (= turnstile turret) (Gebr. Boehringer GmbH, Goppingen).
5.1 Lathes. 5.1.5 Heavy-Duty Lathes
spindle is either driven directly by a speed-controlled main motor via a belt drive or with an intermediate gear mechanism (Fig. 4). The feed slides are positioned by separate path-controlled precision ball screw drives. For longer overall lengths a permanently fixed spindle with a notched drive (drive motor on saddle). or a pinion drive with a gear rack located on the saddle. is normally employed. The position of the angle of the spindle is verified using a rotational measuring system and coupled via the controller to the feed to syochronise the spindle rotation with the longitudinal feed in screw production. CNC continuous-path control systems provide high productivity owing to their ease of operation and programming, such as the provision of subprograms for machining cycles, thread-cutting programs, automatic cut sectionalisation and simulated process graphics on the controller monitor.
is derived from the main drive via complex gearing units (Fig. 6). As modifying the cams to suit the machining task to control tool paths and speeds is a complex business. mechanically controlled single-spindle drives are increasingly being superseded by numerically controlled machines. Swiss bush-type automatic lathes are employed for manufacturing long, slender workpieces (Fig. 7) where the workpiece is fed close to the cutting point. Automatic cross-feed motion is carried out via tool carriers arranged in a turnstile formation in relation to the rotating axis and attached to a permanently fixed rocker column. The longitudinal feed is carried out by a moving headstock in the Swiss system and a moving back rest in the OffenbaCh system. Guide bushes are employed in the Swiss system to increase the degree of accuracy. Both numerically controlled systems and mechanical cam curve systems are in common use.
5.1.3 Front-Turning Machines Front-loading automatic chuck lathes are used to machine disc-shaped workpieces with pitches of up to 800 mm (Fig. 5). These machines are available as one- and twospindle models, permitting multi-point machining, and they have electrohydraulic or numerical program control systems. Capstans, facing slides and cross-slides are employed. These machines are also constructed in the column-type design with a block turret and lateral slides for chucking operations and as facing lathes with a main bed guide wbich is normally positioned at an angle to the horizontal spindle axis for processing larger and bulkier workpieces on faceplates. High output volumes may be achieved with automatic workpiece handling by way of two-arm systems for gripping, loading. turning and unloading in conjunction with workpiece stores.
5.1.4 Automatic Lathes These provide automatic workpiece machining from rod materials or pre-formed chuck parts. One characteristic conunon to the various designs is multi-paint machining which is supplemented in multiple spindle machines by sectionalised machining. The criteria distinguishing these
include the number of spindles, the horizontal or vertical position of the main machine axis, the frame construction, the type of tool carrier, the coordination of the cutting and feed motion and the number of potential feed movements. In automatic machines employed in mass production with mechanical control, control cams and operating cams are used for the following functions: workpiece feed or workpiece advance, movement of longitudinal or cross-slide and tailstock and turret head, speed adjustment and change in direction of rotation, operation of clamping devices, movement of auxiliary or special purpose equipment, and moving headstock or spindle sleeve where required. The main control shaft, which completes one revolution for each completed workpiece, determines the sequence of the productive and idle time motion. It is supplemented by an aUXiliary control shaft to generate independent idle time motion. Mechanical cam control may be supplemented or substituted by electrical, hydraulic or numerical control. Automation of the machine tool handling process, by way of system-compatible loading and magazine devices for rod or chuck parts, also permits a number of machines to be chained.
Single·Spindle Lathes. In mechanically controlled automatic lathes, all feed and switching motion is controlled by control cams and operating cams whose motion
Multiple-Spindle Automatic Lathes. Multiplespindle bar lathes or chuck lathes are used for automatic turned part mass production (Fig. 8).
Classification. There are constructions with or without tool carrier control with rotating workpieces or tools, a horizontal or vertical main machine axis and a variety of types of control. The size may be specified in terms of the spindle capacity diameter or chuck diameter and the number of spindles. Construction. The frame normally takes the form of a single column or double-column construction with high rigidity depending on the working space, tool and workpiece motion. chip clearance and coolant lubrication system; it is also available in vertical construction for larger pitches. The main spindles are driven centrally; only bulky or unbalanced workpieces are processed from one or more sides on machines with rotating tools and a bed-type or column construction, where clamping takes the form of indexing plates or drums. Longitudinal slides and crossslides and all auxiliary motion are controlled mechanically, such as indexing and latching spindle drums, bar feeding, bar stops. collet gripping or chucking and the operation of special-purpose equipment (Fig. 9). Path and switch-
ing data are stored by way of curves and cams on the central or branched control shaft which runs at the feedrate for productive time motion and in rapid gear for idle time motioo. Longitudinal slides and cross-slides are cootrolled in groups or individually. An increase in the working space can be attained with the help of additional equipment for shutting down spindles, variable speed drives. duplex switching with two clamping points, and rod and chuck gripping. Numerical control and independent feed drive applications are made more complex in multiple-spindle lathes owing to the limited space involved. Numerically controlled slides are frequently only provided in critical machining centres for this reason. The two-spindle lathe (Fig. 10) provides the advantage of sectionalised machining and there is sufficient room available for it to be designed as a NC machine. The use of compact hydraulic feed drives permits the conventional structure to be retained and exploits the inherent flexibility of numerical control (Fig. 11).
5.1.5 Heavy-Duty Lathes Large-scale, longitudinally arranged rotating parts are manufactured on special horizontal lathes. The structural
_:..=J
J
J:.~~, ~11
r---~
J
5
2
7
8
Figure 4. NumericaUy controlled lathe (Gildemeister AG - Max Muller, Hanover): a turret design; b tool magazine design; 1 machine bed, 2 headstock, 3 feed drive, 4 cross-slide, 5 d.c. main drive, 6 tailstock, 7 disc-type turret , 8 hydraulic accumulator, 9 chip conveyor, 10 switch cabinet, 11 path measuring elements, 12 operator's panel , 13 longitudinal slide, 14 back rest, 15 tool carrier, 16 tool grip, 17 tool magazine.
b
.., '"
//
J
rr
d
I
a
10
11
~
"
§-
3: ~
(JO
go
c::
~
~
""
•
l~
(JO
5
2
(l
~
~
i
5.1 Lathes. 5.1.7 Flexible Turning Centres
It Figure S. Double-spindle front-turning machine O.G . Weisser & S6hne, St. Georgen): a overall layout , b section of front turning machine.
design of these follows in principle that of single-spindle universal lathes. As the sag of the workpiece has the least effect on the machining diameter when the direction of the tool feed is perpendicular to the direction of the force due to weight, the construction is always of flat-bed type. Additional milling attachments offer overall machining of large-scale pans and produce high rates of metal removal in turning and milling (Fig. 12).
Vertical Boring and Turning Mills. These are employed for machining heavy and bulky workpieces with a small length/turning-diameter ratio. The faceplate rotating around a vertical axis is characteristic as a carrier for the workpiece which is to be machined in the subsequent installation position and clamping device where possible. There are single- and two-column machines with stationary or moving frame components and a number of tool slides on the columns and crossbar (Fig. 1~) . The two-
S.I.6 Special-Purpose Lathes These are used for workpieces where processing on standard machines is either impossible or uneconomical. Their construction depends on the size and type of the workpiece. The structure has to be functional and to conform with standard-sized components hased on the mechanical assembly technique with a variety of sizes, where matching or reconstruction will be carried out as required. Turning machines for roll pins , crankshafts, turbine discs, cylindrical bushes, axles, cams, wheel sets, wheel discs, pipes, sleeves and out-of-true lathes are available depending on the specified manufacturing task. S.I.7 Flexible Turning Centres Numerically controlled machines may be adapted for machining the reverse side for overall machining of a com-
column machine in a double-column construction, or the
prehensive range of workpieces by expanding the kin-
single-column machine with a projecting cross-beam, is employed where there is a turning diameter of over 3 m. Double columns, single columns and faceplate undercarriages may be designed as moving pans to increase the turning diameter or to improve accessibiliry for servicing by crane. The faceplate diameter amounts to between 800 and 5000 mm under normal operating conditions, with a maximum of up to 18 m, while the maximum turning diameter is between 1400 and 5200 mm, or approximately 25 m. The loading capacity of the faceplate determines the permissible weight of the workpiece. Core and ring faceplates may be combined in large-scale machines. The feed is effected increasingly via individual numerical control drives. D.C. motors with drive output ratings of up to 200 kW are normally used as the main drive; positioning control and NC continuous-path control of the faceplate drive are available. Turret or ram slides supply the tool system. Operation is carried out via a suspended panel or in the case of large-scale turning machines via a travelling operating stand. Additional attachments may be employed for specialpurpose operating processes (Fig. 14). The workpiece weight and dead weight of frame components causes tool shifts which should be taken into account in design planning.
ematics system and with the use of power-driven tools and additional attachments. The use of a second slide unit produces a four-axis lathe, which permits an increase in the output volume owing to sectionalised machining. A C-axts is frequently used for the angle positioning of the spindle in comhination with power-driven tools for boring and milling work, while occasionally a Y-axis is also used as the third slide axis. Figure 1 S shows examples of structural forms which may be manufactured with a controUed C-axis in conjunction with tool drives. When the workpiece has been fmished on the front, it may be taken up by a clamping device and the reverse side may be fmished at a separate machining centre, for example in reverse side machin-
ing. Figure 16 shows a two-spindle machine with one stationary and one moving spindle-and-turret system in each case. The CNC machine shown in Fig. 17 with the rotating spindle head permits tool changes to take place outside the working space, thereby reducing non-productive time. Where the turning machine is designed for semi-automatic operation with memory SIOf'dge and handling systems for work pieces and tools, this is known as a turning centre (Fig. 18). Drum, chain and ring magazines are
Manufacturing Systems. 5 Metal-Cutting Machine Tools
Figure 6. Mechanically controlled automatic single spindle short-stroke lathe with auxiliary control shaft and separate main control shaft (H. Traub AG, Reichenbach): a gearwheel layout, b gearwheel arrangement; 1 cam for tool slide 51 (front traverse support), 2 cam for tool
slide 52 (rear traverse support), .3 cam for tool slide S3 (rear vertical support), 4 cam for tool slide 54 (front vertical support), 5 drum cam for drive stroke of turnstile turret, 6 cam for rapid return of turnstile turret, 7 cam for sorting device, 8 cam for gripper device, 9 cam for longitudinal stroke of gripper lever, 10 cam for rotating motion of gripper lever, 11 cam for clamping in gripper lever, 12 cam for front longitudinal lathe attachment, 13 cam for rear longitudinal lathe attachment, 14 cam of clamping drum for tool clamping or rapid clamping, 15 spindle controlling mechanism, 16 mitre gear, 17 main spindle, 18 feed drive, 19.1 rear control shaft, 19.2 front control shaft, 20 auxiliary control shaft, 21 operating shaft for feed drive, 22 worm gear shaft, 23 jaw clutch for workpiece rapid clamping, 24 back-drilling attachment and slotting device, 25 cross-drilling attachment, 26 turnstile turret with rapid return, 27 drive attachment for turnstile turret tools, 28 multiposition switch (multiple control switch), 29 excess load protection for auxiliary control shaft, 30.1 rear guide shaft, 30.2 front guide shaft.
used for external tool storage, and handling is effected via freely programmable systems which are frequently designed in double-column style in the same way as tool handling systems. There are a variety of automatic tool-
change systems for changing the cutting head as a result of wear (Fig. 19). Tool data storage often takes place in pallets or conveyor stores. Further measures for low-operation systems and high
5.2 Drilling and Boring Machines. 5.2. I General
b Figure 7. Tool column and headstock of lungitudina l autntn;llil j;nhc (l-L Traub AG, Reichenbach): a tool or rocker column, b section of headstock design with longitudinal OIOlion ; I roChT, 1 gU ide hush , __j tool slide, 4 collet, 5 relief spring, 6 feed spindle, 7 clamping bush, 8 tensiun ring, 9 spindle bearing fr('C o f shearing load. If) SIXh ' t: ;' ! Hh(~ , I I fork. 12 damping lever.
Figure 8. Six·spindle bar lathe with cross-slide (Gildemeister Bielefeld).
AG,
Figure 9. Gear wheel arrangement of multiplespindle lathe: J drive col· umn, 2 tool space,
3
spindle
column,
4
motor, .5 V-belt, 6 drive
pulley, 7 tubular shaft, 8
central fixed shaft, 9 ccotral gear, 10 spindle gears.
flexibility include the use of systems for automatic change of clamping jaws or the entire clamping device and sys· terns for tool monitoring and workpiece measurement.
5.2 Drilling and Boring Machines ;.2.1 General
Cutting motion and feed motion are coordinated accord· ing to the process, the tool or the workpiece in drilling
and boring machines. Constructions are classified depending on the position of the drilling spindle and type of frame structure (Fig. 20) . Other criteria for classification and selection include the clamping and operating range, rated drilling capacity, speed and feed range, mach· ining accuracy, degree of automation and pOSition and number of drill holes. Twist drills, counterbores, reaming bits, tap drills, drill heads, spade drills and special·purpose drills and boring bars in drilling and boring machines are
Manufacturing Systems. 5 Metal-Cutting Machine Tools
S.2.2 Bench Drilling Machines These are suitable for small-scale drilling operations and workpieces. The torque is transmitted from the motor shaft to the drilling spindle via belt transmission. Speed changes may be made by adjusting the belt or replacing the belt pulley; the feed motion is carried out manually via a start-handle shaft, feed pinion and gear rack drive.
S.2.3 Free-Standing Pillar Machines
Figure 10. Two-spindle lathe.
used. The tools have cylindrical or tapered ends and fit into cylindrical or tapered sockets on the machine. In special-purpose machines arrangements are made for adjusting the length of the drilling.
The frame of these is designed as a free-standing hollow pillar, to which the drive casing and drilling spindle are attached either pemJanently or as part of a height-adjustable boring slide. The bottom section of the column carries the dlill table which is height-adjustable and may be mtated and is not normally supported. Wheel gears or friction gears provide speed adjustment (Fig. 21). Feed is manual, or via a feed cam, or by way of worm gearing attached to the spindle sleeve unit.
S.2.4 Column-Type Drilling Machines These afe suitahle for small and medium-sized workpieces. There is a moving boring slide located at the top of a
Figure 11. Multiple spindle lathe with hydraulic feed drives (Gildemeister AG, Bielefeld); 1 machine bed, 2 spindle column, 3 drive column, 4 spindle drum, 5 main drive motor, 6 hydromotor for spindle drum control, 7 spindle drum latching, 8 hydraulic cocking '--1'1incter, 9 shutdown gate valve, 10 cfoss-slide with
continuous path control, /1 feed drive sliding spool control valve, 12 screw cutting drive, 13 control panel with vnu.
Figure 1:1. Large-scale lathe with milling attachment (Wohlenberg KG , Hanover) .
5.2 Drilling and Boring Machines. 5.2.6 Radial Drilling Machines
a
b
Figure 14. Single-column vertical boring and turning mill with moving column for external machining of reactor pressure vessels. Convertible auxiliary columns for internal machining of vessels on stationary faceplate centre.
,------~~~
I
b
I I
i
~.
I
L------~===l=~r-!~~~ d Figure I". Vertical boring and turning milL a Single-column vertical boring and turning mill with left-hand ram slide, right-hand square turret slide and right-hand lateral slide. b Single-column vertical boring and turning mill with cross-slide and lateral slide and two square turrets. c Two-column machine with moving double column. d Singk-column machine with moving column.
rectangular column and the hottom section supports the drill table, which is designed to cater for heavy workpieces. Tailstock or boring slides execute feed motion, while spindle speed is adjustable via V-belt gearing, charge gears and changeover gears or intlnitely variable drives.
5.2.5 Multi-Spindle Drilling Machines These are suitahle for use with high volumes or frequent repetition of drilling jigs. The frame structure is often identical to that in heavy-duty column-type drilling machines. A mUlti-spindle cutter is used in place of a single-spindle drilling shaft and is driven via one or two shafts from the main spindles. The spindles are driven via gearwheels in machines with permanently fixed multi-spindle drilling heads, and via drive shafts in adjustable, articulated spindle drilling machines. Spindle guides in permanently fixed bearing plates (Fig. 22) have a higher rigidity, hut the adjustable supporting beam guide adversely affects the reproducibility. A number of drilling jigs may he in use simultaneously for the spindle guide. Separate drive trains with reversible-pole motors and
d Figure 15. Boring and milling with C-axis control: a positioning control and line motion control, b line motion control, C C/X interpolation. d e/Z interpolation.
main and change gears permit speeds to be transmitted independently by the distrihutor gear mechanisms, which consist of the central drive shaft, planetary and intermediate gearing. The latter may be omitted, depending on the direction of rotation of the spindle and size of the gearing. Rapid advance motion may be generated by a three-phase motor, and feed motion may be generated via infmitely adjustahle d.c. motors owing to the improved speed synchronisation. Multi-spindle drilling heads are also used in special-purpose machines with horizontal and angled directions of feed.
5.2.6 Radial Drilling Machines Radial drilling or cantilever horing machines are employed for drilling large and bulky workpieces from scratch. The chief components are the baseplate, pillars, cantilever arm, drill slide and drill table (Fig. 23). The baseplate is designed as a rectangular section or in special-purpose designs as an angle plate, cross-baseplate, or double, tum-
Manufacturing Systems. S Metal-Cutting Machine Tools
5.2.8 Turret Drilling Machines The frame structure is similar to the column-type drilling machine. The drive and gearing are arranged either inside moving boring slides or separately. The spindle is driven via drive shafts and a disengaging clutch. Feed motion is normally executed via slides. The turret circuit is designed in the form of a cross-wheel with indexing pins or via cross-toothing (Fig. 25). Turnstile turrets are normally used as tool carriers, while tool changers may also be used. Machines with cross-tables are also designed for light milling operations. They may be expanded in conjunction with numerical control to providing machining centres.
5.2.9 Precision Drilling Machines
Figure 16. Two-spindle turning machine (Boley GmbH, Esslingen): 1 headstock I, 2 turret slide I, 3 moving spindle unit II, 4 stationary turret carrier II.
stile or circular baseplate. The machines are employed in small- and medium-scale serial production as universal machines. Wall-slewing drilling machines have a cantilever arm attached permanently to the wall or to the height-adjustable track.
5.2.7 Jig Boring Machines Jig boring machines may be employed to produce bores, depressions and shallow notches to a high degree of accuracy without tracings or templates. These are employed in single-part production, e.g. in gauge engineering, tool construction and the construction of jigs and fixtures. Singlc-column machines arc employed for small workpieces,
larger double-column machines are available in column or free-standing pillar designs. Single-column machines (Fig. 24) are designed with a cross-table and double-column machines have a longitudinal table. The boring slide is positioned horizontally on a crossbar between the two pillars or columns in backlashfree, anti-friction slideways. The vertical feed is introduced via a pinion on the spindle sleeve. A horizontal shaft drives two worm-type gears synchronously for raising and lowering the crossbar, which is arrested via the self-locking hydraulic flow constriction mechanism.
Characteristics include high static and dynamic rigiclity, high damping factor, high degree of uniformity in motion and minimum change in temperature. As a result, diameter and position tolerances of IT6 to ITS may be achieved and IT4 with the relevant input. The torque is transmitted via a specially positioned belt pulley and clutch onto the verticalor horizontal main spindle to prevent torsional forces and oscillation. Both three-phase motors and infinitely variable d.c. motors may be used. The feed motion is normally generated hydrauliCally and is either executed by the headstock or the carriage (workpiece).
5.2.10 Deep-Hole Drilling Machines These are employed for the manufacture of drilled deep holes with a diameter/length ratio of between 1 : 3 and 1 : 200 and are structured in a similar way to lathes. A distinction is made between short and long bed-type machines, machines with rotating workpieces, rotating tools or counter-rotation of tool and workpiece with a horizontal spindle position and machines with rotating tools and a vertical spindle. The feed is executed via the ballscrew spindle or rack; the tool is guided into the drill bushing or pivot hole via a three-point support on support rails. Continuous supply of coolant and chip removal are required. Machine assemblies include workpiece spindle box, workpiece steady rest, guide slide with coolant lubrication supply, tool clamping bearing and tool spindle box.
5.2.11 Special·Purpose Drilling Machines The Simplest type of special-purpose drilling machine is a serial arrangement of the table-type, free-standing pillartype and column-type drilling machines. The construction is modified depending on the manufacturing task, where
3
Figure 17. CNC chuck lathe (Index-Werke KG, Esslingen): 1 Swivelling headstock, 2 hinged-arm loader, 3 oval-shaped conveyor.
5.2 Drilling and Boring Machines. 5.2.11 Special-Purpose Drilling Machines
Figure 18. flexibLe turning centre (Traub AG , Reichenbach). Components: I line double column, 2 workpiece gripper, .3 tool gripper, 4 workpiece store 1, 5 transfer tube for workpiece store 1, 6 tool-setting terminal, 7 workpiece store 2. 8 transfer tube for workpiece store 2, 9 transfer tube for tool store, 10 tool store for tools, clamping devices and grippers. Operations: 1: Remove tool, place tool in magazine. 2: Remove chuck jaw, place chuck jaw in magazine. 3: Change gripper chuck, place gripper chuck in magazine. 4: Tool monitoring. 5: Tool change. 6: Check workpiece dimensions, set tool offset. 7: Remove ftnished part, feed in unmachined part. 8: Change chuck jaw. 9: Outward transfer of fmished part, inward transfer of unmachined pan . Bar code marker on workpiece pallets, bar code reader for paIJets. 10: Remove unmacltined pan, son finished pan. II: Outward transferlinward transfer of rool holder for cross-machining. 12: Outward transfer/inward transfer of tools. 13: Gripper plate on tool holder in accordance with DIN 69880 with bar code marker, bar code reader for tools.
Manufacturing Systems. 5 Metal-Cutting Machine Tools
Figure 19. Changeover systems for cutting heads (Traub AG, Reichenbach). a Sandvik system, b Widea system, c Hertel system.
Figure 21. Infmitely variable speed control mechanism in freestanding pillar-type drill (WEBO-Ho/heinz, Mascbinenfabrik, Dusseldorf). The control range of the two-step gear is 10.4 (with reversible-pole motor, 20.8). The drive shaft 2 drives the intermediate shaft 8 with geatwheeis 4 and II via the upper friction wheel 1 and the friction ring 2. Torque is transmitted from the transmission gearing shaft 9 to the drilling spindle 13 via pair of gears 5 and 12 or 7 and 6. The feed motion is derived from the pinion 10.
aR b
it lW
Y
II
E
h
L.......C' - - - ' - L - _--'
Figure 22. Spindle cutter of a multi-spindle drill with a stationary bearing plate (Bernhard Steinel, Werkzeugmaschinenfabrik GmbH & Co. , Schwenningen): 1 flange for motor drive, a main gearing, b power takeoff gears (articulated spindle drive) , 2 driving spindle , 3 , 5 joints, 4 driving bush for length offset, 6 spacer, 7 spindle plate, 8 drilling spindle. Figure 20. Schematic diagram of different drill constructions (designation of coordinates in accordance with DIN 66217). a Table-type drill. b Free-standing pillar-type drill. c Column-type drill. d Turret drill. e Multi-spindle drill. f Radial drill. II' Jig boring machine. h Vertical deep-hole drill. i Horizontal deep-hole drill. k Precision drill.
a particular machining sequence is specified with the relevant direction of feed, specifications and optimisation for cutting quality and cycle time. There are single-path and multiple-path machines, rotary cycle machines with transit tables, rotating tables, ring benches or drums. Switched
5.3 Milling Machines _ 5.3.2 Knee-Type Milling Machines
indexing tables are designed for automatic manufacturing processes. With the modular system, special-purpose drilling machines may be built up from the various drilling spindle and feed units. Multi-spindle drilling heads are employed for the manufacture of specific drilling patterns. Routing drilling machines in bridge constructions are designed for processing large workpieces. The bridge nms on rigidly mounted tracks of the desired length. Multispindle drilling units may be employed with moving drilling spindles. One special-purpose construction is the numerically controlled pcb (printed circuit board) drilling machine where performance depends on the number of drilling cycles per time unit which may be achieved.
5.3 Milling Machines 5.3.1 General
Figure 23. Structure of a radial drill (Raboma type radial boring machine , Hennann Kolb Maschinenfabrik, Cologne). The damping plate 1 supports an internal free-standing pillar 2 , on which the pipe sleeve 3, which may be rotated by 360°, is seated on bearings at 4 and 5. The hoisting motor 6 is switched via 7 to lifting gear 8, dlives the lead screw 11 via a gear 9 with a safety clutch ]0 and adjusts the height of the cantilever arm 12 and boring slide 13. The cantilever lifting stroke automatically cuts out at the end position, The slotted cantilever arm clamp J4 opens and closes automatically before and after the height adjustment of the cantilever arm 12 (lifting stroke control). The lubrication of the lead screw I} and the nuts, cantilever arm clamp 14 and clamp drive is via a central oil pump 15. To clamp the cantilever arm 12, the pipe sleeve 3 is held via a hydraulic clamping device (not shown) to the internal free-standing pillar 2. The clamping motor 16 drives a gearwheel pump with an integral pressure relief valve. Piston motion is transformed into the rotary motion of the clamping ring 17, causing eccentrically positioned cam levers 18 to lift the clamping bolts 19, and the double bevel ring 20 to be activated - axial motion prevents any skew in the cantilever arm during clamping. The drilling motor 21, switched via 22 aod controlled with 23, drives the drilling spindle 24 via variable gearing, from which the feed drive is derived. The hydraulically switched changeover and preselection of all drilling spindle speeds and feeds arc adjusted with knobs 25 and 26. 'The hand lever 27 controls the hydraulicaUy switched multiple-diSC clutch with an automatic spindle brake to the home position. The clutch operates with automatic delay when switched on for toothby-tooth gearwheel controL The starting handle 28 operates the feed clutch with overload protection and disengages the feed wheel 29 when switched on. There is automatic disconnection of the feed drive at both ends of the stroke of the drilling spindle 24. The pressure switch 30 releases the drilling spindle 24 from the gearing for tool change. Butterfly-type levers 31 disengage the feed drive on the control head and permit rapid manual adjustment of drilling spindle 24. The graduated dial 32 gives accurate adjustment of the drilling depth with trip range over the entire stroke of drilling spindle (impact boring). The drilling spindle weight is compensated via a counterweight. The boring slide 13 is adjustable via the handwheel 33 on the cantilever arm. Clamping is on a track lined with a hardened steel band 34, and this may be tightened with the clamping shoe eccentrically positioned on the clamping shaft, which is driven by the hydraulic damping unit contained in the boring slide, with an automatic reset device. The clamping device for the rotating motion of the cantilever arm 12 and the slide 13, activated by a pushbutton 35, are interconnected via positively actuated electrical sequence switching. The power supply from the switch cabinet 36 is to the coolant submersible pump 37 and via sliprings 38 to an electrical control unit (not shown) in the rear of the cantilever arm, and thence to operating units and motors. The mushroom-type switch 39 is the ~Main OFF" switch. Illumination of the operating field is by the direct-axis lamp 40.
Milling machines are characterised by three or more axes 0/ motion which are assigned to the tool or workpiece. The position of the axes of motion determines the type of machine. Other criteria determining their classification include kinematic considerations and the frame construction. Technological advantages of specific milling processes and the frequency of their use have led to tried and tested designs (Fig. 26), where their characteristics include the main spindle diameter, table surface area, main spindle location and type of control. A distinction is made between horizontal and vertical milling machines depending on the position of the main spindle. Small workpieces with complex machining operations are carried out on machines with several types of table motion. In the case of large, bulky workpieces. it is preferable for the feed motion to be executed by the tool. A distinction is made accordingly between knee-type and bed-type milling machines. The tools are clamped directly or via a cutter arbor. Special-purpose machining tasks may require modifications using rotary tables, dividing heads, angle cutters, precision measuring devices and digital displays.
5.3.2 Knee-Type Milling Machines These are designed as hurizontal, vertical or universal milling machines (Fig. 27). These are particularly suitable for a variety of machining applications in individual and small-scale serial production because of the simple positioning of the workpiece in all machining directions and favourable accessibility. The machine frame consists of a baseplate and column which accommodates the main drive, main spindle and guide of the knee. Horizontal milling machines have an axially adjustable counter clamp on the column. The knee which carries the cross-slide and milling table is clamped during the milling process. The cross-feed may also be effected ,ia a moving milling spindle box or motion of the spindle sleeve. The individual drive is increasingly tending to supersede the central drive, where the feed is derived from the main gearing and transmission is executed from a telescopic drive shaft to the knee. Heavy-duty knee-type milling machines have permanently fixed main spindles, while universal milling machines have a clamping table which may be rotated about the vertical axis. A vertical or universal milling head may often replace a counter clamp where the basic construction is the sanle.
Control Systems. A distinction is made between manual control by way of selector switches or pushbutton switches, program control with pre-programmed or freely programmable machining processes and numerical con-
Manufacturing Systems. 5 Metal-Cutting Machine Tools
f--- --1360 - - ----;
Figure 24. Single-column jig drill with an eightspindle turret head (Hermann Kolb Maschinenfabrik, Cologne).
machine (Fig. 28). The main spindle is normally positioned vertically, or occasionally horizontally in the mill· ing unit. The machine bed accommodates the cross-table in two inverted guideways or three-way flat gUideways. Modified forms may have the moving machine column in the cross direction and the work table moving in the longi· tudinal direction. The milling slide is guided on the col· umn and is tilt-free, partially supported and fixed with a clamping device.
2
3
Figure 2S. Turnstile turret (Hermann Kolb Maschincnfabrik, Cologne). The drilling spindle drive 6 is via the gearwheels 3 and 4. The cross-toothed rings 1, 5 are for indexing. The hydraulically powered motion is via the gearwheel 3 and the piston 2.
trol with feed motion in all axes effected via d.c. motors and ballscrews. s.~.~
Bed·Type Milling Machines
These are constructed in a number of different arrangements as single-column and two-column machines (Fig. 26c, d), The table is supponed by way of a rigid machine bed, the milling slides are height·adjustable, while the torsionally stiff construction provides for good absorption of the cutting forces and workpiece weights. Load capacity is high, as is the accuracy of table guide.
Single·Column Bed-Type Milling Machine. With its milling unit guide vertical to the column, this is capable of applications similar to those of the knee-type milling
HOrU:ontal Plane Milling Machines. These are intended for the economic manufacture of large and long workpieces and are built up from a number of modules graded according to size. The central module is the long machine bed where the work table moves longways only. One or two columns are arranged along the side of the bed, depending on the level of expansion. Up to four mill· ing machines may be employed for simultaneous machining of a number of surfaces, guided at the columns, with an additional arm or on the crossbar. The individual modules may be combined to produce a variety of different constructions. Modification may be based on standard classes of table widths, clamping lengths and clearance heights and equipment with tlifferent milling units. The double-column construction is characterised by its high rigidity and flexibility. The two columns which guide the crossbar in the rigid or moving double-column design are connected to the bed and traverse within a torsionally rigid frame. This either runs along guides attached on two sides to the bed or is laid completely separately in the base in machines with moving double columns (gantry style), whereas the workpiece clamping table is stationary. Two synchronised feed drives are required for the positioning motion of the crossbar and double column. Other design characteristics of horizontal plane machines include replaceable milling units, weight compensation of the crossbar for sensitive vertical advance, infinitely variable speeds and feeds, motorised tool clamping, automatic clamping devices, central lubrication, automatic tool raising and lowering and a central control desk for all functions. Additional attachments provide for over· all machining of workpieces in a single tool setting. Milling units with drive motors and switchgear are available in a number of designs, e,g. as slide milling units, tailstock milling units and rotating milling units, Feed motion is executed via separate infinitely variable d,c. motors and ballscrews, while table or double-column
5.3 Milling Machines. 5.3.3 Bed-Type Milling Machines
-------,
x
~ .
a
z
Iy
c
b
______ Ji
x
Iz e
d
g ,,..-I
+
I
X \
-',
z m
p
o
n
Figure :16_ Types of milling machine (coordinate designation in accordance with DIN 66217). Knee-type milling machines in horizontal and vertical design: a, b three axes in the knee or two axes in the knee and one axis in the spindle box; c, d bed-type milling machine in horizontal and vertical design; e, f two axes in the cross-table and one axis in the spindle box, or g. h one axis each in table, column and spindle box. Horizontal-plane milling machines: one axis in the table and two axes each in the horizontal or vertical milling units: i single-column horizontalplane milling machine with cantilever ann; k single-column hOrizontal-plane milling machine; I double-column milling machine; m cross-milling machine; n <..-ylindrical milling machine, with one rotary axis for workpiece feed. Special-purpose milling machines: 0 roller milling machine, with two axes and onc rotating motion in milling unit, and two axes (one rotary axis) in table slide; p thread-cutting milling machine with one rotary axis for workpiece feed and two axes in the milling slide.
5
6
7
8
Figure Z7. Horizontal knee-type milling machine, now also available with three-axis numerical control (DIAG (now trading as Werner & Kolb) Werk Fritz Werner, Berlin). The table clamping area is 1500 x 400 nun 2 . The power of the main drive P is 12 (or 14) kW. There are 22 speeds of rotation, with
motion is normally effected via a helix or worm cutting gear rack.
Horizontal Milling Machines. These are designed as Simplified horizontal-plane milling machines. Columns and projecting beds are normally employed which are bolted to a box bed on one or two sides and accommodate horizontal milling units. The main application is for horizontalplane machining with cutting heads at high cutting rates in serial production. Feed motion is frequently limited to longitudinal table direction. Program control is standard for automatic tool advance and raising and lowering in
3
2
Figure :18. Singie-column vertical milling machine (Droop & Rein, Bielefeld). The main spindle diameter is 130 mm. The working space is given by an x-dimension of 2300 mm, a y-dimension of 950 mm and a z-dimension of 800 mm. The table clamping area is 3000 x 1000 mm'. The main drive output rating is 30 kW. The 18 speeds vary from 31.5 to 1600 min~'. The feed is infinitely variable, in three ranges, from 3 to 3000 nun. Frame: 1 machine bed, 2 carriage, 3 table, 4 column, 5 saddle, 6 hydraulic-powered rotating milling unit, 7 chain for weight compensation. Drive: 8 d.c. current main motor, 9 tool clamping motor. Control: alternatives include manual input, programmable control, copy control and numerical controL
Manufacturing Systems. 5 Metal-Cutting Machine Tools
combination with automatic tailstock or slide clamping. Horizontal milling machines based on the modular system may be set up easily to accommodate specific user requirements when using simple main and feed drives. 5.~.4
These are suitable for the manufacture of complex threedimensional shapes which are scanned by sensors from a master or model. The copy control systems differ, depending on the principle of operation or scanning method. Electrical, hydraulic and electrohydraulic systems have been gaining in popularity in recent years. The tool is guided along a curve determined by a model via continuous changes in two or three feed motions arranged vertically in relation to one another. The copy control amplifier generates the feed signals for the actuators from a signal in proportion to the tracer deflection. An automatic line switch with two scan feeds and one sliding feed permits three-dimensional machining with two-dimensional control as well. The model is scanned in lines. After each cycle, it is advanced by the preselected interval between lines. Three-dimensional control is suitable for three scan feeds with either two separate sensors or one special sensor which reacts simultaneously to both radial and axial deflection. Standard bed-type milling machines are frequently fitted with an arm for the sensor device. Special-purpose machines carry out specific copy machining. 5.~.5
Machines for Circular Milling
One of the characteristics of a machine for circular milling is the unit for the circular feed motion for producing cylindrical surfaces. The circular feed may be superimposed by a plunge-cut feed or longitudinal feed. For example, crankshaft circular milling. Circular milling operations may also be carried out on standard vertical milling machines, a rotary table being
required for the circulating feed motion instead of the machine table. The main spindles of the milling unit in machines for circular milling and the workpiece spindles of the vice are arranged parallel to one another. The workpiece is caught in the centred scrolling jaws. Synchronous drive of the two workpiece spindles is carried out by a worm gear mechanism, vice and adjustable transverse milling unit positioned on a common machine bed. An additional slide is provided for lengthwise adjustment of the milling units and back rest. 5.~.6
J
Copy Milling Machines
Universal Milling Machines
These have a broad range of applications in machine tool construction and the construction of jigs and fixtures. They have a wide, finely-stepped speed and feed range, a high degree of machining accuracy and a variable modular program of additional attachments. It is therefore possible to carry out milling, drilling, turning, broaching and grinding within a single chucking operation. A variety of different clamping and parts devices is used to manufacture complex shapes.
Main Characteristic. The simple basic machine is of knee-type construction (Fig. 29). Height adjustment and lengthwise feed are via the knee, cross-feed via the milling headstock. A moving machine column may be added for machining heavy-duty workpieces, also a moving crossfeed milling headstock. All feed motion is assigned to the tool.
9- - --il1lIlII/,
$
6
Figure 2:9. Universal milling and drilling machine tool (Friedrich Deckel AG, Munich). An extended version of the basic model with automatic gear switching and stepper motors for the feed motion, and eqUipped with numerical control (freely programmable memory via keyboard control or continuous-path control) designed for three or four axes. 1 moving vertical milling head, 2 spinille socket, 3 main spindle with extending spindle sleeve, 4 main drive motor (brake motor), 5 inftnitely variable feed motor (d.c.), 6 knee slide with work table guides, 7 table slide, 8 angle table, 9 feed control, reCiprocal interlocking with clamps, 10 control desk and high-reSOlution digital display.
Additional Attachments. These include a counter holder, a vertical milling head, an angular milling head, a rapid advance vertical milling head, a broaching device, a drilling head, a grinding attachment, an angle table (fixed, rotating), a transit table, a rotary table, a rotary table with optical adjustment, a part head, a swaging device, a helical milling attachment, and precision measuring equipment. 5.~. 7
Special-purpose Milling Machines
These are intended for a single purpose, or for machining a limited number of different workpieces. Thread·Cutting Milling Machines. The structure and kinematics of these are similar to those in machines for circular milling. Longitudinal thread-cutting milling machines have the same structure as screw-cutting lathes with an extra milling support. Short thread-cutting milling machines operate with multiple thread section cutters using the plunge-cut process. RoUer Milling Machines. These are used to cut spur gears, worm gears and special toothed gears. Their special characteristic is a roller train of gears and change gear box generating positive motion between workpiece and tool rotation. Engraving Machines. These operate on the basis of a pantograph which may be moved on a plane or threedimensionally. The transmission ratio between the sensor tip and milling head may be set by a slide and transmitted to the workpiece by scanning the guide template. Slotting and Grooving Machines. These produce nuts, keyways, splined shafts and grooves from pre-programmed motions which are normally generated automatically. Rotary Slotting Machines. These represent a special design for the manufacture of winding slots and air flutes in generator rotors and differ from the above-mentioned machines in view of their size.
5.5 Machining Centres
Cam Forming and Profiling Machines. These serve to manufacture control cams with end-milling cutters by superimposing workpiece rotation over the milling table longitudinal motion. Control is effected via template scanning. Cross-<:ut Milling Machines. These are machines composed of two milling units for two-sided machining of unmachined parts to a given length. Simultaneous centring is also an option.
5.4 Horizontal Boring and Milling Machines The construction of these with a number of options for motion and adjustment permits the machining of bore holes in true alignment and of rotating bearing front surfaces and the machining of large, bulky items. A broad field of applications has produced a variety of designs and sizes (Fig. ~O) .
Construction. The drives and main spindles are designed accordingly for drilling and milling machining in the main. High speed and feed ranges are characteristic, with four or more axes of motion and a comprehensive range of additional attachments. Simultaneous execution of machining operations independently of one another is possible with a separate spindle bearing and faceplate bearing. Lathes with a high degree of part accuracy and backlash-free table drive systems permit bore holes to be machined in true alignment in the transition process. The structural design of the main spindle system consists of a main spindle bearing with a faceplate permanently attached to spindle bushing, or else the separate location of the main spindle and the faceplate which are then able to rotate at varying speeds. The machine bed in table drilling machines and milling machines accommodates a fixed column, the cross-slide and a steady rest. The spindle box with the retracting main spindle runs laterally along the side column or centrally along the arc-shaped column, and the height is adjustable. The cross-slide normally supports a rotating table. The drill rod is guided inside the adjustable steady rest for heavy-duty and precision boring
operations with substantial excess lengths. The main drive is designed for a broad speed range and rapid changes in
a
c
b2
Figure 30. Schematic diagram of basic constructions of horizontal dtilling and milling machines (coordinate designation in au.:onlance with DIN 66 2]7): a, table-type drilling and milling machine; b routing drilling and milling machine. 1 standard design, 2. with column moving crosswise and endwise; c two-way design (Planer type).
the direction of rotation, the main spindle for rapid braking. These drives normally take the form of d.c. motors with infinitely variable speed settings and mechanical transmission units switched in sequence. There are individual drives for all axes of motion, each with high rapid advance speed, short acceleration and braking times, a broad speed range and freedom from backlash. The individual machine slides are fitted with damping devices. The use of this flexible type of machine is mainly restricted by the size and weight of the workpieces. Routing drilling and milling machines are suitable for all heavy-duty workpieces. All possible settings and feed motion combinations are available in a column and a spindle box. The workpiece is clamped to a permanently ftxed plate adjacent to the machine bed. There are two different types of column motion. In simple designs, the column is pushed along the bed at right angles to the main spindle axis. In order to guide the tool to the workpiece, the main spindle may be adjusted axially. An extended area of applications is achieved by way of the crossmotion of the column. In bearing sleeve drilling and milling machines, the main spindle is supported within a rigid extending sleeve bearing. As a result, greater absorption of cutting forces is possible with a larger spindle drilling surface area. Angular milling heads may be attached to the bearing sleeve. A wide variety of special-purpose constructions with rotating spindle boxes or turning columns have emerged for special operations. More recent developments in horizontal drilling and milling machines have combined the advantages of these two types. The development of the machining centre is currently underway whereby adaptation to the machining task is carried out by both the tool and the workpiece.
5.5 Machining Centres These are designed for the numerically controlled machining of complex prismatic workpieces in one tool setting. In addition to the various machining operations to be carried out, there are a number of manufacturing processes which are available, i.e. drilling and milling. Machining centres have a high level of flexibility in addition to a high level of automation. Typical is automatic operation encompassing both the machining sequence with all the path and switching functions and the rapid repeatability of individual machining processes and tool change from the corresponding size of magazine. The latest developments in machining centres include pallet-change devices, drilling head changers and tool magazine changers. Machining centres have at least three numerically controlled linear axes, which may be supplemented by two rotary axes (Fig. ~ 1). The type of assignment of the axes of motion to the workpiece carriers (table) or tool carriers (spindle) determines the type of machining centre, A horizontal or vertical position of the main spindle is available. The constructions of horizontal machining centres have emerged as a result of the different assignment of the axes of motion. Figure ~2 shows the most common type~. The main characteristics of the machining centre include the type and number of controlled axes, curting power, speed and feed range, length of operating paths, table area and table load application, resolution, positioning tolerance and number of tools in magazines.
Tool Storage. Turrets or magazines arc available. Up to 40 tools may be stored at full capacity in the disc-shaped revolving magaZines. Larger magazines are normally designed as chaining magazines. In addition to the basic
Manufacturing Systems. 5 Metal-Cutting Machine Tools
f~J ~+-o- . ~ .
b
~
~
Figure 3~. Basic types of tool magazines. Constructions; a revolv-
Figure ~l. Dcfmition of the direction of motion on a four-axis
ing magazine, b chain magazine. Position of tools: c drum-shaped,
machining centre (DIAG (now trading as Werner & Kolb) Week
d turnstile-shaped. Direction of assembly: e parallel, f vertica1. 1 Tool location. 2 TooL 3 Magazine axis. 4 Direction of assembly.
Fritz Werner, Berlin): 1 main spindle, 2 rotating table, 3 tool magazine, 4 tool changer, 5 numerical control.
Assignment of axes of motion 1 1001 axis 2 tool axes
jY
jY
- (j3 -
~
t9 ~ /
Z
-/E3-Z
~
e
3 tool axes
j~x - fl3 -
/' f
@ Z
/ X
IY/X
to be covered during a tool change. Simple changing units handle one tool each and are, therefore, only economic to employ in combination with a tool turret or as auxiliary grippers, as they carry out the entire tool change process, i.e. including the selection motion of the turret without any intermediate storage capacity in dead time. Double tool changers remove one tool from the main spindle and the magazine or intermediate store and reverse the two positions simultaneously. This enables shorter tool change times to be achieved. Figure 34 shows a design where the arms of the tool changer are pOSitioned at an angle of 90° to one another. This enables tools to be changed in a short time from a revolving magazine either attached at one side or on the colunm. In Fig_ 35 a tool change system is shown which is known as a universal unit and employs the principle of the single tool changer in combination with a double one. The tool change may also be carried out directly from the magazine. The correct motion of the magazine in the
/fP.
~
Figure l1:. Construction of horizontal machining centres.
shape of the magazine, the location of the tool and the direction of motion for assembly in relation to the magazine axis is also significant. A basic distinction may be made between turnstile and drum-shaped tool arrangements and parallel and vertical motion for assembly (Fig. 33). The construction and arrangement of the magazine determine the kinematics of the tool change. Tool recognition is carried out via tool coding (the individual tools are coded), location coding (the magazine locations must be labelled with the tools in the correct order), or, since the introduction of CNC control, via electronic accounting (combination of both types of coding). Dead time is affected by tool change time. This is dependent on the number of motions required (degree of freedom) and the size of the paths or angles which have
Fieu:re M. Double gripper for auto matic tool change (Friedrich Deckel AG, Munich). The function sequence of the tool changer 2 begins with the gripper taking one tool each from the magazine 3 and the main spindle 1 respectively. The tools are removed from
the main spindle and magazine simultaneously via motion in the direction of the Z-axis. It then swings through 180 0 to synchronise assembly motion in the opposite Z-direction with the release of the gripper fInger. A Removal and motion for assembly in the direction of the Z-axis, Alan the magazine, Az on the main spindle. B Rotating motion.
5.6 Planing, Shaping and Slotting Machines. 5.6.2 Shaping and Slotting Machines
10
Figure ~S. Universal machining centre (Burkhardt & Weber GmbH & Co. KG, Reutlingen). The housing 1 supports the double rotating head 2, the revolving magazine 3 with space for 30 tools, the tool changer 4, the main drive 5, where P == 11 kW output, and the stepper gear mechanism 6. To change tools, the tools are gripped by the gripper 7 and changed at point 8 between the double rotating head and the magazine. Where a new tool is needed, the double rotating head carries out a rotation about 1800 around the axis 9, so that the positions of points 8 and 10 are reversed. Indexing of the rotating head is carried out via piston 11.
direction of the main spindle axis is a basic requirement for this. Dead times for loading and unloading and setting up and clamping the workpieces may be run parallel to productive time where tool change devices are in use. Where pallet change systems are used the workpieces clamped on the pallet are brought to the working space via linear or rotary motion of the pallet in tum and subsequently taken out of this space. Pallet storage systems enable a number of workpieces to be prepared and thereby to provide for automatic tool changes within longer production periods (Fig. ~6).
5.6 Planing, Shaping and Slotting Machines 5.6.1 Planing Machines Single-Column Planing Machines. These provide for machining of bulky items (Fig. ~7). Constituent modules are: bed, column and height-adjustable arm, with, where required, a removable auxiliary column for support. Workpieces protruding at the side may be held by a roller support bed arranged at the open side. The main drive of the table is by d.c. motor with clutch gearing and gear rack, or else hydraulic. The tool slides are arranged on the arm and column. Two-Column Planing Machines. These have a closed double-column frame with a height -adjustable crossbar (Fig. ~8). The tool slides are recommended for use on the crossbar, although they are also provided on the columns in large-scale machines. Additional Attachments. Milling or grinding attachments (Fig. ~9) may be placed on the crossbar with their own separate main and feed drives. 5.6.2 Shaping and Slotting Machines Horizontal Shaping and Slotting Machines. These are employed primarily for machining surfaces on smallto medium-scale workpieces (Fig. 40). The cutting motion of the tool is carried out by the ram, guided along the top side of the baseframe. The workpiece is supported by a table arranged at the top end of the baseframe 1. The ram 2 is driven either hydraulically or mechanically via a sliding block. The tool is taken up by the rotating bitretaining head 3 which places the tool in a vertical position. The bit vent lifts the tool from the workpiece on
Figure 36. Pallet storage system (Hiiller Hille GmbH, Ludwigsburg). Machining centre 1 is equipped with main spindle and tool 2 and tool store 3. The pallets 4 are taken up and transported internally by the work table in the machine. During pallet Changeovers, the pallet moves with the machined workpiece to point 5 and is shunted onto roller conveyor 6. The next pallet leaves the roller conveyor 7 and is attached to the work table at point 8. The pallets run through the pallet store in one direction only, transported via power-driven rollers. The bulkhead trolley 9 is used to transfer the pallets from roller conveyor 6 to roller conveyor 7.
Manufacturing Systems. 5 Metal-Cutting Machine Tools
Figure ~7. Single-column plan-
Figure~.
ing machine.
planing machine.
DoubJe-co)umn Figure 41. Vertical shaping and slotting machine with singJe-column and bed.
smaller vertical planing machines (Fig. 41) columns and beds consist of one element and in larger machines these are composed of two parts. The workpiece is carried by a cross table or additional rotary table. In smaller machines with stroke lengths of up to 630 mm, the mechanical drive is predominantly used in conjunction with gear indexing mechanisms and crank rotary belts. Hydraulic drives with larger stroke lengths have advantages over mechanical drives with continuous infmitely variable speeds and impact-free control.
Figure ~,. a Planing machine with milling attachment . b Planing machine with grinding attachment.
the return of the ram. The position of the ram stroke in relation to the table may be adjusted via a moving spindle. The sliding block 4 is connected at the top with the ram and at the bottom with a fixed point in the baseframe 1. The sliding block wheel 5 drives the sliding block via a sliding block counterweight 6. The rotating motion of the drive motor is transmitted to the sliding block via a cascade gear mechanism . The workpkce to be machined is clamped to the table 7 fitted with T-bolts which may be rotated around a horizontal axis in relation to the receiving table 8. The feed motion is carried out by the receiving table on the support 9 in a cross direction. The support is height-adjustable. Vertical Planing Macltittes. In these, the ram carries out either venical motioo or motion in two directions. In
5.7 Broaching Machines These are classified as external or internal, depending on the broaching process, and vertical or horizontal, depending on the position of the main axis. Chained and spec/alpurpose machines are considered to be special constmctions (e.g. for use in transfer lines) . Classification is carried out in terms of size, cutting motion and standardised according to main dimensions and supply ratings. Advantages of vertical constmction inclode low space requirement, no torsional stress on broaching tool as a result of dead weight, better coolant lubrication effect and good conditions for integration in transfer lines. Advantages of horizontal constmction are low set-up height, potentially larger stroke lengths, simpler infeed of heavyduty workpieces, and absence of any. requirement for excavated foundations or an operating platform. The machining accuracy of broaching machines depends primarily on sJideway of tool or workpiece. Figure 42
Figure 40. Horizontal shaping and slotting machine (Schlenker & Cie GmbH, Werkzeugmaschinenfabrik, Homberg): J baseframe, 2 ram, 3 bit retaining head,
4 sliding block, 5 sliding block wheel , 6 sliding block co unterweight, 7 table , 8 receiving table , 9 support.
5.8 Sawing and Filing Machines. 5.8.1 General
shows the structural design of a vertical external broaching machine. Broaching machines are employed in mass production. Chaining of broaching machines and transfer lines is possible using automation and feed systems. The frames of broaching machines have to be designed in such a way that they have high static and dynamic rigidity. The requirements are suitable ribbing, cell-type design and welded construction with damping surfaces. In internal broaching machines dual-cylinder construction has the advantage over single-cylinder design in that the propelling force and the guides are on one level, so that the machine is subjected to less bending stress. In external broaching machines, bending oscillation is generated where the colunm oscillates vertically in tertns of the direction of broaching. It is possible to obviate this problem with an extremely rigid joint between the column and the broaching machine. Cutting motion in broaching machines is normally mechanical or hydraulic. Mechanical drives are employed in all chained broaching machines and in some cases in horizontal external broaching machines. A variety of types of table are installed in broaching machines depending on the machining task. Clamping devices are required only in exceptional cases for internal broaching; normally simple receivers or receiving lugs are enough to secure the workpiece. A workpiece receiver 3 is shown for internal broaching in Fig. 4~. The workpiece 2 is pre-centred using three pins 1, while positioning is carried out by inserting the broaching needle. The clamping devices for external broaching are of a more complex structure. The main clamping system has
Figure 43. Workpiece receiver for internal broaching (Kurt Hoffmann, Maschinenfabrlk, Pforzheim): 1 centring pin, 2 work-
piece, 3 workpiece receiver.
to be self-locking, so that the clamping force is not reduced where there is a loss of operating power. A selflocking support system is shown in Fig. 44. The bolt 1 is pressed against the workpiece initially and secured by wedge 2 and bolt 3. Important potentially hazardous areas on broaching machines include the open, moving workpiece, the motion of the slide and parts table, and the motion of the loading and clamping equipment. The tool and cam strip are concealed beneath the working space; an enclosed superstructure is not nortnally required and may cause a hindrance during tool changes. Two-handed operation is common. Accidents due to the uncontrolled motion of tables are prevented by setpoint position control of those machine parts which are moved. Noise abatement is effected via the correct selection of the drive units and active insulation on installation of the broaching machine.
5.8 Sawing and Filing Machines 5.S.1 General These separate and generate cuts and cutouts with flat or uniaxially curved surfaces in workpieces made of metal, wood, glass, ceramics, stone and plastics. Multiple cutting tools are made of tool steel, rapid-machining steel or hard metals. The rotary or linear, continuous or oscillating cutting motion is carried out by the tool, as is the feed motion. The area of application of sawing machines
covers the entire range of single, serial and mass pro-
using vertical external broaching (Kurt Hoffmann , Maschinenfabrik , Pfonheim); I tool slide , 2 workpiece table , 3 rotating table, 4 hydraulic cylinder.
Figure
4~.
machine
as
Overall
an
layout
example
Figure 44. Self-locking workpiece support syslem (Kurt Hoffmann , Maschinenfabrik, Pforzheim): J bolt , 2 wedge, 3 securing bolt, 4 workpiece.
.-..:r:. -.::.:.1:11
Manufacturing Systems • 5 Metal-Cutting Machine Tools
duction. Filing machines are used predominantly in tool, jigs and fixtures and equipment construction. Friction cutting and blanking machines and electrical discharge machining, e.g. hot-wire cutting machines, are only related to sawing machines from the kinematic point of view; however, in terms of productions systems they are viewed as metal removal machines.
C1assific:ation. Sawing machines are classified in terms of cold and hot sawing machines. Circular, bandsawing and fIling and bandsawing and power hacksawing machines for fIling are classified in terms of kinematics. Other characteristics determining their classification include designation of the tool used, type of drive and control and level of automation.
S.8.2 Ciradar Sawing Machines A variety of constructions (Fig. 4S) are categorised in terms of the direction and type of feed motion which may be horizontal or vertical and linear or curved and may be cartied out by a sawing slide. This consists of the sawing blade shaft for carrying the tool, gearing, drive motor, housing and guides, and is normally driven hydraulically via a cylinder and piston or hydraulic motor and ballscrew drive. Automatically activated feed adjustment prevents excess loading of the tool and machine specifically when sawing variable cross-sections, such as T-beams for example, by limiting the pressure in the hydraulic system and thereby the feedrate where the feed force is increased.
a L-________________
~
The self-activating workpiece loading which is required for automatic sawing machines is normally generated hydraulically, where the feed may only be switched on at the end of the clamping process (Fig. 46). In order to be able to cut badly deformed workpieces such as warm sheared billets smoothly, the vertical clamping device is designed as a float device in some machines. Hoisting mechanisms at the sectional clamping point prevent the sawing blade from becoming stuck after completing the cut. The clamping chucks are designed as parallel, horizontal or V-type constructions. Multiple clamping is more economicaL
The material supply (workpiece feed) takes place on rollers or roller blocks which are activated as part of the automatic sequence of motion of the machine, normally against a rotating stop. The largest workpiece dimensions for circular and square-shaped materials and sections are dependent on the diameter of the cutting blade. The range of cutting speeds for steel and cast metal materials is between 5 and 40 m/min with HSS blades and between 60 and 200 m/min for hard metals. When sawing non-ferrous metals, 500 to 1600 m/min is standard. The maximum feedrate for steel and cast iron materials amounts to approx. 1250 mm/min and 2300 mm/min and more for non-ferrous metal machining. The required drive rating for large-scale sawing blade diameters of d = 1120 mm amounts to approximately 55 kW. S.8.~
Baneisawing and Filing Machines
These are designed in horizontal and vertical constructions. The most important attributes include roller guides for the saw or fIle bands and their arrangement inside the machine frame or saw frame (Fig. 47). There are usually two rollers (three for wide clearances), with a coating (e.g. rubber) to preserve the tooth offset, to guide the continuous welded saw or fIle band or file chain. One of the rollers may be rotated to provide smooth band operation within narrow tolerances. The saw Of Hle band nOf-
b L--L________________
~
mally runs between tempered or hard metal guides directly in front of and behind the cut. 1be band speed is mainly iillmitely adjustable via belt drives and achieves rates of between 10 and 1200 m/min and more in universal machine applications. Drive ratings of under 1 kW are required for small cutting speeds at over 1000 m/min and 4 kW and over particularly for frictional cutting. Mechanical or hydraulic and sometimes numerically controlled feed units are employed depending on the machining task.
I ~H---~~TS;;--.;.-~t---- - -
d L-__________-'--' Figure 45. Designs of cold circular sawing machines. a Circular saw cutter with horizontal feed motion (Heller, Ntirtingen). b Longitudinal cross-section circular sawing machine with horizontal feed motion (Treenjaeger, Euskirchen). c Circular sawing machine with ve1tical feed motion (Ohler, Remscheid). d Circular sawing machine with arc-shaped feed motion (Kaltenbach, L6rrach).
---4 I
Figure 46. Hydraulically operated workpiece clamping device for <:old circular saw (Heller, Ntirtingen): 1 hydraulic horizontal damping cylinder, 2 hydraulic vertical clamping '-l'linder, .3 case-hardened, removable clamping chucks, 4 workpiece support.
5.9 Grinding Machines. 5.9.3 Cylindrical Grinding Machines
5.9 Grinding Machines S.9.1 General
Figure 47. Belt saw and filing machine with two belt rollers and vertical belt guide (M6ssner, Mutlangen): / drive motor and belt pulley, 2 belt rollers, .3 belt guide (removable and adjustable), 4 tilting table, 5" adjustment control for belt tension, 6 belt weld attachment.
S.8.4 Machines for Power Hacksawing and Filing There are power hacksaws with horizontal cutting motion and machines for power hacksawing and ruing with vertical cutting motion. Power hacksaws are fitted with a ballscrew drive to generate oscillating stroke motion (Fig. 48). The feed motion is normally generated by force due to weight, or by hydraulics in automatic machine tools. The saw blade is raised during the rapid return with the relevant equipment. The double stroke frequency is between 30 and 150 min-I, with an installed drive rating of up to 6 kW.
Classification. The main criterion is the type of surface produced on the workpiece. On horizontal plane grinding machines (surface grinding machines), flat surfaces are produced, while circular cylindrical surfaces are processed on cylindrical grinding machines. Other constructions include screw thread grinding machines, gear grinding machines and form grinders for machining specific forms specified by forming tools at will, and copy grinding machines for generating any form of surface via mechanical control of the feed motion. Other criteria for classification include the position of the area to be machined, so that a distinction is made between external and internal cylindrical grinders, the most effective surface of the grinding wheel, which gives cylindrical and surface grinders, and the type of feed motion, which includes face and cross grinders (plungecut grinding machines). A further distinction is made, according to the type of tool carrier, between centred grinding machines, clamping grinding machines and centreless cylindrical grinding machines; depending on the area of application, these are known as abrasive cutting machines, grinding machines, finishing machines and roughing machines, where the latter are normally distinguished depending on the machining accuracy which may be achieved and the maximum metal removal rates over time.
5.9.2 Surface Grinding Machines These are designed with a horizontal or vertical main spindle and a longitudinal or rotating table. Sectional wheels are normally used with rotaty tables. The oscillation grinding method is frequently employed. Backlashfree prestressed roller bearings and electromechanical drives with infinitely variable d.c. motors and ballscrew drives are required for full-contour deep grinding operations. Figure 49 shows a longitudinal surface grinding machine with a horizontal main spindle. The machine bed carries the sliding and roller bearing cross-slide for the longitudinal and cross motion. The feed motion is carried out electromechanically via controllable three-phase or d.c. drives which are either stepped or infinitely variable. The longitudinal table is driven hydraulically or electromechanically and is infinitely variable. The rapid-advance feed motion is generated via three-phase asynchronous motors or via solenoids with synchronous or stepper motors for rough or finish feed. The speed of the grinding wheel is frequently infmitely variable via statically or dynamically operating frequency transformeres. Figure SO shows a rotary table surface grinding machine with a horizontal main spindle. Alternatively, this may be constructed using a vertical spindle and sectional grinding wheels. A number of design principles for surface grinding machines with horizontal main spindles and alternative motion assignment are shown in Fig. SI.
Cylindrical Grinding Machines
Figure 48. Drive of power hacksaw machine for curved shaping
S.9.~
and cutting motion (Stolzer, Achern): 1 saw frame, 2 operating crank, 3 guide rollers, 4 guide raiL 5 saw bow, 6 hydrJ.ulic cylinder, 7 point of rotation of saw frame. Kinematic generation: The stroke motion of the saw frame 1 is generated with the operating crank
The Norton process is employed for short and mediumlength workpieces with a vertical grinding headstock and workpiece moving lengthwise along the grinding wheel. Long workpieces are ground on machines based on the Landis process. The headstock with the wheel is thereby moved along the stationary workpiece, which reduces the length of bed required.
2. The frame is guided with the rollers 3 along the guide rails 4.
The pressure in the operating stroke and lift in the return stroke of the saw bow 5 are achieved via the operation of the hydraulic cylinder 6, where the rotation takes place about point 7.
Manufacturing Systems. 5 Metal-Cutting Machine Tools
15 12 16
13
6 5 3
2
Figure 49. Longitudinal surface grinder with horizontal main spindle: 1 machine bed, 2 coolant lubrication pot, 3 cross-slide, 4 grinding table, 5 spray protection, 6 grinding wheel, 7 exhaust device, 8 protective cover, 9 trinuning device, 10 main headstock, 11 feeder arm, 12 main spindle drive motor, 13 coolant lubrication supply, 14 free-standing pillar, 15 control pancl, 16 control cabinet.
The machine bed is frequently designed as a welded construction on small and medium-scale machines and the main spindles are positioned in hydrodynamic multi-surface friction bearings designed for circumferential speeds of the grinding wheel of up to 60 m/s. The machines are also frequently designed for plunge-cut as well as longitudinal grinding; angled plunge-cut grinding is especially common in mass production, where a lengthwise orientation of the workpiece cannot be avoided. A measuring control system is part of the standard equipment in medium-scale serial and mass production. There are a
Figure SO. Rotating table surface grinding machine with horizontal main spindle: 1 machine bed, 2 hydraulic unit, 3 crossslide, 4 spray protection, 5 control cabinet, 6 rotating table, 7 grinding wheel, 8 protective cover, 9 trimming device, 10 main headstock, 11 feeder arm, 12 main spindle drive, 13 free-standing pillar, 14 control panel.
number of longitudinal measuring heads for both automatic and manual positioning. The main headstock or table in universal cylindrical grinding machines may be rotated for tapered grinding and a rotating internal grinding spindle may be accessed for machining drill holes. The external diameter, rough shoulders or internal diameter of a workpiece may be machined on NC cylindrical grinding machines, and a variety of radii and forms may be ground using continuous path control. Figure 52 shows the controlled axes of an NC external cylindrical grinding machine. The machining axes X and Z are assigned to the main headstock, the control of the diameter measuring head is carried out in the Y-axis and the
tool centre point of the grinding space is set via the Waxis. The length measuring head is guided to the shoulder to be measured via the U-axis.
Centreless Cylindrical Grinding Machines. These machines are employed in highly automated environments in mass production. The operating range is between 0.1
Figure St. Design principles of surface grinding machines with a horizontal main spindle. a Support design I, plunging pillar, crossslide. b Support design II, stationary pillar, cross-slide. c Traversing design I, stationary pillar, internal arm. d Traversing design II, stationary pillar, external arm. e Pillar slide design, table base with integrated pillar.
Fipre S2. Controlled axes of NC external cylindrical grinding machine (Schaudt GmbH, Stuttgart-Hedelfingen): X-axis (main headstock), Z-axis (workpiece slide), U-axis (cross-positioning length measuring head), V-axis (control of diameter measuring head), W-axis (longitudinal positioning of diameter measuring head).
5.9 Grinding Machines. 5.9.3 Cylindrical Grinding Machines
and 400 mm. The workpieces lie lengthwise on a receiver between the grinding wheel and regulating wheel. As the machining forces are absorbed efficiently by the regulating wheel adjacent to the workpiece, which is normally the same size as the grinding wheel, long or thin workpieces may also be ground without bending or torsional stress occurring at high rAtes of metal removal (Fig. 5~). The main headstock is bolted rigidly to the machine bed, while the control headstock and workpiece receiver clamp are arranged on a slide which carries out the infeed motion. In designs with a main and control headstock slide arrangement, the grinding slide carries out the infeed motion and the workpiece receiver is stationary inside the machine bed and is only height-adjustable. This principle is applied above all for heavy-duty workpieces, so that the loading equipment does not require adjustment. The regulating wheel and grinding wheel are either floating or reversible. The regulating wheel is rotated by a small angle to generate a longitudinal feed for the workpiece with centreless grinding in continuous operation. Most of the machines are also capable of centreless plunge-cut grinding. The diameters and widths of the grinding wheels go up to 650 mm, the regulating wheel always having a smaller diameter than the grinding wheel. Hydrodynamic multisurface friction bearings are normally employed for both spindles. The spindle of the regulating wheel is frequently driven by an infmitely variable d.c. motor via a worm and
5 6
1
8 9
10
12 IJ
a worm gear.
Automatic feed, loading and unloading devices are also integral to the machine tool system and chaining is also available. Other functions are also automatic, such as trimming the grinding wheel, offsetting the grinding wheel wear on the trimming attachment, and compensating for the trimming factor with the precision feed and measuring control system, in addition to workpiece trAnsportation.
'--- - 1520---
....
L..-----2600-----~
Internal Cylindrical Grinding Machines. The most important modules of an internal cylindrical grinding machine with an additional surface grinding attachment are shown in Fig. 54. The cross-slide 19 for the infeed motion with the drive motor 10 of the main spindles 11 is located on a crossbar 20 rigidly bolted to the machine bed. The lengthwise and feed motion is carried out via the table 1 on friction bearings in the machine hed onto
which the workpiece headstock 3 is rigitlly mounted. The spindle box may be rotated on the backing plate 16 to grind taper bores. The workpiece spindle is driven via an asynchronous motor 24 and sliding gear transmission. The surface grinding attachment 8 may be rotated into posFigure S4. Internal cylindrical grinding machine assemblies (formerly Jung Schleifmaschinen, H. Gaub, Berlin): 1 longitudinal slide, 2 table stops, 3 workpiece spindle box, 4 clamping chucks, 5 trimming device for bore grinding, 6 trimming device for crown grinding wheel (surface grinding attachment), 7 crown grinding wheel for surface grinding attachment, 8 swivel surface grinding attachment, 9 grinding wheel for internal cylindrical grinding, 10 main spindle drive motor, 11 main spindle, 12 grinding support, 13 control desk, 14 feed system for cross-slide, 15 switch cabinet, 16 backing plate, 17 coolant lubrication resenroir, 18 short·stroke attachment, 19 cross-slide, 20 crossbar for cross-slide, 21 electrical switch cabinet, 22 hydraulics cabinet, 23 protective hood, 24 drive motor for workpiece spindle. Figure S~. Centrcless cylindrical grinding machine: I control desk, 2 trimming device for grinding wheel, 3 main headstock, 4 grinding wheel, 5 coolant lubrication supply, 6 workpiece, 7 regulating wheel headstock, 8 trimming device for regUlating wheel,
9 regulating wheel headstock slide, 10 regulating wheel, 1 J supporting arm, 12 machine bed, 13 feed device, 14 grinding headstock slide.
ition for machining, may be moved in a lengthwise direction on a bed baseplate and may be rotated by up to 10° to grind surface areas on workpieces with taper bores. A short stroke attachment 18 is employed to generate a low
Manufacturing Systems. 5 Metal-Cutting Machine Tools
5
]
Figure SSe Flexible cell-type grinding centre for external, internal and form grinding in single chucking (Schaudt Maschinenbau GmbH, Stuttgart): I workpiece headstock, 2 tailstock, 3 external grinding spindle, 4 internal grinding spindle, 5 magazine for internal grinding pOints, 6 grinding point changer, 7 double-column loader for workpiece change, 8 automatic grinding wheel changer, 9 CNC control for up to five axes, 10 measuring system for continuous diam-
eter and length measurement during grinding process.
level of oscillation in the table of up to 6 mm, which, in tum, is operated with an eccentric press driven hy a brake motor via a worm gear.
5.9.4 Screw Thread Grinding Machines Machines are classified according to single or multi-sectional lengthwise grinding and multi-sectional plunge-cut grinding of screw threads.
5.905 Gear Grinding Machines Machines are classified according to the processes to be carried out by them. Part and continuous rolling processes are called after their inventors, i.e. Maag, Niles, Kolb and Reishauer.
5.9.6 Development Trends The introduction to high-speed grinding and the latest technology requires higher static and dynamic rigidity in machines, high-powered drives, improved spindle bearings, coolant lubrication systems and safety devices, such as equipment for rapid balancing at operating speeds. There is a clear trend in favour of plunge-cut grinding with boron nitride and diamond wheels in the grinding machine range. Other development trends include higher, infmitely variable peripheral speeds for workpiece and grinding, improved fme-feed systems, roller and hydrostatic guides, increasing employment of diamond trimming rollers, infeed, loading and unloading equipment designed for the relevant workpiece and improved control systems (measuring controllers, NC controllers, AC systems). Machines for external and internal grinding have emerged as part of the overall machining process, as is the case with the machine shown in Fig. 55, where both headstocks are arranged parallel to one another on a rotating table.
manual honing machines. The rotating motion is executed by the honing tool and the typical stroke required for honing is carried out by hand. Production times are minimised by way of mechanically generated stroke motion . Honing of external cylindrical areas is performed by rotating the workpiece via the main spindle. The stroke motion with an external honing tool may be manual or mechanical. The horizontal arrangement is also employed for machining large-scale workpieces, especially for internal honing of long pipework. The rotating and stroke motion may be aSSigned to the workpiece or the tool, or both. The required motion in vertical long-stroke honing machines is executed by the tool. The main spindle drive is powered from the three-phase motor by way of an infinitely variable frictional gear mechanism and V-belt. The stroke motion is carried out hydraulically via a cell pump with a variable volumetric flow. The stroke length, stroke position and stroke rate are infinitely variable. Standard peripheral speeds for a honing tool are between 15 and 40 m/min; axial speeds are between 12 and 25 m/min. The corresponding honing stone contact pressure is between 20 and 200 N/cm' for honing stones made of aluminous abrasive or silicon carbide, or between 200 and 350 N/cm' for cuboid crystalline boron nitride (CBN) and between 300 and 600 N/mm' for sintered diamond honing strips. Long-stroke honing machines are designed for
tt-ib
Long-stroke honing machines Machining
!ask
In emal machining
Position of main spindle
Horizontal
5.10 Honing Machines 5.10.1 Long-Stroke Honing Machines
Number of main spindles:
SiilQle spindle
Machining:
EJtemaI machirMng
Vertical
I
Mufti. spincle
Sifl!je spindle
MuttiJ
spindle
Figure 56 shows the classifications of long-stroke honing machines. The horizontal design is normally employed for
Figure S6. Classification of long-stroke honing machines.
5.10 Honing Machines. 5.10.2 Short-Stroke Honing Machines
machining drill hole diameters of up to 1200 mm and lengths of up to 12 000 mm (24000 mm with repositioning) . Mechanical, pneumatic, hydraulic or electrohydraulic motion may supply the infeed of the interlocking honing tool. Graduated feed initiates machining of the tightest points and areas of a bore first. Infeed time intervals are variable and may be adapted to the prevailing conditions. Compliance with specific machining diameters is ensured by using a programmable interval timer or mechanical or pneumatic measuring devices. These measuring devices have the advantage that the wear of the honing stone is taken into account. Clamping of the honing tool and workpiece is carried out in such a way that there is offset motion for minimal axis displacement. This may also be achieved by suspending the honing stone in an oscillating position. This is in a "floating" position in small or medium-sized workpieces or where the workpiece can withstand slight pressure, and the honing stone is connected rigidly to the honing spindle. Infeed attachments for the workpieces simplify loading and chaining in long-stroke honing machines in transfer lines. The external honing tool is clamped to the operating table and the workpiece is connected to the main spindle for external honing with longstroke honing machines.
5.10.2 Short-Stroke Honing Machines Short-stroke honing machines (Fig. 57) are also known as superfinishing machines, superhoning machines or orbital grinders, and are used for machining internal and external surfaces. Figure 58 shows the structure of a centreless shortstroke honing machine in continuous operation. An oscillating head 1 is used to generate the sinusoidal oscillating motion of the honing stones and is guided by a too!holder to two pillars connected with a crossbar. Height adjustment of the oscillating head is by means of a ballscrew. In continuous operation, machining is normally carried out with a number of honing stones of different grades and hardnesses positioned behind one another so that the surface quality is improved gradually. The honing stone is
Continuous operation Shott-stroke honing machines with wor1
w~ With longitudinal feed
PlullQ6-Qll process ~;
~PlullQ6-Qll process
Short-stroke surlace hotting machines
-~ Short-stroke fann honing macllines
wo~ Inlemal shott-stroke hotting madlines
Figure S7. Classification of short-stroke honing machines.
rise and fall of the honing stone. Control may be pneumatic or hydraulic. Contact pressure is individually infinitely variable, depending on the machining task. The workpieces are moved with a defined feed motion running parallel to the oscillating axis below the stone guides via transport rollers 3 moving in the same direction. Parallel continuous operation is made possible by the form and
carried and guided in stone guides 2 which are installed
adjustment capacity of the transport rollers (angle of
in the required number on the oscillating head. Each stone guide consists of a cylinder via which a piston effects the
incline 0.5 to 2°). In special cases, e.g. short-stroke honing of tapered rollers, the transport rollers are not angled back
Figure S8. Structure of centreless continuous operation short-stroke honing machine (Supfina, Remscheid): 1 oscillating head, 2 slotted link, 3 transport rollers, 4 mechanical oscillation generation.
Manufacturing Systems. 5 Metal-Cutting Machine Tools
Figure S9. Oscillating head of pneumatic drive driven via two exciter pistons (Supfina, Remscheid).
to back and continuous operation is provided via a suitable roller form. Oscillation frequencies are between 4 and 45 Hz, and stone contact pressure is approx. 25 to 100 N/cm'. The oscillating motion may be generated mechanically by translating rotary motion into linear motion via an eccentric shaft. Another principle is shown in Fig. S9. The oscillating head thereby consists of a triple-mass oscillating system with a pneumatic drive driving two selfsynchronising exciter pistons. Frequency and amplitude adjustment is carried out by way of a pressure relief valve with a vibration damper downstream. The stroke length is infmitely variable within a range of ± 15 nun. Internal short-stroke honing machines are normally equipped with clamping devices which are modified for tbe workpiece and machining task. There are also specialpurpose machines for short-stroke honing of flat and curved surfaces. There are short-stroke honing machines which may be designed for attachment to other machine tools, e.g. turning machines for precision machining in single-part manufacture and small-scale serial production. These additional attachments are equipped with one or more stone guides and clamped to the tool carrier instead of the tool. Longitudinal and plunge-cut machining is possible. The generation of the cutting motion is carried out on the basis of the principles described.
5.11 Lapping Machines S.ll.l General
Mechanical lapping is classified as shown in Fig. 60.
Type of machine
A~angement
.01main spindle
Sing e-wheel Vertical lapping macl1ine
ensures even wear of the lapping tool. The abrasive
wheels also serve as carriers of the workpieces, which are laid in the machine without any clamps. Where necessary, the workpieces are hydraulically or pneumatically loaded with additional weight to maintain the desired pressure for the lapping process. The infeed of the lapping abrasive is carried out via one or more supply reservoirs. Turbomixers and agitators provide a homogeneous mix for the lapping abrasive (normally a suspension of water and
CWt
WOI1
Chaiacteristlc machine size lapping wheel diameter
StJrlace lapping Twirl-wheel Vertical lapping machine
T~eeI
Vertical
lapping machine
.,~ Wlih parallel laces
~
laleraf, exlemal cirrumlereotiallaj>-
Iping witI1 line contact Sp/lencaI lapping machine
Vertical Horizontal
----
WOIk-. . .
~
Lawing 01 ball bearings Inlemal and
exlemal cytindrical
,awing machines
Vertical
S.11.2 Single-Wheel Lapping Macltines See Fig. 61 for the basic structure. A baseframe accommodates the lapping table, consisting of an underbody fitted with a lapping wheel or lapping segments, depending on the size of the machine. Abrasive wheels supported by lateral guide arms run along the lapping table. These sharpen the lapping wheel continuously during the main operating process as a result of the friction conditions created by the turning motion (friction coupling). This
MacI1ining task
Horizontal
Diameter and lengIh 01 inlemal (J( exlemal surlace Circumferential exlerna 10( lapping Nlin
w~
~
ircumferenlial exlerna cylindricaJ 'awing with surlace contact Oscillating lapping machlne
Vertical
Horizontal
W~ece
Travel paths
~
Internal and exlernal surtaces, as desi'ed
Figure 60. Classification of lapping machines according to DIN 8589 Tl5.
5.11 Lapping Machines. 5.11.3 Twin-Wheel Lapping Machines
9
/0
13
5
Figure 62. Structure of twin-wheel lapping machine (Peter Wolters Maschinenfabrik GmbH & Co., Rendsburg): 1 motor for top
Figure 61. Single-wheel lapping machine (Waldrich Coburg, Coburg).
lapping powder, i.e. aluminous abrasive, silicon carbide or boron carbide). Control devices and jets in the lapping abrasive circuit provide a metering system and may be adapted to suit the individual machining task. Single-wheel lapping machines are designed with lapping wheel diameters of between 350 and 5000 mm. S.l1.~
Twin-Wheel Lapping Machines
These are used for surface lapping, lapping with parallel faces and external cylindrical lapping (Fig. 62). Lapping wheel diameter is between 250 and 1000 mm. lip to four separate coupled or decoupled speed drives are available for all types of motion. Speeds may be stepped or infi-
lapping wheel drive, 2 gear mechanism, 3 toothed belt drive, 4 top Japping wheel, 5 motor for bottom lapping wheel drive, 6 V-belt drive, 7 worm gearing, 8 bottom lapping wheel, 9 motor for workpiece drive, 10 worm drive, 11 toothed belt drive, 12 internal gear ring, 13 motor for lowering external gear ring, 14 external gear ring lowering device.
nitely variable. The base contains the drives for the bottom lapping wheel 8, for the internal gear ring 12 and for lowering the external gear ring 14. The bottom lapping wheel is arranged on a thrust bearing which is prestressed via grooved ball bearings and cup springs and is driven by a three-phase motor 5 with a V-belt 6 and worm gearing 7. The top lapping wheel is arranged to oscillate, and the top main spindles are arranged on a pivoting arm. Workpiece drive is for positively driven operation of workpieces, resulting in even wear of lapping wheels and a high degree of uniformity of workpieces 9, 10, II. The
i¥Jf0~ I I a
b
c:
. I ~
I
Figure 63. Arrangement of modules in special-purpose machines. a Single-path machine, horizontal, with permanently fixed table or with horizontal linear feed motion. b Single-path machine, vertical, with permanently fixed table or with horizontal linear feed motion. c Two-path machine, hOrizontal, with permanently fixed table or with horizontal linear feed motion. d Two-path machine, horizontal and vertical, with permanently fixed table. e Three-path machine, horizontal, with permanently fixed table, f Three-path machine, horizontal and vertical, with permanently fIxed table. g Three-path machine, horizontal and vertical, with vertical rotary feed motion. h Five-path machine, horizontal and vertical, with horizontal rotary feed motion with five work stations and one loading/unloading station.
Manufacturing Systems. 5 Metal-Cutting Machine Tools
I--_ _ _ _ _ __
_ _ __ __ _ Awroximalel' 23250 _ _ _ __ _ _ __ _ _ _ __ __ _---;
II
B B :;;:
18 ~
fJ
.5
e
§:
D
og
8FR,-----------,-1
Station Figure 64. Fourteen-station transfer line for machining articulated shafts (HOller Hille GmbH, Werkzeugmaschinen, Ludwigsburg).
workpiece carriers arranged between the lapping wheels are driven by the internal gear ring and roll around the external gear ring, which is normally stationary. The workpieces are thereby moved onto cycloidal paths between the lapping wheels. The drive shaft for the internal gear ring is guided by two angled ball bearings arranged around the Z-axis in conjunction with a needle-type bearing. The external gear ring is lowered to change the workpiece carrier. The lapping pressure is built up hydraulically or pneumatically. The pressure level may be varied via a PLC system. Specific workpiece dimensions and tolerances are normally maintained using a programmable interval timer system. The lapping time required is set on the basis of his-
toric values. An additional further option is an indirect measuring control system. Here the distance between the two lapping wheels is measured and the machine is switched off having reached a set level. This type of control is often employed in association with greater admissible workpiece tolerances, because the lapping abrasive rum thickness and the lapping wheel wear are included as the measurement.
S.11.4 Triple·Wheel Lapping Machines Triple-wheel lapping machines have two bottom lapping wheels arranged adjacent to one another, with a diameter of between 380 and 810 mm. The top lapping wheel may be accessed and rotated in tum over both bottom wheels.
Figure 6S. Flexible manufacturing system with two machining stations for prismatic workpieces (Fritz Werner, Berlin).
6.1 Arc Welding Machines
When a pair of wheels are in operation, the disengaged lapping wheel may be loaded and unloaded.
5.11.5 Spherical Lapping Machines Spherical lapping machines are constructed in much the same way as twin wheel lapping machines. Only one lapping wheel is driven in this design, with a diameter of between 100 and 1200 mm; its arrangement may be vertical or horizontal. One of the two lapping wheels normally has concentric V-shaped grooves, whose dimensions match the dimensions of the ball bearings to be lapped and semi-circular wear is generated during lapping. As the ball bearings have to pass between the lapping wheels several times in a different groove each time, infeed attachments are required which also perform a mixing function at the same time. Preliminary machining operations employ bonded abrasive wheels, while precision machining uses cast lapping wheels and freely flowing lapping abrasive suspension.
5.12 Multi·machine Systems Specialpurpose machines and transfer lines are employed to obtain high productivity in mass production. The structure takes the form of modular components with standardised main and ftxing dimensions, which may be combined to produce special-purpose machines of differ-
ent designs in horizontal, vertical or any angular arrangement. Figure 6~ shows a number of forms of single and mUlti-path machines, machines with stationary tables and with linear or rotary workpiece infeed motion. A transfer line is shown in Fig. 64 for machining articulated shaft parts. The tools pass through the machining, washing and dust extractor stations in continuous operation and trans·
port from station to station is simultaneous and fully automatic. The cycle time is determined by the station with the longest machining time. With the growing variety of parts and falling unit volumes in serial production, requirements for flexibility in manufacturing systems are increaSing. Manufacturing machines with tool magazines and tool change systems are employed in stations in the flexible transfer line, as a result of which production of a wide range of parts may be carried out, instead of single-purpose machines with a ftxed arrangement of tools. Flexible cell-type manufacturing systems are employed in serial production of average lot sizes which contain workpiece and tool storage systems as well as handling, measuring and monitoring systems in addition to the numerically controlled machining centre for prismatic or rotary-based parts. Flexible manufacturing systems are created by networking these types of cell-type manufacturing system, where work stations substitute or supplement one another and provide non-cycled manufacture independently of lot sizes. Figure 65 shows a flexible manufacturing system with two work stations for prismatic workpieces.
~Welding and Soldering (Brazing) M a c h i n e s _ L. Dora, Berlin
(For welding and soldering see Fl.l and F1.2.)
6.1 Arc Welding Machines Requirements. A number of specific electrical requirements have to be satisfted relating to the welding power source for arc ignition and maintenance of the welding arc in welding operations: High no-load voltage compared with arc burning voltage (safe ignition). Rapid restoration of supply following drop short circuits (rapid reignition). Short-circuiting current slightly higher than welding current (low spatter welding).
Static Characteristic. This describes the change in the source voltage U in relation to the level of the welding current I (Fig. 1). The volt -ampere settings (operating point A) generated during welding correspond to the point of intersection between the static characteristic setting (1, 4) and the arc characteristic (2, 3) rising as the arc length increases. Where a steeply falling curve is produced (4), changes in arc length (voltage) produce small changes in current. This assists in providing an even heat supply in manual arc welding. In internal submerged arc welding using thicker wire, the voltage change is exploited to maintain a constant arc length with an adjustable feed motor (known as "external contro!").
With a shallow volt -ampere output curve 1, small changes in the arc length (voltage) cause large changes in current. Where consumable electrodes with a higher electrode current density are employed, the melt rate changes according to the heat output of the arc, and consequently so does the arc length with a constant-speed wire feed. This is used in gas-shielded metal-arc welding
and submerged arc welding with thinner wires to maintain a constant arc length with variations in the torch-to-
work distasnce (known as internal control).
Dynamic Characteristic. This describes the power source characteristic with different short-term load characteristic changes, such as those generated on ignition and in drop short circuits (Fig. 2). In power sources with excessive initial current surge I k ", the electrode tends to adhere to the workpiece on
ignition, and with drop short circuits large amounts of
Voltage U Uo . - - - _ Uo
Welding current Figure 1. Change in operating point by increasing arc length with steep and flat static power source characteristic: I flat characteristic, 2 long are, 3 short arc, 4 steep characteristic.
Manufacturing Systems. 6 Welding and Soldering (Brazing) Machines
"11111111111111111 No-load voltage
1" tlllllllllllilld, """" ,. Drop short circuit
Arc buming phase
Ign"ion short circuit
the welding duty cycle in an operation differs from this, the maximum current may be obtained as follows: I, = I, ~EDl/ED2'
i
II,~ =~~IkSI ~lUIlllljllllllllllllllllllllll 1k
""lJJ.W.Lll.lJ..l
Figure 2. Time-based progression of current and voltage with short-term changes in loading as a result of contact ignition and drop short circuits.
Admissible No-Load Voltage. No-load voltage is limited on the grounds of safety, namely to 113 V peak value in d.c. sources and 113 V peak value in welding transformers, and 80 V actual value. No-load voltages of no more than 68 V peak value and 48 V actual value are admissible for welding in boilers and restricted spaces using welding transformers. In fully mechanised welding plants, 141 V peak value and 100 V a.c. actual value are permitted, by comparison. Design The characteristic attributes and examples of welding current source applications are compared in Table 1.
spatter are generated; also, where the initial current surge is too low, the heat generation is not adequate to ensure reliable ignition. The voltage on restoration following dipping should return to the full no-load voltage Uo as soon as possible in order for the power source to provide reliable reignition. In power supplies with high inductance levels in the welding current circuit, e.g. welding rectifiers with constant current booster, there is a rapid rise in voltage to above Uo, whereas in rotating welding generators the no-load voltage is only reached after a delay. However, it is also possible to generate a short-term voltage surge Uz directly on restoration following dipping with structural modifications, thereby guaranteeing rapid reignition. In a.c. welding the arc is interrupted in rapid succession even without drop short circuits, and has to be reignited. The rapid voltage rise required for this may be achieved using welding transformers with open-joint restrictors or special magnet core combinations which produce a rectangular-shaped voltage curve in contrast to any of the usual sinusoidal forms. An extremely rapid volt-
age rise is achieved by superimposing high-frequency, high-voltage pulsing (3000 V) using arc ignition and stabilisation equipment.
Setting Range of Welding Current. This is obtained from the point of intersection between the arc characteristic and the static characteristics at the highest and lowest setting level. The following approximations apply for the arc characteristic in the form of equations with numerical values:
= 20 + 0.041.
Coated electrodes
U
TIG welding
U = 10 + 0.041.
MIG welding
= 14 + 0.051. = 20 + 0.041 or U = 14 + 0.051. U
Submerged arc welding
U
Welding Current and Welding Duty Cycle [EDl. The power supply is not norntally permanently under load during welding, instead the loading is discontinuous, corresponding to a specific relative setting time: welding duty cycle time under load
= time under load + break time x
100%
This is why the admissible welding current for the machine heating system is not specified just for a welding duty cycle of 100% (continuous operation), but also for a welding duty cycle of 35 or 60%, in accordance with standard practice in manual welding operations. Where
Welding Transformers. These are normally connected to two external conductors in three-phase current supply; for smaller outputs they may also be connected to one external conductor and the mid-point conductor. The falling characteristic is produced by changing the coupling between the coils (stray field transformer), generating a secondary magnetic connection by introducing a plunger core (stray core transformer), or by changing the inductance using a control throttle downstream. A stepless, remote current control is provided by way of adjustable throttles (transducers) using a pre-magnetised iron core.
Welding Rectifiers. TIle mains voltage is transformed via a three-phase transformer and set to the desired characteristic via a control throttle or transducer. A semiconductor rectifier set rectifies the welding current. Additional current damping throttles are frequently installed to improve the dynamic characteristics.
Motor-Generator Welding Set. TIle motor-generator welding set is designed for a rotating d.c. generator driven by either an electric motor or a combustion engine. Motors and generdtors (including fans) are arranged around a common shaft in so-called single-shaft converter sets. Generators are used primarily on construction sites due to their mains-free operation. Differentially wound generators, quadrature-axis generators and stray field generators all generate a sharply falling characteristic in different ways, where the main field (and thereby the induced voltage) is weakened by an opposing field increasing in line with the welding current. More advdnced brusWess generators require very low maintenance because slidingaction contacts and commutator segments are no longer used.
Multiple Spot Welding Units. Where a large number of welding units are each operated with a very low duty cycle, supply via a multiple spot welding unit may be economically advantageous. In contrast to the setting equipment, this employs generators or rectifiers with constant voltage characteristics and high electric output ratings. The individual setting of the desired falloff characteristic at the welding spot takes place via leakage or multistep inductors for a.c. current, and for d.c. current via multi-step, high-wattage resistors. Electronic Welding Current Sources. Thyristor or transistor current sources are characterised by their accur-
ate and rapid controllability and wide range of potential applications, e.g. pulse-welding.
6.2 Resistance Welding Machines
Table 1. Mains supply ratings, operating characteristics, costs and areas of application of welding current sources Welding transformer
Welding rectifier
D.C. converter
Mains supply
Single-phase
Three-phase
Three·phase
Mains loading
Asymmetrical
Symmetrical
Symmetrical
Feedback to mains
Undamped
Undamped
Damped
Effect of mains fluctuations
Proportionate
Proportionate
Insignificant for partial loads
Degree of effectiveness
80 to 90%
65 to 85%
45 to 60%
No-load pickup
0.06 to 0.15 kW
0.1 to 0.25 kW
1 to 3 kW
Output rating, without compensation
cos cp 0.4 to 0.6
cos cp 0.5 to 0.75
cos cp 0.85 to 0.9
Ignition characteristics
Satisfactory
Good
Good
Weld characteristics
Satisfactory to good
Good to very good
Good to very good
Application potential
limited
Universal
Universal
Arc blow effect
Insignificant
Extreme
Extreme
Power loss in welding line
High
Low
Low
Noise emission
Insignificant
Low
High
Comparison of acquisition costs
50%
80%
100%
Operating costs
Minimal
Low
High High
Maintenance and repair costs
Minimal
Low
Electric welding
Occasionally for B and C types
All types of electrodes
All types of electrodes
Submerged arc welding
No high base flux
All flux types
All flux types
MIG/MAG welding
Not conunon
Reverse polarity welding
Reverse polarity welding
TIG (tungsten inert gas) welding
light metals
Steel, non-ferrous heavy metals
Steel, non-ferrous heavy metals
6.2 Resistance Welding Machines Resistance welding machines may be stationary machines (Fig. ;') and mUlti-point equipment including portable weld guns and gun welding heads. These are classified as spot, projection, seam-welding and butt welding machines depending on the process. The task of supplying electrode forces and welding current requires a combi-
nation of mechanical and electrical functions.
MechllAical Functions. Electrode forces are variable over a broad range, thereby providing the best solution
6
5
Fipre 3. Schematic layout of spot welding machine: 1 transformer, 2 busbars, 3 live spring, 4 lower ann holder and lower arm, 5 upper arm, 6 compressed air cylinder and ram guide, 7 electrode holder with electrodes.
for the relevant welding task. The machine frame and electrode anns are to be designed with high rigidity. Bending under the electrode forces would cause a reciprocal shift in the electrodes during the expansion phase of the material. The rigidity is particularly important in projection welding machines in order to provide an even current distribution simultaneously over the projections to be welded. Rapid closing motion of the electrodes is required to optimise production rates. The electrode should make full contact to minimise operating noise and electrode wear. The adjustable electrode system is to be designed to be as lightweight as possible, in order that the low-inertia electrode can follow the elastic material in the expansion phase. Short-tenn release of interlocking between electrode and workpiece may have undesirable effects due to overheating.
Electrical Functions. The welding machine is to discharge the highest possible output current within a short period. The maximum short-circuit current measured where the electrodes collide is an extremely important characteristic. At the same time, low values are desirable for the output current to maintain the lowest possible installed load. This means that energy losses from the transfonner, output circuit and welding current control should be minimised. A high continuous load capacity of the machine, specifically of the welding current transfonner, should be guaranteed in serial production. Continuous output, i.e. average output which can be taken up in the long tenn without excessive temperatures being recorded, should be taken into consideration when selecting seam and butt welding machines.
'WI.i.'
Manufacturing Systems. 7 Industrial Robots
Welding current control is designed to generate brief welding periods of a few hundredths of a second and maintain these accurately. In addition, the full electrode force must be applied at the onset of the flow of current.
Three-Phase Welding Machines. Three-phase machines based on the principle of output side rectification or input side frequency transformation are becoming more popular than traditional single-phase types (Fig. ~), owing to the more reliable mains supplies and improved welding characteristics in some cases.
of > 100kHz are more suitable for thin parts. Coil and panel inductors may be employed as inductors, depending on the shape of the workpiece. The structure of resistor soldering and brazing machines corresponds to that of welding machines to a large extent. The soldering heat is generated in the workpiece itself with internal resistor heating. Copper electrodes are employed with shon heating times. With carbon resistor heating, the heat is retained for longer and is preferably generated in graphite electrodes.
Welding Guns. These are a form of mobile welding
Oven Furnace Soldering
equipment which is guided either manually (normally with weight redistribution) or by industrial robot.
The soldering furnaces are either gas or oil-fired, or electrically-heated. The latter has the advantage of maintaining even temperatures accurately with a defined inen gas atmosphere in the furnace area. Funbermore, a distinction is to be made between discontinuous operation furnaces, such as reton-type furnaces, pit-type furnaces and bell furnaces, and continuous operation furnaces. Flux may be omitted in cenain circumstances where inen gas reduction is carried out, e.g. H2 -CO/C02 mixtures. Either hot wall furnaces or cold wall furnaces are used with heated resistors or induction heating for flux-free vacuum brazing with pressures of berween 10- 1 and 10-6 mbar.
6.3 Soldering and Brazing Equipment MecJuudsed Soldering and Brazing Plants The soldering and brazing process may be mechanised Simply with suitable solder feed mechanisms, e.g. as brazing aUoy preform, solder powder, soldering paste or as braze-bonded cladding. Normally, rotating tables or conveyor slides are employed for guiding the workpieces through the heating cycle. The heating is carried out by gas torches, induction coils, infrared radiation or by way of electrical resistors. Rotating current transformers with mean frequencies of berween 1 and 10kHz are preferred for induction brazing of thicker parts. Core generators with frequency ranges
Dip Soldering and Wave Soldering These are processes for the soft soldering of electrical component terminals onto the circuit-board conductors of printed circuit boards via immersion in a solder bath, or a solder shaft which simultaneously carries out the tasks of feeding and heating the solder.
Industrial Robots • • • G. Spur, Berlin
7.1 Systematics of Handling Systems Handling systems consist of production machines which are designed to handle objects with suitably dedicated
devices, such as grippers, tools, or baffle plates (see K6.2.2). Computer-controlled universal handling systems with programmable sequences of operations are more popular in extended operations than other constructions. This group may be classified in terms of systems with continuous patb control, such as cams, limit switches or fixed stops, and systems with point-to-point control via controlled setpoint input. Continuous-path control devices only allow for actuation in rwo different positions in each axis of motion. On the other hand, with point-to-point control systems, as many positions as are required may be actuated for each axis of motion. The number of positions is limited only by the capacity of the setpoint memory in the controller. In computer-controlled handling systems flexibility is tending to increase with the improved programming facilities. Flexible handling systems (industrial robots; Fig. 1) are production machines designed for automatic object handling with suitably designed tools and may be programmed in a number of axes of motion in terms of orientation, position and sequence of operations. As the mechanical structure of handling systems may
be represented in the form of kinematic chains, the number of degrees of freedom within a handling system is equal to the number of links in the kinematic chain which may be moved independently, assuming that each anicuIated joint may only have one single degree of freedom. links, levers and drives are employed as elements in the creation of kinematic systems. Each individual combination of link, lever or drive is described as an axis of motion. Each axis of motion corresponds to a degree of freedom in the kinematic chain. Rotary or linear axes are generated as a result of the use of revolute or prismatic links (see F9.1). Handling systems are used to change the orientation and position of three-dimensional objects. Six independent movements (degrees of freedom) are required to adjust the position of a three-dimensional object. If this is based on a closed Cartesian system of coordinates, three linear degrees of freedom are required to adjust the position of an object and three rotary degrees are required for the orientation of the object. Three-dimensional motion as is required in tool handling systems may be generated via linear motion and rotary motion. Three degrees of freedom are required to position an object within a specific three-dimensional Cartesian system of coordinates. The independent motion required for this may be achieved by way of a suitable arrangement of a minimum of three controllable axes. The combination of linear and rotary axes and their arrangement determines the working space in the tool handling system and may be defined by way of the system
7.2 Components of Robots
Coomnate designation
Axis combina~on
'WI,II
Wor1
Axis designation 3 linear axes
Xl
2 linear, 1 rotating axis
I
•,
i
iG
~
Xl
~
llilear, 2 rotating axi
1 B
I
I
3 rotating axes
1 ~
."
~
0::
~~
mlinear, a'E n rotating axes ·il §
~ J!! '
~1
,
,,
/
i
,
'LL--
~
/
"' 8
I
~
A ,
,
II d;.
,
,
/
I ~8____
~
~
>-
!Il
-~
t~ A~ , "
~J
l;
Figure 1. Working spaces and coordinates of motion in handling systems.
of coordinates of motion or the geometry of its areas of contact. Figure 1 shows these relations in these basic three-axis systems. More advanced configurations with more than six controllable axes have been introduced very recently to add to these four basic types [I J. Examples of these are the vertical column or trackmounted robots and robots on rotary bases. These additional axes of motion extend the working space in the handling system and interference prevention and optimised motion sequences have also been developed, now that degrees of frequency have been largely superseded. Robots may also be set up as systems with distributed axis control, with a combination of six-axis robots and rotating and tilting benches for seam welding applications, for example (Fig. 2) , or multiple articulated joint handling with cooperating robots.
7.2 Components of Robots The drive units of the active articulated joints must satisfy the high requirements which it is somewhat difficult to combine in view of the dynamic motion characteristics of the robot. These requirements include low inertia, low
Figure 2. Robot systems with distributed axis control (KUKA, Augsburg).
output weight, high pulse rate, high short-tertn overload capacity, and high resolution both over the path setting range and speed range. Electrical, hydraulic and pneumatic drives are commonly in use (see Ll.2.1) . D.C.-powered machines are mainly employed in disctype rotors and squirrel-cage induction rotor constructions. More recently, there has been a trend in favour of the use of brushless, wound d.c.-powered motors with rare-earth element solenoids, and asynchronous motors. These machines require very little maintenance and have a better power-to-weight ratio. The input required in tertns of power electronics is nortnally compensated for by the advantages specified. Electric motors are used in combination with high reduction gearing, such as hartnonic drives, wortn-type gearing, or planetary gearing (see Ll.2.2). Non-reduction direct drives are in limited use in a few specific applications. Hydraulic cylinder drives cater for extremely high driv-
ing forces with relatively low dead weight. They are recommended for driving large-scale kinematic systems for this reason. Pneumatic cylinders are also designed for extremely high speeds, but the air compressibility factor prevents them from following accurate paths. These drives are employed mainly in supply and loading systems with adjustable stops. The internal robot sensors provide accurate and rapid data logging of the actual position of the articulated joint. The robot controller thereby obtains the feedback difference between the actual and setpoint setting, as a result of which this is translated by the position feedback control system into the appropriate manipulated variables. Internal sensors may be classified in tertns of test results as linear or rotary values, or in tertns of data logging as digital and analog, and in terms of the measurement process as incremental and absolute values. Synchro-resolvers and digital recorders are employed in practice in industrial robot applications. Synchro-resolvers are measuring systems based on the induction principle with direct angle data logging. These may also be used as indirect path measuring sensors. Specific advantages of the resolver include their compact design, high resolution capacity and resistance to wear.
'WIlli
Manufacturing Systems. 7 Industrial Robots
Digital locators record the measured value as a whole number multiple of the angle or path increment. A distinction is made between incremental (relative) and encoded (absolute) measuring systems, depending on whether the value is obtained as an increment or from a linear encoder readout. Absolute digital signals are .relatively costly to construct if they are to provide long traverse paths with high resolution. Incremental digital signals are cheaper than absolute ones and have an unlimited measuring range in principle. The cumulative results of measuring errors and loss of the reference point during power loss are two of the main disadvantages of incremental measuring systems.
7.3.2 Dynamic Model The system of differential equations of motion for the robot may be produced in the form of the Newton-Euler equations for bolonomous systems, H(q)q + h(q)q
+ G(q)
= P.
Here H = (H'j) represents the n-dimensional inertia matrix, h = (hi . .. h n ) represents the vector of generalised constraining forces (centrifugal and compound centrifugal forces), G = (G I . . . G n ) represents the vector of generalised forces due to weight and P represents the vector of driving forces [2]. There are consequently two basic dynamic problems in connection with this model:
Direct Problem. The robot motion q(t) has to be obtained for a given driving force P(t).
7.3 Kinematic and Dynamic Models 7.3.1 Kinematic Model Six coordinates which are independent of one another as a function of time are required to describe the general three-dimensional motion of a rigid end effector. Three of these coordinates determine the original position of the effector coordinates. The remaining three determine the orientation of the effector coordinate system in relation to the fIXed frame of reference. These coordinates are designated as external X-coordinates. The joint coordinates (inner coordinates) are selected as generalised coordinates in structures without branches or links in the kinematic chain of the robot members. The relationship between external and internal coordinates is defmed by a non-linear vector-type pattern X = f(q). The concrete form of the vector function f is dependent on the selection of the internal and external system of coordinates and the structure under observation. In practice, both coordinate systems are in common use. On the other hand, robot motion is normally planned in tenus of external
coordi~
nates, control and monitoring being carried out in terms of joint coordinates.
Basic Kinematic Problems. There are basically two problems associated with the kinematic analysis of the robots, as follows. Direct Problem. The corresponding position and orientation of the end effector has to be obtained in terms of external coordinates for given joint positions. Inverse Problem. In order to generate the desired path in terms of external coordinates, the required internal coordinates have to be defined. The Denavit-Hartenberg convention for describing and modelling the kinematic robot structure is employed to solve both of these problems. Each link in the kinematic chain is then assigned to an object-related system of coordinates, as follows. The transformation of coordinates between two adjacent joints depends purely on the coordinate of the connecting member, the transformation f(q) between the base and the effector joint of the unbranched kinematic chain of all member coordinates. The solution to the inverse problem, q = .r1(X), is normally ambiguous and may only be obtained in a closed form for specific robot structures. Explicit solutions in a closed form are possible, for example, for structures with "spherical manipulation·', where the axes for articulated hand operation intersect at a point [1]. Numerical solutions should normally be applied.
Inverse Problem. The required driving forcesP(t) have to be determined before the desired robot motion q(t) can be generated. Manual calculation of the dynamiC equations for industrial robots is extremely time-consuming and susceptible to errors. For this reason, a number of algorithms and correspondingly efficient numerical or mathematical program packages have been developed to create mathematical models for robots. Mathematical programs provide the equations of motion for the relevant task in hand in a form which minimises the input required for their arithmetic evaluation.
7.4 Characteristics, Accuracy The specific attributes of the robot machine tool in terms of its design and sequence of motions require that specific characteristics and the process associated with their designation are defined in order to make robots comparable and to enable modifications to be estimated, e.g. as a result of wear. The following characteristics and the
respective designation processes specify characteristics for industrial robots in the VDE Guideline 2861, as follows: Working space, useful load, speed, acceleration, travel time, repeatability, reproducibility, contouring accuracy, programming accuracy. ISO Standard DP 9283 represents a further development. An increase in the degree of accuracy of robots which is currently around a factor of 10 less accurate than machine tools will open up opportunities for numerous new applications in sectors with considerable potential for future expansion such as assembly work, for example. As any increase in the degree of accuracy by way of modifications on the part of the robot mechanics would lead to a considerable increase in costs, efforts have been made to increase the degree of accuracy over the past few years by identifying systematic errors and taking these into account in the mathematical and physical models on which the control is based or in compensation processes. This enables minor discrepancies in the geometric/ kinematic data, elasticity values, or transmission-influencing factors to be accounted for numerically and be taken into consideration in the transformation calculations
between external (effector position and orientation) and internal (link position) coordinates [1 5].
7.5 Industrial Robot Control Systems The task of an industrial robot control system consists of controlling one or more handling systems in accordance
7. S Industrial Robot Control Systems
with the handling and processing task required for the technical process. Motion sequences and actions are specified as part of a user program which is executed by the controller. It obtains process data via sensors and is therefore in a position to adapt the predefmed processes, motion and actions to the changing or environmental conditions or those previously unknown to a certain extent. In addition, an industrial robot control system has to satisfy specific requirements as regards operating modes, user operation and programming, as well as monitoring and safety functions [I, S]. Industrial robot control systems are based on microcomputers to a large extent nowadays, sometimes using multiprocessor technology (see L2). Interfaces to manufacturing communications systems, e.g. MAP (Manufacturing Automation Protocol). are available for connection to primary control and programming systems. More and more links are being established using serial bus systems, e.g. field bus and bit bus, to peripheral processes in the same way, e.g. in welding control and conveyor systems, also providing a means for external sensor analysis.
Software Components of a Robot Control System (Fig. ~ ). Data transfer is carried out via the communications module to other control systems (industrial robot controllers, cell-based computers, primary computers). Specifically, loading of user programs to the robot controller and the transfer of status data and reports to other control systems (DNC operation) are carried out in this way. A common communications standard was laid down in ISO 9506, known as the manufacturing message specifIcation (MMS) for the various classes of equipment [3,4].
Sequence Control. The user program of an industrial robot contains motion commands. effector commands, sensor commands, program sequence control commands, arithmetic commands and technological commands. The execution of the user program is organised in sequence control which is normally identical to the so-called interpreter. This means a program which reads the instructions of the user program or a relevant code, e.g.
(Primary controller)
(User)
Figure 30. Software components of industrial robot controller.
'W"@'
IRDATA [6], or decodes the instructions generated by a compiler and accesses and coordinates the respective execution routines or segments. The task of motion control is to generate the relevant reference variables for the servomechanism to drive the handling systems involved in the motion, i.e. robots, rotary table. kinematic elements and other auxiliary axes, using the given motion sequence provided by tht' user program and user data. Point-ta-point control (PTP) enables a sequence of discrete three-dimensional points to be logged and actuated. The path of motion of the endeffector between these three-dimensional points is not explicitly specified. This enables extremely efficient timecontrolled motion characteristics to be achieved. This method is used for tasks where the precise path progression is not important, e.g. for handling and spot welding tasks. Continuous-path control (CP = continuous path) offers the option of traversing mathematically defined paths of motion within the task range. The continuous-path control computer (interpolator) obtains a number of intermediate values on the three-dimensional curve programmed in terms of the points corresponding to a given path function (straight-line, circle or higher polynomial) and speed function and reports them to the servomechanism at a given rate. Continuous-path control is used, for example. in track-mounted robot welding and deburring applications. The servomechanism has the task of traversing the robot axes in accordance with the current position setpoints. Axis-related intermediate points are recomputed within a narrower time grid within the axis controller (precision interpolation). The values of the axis angles or paths are translated into motor currents, voltages or increments and output to the actuators. Actuation of the axis positions is monitored and computed using the actual position feedback reported by the path and angle monitoring systems. Sensor data processing involves the reception of signals and data, e.g. coordinate values of objects, from sensors inside the robot (path and angle monitOring systems, force/torque sensors) and outside it (proximity sensors, recognition system). These data are required and processed by different levels of the robot control systems, i.e.
sequence controL motion control and axis control. The shortest response times to external conditions may be achieved where the relevant sensor data are reported at axis control level (e.g. force/torque sensor analysis, collision monitoring). Action control executes the action commands of the user program which normally relate to objects to be handled or the control of peripherals. Depending on the action conunand, it executes a combined logic operation consisting of internal and external control process signals (i.e. motion status, limit switch, light barriers, responses from other controllers) and produces the drive signals for the binary actuators (such as the contactors, single drives and values) or generates commands for peripheral manu(such as start, stop, and facturing devices synchronisation) . The operation control segment supports functions such as selection of the operating mode, input of operating parameters, program start/stop, loading/storing programs. A robot controller can be operated in two basic operating modes which are normally divided up into further submodes. The user may access all operating elements in the controller which are required to operate the robot and to create or modify handling programs in "set-up" mode. Onl), a few simple operating functions may
IWI,EI
Manufacturing Systems. 7 Industrial Robots
be executed in "automatic" mode, e.g. program selection, start, stop, and resume. Information on the current handling program and operating instructions and error messages are displayed via the current handling program. Programming components are used for user program generation, maintenance and management. The function modules required for program generation, such as the editor, debugger and compiler may form part of the control system or may be exported to another computer, e.g. a Pc. An interactive programming component enables programs or selected positions for the robot or the effector in the teach-in process to be entered or programs of motion to be tested.
7.6 Programming 7.6.1 Programming Procedures Programming procedures constitute the planning procedures for generating user programs. According to VDI Guideline 2863 (IRDATA), a user program constitutes a sequence of commands aimed at executing a given manufacturing task [7]. Programming procedures enable user programs to be developed and provide the relevant programming aids for this purpose (see L2). Programming procedures may be divided up into direct programming (on-line systems), indirect programming (off-line systems) and hybrid programming [8].
Direct Programming. This is characterised by the fact that user program generation is carried out using the robot system. The consequence of this is that the manufacturing system is not available for production during the programming or testing period, resulting in high idle times during installation. Integration of operational, computerbased data systems is only possible on a limited scale. The quality of the user programs is dependent on the experience of the programmer to a large extent. There are subSidiary categories within these procedures, known as play-
back, teach-in and sensor-based processes: In playback programming, programming of an operational sequence is carried out by manually guiding the robot along the desired three-dimensional curve. The actual positional values (axis points) within a defined time or path scale are thereby carried over into the user program. A special lightweight programming arm may be employed to program motion, which is extremely similar to human working methods in dynamic terms. One typical application is programming paint spraying robots [9]. In teach-in programming the motion data are generated by tracking the desired three-dimensional coordinates with the aid of a hand-held programmer or operating field, and these points are subsequently programmed by activating a function key. Other motion commands may also be input via the keyboard, such as speed and acceleration data or type of control (point-to-point or continuouspath control). Sensor-based programming, which is now becoming popular, may be categorised in terms of automatic sensor control and manual sensor control [10]. The former is based on rough motion data (such as start and target point) and the workpiece is scanned automatically by the robot using sensors. In the second system the robot is guided along the desired three-dimensional curve by the operator using a sensor or light pen. In contrast to playback programming, where the robot plays a passive role, in this case sensor signals are fed direct to the control circuits in the robot control systems. These initiate the
active sequence of the operating tasks. In sensor or manually guided programming, the path travelled is automatically stored in memory. This occurs by programming path interpolation points in accordance with given criteria, such as the desired degree of accuracy [11].
Indirect Programming. This is characterised by the fact tbat the user program generation is carried out separately from the robot system on independent computer control systems. It requires a computerised model of the robot system and the system environment. Programming and testing of the user program is reassigned to operational preliminary planning and therefore becomes a part of the production planning process. Integration of operational data systems and intelligent, computer-based aids are provided as support for the programmer. A distinction is made between textual and CAD-based processes in indirect programming employed in off-line
programming procedures. Textual programming procedures request geometric data input via a keyboard, as is standard practice with computer and NC programming languages. A further development has taken place with textual input and graphics support. More advanced programming procedures provide direct CAD support for geometric definition and motion. CAD-based programming procedures support geometric models of the components involved in the manufacturing process. Geometric model generation is carried out via CAD systems. Functions are provided on the graphics screen for defining positions to be covered and travel paths. Integrated simulated modules allow for visual display of the motion executed by the robot. CAD-based programming procedures are therefore characterised by their graphics representation. A further distinction is made between motion-based (explicit) and task-based (implicit) programming procedures [12]. In explicit programming procedures, all the actions of the robot, especially the motion, including execution parameters (e.g. speed, acceleration), are specified by the programmer. Thus all travel paths and positions to be covered require designation and should take into account freedom from collisions. In impliCit programming procedures, programming is not carried out by deSignating the travel path, but by describing the handling task. The path data are derived from the programming procedure automatically using a
Table 1. Attributes of direct and indirect programming procedures Direct programming
Indirect programming
- Real-time robot system and system environment required
- Computer model of robot system and system environment required
- Production system not available during programming
- Programming during
- User program testing on real system
- Simulated program testing
- Umited integration with operational data systems
- Full integration with operational data systems
- Quality of user programs dependent on experience of programmer
- Programmer support via intelligent, computer-based help utilities
operational planning as part of production planning process
7.6 Programming. 7.6.2 Off-line Programming Procedures
'WI••
model of the robot cell (model environment). Programming procedures based on such systems aft' currently undergoing research.
cedures according to modem standards. Robot programming languages have been classified based on an analysis of the language and system stmctures [131.
Hybrid Programming. This represents a combination of direct and indirect programming procedures (Table 1). The progrdm sequence is subsequently determined by the indirect programming procedure. The motion control section of the program may be defmed using the teach-in or playback system or by sensor-guiding.
Local Control Off·line Programming Procedures
7.6.2 Off·line Programming Procedures In 1986 the US National Bureau of Standards published a study on the existing off-line rohot programming procedures all over the world. Approximately 9S different robot programming languages were named, not all of which may be employed in oft'line programming pro-
Table 2.. Requirements of robots systems in forming and shaplllg Requirements
Program generation with systems of this type is carried out in the same programming language as is employed in the robot controller. Textual program generation is simplitied with the use of user-friendly editing programs (e.g. menu-driven) and improved communications facilities. Such off-line programming procedures enable a complete program to be developed which describes the sequence of the handling or machining task, communications with process peripherals and their synchronisation with the motion of the industrial rohot. The definition of geometric data from the program of motion, i.e. the position and orientation of the end-effector system, is carried
procc~~es
Areas or application for integrated rohot manufacturing stations
Forming
Shaping
Preforming
Re-forming
Cutting
Deleted
limited applications due to high process forces
Grinding, brushing, laser beam
Not known
limited application due to high reactive foret'S. exception: jOinting in reforming process
Grinding spindle, brushing tools, drill unit, milling unit, laser device, fluidised jet head
Loading deleted
Pallet transportation of workpieces to forging presses, bending machines
Pallet transportation of work pieces to machine tools
Unloading Destacking workpieces to diecasting, gravity diecasting, injection moulding machines and sinter presses
Substantial geometrical modification
Geometry Process material testing, e.g. blowholes, surface quality
Geometry Dimensions, e.g. wall thickness, transition radius
Process implementation Workpiece handling: - End-effector and workpiece location and takeup of process forces - Protection from environmental effects, process residues Motion suffiCiently accurate for machining purposes
Tool handling: - Tool location and takeup of reactive forces - Energy signal guide to end of arm - Continuous-path control, teach-in programming and process data feedback normally required Object handling
- High process speeds - Average positioning accuracy - large working space - PoinHo-point control frequently adequate - Object presence cheel< - Low number of axes - Flexible gripper system
Swage, cooling water, burring
Testers - Tactile and visual sensors - Program synchronisation with test results - Compliance with specified environmental conditions
Geometry Dimensions, surface quality, process materials
, ...11
Manufacturing Systems • 7 Industrial Robots
out subsequently either in the form of numerical specifications or by way of subsequent teach-io programmiog on the actual robot. All systems within this category carry out a syntactical examination of the user programs which are developed, so that only user programs with the correct syntax may be loaded into the robot controller. Although a number of systems check for compatibility of the specified endeffector positions with the robot kioematics (compliance with the axis travel ranges), no verification may he made to ensure proper function of the user programs which have been developed on the real robot system (e.g. relating to collision). This is why program simulations with the relevant graphics support are required, although these do not normally form an iotegral part of systems of this type. Owiog to the lack of facilities for geometrical definition and the limited simulation and test support in a large number of cases, the use of local control off-line programming
procedures is only foreseen withio the framework of hybrid programming.
CAD-based Off-line Programming ProcedUl'es. Simulated graphics support for programming and user program testiog is, however, a characteristic for this category. Systems of this type are either based on existing CAD systems, expanded to encompass a robot-specific module, or on special developments with integrated graphics functions [14]. The functional capacity of these systems is not restricted to the actual programmiog of an industrial robot, but provides options for modelliog, programming and simulation of the entire production centre. They therefore constitute a planniog tool for robot-based systems. Such system applications are simplified thanks to the existence of libraries for robot and controller models. In addition, aids for the detlnition of new robot models or modifications
Table'. Requirements of robot systems in treatment and assembly processes Requirements
Areas of application for integrated robot manufacturing stations
Assembly
Treatment
Changing material characteristics
Coating
Jointing
Workpiece handling:
Magnctising
Object replacement
- End-effectors and workpiece location and takeup of process forces
Laser heat trcatmen!
- Contact pressing and press-fitting object
Process implementation
- Jointing in re-forming process
- Protection from environmental effects, process residues
- Folding - Inserting
- Motion sufficiently accurate for machining purposes
Tool handling:
Magnetising button
- Spray painting
- Tool location and takeup of reactive forces
Laser device
- Fibreglass plastics
- Energy signal guidance to end of arm
Nitration unit on robot
- Polyurethane coating - Cement application
- Continuous-path control, teach-in programming and process data feedback normally required
- Plasma spraying
- Welding with tongs, burners, laser - Riveting, clamping - Bolting, bonding - Gauge needle - Sensors
Object handling
- Large working space
Loading and unloading of tunnel-type furnace trucks and heat treatment equipment
- Point-to-point control frequently adequate
- Induction system
- Object presence check
- Immersion baths
- Low number of axes
- Furnaces
- High-speed system - Avet"dge positioning accuracy
Loading and unloading trays and drying shelves in galvanising and enamelling processes
- Pallet transportation of small assemblies - Loading and unloading automatic assembly machines - Jointing machines: • Riveting machines • Automatic welding machines • Bordering presses
- Flexible gripper system
Testers - Tactile and visual sensors - Program synchronisation and test results - Compliance with specified environmental conditions
Material properties, e.g. hardness
Thickness of coat, surface quality
Force and torque adjustment Inspection for compliance
'WI.M
7.7 Main Applications and Selection of Robots
7.7 Main Applications and Selection of Robots
to existing models are frequently provided. Progtam generation is generally effected either in system-specific advanced programming languages or in robot controlbased languages. Post-processors carry out translations of programs for the various controllers. The geometric data for the user programs may be derived from the workpieces and other relevant components using CAD models. CAD-based systems provide the most flexible means of support for this. Simulated testing of user programs generated off-line requires a computerised model (model generation) of the robot and its operational environment as regards all the relevant aspects, such as control, kinematics, fono and communications models. The aim is to ensure that user programs of the type undergoing testing can be operated on the actual system with the least possible amount of modification. The operation sequences are presented in the form of a visual display via computerised graphics to assist the user. The robot and associated system environment reference data are taken as the basis for programming and simulation. The two systems tend to be subject to tolerance errors and other errors, so that a user program generated off-line cannot normally be expected to be fully operable immediately. Robot and system offset measurement and error compensation are requirements for this [15].
Table 4. Requirements of characteristic Industrial robot application
~l'stem
Industrial robot applications are mainly centred around spot welding, track-mounted robot Welding, coating, assembly and machine loading technology [16). The process-related requirements for industrial areas of application are compared in Tables 2 and l [17]. The most essential system attributes include adequate working space, number of additional axes, rated load, traverse rate and scope of functions to be executed. The latter encompasses the type of control (point-to-point or continuous-path), program length, and required number of signal inputs and outputs. Requirements for characteristic system attributes for industrial robots in a variety of applications are shown in Table 4. The sheer volume of influencing factors to be taken into account in terms of the manufacturing process and handling activities when planning an industrial robot applications means that a systematic, methodical process of elimination is essential [18]. Planning tasks are processed in stages with a certain number being executed simultaneously, however. The specification has to contain the basic planning data requirements in the fono of fixed values for all those affected; in the final design of the lay-
attributes for industrial robots in a variety of fields of application
Requirement Load-
Number
bearing
of axes
capacity
Max. feedrate
Repeatability Class of (±mm) control
(mm/s)
Sensor function During
(kg)
Intermittent
Required time
for sensor data processing (ms)
movement
Handling:
Machine loading
10 to 40
4 to 5
Pallet transportation
10 to 40
_~
2 to '; 5 to 10 40 to 90 5 to 10
4 to 5 5 to 6 5 to 6 6
Continuous-path treatment processing: Coating 5 to 10
Assembly Small part assembly Adhesive application Spot welding Track-mounted robot welding
Shaping: Milling Water torch cutting Belt grinding, rotary grinding Laser beam machining
10 10 10 10
to to to to
60 60 60 60
0.5 1.5
to .,
PTP
B B,IDP
PTP
PTP
B
B, IDP B B
20
0.1 0.5 1.0 0.5
Circ.-CP
S, F, T
B
Sto7
1500
2.0
PTP (CP) B
6
40 20 60 100
0.2 0.1 1.0 0.05
Circ.-CP Circ.-CP Circ.-CP Circ.-CP
6 6 6
600
B: Binary signal processing Circ.-CP: Continuous-path control with circular interpolation CP: Continuous-path control F: Contour sequence functions geom.: Continuous-path geometry adjustment IDP: Image data processing PTP: Point-to-point control S: Search functions T: Technological value adjustment () Requirement in certain cases
to to to to
100 250 120 300
CP Circ.-CP
B, (F) B
S, F, T (F) S, F. T S, F
< 70
B
B, geom. B, (lDP)
B, geom. B, (lDP)
< 30
«
25)
< 20 < 15
1 •••f:1
Manufacturing Systems • 8 References
out planning phases, equipment selection and peripheral configurations have to be run througb individually until a satisfactory solution is reached. A preliminary decision should be made on the basis of "independent design requirements". The ultimate selection of the robot and detailed definition of all peripherals may only be carried out during the course of the layout plamting. The individual optimum arrangement of all system components may then be obtained with the aid of selected industrial robots during this planning phase. Safery, compact design, delivery, accessibiliry for main-
tenance and repairs and modification and expansion costs should be taken into account in the selection of robot systems. Computer-based planning systems have been developed to enable alternative solutions to be provided relating to operating materials, accessibiliry of three-dimensional points, collision hazards and display of execution time, sizing of memory and options for optimising performance in relation to the sequence of installation, parts and tool provision, and flow of materials based on CAD systems and simulated systems [19].
References L1 Machine Tool Components [1] Stute G. Regelung an Werkzeugmaschinen. Mtinchen: Hanser, Munich, 1981. [2] Weck M. Werkzeugmaschinen, vol. 3, Automatisierung und Steuerungstechnik, 3rd edn. VDI-Verlag, Dtisseldorf, 1989. - [3] Weck M, Ye G. Elektrische Stell- und Positionsantriebe - Systemaspekte und Anwendungen bei Werkzeugmaschinen. ETG-Fachber. 27. Berlin: VOE-Verlag 1989,pp217-231. - [4] BederkeH] ElektrischeAntriebe und Steuerungen. Teubner, Stuttgart, 1969. - [5] Birett H. Der elektrische Antrieb von Werkzeugmaschinen, 2nd edn, WB 54. Springer, Berlin, 1951. - [6] Dutcher JI. Maschinengestaltung und Regelantrieb fur numerische Steuerungen. General Electric, Waynesboro, 1963. - [7] Lehmann W, Geisweid R. Elektrotechnik und elektrische Antriebe, 7th edn. Springer, Berlin, 1973. - [8] Opitz H. Auslegung von Vorschubantrieben fur numerische gesteuerte Maschinen. RWTH Aachen, Werkzeugmaschinenlabor, 1969. - [9] Vh"M-Handbuch. Die Technik der elektrischen Antriebe. VEB Verlag Technik, Berlin, 1964. [ 10] Kennel. Ein mikroprozessorgeregelter Hauptspindelantrieb mit Feldorientierung fur Werkzeugmaschinen.
Bosch Kolloquium Antriebstechnik, Solothum, 1986. [11 J Blaschke F. D.s Verfahren der Feldorientierung zur Regelung der Drehmaschine. Diss. TV, Bnmswick, 1974. [12] Vogt G. Digitale Regelung von Asynchronmotoren fur numerisch gesteuerte Fertigungseinrichtungen. Springer, Berlin, 1985. - [13] Pfaff G. Neue Entwicklungen bei elektrischen Servoantrieben. tz fur Metallverarbeitung (1984) H.8, 15-21. - [14] Zimmermann P. Btirstenlose Servoantriebe fur Werkzeugmaschinen. wt-Z. indo Fert. 73 (1983) 629-32. - [15] Henneberger G. Servoantriebe fur Werkzeugmaschinen und Roboter. Stand der Technik, Entwicklungstendenzen. Conf. Proc. ICEM, Munich, Sept. 1986. - [16] Polymotor. Firmendruckschrift FASTTACT, Genoa, 1987. - [17] Wolters P. Lageregelung fur asynchrone Linearmotoren. Ind. Anz. 99 (1977) 129ff. - [18] G6tz FR. Hochdynamische Antriebe Umrichtergespeiste Drehstromservoantriebe und Linearmotoren. ETG-Fachber. 27, Berlin: VDE-Verlag 1989, pp 159-167. [19] Stute G. Untersuchungen tiber die Verwendbarkeit von Gleichstromnebenschlussmotoren als Vorschubantriebe fur numerisch gesteuerte Werkzeugmaschinen. VDWBer. TU-Stuttgart, Inst. fur Stcuerungstechnik, 1971. [20] Wilhelmy I. LongLife-Tachody-namos im Vergleich zu anderen modemen DrehzaWistwertaufnehmem fur die Antriebs- und Regelungstechnik. ETG-Fachber. 27. VDEVerlag, Berlin, 1989, pp 147-57. - [21] Backe W. Umdruck zur Vorlesung "Gnmdlagen der Olhydraulik". Inst. fur hydraulische und pneumatische Antriebe und Steuerungen, RWTH, Aachen, 1988. - [22] Backe W. Umdruck zur Vorlesung "Servohydraulik". Inst. fur hydrau-
lische und pneumatische Antriebe und Steuenmgen, RWTH, Aachen, 1986. - [23] Backe W. Fluidtechnische Realisierung ungleichmassiger periodischer Bewegungen. Olhydraulik und Pneumatik Mai (1987) 22-8. - [24] Backe W. Neue Miiglichkeiten der Verdriiugerregelung. Tagungsunterlagen zum 8. Aachener Fluidtechnischen Kolloquium, vol. 2, 1988, pp 5-59. - [25] R6gnitz H. Stufengetriebe an Werkzeugmaschinen, 4th edn., WB 55. Springer, Berlin, 1965. - [26] Riegel F. Rechnen an spanenden Werkzeugmaschinen, vol. 1, 5th edn. Springer, Berlin, 1964. - [27] DIN 781: ZahnezaWen fur Wechselrader. Beuth, Berlin, 1973. - [28] Scb6pke H. Grundlagen der Konstruktion von Werkzeugmaschinengetrieben. Westermann, Brunswick, 1960. - [29] R6gnitz H. Getriebe fur Geradwege an Werkzeugmaschinen, 2nd edn, WB 101. Springer, Berlin, 1964. - [30] Dilrr A. Wachter 0. Hydraulik in Werkzeugmaschinen, 6th edn. Hanser, Munich, 1968. - [31] Ebertsbiiuser H. Bauelemente der Olhydraulik, O+P TB 3. Krausskopf, Mainz, 1974. - [32] Krug H. Fltissigkeitsgetriebe bei Werkzeugmaschinen, 2nd edn. Springer, Berlin, 1959. - [33J Wiessner H. Ober pneumatisch-hydraulische Vorschubeinheiten. wt 55 (1965) 163ff. - [34J VDI-Richtlinie 3230. Technische Ausfuhnmgsrichtlinien fur Werkzeugmaschinen und andere Fertigungsmittel, H-Hydraulische Ausriistung. VOl-Verlag, Dtisseldorf, 1967. - [35] VDI-Richtlinie·3229. Technische Ausfuhrungsrichtlinien fur Werkzeugmaschinen und andere Fertigungsmittel, P-Pneumatische Ausriistung. VOl-Verlag, Dtisseldorf, 1967. - [36] Weck M. Werkzeugmaschinen, vol. 1. Maschinenarten, Bauformen und Anwendungsbereiche, 3rd edn. VOl-Verlag, Dtisseldorf, 1988. - [37] Sahm D. Reaktionsharzbeton fur Gestellbauteile spanender Werkzeugmaschinen. Diss. RWTH, Aachen, 1987. - [38] Weck M Werkzeugmaschinen, vol. 2, Konstruktion und Berechnung. VDI-Verlag, Dtisseldorf, 1990. - [39] Rinker U. WerkzeugmaschinenFtihrungen, Ziele ktinftiger Entwicklungen, VOI-Z 130 (1988). - [40] Weck M, Rinker U. Einsatz von Geradfuhrungen an Werkzeugmaschinen. Ind. Anz. 79 (1981). [41] DIN 50 320: VerscWeiss, Begriffe, Systemanalyse von VerscWeissvorgangen, Gliederung des Verschleissgebietes. Beuth, Berlin, 1979. - [42] Weck M, Rinker U. Reibungsverhalten von Gleitfiihnmgen. Einfluss der Oberflachenbearbeitung. Ind. Anz. 28 (1986) - [43 J Untersuchungen am Lehrstuhl fur Werkzeugmaschinen. RWTH, Aachen, 1988. - [44] Weck M, Miessen W. Optimierung und/oder Berechnung hydrostatischer Radial- und Axiallagerungen. KfK-CAD 77. Kemforschungszentrum, Karlsruhe, 1979. - [45] Peeken H, Benner] In: Goldschmidt infonniert. Aus der Arbeit der Th. Goldschmidt AG, no. 52, 1980. - [46] VDI-Richtlinie 2201: Gestaltung von
Manufacturing Systems • 8 References
Lagerungen, vols 1 and 2. VDI-Verlag, Dtisseldorf, 1975. [47] VDI-Riehtlinie 2202; Schmierstoffe und Schmiereinrichtungen fur Gleit- und Wiilzlager. VOl-Verlag, Diisseldorf, 1970. - [48] VDI-Riehtlinie 2203 Gestaltung von Lagerungen, Gleitwerkstoffe. VDI-Verlag, Dtisseldorf, 1964. - [49] VDI-Riehtlinie 2204. Cileitlagerberechnung hydrodynamischer Gleitlager fur stationare Belastung. VOl-Verlag, Dtisseldorf, 1968. - [50] Briiuning H. Einsatz mehrflachiger hydrodynamischer Gleitlager in Schleifspindellagerungen. Ind. Anz. 59 (1973). - [51] Week M, Koch A. Vergleich von Hauptspindel-Lager-Systemen. Vortrag am Lehrgang: Konstruktion von Spindel-Lager-Systemen fur die Hochgeschwindigkeits-Materialbearbeitung an der TAE-Esslingen, 1988. - [52] Korrenn H, Kleinhenz G, Voll H. Spindellagerungen in Werkzeugmaschinen. Tell I: Wiilzlager; Tell 2: Gleitlager. Klepzig Fachberichte l2 (1972) and 3 (1973). - [53] Briindlein j. Eigenschaften von walzgelagerten Werkzeugmaschinenspinde1n. FAGPublikation no. WL02113 DA. - [54] Ophey L. Entwicklung schnellaufender, wiilzgelagerter Hauptspindeln fur Werkzeugmaschinen. VOW Forschungsberichte 1986. [55] Week M, Koch A. Experimentelle Untersuchung von Hochgeschwindigkeits-Spindel-Lager-Systemen mit Walzlagem. Vortrag am Lehrgang: Konstruktion von Spindel-Lager-Systemen fur die Hochgeschwindigkeits-Materialbearbeitung an der TAE-Esslingen, 1988. - [56] Week M, Ophey L. Walzgelagerte Spindel-Lagersysteme fur die Hochgeschwindigkeits-Hochleistungs-Bearbeitung. Ind. Anz. 37 (1987). - [57] Week M, Steinert T Konstruktive AusJegung der Walzlagerung schnellaufender Werkzeugmaschinen-Spindeln. Vortrag am Lehrgang: Konstruktion von Spindel-Lager-Systemen fur die Hochgeschwindigkeits-Materialbearbeitung an der TAE-Esslingen, 1989. - [58] Giehner E. Die AusJegung von Arbeitsspindellagerungen. SKF Publikation no. WTZ 83 06 20. - [59] Fritz E, Haas W, Muller HK. Abdichtung von Werkzeugmaschinenspindeln. Konstruktion 41 (1989). - [60] Week M. Werkzeugmaschinen, vol. 4, Messtechnische Untersuchung und Beurtellung, 2nd edn. VDI-Verlag, Dtisseldorf, 1985; [61] Week M, Teipel K. Dynamisches Verhalten spanender Werkzeugmaschinen. Springer, Berlin, 1977. L2 Control Systems [I] Berthold H. Programmgesteuerte Werkzeugmaschinen. VEB-Verlag Technik, Berlin, 1975. [2] Week M. Automatisierung und Steuerungstechnik. Werkzeugmaschinen, vol. 3. VDI-Verlag, Dtisseldorf, 1989. - [3] Herold H-H, Massberg Iv, Stute G. Die numerische Steuerung in der Fertigungstechnik. VDI-Verlag, Otisseldorf, 1971. - [4] Simon Iv, Die numerische Steuerung von Werkzeugmaschinen. Hanser, Munich, 1971. - [5] Ammann J. Cirundlagen der Pneumatik und Hydraulik, 3rd edn. Halscheidt, Heidenheim, 1973. - r6] Durr A, Wachter O. Hydraulik in Werkzeugmaschinen. Hanser, Munich, 1968. - [7] Hemming Iv, Steuem mit Pneumatik. Archimedes, Kreuzlingen, 1970. - [8] Binder D. Interpolation in numerischen Steuerungen. Springer, Berlin, 1979. - [9] Schmid D. Die numerische Bahnsteuerung. Springer, Berlin, 1979; - [10] Walker B. Konfigurierbarer Funktionsblock Geometriedatenverarbeitung ftir numerische Steuerungen. ISW Forschung und Praxis, vol. 68. Springer, Berlin, 1987. - [11] Pritschow G. (ed.). Die Lageregelung an numerisch gesteuerten Maschinen. SeJbstveriag FlSW-GmbH, Stuttgart, 1986. - [12] Stute G. Regelung an Werkzeugmaschinen. Hanser, Munich, 1981. - [13] Egner M. Hochdynamische Lageregelung mit elektrohydraulischen Antrieben. ISW Forschung und Praxis, vol. 74. Springer, Berlin, 1988. - r14] Speicherprogrammierbare Steuerungsgeriite. VDl-Ber. 481 (1983). [ I 5] Storr A. Planung und Steuerung tlexibJer Fertigungs-
'WI.g
systeme. Selbstverlag ISW, Stuttgart, 1984. - [16] Spur G, Krause ILL. CAD-Technik. Hanser, Munich, 1984. - [17] Spur G, Stute G, Week M (eds). Fortschritte der Fertigung auf Werkzeugmaschinen, vol. 4 (6). Rechnergefuhrte Fertigung (Beitrage zur Weiterentwicklung der Automatisierungs-technik). Hanser, Munich, 1977 (1983). - [18] Walcher H. Digitale Lagemesstechnik. VOl-Verlag, Diisseldorf, 1974. - [19] Philips AG. Philips-Linear-Mess-system LM SIV. Company document, Eindhoven. - [20] HewlettPackard. Laser Transducer Systems Computer Interface Electronics. Company document, California, USA, 1976. [21] DIN 19237. Steuerungstechnik, Begriffe. - [22] DIN 19239. Speicherprogrammierbare Steuerungen, Programmierung. - [23] DIN/lEC 65A (SEC) 67. Speicherprogrammierbare Steuerungen, T.3: Programmiersprachen. [24] VDI/VDE-Richtlinie 2600; Messtechnik, paper 2. VOl-Verlag, Dtisseldorf, 1973. - [25] DIN 1319; Grundbegriffe der Messtechnik. L3 Shearing and Blanking Machines [I] Hellwig Iv, Automatisierung in der HochleistungsStanztechnik, VOlBer. 694 (1988) 251-73. - [2] Hellwig Iv, Entwicklungsfortschritte in der Stanzerei. Bander Bleche Rohre 31 (1990) 1,73-8. - [3] Oehler/Kaiser: Schnitt-, Stanz- und Ziehwerkzeuge, 5th edn. Springer, Berlin, 1966. L4 Presses and Hammers Jor Metal Forging [I] Lange K (ed.). Umformtechnik - Handbuch fur Industrie und Wissenschaft, vol. 1: Grundlagen, 2nd edn. Springer, Berlin, 1984. - [2] Doege E. u.a. Tiefziehen auf einfachund doppeltwirkenden Karosseriepressen unter Beriicksichtigung des Gelenkantriebs. Werkstatt u. Betrieh 104 (1971) 737-47. - [3] Siegert K. Einfachwirkende mechanische Karosseriepressen mit hydraulischer Zieheinrichtung im Pressentisch. ZwF CIM-Zeitschrift fur wirtschaftliche Fertigung und Automatisierung 83 (1988) Sondemummer 24-6. L7 Industrial Rohots [1] Spur G, Auer BH, Sinnig H. Industrieroboter. Hanser, Munich, 1979. - [2] Vukobratovic M, Kircanski M. Scientific fundamentals of robotics 3: Kinematics and trajectory synthesis of manipulation robots. Springer, Berlin, 1986. - [3] ISO. Manufacturing message specification (MMS). ISO 9506,1989. - [4] ISO: Robot companion standard to MMS. ISOjTC 184jSC 2/WG 6 N6&, 1988 (Draft). - [5] Duelen G. Robotersteuerungen. Automatisierungstechnische Praxis 30 (1988) 410. - [6] VDI: Industrial robot data (IRDATA). VOl 2863. 1987. - [7] VDI-Richtlinie 2860: Handhabungsfunktionen, Handhabungseinrichtungen, Begriffe, Definitionen, Symbole. VDI-Verlag, Dtisseldorf, 1982. - [8] Spur G. Stand der Programmiertechnik fur Industrieroboter. Vortrag am FTK '88, Stuttgart, 5-6 Oktober 1988, und Werkstattstechnik, Sonderheft FTK 17, 1988. - [9] Prager K-P' Kopplung extemer Programmiersysteme fur Industrieroboter. Reihe Produktionstechnik Berlin, vol. 33. Hanser, Munich 1983. - [10] Pritschow G, Gruhler G. Selbstprogrammierung von Industrierobotem durch Ftihrung im geschlossenen Sensorregelkreis. VOl·Ber. 598. VDI-Verlag, Dtisseldorf, 1986. - [II] Balling G, Fuehrer D. Einfache Programmerstellung fur Roboter durch sensorgesteuerte Raumpunktgenerierung, Energie & Automation 9 (1987) 12-14. - [l2] Rembold U, Frommherz B, Hormann K. Programmiertechnik fur Industrieroboter - Stand und Tendenzen. Techn. Rundsch. (1986) H. 25, 96-108.[13] Hocken R, Morris G. An overview of oftline robot programming systems. Ann. CIRP 35 (1986). - [14] Spur G, Kirchhoff U, Bernhardt R, Held .H. Computer-aided application program synthesis for industrial robots. In CAD-based programming for sensor-based robots. Nato
,w.".
Manufacturing Systems • 8 References
Advanced Research Workshop, 4-6 July, II Ciocco, Italy, 1988. - [15] Due/en G, Held j, Kirchhoff U. Approach for the estimation of kinematic parameters and joint stiffness of industrial robots. Robotics and flexible automatization. Proc. 5th Yugoslav Symp. Applied Robotics and Flexible Automatization, Bled, Yugoslavia, 1-4 June 1988. - [16] Spur G et al. Anforderungsprofile fUr die Weiterentwicklung der Robotertechnik. 4th Conference
"Jurob 88", Ljubljana, Yugoslavia, 11-12 April 1988. [17] Severin F. Planung der Flexibilitat von roboterintegrierten Bearbeitungs- und Montagezellen. Hanser, Munich, 1987. - [18] Furgac 1 Aufgabenbezogene Auslegung von Robotersystemen. Hanser, Munich, 1985. - [19] Deutschlander A, Severin F. Rechnerunterstiitzte Layout-Planung fur Industrieroboteranwendungen. ZwF 81 (1986) 51522.
Index
Abbe coefficient L49 Abrasives, grain size KI14 Absorption number C34 Acceleration instantaneous centre of A25 particle AI9 natural components A22 Accident prevention, see Safety standards Accuracy, machining L I Ackeret - Keller process C20 Acme thread F36 spindles L16 Acoustics basic concepts J29- 31 power level J30 see also Noise Acrylate rubbers (ACM) D59 Acrylonitrile butadiene rubbers (NBR) D59 Acrylonitrile butadiene styrene (ABS) D56 Action control, robot L 103 Adaptation of work to people, manufacturing processes KI Adhesive bonding F21 Adiabatic systems C2 changes of state C5, CI4 entropy changes C7 exponent C II Adjusting, assembly K93 Aero·emulsions D74 Aerodynamic drag A6S Aerofoils A66-68 Aesthetics, design for E 19 Aggregation states K3 Air cooling, indirect H27 Air ejectors H26- 27 Air entrainment, luhricating oils D74 Air pumps H26 Air resistance, of vehicles A65 Aircraft, steels for D40 Airy's stress function 837 Alitising D32 Aluminisation D32 Aluminium and aluminium alloys D45-46, Dl03-109 as alloying element D33 aluminium bronze D44, DI02 casting alloys D46, D109-110 weldability F8 Amino plastics (MF, UF) D57 Amplitude-frequency response JI4 Analog data logging LSI-52
Analysis process, problem-solving E4 Anchors, piping HI3 Anergy C8 Angular momentum conservation law A30, A32, A38 equation A30, A32 Anisotropy K22 Annealing D30 coarse-grain D30 recrystallisation D30 soft D30 steel strength values D80 Annular discs, under in-plane loads B41 Anti-vibration mountings F61-62 Antoine's equation CIO constants table C36 Anvil hammers L62 Arc welding machines, see Welding machines, arc ASA (plastic) D56 A-S-I method (adaptation, substitution and/or integration) KI Aspect ratio A68 Assembly process K91 automated K96 design for E19 planning K93 systems K95-96 Asynchronous control L35
Attenuation measurement D24 Austenite D26 Autocatalytic plating KI9 Automation assembly K96 materials handling KIOI-102 Avogadro's constant C9 Axial locking devices F32-33 Axis coordinates L47
Bach's correction factor B6 Backlash LI8 Ball bearings, see Bearings, ball Ballscrew mechanisms L 16 Ballvalves H17 Bandsawing K48 machines L88-89 Bar stress constant axial load B7 notched bar B7 temperature variation 87 variable axial load B7 variable cross section B7 Basic design E 13
Bauschinger effect (fatigue strength) B 56 Beam cutting (material removal), see Jet cutting Beam elements, use of FEM B52 Beam welding F3-4 Beams arbitrary loads B 11 bending stresses B12-16 continuous B33-34 deflection B20- 26 equal bending loads BI6 forces and moments B8, B 11-12 highly curved B19 lateral buckling B48 line loads B 10 plane angled B II plane curved B II single loads B 10 single moments B\o slightly curved B26 Bearing pressures, see Stress, contact Bearing seals F98 Bearing shells/liners F97 -98 Bearings ball F77-78, L16, L33-34 dry F99 feed drive Ll6- 18 hydrostatic jacking systems F99 hydrostatic journal F99-100 hydrostatic thrust Fl 00-10 I lobed F98 machine tool Ll, L25-26, L3134 materials F97, F98, F191 multi-pad plain F98 oil flow rate F92 plain design F89 form design F96-98 friction F89 lubricant supply F96-97 plain journal steady radial loading F89-92 variable radial loading F92 plain thrust F92-96 reactions A34 roller F78-79, L16, L32-33 steels for D39 rolling assembly design F87 -89 fatigue life F79-82, F83 friction F87 fundamentals F75 heating F87 limiting speeds F83 load capacity F79, F82
2
Index. C
Bearings (cant,) lubrication F84-86 service life F83 types F77 - 79 sliding L32 temperature calculations F9 I, F189-190 types of A7 Beat (vibrations) A42 Belleville springs, see Conical disc springs Bellows compensators HI4 Belt drives FlOl flat FlOI-I07 calculations F107, FI92 constant speed ratios F104105 materials F106 step-cone pulleys F103-104 tensioning F105-106 variable-speed drives FI03104 machine tool L13 synchronous F109, F193, LI8 toothed, see belt drives, synchronous v- F107-109 calculations F108-109 types and sizes Fl08, FI92 Belt grinding K54-55 data table K115 Beltram's equations B36 Bending arrow B45 Bending line, effect of shear strains B25-26 Bending sheet forming process K29-30 methods K30-31 Bending strength test method D20 plastics D60 Bending stresses double BI4 graphical determination of moment B11 highly curved beams B 19-20 oblique loading B 14 simple loading B 12-13 straight beams B12-16 Benedict-Webb-Rubin equation (gas properties) C9-10 Bernoulli's criteria B12, B56 Bernoulli's equation A51-52, A53, A62, A66, 61 Bessel functions A47 Betti's Law (equilibrium conditions) B54, B55 Biaxial moment of area B 12 Billet shears/parters L52 Bingham Medium (fluid flow) A59 Biot index C28 Black body C33 Blade rows, see Cascades Blades, see Aerofoils Blanking machines L53-54 Blanking processes K61, K62 defects K64 sequence K62-63 special K65-66
tools K64 Blasius formula (fluid flow) A53 Blister packing method D65 Blockboard D53 Blocks, modular E22-23 Blowing (of plastics) D63 Body measurements, representative K107-108 Boiler plate steels D36 Boiling point C2 Boiling process C32 Bolted connections, see Connections, bolted Bolts and nuts F36- 38 large F45 resilience F39 steels for D39, D85-89 stress in B8 in thermal apparatus H8 Bonding adhesive, see Adhesive bonding of plastics D65 Boring machines L7S-79 horizontal L83-84 Boron (as alloying element) D33 Boron nitride (PBN) , cutting material K50 Boronising D32 Boundary element method (BEM) B53 beam B54 discs, plates and dishes B54 Boundary layer hardening D31 steels for D35 Boundary layer theory (fluid flow) A63, A64 Boundary zone, heat transfer C28 Boussinesq's formulae (stresses)
B37, B38 Brainstorming method (for problem solving) E5 Brakes F73 for presses L58 Brasses, see under Copper and copper alloys Brazing, see Soldering Brazing equipment LlOO Bredt-Batho theory (thin-walled tubes) B28 Bred!'s equations first B28 second B30 Bricks D49-50 fireproof D50 Brinell hardness test D20- 21 Brittle materials, stress distribution B5 Broaching machines L86-87 Broaching process K47 -48 Bronzes, see under Copper and copper alloys Buckingham'S II-theorem A 71-72 Buckling bars B45-48 beams B48 columns B47 domed end closures H8 energy method B46-47 frames B47
inelastic (plastic) B46, B50 plates B48-49 rings B47 safety value B45 shells B49-50 torsional B47 -48 Building board D53 Buildup technique, see UGA process Burmester circle/centre point curves FI67 Bum characteristics, plastics D62 Butterfly valves, see Valves, flap Butyl rubbers (IIR) D59
C-ftames, press L57 Cables A12-13 point load AI3 uniform load A13 Calendering, of plastics and rubbers D63, KIS Caloric state variables C9 table of C49 Calorilic value fuels C49, C50 gross CI7 net C17 Calorisation D32 Cam control L38 Cam mechanisms kinematics FIS9, FI68 press LS7 Cam-forming machines L83 Cam-throttle control L37-38 Capacity utilisation K 100 Capillary tension A49 Carbides, cemented K49-50 Carbo-nitriding D31 Carbon (as alloying element) D33 Cardan shaft F65 Carnot cycle C19-20 Carnot factor C8 Cartesian coordinates L47 Cascades A68 Case hardening D31 steels for D35, D80 Cast iron, see Iron, cast Cast parts, manufacturing K4-12 Cast steel, see Steel, cast Castigliano's theorem B26 Casting process aluminium alloys for D46, D 109111 centrifugal K 11-12 cleaning operations K 13 continuous K4 guidelines for design K12, K14 ingot K3 inspection and testing KI5 machines for K4 manufacturing of parts K4-12 melting of materials KI3 plastic sheet KI5 semi-finished products K3-4 steel D28 strip and wire rod K4 Catenary AI2-13 Cauchy number E20
C. Index
Cauchy-Riemann differential equations A62 Cauchy's Similarity Law A70 Cavity sinking, ultrasonic K85 Cell control L36 Cell-type manufacturing systems L97 Cellulose derivatives (CA, CP, CAB) 056-57 Celsius Scale C2 Cementite 026 Cements 051,0112 Centre of gravity A13- 14 Centre of rotation, instantaneous A24-25 Centre-of-mass law A30, A36 Centrifugal casting K 11- 12 Centrifugal clutches F74 Centrifugal stresses, rotating components B43-44 Centroid gyro A38 Ceramic materials 049-50, 0111 inserts K50 moulding K7-8 Certnets, see Carbides, cemented Chain dtives FIOI, FI09-110, FI93 Chain transmission, machine tool LI3-14 Chains, see Cables Changes of state adiabatic C5, CI4 gases and vapours at rest C14-15 in motion C15-16 irteversible C4 isentropic CI4 for primary shaping K3 Channel bending, see U-bending Chatter, machine tool L4 Chemical analysis methods 023 Chemical industry, use of condensers H24, H25 Chemical milling/machining K57, K61 Chemical vapour deposition (CVD) 048, K49, K86, K91 Chill casting KlO low pressure KlO Chip formation grinding process K52 machining K35-36 non-defined tool edges KS I turning K38-39 Chipboard 053 Chipless machining laser beam K83 processes classification K85 Chromating 049 Chromising 032 Chromium (as alloying element) 033 Churchill & Chu equation (free convection) C32 Circuits, hydraulic G12, GI5-16, L61 Circular sawing K49 machines L88-89 Circular shears L52-53 Clack valves, see Valves, flap
Clamp joints F27-28 Clamping/support mechanisms, lathe L67 Classification systems, item E30-31 Clausius-Clapeyron equation (entropy of evaporation) CI2 Clausius-Rankine process (reversible cyclic) C21 Clausius's inequality C7 Clevis joints F29 Clocked control L35 Closures domed H7-8 flat end H7 Cloud point, lubricating oils 073 Clutches F65, F69-73 automatic F74-5 dog F69 freewheel F75 friction F69-70 engagement slip F70-72 layout design F72 operating methods F73 size selection F72 type selection F72 overrun F75 press L58 slip clutches F74 Co-generation, see Combined power and heat generation Coating processes K86-87 laser beam K83 on metal 048-49 non-metallic 048-49 plastics on metals K86 surface K90-91 Cobalt (as alloying element) 033 Cocks (piping) H17 Cold forging K34, L61 Cold forming processes K20 effect on material properties 09 Colebrook-Nikuradse diagram A54 Colebrook's fortnula (pipe friction) A54 Collision forces, coupling L20 Collision time A39 Colour assessment, of plastics 062 Combined power and heat generation C23 Combined stresses axial load and torsion B32 bending and axial load B31 bending, axial load, shear and torsion B32 bending and shear B32 bending and torsion B32 shear and torsion B32 Combustion processes C 16-18 temperature in C 17-18 Commutators, d.c. motor L8 Compacting, hot -cold (heat treatment process) 032 Compensation effects, overload E 14 Complex velocity potential A62 Compliance frequency response L4 Components, relative expansion El7
3
Composite design E19 Composite (integral) casting K12 Compressed air, pneumatic installations G 17 Compressible fluids, see Fluids, compressible Compression bars (spring) F52 Compression moulding K16 Compression refrigeration, see Refrigeration process, compression Compression strength test method 020 plastics 060 Compressors G16 Computer aided design (CAD) systems Ll05-106 Computer aided production planning K98 Computer aided quality control (CAQ) KlO4 Computer integrated manufacturing (CIM) system L37-38 Computer numerically controlled (CNC) systems L68 Conceptual design E 11-12 Concrete OSO blocks and slabs OS I lightweight 051 prestressed OS I reinforced 050-S 1 Condensation process ClO, C32 principles H23-24 Condensers air-cooled H26 in chemical industry H24, H25, H26 direct-contact H25 injection H25 low-pressure saturated vapour H25 for power stations H26 in steam power plant H24 surface H24 Conduction, thermal C26-27 Configuration guidelines D66 Confrontation, problem-solving E4 Conical disc springs F55-56 Connecting rods F173, F174, F176 Connections adhesive F21 bolted F34 loading F40-45 strength F42-45 tightening F38-40 elastic, see Springs friction F23-28 positive F28-34 type selection F47-50 Consolidation index (material hardness) K21 Constant-power control G14 Constraiitts, task E4 Constructional interrelationships E4 Contact conformal 068 contraformal 068 stress B8
4
Index. 0
Contact controls L40-41 Contact corrosion E18 Contactor L40 Continuous casting processes K4 Continuous-path control L46, L58 robot LI03 Contraction coefficients D77 Control data processing L44-45 Control programs L35-36 Controls asynchronous L35 cam operated L38 clocked L35 electrical L40-4I electronic L40-41 hydraulic transmissions G 13-14 levels of L37 means of operating L3 7 - 38 organisation of L36-37 robot L102-104 safety standards L38 signal processing L35 system structure L34-35 Convection heat transfer C26, C3033 Coolants, refrigeration C22 Cooling lubricants, machining processes K52 Cooling ponds H27 Cooling towers H27 design calculations H27 - 28 Coordinate conversion L47 Coplanar forces A2 equilibrium conditions A6 Copper, weldability F8 Copper and copper alloys D33, D43, D93, D94 -aluminium ailoys (aluminium bronzes) D 102 -lead-tin alloys D103 -tin alloys (tin bronzes) D101102 -zinc alloys (brasses) D43-44, D95-98, D99-1oo bronzes D44 Core zone remelting D28 Coriolis acceleration A25 Correction values table (cutting processes) KI09 Corrosion design for avoidance E 17-18 types D5, El7-18 Corrosive media, effect on material properties D 10 Cost accounting KI07 Cost centre calculation E8 Cost location accounting K106 Cost type accounting KI05 Costing E9-1O operational K105-107 Costs, types of KI05 Cottered joints F30 Couette flow A63-64 Couette viscosimeter A64 Coulomb's law of friction A42, K23 Counterblow hammers L62, L63 Couples Al Couplings L20, F64-65
elastic F65-69 resilient metal F68 rigid F65 self-aligning F65 torsionally stiff F65 CR, see Polychioroprene rubbers Crack ductility, test methods D22 Crack formation/propagation D3-4 weld joints F6 Crank mechanisms A24 components F1n-l76 designs F 168 dynamics FI70-172 kinematics Fl68-l70 machine tool LI4 press L57 Crank-rocker mechanisms L14 Crankshafts FIn, F174, FI75 oscillatory system J11, J19 Creep 056 strength calculations DIS Cremona force diagram All Crevice corrosion E 17 Critical loads, approximation methods 046-47 Culmann's auxiliary force A3 Curing process, thermo-setting plastics K3 Curved shears L52-53 Cutouts, pressure vessel H8 Cutter location data (CLDATA) L44 Cutting off process K62 Cutting processes K35 speeds Ll steels for D38 Cutting tools, materials for K49, K50, K1l7 CycliC process C 19 Cyclo gear mechanisms L19-20 Cylinders bound hollow 042 elliptical 042 pneumatic LIS rotating 044 thick-wall 043 Cylindrical shells under external pressure H6-7 under internal pressure H6 Cylindrical ways L28 Cyril-Bath process (stretch forming) K28
D'Aiembert's Principle A29, A3637,043 constrained motion A3I hydrostatic paradox A62 Dalton's Law C23 Damping machine tool L4 modal J12 vibration A42-45, J12 Damping couplings F67 Databases, manufacturing system L37-38 Debye temperatures C 13 table of C49 Decimal-geometric series E20 Deep drawing
plasticity theory K27 - 28 technology K28-29 Deep-hole drilling machine L 76 Deflection beams 020-26 oblique bending 025 slightly curved beams 026 Deformation E15, E16 material flow K20- 21 plastic H7 technology 056 test methods for determining D19 Deformation energy theory D5 Degrees of freedom A23 Delphi method (for problem solving) E5 Demand calculation, material K9899 Denavit - Hartenberg convention (robot structure) LI02 Density A49 DeSign process E 10-11 basic E13 conceptual, see Conceptual design detail, see Detail design embodiment, see Embodiment design redesign E 13 variants E 13 Detail design E 12 Deutsche Institut fUr Normung (DIN) Standards, see DIN Standards Development and design, problemsolving E4 Diathermal walls C2 Diathermic body C33 Die, wire drawing K25 Diesel engine C18 Differential design E 19 Diffusion coatings D48 Diffusion mechanisms, sintering K3 Diffusor flow C 16 drag coefficient A56 Digital absolute measuring systems L50-51, Ll02 Digital data logging L49-50 Digital incremental measuring systems L49-50, LI02 Digital locators LI02 Dilatant liquids A59 Dilatometer measurement D24 Dimensions, types of E25 DIN (Deutsche Institut fUr Normung) Standards E2324 Dip soldering L 100 Dipole current flow A62 Disc friction A65 Disc model, plasticity theory K25 Discs rotating 043-44 rotation of two A26 skewed angle rotation A34 under in-plane loads 041
E. Index
Discursive methods, problem-solving E5 Displacement control, motor LIO Displacement method, FEM B50 Displacement power, pump G5 Distortion B3 Distribution laws, sample evaluation D17-18 Dittus-Boelter-Kraussold equation (heat transfer) C31 DNC operation, robot control L103 Double-acting hammers L62, L63 Drag, solid bodies A65 Drag coefficient ASS-57 Drag force A67 -68 Drawing process L S9 Drilling machines L7 3 bench L74 column-type L74-75 deep-hole L 76 free-standing piller L7 4 multi-spindle L75 precision L 76 radial L7S-76 special purpose L 76-79 turret L76 Drilling process K41-43 laser beam K83 short-holes K43 Drives control of L7, L9-IO, L13 hydraulic G15-16, L61-62 control G13-14 machine tool LI, L4 pneumatic, see Pneumatic drives press L57-58 Drop forging K33 Drop hammers L62-6; Drop short circuits, welding L97 -98 Ductility D7 at fracture D3 crack D22 processes affecting D8-10 Dulong-Petit rule Dynamic loading, strength calculations D 13 Dynamic similarity A70-71 Dynamic stress, weld joints FI6 Dynamic viscosity AS3
cn
Economic efficiency, manufacturing processes KI Effective motion L1 Effective principle E2, E4 Efficiency A28 heat exchangers H 4 Eigenvectors J12, JI3 Elastic buckling, see Euler buckling Elastic couplings, see Couplings, elastic Elastic expansion, scale factor A69, A70 Elastic hardening material B56 Elastic limit D 19 Elastic modulus, test methods for determining D 19 Elasticity B 36
Elasto-plastic material B56 Elastomer couplings F68-69 Elastomers D54, D58-59 polyurethane (PUR) D59 thermoplastically processable (TPE) D59 for vibration reduction F66-67 Elastoviscous substances A59 Elbow pieces, drag coefficient A55 Electric slag remelting D28 Electrical discharge in gas cutting Fill Electrical machines, steels for D40, D41, D94 Electrical perturbation moments, excitation by J20 Electrochemical machining (ECM) K57, K60, K85 Electrodischarge machining (EDM) K57-59, K85 cavity sinking K58, K59 cutting by K 59 material pairs table K1l5-116 thread production K69 Electroforming KI9 Electrohydraulics L40 drives LlO-ll Electron beam processing applications K83-84 inscriber K88 lithography K1l4, K87 micro-analysiS method D23 principles K83 vapour deposition K84 Electroplating K91 Ellipsoid of inertia A34 Elongation B3, B4 permanent limit B5 Embodiment design E 12 basic rule E 13 guidelines E 16 principles E 14-16 Emission index C34
table of C53 Empirical temperatures C2 EN Standards (European Standards) E23 Enamelling D49 Encoders, digital absolute measuring systems L50 Energy C3 in closed systems c4 dissipated Cll forms of C3-4 internal (stored) C4, Cll, C12 in open systems C4-5 Energy conservation law A36, A38 Energy efficiency, overall C18 Energy equation A29, A30 Energy method, buckling load B4647 Energy transfer, fluid drives G I Engesser-von Karman's curve (buckling) B46 Engineering drawings conventions and scales E27 dimensioning E27 lines and lettering E27
5
sizes and formats E27 types E25-27 Engineering surfaces E24-25 Engines diesel C18 fan-type, force calculations J8-9 flat, force calculations J9 internal combustion CI8 machine vibrations JII pressure graph JI radial, force calculations J8-9 two-cylinder, force calculations J5 V-in-line, force calculations J8, J9 Engraving machines L82 Enthalpy C5, CI!, Cl2 humid air C24 Entropy C6, CI2 generation C6 Entropy flux C6 Environment, effect on material properties D 10 Epicyclic gears A26 Epoxy resins (EP) D58 EPS, see Polystyrene, expanded Equal embodiment strength principle EI6 Equations of motion machine vibrations JI!, J25-29 see also Motion Equilibrium conditions coplanar forces A6 forces in space A5 thermal C2 types of A7 ErgonOmics E18, K107-108 Ericson process C20 Error technology, manufacturing processes KI ETFE (plastic) D57 Ethylene propylene rubbers (EPM, EPDM) D59 Euler buckling B45 Euler datum system (fluid mechanics) CI Euler's equations of motion A23, A37, ASI, A60 for stream filament A51 Euler's rope friction formula Al718 Euler's similarity parameters A 71 Euler's velocity formula A23 Evaluation procedure, solution E6, E7, E8 Evaporation process C33 Excitation spectrum, noise J31, J34 Exergy C7 closed system C7-8 and heat C8 losses C8, H5 open system C8 Expanding process, plastics forming KI6 Expansion thermal CI2 design for El7 solids C12, C48 table of coefficients C48
6
Index. G
Expansion compensators H13 bellows drag coefficient A56 loops H13 Exponential equations, use of E2122 External forces Al Extruders (machines) L61 Extrusion processes K15 direct K34 hydraftlm K35 hydrostatic K35 indirect K34-35 plasticity theoty K26-27 plastics D63
F
actors of safety, see Safety factors Fahrenheit Scale C2 Fail-safe principle E16 Failure causes of D2 stress failure criterion 86 under alternating stress D3-4 under complex stress conditions D4-5 under continuous stress D2-3 Fatigue crack corrosion E18 Fatigue endurance limit D6, D13 Fatigue life, rolling bearings F79-82, F83 Fatigue notch factor D12 Fatigue strength D6 Fatigue test, plastics D61 Feed drive components L 16 Feedgear mechanisms L18-20 Feedscrew mechanisms LI6 FEP (plastic) D57 Fibre-reinforced components,
manufacture D63 Field-vector control L5 Filing machines L88-89 Fine machining K81 Finishing, plastic components D67 Finite element analysis, drive mechanisms FI75 Finite-element method (FEM) 85053 applications 852-53 displacements 851 elongations 851 machine tool design L25 oscillatoty systems J22 shears 8S1 stresses 8S2 Fit systems E2S Fits and tolerances E21, E25 Fixed points, temperature scale C3, C34, C3S Flame hardening D31 Flammability, of plastics D62 Flange joints F42 pressure vessel H8, H8-10 Flashpoint, oil D73 Flat-bed frames, machine tool L21 Flexibility matrix method, FEM 8S0 Flexible manufacturing systems L97, K102-103 Flexible turning centres L94 Floating bodies, stability ASO
Flow criteria for material flow K21 incompressible fluids A60 laminar AS2-S3 multi-dimensional inviscid fluids A60-63 viscous fluids A63-68 non-Newtonian fluids AS9 non-stationaty ASI pipe losses HI0 plane potential A62-63 process theory 6 I stationary ASI twisted bar analogy 831 Flow controllers 611 Flow curve, material K21 Flow distributors 6 II Flow of force, see Force transmission Flow law, isotropic materials K21 Flow regulators 611 Flow straighteners, drag coefficient AS7 Flow stress K20 Fluid drives/transmissions 62-3, L40 Fluidics L39-40 Fluids compressible A53 incompressible A60 Fluorine rubber (FKM) DS9 Flywheel design JI-3 FMEA (Failure Mode and Effect Analysis) K104 Foam moulding, thermoplastic (TSG) DS8 Foaming process D64 Fog range C24 Folding (metal bending) K30 Force compensation principle E 15 Force of gravity, work A28 Force transmission E 14 by friction F23-28 diagram LI2, L13 Forces Al combinations of concurrent A2-3 non-concurrent A3-S coplanar systems A2, A4 excitation of out-of-balance J 19 method for statically indeterminate systems 833 resolution of concurrent A3 non-concurrent A4 spatial systems A2, AS Forging processes K32-34, L61, L63, L6S-66 aluminium alloys D4S steels for D38 Form drag, see Pressure resistance Form hardening, austenitic D32 Form milling K47 gear cutting K71 Form numbers D78 Formability K22 Formed parts, design and tolerances D66-67
FOrming cold, steels for D38 hot, steels for D38 powder metallurgy K18 theory K23- 2S FOrming limit curve K23 Forming machines LS4-S7 Fourier analysis non-periodic processes J17 periodic oscillations J17 Fourier index C28 Fourier's Law A7l, C26, C27 Fracture mechanics D2, D3-4, D7 test methods D22 Fracture strain, test methods for determining D 19 Frames machine tool L2, L21, LS7 embodiment design L23-2S materials L22 Framework, pin-jointed AlO-12 Freedom of motion, mechanism FlS8 Freeing principle A 7 Freezing point C2 Frequencies natural JIl-13 response functions J13-14 Friction sliding AlS static AlS systems analysis D7l-72 tube AS3-SS types of D67, D68 Friction corrosion E18 Friction coupling L94 Friction drives calculations FI12-115 definitions F 11 0 types Fl11-112 use and operation FllS-1l6 Friction losses, heat exchanger HS Friction of motion,
see
Friction,
sliding Friction pairings, table F182 Frictional force, work A28 Frictional gears L14 Frictional resistance A6S Froude's Model Law A70 Froude's Similarity Law A 70 Fuels calorific values table C49, C50 composition C50 liquid C50 Full mould casting K9 Function control L34 Functional interrelationships E2, E4
Gallery method (for problem solving) E5 Galvanic coatings D48 Gamma-ray test procedure D25 Gas constant, universal C9 Gas cutting processes F17-18 Gas fusion welding F2 Gas lubrication, bearings L30 Gas phase deposition coatings D48 Gas turbine system
H • Index
Gas turbine system (cont.) closed C20 open C18 Gas-shielded arc welding F2-3 Gases combustion characteristics C50 ideal C9 caloric properties C 10-11 flow CI5-I6 mixtures of C23, C24 physical properties C52 radiation of C34 real C9-JO caloric properties C II Gaskets H9 Gate valves H17 GE hypothesis (deformation) K21 Gear cone transmission units L 11- 12 Gear cutting by hobbing K73-7 4 by planing K72 by shaping K72-73 form-cutting processes K70 generating processes K70 worms K77-78 Gear manufacture fmishing K74-77 grinding K76, L92 materials and heat treatment F126-128 Gear pairs F 116 Gear ratio FI17 Gearbox, three-speed L II Geared transmissions bearings F 156-157 cases/housings F154-156 connections FI54 design FI54 types F153-154 Gears bevel A27 cutting by generating methods K79 by grinding K81 by milling K80-81 by shaping K78-79 straight F 13 7 - 138 bevelled spur F138-139 circarc FI22 compound planetary trains F150-152 cooling F126 crossed helical F139 crown FI38 cycloid FI22 cylindrical K70-77 efficiency F 117 epicyclic F143-152 frictional L14 grinding machine for L92 helical FI28-137 hydraulic L 14-15 involute F120-122 lubrication F125-126 machine tool Ll mangle F124 mechanical L II noise behaviour F116
planet, see Gears, epicyclic pneumatic LIS spiral bevel FI38 spur FI28-137 superposition F149-150 tooth geometry F117-124 Wildhaber-Novikov (W-N) F124 worm FI39-143, K77-78 Gearwheel an-mgement L 12 - 13 Geometrical data processing (GEO) L45 Geometrical features E2 Geometrically similar series E21 German Standards, see DIN Standards; VDE regulations; VDl guidelines Gibbs's fundamental equation C6 Glass 051-52 Gnielinski's equation C31 Gottingen-type profile (polar curve) A68 Graetz solution (infmite series) C31 GRAFCET (programming standard guideline) L42 Grain characteristics, for machining K50-51, K114 Graphite-bearing metals, bearings F98 Grashof number C30, C32 Grashofs criterion (linkages) F158 Grdtings incremental measuring systems L49 incremental scanner L49 Gravitational forces, scale factor A69, A70 Gray code, digital absolute measuring systems L51 Greases, see Lubricants, greases Green's function (influence coefficient) B54 Grey cast iron, weldability F8 Grey radiators C33-34 Grids and screens, drag coefficient A57 Grinding machines L89 centreless cylindrical L90-91 cylindrical L89-90 developments L92 gear L92 internal cylindrical L91 screw thread L92 surface L89 Grinding processes belt K54-55 gear finishing K76 inside-diameter cutoff K57 rotating tool K52-54 thread production K69 Grinding wheels K52 dressing K53-54 Griibler's equation (degrees of freedom) FI58 Guided gyro A38 Guides linear L2S-31 machine tool L 1 rotary, see Bearings
7
Gyroscopic motion <\.38 zero-force A38
H-frames, machine tool L21 Hacksawing K49 Hacksawing machines LB9 Hagen-Poiseuille formula (volume flow) A53 Haigh fatigue strength curve 06, 077 Hamilton's Principle (dynamics) A32-33 Hammer forging K33 Hammers L54, L61-63 Hand moulding K6 Handling systems L1 0 1 computer controlled LJOI Hard soldering FI9 Hardening treatment 029 Hardness, of plastics 060 Hardness test methods 020 Harmonic drive gear mechanisms L19 Harmonic oscillations J14 Hausen's formula C31 Head loss A53 Heat C4 Heat conductivity C26 table of values A52 Heat conductivity resistance, see Thermal resistance Heat exchangers characteristics HI efficiency H 4 finned-surface H22 flow arrangements H4 operating characteristics H4 plate-type H22 recuperative, see Recuperators regenerative, see Regenerators shell-and-tube H21- 22 spiral H22 tube-bundle H21-22 Heat flux C26 Heat penetration coefficient C29 Heat pump process C 19 compression C22-23 Heat transfer C26, C27-28, HI cooling tower H27 with phase change C32-33 without phase change C31-32 Heat transfer coefficient C27-28, HI, H24 Heat transmiSSion, instationary C2830 Heat treatment processes 030-32, K15 effect on material properties 09, 029 Heating index C23 Hedstrom coefficient (fluid flow) A59 Heller process (condensation) H25 Helmholtz equation for vorticity A61 Helmholtz Laws A61 Hencky's Law (stress/strain) B57
8
Index. L
Henneberg's Rod Transposition Process All Henry-Dalton law D74 Herpolbode A25 Hertzian contact stresses B38, DIO,
069 arbitrariIy curved surface-plane contact B38-39 arbitrarily curved surfaces contact B39 cylinder-cylinder contact B38 cylinder-plane contact B38 sphere-plane contact B38 sphere-sphere contact B38 Hertzian fundamental bending frequency A43 Hertzian theory (compression behaviour) F75 Hierarchical vieWpoint E5 High vacuum casting K8 High-polymer materials, state for shaping K3 High-speed steels K49 Hobbing process K47 gear cutting K73 Hole-based fit systems E25 Homogenizing D30 Homokinetic joint F65 Honing K55-56 Honing machines long-stroke L92-93 short-stroke L93-94 Hooke's Laws B36 Hooke's Model Law (for similarity) A70 Hot forming processes )(20 temperature-controlled (heat treatment process) D32 Hot-dip coatings D48, D49 Humid air C24 changes of state C25, C26 table of characteristics C51 HydraIllm process (extrusion) K35 Hydraulic circuits G12, G15-16, L61 Hydraulic fluids/oils G2 characteristics G2, G18 compressibility G 1 Hydraulic gears, see Gears, hydraulic Hydraulic linear motors, see Hydrocylinders Hydraulic systems, water G 17-18 Hydraulic transmissions, see Transmissions, hydraulic Hydrocylinders LIO Hydrodynamic lubrication slides L28-30 Hydrodynamics A51 flow C31 Hydromechanical efficiency, pump G5 Hydromotors L9-10 Hydrostatic lubrication slides L30 Hydrostatic transmissions, see Transmissions, hydraulic Hydrostatics A49-50 Hysteresis loops B56, D6
10
cutoff grinding K57 Ideal mirror body C33 Identification (ID) number, item E30 IEC recommendations (International Electrotechnical Commission) E23 II-theorem, Buckingham's A71-72 Impact A39 eccentric A40 normal A39 oblique A40 rotary A40 Impact coefficient A39 Impact tests D 19 on plastics D60 Impetus A29 In-line machines, force calculations
J4
Inclusions, effect on material properties D8 Incompressible fluids, see Fluids, incompressible Indentation, elastic H7 Induction heating D31 Inductosyn scale L51-52 Industrial robots, see Robots 1nfiltration alloys D41 Influence coefficient, see Green's function Influence lines All Informational work, human response K108 Ingot casting processes K3 Injection moulding K16 plastics D62 Insertion devices, mechanical handling KIO 1 Inspection and testing, casting process K15 Insulating materials, heat conductivity values table C52 Insulation losses, heat exchanger H5 Integral design E 19 Interference fits F24-27 Internal combustion engines C 18 Internal forces Al Interpolation, see Position setpoints, generation Interrelationships, engineering systems E2, E4 Intuitive methods, problem-solving E5 Inviscid fluids A60-63 Ion implantation K86 Iron aluminium D43 cast D41-43, D92 austenitic D43 gray D42 lamellar-graphite D42 malleable D42 pipes and fitrings H 11-12 spheroidal-graphite (SG) D42 white D42 chromium D43
silicon D43 sintered D41 Iron base materials D26 Iron-carbon diagram D26 Irregulatory factor, flywheel inertia
J2
Irreversibility principle C4, C6 Isentropic changes of state C 14 ISO recommendations (International Organisation for Standardisation) E23 ISO Standard DP 9283 (industrial robots) L103 ISO thread metric F35-36 force and torque table F181 tables of sizes F179, F180 Isobaric (constant pressure) changes of state C14 Isochorous (constant volume) changes of state C14 Isothermal transformation (heat treatment process) D32 Item numbering systems E30-31
Jager's load curve (bearing stress) B46 Jet cutting F18, K66 machines L54 Jet flow A58-59, C16 impact force A60 Jig boring machines L76 Job distribution K100 Job production control K98-100 Job production planning K%-98 Job schedule K%-98 management K98, K100 strategy KIOO Joining processes K91-92 see also Bonding; Connections Jointing forces, flange H9 Joule process (reversible cyclic) C21 Journals, stress in B8
Kelvin Scale C2 Kepler's Second Law A30 Kienzle's approach drilling forces K43 machining forces K46 Kinematic pair F157 Kinematic viscosity A53 Kirchoff boundary transverse forces B54 Kirchoffs Law C33-34 Knee-type frames, machine tool L21 Kutta-Joukowski equation (lift force) A63, A66
Labour costs E8 Ladder stabiliry A16 Lagrange datum system (fluid mechanics) C1 Lagrange's equations (dynamics) A32
M. Index
Lambert's cosine Law C33 Laminar flow A52-53 Laminating process K 16 Lancaster drive JI0 Landis process (grinding) L91 Langmuir-Blodgett coating process
K86 Laplace potential equation A62 Lapping machines L94 single-wheel L94-95 spherical L97 triple-wheel L96-97 twin-wheel L95-96 Lapping process K56 Laser beam processing applications K82-83 cutting processes F18, K59-60,
K66 principles K82 suriace hardening D31 Laser chemical vapour deposition (LCVO) K83 Laser cutting tool equipment L54 Laser interferometer L52 Lathes L66 automatic L69 Caxis L71 classification L66 construction L66 copying L67 facing L69 four-axis L71 front-turning L69 heavy-duty L69, L71 multi-spindle L69, L71 numerically controlled L69, L7375 single-spindle L69 sliding and screw-cutting L67 special-purpose L 71- 72 turret L67 two-spindle L69 universal L67 Y-axis L71 Laval's compression ratio C15, C16 LO converter process (steelmaking) D27 Lead and lead alloys D47, D111 as alloying element D33 lead bronze D44 Ledeburite D26 Lever Law, see Moment-ofMomentum Law Lewis number (heat transfer) H28
Lift aerofoils A66 hydrostatic A50 UGA process (build-up technique) K89 limit cooling temperature C26 Limit slenderness B45 Limitation method (sample evaluation) D18 Line motion control L46 Liquid fuels, combustion characteristics C50 Liquids ideal A51
inviscid A60-63 Newtonian A51 non-ideal A51 physical properties table C52 pressure distribution A49 viscous A59 Lists, use of £5-6 Lithography processes K87 Ljungstrom-type regenerator H3 Load conditions fundamental D 1 pressure-thrust D2 redistribution £ 14 Load-time functions Dl Loadbearing capabilities random loading D14 single stage dynamic loading D13-14 static loading D 13 under creep conditionss DIS Logic control L35, L42 Logical functions £2 Long-duration tests D25-26 Loss factors A54-57 Love's stress function B36 Lubricant friction A64 Lubricants additives D74 choice of, table F185 consistency classes D 120 fIlm thickness D69-70. FI26 greases D75 rolling bearings F85-87. F 188 table F187 oils D72-75 table FI86 SAE viscosity categories D 120 selection £ 18 solid D75 Lubrication aerostatic L30 elastohydrodynamic (EHO) D6870 gears FI25-126 hydrodynamic D69, F89 rolling bearings F84-86
MaChinability D24, K38 Machine dynamics J14-16 Machine enclosures, noise reduction J35 Machine equipment, mechanical equivalent system J23, J25 Machine noise generation of J31-35 shielding J31, J35 Machine tools design of L25 frames L21-25 systems function structure LI-2 systems mechanical characteristics L2-4 thermal characteristics L4 Machine vibrations J14-16 forced J16-21 free JI6 paramter-excited J18 self-excited J18
9
Machine-hour rate calculation KI06-107 Machining forces drilling KIlO milling K45-46, KIll table KllO-l11 Machining indices K46 Machining processes chipless K57 with defined tool edges K35-50 numerically controlled centres L83-85 steels for D38 without defined tool edges K5057 Machining tools, materials for K49 Mach's Similarity Parameters A71 Macrostructure, investigation of D24 Magnesium alloys D46, D107, Dll0 Magnetic fracture testing D25 Magnetic moulding KIO Magnetic properties, steel sheets D89 Magnetic sensor pulse transmitters L49 Main technology, manufacturing processes Kl Maintainability, design for £ 19 Malleable cast iron, weldability F8 Manganese (as alloying element) D33 Manufacture-orientated design £ 19 casting K12, K14 Manufacturing Automation Protocol (MAP) LI03 Manufacturing communications systems L 103 Manufacturing costs £8 cost sheet (MCS) KI06 Manufacturing message specification (MMS) LI03 Manufacturing systems KlOO-103 flexible L97, KlOZ-J03 function structure L 1- Z mechanical characteristics L2-4 multi-machine L97 process classification KI Martens thermal dimensional stability D61 Martensite hardening D32 Massive forming L58 Material service life. see Fatigue limit Materials characteristics D78 costs £8, £9 design values D5-8 features £2 flow characteristics K20 for machining tools K49 selection of D53-54 state for primary shaping K3 Materials handling K92-93 automation of KlOl-102 Materials management, manufacturing processes K98-JOO Materials testing fundamentals D 17-18
10
IndexeN
Materials testing (cont.) methods 018-26 Mathieu's differential equation A49 Maurer diagram (cast iron structure) 041 Maximum distortion energy theory K21
Maxwell medium, see Elastoviscous substances Maxwell's law A45 Measurement direct/indirect L48-49 as part of inspection K93 Mechanical eqUivalent systems 111, 121-22 examples 123-5 parameter definition 123 structure definition 122 Mechanical moulding K6 Mechanical work C3 Mechanisms crank-and-rocker FI66-167 dynamic properties FI66 kinematic analysis FI61-163, FI65-166 kineostatic analysis FI64 running qualiry F 166 synthesis FI66-168 systematics F157-161 see also Crank mechanisms Melting curve C13 Membrane stress theory B41, B42 Membranes, vibrationnuss A47 Merkel number R28 Meshed gears L11 Metal cutting kinematics K44-45 Metal etching K86 Metal forming processes classification K19-20 fundamentals K20-23 Metallic materials, state for shaping K3 MetaUographic investigation methods 023-24 635 method (for problem-solving) E5 Methods time measurement (MTM) K97 MF (plastic) 058 Michaelson interferometer L52 Microminiaturisation KB2 Microstructures investigation of 024 manufacture of K88 Microtechnology (in manufacturing) KBI-82, K88 MiUing machines L 71, L 79 bed-rype LBO-81 circular operation LB2 copy system LB2 horizontal LBI-82 horizontal plane LBO knee-type L79-80 special-purpose LB2-83 uttiversal LB2 MiUing processes K43-47 guide values for ferrous materials K112-113
vibrations K46 Modal analysis 113 Modular network, gearing design L12 Modular parts lists E30 Modular systems E22-23 Modulus of elasticity B4, 05 plastics 060 rypical values 077 Mohr's circle B3, B4 Mohr's failure criterion B6 Mohr's method, beam deflection B22, B25 Mol (unit symbol) C9 Molecular heat, gases C37 Molecular weight C9 MoUier curves C 12 humid air C24-25 Molybdenum (as aUoying element) 033 Moment of area B12 Moment of displacement Al Moment of impulse, see Angular momentum equation Moment of inertia mass A36 rotated axes A34-35 Moment of linear momentum, see Angular momentum Moment-of-momentum Law A34, A36, A37 Moments Al Momentum equation AZ8, A31 incompressible fluids A60 Momentum Law A31 Motion of the centroid A30 compound AZ3 in cylindrical helix A22 equations of A37 non-uniformally accelerated AZO particle A19-22 plane AZO-22, AZ4 circular AZ2 in polar coordinates AZI relative AZ5-27 rigid body A22-27 rolling A36 rotary A34 spatial AZ2, A23, A37-38 spherical AZ4 uniform A19 uniformally accelerated A 19 Motion control machine tool LI robot LI03 Motor control L5-8 Motor-generator welding set L98 Motors asynchronous L4, L7 d.c. L7-9 hydraulic 68-9 machine tool L4-1O pneumatic 616 slipring L6 squirrel-cage L 5 stepper L9 e1ectrohydraulic L 10
synchronous L6 Moulding processes K4-6 ceramic K7-8 hand K6 high vacuum KB magnetic KIO mechanical K6 plastics 063 shell K7 suction K7 vacuum KB Moulds, casting K3 Multi-component mixtures, condensation R24 Multi-degree-of-freedom systems, vibration A44-47 Multi-machine systems L97 Multi-point machining L69
NatUral frequency A45-46 Navier-Stokes equations A63 Navier's equations (motion) B36 Needle bearings L16 Neutralisation capacity, lubricating oils 073 Neutralisation number (NZ) 073 Newton-Bertrand similarity law A70 Newton-Euler equations, robot movement LI02 Newtonian fluids/lubricants 073 Newton's law of shearing stress A49 Newton's laws of motion A28-29, A30 Newton's similarity law A70 Nibbling machines L53-54 Nibbling process K62 Nickel, weldability F8 Nickel and nickel alloys 047 as alloying element 033 Nikuradse'~ formula A54-55 Niobium (as aUoying element) 033 Nitriding 031-32 steels for 035, 080 Nitro-carburisation 031 No-load voltage, welding L98 Node intersection procedure A12 Nodular graphite cast iron, weldability F8 Noise 129 airborne 131-33 control regulations L66 damping 131 machine 131-36 shielding 131 spectra 131 structure-borne 130-31 Nominal pressure (PN), piping RIO Nominal size (ON), piping RIO Non-destructive testing 024-25 Non-ferrous metals, physical characteristics table 0115 see also under specific names Normalising 030 Norton process (grinding) LB9-90 Notch concept (fatigue life) 06 Notch effect, constructional design Oll,012
Pelndex
Notch impact bending test D7, D21-22 plastics D60 Notch impact ductility D21 Notch sensitivity factor D12 Notch stress conditions D 11 Notching process K62 Nozzle, drag coefficient A56 NR (natural rubber) D59 Nuclear energy plants, steels for D40 Number of transfer units (NTIJ) H28 Numerical control (NC) basic functions L34, L45-46 data processing L45-46 definition L42-43 of lathes, see Lathes, numericallycontrolled workpiece programming L43 Nusselt number C30, C31, C32 Nusselt's similarity law A71 Nuts, see Bolts and nuts
O-frames, press L57 O-rings H19 Objectives, task E4 Octahedral stress B3 Offenhach system (lathe) L69 Offsets, forming machines L55 Oils hydraulic, see Hydraulic fluids/oils lubricating, see Lubricants, oil Oldham coupling F65 Omega method B46 One-degree-of-freedom systems, vibration A40-44 Open arc welding F2 Open channel flow A58 Operability, design for E19 Operation control, robot L103-104 Operational costing K105-107 Optical diffraction systems L50 Orbital grinders, see Honing machines, short-stroke Order calculation K99 Organisational plans E5 Oscillating extension test D26 Oscillation of machine tools L3 pendulum A41 rotary A41 Oscillator A32 Oscillatory signals J16 Oscillatory system block diagram
J11, J14 Oseen formula (fluid resistance) A64 Oxidisation (as coating) D48-49 Oxygen cutting F 17
Packing rings H20 Packing (stuffing box seals) H 19-20 Pain threshold, noise J30 Painting D49 Pallet storage system 1.85 Parallel key assembly F30-31
Parallelogram of forces A5 Paris equation (crack propagation) D4 Particle dynamics AI9-22, A28-30 on helical curve A29-30 on inclined plane A29, A30 suspension velocity A65 Parts lists E27, E28 Pascal's law A49 Patenting (heat treatment process) D32 Patterns, use of in moulding K4, K6 Pearlite D26 Peclet number C30 Peclet's similarity law A71 Pendulum oscillation A31-32, A41 Permanent moulds KIO-12 Petukhov's equation, modified (heat transfer) C31 PFA (plastic) D57 Phase-frequency response J14 Phenolic resins (PF) D57 Phosphating D49 Photoelectric pulse transmitters L49 Photolithography K87 Physical analysis methods D23 Physical effects E2 Physical operative principle E2 Physical phenomena investigation of E5 table of properties C52-53 Physical vapour deposition (PVD) D48, K49, K86, K90 Pin-jOinted frames planar AI0 spatial A12 Pinned joints F28 Pipe flow calculations drag coefficient A56 inviscid liquids A60-63 losses HIO Newtonian liquids A52-59 non-Newtonian liquids A59 Pipes/piping diameter HIO expansion compensators H13 fittings Hl1-13 heat conduction through wall C27 heat transfer C31-32 materials H 11 standards H 10-11 steels for D40 supports H14 Piston machinery, excitation forces
J19 Pistons FI73-174, F175, F176 movements F169-170 Pitot tube A52 Plain face milling K44-47 Planck's radiation Law C3 Plane stresses B37 Plane surface structures, production of K87 Planing machines 1.85-86 Planing process K47 Planning systems, computer-based LI07-108
11
Plastic hinge B56 Plasticity theory B55-58, K23-25 Plastics D54 for bearings F98 chemical characteristics D61 elastomers D58- 59 electrical characteristics 061 forming D62, D63-66, K3 calendering K15 characteristics D61-62 compression moulding K16 design D66-67 expanding K16-17 extrusion K15 injection moulding D62, D63, K16 laminating K16 sheet casting K15 transfer moulding K16 mechanical characteristics D6061, D116-120 physical characteristics D115118 structure and properties D54-55 testing D59-62 thermal characteristics 062 types cellular D58-59 fluorinated D57 foamed D58-59 thermoplastic D55-57 thermosetting D57-58 Plastics parts finishing D67 testing D62 Plate elements, use of FEM B53 Plate shears L52 Plates B39 buckling B48-49 circular B40 elliptical B40 rectangular B39 thermal stresses B41 thickness/die clearance table K117 triangular B40 under in-plane loads B41 Plating techniques D48 Plywood D53 Pneumatic drives/transmissions G2, G16-17 graphical symbols G3, G19 Pohlhausen formula (laminar flow) C31 Poinsot's law (gyro setting) A38 Point-ta-point control L46 robot LI03 Poiseuille parabola C31 Poisson's ratio B4, D5 Polar coordinates, plane motion A21 Polar curve A68 Polhode A25 Polyacetal resins (POMS) D55 Polyacrylate (PMMA) D56 Polyalkylene terephthalates D55 Polyamide (PA) D55 Po1ycarbonate (PC) D55
12
Index. R
Polychloroprene rubbers (CR) D59 Polycrystalline diamond (PCD), cutting material K50 Polyester resins, unsaturated (UP) D57 Polyesters, linear (PET PBT) D55 Polyethylene (PE) D56 Polygon of velocities, see Centre of rotation, instantaneous Polygon-type connections F32 Polyimide (PI) D56 Polymer concrete, uses of L23 Polymers D54 Polyolefins D56 Polyphenyl ether, modified (PPE) D55 Polyphenyl sulphide (PPS) D56 Polypropylene (PP) D57 Polystyrene (PS) D56 expanded (EPS) D58 Polysulphone (PSU/PES) D56 POlytetrafluoroethylene (PTFE) D57 as lubricant D76 Polytropic changes of state C14 Polyvinyl chloride (PVC) D57 plasticised (PVC-P) D57 unplasticised (PVC-U) D57 Position control L48-49 fluid transmissions G3 measuring systems for NC machines L48-52 setpoints generation L46-47 Potential flows A61-62 Potential vortex, see Vortex line flow Pour point, lubricating oils D73 Powder metallurgy D28-29, K17-19 coating K91 for cutting steels K49 Power A28 Power shift gears L11 Power stations, condensers for H26 Power transfer/transmissions, fluid drives G1, G2-3 Prandtl number C30, C32, C52 Prandtl-Eyring formula A59 Prandtl-Reuss's law (stress/strain) B57 Prandtl's Similarity Law A71 Prandtl's soap film analogy B31 Precession cone A38 Precision engineering KSl-82 casting K8-9 forming machines L56 Preferred numbers, see Series, preferred numbers Presses L54, L55-57 flywheel spindle L64-66 hydraulic L60-62 mechanical L55-60 pneumatic L60-62 safety regulations L65-66 screw, see Presses, flywheel spindle wedge L59 Pressing plastics D63 steels D38 Pressure A49
in container A49 cutoff control G 14 loss of A53 resistance A65 Pressure die casting K11 Pressure welding F4-5 Prestressed shaft-hub connections F32 Primary functions E2 Primary shaping K2- 3 tools for K3 Prime cost E8 Printing processes KS7 Problem-solving E4 solution evaluation E6-1O solution principles E5-6 Process control L42 Process quality index H4-5 Production control K98-100 Production planning K96-98 computer-aided K98 estimation of costs E8 sequence K100 Production structure E4 Production supervision K100 Production technological tests D24 Profiling machines L83 Program simulations L105-106 Programmable logic controller (PLC) L41-42 programming L42 Programming procedures L 104-1 06 control systems L34, L36 direct L104 hybrid L105 indirect L 105 off-line Ll05-106 Projection, drawing standard E30 Protective systems E 16 Pulleys resistance at A18 rope friction A17 Pulsation test, longtime D25 Pulse transmitters, photoelectric L50 Pumps centrifugal J24-25,J28-29 condensate H27 cooling water H27 gear-type G6 hydraulic G4-5 lift G5 piston G7 port-plate G8 power rating G5 rotary G5 rotary piston G6 screw G6 swashplate G8 tilting-head G8 vane-type G6 variable displacement G7, G14 Punching K62-63, K65 steels for D38 Punching machines, automatic L53 PUR, see Elastomers, polyurethane PVDF (plastic) D57 QUality assurance K103 Quality engineering K103-105
Quality systems K103 Quantity lists E28
Rack-and-Pinion drive L18 Radiation coefficient (heat) C33 Radiation coefficient (noise) J3031, J32 Radiation exchange number C34 Radiation process C26, C33-34 effect on material properties D 10 of gas C34 Radius of inertia A36, B13 Random loading, strength calculations D14 Rapid-traverse gears L15 Ratio, choice of, see Series, preferred numbers Ratio traction drives, see friction drives Rationalisation, manufacturing processes K 1 Rayleigh number (heat transfer) C32 Rayleigh quotient (buckling) B46 Rayleigh quotient (vibrations) A46 Reactance circuit LIS Reaction equations, combustion C 17 Reaction foams (RSG) D58-59 Reciprocating machines multi-cylinder force calculations J4 moment of inertia J4 torque fluctuations Jl-3 Rectangular thread Al 7 Recuperators, thermodynamic design Hl-3 Recycling, design for E 19 Redesign process E 13 Reduced mass A36 Redundancy principle E16 Reference surface/profile E24 Refrigeration process C 19 compression C22 Regenerators, thermodynamic design H3-4 Regression analysis, cost estimation using E9 Reinforcing overall effect E 14 Relative costs E9 Relative humidity C24 Relative motion A39 Relaxation, creep B56 Relaxation test D25 Relay, see Contactor Reliability prinCiples E15-16 Remelt coating, laser beam K83 Requirements list, design E 10-11 Resistance rolling A18 tractive A 18 Resistance fusion welding F4 Resistance pressure Welding F4 Resistance welding machines, see Welding machines, resistance Resonance A42-43, E17 design for avoidance E 17 engine compensation J9-1O
Seindex
Response functions, frequency J1314 Reverse side machining L71 Reversible changes of state C4 Reversible processes C6 Reynolds numbers A53, A63, C30 Reynolds similarity law A70, A71 Rheopexy D73 liquids A59 Rheostatic control, motor L 10 Ribbing, rigidity requirement L23 Rigid body dynamics A33-38 acceleration A37 general plane motion A36 main axes A..~4 moment of inertia A34 plane motion A36-37 rotation about a fIXed axis A33 straight-line motion A28-30 systems AIO Rings, rotating B43 Ritter's Method of Sections All Ritz fortnula (vibrations) A46 Ritz fortnulation (buckling) B49 Riveted joints F33-34 Robots accuracy characteristics L 102 applications and selection K102103, Ll07-IOB components LIOI-I02 control systems L102- 104 dynamic model LI02 kinematic model LI02 programming LI04-106 Rocket drive equation A33 Rockwell hardness test D21 Rod transposition process All Roller bearings, see Bearings, roller Roller milling machines L82 Roller slideways L30-31 Rolling motion inclined plane A36 resistance A 18 Rope friction Al7 -18 Rotary slotting machines L82 Rotating components centrifugal stresses B43-44 slide crank A27 Rotating tube, motion in A26, A39 Rotation law of motion A34 rigid-body A22 about a point A24 plane representation A23 vectorial representation A22 Rotors balancing JI5 motor L9 shaft mechanical equivalent system J23, J28 Rotscher cones F38 Roughness, tube A54 Roughness values, surface E24-25 Roulette A25 Rubber D58-59 bearings F98 properties F59, F61 springs, see Springs, rubber
S
AE viscosity categories D 120 Safe-life principle EI6 Safety factors strength calculations D 16 thertnal apparatus H5-6 Safety principles E15-16 buckling B45 design for EI8 Safety regulations/standards E24 automation system L37 health KI08 laser cutting K60 press working L65-66 Safety stress, see Stress, permissible Safety valve HI6 Saint-Yenant's theory (warping) B30 Sampling methods, cast metals DI7 Saturation, vapour C 10 Sawing machines L87-89 Sawing processes K48-49 Scale factors A69 Screen printing K87 Screw motion F35 Screw thread grinding L92 Screwed connections, piping H12 Screws AI7 Seals HI8 bearing F98 dynamic contact H19-20 rolling bearings FS8-S9 rotating mechanical H20 static contact H18-19 Seamiess tube steels D36-3 7 Second moment of area, beam bending B13-14 Secondary functions E2 Section tables, pipes and rolled sections B59-75 Sectionalised machining L69 Seiliger process J I Selection procedure, solution E6 Self-help principle, system E14 Self-locking epicyclic gears F147-149 screws AI7 Semi-finished products, manufacturing K3 Semi-infinite bodies, heat transfer C28-29 Semi-similar series E21 Sensor data processing L 103 Sequence control systems L35 cascade representation L42 robot Ll03 Series, preferred numbers E20-21, E24 Service life experimental determination D 15 prediction D 15 rolling bearings F83 Servodrives L5 Servohydraulics L40 Servomotor L5 Severing processes classification K61-62 forces and energies K63-64 technology K62-63
13
tools K64 workpiece properties K64 Shaft-based fit systems E25 Shafts critical speed A43-44, A45 torsional buckling B47 Shaping machines L85-86 Shaping process K47 primary K2-3 Shaving (gear finishing) K74-75 Shear centre B 18 Shear stability, lubricating oils D73 Shear stress B I defortnation B25-26 distribution B 16-19 in fasteners BI8 forces BS, BI6 transverse B7 Shearing machines L52-53 Sheet cutting, laser beam K83 Sheet fonning metal L59, L61 superplastic K31 Shell elements, use of FEM B52, B53 Shell moulding K7 Shells B41-42 bending rigid B42 buckling B49-50 under internal pressure B42 Sherardisation D32 Short-circuits, shaft-loading by J21 Siemens-Martin process (steelmaking) D27 Silencers J34 Silicon (as alloying element) D33 Silicon technology K88-89 anisotropic etching K88 Silicone rubbers (YOM) D59 Siliconising D32 Similarity mechanics A69-72, E20 cost estimation using E9- 10 thermal A71 Sintered materials D41, D91 aluminium alloys D46 metal bearings F98 Sintering process K3, K17, K18-19 fluidised bed K91 635 method (for problem-solving) E5 Size effect, constructional design Dll, DI2 Skin packing method D65 Skull (extrusion product) K34 Slider-crank mechanisms Ll4 press L57, L58 Slideways L26-31 Sliding pairs, oil-lubricated D69 Sliding-gear transmission units Lli Slip couplings L20 Slitting K62 Slotting/grooving machines L82, L85-S6 automatic L53 Smelting process, steelmaking D2628 Smith creep-strength diagram D6, D9, D76
14
Index. S
Soldering FI8-20 equipment for LIOO soft FI9 Solid body ftlling, flow through A57 Solid fuels, combustion characteristics C50 Solid lubricants, see Lubricants, solid Solids caloric properties C12-13 heat conductivity values table C52 physical properties table C53 Solution heat treatment D30 Solutions, evaluation procedures E6-1O Sommerfeld coefficient (hydrodynarrricioad capacity) F90 Sound,.velocity of CI6 Sound errrission analysis D25 Sound-deadening measures J34-36 Soundproof boxes J36 Space frames AI2 Spacecraft, steels for D40 Spark ignition engine CI8 Special-purpose machines, mass production L97 Specific heat capacity CII of air C36 measuring C 13 Spectral analysis method D23 Speed flow diagram, gearing design
Ll2 Spheres, thick-wall B43 Spindle bearings L16-18 Splined joints F31-32 Spot Welding units L99 Spray coatings D48 Spring characteristic, linear A40 Spring force, work A28 Spring rate A41 Spring steel, see Steel, spring Spring-back, elastic K30 Spring-mass system A40-41 Springs F50-51 fibre composite F62-63 gas (air) F63-64 helical F57 -59 leaf F53-54 metal F51-59 ring F53 rubber F59, F61-62 spiral F54-55 Sputtering K86 Square-threaded screw AI7 Stability A7, EI6-17 design for E 16 floating bodies A50 Stainless steel, see Steel, stainless Staircase method (sample evaluation) D 18 Standardisation process E23-24 Standards basic E24-25 European E23 German, see DIN Standards industrial E24 international E23
sources of E23-24 use of E24 Stanton number C30 State variables tables ammonia C43 carbon dioxide C34 difluorodicWoromethane C46 difluoromonochloromethane c47 monofluorotricWoromethane C45 water and steam C40-42 Static loading, strength calculations DI3 Static rigidity, machine tool L3 Static similarity A69-70 Statically determinate systems AIO Statically indeterminate systems AIO, B32-33 annular beams B35 annular frames B35 closed rectangular frame B34 constrained frame B34 continuous beams B33-34 double jointed frame B34 frame with serrri-circular curves B35-36 projecting frame B34 Statistical process control (SPC) KI04 Steady state processes C4-5 Steam superheated H24 heat transfer H24 Steam power plant C21 use of condensers H24 Steam table C38-39, C40-42 Steam trap HI6 Steel alloying elements in 033 boiler-plate D36 case-hardening D80 cast D28, D40-41, D90 heat-resisting D40-41, D90 wear-resisting D41 cryogenic, see Steel, lowtemperature designation/classification D32-33 forged D33 peariitic D34 free-cutting D35-36 heat-resisting D37, D81, D82, D83 high-alloy D33 high-temperature D36-37 low-alloy D33 low-temperature D38, D81 magnetic D90 maraging D35 nitriding D35, D80 non-ageing D34 pressurised-hydrogen-resisting D37 rolled D33 sintered D41 spring D39, D87-89 stainless D36, D81 structural D34, D79 special D34 weatherproof D34
weldable D34 tool D38, D39, D84 Steel pipes/tubes HIO connections Hll-12 Steel-Iron Test Sheet 1570-1 D8 Steelmaking processes D26-29 effect on material properties D8 Stefan-Boltzmann Law (radiation) C33 Steiner's Principles (parallel axes) A35-36, B13 Stephan & Preusser formula (evaporation pressure) C33 Stephan formula (Iarrrinar flow) C31 Stepless drives L13, LI4 Stepper motors, see Motors, stepper Stereo lithography KB3 Stock management KlOO Stock-keeping policy K98-99 Stokes resistance formula A64 Stoneware D50 Strain B3-4 Strain energy B4, B26 Strain-time functions DI Stream ftJament A 51 Stream tube A51 Streamline A51 Strength properties annealing steels D78 dynarrric DI2 evaluation of D17-18 static Dll typical values D79 Strength theories D4 Stress BI alternating conditions D3-4 axisymmetric B36-37 bending, see Bending stress combined, see Combined stresses contact B8 continuous conditions D2-3 distribution of B5 internal conditions D2 one-dimensional Bl perrrrissible B4- 5 plane B37 residual B 56 static conditions D5 three-dimensional (spatial) B2 two-dimensional (plane) B 1-2 see also Bar stress Stress crack corrosion E18 Stress dynarrric conditions D5-6 Stress failure criterion maximum principle B6 maximum shear strain energy B6 maximum shear (Tresca) B6 Mohr's B6 Stress relief, heat treatment D30 Stress-orientated design, casting K13, K14 Stress-tupture test D25 Stress-strain diagrams, closed, see Hysteresis loops Stress-strain laws, plastic theory B58-59 Stress-time functions DI Stretch forrrring K28
Telndex
Stribeck curve (friction state) 069, F89 Strings, vibrations A47 Strip model, plasticity theory K24 Strip shears L52 Strip and wire rod casting process K4 Stripping strength, thread F 44 Structural parts lists E28, E30 Strutt's map (vibrations) A49 Styrene acrylonitrile copolymer (SAN) 056 Styrene butadiene rubbers (SBR) 059 Sryrene butadiene (SB) 056 Subfunctions E2 Sublimation curve C 13 Subsystems E 1 Suction moulding K7 Sulphur (as alloying element) 033 Superfinishing machine, see Honing machines, short-stroke Superhoning machines, see Honing machines, short-stroke Superplastic forming K31 Support, types of A 7 Support reactions planar A7-9 spatial A9 Surface coating processes K90-91 Surface conditions, effect on material properties 09-10 Surface equation A30 Surface pressure B38-39, 010, 069 table FlBO Surface roughness, see Roughness values, surface Surface tension A49 Surfaces, engineeting, see Engineering surfaces Swiss system (lathe) L69 Swivel mechanisms, press L59 Synchro-resolvers L51, L101-102 Synchronisation faults, shaft-loading by J21 Synthesis process (for problemsolving) E4- 5 System programs L36, L42 Systematic approach E4-1O Systems composition E1 coupling El elements El state of Cl
Tantalum (as alloying element) 033 Taper-pinned joints F28 Tapping (spiral drilling) K68 Tasks, division of E 14 Taylor function (tool life) K41 Technical work C4 Techn
equalisation C29-30 mixing Temperature profile C29-30 Temperature scales C2-3 international practical C3, C34 Tempering treatment 030 steels for 035 Ten-point height E25 Tensile test method 018-19 hot 019 for plastics 060 Tension bars (spring) F52 Testing materials, see Materials testing as part of inspection K93 of plastic parts, see Plastic parts, testing quality system KI04-105 Tetmajer's method (inelastic buckling) B46 Thermal apparatus, design of HI-4, H5 Thermal deburring K61 Thermal efficiency, engines C18-19 Thermal inverse mixing H5 Thermal machining K57 Thermal power plants C20-22 Thermal resistance C26, C27 Thermal state variables C9-1O Thermoana!ysis 024 Thermochemical treatments 031 Thermodynamic temperature scale, see Kelvin Scale Thermodynamics First Law of C3-5 processes C 1, C2 Second Law of c6-7 Thermomechanical treatments 032, K57, K61 Thermometer C2 fixed points C3 Thermoplastic foams (TSG) 058, 064 Thermoplastics 054, 055-57 aggregation state K3 hot fonning 064-65 Thermosets, see Thermosetting plastics Thermosetting plastics 054, 05758 aggregation state K3 Thermosymmetrical construction, machine tool L4 Thixotropy 073 liquids A59 Thomas process (steelmaking) 027 Thread locking devices F46-47 Thread production chasing K67 die cutting K67 EDM K69 forming K70 grinding K69 milling K68 pressing K70 rolling K69 tapping K68 turning K67
Cn
15
whirling K68 Thread run-out F45 Thread-cutting milling machines L82 Threads fatigue loading F45 types F35-36 Three-jaw chuck L67 1bree-jointed arch AIO Three-pole theorem (kinematics) FI60 Threshold strength 06 Throttle devices, drag coefficient A56 Throttling, flow C5 Thyristor drives L14 Tightening procedures (bolted connections) F38-40 Timber, see Wood Time characteristics, forming machines L56 Time management control L34 job schedule K97 manufacturing processes Kl00 Tin and tin alloys 047, 0111 tin bronze 044,0101-102 tin-lead-bronze 0103 Titanium alloys 033, 046-47, 0110 Tolerances, see Fits and tolerances TOOl handling automated K102 carrier L1 change time L84 magazines/changers L83-85 systems LI0l Tool life K41 Tool steel, see Steel, tool Tooth accuracy, gears F124-125 Tooth profiles, gears F118-119 Torque, work A28 Torque fluctuations, reciprocating machines JI-3 Torsion arbitrary cross section bars B30 circular bars B27-28 thin-walled rubes B28, B30 with warping constraints B31 Torsion bar springs F56-57 Torsion limit 05 Torsional oscillations, examples
J25-29 Total base number (TBN), lubricating oils 073 Tough materials, stress distribution B5 TPE, see Elastomers, thermoplastically processable Tracer control L38-39 Tractive resistance A18 Trajectory, particle A19, A20-21 Transfer functions F161 Transfer law A69 Transfer lines automated KI02 mass production L96 Transfer moulding K16 Transformation operations E 1
16IndexeW
Translation, rigid-body A22 Transmission angle, kinematics Fl66 Transmission components, machine tool Ll Transmission electron microscopy D24 Transmission ratio, gears F117 Transmission units LI1-15 gearing L12 Transmissions geared, see Geared transmissions hydraulic GI-2, G4-9, L30-40 graphical symbols G3, G 19 operation G12-l3 pneumatic, see Pneumatic transmissions TranSmitted-light optical diffraction systems L50 Trapezoidal threads AI7 Tresca failure criterion B6 Tresca flow conditions B56 Tresca's shear stress hypothesis K21 Triangular threads AI7 Tribology D67, EI8 system characteristics D71- 72 Trinuning K62 Triple point CIO, Cl3 TSG (plastic) D58 Tube plates H7 Tubes bends B42 use of FEM B53 friction in A53-55 model plasticiry theory K25 steels for D40 under internal pressure H6 Tungsten (as alloying element) D33 Turbo-generator, mechanical equivalent system 123-24, 126-28 Turbosystem, vibrations 126 Turbulent rum flow, bearings F92 Turbulent flow A52-53 Turning centre L 71 flexible L71, L77 Turning mills L71 Turning process K38-41 Turret drilling machines L 76 Turrets (tool carriers) L67, L68 TOY (Technical Monitoring Associations) regulations E18, E24
U-bending (metal fomting) K30 UF (plastic) D58 Ultrasonic processing applications K84-85 principles K84 testing D25 Universal joint F65 Unsteady state processes C5 UP (plastic) D58 Upsetting process K32 cylindrical parts K25 square parts K25 User programs L36, L42 generation of LI04-106
User safery, see Safery regulations/standards
V-belt drives, see Belt drives, VV-bending (metal fomting) K30 V-scanner, digital absolute measuring systems L51 Vacuum fomting, thermoplastics D64-65 Vacuum metallising K86 Vacuum moulding K8 Vacuum-pouring technique D28 Value analysis E7, EIO Valves control H14-18 directional control G9, G 10 drag coefficient A56 flap HI8 flow control GIl gate H17 hydrauliC G9-11 materials for D39 pneumatic GI6-17 pressure control G 10-11 proportional Gil rotary gate H 17 seat G9 shutoff, see Valves, control shuttle GIO slide GIO throttle GIl types H15-17 Vanadium (as alloying element) D33 Vapour depOSition, electron beam K84 Vapour pressure curve CIO, C12, C13 Vapour Tables CI2 Vapour-jet air ejectors H26-27 Vapour-liquid critical properties C35 Vapours CIO caloric properties C II mixture with gases C24 Variable mass systems A33 Variant design El3 value concept E8 VDE (Association of German Electrical Engineers) regulations E24 VDI (Association of German Engineers) guidelines E24 Velociry of sound C 16 Venturi tube A52 Vertical boring mills L71 Vessels, flow from A57-58 Vibration isolation, see Anti-vibration mountings Vibration model, see Mechanical equivalent system Vibration reduction F66-67 Vibrations extensional A46 flexural A45-46,128-29 forced 126,127,129 forced damped A43, A45 forced undamped A42, A44
free damped A42, A44 free undamped A40-42, A44 machine 110-11 fundamentals 111-14 problems 114-16 membrane A47 milling process K46 non-linear A48 parametrically excited A48 plates A47 representation in frequencydomain 117 representation in time-domain 116 string A47 torsional A46 Vickers hardness test D21 Vincent friction screw press L65 Virial coefficients C9 Virtual work principle A6, A29, B26-27 Viscoelasticiry theory B56 Viscosiry hydraulic fluids/oils G2, G18 lubricating oils D73-74, D119, D120 Viscosiry index D73 Viscous flow A52-59 Volume strain B4 Volumetric work C3-4 VOM, see Silicone rubbers von Mises's failure criterion B6 von Mises's flow conditions B57 Vortex lines A61, A62 Vortex tubes A61 Vulcanisation D59
Wall, conduction through C2627 Ward-Leonard speed control system LB, LI4 Warping, twisted bar B30 Washers F37 Waste heat utilization HI Water content, humid air C24 Water hydraulic systems G17-18 Water torches L54 Water-jet air pumps H26 Wave soldering LlOO WCfFE (plastic) D57 Wear process D5, D68, D70 design to limit E18 during milling K47 resistance tests D24 systems analysis D71- 72 rypes of K37 Wear safery, plain bearings F90 Weathering tests, of plastics D62 Weber's similariry law A 71 Wedge friction A16-17 Wehnelt cylinder K83 Weld joints F9-11 safe loads Fl3-16 strength calculations Fll-13 Weld positioning F6 Weldabiliry F5-6 Welded connections, piping HI2 Welding current L98
Z • Index
Welding current (cont.) electronic sources L98 Welding duty cycle L98 Welding guns Ll00 Welding machines arc L97-98 butt L99 projection L99 resistance L99-100 seam L99 spot L98, L99 three-phase LI 00 Welding processes F 1-S graphical symbolS Fll, F176 laser beam K83 plastic materials D6S reliability F6-8 rules and standards table F 177178 Welding rectifiers L98 Welding transformers L98 Whitworth thread F36 Wildhaber-Novikov gears F124 Wind pressures, on buildingS A6S Wire drawing, plasticity theory K25-26
Wohler curves (fatigue endurance) DIS, D2S Wood DS2 permitted loads D 113 protection of D52 strength characteristics D 113 Wood-based materials D53 Woodruff keys F30-31 Work A27, <:3 displacement <:4 of dissipation <:4 force of gravity A28 frictional force A28 spring force A28 technical <:4 torque A28 total A28 Work factor K97 Work Law A30, A34 Work-hardening K20 Working environment K108 Working space, machine tool Ll Workpiece embodiment design E19 Workpiece handling automated KI02 carrier Ll
17
Workplace design K107-108 Worm gear cutting K77-78 Worm-rack gearing LI8
X-ray fluorescence analysis method D23 X-ray lithography K87 X-ray test procedures D2S
V
ield strength, see Flow stress Yield stress, test methods for determining D20
Zero-flow control G 14 Zero-order transfer function F161 Zeroth main law of thermodynamiCS <:2 Zinc and zinc alloys D47, DIll Zweifel's cascade results A68-69