mm HI
MBBfflffl
.m
WlfflfflHI
HSU
MM
m
I
hHHHHBI Hi HH bHrKBH HGH
fflBflfflmB
j
1
«
Bra HH
RHMMMlH
I5H
•M,*
Hi HH 1 1HH HH H
wHBm
Mi
1
BH
H
MOW
1
§1
H
H
bmmmbm
HB Hkbh
I
A'
^
v.
>
^
X*
A
7*.

i
o
0'
%<
>
u
^
N
c^
\^.
^
\
/
^
frV


V
X*
^
^, v
'OO
x
'
pq.
8
1
^.
c
A


.0
V
\
>
o
ELEMENTS
OF
MACHINE DESIGN
BY
DEXTER
Professor
S.
KIMBALL,
Sibley
A.B.
Cornell
University,
of
Machine
Design
and
Construction,
College,
Formerly
Works Manager, Stanley Electric Manufacturing Company. Member of the American Society of Mechanical Engineers.
AND
JOHN
Cornell University.
H. BARR, M.S., M.M.E.
Formerly Professor of Machine Design, Sibley College,
of the American Society of Mechanical Engineers.
Manager, Smith Premier Works.
Member
FIRST EDITION FIRST THOUSAND
NEW YORK
JOHN WILEY & SONS
London:
CHAPMAN
& HALL,
Limited
1909
Copyright 1909
BY
DEXTER
S
KIMBALL
and
JOHN
H.
BARR
n. K
2465*64
17
SEP
1909
Electrotyped and Printed by Publishers Printine Co..
i\M
w
New York,
U.
S.
A.
PREFACE
This book
College,
is
the outgrowth of the experience of the authors
in teaching Machine Design to engineering students in Sibley
Cornell
University.
It
presupposes a knowledge of
Mechanism and Mechanics of Engineering. While the former subject is a logical part of Machine Design, it may be, and usually
is,
for convenience, treated separately
and
in
advance of that
portion of the subject which treats of the proportioning of machine
parts so that they will withstand the loads applied.
logical order is usually followed in
The same
as
it is,
actual designing,
ordinarily, necessary
before proportioning the various
and convenient to outline the mechanism members.
of designing a
With the mechanism determined, the remainder of the work machine consists of two distinct parts: (a) Consideration of the energy changes in the machine, and
the
maximum
(b)
forces resulting therefrom.
Proportioning the various parts to withstand these forces.
logical procedure,
first
This
in
and the fundamental principles underseldom made clear
to
lying the
part
(a),
are
the student,
is
works of
this character;
and such information
is,
as
given on
energy transformation in machines
to special cases or types.
in general, that relating
A
thorough understanding of these
general principles
ful
is,
however, in most cases, essential to success
design,
since
a consideration of the machine as a whole
consideration
of
details.
necessarily
precedes
A
is
very
brief
discussion of typical energy and force problems
fore, in
given, there
Chapter
II, in the
hope of making
this
important matter
somewhat
clearer to the beginner.
While the treatment presented presupposes a knowledge of
Mechanics of Materials, a
straining actions
is
brief discussion of the
more important
given in Chapter III, partly to
make
the appli
cation of the various formulae to engineering problems
somewhat
Considerable of the matter contained in the book has already been published. in no sense. as it The in treatment of all topics is necessarily brief. Barr. specially for the use of students in Sibley College. with due regard to constructive considerations and other practical limitations. It is hoped that the tical illustrative numerical examples which are intro duced throughout the work may. neither is it a manual for the drafting room. more assistance in selecting and partly to present such rational theory as is of working stresses and factors of safety. was desired to obtain a textbook which could be conve niently covered one college year and yet present the salient by the student as a preparation more advanced work. suggest proper treatment of practical prob lems in design. while the discussion is necessarily brief it is to the engineer. believed that the fundamental principles are fully covered. under the title of " Special Topics on the Design of Machine Elements." by John H. a handbook. but to reduce the analysis to such forms and terms that definite numerical results can be obtained in concrete problems. It has been the endeavor in the preparation of the book not only to develop rational analytical treatment. but is a discussion of the fundamental principles of design. The general principles of lubrication and efficiency are Both of these are of prime importance discussed in Chapter IV. in conjunction with the analy methods given. to show why certain equations have and also to collect in concise form the more important equations relating to stress and strain with which the designer needs to be familiar. and also in " Elements of Machine . of the book is devoted to the discussion of more important machine details. and only such practical data have been The remainder of the some collected as are needed to verify or modify logical theory. The treatise is. While intended primarily for engineering students it is hoped that it may also prove of some features of the subject needed and basis for interest to the practising designer. and This discussion serves also been selected in preference to others.iv PREFACE definite. with a view of showing how the theoretical considerations and equations discussed in the first part of the work are applied and modified in practice.
." Part I. Ithaca. Cornell University. Briggs. 1909. The of authors are especially indebted to Professor G.PREFACE Design. D. Y. A. for many helpful suggestions and and criticisms. all and C. Albert of whom also have given instruction in the course at various times. J. N. D. Barnard. L. B.. Professors W. selves freely of "the work of The writers have availed themmany others in the field. V by the Authors. for which due credit is given in the text. N. H. J. to Mr. Blessing Swarthmore College. S. of Sibley College. K. A. They will be very grateful for further suggestions or criticisms which will improve the book. June. F. Darling.
.
of Machine Design. . 114 CHAPTER Riveted Fastenings. and Force Fits. .156 CHAPTER Keys. 6 CHAPTER III Straining Actions in Machine Elements. Fundamental . . and Thin Plates . IV 96 and Efficiency. 31 CHAPTER Friction. . Flues. Consideration of Machines as a Means of Modifying Energy. Cotters. VI 136 CHAPTER Screws and Screw Fastenings. . VIII 190 CHAPTER vii IX 211 Tubes. . VII . . Formulas for Strength and Stiffness.CONTENTS CHAPTER Introductory. CHAPTER V Springs. I PAGE Definitions and Fundamental Principles i CHAPTER II The Energy and Force Problem. Lubrication. Pipes.
406 CHAPTER XVI Machine Frames and Attachments. and Couplings. CHAPTER XIV Toothed Gearing. Shafting. . Bearings. 428 . Spur.285 CHAPTER XII 309 Belt. Bevel and Screw Gears. . Friction Brakes. . CHAPTER Applications of Friction. and Chain Transmission. XIII Friction Wheels. Roller and Ball Bearings. . Sliding Surfaces. . 364 CHAPTER XV Flywheels. 350 and Clutches. . . . . 232 CHAPTER XI Axles. Rope. Pulleys and Rotating Discs.viii CONTENTS CHAPTER X PAGE Constraining Surfaces. Journals.
Mechanics of Machinery is that portion of pure mechanics which is involved in the design. the stationary frame and the nature of the connections between them. Machine Design therefore may be defined as the practical apMechanics of Machinery to the design and construc tion of machines. solid. The constrainment of determined by the moving parts. I INTRODUCTORY i. of machinery is to transform energy obtained for directly or indirectly from natural sources into useful work the laws that govern motion human force. hence is the basis of Machine Design useful and The term and that definite force. for urable character. construction. its work carries with it the idea of definite motion work itself is always of a definite or measAn examination of any machine will show to parts are so put together as to give definite constrained motion suitable for the work motion is be done. Mechanics is the science which treats of the relative motions of bodies. and operation of It machinery. and of the forces acting upon them. liquid. or gaseous. has been noted that the consideration of a ma chine involves constrained motion. their practical which machine action application brings in many modifying plication of conditions. Useful work involves both motion and force.MACHINE DESIGN CHAPTER The purpose needs. of Mechanics of While the laws Machinery give us the underlying principles on rests. . hence that portion of pure me chanics is mostly needed in Machine Design which deals with stationary structures and constrained motion.
with regard to the amount of energy to be transmitted. so far as motion concerned. relative proportions of its parts can be designed to suit the work . however. Proportioning of the parts so that they will carry the necessary loads due to transmitting the energy. bination of mechanisms which A mechanism or com constructed not only for modify ing motion but also for the transmission of definite forces and for the performance of useful consists of one or not necessarily work is called a machine. and are used only to modify motion as in the case of etc. the solution of the motion problem borders on or may indeed be of the nature of invention. vided into two main parts (i) (2) Except for the limitations of such cases as those just noted. models.: 2 MACHINE DESIGN A Mechanism is a combination of material bodies so con nected that motion of any member is involves definite. A machine more mechanisms. or in general mechanisms as centrifugal gover where inertia or other kinetic actions affect constrainment of motion. Many mechanisms transmit no energy except that required to overcome their own frictional resistance. constrained motion of the other members. (1) is The design or selection of the mechanism for a machine is governed by the manner in which the energy supplied and the character of the plied in one work to be done. a mechanism. While it is true that in most cases the mechanism and the will with quite a different one. But if a new type of machine is to be built. is a machine. most engi neering instruments. for energy may be to sup form of motion and the work may have If be done mechanisms already exist which accomplish the desired result the problem is one of selection and arrangement of parts. of course. watches. This. without undue distortion or practical departure from the required constrained motion. A brief reflection will show that the same mechanism will serve for different machines (see any treatise on Kinematics) and for a given is within limits the design of the mechanism machine lit may tle usually be carried out. relative. does not apply to such nors. the design of any machine may be di Design of the mechanism to give the required motion. or a new mechanism is desired.
and such is the case when the stresses are simple and the conditions fully known. is necessary to know something about the energy trans mitted before any definite dimensions of the parts of the mechan ism can be fixed. of the energy and force problem in the mechanism. When the type and proportions of the relative velocity of If been fixed the may be transmit found. the theoretical design is not always clear. the forces acting on a machine member and the manner in which it is connected are known. or the product of velocity multiplied by force If constant throughout the train. these may serve as a basis for the assigning of definite dimensions to the part. anism is determined. (b) Assigning of dimensions to the various parts based on the forces acting (a) upon them. (2) necessary to keep the latter con The problem of proportioning the various parts of a ma chine so that they will carry their loads without excessive or un due deformation may conveniently be divided into two parts: (a) Solution as a whole. on a machine member can be deterwould seem easy to choose the material and assign proportions to it based on the laws of Mechanics. and frequently before the nature of the mechFurthermore. But in many cases the forces acting are very complex. in general. A fuller discussion of this important principle is given in Chapter III. and our knowledge of materials . it is ing at any point since the law of Conservation of Energy underlies all is machines. is mechanism have mechanism then the energy which the mechanism must any point in the possible. can be intelligently proportioned in this manner. to find the forces act known. (b) If the forces acting mined it Thus a machine member subjected to simple tension within known limits.INTRODUCTION to it 3 in general be done without reference to the energy transmitted. and in general while the it is mechanism can be from the energy problem stantly in mind. Thus be be in designing a must first fixed before the length of crank and connectingrod can treated apart fixed. the methods and available steam engine the size of the cylinder facilities of construction control the design to a large extent.
(IV) and Drawing. (b) Modify such design by judgment and a consideration of the practical production of the part. the logical steps in the design of a machine are as follows: (I) Selection of the mechanism. experience. as the writing of specifications presup poses the most intimate knowledge of theory of design. Recourse must there fore often be made to judgment or to empirical data. sign tioning machine elements where theory as follows is applicable is. It is. etc. written specifications useful but necessary. (II) Solution of the energy and force problem. making a record of what has actually been done. and Drawing. Summing up then. therefore. judgment and empirical data are commonly the best guides. It is not machine design of itself. is The last step. (a) Make as close an analysis as possible of all forces acting and proportion parts according to theoretical principles. Specification a necessary and a powerful aid of important adjunct to the process of design. (III) Design of the various machine members so they Specification will not unduly distort or break under the loads carried. theoretical demust always be tempered with practical modification and by The logical method of proporconstructive considerations. From the foregoing it is seen that the part of generally is Machine Design for convenience included in Mechanism can be and .: 4 MACHINE DESIGN their laws is limited in and many respects. to the designer's it is mental process and is to best way showing the workman what and also of be done to construct the machine in ques tion. and selection of materials. Very often accompanying the drawings are not only In fact the highest skill on the part of the designer just is often needed to clearly and fully specify in writing what is to be done. as machines may be designed and built without any drawings. is the. In the case of details and unimportant parts. nevertheless. an indispensable part of the designer's equipment. however. the result of Even when the conditions are clear.
Mechanics of Engineering. and the Design Machine Elements. The chapters that follow deal therefore with the solution of the Energy and Force Problem. and the student is 5 expected to have a knowledge of Mechanism. and Materials of Engineering as a preparation for the work contained of in this book. . Mechanical Drawing.INTRODUCTION taught as a separate subject.
Thus the energy supplied during a number of cycles may be some heavy moving part and then be given out during some succeeding kinematic cycle. The slidercrank mechanism of the engine considered separately the crank. the useful work done must always be less than the energy received. From the law of Conservation of Energy it is known must be that energy can be transformed or dissipated but not destroyed. occur in all machines. frictional losses are neglected. makes a complete kinematic cycle every revolution of The engine makes one energy cycle every two revo6 . pass through a series of energy changes. unless otherwise stated. Since frictional resistances. An Energy Cycle is made by a machine when its moving stored in parts start from any given set of simultaneous energy conditions. The energy received by a machine during a kinematic cycle may or may not be equal to the work done plus frictional losses. Thus the complete mechanism of a fourstroke gas engine makes one kinematic cycle every two revolutions of the crank shaft. and ultimately return to their original energy conditions. In the discussion which follows in this chapter. Therefore all the energy supplied to any machine ex pended as either useful or lost work.CHAPTER II THE ENERGY AND FORCE PROBLEM 2. pass through all positions possible for them to occupy. A Kinematic Cycle is made by a machine when its moving parts start from any given set of simultaneous positions. as in the case of a punching machine with a heavy flywheel. and ultimately return to their original positions. ent for different machines and is evidently a fraction or less than unity. and frequently other losses. The useful work delivered divided by the energy received is This expression is differcalled the efficiency of the machine.
during a kinematic cycle. A ficient knowledge of the quantity of energy required desired to do the work during a complete energy cycle is not always sufinformation upon which to base the design of the machine its or the capacity of driving device. And during an energy cycle. determine the capacity of the driving device. obtained by multiplying together simultaneous values of velocity A making any energy transformations both the force and the velocity factors must be kept in mind. for any member it is therefore necessary to make a complete driving device. it follows that in source of energy into the desired motion. on the maximum force it must carry. it will be making a complete kinematic cycle every stroke and a complete energy cycle every four strokes. to select the known and the problem mechanism which will trans form the motion to of the source of energy into the required motion. and in such a case may be greatly in excess of any direct force that the driving deBefore this maximum force can be determined vice may deliver. 7 If a punching machine driven by a belt and running continuously. it may not necessarily . as regards strength and rigidity. Generally speaking. machine may receive energy at either a uniform or variable rate and may be called upon to do work at either a uniform or Power or rate of doing work being the product variable rate. Energy received = useful work + lost work. While the mechanism chosen may transform the motion of the and force. maximum force may be due to the direct action of the it may result from the inertia effect of some member which has a capacity for storing energy. Therefore. and to proportion the machine members. Energy received = useful work + lost work ± stored energy.THE ENERGY AND FORCE PROBLEM lutions of the crank. The and this proportions of any machine part depend. punches a hole every fourth stroke of the punch. or solution of the energy problem including the determination of the driving device. therefore. the useful work to be done and also the character of the source of energy are of design is.
thus reducing the operating force. will that the capacity of the driving device be greater in such a case than based on the average rate If of energy required per energy cycle. and during one part of the cycle they mately fulfils required may absorb energy. recourse must be had to a mechanism or to the use of flywheels or other means storing and redistributing energy. and out when the effort the capacity of the driving de vice need only be such as will supply during the energy cycle an amount of energy equal ing that cycle. may give up energy. If simultaneous values of the force and velocity at the working point are multiplied together. (a) In many machines under it continuous operation. the driving device under the above conditions should be too large or expensive. If it is predetermined that some device is to be used for for giving it storing energy when the effort is deficient.8 MACHINE DESIGN so modify the energy as to give a distribution of force at the point where work the is being done which exactly or even approxi conditions. consider any . and may materially modify the design. thus increasentirely Such a condition may make an different distribution of the forces acting on the members of the mechanism. where flywheels are not desirable. as is liable to different of be the case in large work. their product is the rate at which work will will be done at the point considered. The maximum which work product thus obtained will be the maximum rate at be done and also It is evident at which energy must be supplied by the if source. Two such cases may be noted. the force at the operating point is sometimes greater and sometimes less than that required. some of the moving machine parts may have to be very heavy in order to carry the required loads. Again. (b) Again. while at another part of the cycle they ing the operating force. desirable and in to the useful work and lost work dur But in many machines such devices are not many others they cannot be applied. is in excess. from that which would occur were the parts light or the motion of the machine very slow. is found that if the driving device is proportioned so as to supply energy at a uniform rate equal to the average rate required throughout the energy cycle.
Case received at As an example of this case. Evidently the rate at which energy * In certain hoisting devices friction is utilized to sustain the load or prevent overhauling. the driving device chosen might have to be excessively large this condition might preclude the use of the driving device first selected. at a variable rate and work be done at a variable 3. but sustain the load at any point. must be at least equal to that of the load. where energy is a constant rate and work done at a constant rate. at a constant rate and work be and work be done at a constant (b) Energy may be received rate. the torque of the driving device must It is also be able to start and evident that in such cases on the hoisting drum. Not only must it the driving device supply during the cycle of operations (the raising of the load) energy equal to the work done. In any of these cases. before apply ing the above principles to the discussion of illustrative problems. and may be well to enumerate the combinations that can occur. consider a steam turbine running a centrifugal pump raising is water to a fixed level. . be noted that the choice of mechanism and the capacity governed largely by the relative manner it of the driving device are in which energy is to be received and work done.* minimum and torque should be small compared to the maximum. (a). the maximum force that may come on any member may It is to also be determined.THE ENERGY AND FORCE PROBLEM hoisting Q mechanism. Energy may be received rate. this statement does not apply broadly to such cases. at a constant rate done at a variable (c) Energy may be received at a variable rate and work be done at a constant (d) rate. In any machine under continuous operation energy received and (a) may be work may be done in one of the following ways: Energy may be received rate. its minimum torque must be equal If this to that of the load when referred to the same shaft. and if the torque of the driving device should be variable. after the form and capacity of the driving device have been determined.
During the re turn stroke. The machine runs is idly a large portion of the time. During the return stroke frictional flywheel by bringing its velocity resistances only are to be overcome. 1 has been . Case (b). The total capacity of the driving belt need only be sufficient to supply. very easily determined. which is of considerable magnitude. As a second example of these conditions. such as a flywheel. is delivered in If the driving belt were designed with sufficient it capacity to force the punch through the plate by direct pull would have to be very large. the parts. the wheel giving up stored energy. the work done at a variable rate. an amount of energy equal to the useful work plus the lost work. at a The machine is driven by a belt which can supply energy is uniform rate and. during the energy cycle. termittently. Here the driving belt can supply energy at a constant rate while the useful work. however. The maximum force that may be transmitted by the machine members will be based on the maximum force at the tool and will be transmitted only by the members that lie between the tool and the flywheel. and the capacity of the turbine 4. Here the useful work is done during the forward stroke of the ram. The resistance of the cut is during the forward stroke limited is uniform and the speed of cutting by the character of the metal to be cut. the velocity limiting velocity depending may be greatly increased. devised to meet these condiSuppose a mechanism such as shown in Fig. for any given period. as on the mass of the moving these should be brought to rest at the end of the stroke without shock. can be used to advantage. As an example of this case (energy received at a constant rate and work done at a variable rate) consider the case of a machine for punching holes in boiler plate. as noted above. take the design of a small shaping machine. and in a machine of this kind a device for storing energy. During the time that the machine is running idly the belt can store up energy in the up to normal. Numerous mechanisms have been tions. while the plate being shifted. When a hole is punched the velocity of the wheel is reduced.IO MACHINE DESIGN work is supplied must just equal the rate at which tional done plus fric and other is losses.
so The crank pin on a the outer position as that the vibrator e can be made to give the ram R any length of .THE ENERGY AND FORCE PROBLEM selected. D x is also on the smallest step of the countershaft is adjustable and can be moved from shown toward the centre of the crank. On it the countershaft overhead is a mating stepped pulley so placed that when the belt is on the largest step of the machine pulley. is The gear b is in which is rigidly attached to the shaft On the other end of d a stepped pulley having diameters el. length of stroke should be 3 Continuous rotary motion is imparted to the crank it minimum a through the gear b of which turn driven by the pinion d. c forms a part. Diagram Fig. D D D X 2 Z Z> 4 . II The maximum be done and the length of the stroke is fixed by the work to or 4 inches.
The diagram is for the backward stroke is not drawn since it not needed in the solution of the energy problem. is and the problem is therefore to design the driving belt and proportion the machine members on the basis of the maximum pull which the belt may be able to exert. Jones. . If the belt supplies it energy at a constant rate the force its which can deliver at the tool will cutting velocity. the inertia of the moving parts small. may not give the distribution To design the driving device (or belt) for such a mechanism of * For a full discussion of these so called quickreturn mechanisms and the methods drawing velocity diagrams see "Kinematics of Machinery" by John H. Consider the crank pin at its extreme outward position and the belt on for full forward stroke D x .12 stroke from the MACHINE DESIGN maximum (20 inches in this example) to a minimum for of 3 or 4 inches. " Machine Design " by Smith and Marx and " Machine Design. so that at any stroke It is variable speed may be obtained to suit the metal to be cut. The range of velocity of the tool it any length of stroke must be such that can be lowered to the cutting velocity of hard cast iron or tool steel and raised to the economical cutting velocity of brass. The mechanism transforms the uniform rotary motion of the line shaft into the required reciprocating motion. If the belt is shifted to a step as D 2 the velocity of the ram will be increased. D the ram will is be a maximum for that position of the As the crank drawn toward the centre (the belt re ram is obvinow smaller ously decreased. by F. R. The cutting resistance. The velocity diagram under these condi tions is shown. is uniform so that while the mechanism produces the it desired transformation in motion of force desired. the ordinates of the diagram* representing the crank length represents velocity of the ram to the scale that the the uniform velocity of the crank pin. Barr. however. but it should in general be drawn to make sure that the change in velocity at the extreme is ends of the stroke vary inversely as not excessive. With the pin in its x extreme outer position and the belt on the large step speed of the belt.. maining in its original position) the velocity of the not desirable to use a flywheel." Part I.
it will is If then the belt is is when the ram making full on D lf and hence at the lowest belt velociIf a have excess capacity when in any other position. 12. thus keeping their product approximately constant. must be increased by shifting the belt to a smaller step of the machine cone. the belt remaining in the same position. but this can only be done by shifting the belt to a position where its and hence its capacity will be greater. The belt should then be on the smallest diameter D 4 and hence at its highest speed. Since 7\ — T 2 has. shortest stroke . required for small machine tools ' The maximum power approximately constant is at all speeds. designed to have sufficient capacity stroke and the belt ty. it follows that as the cutting speed increased the resistance of the cut must be decreased and vice versa. (See Church's is The power* that a belt can give out V (7\ — T 2) where V is the velocity of the belt. the effect of moving the crank pin inward. The belt receives its energy shaft running at constant speed and when the it belt is smallest step of the countershaft cone largest step will also be on the D 1 of the running at its lowest velocity. is to decrease the average velocity velocity of the ram. the power that a belt can deliver will vary directly with its velocity. is softer metal to be cut the velocity of the ram may be increased.3 THE ENERGY AND FORCE PROBLEM the operating conditions of the machine its 1 when the belt has both maximum and minimum maximum pull which a belt ''Mechanics. As before noted. a constant maximum from a on the value for a given belt. is 7\ — T 2 where 7\ The is the allowable tension on the tight side of the belt. to maintain a given cutting speed. its The other limiting condition is when the ram is making and giving a cutting velocity high enough for the softest metal to be worked.) therefore velocity can give must be investigated. at all moderate belt speeds. Therefore as the stroke is made shorter the velocity of the crank." page 182. * A full discussion of the power transmitted by belting is given in chap. for since the heating effect tool is proportional to is which governs the cutting capacity of the the work done. machine cone and will in consequence be under which condition its capacity for delivering energy is a minimum. .
cutting speed on full speed = 25 per min. for a given belt pull since no flywheel used. the capacity of the belt when running on D is made great enough to give a force at mid position of the ram equal to the required cutting force. linear velocity of ram max. and if this condition does not give too large a belt the driving device will be satisfactory. and the resulting stresses may be easily computed. Example Let the greatest resistance of cut " " " = 800 20 4 lbs. The maximum force be based on the that any member may have belt. X velocity of belt = x force at tool X velocity of tool. there ram making full stroke. .: 14 MACHINE DESIGN An full is inspection of the velocity diagram its stroke shows that velocity is ram is making a maximum when the ram when the will in mid position. Neglecting friction and inertia. however. (i 6y 2 iH ic " " max. inches. In general. " " " " " = maximum stroke of ram = minimum stroke of ram = maximum length of crank " " minimum " belt speed inches. Then in general. to sustain will will maximum torque of the which occur when this it is running on D x \ for since the inertia forces are small torque will be transmitted directly to the members. stroke and " " max. the velocity of the ram is a maximum at any given is because. ft. ordinate of diagram* is length of crank Hence in this example when the ram making full stroke at lowest speed. cutting speed on shortest stroke and highest lowest belt = 60 ft. it will fore. the force exerted on the ram be a minimum where belt velocity. force at belt If. which here are small. per min. linear velocity of crank max. the diagram must be drawn in order to locate the maximum ordinate. * In the mechanism here chosen the position of the ram for maximum velocity can be located by inspection and the value of the velocity determined without drawing the complete diagram. with the have excess capacity at any other position.
'. approximately. of crank per min.5 ft.M. the machine be 85 per rate of doing work 800 — X • at this position of belt is the cutting resistance multiplied by the maximum 25 velocity of the ram. . per min. A=A^ Max.P. In a similar way when the ram at highest speed. velocity of belt on low speed ft. as is If the cone pulleys on machine and countershaft are the usual case in metalworking tools. divided ft. width of belt = — = 2W" 45 of of of of nearly.P.500 lbs. R.M. or = 85 23.1. R. _ . . R. is of shaft d = 55.M.P. = 14 X 7T 55.M. of crank 23 5 ' 2 X *2 = X x 6J 7T 6.M.M. ' shortest stroke Linear velocity of crank = 42. Then the minimum and maximum R.8 5" .P. The effective pull of singleply belt per inch of lbs. cent. then A A and hence.9. ° Let the gear ratio be 8 to 1. * • Dv 1 . R. width may be taken at 40 to 45 . per minute.2 — maximum by the 12 = 204 If the efficiency of the per min.P. is making the ft.P.2 and r i_ 1 i 432.P. in the m ' example A= i 4j A= x4 \ \ ^' 2 = 432. = 42 5 2 X 7T X " = X ii 54. Max.'. alike. of 15 crank = = 2 cf v 6 = 23. efficiency. \ effective pull at belt = = 115 lbs.5 Therefore. nearly.8 respectively. A X 14" pulley a convenient diameter for 1 . R. if Min.THE ENERGY AND FORCE PROBLEM Linear vel. J Machine Cone Machine Cone Machine Cone Machine Cone # " ^Min.M. R.
For every cycle of the machine the shaft d makes 8 revolutions. 5. the maximum is load applied in entirely different ways. . But when the power needed is great. when the belt The difference on D the largest step of the in this respect between this case and the while punching machine discussed above should be noted. sometimes making it introduce a flywheel to reduce the size of motor. One of the best examples of Case (c) where at a energy received at a variable rate and is work is performed uniform rate since this it found in the reciprocating steam engine. mechanisms of both can deliver energy at a uniform and while both do work at a variable rate. and is machine of such great importance to the engineer detail. occurs cone. During the complete energy* cycle of the machine the total work done. and after cutoff occurs and the steam expands *The in the cylinder the pressure falls from kinematic and energy cycle are. hence the if belt could deliver amount of energy that the work were done uniformly during one cycle is 8 X — it X 115 = 3370 is ft. lbs. in fact. lbs. This. this added first cost is not serious. Where a small machine is beltdriven. such as a flywheel. or 800 X 20 — = 1333 ft.6 1 MACHINE DESIGN The maximum force that may x be applied to any member which machine will be based on the maximum is torque of the driving belt. in this case. Case is (c). neglecting friction. will be discussed somewhat in Here the energy is supplied in the form of steam pressure. had been used. as in the case discussed. the additional cost of a driving device so greatly in excess of average requirements needs to be carefully considered. the driving rate for. is equal to the length of stroke multiplied by the uniform resistance of the cut. simultaneous. The great as capacity of the belt therefore two and onehalf times as would need to be if a device for equalizing the energy. is one of the most important desirable to elements to be considered in fixing the size of motors needed for directdriven machine tools. or in such cases as direct driving by electric motor.
Stroke— Fig. The energy therefore supplied at a But the engine is required to deliver energy at The mechanism used will the driving belt at a uniform rate. . varying rate.THE ENERGY AND FORCE PROBLEM the "initial" or boiler pressure to 17 somewhat above exhaust or is atmospheric pressure. 3.
2 (2) pressure* on the other side of the piston. The reciprocating parts. in above atmospheric pressure noncondensing engines. be greater and sometimes less than the resisting effort of the driving belt and the machine will stop unless a redistributing device. thus outlining the curve whose ordinates represent the velocity of the crosshead at any point. and their maximum velocity is considerable.8 1 MACHINE DESIGN produce the required transformation of the reciprocating motion But the of the piston into the rotary motion of the crank shaft. distribution of the driving force in the gential effort will not form it of torque or tan be uniform but will be a maximum somewhere near the position at which the crank is at right angles to the connectingrod. velocity of the crank pin be represented crank. simultaneous velocity of plot the crosshead to the same These intercepts may be ted at the corresponding positions of the crosshead. forming a polar the diagram of the required force at the pin. If plotted radially from the crank circle as a base. and it becomes zero when the crank is on The turning effort will therefore sometimes the dead centre. hence the forces due to their inertia cannot be neglected. such as the piston and crosshead. is used. Referring to Fig. is equal to equivalent that at the This required driving force at the crank pin may be circle S. * This generally amounts to 2 or 3 pounds per sq. the ordinates of the curve T. belt. actin. . 2 (a) the crank a is required to rotate around the center O with uniform velocity and to give a uniform force are heavy . such as a flywheel. Fig. the intercept tical Op by the length of the made by the connectingrod on the verthe through O will represent scale. The moment at the driving belt the average moment at the crank pin. hence the uniform force at the crank pin may be derived from at the driving belt. and also the connectingrod. (1) The steam The back pressure which is represented at any point by (b). as shown by The crosshead C moves at a varying rate of speed. The forces acting upon the piston and which must be trans mitted to the crank are.
They can be represented first graphically by such a curve as V. and V will give W. is also known the force necessary to produce the known since accelerating force = mass X acat and the force any point (reduced sq.9 THE ENERGY AND FORCE PROBLEM 1 ing against the steam pressure. page 71. by the connectingrod crank pin The pressure of the rod against the may be resolved into two components. U. due to accelerating and retarding the heavy reciprocating During the first part of the stroke these inertia forces tend to reduce the effective pressure transmitted to the crank pin. * For a full discussion of this matter see "Kinematics of Barr. hence the inertia forces increase the effective crankpin pressure from that point on. The two curves can be found by the wellknown methods of drawing indicator cards. When the reciprocating parts reach their maximum velocity their acceleration is zero. It is believed that the analytical method is the most satisfactory. H. its ordinates must be subtracted from the forward algebraic The a resultant pressure curve sum of the curves T. of piston) may be plotted as to pounds per shown by curve V. The compression curve (U) tends to decrease the effective pressure on the piston and hence pressure. and represented by the exhaust pressure line z z and the compression curve U. This point is the crank and the connectingrod are at right angles error introduced ratios of and the by assuming this to be so is small with ordinary the recipro crank to connectingrod length. . one tangential Machinery" by J. and during the latter part they increase the effective force on the rod. and such a method is presented in a suc ceeding article. 2 (c). (3) The inertia forces parts. Beyond g cating parts are retarded. Fig. This effective pressure is transmitted to the b. Fig. in. paragraph 42. whose ordinates at any crank point represent the effective pressure acting on the piston rod at that point. and the third can be found either by mathematical deduction or by graphic methods* based on the velocity diagram. 2 (b). hence the curve of acceleration forces very nearly at the position where crosses the axis at a point g corresponding to the point of maxi mum velocity. If the acceleration is acceleration celeration.
2 (a) represent the velocity of the crosshead to the same scale as the length of the crank plotted radially on the crank circle as a base. give the simul taneous values of force and velocity at every point of the stroke. 2 (c) . — = Lay rm. off Oi = e 1/ 1 and draw i k parallel to ej. In Fig. thus giving (c).— 20 to the . represents the velocity of the crank pin. The construction for the return performed in a similar manner. (c) Now and 2 respectively. * . MACHINE DESIGN crank circle and tending to produce rotative motion. and to one radial along the crank tending tension in the crank and produce compression or friction in the main bearing. be noted that the distribution of force as represented is less It will by this diagram uniform than the original curve of press . If friction Only the tangential force can do useful work. j as Therefore stroke O k may be laid off radially from an ordinate of the required curve as is j k'. . Fig. and this may be what is Curve X. It will be remembered that the ordinates of the velocity diagram (R) as drawn in Fig. the connect ingrod extended. hence the velocity of the crank pin multiplied by the force crank pin is equal to the velocity of the crosshead multi plied x by the force x at the crosshead. Neglecting friction. Therefore O h = velocity of crosshead when O j = velocity of crank pin. 2 . 2 (a) being done at the piston. called a radial crankeffort diagram. Ok prr = Oi — * Therefore. If such simultaneous values be multiplied together and divided velocity of the crank (all by the uniform the quotient is in the proper units) the tangential force at the pin. . the rate of work at the at the crank pin is equal to the rate of work at the cross head. if necessary. cuts the perpendicular through O in the point h. tangential force = ejiXOh b. be neglected crank must the rate at which work is done by is this force at the equal the rate at which work the curves R and W. Ok = m OiXOh —— (J ] = XOh — u 1 tangential force. Fig. These values of the tangential force can be found more easily graphically. Then. or the tangential force XO j = e f xOh.
I v Kv etc. Fig. J. Since the abscissas represent space and the ordi nates represent force. howmechanism must produce a uniform turning effort at the driving belt or such as would be given by a crankeffort diaA flywheel must therefore be used to gram like S. on which these graphic methods depend. 2 (a) store energy when the crank effort is in excess and to give out energy when the crank effort is deficient. Fig. are open to criticism on account of their inaccuracy be cause the tangents or subnormals to the curve. of the force diagram W. The base YY is equal to the circumference of the crank circle and the ordinates of the line / m are equal to the ordinates of the required uniform crank effort curve S. veloped above will enable the designer to determine the maxi mum straining action The graphical method of finding the inertia curves. are difficult to construct with accu racy and are at some points indeterminate. one is more Referring to Fig. the areas /. . The work it represented by K+K t is that which the l wheel must absorb and the area represented by I + J + I 1 which must give up in one revolution. fore. length of crank in " feet. 21 By the conditions of the problem. The maximum force that K+K may come upon the engine is the crosshead can It is to be seen from an inspection noted in this regard that cutoff. it In general. Manifestly + J that I+J+I \1 J x must equal A full discussion of the design of the v flywheel will be given in a later chapter. The method of analysis deon any member of the mechanism.THE ENERGY AND FORCE PROBLEM ure at the crosshead. while con venient. " connectingrod in feet. K. there is thought that the following method or some similar satisfactory. 2 (d) shows the crankeffort diagram rectified with rectangular ordinates equal ever. be if designed for variable an indicator diagram at late cutoff should be drawn for the purpose of locating this will not give the maximum force. represent fly work. as an earlier cutoff maximum value. 3 (page 17). Let a " u — R = L = acceleration at any point. the . to the polar ordinates of curve X..
per min. differentiating with reference to t . dv = j. VL —R 2 sm2 sin o 2 = # 2 — J— \ R 2 sin 2 RV n — . velocity angular & ^ in radians per sec. from therefore mid — . Let k Let x = distance from = centre of crank shaft to mid position of crosshead. Then x + But Lcos = = OB 2 + BC 2 = R. x. displacement of crosshead from mid position. = 1 j?=. U " v ~ = R' velocity of crosshead at any point v. " " t .. and <p = angles made with centre line by the crank and connectingrod respectively at any position measured from the crank position O r. /2 7riV 2 \ / COS2^\ (5) . r hencea = (^J^^co5^ + ^J.. v = dx 7 = = / —_( ic I l sin * + / sin 2 <9\^6» sin2#W0 2^r^ + cos2 0'\ •• (dd\ 2 •  G) The acceleration a . x + k =R cos e + in — J (2) Now x = the distance stroke and velocity (2) moved through by dx and at x = the crosshead.R 1 • d2 x \ COS0 ^r)\Tt) 27zN 60 ' (4) but de — = d/ .22 MACHINE DESIGN Let " N = Rev. cos e +L 2 cos <p.== o) time elapsed corresponding to = k <p angular velocity in radians per second. x + k = R (costf + Vn — sin o) (i) all Expanding the radical by the binomial theorem and omitting error with the limiting proportions ordinarily used) equation terms beyond the second (which can be done without appreciable (i) becomes.
(6) ' or reducing. from Mechanics it is known that the force necessary to produce an acceleration (a) is P = p W a where g = 32. . . In general. = WrN 2 .— THE ENERGY AND FORCE PROBLEM which is 23 the general expression for acceleration of the recipro cating parts. . equal to o By means of equation (7) all points on the acceleration curve and it is could be found and plotted. (7) J When when the solution of the above expression gives a negative result the force of inertia is acting positive. and a curve drawn through the three points thus located is sufficiently accurate for all ordinary purposes. . m ( I0 ) The piston is not at half stroke. . (Q ) e oo° or 270 * P = (^1^1) ^ 35200 W . however. P = — fWr N [ V \ ( ) 35200 / \ I+) n' (1 v 1 \ . In cases of extremely high speed be desired. is in ft. 2 It is also to be noted that the of expression W R (2xN ( \ ) is the centrifugal force a weight equal to that of the reciprocating parts concentrated at the crank WRo pin since centrifugal force in general is 2 . If the weights of parts be called W.200 V n where r is in inches. i n ) where R _ • . 2 with small ratios of connectingrod to crank a more accurate determination of the curve may When When When e = = = * 0. 35. the exact sufficient to characteristics of the curve are not essential make the three most simple solutions as follows. (8) 180°. therefore —™ r f2N — ^ 60 —M 2 \f ) COS2 0\ cos H . away from the crank and toward the crank. . P = + (El^!) \ / 35200 ' — ) »/ (L) / . . .2 — n R in English units. ( cos + cos2 0\ ) .
Cutoff at y& stroke. Ratio of crank to connectingrod =1 to 5.24 MACHINE DESIGN If the inertia forces are to be combined with the steam press ures.75 = The of pressures taken = 70 X = 70 lbs. therefore the area between z z and the curve T represents the work done by the steam pressure during the stroke.P. in.875 = 62 lbs. to pounds per square inch of piston to give correct An example may the following data: serve to make these points clearer. = strokes per minute multiplied by length of = 640 ft.875". = 2". Area under ordinate T minus 1 area under . they must be reduced diagrams.P. Here P. before definite dimensions are assigned to the various Let a theoretical indicator card be drawn as in Fig. Here something must be known about the members. = horse power required. H. with Steam pressure = 90 Piston speed stroke lbs.75 scale There fore mean is = . U = 1. . Then H. 2 (b). Let it be required to design a steam engine to deliver 150 H. are known. size of cylinder necessary. z z In the case given sq. = 2 N P LA 33000 N . The difference of these areas is the net work done per stroke of piston and the mean ordinate corresponding proper scale the average pressure to this area represents to the per sq. *P = mean effective pressure per L = length of stroke in feet. as shown graphically in Fig. gauge. 2 (b). in. . 1" Therefore mean pressure during stroke Let A = area of piston.P. In a similar way the area under curve U represents the work of compression due to back pressure.P. sq. inch on the piston during stroke. neglecting for the present the inertia curve V since this its only tends to redistribute the energy and does not affect quantity.N X L and H. The distance z z represents the piston travel and the ordinates of the curve T represent piston pressures. = number of revolutions per minute.
2 = 24. E = energy to be absorbed. =i ft. N = i6o R. on which the de sign of the flywheel can be based. lbs. If the stroke about twice the diameter of the cyl inder.5 sq. of area = 70 lbs. curve can be rectified as in Fig.5 () =6. The area I + J=K will be proportional to the energy be absorbed and delivered by the flywheel. in. is to locate the inertia 5> curve. . 2 (d) and the (c) can now be The crankeffort mean ordinate Yl . . to locate the curve as in Fig. The general expression for the inertia force for e = o. therefore one sq. and 90 or 270 (Fig. One inch of ordiof abscissa nate here = 70 lbs. C is a constant and here equal 4 Therefore. In the above example W = 35> n = an d N 1 ) = 160.5 (1 — ) Fig. area of if ^ = . ^^ A = H.THE ENERGY AND FORCE PROBLEM 25 . drawn and drawn. When When These values serve = = oo°.5X70X132=4. curve W. of piston and one inch ft. P= W rN 35. Hence (a since 2L X ^ = 640.200 When = P = 30. to the crankeffort diagram X plotted. in. per sq. £ = .5 (1 + J = 36. the proportions will be good. =1 ft. * Reduced in reproduction about onehalf. 2 and c). (2). The resultant of T U and V.4 lbs. 35. 3. P = .200 2 ( 1 1 v \ + / = C / I 1 n v \ + 1 ) n' where <. in.1 180 P = 30. and* area of piston = 132 sq. in. As before where sufficient is the location of the three points. 180 . or say 24 inches. The mechanThis has been done in Fig.P. 3).P. 30.620 ft.* the space scale being i" stated. lbs. The Hence. respectively o°. 12 X 160 2 to = ^o. ism can now be laid out to scale. namely. is.M.6 lbs.5 X o°.X 33000 PX2NL be taken at = 150X33000 =I32 62X640 SqUare mcheS ' or a diameter of cylinder of 13 inches.
equal the efficiency. . pressure in the air The area of the cylinders are. be assumed.26 MACHINE DESIGN The maximum pressure that can occur on the piston is the initial or boiler pressure as the ordinates of W are at all points will less than those of T. Thus the loss on the steam side can be represented by the line a b. will occur where the pressure is a however. The mean effective pressure of the air cylinder multiplied by the area of the air cylinder. The steam cylinder takes assumed to be for simplicity. found in a direct Here the varying steam pressure air in the steam cylinder cylinder as is opposed by a varying . It can to coincide more closely angular distance apart. the diagrams which represent pounds per square inch of piston area would not have a ratio equal to the efficiency. when running. 5 (a) equal. taken at 80 per cent. the parts be subjected to less load than in starting up. pressure and the air compressor cylinder delivers The efficiency of the system shown is air at 100 lbs. ure may be 6. pressure. etc.* If both the pistons were rigidly attached to the same rod is it is evident that the air maximum minimum. The Part of this is lost loss by friction. that the losses can be evenly divided between the two slidercrank chains and also that the uniform rate throughout the stroke. the maximum and minimum pressures of the cards may be made by placing the crank pins at the proper In other words the mechanism may be so designed that energy will be delivered at the working point more nearly at the rate required by the work to be done. in this case. Fig. applied before inertia forces when become is full boiler press noticeable. each cylinder independently connected to a com mon shaft by means of a crank and connectingrod mechanism. would. shown in Fig. Hence. at a vary Case D. of the steam card.. is about 20 per cent. steam pressure If. divided by the mean effective pressure of the steam cylinder multiplied by the area of the steam cylinder. where the cylinders are of different diameter and area. steam at 80 lbs. without great loss is at a error. and hence the area of the compressor card is 80 per cent. on the steam side and part on the aircompressor side. A good example of energy supplied ing rate and work done at a varying rate driven air compressor. which reduces the effective pressure at every point by a fixed * In the general case. 5 (a).
the effect of inertia cannot be neglected. c 27 d increase the effective resistance of the air The all area of the diagrams modified in this way will be equal and energy supplied will be accounted for. * 1 x x Y w Yf 4— V K> \ Kl > 'T ^ j. the . 5 (c). Fig. 5 (b). 5 (6) and steam cards are shown with the inertia curve. In a similar way ordinates to the line diagram. Fig. Since the moving parts of both slidercrank chains will be heavy. THE ENERGY AND FORCE PROBLEM amount. the air In Fig. 5 (d). Fig.
but will certain maximum capacity for delivering power. 5 (c) shows the resultant pressure curves. But on a steep hill the gears must be shifted so that the engine has a greater mechanical advantage. terminal air while on the air side pressure. as before. belong to one of these is Where the action of the machine intermittent or irregular. the curve pressures being plotted below the base line for conveniThe crankeffort curve of the steam cylinder is reprerepresented by Y.28 friction line. 6. a in practice. these general solutions will not always hold and the design of the machine cannot be based on the energy given or depend on the maximum force_ or maximum torque or. Fig. be still the excess and deficiency in Fig. that this not the most is advantageous angle. sented by X. of air ence. on the rectified curve of crank effort. and the resisting crankeffort curve of the air cyl inder is The cranks are here placed 90 apart. Here the area Ii K+K the x + Ji revolution. on the mechanical advantage which Thus the motor on an auto car has a the motor must possess. at a On a level road it can propel the car high rate of speed. will and all machines which operate continuously classes. for to correspond with the point e on the air curve 5 (c). that it due to the initial steam pressure. Another example of this is the case of hoisting mechanisms already discussed somewhat (see article 2). received. the engine making only a few turns to every revolution of the wheels. amount of energy to be absorbed and I + J + amount to be given up by the flywheel during one In the steam slidercrank mechanism the greatest is the pressure is. of effort will further reduced. made / on the steam curve. the engine making many revolutions to every one of the wheels. MACHINE DESIGN and the compression curves in their correct relationFig. ship. the steam crank being in advance.1. in other words. 5 (d). An en gine or a motor might be capable of giving out energy at a rate . if common arrangement is however. It is evident. and gives a greater torque on the axle. This would This is even more clearly shown place the cranks at 45 apart. will be that due to the In the four cases discussed above the action of the maall chine has in instances been supposed to be continuous.
the design of the hoisting devices becomes quite complicated and It will is beyond the scope of the present treatise. hoisting devices and similar machines designing in mind in which act intermittently and slowly. in hydraulic distribution. however. Redistribution of Energy and Inertia Effects. the flyits wheel absorbs the excess and thereby has velocity (and hence . or by accelerating and retarding some heavy moving part. 6. the by compressing a gas. or where redistributing devices are undesirable or impossible. by using a spring. to raise the load to the required height. In electrical distribution a storage battery sometimes used for the by compressed distribution is air a large reservoir is same purpose. of Thus.THE ENERGY AND FORCE PROBLEM equal to that required to lift 29 the load in a given time. minimum torque of the motor or engine must always exceed the maximum torque of the load when reThis general principle must be kept ferred to the same shaft. Where the torque of the load is constantly changing.1 1 e A B ^m^— r~ Fig. In transmission of power sometimes employed as a re storehouse of energy. Thus in the steam engine the piston compresses steam in the clearance space at the end of its stroke. that in such cases the be noted. Devices for storing and redistributing energy are very common in transmis g f 3 . A' lAA/W aa 1 sion systems. the excess supply power is stored in an accumulator. running continuously. and the energy so absorbed is returned to it during the next stroke. and it might be able. effected In the case of a single machine. as in deep mine hoisting.2. is Again. But its ability to start it and sustain the load at any point will depend on whether at that point has a mechanical advantage and not on its capacity. 5 (e). and given out again when the supply is is deficient. when the energy supplied by the steam in excess of the effort required.
wheel gives up the stored energy It at the expense of all does not necessarily follow. S. cylinder. the removed at^4' and the "driving" belt applied. were fitted to the machine. till its kinetic energy is equal to the rectangle e g h if c. however. the its velocity. and suppose its to be making return stroke. In Fig. The force just necessary to slowly move the platen may be represented by the vertical ordinates of the diagram that a greater force is abed. that heavy moving parts simply redistribute the absorbed energy as useful work. Evidently the platen will not stop at the end of the stroke the actuating force be removed at A'. and the platen would stop stored in the spring at the end of the action is The energy be returned to the platen on the forward stroke. this last case. the energy would then This latter identical with that of compression in the steamengine under the curve the U being returned to the reciprocating parts on the next stroke. that It is to be noted in even to if work of compression is not quite latter part of the equal to the energy stroke. as the work of friction during the remainder If. erly chosen the platen will just stop at the end of the stroke and the energy absorbed by the belt will equal the area / g h b. 2. of the stroke is less than the stored energy. applied. If a spring. as what is not absorbed by compression useful effort. Suppose now. moving from left to right. the return belt could be thrown stroke. so that at the position A' the platen has been accelerated . as the action may be a positive source of loss. When the effort is in excess. absorbed at the crank pin in . in order to hasten the operation. therefore. there is be absorbed during the energy is no loss of (friction neglected). at that position. the latter will slip upon the driving pulley till the excess of energy If the point A' has been propis absorbed and dissipated as heat.30 its MACHINE DESIGN kinetic energy) increased. off at A'. so that the work of com^ pression from the position A' to the end of the stroke just equalled "return" belt is the excess kinetic energy of the platen. 5 (e) let it A be the platen of a large planing machine. Fig.
Centrifugal or inertia forces. applied and removed. The weight of the part itself or of other partSc (d) (e) (f) Inertia forces due to change of velocity. A steady or dead load is one which always applied steadily in the same direction. A live load is one A suddenly applied initial velocity.CHAPTER III STRAINING ACTIONS IN MACHINE ELEMENTS 7. Nature of Forces acting in Machines. most important. may be applied to a machine in several They may and then act steadily in one direction. The pro portioning of machine elements as dictated by various methods of loading is therefore this chapter. to move the in fixed paths it is important that they should neither strong but also break or be distorted appreciably under the loads carried. may be applied first in one may be applied gradually. (g) Magnetic attractions. These ways. and will be considered in The forces acting on a machine element may be one or several of the following: (a) The useful load due to the energy transmitted. Forces due to change of temperature. members must be not only stiff. 31 which load is is alternately one imposed instantaneously but without If the load is ap . clear that From the fore machine members which transmit going chapter it is energy are subjected to forces of a varying character and intenSince the various parts of a machine must be constrained sity. they they may act in termittently in one direction. as in forces or loads electrical machinery. or they direction in the reverse. (b) (c) Forces due to frictional resistances. that is. is or suddenly in the nature of a shock.
are set is up within the material and resist the deformation. sets of interacting molecules.32 MACHINE DESIGN blow from a falling plied with initial velocity as in the case of a body. Since more or less elastic a machine element must change its form to some extent whenever subjected This change of form may be very small and tempoto a load. discussing homogeneous metals. and tion. between which the stress is should be noted that the term fibre used in a conventional sense when steel. flexure. Torsion is a Owing is to the frequent occur convenient to treat these as elementary forms of stress. the 8. and Strain. or if the load applied be heavy enough the element may even be ruptured. A given load may produce tension. and flexure are essen molecular actions normal to the planes separating adjacent that is. rary. member is subjected to impact. pressed. The primary is straining effect of shearing displacement of * It and torsional actions adjacent molecules. shearing. The results character of the straining action and of the stress which from a given load depend upon the direction and point of upon the form. is Stress the term applied to this internal reaction and to be clearly distinguished from strain. application of the load (or forces). stress being in the nature of a force and strain being a dimension. the stresses increase or de crease the distances between these molecules along lines connect ing them. The tially stresses due to tension. strain. Such Nature of Straining Actions. compression. compression. Stress. is called a When a machine member is thus distorted under a load certain molecular reactions. all materials of construction are change of form. such as iron and . it may be a permanent distortion. rence of flexure and torsion it same body. the posiand the arrangement of the supports of the member. in different fibres* of the special form of shearing stress. stresses Flexure is a combination of tensile and compressive between different sets of molecules. equal and opposite to the load applied. pression cannot both exist at the Of course tension and comsame time between any pair of or. whether temporary or permanent. or torsion or a combination of these. as it is often ex molecules.
These relations have been deter mined experimentally neering. normal or direct. Such stresses are called Compound Stresses is and will be more fully treated later. its original shape. uniform shear the interacting molecules atively with a rectilinear translation. move or are strained rel In torsional action the ad jacent molecules each side of a plane of stress have a relative motion or strain about an in reality only axis. But above a strain usually increases* faster than stress. When a load applied to a piece of material the strain which results is a function of the load terial involved. page 201. the curve so formed is sensibly a straight line. A brief reflection will show that two kinds of strain exist. 6. (See Church's Mechanics. is finally Ordinary rubber stress increases. an exception to this general rule. desirable to treat the special cases previously men elementary stresses. If axes O X and O Y are chosen and the stresses plotted as ordinates and strains as abscissas. namely. ject fully. is and of the character of the ma In general for a given loading the deformation different for different materials but constant in its relation to stress for any one material. either in tension or compression. it will be found that up to a If a bar of metal the strain caused certain point as a. this relation till to strain.) Machine members are often subjected to combinations of these simple stresses. such a diagram is called a stressstrain diagram. as flexure and torsion. strain ckcreaSi k as 3 . for all the ordinary materials used in engiof materials treat of the sub and works on mechanics Enough will be inserted here to make the discussion complete. separately as But for convenience it is much more tioned. that is. and tangential or shearing. In a similar way only two corresponding kinds of stress are met with.STRAINING ACTIONS IN MACHINE ELEMENTS acts. is tested under an increasing tensile load and by each successive load is accurately observed the relation between stress and strain can be shown graphically as at O ade Fig. stress is proportional stress is released. Further. tangentially to the 33 In planes separating such molecules. elongation (con traction if negative) and shearing. if at any point below a the the piece returns to ceases. namely.
it is MACHINE DESIGN If at any point beyond a the stress is released. If sufficient tensile stress is applied to a test piece its elonga tion increases until finally it " necks down" at its weakest point and rupture occurs.34 rupture occurs. however. breaks is The is load per unit area under which a bar called its ultimate strength and the corresponding stress or load per unit area called the ultimate stress. If at any point on the curve below a the is stress be divided by all the strain a ratio obtained which is is constant for points of below a. found that the piece no longer returns to its original dimen sions but has been permanently distorted. The stress at which a member is designed to . this modulus a given material. Cast iron has. providing ing to the point a. Similar phe nomena are observed when a piece is tested in compression or It is evident that the working stress of a machine member must be less than the elastic limit if the piece is to retain permanency of form. limit and little permanent elongation. called the modulus or coefficient of elasticity is elasticity. the strain corresponding to known for any given load may be calculated. therefore. 6. no welldefined Fig. This ratio If. Materials of this kind are said to be brittle. is it does not exceed the value correspond The point a called the elastic limit and is welldefined in elastic most materials.
the formula for strength dictates a piece of uniform cross section without regard the Hence to any particular form. stress be especially noted that not only must the working in the member be kept below be. Let p be the stress in the section. P the load. exist A short discussion will stress. now be the relations which between load. is The relation which between them in simple tension I > if = A T and I W the length of the And E be the coefficient of total elongation elasticity is member. avoiding . most convenient or cheapest form would be used. In this discussion it will be assumed that the load is a dead load applied without shock. of the piece. to carry the load with perfect safety. the given by the equation ifi The elongation per unit of length or the strain <> = A y. it but also so low that the resulting strain. and the modifying effect of suddenly applied and repeated loads will be considered after the fundamental relations between load and cases most often stress are established. and yet distort so badly under the load as to it Both strength and will dictate the stiffness should therefore be kept in mind in designing a machine part. and strain for the met and of their bearing on the selection of the form and size of a machine member. a tension member is to be designed to join two machine parts. If. 9. shall the value where permanent deforma tion takes place. as sometimes one and sometimes the other dimensions to form and given of be used.STRAINING ACTIONS IN MACHINE ELEMENTS be operated It is to is 35 called the working stress is and the ratio of the ultimate stress to the working stress called the factor of safety. then. exists and A the area of cross section. Tension. whatever may be so small as not to destroy the proper alignment through distortion. or cause unnecessary friction A machine member may be amply strong enough render unfit for the service' desired.
000 to lbs. and load are similar to those for tension or >s=T 12.. Torsion.000 In general. the remarks of the last paragraph apply equally well if the member can be is considered a short its column. .000 = .PI — A  = X 40 h. let the allowable stress p = = £ 30.. If the member subjected to simple shear the expressions for the relations existing between the stress. If the design is based on allowable stress alone. where tension members are of any considerable length and distortion under load is of importance. = arm of load in inches. is often the case in the two surfaces within machine tools where accuis If the tension member long it more than below the reduce desirable. in. one whose length If longer not greater than six times it least diameter. . Shear. i. Suppose racy is it is to hold certain limits. If the < C) member is subjected to a torsional stress.000 10. elastic limit though the working stress and a greater area may be necessary may yield may be well to a to the desired value. (See Art. therefore. 10.. let Let P = 20. as desired. they should be checked as above. 2 square inches. area. the following relations exist Let P a =load applied in pounds. Compression.000 = . .001".000. let = 4o".000. A = _ But for A P = — p A = 20.000. 10. 30.001 X = 26 sq. Example.000 lbs.e.) is 11.: 36 thin. 20. If the member under consideration be subjected to compression. and let it be required / keep A within .001. MACHINE DESIGN wide sections where concentrated required is stress at the edge might cause undue strain. than this must be considered its as a long column and the conditions governing design will be more fully treated hereafter. 20.
p s = shearing stress in lbs. fibre in inches. (D) **£ — df) T= p^W i64 » {F) For a hollow circular section whose outside and inside diameters are d x and d 2 respectively. hence torsional stiffness directly proportional to the fourth power of the outer diameter. Pa=T = £* For a circular shaft of solid section. . . . and for a hollow circular section. 32  r/ £ 8 (fli —d 4 2 ) An inspection of equation is (Z>) shows that the torsional resistance for a given stress of inertia divided proportional to the polar moment outer fibre. which is common case. . . for Examination by the distance from the neutral axis to the of equations (E) and (F) shows that strength is circular sections torsional proportional to the third power of the outer diameter. T— twisting s moment applied to member in inch pounds. For deformation under the most stress for a solid circular section.STRAINING ACTIONS IN MACHINE ELEMENTS 37 Let / p = polar moment of inertia of the section in biquadratic inches. at outer fibre. is Equations (G) and (H) show that torsional deformation inversely proportional to the is fourth power of the outer diameter. E = transverse coefficient of elasticity. per unit area e = distance from neutral axis to outer / = length of member in inches. Then for torsional strength in general. = angle of deformation in radians. For a given amount of material that section in which this ma terial is distributed farthest from the gravity axis will be strongest .
a Example. bar A circular cast iron boring bar 60 inches long carries a solid circular boring is head 60 inches in diameter. 8. The cost of pattern work little is about the same. The subjected to a torsional moment of 60. while the greater. in general." Fig. work in the foundry is. The thickness of the walls being thinner in hollow than in solid also lies. in general. sional strains. or shearing. 7. compression. 7. the box section flexure well adapted to resist sides of a combined and torsion. for hollow sections as for I or other sections. and the box section. The hollow circular and hollow rectangular sections. while equally good for simple stresses. Fig. tion of stresses.38 MACHINE DESIGN stiffest as long as the walls of the section do not become so and weak as to yield locally from other causes.000 inch pounds . is. The box section is peculiarly useful in machine construction. therefore. The box section also afford facilities for attaching auxiliary parts and its appearance is one of strength and staflat bility. as many machine members must carry a combinaMachine frames may be subjected to tension. as has been noted. more skin where the greatest strength to of cast iron An advantage not be overlooked in some lines of work which hollow sections can be strengthened by increasing the thickness of the walls by changing the core withis the ease with out changing the external dimensions. forms insures a better quality of metal in castings and surface. are best adapted. to resist tor Fig. is Furthermore. combined with torsion. vastly superior in torsion. commonly called the thin and "box section.
000. a* = 32 X 6.STRAINING ACTIONS IN MACHINE ELEMENTS which is 39 applied at one end. from G. The * area of the hollow shaft = 31. in order to prevent chattering What should be the diameter of the bar if the working stress be taken as 3. its entire length.6 rr hence d2 sq. dS Substituting in = ^4" and df —d = ^f d* 256 256 2 H.000 60 X 60.000 pounds per square inch and E s be taken as 6. For torsional strength from formula E. 100 .000 X 16 = 3.6". .*. = — 3° = r giro" since is in radians and the length of an arc = where = • radius. .\d = For torsional stiffness * r0. dt 4 4 4 whence.000 — X— — — x = 5. = 7.550 and d x = 9.2o6X (Angular deflection in radians). d 3 = 60.000.000.000x60 itit^u256 x X 6. dS = 8.000.67 of a shaft while the area of The angular deflection or twist in degrces=57. In this case either the inside or outside diameter or the ratio between them can be assumed. It is desired to keep the torsional deflection of the tool below gV" when the bar is transmitting power through of the tool. . gio .870 cnhr hence if </ = 8. " y* 60.000 . If the section is made hollow Let less metal can be used. therefore the last value would be used.000 X4 .2".8". It is evident that the shaft will be amply strong designed for stiffness.. in.
In the cases of simple loading just discussed only one form of stress is brought on the safely based member on this and the design of the crosssection can be stress. When a beam is subjected to simple bending (a) the principal stresses that are induced are side of the neutral axis. however. inch. stiffness. Compound Stresses. A = deflection at any point in inches. by using the hollow section. beam at right Generally speaking. that they may be be borne in mind. but regard must be had to the combinaIn many cases one or more of tion of stresses that may occur. so that with a small increase in diameter one half the metal secures. based may be called the predominating or primary stress and may be a simple stress or a combination of simple The latter will be called a Compound Stress.— 40 the solid shaft MACHINE DESIGN = 60. The predominating stress will be the tension or compression depend ing on the material and the form of section. the loads applied induce stresses of several kinds. Let When a beam is subjected to simple flexure. in designing the is such.. the shearing stress small com pared with the tension or compression and can often be neglected. and where the beam is designed to withstand the bending moment only. (b) a a tension on one compression on the other side of angles to the tension and comis the neutral axis. When. / = moment of inertia of section in biquadratic inches. It must never be forgotten. 14. per sq. = maximum E — coefficient stress at outer fibre in lbs. care should be exercised that the sections which are subjected to a small bending moment are not made so small as to yield under shear. . though they should always The stress on which the design of the memstresses. of elasticity. Flexure. e = distance from neutral axis to outermost fibre in inches. in. ber is it member. it is no longer possible in general to base the design on any one stress.84 sq. M = bending moment at any section in inch pounds. the stresses are so small. P p —load applied in pounds. however. the same 13. or their action neglected. and (c) a shearing stress which acts on every section of the pression.
whatever the shape of beam may be. in general. as in other classes of machine members. depending on the which it is supported. used as a track for a crane trolley carrying 4.000 From Table I. M must be expressed The ordinate in terms of x and the expression integrated twice. can then be found for any value of x and its greatest value is the maximum deflection. d2 .000 X 240 3 16 whence * I = 4. and a maximum more than 8. .000 — E I 48 X X 240 X 16 X 3 30. which is the deflection.y r (If Xz = ^— Ht 1 M To y. . pi * M= — e . The the general equation of the elastic curve. This integration in practice.000. within the elastic limit. ends is A steel / beam 20 ft. the design of the part may be based on strength or stiffness.— STRAINING ACTIONS IN MACHINE ELEMENTS .000 Select a standard rolled / beam that will carry the load with a at the centre deflection of not stress of not more than yV lbs. 41 Then for strength. . The curve assumed by a beam loaded within the elastic limit is called the elastic curve and is of shape of its crosssection. depending on the conditions. or manner any of support is. long and supported at the lbs. however. is applicable only to symmetrical sections as e may have two values for other — * / . 30. the material. that this ex pression sections. Here. should he noted. 3" PP 48 3 4. U) Every beam when loaded deflects somewhat. and the load applied. . . is termed the resisting moment.000 — sometimes called the modulus of the section and It is generally indicated by the letter Z. has been performed for all the cases usually met with It is to and the results are tabulated in Table I. and in general both should be considered. Example.000. find the particular equation for case. the in way course different for different combinations of the above conditions. the load.000 X / 48 The expression X is = 20<:. be noted that for stresses this tabulation is for beams of uniform section and within the elastic limit. .
per foot is Let such a beam be chosen.* is Sometimes desirable to have the crosssection uniform. to resist the bending moment and no forms most usually met In the latter cases the shearing stress must be looked after Table II gives a few of the and an example may make their application clear. varies and therefore. .= / 205 is satisfactory.000 other way graphical Then when # = 30. carefully. What is its vertical outline so as to have uniform strength ? Let the thickness = b and the variable height = y. hence Px = which is —— — = pby pi e 2 6 or J y = 6Px — pb is at the equation of a parabola whose vertex let 6 the outer end let of the beam. which if. is The values in Table I all of uniform crosssection. the beam is made strong enough at its most strained it section and uniform in crosssection throughout its length will it have an excess of material at every other section. In the problem assumed = 1. and stress increases as the crosssection of the hence the shearing * This of course does not cover the possible case where the effect of shearing or other stresses may exceed that due to flexure.000 lbs. Px.000X10X12X6 = = —r. y = h = 5.5 inches and /> = 4. Example. must be equal to the resisting moment 2 of the section at each point. . nearly. section therefore 15. „ 7.5 lbs.— MACHINE DESIGN 42 From handbooks on structural shapes it is found that the moment of inertia of a 12" I beam weighing 31. lbs. In nearly cases the bending moment. The shearing load at any point is P.000 lbs. Then from formula J. usually the basis of design. form thickness. refer Beams to beams of Uniform Strength.$".8. while in other cases the metal can be so distributed that every section shall have the necessary strength more. I. the 215. The _. Table Then 2) is the moment and this at any section at a distance x (Fig. stress p Me 2. A cantilever of rectangular section 30 inches long It is to have a unicarries at its outer end a load of 1. In a similar points may be found or the curve may be laid out by method.
Combined Flexure and Torsion. therefore the outer end of the when x member Refer must be modified so as ence will be to safely carry the shearing stress. It can be shown that if a bar or rod subjected to a longi tudinal tensile or compressive stress and at the same time to a shearing stress at right angles to these stresses its length. Fig. that. 16). Let the force P. y very small.. and can be neglected. thickness of the Thus in the cantilever example above. the theoretical out do not apply. 16." page 317. if the beam is not kept uniform the outline for uniform uniform strength is not a parabola. made to this again under the section dealing with It is to machine attachments (see chap. the predominatis ing stress therefore that due to the combined action is of the bending and twisting moments. by bending moment and shearing load the correct will depth of section can be found for a number of points and a curve plotted that strength. y = o. STRAINING ACTIONS IN MACHINE ELEMENTS 43 beam is decreases. is When v = o. and in general small. when the thickness is not The mistake of using a parabola is often made when I ox T It is sections are used instead of uniform thickness or depth. be especially noted that these theoretical shapes are based on certain assumptions and unless these are observed lines in the design. upon a rod with an arm a at a distance from the support equal answer the requirements of uniform act to /. evident. Then (a) the stresses induced in the section close to the support are flexure due to the bending moment PI twisting direct load (b) (c) torsion " " " " " Pa and equal to shearing " " is P — due The to shearing stress usually very small compared to that bending and twisting. . the combination of may produce similar stresses greater than either and acting along planes other than those along which the original stresses act. whatever means of the may be the form of section adopted. 8.* * Church's " Mechanics.
P= concentrated load. Bending Moments and Shear Greatest I BEAMS OF UNIFORM SECTION Location Greatest Location of Bending Maximum Section Moment Deflection M M A of Shearing where Shear Force is Max. W= total distributed load. PI PI 3 Any 3EI PiMP. .44 MACHINE DESIGN TABLE Diagram of Loads.451 BtoC EI from +— PI' A CtoA wl 128 2 > c 3wl wl 3 185 EI 8 * B 5wl •w=load per unit length.is ii Px + P2 From CtoB  >oooooocpA w 1 VI 3 wl = W 8EI wi 1 +P1 ElL3 + wl 8j +P _ 5 PI Pt 3 At 107 0.
Bending Momenta and Shear Bending Moment Greatest Location Greatest Location of of Deflection Maximum Shearing Force 3ection M M A is where Sheor Max.4231 from B T J T t. g. P= concentrated load.I. 5 ^ <D I PI I 1 At C CtoA l 3IEI Not Max. . W= total distributed load. PI VII 1** ooo AandB 12 Equal 2 E] wl 3 At B VIII ^£ ®^s • wl At At A Wl 24 3 ooo I_ EI W 0. a j£ . <g> IX 4) O PI 4 Pi 43 BtoC CtroA EI PI2 BtoC 3 O Pill. 51 CO * O fl B C .STRAINING ACTIONS IN MACHINE ELEMENTS 45 TABLE I—Continued Diagram of Loads.3 PA PI) BtoC Centre C to D 24 [31 EI 2 XI 83 1 4lf] DtoA B ooorfoonnA wl 1 swr Centre wl Centre 3*4EI AorB xii L * o u'=load per unit length.
2 a o a P " XIII a s 02 48"EI 2 + W 2 ©. TF= total distributed load. E m A ^N 9Q00 ooooofo 9 ?M Wa J E CandD W S5 XVI h at 131 "Ii PI A. atB s3 at^C 2PIilo 3EI(It2.46 MACHINE DESIGN TABLE 1— Continued Diagram of Loads Bending Moments and Shear Greatest Loca Greatest Loca Maximum Shearing Stress Bending tion of Deflection tion of Moment M M A B OOOOiOOOO <P A (p+fw)x [P+4] J . i A 2 >=load per unit length. and C.l) at XVIII I 3 pM. . ka> — 1_.B. P= concentrated load. PI 192 EI 3 nooo^oooo A r 1 3 Hi AorB 12 384 wV EI i XVII Phi. i C 5 Pa DtoC XIV bf B D.
' Centre ' 3P1 J SEbd^ 2 Diagrams same as No.Maximum Section Bending tion of Deflection tion of Shearing where Moment Stress Shear is Max. !t! o i flP P A $ = PI 3 2Ebds X Table I Centn as No.1 o. IX Table I w=\oa.STRAINING ACTIONS IN MACHINE ELEMENTS 47 TABLE Outline of II BEAMS OF UNIFORM STRENGTH Greatest Beam Loca Greatest Loca. Diagrams as ajp in No. P= concentrated load. W= total distributed load. Diagrams same XII Table 2i= I r VII I. . Plan 5 SO 3 3 a g fcl 6P1 3 Pi Ebh 3 Diagram of s3 Moments and Shear same as I a. M M A A B u *—'& Is « c3 3 +3 «o 8P1 3 PI ® o <o3 a Ebh s Pi © O 11 03 i h o . Diagrams as in I 3 >.j. < a> b'i e 0. Ill Table I t^MI Diagrams same as No.d per unit length. ipoopooooc) / fi r3 1 : a O 2 w when 1]*=1 .
.. v 3^0.000 v. since p = 2.600 lbs. the crosssection of the shaft. per stress square inch and the allowable shearing required to design p e is 1. then the maximum compressive stress p due to t and s is given by the following equation p= P. twisting A certain section of a circular cast iron shaft subjected to a bending moment M of 10. as it usually the case.000 inch lbs. But should the allowable shearing strength of the is material be less than the tensile strength. I = s Me = T / 32 M 3 Td? 16 = 32 X 7t 10./> = 300. The It is allowable tensile stress p.000 = 100.000 _ .000 or d = . and a moment T . and s is (i) and the maximum shearing stress I to / = V~f+^~s2 (2) It is evident that the numerical value of p will always exceed thai of p a and therefore if the material used has approximately the same tensile and shearing strength the design can be safely based on (1). 1. h[t + VT+~4?\ p due s . . d3 300. of 60.600. From .000 hencefrom(i). may happen that the shearing stress p s as found by (2) would (1).000 . and from D.600 . per square inch. .. From(2). = Te p T —=— „ = *d 16 X X 60." — — and . is Example./>. /. .000 lbs.55" _ . .000 inch lbs. . 5..: 48 If / MACHINE DESIGN be the greatest tensile or direct tensile or compressive stress and s the greatest direct shearing stress applied to the bar. ds since re p  = . dictate a larger section than that required If the tensile stress is by p as found by due to a bending moment and the shearthe values of ing stress is due to a twisting moment be found from equations J and obtained as above in equations (1) D respectively 5 and / can and p and p i and (2) respectively.000 = ~^ — and .000 = ^— 1. is 2.d3 r„ d3 ~J 3 d 2..
(Since I p T = — r  = r = 2l for circular or other sections for which the mo ments of inertia about two perpendicular axes are equal. . where any particular problem be a known quantity. • (KJ The quantities M x and for T are usually large and the numerical will work involved in solving K and K { can be simplified by writing M=x T. M= T = = the bending the twisting moment producing the direct stress moment producing the shearing stress /. (K) In a similar manner an equivalent twisting moment can be deduced from (2) thus. lt P 2 L r \ e r 2 r yi * .. M = — and M // r = fe £. From J. Equations (1) and (2) are general. whence P1 r *P 2 7 . TTy/j +1 (/Q . but for circular shafts operating under conditions that or ideal bending as follows. produce both bending and twisting it has been found convenient to make use of what may be called an equivalent moment which may be derived from equation (1) Let Let " M = the t equivalent bending which will produce the maximum direct stress p.\d = $. 5. (^2) reduces to. It is that the last value should be taken. .8" or 49 evident % n greater than that given by (1).) Multiply equation (1) 2 through by . 2pJ r VM +T 2 2 .. . Whence and K reduces to. pi = M = yM + v M + T 2 2 . M K { e = y 2 T[x + v/?+"i] . and applicable to any and all sections.. . " r radius of shaft. r pi and from D.STRAINING ACTIONS IN MACHINE ELEMENTS ..
whether equation (3) or equation (4) will give the greater value of d without the necessity of solving both equations. It is.. desirable to know. (3) In a similar way from E Since in any given problem M and T are always known. /.. for any since the allowable values of set of Conditions. M K s e and T e can always be found from and K x (or K 2 and K 3) and p and p can always be assigned. conditions each should be used for designing in order that the maximum From diameter of shaft shall be found in all cases. and it remains to consider their applica It has been pointed out that by equation (i) does not necessarily indicate that a larger section will result from its adoption than would result from the use of equation (2). the greater numerical value given M e . the diameter of the shaft d can always be determined from both equations (3) and (4) and the larger value selected as in the problem previously solved. For the same obtained from K may reasons the greater numerical value of not give a larger section than would be obtained from T by applying K It is necessary therefore to determine under what tion to the design of new shafts.50 It is to MACHINE DESIGN be especially noted that M e and T e are equivalent mo ments in a numerical sense only. M d3 e = 32 Whence = pnd V — =V M =—X pI r 3 32 e 16 2 M e . that is. however. . and Te is moment which a stress equal to the maximum shearing stress ps reference being made to the same section.. will give a stress equal to the maximum a twisting resultant resultant tensile or compressive stress p. if a bending moment M and a twisting moment T are applied to a shaft. . tensile stress t producing a and a shearing stress s respectively. then M e is a bending moment which will give . . e x . The application of these equations to the investigation of any existing shaft subjected to a bending moment M and a twisting moment T is obvious.
the relations as given in K 2 and K z be substi tuted for those in K and K y x the equation = = = becomes (6) which is the equation of a curve expressing all the simultaneous values of y (or — S J and x (or —) for which equations K 2 and K^ will give It is equal diameters of shaft. desirable before plotting the curve to examine the limits between which x and y may vary. STRAINING ACTIONS IN MACHINE ELEMENTS It is 5 evident that in order that equations (3) and or 2 (4) ° may give the same diameter of shaft —^— must e equal —i p 6 P M = e — p  and that for conditions other than these. " give values of d equal H when P» — = s T e ' p 2M or e if p — s 8 be called then p 2 2 T J 2 is e I e VM +T M + VM +T 2 2 ° the simultaneous the equation of a curve which expresses all values of p — s and d. either equation (3) or equation (4) may give the greater diameter. It is therefore neces sary to investigate the relations existing between T ° and — p for three sets of conditions. M and T is to equation (5). (1) (2) When " equations (3) and (4) will give equal values of d. It is clear that for x = o. and . —T ~ e for which equations (3) and (4) will give equal values of The value of either M or T in equation way plot their ratio. (1) It has already been shown that equations (3) and (4) will y. and for T = o x= 00 hence the limits of x are o.   1 . M=o <*> . 5 may vary from zero to infinity and the most convenient simultaneous values of in of plotting If then. equation (3) will give the greatest value of d.
1 2.9 1 1.7 1.8 .4 1.6 . o. on them will give the largest value.1 1.2 A .9 2 2. 9. For has just been this shown that M e can never be less than T — 2 e and only equals .5 1.6 1.2 1. Further.52 MACHINE DESIGN for Using these same limits that M and y T in equation (5) it is found when and when M= T= o. 1 The curve giving these simultaneous values shown (2) in Fig.7 . for values of P — equal to or greater than unity. M y = \ and M = 1 and T =— e f e = T e That is for all materials where the 1 ratio of allowable shearing to tensile stress lies between neous values of equal values of is M and T for d. ratio of M be — taken greater than the simultaneous value given by the curve (or in other words if the coordinates chosen intersect above the curve) equation (3) will give the largest value of d. equait P tion (3) will also give the largest value of d.8 1. 9 and has been plotted from equation If for any given value of — within the limits c f and /a x 2 Field r K 2 M e~ 4U F: wx + I vr eld of K> lii I.rV .2 x T Fig. which equations and yi there are always simulta(3) and (4) will give (6).3 1.1 I .5 . For the value of relatively to — can be increased only by making M greater 1 and an examination of K and K shows that inHence increases K more rapidly than it does K creasing in such cases K (or K 2 ) applies and equation (3) which is based T M x .
that the pressure on the crank pin may be equal to 100 unbalanced pressure per square inch of the piston when is the connectingrod perpendicular to the crank radius. which greatest value of d. 16" in per square The centre of the crank pin overhangs the centre of the parallel to the axis of shaft). Equation K . which based on K (or (3) K 2) . (piston lbs. An engine cylinder is i6"X24 diameter and stroke of 24"). therefore. is Equations K 2 and K 3 are the most convenient It forms of equivalent moments and will be used in this work. Allow . steam pressure = 100 inch. to be particularly noted that they are applicable only to circu lar or other sections where the polar moment of inertia is equal to the sum of the rectangular dicular axes (see page 50). is moments of inertia around perpenWhere the section to be considered more complex the solution must be based on the original equations (1) and (2) in a similar manner to that employed in the example on page 49. based on K P x (or K 3 ).STRAINING ACTIONS IN MACHINE ELEMENTS 53 when M= o. 3 should be used where the simultaneous values of p — s ! and P the curve. and evident from equations e  and that for values of p s = p and T equation M>— e (3). In a similar l KIT way it can be shown that for all simultaneous values of — and = which intersects below the curve and within T P the limits y =i and y = >£. For all finite it is values of M. Equation K 2 should be used where the simultaneous values of M intersect above the curve which — and — T Ps P is always the case whenever Ps — > 1. (3) M is t must (4) be greater than — . M intersect below — T // which is always the case whenever Ps — <. or for is all materials where Ps — i is less than }4 } equation (4). will give the greatest value of d. P Example 1. will give the Summary. main journal by 15" (measured Assume lbs.
Area a of piston = 200 sq.8y. and . arm of bending moment lbs.54 MACHINE DESIGN ing 8. however.25 the intersection of the ordinates falls below the curve. that K 2 should be used. 250.400 p s + = 1.000 y° D = 257.25 the ordinates intersect above the curve.500 X 1. —A circular cast iron shaft subjected to a twist inch lbs. inch stress 1. M 3 M d e = = J [1. moment of 250. and a bending moment of 62.000 inch 17 From / \ (3).8 referring to Fig.000 inch lbs.25) + 1] 250. & + . hence K z should be Used.7 From # = the curve.000 it —= .7 and x is = 62.500 d = I6X 257. 6. 2 e Then T = [v d3 7T (.25 + \/ 1.500 2. per sq.400 as the maximum allowable shearing compute the diameter of the shaft. Vv3/ ' = = — 342.000 8.000 6.000 lbs. inches.500 The allowable tensile stress is 2. and the allowable shearing diameter of the shaft.000 seen that for y Ps . Fig. 9. radius of crank (arm of maximum twisting moment) 100 r = = = 12". 9 it is seen that f or y Ps = — . ^ 758^ is Example ing 2.= = — .= and # = P 1. .400 — 9. Suppose. .25 2 lbs. From K 2.25.000 pounds as the maximum allowable direct stress and stress.400 = . = 15" T= 200 200 15 M= M * = = By X X 100 12 X X 12 15 240.400 lbs.75 inches.000 X — 32 = X * 435 . 1] 240.000 342.'.000 inch Also 300. hence K 2 should be used.25. Determine the Here y A = —— p 1.
For equation K lay off be = Ob = x along Then ta + a b = ac = v + V.25 + V(. — Equation K for is sometimes trans formed into an equivalent twisting moment.STRAINING ACTIONS IN MACHINE ELEMENTS Then. unity. Since in general stress (that it is M is. . 2 2 ab. the extension of 2 . then ab 2 e 1 . yi T (a twisting moment equivalent to the combined bending and twisting moments) provided the same allowable . Other Formulas.4 inches or . 32 X l60 00 ° 2. 17..000 *= .000 = 160.35" less than the value given by K 2.v + 1 Hence M = K [>c X T]. r = 2 M = M + s/W+T (K<) . direct stress is used with T in solving for the diameter of shaft.*. A 9 (a) convenient graphical solution of K 2 and K s is shown in Fig. equivalent e combined bending and twisting moment). lay off Ob = x to the it vertical Draw ab extending same beyond scale b on a for length somewhat greater than x.000 = 830 d = 9. = \/ Ob + Oa = Vx + hence T = a b X T.25) + 2  1] 250. which may be used as follows: 3 For the K make Oa = axis. = p) T = 2M for the same If therefore con sidered more convenient to use an equivalent twisting moment instead of an equivalent bending stitute for moment it is allowable to subto the M t (the bending moment. 2. = pB pi — and r T = 2  PJ an equal intensity of section. 55 M e = J [.
The same little evidently applies to equation (8) which differs but from Grashof's.35 M + o . is when the shearing stress predominates. . except where the bending moment is very small. and any variation is more than covered by the factor of safety which must be used. It will (8) be found that for the range where equations and be give values greater than K K 2. Grashof's formula gives a value of Af e 25 per cent greater than that given by it K 2. Take for example steel where — : = .. Other authorities gives (7) give slightly different coefficients.88 J seen that the difference between d as determined is by K 3 and Grashof's formula negligible. M is equal to zero. the difference between these equivamoments decreases. But may be noted is less that in general for all materials whose shearing is is strength for in than their tensile strength (and this the case most materials used in engineering) that when other words. MACHINE DESIGN formula for K 2 K 3 and K are all combined bending and 4 different forms of Rankine's twisting. Thus Grashof K While others give = =  M + I VM K 2 + T 2 . + T 2 .84 * \ p — T P* — — from Grashof's formula from which it is d 1. safer to use K 3 in preference to (7) K 2 . M small it or.8 and x in — which down close to the limit to where Grashof's 2. . 2 M e 0.77 3 from 3 d = = 1.i 9 formula K is 3 .65 VM 2. from K K «y ^ N p \ 2 d = 1.. .56 Equations . As the value lent bending of x increases. formula gives the greatest value compared terms of K Expressing d 3 T as in equations (3) and 3 (4). that these values will still less than those obtained from 3 or at least not enough greater to warrant the use of a different in place of .. (8) The diameter of not differ much from shaft given by equations (7) and (8) will that given by for any set of conditions. At the limit where the bending moment .
Propeller shafts and vertical shafts carrying considerable weight are subjected to combined twist and thrust. action of the thrust frequently so small that the shaft may be considered as subjected to simple compression. or distance is between bearings. T = the twisting moment on the shaft. maximum resultant com and p the t maximum _j_ resultant shearing stress. and 5 — = = the 2 / the intensity of shearing stress due T  — r P . 4 I ^ l6 T * d3 for solid circular shafts. r = % d radius of the shaft. then p = _ c+ 2 vV 2 ^ s 2 ancj p 2 N /r + 4 ^ 5 2 . 57 of steamers Combined Torsion and Compression. p c the pressive stress. be the shearing stress due to T. is so far as the concerned.STRAINING ACTIONS IN MACHINE ELEMENTS 18. is The intensity of this compressive stress in such cases in which If P = the thrust. or be the compressive stress due to P. / = the polar moment then of inertia (= 2 times the rectangular moment to T. and d = the diameter of the (solid cir cular) shaft. hence equations combined bending and twisting and (2) of the preceding article apply and may be used to find the if c maximum compressive or max5 imum shearing stress. of inertia). 17). The resultant maximum stresses are those due to the com bined actions of a normal stress (compression) and a tangential stress (shear) as in the case of (1) (Art.*li ~ d S ' ' * IA. The span.
. The line distribution of these in Fig. and pc and p the 20). In the case shown the combined compressive or combined tensile stresses at X are the greatest which may come on the section. both cases is shown graphically above the where tensile stresses are plotted UV and the compress ive stress below. the trial diameter of the shaft compressive solving for may be shaft taken so as to bring the mean after stress c well a below the allowable value. the stresses induced in the section will be:— (a) A uniformly distributed stress will due to the load P and equal to P — A. Fig. stress on the side toward the load The maxi mum stress will be the greatest algebraic sum of these combined stresses at the outer fibres at stresses for X or F. tensile. 10. per unit area. may it also it be checked as a long is column stress (Art. and it assume a and p to s trial see that they diameter d and then check for the values of pc do not exceed the allowable compressive and shearing If the stresses of the material under consideration. 10. will the ordinates under r s representing the flexural and those under m n the show where the algebraic sum direct stresses. depending on the . stresses. (b) A flexural stress due to the bending moment P a. This other. induced in the section by the load P is is then the flexural stress on the side toward the load If the direct stress induced is is compressive.58 MACHINE DESIGN d for a given value of p c or p is much more convenient to s It is difficult to find the value of from the above equations. flexural stress will be a tensile stress on one side of the gravity axis which is at right angles to a. is An inspection greatest. be acted on by a force P at a distance from its 19. is span of the shaft between bearings so great that the shaft must be considered as a column likely to buckle. If the section XY. but s well to check the shearing p against the allowable stress by applying (Z. In steel shafting is necessary usually to apply equation (L) only. depending on the direction of P. the flexural compressive. and compressive on the If the direct stress tensile.). Combined with Direct Stress. This be tensile or compressive. Flexure gravity axis O equal to a.
e This increases on the compression side. but If the material equally strong in tension and compression the gravity axis should not be far it from where cast iron is used is advantageous to distribute the metal more toward the tension side. thus drawing the gravity axis toward that side. The form of section and location of the gravity axis should be fixed with reference to the relative tensile Let and compressive strength p' = the direct stress due of the material used. io (b). p = Therefore. will 59 This if is not necessarily so. as a brief reflection show that O be located near enough to X the reverse of the above conditions may exist. Let p" Let p = = the tensile or compressive stress the due to P a. Fig. to P.— STRAINING ACTIONS IN MACHINE ELEMENTS direction of P. where used e is the distance from is O to the outer fibre at either X or Y.f = — ." . P ae P — and from formula J. P = f+ P Pae*   (M) Fig. 36. *See Church's "Mechanics. and hence increases the p. io (a). depending on which is under consideration. maximum stress in the section at f X or Then from formula A. central. Y.
A punching machine Maximum force P acting 12) has a reach of at the punch is taken at fibre 70. the neutral axis is found to be 8" from n. pAI n P_ I + Aae 3. is or in general where only one dimension unknown.000X35X7. or circular sections. clear swing of 28 11 (b) . The section at m n is shown by Fig.6o compressive stress. is but the algebraic ex pression to a troublesome cubic equation.'. . The practical way is assume a trial section and check this for P or Example i. not practicable. Design the section mn so that the maximum stress at n (tension) shall be about 2. in general. A small crane (Fig. With solid. 7*3 9. MACHINE DESIGN It decreases e stress. square.000 lbs. per square inch at n. to solve equation (M) for th& It is direct determination of the dimensions of a cross section to sus tain a given eccentric load p.. load corresponding to a p. P with an assigned intensity of stress because both A. 12 (b). . and check the compressive stress at m. p = Pae P + I A 15 — . 2. It is also found that A — 216 and I = 7. and hence decreases the tensile Cast iron much stronger in compression than in tension. Find the maximum fibre stress (compression) of 9. e are and tions result. 12 (6).400 pounds per square inch. functions of the required dimenand with any but the simplest sections complicated funcI. The general form of section best adapted to this case is that shown in Fig. sions. = 3. . / = 28 + 2=30 = _i_(2 X 64 A = 2 X 4 — 15 X + 27) = 7.000 lbs.5 rp _ Example 22 inches.3 . on the tension is side. Taking the trial dimensions as in Fig.680. and therefore a greater moment can be withstood by a given crosssectional area when distributed in this manner. n) has a inches.3 =I o6olbs + 35X3° X 2 (Fig. a = 22 4 8 = 30. it is possible to reduce M to a form which can be solved.
J. Recent investigabe have pointed out the fact that curved beams. the above equation does not apply. If. 7. The above tions is theory regarding the position of the neutral axis given that in general use for such cases.STRAINING ACTIONS IN MACHINE ELEMENTS at n. Stresses Columns or Long When a short bar in is subjected to an axial compressive load the stress induced is each section simple compression (see Art.000 70. however.000 X 30 X 8 = 216 325 + 2. as the bar . 61 P and P * = at 70.200 = 2.000 1 70.680 m. Struts. this theory is not absolutely it rigid for For the usual case of design accurate.. 11. Fig. It massing the metal compressive stress evident that this can be accomplished by further toward the tension side as the still is very low.000 ' X 30 X 11 216 tensile = 7.680 slightly —3 25 + 3> 000 = 2 > 6 75 lbs  The stress is greater than the limit assigned. 20). and the value given by formula (A) or of the stress p is W P. If this excess is not considered permissible is a stronger section must be taken. the bar is more than 4 to 6 times as long as its least will. Fig. if diameter. 12. 70. is lieved that the above in is sufficiently 20.525 lbs.
62 MACHINE DESIGN will proportioned as above. m= Then a coefficient for the end conditions as shown in Table 3. the least moment of inertia of crosssection. deflect laterally under the load and ultimately break under a and lateral bending. compound stress due to compression Such a member is called a column. deduced from the theory of The coefficient m is also . the least radius of gyration of crosssection. in pounds. or the unit breaking load.. Other formulae were later developed ex perimentally by Hodgkinson and Tredgold. first Theoretical equations for the design of columns were developed by Euler. or unit stress at the yield point. r n = This P h p = the intensity of working is stress in the column ( = p the maximum intensity of intensity of stress in the column when the mean stress is p/. This is the maximum intensity of stress in the stress is column when the mean intensity of p '. Let = P = I = A = P = p = I c ' the length of the column in inches. is Euler's formula for long columns 2 7T EI It is to be especially noted that Euler's equation is rational and elasticity. =P +A c pc = the crushing strength of the material. the breaking load c the mean intensity on the column in pounds. c n = the factor of safety. for the design of this class of mem The student is referred to any good treatise on the Me chanics of Materials for a fuller discussion of these expressions than can be given in this work. the working load on the P = p' column A. = sc pi n). of stress under the breaking load. P 5 c n. the area of the crosssection in square inches. = the mean intensity of working stress. or unit working load. Gordon and Rankine have also proposed equations bers.
are to homogeneous.STRAINING ACTIONS IN MACHINE ELEMENTS rational will 63 and applicable to other forms of column is strictly formula. or yield point. As be shown later. if it is an absolutely concentric to load (that an ideal column) there seems be no reason why its strength should diminish rapidly with an increase of length. the other guided. or only slightly in excess of.'. it for when this stress is reached the metal flows. "Pin & Square' One end nixed. Ended" Both ends fixed. Very long columns may approximate the resistance as given by Euler's formula. ends free. increased beyond this limit and a deflection load is caused in any way. Very short compression members. m =. free at other. the TABLE 'VALUES OF III m FOR DIFFERENT FiND CASE. However. perfectly weaker than the former and a column is initially exactly straight. and subjected is. If a deflection is caused while the column is under a load less than this greatest load consistent .II "Pin Ended" CONDITIONS CASE. the equation applicable only to very long columns. the deflection will increase until the stress due to flexure pro duces failure of the column. Columns flexure If of lengths intermediate between compression members which yield by simple crushing and those which fail by pure stronger than the latter. of ductile material. Both. IV "Square CASE I CASE III Fixed at one end.\ m 9/i m=4' '>/////////// apparent elastic limit. but guided. although does not actually break. fail under stresses corresponding to. other conditions remaining the same. even an ideal very long column would reach the condition of unstable equilibrium when subjected is to a certain critical If the load (the greatest load consistent with stability).
x = I * P = I  \/ /  A. Other authorities give somewhat agree that most but nearly all columns in ordinary structures and ma chines are intermediate between simple compression members and those great to which Euler's formulae apply. is There have been a graphical represenin Fig. crosssection. section.000 pounds per square inch. abscissas represent ratios of the length of column to the least radius of gyration of the crosssection." it member be may be treated Johnson gives as which will yield by simple compression. formulae proposed. the stress is not uniform across the The maximum intensity of stress must be kept within the compressive strength of the material. working of moderate length. . hence no for mula should be used which would * give a value of p ' e in excess of Owing to the flexure of the long column.* terial. MACHINE DESIGN the elasticity of the material tends to make the column regain its normal form. than short struts of similar material and crosssection. The could not be expected to have a greater strength. for the ideal conditions are not realized in practice. in other words. or nominal.000 for a very short compression member. and a modulus of elasc value of p f c is 36. and y = p\ = P  A. many column A tation of several of these formulae shown In this diagram. if of ductile maif the flow at the yield point causes buckling. Merriman as one says that the length of a compression its least only from four to six times "diameter. limits within I f which the Euler formula should not be applied p= 150 for pinended. hence long struts fail under smaller loads Or. of 29.400. Even in columns and lower mean. different limits. hence the mean stress is less than for shorter compression members.000. 13. p (yield point) of 36. Or.64 with stability. The diagram is drawn for the ultimate resistance of pinended columns with a material having a crushing resistance. for equal safety under a given load long columns must have a greater stress. and it is evident that a long column ticity. in which the mean stress is more nearly equal to the maximum. and the ordinates represent the nominal (mean) intensity of compressive stress. Initial defects in the form or structure of a column or eccentric application of load tend to produce such a deflection. E. and of the = 200 for squareended columns.
13.N sRs V \H sVi V \ ^K *\ X v^L? f^ F x = 1^P 1 £ 10 K 2 100 200 300 Fig. it pear that the Euler formula (represented by the curve EE E 1 2) cannot apply in to pinended columns 90. columns stress considerably less moderate length fail under a mean than the simple crushing resistance of the material.p. not reasonable to expect such an abrupt change of law in (/ s passing this limit of = 90) . and. If (of this particular material) which I * p < columns with a ratio of / to p less than this limit yielded by simple crushing. Johnson has developed a formula which is based on the assumption that the strength of the column may be taken inversely as / s. as already stated. Thomas H. . 65 will ap Referring to the diagram. or the strength F F. and those with a greater formula.STRAINING ACTIONS IN MACHINE ELEMENTS the crushing resistance p c . This expression is tc A lc p (1) in which the coefficient k has the value. the straight line ratio of / to p followed Euler's FF x and the curve It is F E E 1 t 2 would give the laws p for all lengths of columns. of columns is inversely as some function of the length divided by the "least diameter. ! >K l\ 30 "T^k A\ >i \ ^ 20 ^\B R." Mr.
and with vertex in the axis of ordinates at F. . n. the Euler formula is be employed. This curve its a parabola tangent to the Euler curve. Johnson has derived a formula from the results very careful experiments of Considere and Tetmajer. . but agrees remarkably well with very refined experiments on breaking loads. should the columns exceed the length corresponding to this point of tangency (/ +p > 150). It does not appear to have any advantage. when some particular value of k does not apply to several computations. B. For determination of nominal working above) f stress. pj (as computed pc r may be divided by a suitable factor of safety. is: His formula **$*<§ for (3) pinended columns. be noted that this line tangent to the Euler curve at J v and the equation of the latter is to be used.tJ_*i. and in structural it is employed work to a considerable extent. . probably because more refined tests.L*l. 13) represents this expression. on the ground of simplicity.L > 3?wj E n np 3 : . considerably higher values for allowable stress than other generally accepted formulas. Or if n = p' the expression may be put in the following form for } direct computation of mean working stress. the direct material. very convenient in making a large number of computations for columns of any one material. The is curve FB J 1 (Fig. Professor Johnson says ("Materials of Construction. or based upon from removed upon conditions further it is those in practice.66 This formula 13. it The formula It gives of Professor Johnson's is empirical. This expression It is is very simple. It will is MACHINE DESIGN represented by the straight line is THJ 2 in Fig. crushing stress of the For columns having to I t p x greater than the value corresponding to the point of tangency J (should such be used). (2) p of the Professor J." pages 301302) that both Bauschinger and Tetmajer "mounted their . after k has been determined.
for the lack of homogeneity of the material. with no lateral It movement whatever. . The * this "This precaution is essential to a perfect test of the material. requiring a higher contingency plotted in Fig. factor in the latter for safety." [J. B.: ' STRAINING ACTIONS IN MACHINE ELEMENTS columns with cone or knifeedge bearings axis."* this would appear that precaution tends to make the test one of the material and not of a long strut. the formula may be put in the following form for computing nominal working stress: 'to®'The Rankine or « It Gordon formula (see Church's "Mechanics. .] . in a measure. and this fact limits its usefulness. J. 13. value of p e ' (as computed from the above form of Johnson's expression) should be stress. would probably be still nearer the line F F iy and the difference between these test columns and columns as used in practice would be greater. and arranged a very delicate electric contact at the side so as to indicate a lateral deflection as small as 0. the results of the tests. Only in — . Considere mounted his with lateral screw adjust ments. 67 at the computed gravity while M.5> ' ' based upon experiments on the breaking coefficient P it The is purely empirical.001 mm.77T7(Lf m v (i . way can other sources of weakness be eliminated. correction been if Had the made under greater load. may be expressed as follows '' The above formula strength of columns." pages 372376) has been extensively used for columns. Or. He then applied moderate loads to the columns and adjusted the end bearings until they stood under such loads rigidly vertical. the For determining the working divided by a suitable factor of safety n. for the eccentricity of the load (relative to the nominal geometric axis) compensates. is ^. for to leaves much uncertainty as how this coefficient should be modified for materials different from those which have been actually tested as columns.
The ratio p this p is the factor For ultimate strength. the graphical representation of the last expression. Thei most successful attempt is that of Ritter." page (129) "Several attempts have been made to establish a formula for columns which shall be theoretically correct. p f . who. is N) is the more important." . the formula m E it \ p / "The form formula.: 68 MACHINE DESIGN intensity of f mean working stress. but the 1897 later than He gives Ritter sole credit for the formula in edition of his "Mechanics of Materials. eq. of safety. . . . The form is of the Rankine expression his " is rational. article. The curve R T R x (Fig. iV 1 . .* Merriman P "*" c gives the Euler formula for a factor of safety of is n = P> which * Professor Ritter. or the expression can be written p i 1 > () + — \p/ m \p (6) but it is not entirely satisfactory to assume the action for stresses within the elastic limit. but the coefficient P not. Merriman developed equation Ni independently. Professor Merriman says. of this formula is the same as that of Rankine's it but it deserves a special name because by finding com pletes the deduction of the latter formula for p a value which is closely correct when the stress elastic limit p " c The above notation is p does not exceed the changed to agree with' *c that previously used in this. proposed . formula might be written: mr?E \p but the first 2 form 13) (eq. . in : Mechanics of Materials. might be inferred by divi ing p c by n. from the results of tests for breaking strength. in 1873.
If it is desired to design a castiron Pc column with great accuracy values of m J= 2 E may be taken which will give results in ac cordance with experiment and which practically transforms the equation into Rankine's formula. all justify its /^/> x x 2 use. when / h p = infinity. for cast Since castiron in may it vary from 15. that the two curves.000. plies to the use of both of which have no But the same criticism apelastic any rational formula founded on the iron is theory.000 to 20. definite fixed values for cast iron.000 5. T R 2 are 2 and R X . 13 that the Ritter for the and Rankine formulas agree very material taken for illustration. infinity. and that it boundary F F E E (Fig. which it the ordinary formula for short compression members. The Ritter formula (eq. 13). the Ritter formula reduces to p'=P±A. Exception may it be taken to the use of the Ritter formula for involves the use of the stress at the elastic cast iron.0 for one end fixed and the . gives the The lies facts that this formula is rational in form.000. columns designed simply for strength are very rare design it machine therefore seems best to use the formula since otherwise fulfils all needs better than any other. and it will be adopted in this work. since limit. but the fact that the curve of the latter crosses the Euler curve near the righthand limit of the diagram indicates that its constant /? is not theo retically correct. this is only the case when I s p = Professor mathematically.r~= E 1 . It will be noted from closely Fig. tangent to each other If is /=/> EE E x Merriman also shows. and the coefficient of elasticity. Thus the expres sions for deflection in simple iron. that correct values at the limits wholly within the = ooand l + p = o. then for cast columns with fixed ends q = . N) reduces to this last expression for columns so long that the term unity strictly in the denominator is negligible. speaking. as far as cast concerned.STRAINING ACTIONS IN MACHINE ELEMENTS [Failure occurs if 69 p >_ p c . — P L = q. beams contain E which.000. = o. ^ iron If m.
Barnard has devised a diagram which steel or is very convenient for these computations for It is wroughtiron columns. corresponding to a maximum working stress p. x= 20 40 60 80 100 120 140 Fig. 13 (a). All of the above formulas give the value of the stress (p e f mean ultimate h =P 4c A). the ordinary problem Stress in 1000 Lbs. and for both ends free but guided In addition the student should consult treatises on 5.Mi I. U=h f +p "N n. N. J 11 / '// i//t %VAV 9 i I w \//B 1 10 // v/ x A/ / // 7/ '/ys %> III. for the area in such expressions as those given in this article as p' (or p) and p are both functions of the area of the crosssection. of design is to assign proper dimensions for the member under the given load. 7 8 . It is not practicable to solve directly. demanded and then to check for the ultimate load P./ 7o other free but guided q A MACHINE DESIGN = I. to a reduced scale. IV. // A/ / / // / k s* X H=4 Y 1 Jf // . ultimate stress pc or a maximum However. The four curves are for the four end conditions . It is usual to assume a section somewhat larger than that for simple crushing. Professor W.M =u . in Fig. or the working load P'.000 the strength of materials treating fully of this subject. 13 (a). or the mean working stress (p' = P A). respectively.^jy 10 12 14 / '/. shown.
1 STRAINING ACTIONS IN MACHINE ELEMENTS given in Table III. take this quotient on the lower scale and pass directly upward to the proper curve for the given end conditions.). 7 They are plotted for a inch. which fixes p.500. for an example which /^/>=8o and the maximum working stress= 14. etc. be used for cast iron. square sections.* stress (p is If this value of p f agrees sufficiently well with the trial area. page 63. etc. In this case. provided these bear symmetrically with reference to the axis of the column the .900. but such cases may be solved directly by equation N. In the case of a squareended column. In case of a pinended column. . then is pass horizontally to that one of the radiating diagonals which numbered point pass to correspond with the selected stress. If the crosssection of such a column has equal dimensions in these two planes (circular. If the pinended column has an oblong section rectangular but not square. plane.000 (pinended). maximum of working stress of 10. p' found to be about 7. or when the supporting action of the ends is equal in all possible planes of flexure. may be weaker in * in The method is of using the diagram is indicated by the arrows. or other materials where the ratio t E ^ will give values far different from the above. however. the cylindrical supporting pins make it equivalent to a squareended column against flexure in the plane of the axes of the pins. value of p.000 should not. or to take p for the axis about which the section is weakest. it for the latter (elliptical. as a connectingrod. The curves 29. while column is pinended with reference to a plane perpendicular to the axes of the pins. from this to the horizontal scale at the top of the f last upward diagram. wood. They may be used for any other stress by proceeding as follows: Divide / by this Assume a trial crosssection. it is sufficient to take the least radius of gyration of the section.).000 fy ' pounds per square and a value a E h 1 __ ° OOO which is an average value for steel. quotient of the load divided by the the section may be considered as satisfactory. ) where the value of the unit load or mean working read off. the column need only be computed I section.
In the preceding discussion. the various formulae have been given both for breaking and for working loads.000 lbs. If the m = i. feet The is connectingrod of a steam engine per the 5 long and subjected to a load of 20. like the J. must be greater than * Assume 2]/ 2 inches as a trial diameter. or elliptical. and column has a the section of pins is vice versa. are. except for quite long columns. The Euler and hence Ritter formulae are derived from the theory of elasticity. in which the dimension in the plane more than onehalf the dimension in the plane perpendicular to the pins.2 sq. less rigidly applicable to working loads stresses. for similar reasons. are. A = it — 000 9. in this respect. 20 rod were designed as a short column. in. a circular section the at centre of the Take £ = 30. lbs. unless the ob viously stronger in one of them. The Euler formula is not applicable for practical applications. in which the stress should never exceed the elastic limit. the required area would be inches.000.72 either of MACHINE DESIGN these two for planes. to notwithstanding the difference in end conditions relative them. it will suffice to compute as a pinended column against flexure in the latter plane.000 The rod may be considered a pinended column. or a diameter of ifj and evident that for a long column the diameter this. common beam Such formulae as Rankine's and B. and it may be section necesis sary to compute both planes. in. .* It does not follow that these two rational formulas will agree with experiments on the ultimate resistance of columns or for materials which do not follow Hooke's law strain. and Hence elastic limit ^ = 36. sq. deter mine the diameter per sq.000 lbs.000 is = 2. these are proper for computations pertaining to working loads. and is Example. ins. Johnson's. of proportionality of stress to These expressions formulae.. stress is If the maximum rod. allowable of 9. If a rectangular.000. derived from tests of ultimate resistance of columns.
if the ratio P / be less than about in para 25.9 = 21. it ing discussion of columns has been assumed that the load has is been applied axially.000 6O = 4. and the arm a is called the column be less than 4 or 6 times its least diameter.000 / 1 + I 36. it due to the load is has been shown by the discussion on long columns that such not the case. This obviously the best way of applying at a dis the load. that is. the member may be and formula treated by the method outlined p graph If. The student should also follow the solution through 21. (10) M will apply or Pae P =— + I A  however. is In such a case the column said to carry an eccentric load.000. whence in JV 0.300X4. but cases often occur where it must be applied tance a from the axis of the column. p =.9.= ^ o 4 d S / = 60". In the preced Loading of Long Columns. Eccentric on the diagram.3°° X 2 7T X 30. which is a little more than the required load and the section will fulfil the requirements. as P — A. If the length of the eccentricity. adding to arm a an therefore additional amount due to this de The column (a) stresses acting on an eccentrically loaded the are A compressive stress p„ such as would be induced if load were axial. can no longer be assumed that the direct stress uniformly distributed over the section.— STRAINING ACTIONS IN MACHINE ELEMENTS 73 Then A = 4. .000 . the original lever flection. is In addition. the column be longer than 4 to 6 times it its least diameter. it it is obvious if column will deflect somewhat more than This will would the load were applied axially. that the if the load is applied eccentrically.\P = 4.070 lbs. have the effect of «.
500. and working such as must be used in machine design. If e' be the distance from the neutral column is made of a material. the character and magnitude of the stress on the convex side should be investigated. tensile or compressive.. always be less than the stress on the con For materials whose elastic strength is is about the same in either tension If. p = 3. page For the ordinary cases of machine design this refinement 217. can be determined by the theory of see For a full discussion of the manner of computation Merriman's "Mechanics of Materials. such as cast iron.) * The stress induced on the convex side of an eccentrically loaded column but will may be either cave side. however. may be Example. . to the eccentricity {a and propor tional to the bending moment P + «) For the first from Ritter's formula (N) and for the second from (/) (a P h= Therefore the + 1 «) e = P (a+ a) e A? is maximum compressive stress* in the section pT For columns whose stresses ratio A of fl\ 2 / (a+a) e \ — is less than ioo. the whose tensile strength is much less than its compressive strength. then the stress (p) on the convex side is. 74 (b) MACHINE DESIGN A flexural stress p 2 due . of 10 inches. a elasticity. p negative the stress is compressive.000 pounds c . omitted. A circular to carry a transformer wooden pole 30 feet high is required weighing 800 pounds. the deflection neglected. or compression the stress on the convex side of no importance. or where the stress is necessarily high. P = A if [_ ^~E \p) is ~ 1 + ^~J 2 J P 1S P ° SItIVe e Str6SS 1S t6n " sile. with an eccentricity the diameter at the middle in order What must be due Let that the stress to this load shall not exceed 500 pounds per square inch? 1. a may be For columns longer than this." 1905 edition. per square inch and E= Also m=— 4 (See Table III.000. axis to the outer fibre on the convex side.
500. Then /> = 2 and A = 50 /^6o\ 2 Whence p r = 800 50 r L 1 + 1 X r ?. for a rise is The amount which of one degree in temperature. E p = C C / strain Example. or elongated. Stress metals expand Due to Change of Temperature. cannot expand.0000062 Wrought Iron If . a bar expands per unit of length. t E = stress P.j. Practically all when heated. what stress will be induced in and what force must oppose it to prevent expansion? . called its coefficient of linear expansion.. 22.0000065 Hard Steel C= C= Soft Steel Cast Iron C= C= stresses are . Let p Since = induced per unit area. and contract again when cooled.: STRAINING ACTIONS IN MACHINE ELEMENTS 75 Assumed a diameter of 8".000 V 2 / 1 10X4I ^ 4 J 4 800 _i + 26 c + 10J = 592 pounds If this excess is per square inch. Let t = change stress in temperature in degrees. A bar of wrought iron 2" square its is raised to a it temperature of 100 degrees above normal. a second approximation can be made. and will be denoted by C. considered too great.0000074 . values of The following table gives C for various substances for one degree Fahrenheit . so as to prevent from in it which are same as though the bar had been compressed. an amount corresponding to its expansion or contraction due to the change in expanding or contracting. If held so that it.ooo 6 ' 2 7T (*—) + X 1.0000068 it a bar of metal is held at the ends. produced called temperature stresses the effect being the temperature. n *j? — = —. .
wiredrawing. subjected. all In the previous discussions on the vari ous straining actions to which a applied without member may be or impulse. as it is deformation.000. limit.000. the energy expended called elastic resilience* * is If the piece is ruptured. as it already pointed out. and the total opposing force P = 23. it has been assumed that the load was a simple dead load and initial velocity But.400 lbs. to be understood. or not stored but has of form. or may be applied in any way. the load may be applied impulsively.000 = 20. produced equal to is the deformation at the true elastic limit. and removed and applied again and again repeatedly.0000068 100 X 30. and the energy required to cause the plastic not recovered. is to that expended producing elastic deformation. elastic strain. but does affect the amount of stress or strain.. .400 X 4 = 81. a material is distorted doing a certain amount of If the by a straining action. the shaping of ductile are processes in metals by forging.000 p =C E = t . it is capable of work as it recovers its original form. it returns beyond the in work only equal set. been expended in producing such permanent change etc. Resilience. Ordinary springs illustrate the first case. the discussion of repeated loads will be given in a succeeding If section. rolling. In order to more clearly discuss the it effect of impulsive effect of loading will be necessary to consider the straining a load somewhat more fully. the energy elastic resilience When the term resilience is used without qualifying context. this deformation does not exceed the is amount elastic of work equal to the work done upon the material in producing If the material is strained such deformation. which nearly all of the energy is expended in producing perma nent deformation. or the repeated application of a load.76 MACHINE DESIGN Let E = X 30. The work the required to produce a strain in a If the strain member is is called work of deformation.600 lbs. The application of a load in an impulsive manner. P will be 20. does not affect the character of the straining action.
A more comparison of these illustrative stressstrain diagrams (for quite different materials) also shows that. less rigid material may have the greater resilience. the curve divided by the maximum ordinate be stress. the work action of deformation equals the ultimate strain multiplied by k times the maximum It is evident that for a straining beyond the curve elastic limit. 6) is the stressstrain diagram for a given mate rial. as in . the ductile. less total and tion. work of deformation per cubic inch of the In such materials as have wellmarked elastic limits (proportionality line between stress and strain through a definite range) the Oa is a sensibly Oaa' = ^aa'xOa'. OADEE' represents the stressstrain diagram of a material having higher elastic and ultimate strength than the The greater inclination of the elastic line (OA) with the axis of strain of elasticity. in the second case. in the sec Oa' ond case. E = 2 A A' OA' diagram The rials stressstrain OADEE' and shows that of two mate one still may have have curve both the higher elastic and ultimate strength. and the elastic resilience the elastic resilience equals the elastic strain (Oa') multiplied by onehalf the elastic stress (% aa'). k > X and k < 1. but less total work of deformation than the first. straight line. a higher modulus this modulus equals the elastic stress divided In the first the by J elastic strain. for a given stress.STRAINING ACTIONS IN MACHINE ELEMENTS expended II 77 in breaking it is called total work of deformation. less elastic resilience work of deforma If the O a" d" e" is the stressstrain diagram of a of elasticity third material first). The area Oadee' equals the base (Oe') multiplied by the of the curve Oade. Hence. as (OX) shows. the area O a a' represents the elastic resilience. O a d e (Fig. it (having a modulus similar to the appears that this third material possesses greater elastic resilience. when a member must absorb considerable energy. The former. or. or. if the quotient of mean ordinate (y) this mean ordinate of called k. case E = x a a' 7—. and O a d e e' represents the total material.
: 78 MACHINE DESIGN weak yielding material case of severe shock. if loaded to a" the permanent set is O O". Shock. a comparatively may be safer than a stronger. effects of strain O" a" d e. This is frequently similar to recognized in drawing specifications. elastic limit a" of the overstrained material is evidently higher than the original deformation. The beyond the elastic limit are thus seen to be I.e. but beyond the point a' its stressstrain diagram will fall in with the curve which would have been produced by continuing the first test (i. a load impinging on member. elastic limit. is while the original total work of O a d considerably greater than the total work of deformation of the overstrained material. or in having greater resilience for a given If a material is strained much f maximum beyond its elastic limit. Reduction of the facts total work of deformation. a'de). and torsion all except (5) apply directly to cases of and flexure. Elevation of the elastic strength and increase of the elastic resilience. Upon again applying load. In fact springs differ from the socalled load. Similarly. but the formulae will apply equally well to uniform compressive and shearing stresses. its elastic curve will be 0' a'. . The 13 b). perhaps be well the to first consider the general case of initial velocity. stiffer material.. the stressstrain diagram becomes O" a" d e. rigid members only in the degree of distortions under loads. ing to a free through the height For simplicity. as to a (Fig. Impact. and upon again applying load. Suddenly applied Load. 24. overstraining The above noted elevation of the elastic limit can usually be largely or by wholly removed by It will annealing. II. with an fall this velocity (v) correspondh. upon removal of the load it will be found to have such a permanent set as O O'. . e. a. The principle is that involved in the use of springs to avoid undue stress from shock. to re These have an important influence on resistance peated shock. the dis cussion will be confined to a load producing a tensile stress.
13 (b).STRAINING ACTIONS IN MACHINE ELEMENTS 79 W= d static value of load applied to member. but i. The energy k Po. L. = total distortion of member due to impulsive load.) I. Resilience. x= ft ^r X &=a (for convenience). Fig. work of deformation (See preceding article. A = area of crosssection of the member. 13 (c). stress due to load as applied suddenly. + >>) W (h n = P = ps =HP& (1) d:X::P:W w = (2) 2Wh — W —+ _ 2 2W 2 h „. . J^^ Fig. Case — Maximum Stress within Elastic Limit. h = height corresponding to velocity with which load is applied. p = maximum intensity of resulting stress. px + 2W (3) . . is exceeded k K > y 2 if E. P = p A = total max. to be absorbed by the is member due the to the impulsive is application of the load W (h + d). value is E. W. X = total distortion of member due to static load. L. if and k < 1 1 «W'* 1 1 ^ . its constant. is not passed.
'. or strain. Thus (Fig. hence is the stress produced by a given velocity of impact (height h) sible. elongation at the elastic limit equals of elasticity F+E. A = WL ^ A E. whence p = w p (1 + vTTo) = 2 2 w (30 *=a=T__ = + V + — Case II.P 2 .— 80 2 MACHINE DESIGN 2 W h =T. (4) P d = + J=J(l+Vl A = A (1 P_ w (5) =— + V 1 + 2 x) (6) The limit. (or corresponding value of h) to produce a given . reduced by using as long a member as pos If the ity. 13 c). When (a) the stressstrain diagram : known. It and 8 are directly proportional to the length of the member. work is and this area must equal resili the area ence.'.+2WP . the constant k of equaexact between % and 1 (see Art. but without initial veloc h = o and x = o.L) (F . .E) : the : member. and than the total of deformation O ade e'. the latter being greater than the elastic less O a a'. 23. load is applied instantaneously. represented by the rectangle m n ca.P = W(l + 2X) A/I + 2X) . and its value cannot be determined in the absence of the stressstrain dia gram is for the particular material. in which E= modulus If and F = intensity of stress at the elastic (A L = length of . . in this illustration. O a b c. Resilience).'. (7) As will A is small for metals (except in the forms of springs) a moderate impinging velocity be evident that A may produce very severe stress. the following prob lems can be readily solved Determination of the velocity of impinging of a given load stress. W {h + 8). d w <s0 2 A = x (1 1 o) (60 If the maximum tion (1) is stress exceeds the elastic limit. Maximum Stress Beyond the Elastic Limit. A) : : (W + F.
8' the distortion per unit of length of the member due to impulsive load. or strain. stressstrain diagram for stress per unit of sectional area strain per unit of length of the W (a) : W w : (h' + V) = kp d' = R'. as almost always desired keep the maximum intensity of stress. or the modulus of resilience. P f A. but can be easily accom plished. raises the elastic limit 13). The limit to is case in which the maximum stress is within the elastic it is by far the most important. ticular stress. the d. known If the or strain.1 STRAINING ACTIONS IN MACHINE ELEMENTS (b) 8 Determination of the load which Determination of the will produce any parvelocity. or strain. because of reduction in cause the ultimate resilience. as the preceding. (Fig. (7) W ' = JPTs The solution of this problem is (8) (c) not quite so definite. in (a) is and (b). and R r the resilience for unit of volume. let be the load per unit of sectional area. and decreases the The effect of a is shock which strains a its member beyond for the elastic limit to reduce margin of safety its subsequent similar loads. especially as every overstrain (beyond this limit) total resilience (see Fig. parallel to O e f . the shaded area area/zg = gOt. . Numerous actions successive reductions of the total resilience finally by such may 6 member to break under a load which it has often previously sustained. • . 13 c) it Draw the line g q (indefinitely). and at a distance from equal to W\ take out the value of o. Whatever the <$'. W the height due the velocity of impinging divided by the total acting length of the member. with sufficient accuracy. (c) when impinging with a given stress. Let the work of deformation corresponding stress. in it the general case. hence the unO c qfi gO = W shaded area under the few stressstrain curve must equal W hf . . W —V . . be called R=k P and member. h' = J. . produced by a given to load impinging with a given velocity. A trials will suffice to locate give/i b qf = m O = 11 t W the limiting line b q c which will h'. graphically. within the elastic limit.
possibly owing to slight departure from perfect elasticity under stress much below crystalline that ordinarily designated as the elastic limit. which ing article. although some authorities restrict this term to the kind of deterioration already referred to as the simple result of a decrease of resilience. It is well known that an appreciait necessary for a ductile metal to flow. a decided de terioration of the material very frequently occurs. This effect has been called the Fatigue of Materials. as this cause. and frequently repeated. in part at least. that materials to which are subjected continuous variation of load cannot be if depended upon to resist as great stress as they will carry applied but once. to the elevaof the ultimate resilience. provided the application of load is not so rapid as to become impulsive. . When the load is suddenly applied. The term fatigue implies a weakening of the material due to a general change of structure. the decline endurance of strength or of the power of may perhaps be ascribed.82 MACHINE DESIGN No doubt many cases of failure is effects just discussed. greater apparent strength it than by applying 25. even in the absence of appreciable shock. 24. Fatigue of Metals. as is does flow when show its section changed under stress. It was formerly supposed that the repeated variation of stress caused such change of the general structure. has been found by experience and experiment. or only a few times. The kind of failure which is the subject of the next topic is due to a real permanent deterioration of the metal. this The but appearance of the fracture sustained tests of pieces numerous from a member ruptured in way. but there can be accounted for by the another and quite different is kind of deterioration of material. tion of the elastic limit and reduction discussed in Art. But apart from with repeated loads. in time. Thurston has shown that the prolonged application of a dead load may produce rupture. with an intensity of stress considerably below the ordinary static ultimate strength but above the ble time is elastic stress. It On the Peculiar Action of Live Load. treated in the follow Dr. and it is due to distinctly different causes from those mentioned above. this view. a test piece will by quickly applying the load more slowly. hence.
STRAINING ACTIONS IN MACHINE ELEMENTS (taken as near as possible to the break) fracture. and the product of the best steel makers of today is much stronger and more reliable than wrought iron. However. Professor Johnson suggests: "the gradual fracture of metals" as a more appropriate term than " fatigue. each grain consisting of many crystals. in which a fracture in one wire does result of unskilful manipulation of this . and that when deformation takes place in a metal these crystals move relatively to each other along is u gliding planes.under variation of stress. in time. often originally too small to be detected by ordinary means. and may. It is probable that the prejudice against steel is largely the more sensitive material." Many men of large practical experience still prefer wrought iron to mild steel for various members which are subject to constantly reversing stress. The "gradual seems to fracture" through extension of "microflaws" accord with the observed facts more closely than the older theory of general change of structure. . S3 and it is difficult to test pieces show such crystalline reconcile the normal appearance and fail to behavior of such structure. These " microflaws " tend to extend across the section . Wrought iron may be likened to a wire rope. . 1906) will be found an account of recent researches tending to show that metals are made up of grains. not directly extend to adjacent wires." If the stress producing such sliding the gliding planes series of repeated often enough the contact at into a crack or weakens and finally passes cracks which extend across the section. In the American Machinist (Sept. reduce the net sound section so greatly that the intensity of stress in the fibres which remain intact be comes equal to the normal breaking strength of the material. it is just possible that the very lack of homogeneity in wrought iron renders it safer under varying stress {other things being equal) as the fibres are more or less separated by the streaks of slag. 27. with the theory of general change of that every piece A theory which has been largely accepted is of metal contains innumerable minute flaws or imperfections. and a flaw is less apt to extend across the entire section than it is in the continuous structure of steel.
" (Du Bois) : " Structure of Iron and Experiment has shown that the breaking strength under peated loading. it not probable that such rational analysis can ever be satisfacAll of the formulae that have been derived for computation of torily applied. (c) The breaking strength or the first static elastic strength. The systematic experiments upon the effect of repeated loading were conducted by for example.84 MACHINE DESIGN The theory of the subject is. The number (b) allowable working stress usually depends upon of applications of the load. in a many machine members. This is frequently either from zero to maximum. function different for different materials." Unwin's "Testing of Materials." is re a function of the magnitude of the variation tions of stress and of the number this of repetiis of such varying stress. and if the "microflaw" theory is correct. sile Wohler [1859 to 1870]. or stress. breaking strength under known variations of load. or between equal plus and minus values." Weyrauch Steel. general. was ruptured by: . Consult: Johnson's "Materials of Construction. that a bar of wrought iron. The range of load. on record which go to and there are authentic observashow that. subjected to ten stress varying from zero to the maximum. Furthermore. or the "carrying strength. the carrying strength under repeated loads a function of the static strength. He found. are fit empirical ones which have been adjusted to ally the experiment determined facts. as yet. : (a) The This should be considered as indefinite. or practically infinite. too incomplete to permit of derivation of rational formulae to account for the effects of re peated is live loads. as between different mateIn is the one with the higher static breaking strength does not always possess the greater endurance under repeated loading." Merriman's "Mechanics of Materials. however. tions rials.
STRAINING ACTIONS IN MACHINE ELEMENTS
800 repetitions from o
107,000
" "
" "
to 52,800 lbs. per sq. in.
85
o to 48,000
o to 39,000
"
"
450,000
10,140,000
"
o
to
35,000
—Merriman, page 191.
up to something less than the elastic limit an indefinite number of times (several millions) before rupture occurred; but with complete reversal of stress, or alternate equal and opposite stresses, (tension and compression), it could be broken, by a sufficient
It
was
found that the stress could be varied from zero
60000
„   ' '^'^^jS
40000^
s'yp>^
^'^^ y/
D
^,~Z*^*^
S>
20000
r
I
/
O
Tension Compression.
1
/
number
of
20000
Fig. 13 (d).
applications,
when
the
maximum
to
stress
was only
about onehalf
to twothirds the stress at the elastic limit.
A number of efforts
of
have been made
deduce from the experto the design
iments of Wohler, formulae which could be applied
machine members
is
(see
Unwin, page
it is
36).
One
it
of the best of
easily applied
these formulae
that of Professor
Johnson as
is
to all cases that will arise;
simpler than most of those previ
ously proposed; and
it is
probably as reliable as any yet offered.
Launhardt's for
Two
formulae which have been very generally accepted for
are:
computing the probable carrying strength
:
86
MACHINE DESIGN
varying stress of one kind only, and Weyrauch's for stress which
changes
60,000
sign.
Suppose a material
lbs.
to
have a
static ultimate strength
u
of
per sq.
in.
If the
minimum
unit strength be plotted
as a straight line,
A OB
is,
(Fig. 13 d), the locus of
is
the
maximum
tensile
unit stress, from the
Launhardt formula,
for example,
the broken curve
from
B
to
D.
.That
when
the
minimum
stress is 12,500, the
maximum
tensile carrying stress
about 40,000;
indefinite
or the material could be expected to
of loadings
if
would be stand an
In a
number
the range of stress did not
exceed 15,000 to 40,000 pounds per square inch in tension.
similar way, the
broken curve from
D
to
C
is
the locus of maxiof
mum
line
tension,
from the Weyrauch formula, when the locus
compression)
line
is
minimum
stress (negative tension, or
It will
the straight
A
O.
appear that the straight
CDB
it
agrees
fairly well
with these two curves.
Inasmuch
is
as
seems un
reasonable to expect an abrupt change of law
stress passes
when
the
minimum
for
through zero, and as there
no rational basis
Launhardt and Weyrauch formulae, it appears reasonable to adopt the upper straight line as the locus of the maximum stress.
the
Owing
to the discrepancies in the observations
(which must be
expected from the probable cause of the deterioration of the
metal), this straight line
may be
accepted as representing the
line.
law as accurately as could be expected of any empirical
These
are, in substance, the
reasons given by Professor Johnson
line
for basing his
formula on the straight
C D B.
For
full dis
cussion and derivation of the following formula, see Johnson's
" Materials of Construction," pages 545547.
Let p2
— maximum intensity of stress. p = minimum intensity of stress. u = ultimate (static) intensity of stress.
x
Then
in general
h=
As
~p[
(l)
the expressions contain the ratio of the
minimum
to maxi
mum intensities of stress,
instead of their difference, they are ap
STRAINING ACTIONS IN MACHINE ELEMENTS
plicable
for
87
when
the area of crosssection of the
member is unknown;
same
as the
whatever
this area, the ratio of the stresses is the
ratio of the loads
producing these
stresses.
In substituting values
of
p and p 2 care must be taken
x
,
to use
proper signs;
is
thus,
if
tension
is
taken as positive, compression
negative;
is
or, if the
stress varies
/>,
between tension and compression p 2
positive
and
is
negative.
x
For dead load, p
= p
2
;
P2
For repeated load when p
x
=
o,
p — =

o
.,A£;X.
For complete reversal
of load,
!.....
p
x
(3)
=
^ pz
(
^T^"^*"*"
2
y
2
u
y
2
u
+
P2
The
three special cases (2), (3),
in designing,
of.
and
(4),
are those most
com(1)
monly met with
Example.
is
but the
general expression
should not be lost sight
A
bar of
steel,
whose ultimate
static tensile
strength
70,000 lbs. per sq. inch, is subjected to a
repeated load whose
value.
minimum maximum
number
value
is
one half the
maximum
What
is
the
stress that
can be carried by the bar for an indefinite
P —
of repetitions?
to the load
Since the stress will be proportional
p1
=
Hence substituting
....
.
in equation (1),
p.,
=
A
l
x
I
70,000
P2
—
2
=
47,000.
2
It
is
2p
to
be noted that the allowable
maximum
if
stress is
above the
original elastic limit of
to
most
steel,
and
the piece were designed
be stressed
to
47,000
lbs. the
result
would be that the
first
application of the load would raise the elastic limit to that value.
88
MACHINE DESIGN
of in
But the piece would take permanent set and be in most cases no further use. A factor of safety must therefore be used
order that the
limit.
maximum
stress
may be
well below the elastic
The
rials;
experiments of Wohler, and his successor in the
field,
Baushinger, were conducted on a very limited variety of mateso that while the above discussion points out
what may be
expected in a general
ficiently
factor of
way from most materials, they are not sufconclusive to make it possible to pick out the exact safety to be used in all cases. They do, however, throw
on the apparently high factors of safety which must for which no other satisfactory explanaof Safety.
much
light
sometimes be used, and
tion has
26.
been found.
The Factor
The
preceding paragraphs
(arti
cles 9 to 26)
have considered the
effect that different
methods
stress
is
of applying the load will
have on a member, and the relations
the load
which
exist
between a given dead load and the resulting
It
and
strain.
has been shown in Art. 24 that
if
applied suddenly the resulting stress and strain will be twice as
great as for a dead load.
And
finally in
Art. 25
it
has been
shown
that the
maximum
stress that
can with safety be induced
It
repeatedly in a member, will depend on the range of stress.
would seem as though a member designed
these logical theories
in
accordance with
would be
satisfactory.
But
it
must be
very
remembered
incomplete;
that these theories are not absolute, that the inis still
formation regarding the characteristics of materials
that flaws
and hidden defects always exist; and finally that there is always danger of accidental overloading. In addition, it is generally essential that a machine member be not only strong enough to avoid breaking under the regular
maximum working
permanent
if it
load, but also that
it
shall not receive
a
set;
for a
machine member ordinarily becomes useless
it
takes such set after
has been given the required form.
In
many
cases a temporary strain, even considerably below that
limit,
corresponding to the elastic
would
seriously impair the
accuracy of operation;
and
in such cases the
member
often reIt
quires great excess of strength to secure sufficient rigidity.
STRAINING ACTIONS IN MACHINE ELEMENTS
follows, therefore,
89
from these considerations that
the
if
the design of
a machine
member were based on
maximum
allowable stress,
as indicated
fied
by Wohler's experiments (such
stress
being modi
by the theory of suddenly applied loading, should it be would be no margin to allow for the uncertainties and unknown defects enumerated above; and in many cases leave no assurance that the elastic limit would not be expresent), there
ceeded.
theories
So that while stresses fixed in accordance with these
form a good
basis, they
must
in general
be reduced by
means
of a factor of safety so that the
working
stress is
enough
lower to provide for these uncertainties.
The
factor of safety
is
generally defined as the quotient of the
ultimate static strength divided by the working stress.
A
con
sideration of Wohler's experiments shows that such a definition
is
misleading.
For a factor of safety of
2,
for instance,
might
with
be perfectly safe for a dead load; but
stress in
for a repeated load
one direction
it
would leave no margin
factor of safety
at all for continto
gencies.
The apparent
and the
would seem
be a
better term,
real factor of safety
quotient of the carrying strength, or
as given
may be defined as the maximum allowable stress
stress.
by Wohler's experiments, divided by the working
too often applied,
is
The
and, as
factor of safety has been called the "factor of ignorance,"
it
is.
it
perhaps
little
else.
Thus
shall
very often
it is
specified that all the
members
of a
machine
be designed with a certain fixed factor of safety without regard
to the conditions
act.
under which the various members
applied in this
It is
may have
is,
to
A
factor of safety
manner
generally
speaking, a factor of ignorance.
of safety will always retain
probable that the factor
it
an element of ignorance, for
can
hardly be hoped that the powers of analysis will ever permit the
prediction of the exact effect of every possible straining action,
due
to regular service
and accident.
Neither can
it
be expected
that the
methods
of manufacture,
and inspection,
will
become
so
perfect as to eliminate or
in materials or
measure precisely every possible defect
workmanship.
(at
But a careful study
least
of the condito the effects
tions of each particular case
and a proper attention
which may be weighed
approximately) should, with
90
the knowledge
fairly
MACHINE DESIGN
now
is
to
be had, enable the designer
to
make
a
accurate application of the factor of safety, an intelligent
the most important part of design.
choice of which
Most
of the formulae of
Mechanics which are applicable
to the
design of machine members, are based on theoretical treatment
induced by the action of given forces within the upon the member under consideration; and the theoretical conclusions so reached are amply verified by practical experiment. When, therefore, the conditions under which the member is to work can be analyzed, and the laws of Mechanics applied to its design, such methods as outlined in this chapter
of the stresses
elastic
limit
are perfectly rational,
gencies.
if
intelligent allowance is
made
for contin
Many machine members,
and
less satisfactory
however, are subjected to
that analysis cannot be
such a complicated system of
strictly applied,
stress
approximations or assump
tions are unavoidable in the present state of knowledge.
When
such
is
the case, the designer must either base the design on the
stress, if
predominating
there
is
such, allowing such a margin or
factor of safety as experience or experiment
may
show, to pro
vide for the minor uncertain stresses; or,
if
the case considered
to
be beyond such treatment, recourse must be had
empirical
methods or judgment.
as a
(See Art.
i.)
While therefore mathematical treatment of any case
will serve
good guide
to correct proportions,
such treatment must
other branches
always be tempered with judgment,
a high development of
all
which
is
necessary to successful design, as in
of engineering.
While,
also,
no
fixed
rules
for
selecting
the
factor
of
safety can be laid
down, a knowledge of Wohler's experiments,
suddenly applied loads, will greatly aid the
and the
to
effect of
designer in the matter.
Thus when
it is
known
that the load
is
be a dead load, an apparent factor of safety of 3 will, for wrought iron, or steel, bring the working stress well below the
elastic limit
and allow something
for contingencies.
If,
however,
to a
the load be a repeated load, the stress varying
from zero
maximum
must
tensile stress, the apparent factor of safety for steel
at least
be
5, to
allow a good margin below the elastic limit;
STRAINING ACTIONS IN MACHINE ELEMENTS
and
in either case,
if
91
in addition the load is to
2 to
be suddenly applied,
these factors
must be multiplied by
insure safety.
Example.
A
steel
beam
is
subjected to a suddenly applied
load which alternately induces an equal tensile and compressive
stress;
if
the ultimate strength be 60,000 lbs. per sq.
in.,
what
apparent factor of safety should be used, and what
real factor of safety?
will
be the
Since the stress
is
a reversed one, the
maximum
allowable
stress or carrying strength is
of
by Wohler's experiments onethird If the working stress is oneultimate strength or 20,000 lbs.
it
half of this value or 10,000 lbs.,
will leave a
good margin
(10,000)
for
contingencies, disregarding the impulsive effect.
is
But the load
applied suddenly and, by Art. 24, this value
2,
must
be again divided by
sq. in
making the working
stress 5,000 lbs. per
is
Therefore the apparent factor of safety
60,000 —
=
12,
5,000 20,000
while the real factor based on Wohler's law
If the
is
4
5,000
member should have
or
if
to
work under extremely trying
still
conditions,
shock or other stresses which could not be
further
analyzed were present, this value might have to be
reduced.
TABLE
IV
FACTORS OF SAFETY
Repeated
Stress in One Direction.
Repeated Reversed Stress.
*T3
Character of
Material.
Dead
Load.
•Sag
(A
Sag
CO
Wrought
or
Iron, Steel,
other
Ductile
3
5
Metals
Cast Iron, or other Brittle Metals
IO
6
12
4
6
12
10
20
Table 4 contains factors of safety such as are used
in practice
and which agree
are, of course,
fairly well
with the foregoing theory.
They
average values and must be used with judgment;
92
MACHINE DESIGN
its
but in the absence of trained judgment, or as an aid to
velopment, they
de
may be found
useful.
Table
5
contains values of the ultimate strengths and elastic
limits of the materials
most used
in engineering.
They,
also, are
average values such as the designer must use in the absence of
exact information regarding the material to be employed, and in
general such exact information
It
is
lacking.
may be
observed that an increased factor of safety
may
not
If
always in the case of cast metals give a stronger member.
results, the gain in strength
the increased dimensions give sections so thick that sponginess
may be
negative; and
when
internal
pressure, such as
it
is
found in hydraulic work,
is
often necessary to
is to be withstood, do with a smaller factor of safety to
insure soundness.
STRAINING ACTIONS IN MACHINE ELEMENTS
of
93
o o o
o"
Elasticity.
o o
o"
ooo'ooo'oi
o
Transverse
Coefficient
o o
ooo'ooo'oi
ooo'ooo'oi
ooo'ooo'oi
ooo'ooo'oi
ooo'ooo'oi
o
pT
c
o o o
o"
o
o^
o^
CM
ooooooooooooooo ooooooooooooooo ooooooooooooooo d d d d d d d d d d d d d d ooooooooooooooo
o"
o o o
o"
o
to
.„
oo"
o^ o"
o^
o^
o^
o
POPOfOPOPOPOrO<N POM
o"
o"
o*
o^ o"
o^ o"
o^
o^
o^
o^
o^
o^
o^
o^
o^
h
i^
h
n
io o" m" ^t
m
m
M
m
P
!3
u
o o o
OO"
o o
o^
N
P)
s H •J Z a =s H
CO
6.
o o
o^
o o
o^
oo"
P)
o o
o^
"3"
o o
o^
"TT
o o
o^
o o
o^
o o
o^ n"
o o o
o"
o o o
o"
»o
PO
o o
o H
t/2
E o
vo
O" no"
rO PO LO NO t^
oo
po
•*
w
o
'35
c
o o o
o"
oooooooooooo oooooooooooo
o^ o"
c S
Eh
d d d d PO^l^iOLONO PO 00 PONO
p*"
o^
o^
o^
o^
o^
o^
o^
o^
o^
o o
O^ o"
cs
o o
O^
o^
o^
iO
no"
o o
oo"
po rC
d\
no" no"
io to
3 o
CO
ooooooooo ooooooooo
o N o o
o
o"
o"
P»
o" oo" <t 't
n "O^O
o^
o_ o"
fO oo" OO 00 •sf
q^ o"
o^
o^
o o o
o"
o o
o^ pT
o 2
H
Pi (
a H <
O
d
o
B
o o °
o o °
o o o
NO
o"
o o
o^
o o
o^ pT
to pT ON ^t"
3
73
ooooooooooooooooooo ooooooooooooooooooo
o o
o_
o^
o^
o^
o^
o^
o^
o^
o^
o^
c y
r
o"
pi
iO to
i/> it,
fO o" C> fO CO* o" no" 00 00 O ii NO n
o" o" VOO
o^ pT
N
o^ hT
fO
o^ o"
o^
o^
w
io o"
roiooo n
o^ o"
o
in
<
<fi
c
«:
i
c c 
c
3
c
i
c
r> ~
,2
R
c
c c h
c
: _:
B a
b
s
]
c
p
c
1/
u
01
1
)
s
)
£
1/
L u c
G
DC
_
1
b
7

c
7
p
•s
c
u
i c c
c
p
.
c
1/ 5
NO
J;
'
J 1
c 
7
b
s
c
c
c
"a
c
"c pc
"a
2
'Z
C  7 7 u u u i c a
a a

7 3
_t
C/3
C
C
L
7
c7
7 7 7
i
I
9.
p
1 7
L.
"
~z
K
N c c 
g L
7 H <
1
I
94
MACHINE DESIGN
TABLE
Character of Stress or Strain.
VI
Formula.
A
B c
Stress in
Ten. or
Comp
P
Pl A AE
Strain in Ten. or
Comp
P
D
E
F
Torsional Stress,
Solid
Cir
Pa
T
Palp
e
10
Torsional Stress, Hollow Cir
Pa
4 '.*> r _A(<.
IOfli
G
Torsional Strain, Solid Circular Shaft Torsional Strain, Hollow Circular Shaft
Deflection in Bending
Stress
6=
e __
*
2Tl
TrEsd*
32
H
I
T
TrEaidS
See Table
—
I.
d<f)
J
due
to
Flexure
M
e
= tL
e
See Table
I.
K
K,
Combined Bend'g and Twist'g
« "
2
K
K
L
"
"
«
3
M = }4M+y V M*+7 J T =V M M = }4[x+V X \l]T r =rV^ +
2
2
2
e
\2
2
e
2
e
i
Combined Torsion and Compression
*iV/Vf,+fi £]
*7*f"+*r
•
Lx
Combined Torsion and Compression
M
N
Combined Flexure and Direct
Stress
>'
P=
F
A+
P
,
Pae
I
P
m
O
Eccentric
7r
2
E\
'
P J
Loading of Long
Columns
*.
rf, A
L
1
*
(i\', («'+«)«'l
p"
'
m*>E\p)
J
STRAINING ACTIONS IN MACHINE ELEMENTS
95
TABLE' VII
PROPERTIES OF SECTIONS
Shape
of Section.
Moment
of
Inertia.
I
Modulus
of
Section.
I
Square of Radius of
Gyration.
"Eolar
Moment, ofi
Inertia.
p'A
.09SD 3
J
TTd*
_
.049d
4
7TD
3
D
.16
2
ttd 4
32
61
IS
m^\
12
D 2+d2
32
7T(D 4d 4 )
32
L
D
2
BH
*BIhlk
t..
::.
B:HCB a+H.2 )
12
12
£
[BH?bl. 3 ]
g^rf"a
j^ f
12
BH 3— bh ] [bh— bh. J
3
i[BH^bh 3]
^[BH 3bh3]
1
12
rBH 3 bh 3 1 [BH bhj
b h.
i[BH bh ]
3
3
^[BH bh uH
3
3
]
3 3 1 ~BH bh 12_BHbh J
~

hH
I
_ (BH 2bh~)4BH bh (Hh)
)
2
fcS
= (B H bh 1
2
B•
b
1 BH bh (Hh) 2 12(BHbh)
2
)
6(BH 2 bh 2 2 2 2 I _ (BH bh )— 4BHbh(Hh) e2 6(BH 2 2bhH+bh 2
ei
)
^[bH'+Bh
BH°
3
]
^[bHW]
e!
I
12(bH + Bh)
tffej
* bH
=
24
!
^BH
12
e2
H
7rBH
64.
;
ttBH
tt(BH 3+HB
64.
;
Eh
or lubricants of any kind are interposed between the surfaces. has been found that many of the older theories 96 based on ex . is ance to relative motion due to the interlocking of the minute de pressions and elevations which exist even in the smoothest surfaces and will. etc. screw fastenings. to a considerable extent. the resistance to relative motion is. of course. are of prime importance to engineer.CHAPTER IV GENERAL THEORY OF FRICTION. keys. fitted together without any foreign matter in between. Friction in General. brake straps. which If oils further increases the resistance to relative motion. great utility and every effort is made the to insure its presence. any effort tending to relatively to tially to move them This resist each other is met by a resisting force acting tangen the surface of separation of the two bodies.. AND EFFICIENCY 27. If the two surfaces are very carefully will. In the case of bearing and rubbing surfaces generally. It experimental work has been done. they in the case of still many substances. unsurfaced cast iron will show a very great resistance to relative motion. while in the case of friction clutches. and the manner of their application therefore. while two hardened and ground surfaces of steel will move over each other with much more ease. overcome. This tendency feature to resist relative motion is sometimes a desirable and sometimes not. The laws of friction. Thus. When two solid surfaces are held in contact by any appreciable force. such frictional resistances result in loss of power and should be reduced frictional resistance is of to a minimum. adhere firmly together. vary with different properties of materials and different qualities of finish. These laws are though considerable at present rather imperfectly understood. LUBRICATION.
fore flat The work of energy is if often there an important factor plates the foot in the design of rubbing surfaces. and • / = c length of bearing. then the intensity of pressure per • unit of projected area. For circular surfaces. and Thurston's "Friction and Lost Work. 40. the distribution of the normal pressure is variable and dependtogether.GENERAL THEORY OF FRICTION perimental work are true 97 of only for the range conditions covered by the experiments. such as journals and bearings. where V the velocity in feet per minute and P pounds." page 7 . N be the number of revolu tions per minute. f == F — or F = f P. ent on the manner it in which the surfaces are fitted In such cases is customary for convenience to define the co efficient of friction in a similar way as in flat surfaces." page 938. and that conditions different from these show entirely different results. then for flat surfaces. if sure per unit of projected area. For is is pounds absorbed per minute E = fP in T. and con sider that it has special values for circular surfaces. or. * E = ldNP = 12 26lS d Np For other forms of surfaces sec Kent's "Engineer's Pocketbook. is be denoted by/. d = diameter of shaft. The energy absorbed by into heat which is frictional resistance is transformed conducted away by conduction and radiation kinds of bearings. The co/«. to the . as before it The normal pressure on circular stated. called the coefficient of friction or if this ratio normal load P. is difficult of accurate determination and is therefore customary to take as the normal pressure the intensity of presOr. In the case of circular surfaces. The ratio of frictional resistance F. surfaces. intensity of F = a P. efficient of friction for circular surfaces will be denoted by Hence as before. and d ^' the diameter of shaft in inches. in the case of certain circulation or other means. by water of friction is to the air. A w = P — .
the frictional resistance and the energy absorbed for any load P may be calculated. except at very low speeds. Rennie. for a full discussion . Coulomb* and many others. furnish the fol lowing laws for dry or very slightly lubricated surfaces. these values from even slight contamination by must be used with judgment. tally for many Values of/ and n have been obtained experimenof the materials and conditions met with in engistill neering. was formerly supposed that an abrupt change took place the value of / when the body passed from a state of motion It in to one of of rest rest. (1) The The frictional resistance is approximately proportional to the normal load. Friction of Lubricated Surfaces. The following values of/ must. and also because of variations which come with slight changes of conditions. The experiments of Morin. of As the velocity increases. the value / materially decreases and this must be taken account of friction is in designing machinery where involved. that while the coefficient in general greater than that of motion. Unfortuis nately the information regarding high or even moderate speeds also very incomplete. (2) frictional resistance is approximately independent of the extent of the surfaces. (3) The frictional resistance. the is change in value is gradual and the value at rest not far different from that at very slow motion. for any pair of rubbing surfaces. is It seems now. but the data so far available are incomplete. however. The consideration of the laws of friction. (a) (b) Friction of Dry or Unlubricated Unlubricated Surfaces. be looked on as approximate values only. in view of the incomplete infor mation. de creases as the velocity increases. as applied to ma chinery. of 28. Unless it is positively known that the surfaces will be kept free oily substances. * See "Lubrication and Lubricants": Archbutt of these points. Friction Surfaces.98 If MACHINE DESIGN then /or \i be known. & Deely. naturally divides itself into two parts.
serve as what may be expected to occur. to If r be expressed in inches k found have a value of about . for the decrease in the value of the coefficient. The most important is use of rolling friction in as far as the present discussion concerned connection with roller bearings for shafting. a coefficient to be deis has been found that P F = — k r where k termined experimentally. . the frictional materially reduced because the surfaces are wholly or . Neither the coefficient k nor the exact theory of rolling friction is at present very accurately known. . . . It is to be parfor cast iron ticularly noted that.02 for iron or steel rolling on iron or steel. .06 in the There are no experimental data giving the decrease value of/ at high speeds. Dry Rolling Friction. 3 to . for combinations such as wood or leather on metals. Friction of Lubricated Surfaces. The data a rough guide to on steel will. and it r = is radius of rolling body.32 .2 = = " " " .5 2 to . " " . in designing brake shoes or other friction allowance must machinery where great be velocities are involved. GENERAL THEORY OF FRICTION Coefficients of 99 F vict ion (J) for Dry or Slightly Lubricated Surfaces..3 —velocity — " " 440 2640 5280 minute. frictional resistance has been found that the socalled due to the rolling action is much less that due to sliding. F = horizontal force required at the axis of a circular body to pro duce and sustain uniform motion. it When a curved body rolls upon than the a plane or curved surface. . for the same load. made 29. is. When a lubricant is interposed between resistance is a pair of rubbing surfaces. however. and a fuller dislater.5 Wood " " " " feet per Metal on Metal Cast Iron on Steel " " " " " " " " " (average) . If P = the load. . cussion of these will be given 30.6 Wood Wood on Wood — Static " " " on Metals or very low velocity " " " " Leather on Metals Leather on " " " " " " " 3 to .6 3 to . " .
IOO partially The If the lubricant may be fed to the surfaces in a number is motion oil is intermittent. such as greases. . Fig. The effect of friction. socalled compression cups are often used and are constructed so as to force the lubricant in between the surfaces. Static Friction and Lubrication. thus lubricating the whole length of the journal continuously. a simple If the hole leading to the rubbing surfaces is often used. 14 (b) shows a called sight feed cup where the it oil falling by gravity from the also used to cup can be seen as lated passes the hole e and the flow can be regu by the screw oil d. Sometimes chains are used instead of solid rings. where very heavy pressures are carried on the rubbing surfaces. (b) Friction of (c) Imperfectly Lubricated Surfaces. by This so sometimes called feed. may conveniently be treated under three heads: (a) Static Friction of Lubricated Surfaces. oil hole. The ring r running loose on the shaft s dips into the pocket below the shaft. The friction of the ring on the shaft causes it to rotate and draw up oil from the pocket. For the most efficient lubrication the journal itself in the bearing so that a made runs in a bath of pressure. and other conditions oil will allow. motion tinuous continuous. the pressure tends to slowly expel the lubricant from between the surfaces. of ways.— MACHINE DESIGN separated from each other by the lubricant. Sometimes an opening pad saturated with lubricant can be kept pressed up against the moving surface. 15) or is flooded with oil supplied under of sup The relative merits of these various methods plying the lubricant will be more apparent after a discussion of the general laws of lubrication. some form of cup which (a) will give a con supply is better. and the efficiency of lubrication of so called lubricated surfaces. For heavy lubricants. 14 (c) shows a "ring oiled" bearing. Centrifugal action is some is extent to feed to rotating parts. 31. When a pair of lu bricated surfaces are pressed together by a load. Fig. 14 shows a cup of oil the simpler type where a wick of cotton or wool draws up the capillary attraction is and feeds siphon it slowly into the Fig. oil (Fig. Friction of Perfectly Lubricated Surfaces.
will Longitudinal Section Fig. metal surfaces and pressures ranging from 75 to 500 in. if not oiled before starting. this action is sufficient to expel so much of the lubricant that may have been between the sur faces while running as to allow the metallic surfaces to less in contact. 15.* per sq. If it IOI is very difficult even with limited areas and heavy pressures completely to expel the ordinary machinery. at will be seen presently that even their low velocities the surfaces tend to It is draw in the lubricant by motion. however. a wellknown fact that heavy machinery always offers a great resistance to starting after lying idle a short time and often the rubbing surfaces.GENERAL THEORY OF FRICTION Experiments and experience show that lubricant. The coefficient of static friction for lubricated surfaces is not very accurately known and it varies somewhat with the pres sure and character of the lubricant. grooves should be carefully provided so that lubricant can be applied as near the point of greatest pressure as possible before motion begins." page. is . .15. See Thurston's " Friction and Lost Work. is allowed to stand at rest for a short period of time. abrade each other before the lubricating action due begins to take effect. to running The and oil materials therefore for the rubbing surfaces of heavy machinery should be carefully chosen for their antifriction qualities. is come more or cated surfaces The static coefficient of friction very hence much higher than that of for it of lubri surfaces which move even very slowly. A fair average value for lbs. 316317.
velocity. or the supply of lubricant limited. in between the Thus the layer of oil it which touches the surface of a journal adheres to and is carried along under the bearing. as even after the machinery has metallic contact been successfully set in motion may occur between them. but what data are . the load remaining the same. When one lubricated surface slides over another. The is action of the sliding surfaces in drawing in the lu bricant less similar to that of the rotating journal. The is formation of this separating film with in creasing speed probably gradual and the character of the contact most probably passes through a gradual change. is more and more lubricant till. thrust between the surfaces by the action noted above at a point depending on the pressure. the moving surface. tends to carry the lubricant. fric This last state is known as perfect lubrication. The exact point at which perfect lubrication occurs for any given load.102 32. and vis cosity of the lubricant. velocity of rubbing. If the velocity of rubbing and the supply of lubricant be in creased. if properly applied. Imperfect MACHINE DESIGN Lubrication. the quantity of lubricant that is carried in very small and the surfaces in contact are very slightly lubricated and metallic contact. for the rubbing sur faces of slowmoving machinery should also be carefully chosen for their antifriction qualities. ing action which results between layers on account of the action of the moving surface is it In plane sliding surfaces the lubricant generally applied to the stationary surface and in spite of the tendency of the slider to rub tends to cling to it off. or the is pressure very high. and lubricant is not accurately known. therefore. surfaces. from contact which tially is nearly metallic through successive stages of par fluid contact to complete fluid separation. may even be in actual The materials. If would naturally be expected from the nathe velocity of rubbing be very low. in turn tends to carry along the layer which next ad because the viscosity of the lubricant opposes the shearof the journal. the metallic surfaces are completely sep arated and the friction becomes only that due to the fluid tion of the lubricant itself. This layer joins it. even at low velocities. but in a much marked degree as ture of the case.
that and pressure remain constant. oily pads. however. as in the case where a journal runs It is an oil bath. 16).e. oil i. with either varying pressure is or velocity. and with good are as follows * : supply from cups or pads. the velocity of rubbing. oil is also found to be modified by the rate at which at supplied. experimental results are very discord ant. The law of variavelocity tion of the coefficient of friction. for instance. the emphasize the great variation It is results only serve to in conditions with change of these variables. and the temperature of oil. f supply) a minimum reached With further increase of velocity the and Thurston's "Friction and Lost the See Archbutt and Deeley. in any way must be between considered as imperfectly lubricated. It is known. to lubricate the greater part of the rubbing surfaces of machines in this manner and. if evident. normal temperature. impossible or inconvenient. restricted etc. the exact condition which the supply will exist such surfaces depends on the pressure. The generally accepted theories for imperfectly lubricated bearings running under average conditions. all surfaces lubricated holes. (a) Starting from rest with constant load. the coefficient of friction first increases slightly with increasing velocity and then feet decreases. to be noted that this discussion and coefficients given refer to circular bearings and friction of rotation. As already noted. however. therefore. and character of the lubricant. and while an immense amount of work has been done. that perin cannot be obtained without a plentiful supply of the lubricant. . until at a velocity ute somewhere below 200 oil per minvalue is (and depending upon the (see Fig. the bearing as affecting the viscosity of the so Naturally where many variables exist. or is is supplied by socalled forced lubrication where the lubricant delivered under pressure. page Work." pages 296312.GENERAL THEORY OF FRICTION 103 available will be given in connection with the discussion of perfect lubrication fect lubrication which follows... almost any condition may be produced from metallic contact to perfect lubrication simply by varying the supply of lubricant. is by such means as simple oil where the supply of lubricant oil cups. fit is 58.
how may and be expressed as follows: every combination of presvelocity requires a lubricant of a certain viscosity for sure best results. usually met with in machines. in the friction. Its general characteristics. (b) With constant velocity is and very light loads (see Fig. decreases very rapidly at first.015 . 17) the coefficient of friction the coefficient As the load is increased. and its fluid fric which constitutes the greater part be less of the resistance in such oil.io4 coefficient increases MACHINE DESIGN till the temperature affects the viscosity of the lubricant to such an extent that abrasion and failure occur. will tion. an increase of temperature decreases the viscosity and may. At high speeds and light loads.01 . owing in friction.025 Coefficient of Friction Coefficient of Friction Fig.100 / .025 . be readily drawn in between the bearings. will than that of a heavier its Increasing the temperature of a lubricant decreases viscosity and. to the expulsion of the lubricant.01 . Fig.005 . i 6. . cases.005 Ah . for sometimes a change of lubricant is suffi . 17.03 .. would cause a decrease in In the case of the heavier loads and lower velocities.02 . slowly till pressures of about 100 to 200 lbs. coefficient again slowly increases.015 . and then more very high. thin oil 600 500 / • 500 3400 a / / c m u i 300 ft / &300 m C / 8 200 g200 100 \\^ . per square inch are obtained (c) when the The law of variation of friction with temperature is very complex and not well defined. a light. give an increase Care should therefore be used case in to obtain an oil suited to the hand. above case therefore. ever.02 .
02 to .. the average value of is . the velocity is exceedingly low. With good lubrication and moderate velocity it may be as low as .005. 17 it must be kept in mind that. are not to be taken as giving exact values of the coefficient but serve to show graphically the general laws by In interpreting such curves as Fig. 105 to on the other hand. load and the coefficient of friction. is evident that a state .008 and. The curves taken from a number Figs. From Figs. When considered that the temperait ture also greatly affects these relations.GENERAL THEORY OF FRICTION cicnt to cause great trouble or. the coefficient approaches that of static friction of lubricated surfaces.. the coeffi cient decreases as fast as the load increases. The curves show. for pur poses of design of ordinary machinery. A fair average range for pressures from 50 to 500 and velocities from 50 is to 500 ft. is which it varies. is from . which between is coefficient of friction. The results of lubricated bearings generally from the lowering of the viscosity by increased temperature. and again with low velocity and poor it lubrication may rise to . // 16 and 17 show the variation of for varying velocities and pressures. the frictional resist ance will remain constant. for instance. light and throw some of the relation on proposed changes in running speed of and the machinery already installed. where best results may be expected when designing new machinery.05 or more. 16 and 17 are com of actual experimental re They /'. is reduce the failure temperature imperfectly of a bearing that heating. the actual frictional resistance like decreasing or in may is not be changing in the product of the manner. however. and more dependent on the character of the rubbing surfaces. exists They it also indicate the complexity velocity. When which lbs.015.15. posite curves sults. to may be is taken at . pressure. while the coefficient creasing. It be noted that with imperfectly lubricated surfaces and low the coefficient of friction less velocities dependent on the character of the lubricant. so that the oil film is no longer maintained and metallic contact and abrait is sion ensue. The frictional resistance If. per minute. the foregoing evident that the coefficient of friction for imperfect lubrication will necessarily be a variable quantity.
a maximum oil pressure of was found near the middle point of the bearing. Beaucamp Tower experimenting with journal that with a journal friction (see Proceedings of Institution of Mechanical Engineers. draw more or less of the lubricant in between the journal and bearing. impracticable. per sq. it is found that this action is so marked that the rubbing surfaces are completely separated by a thin film of lubricant friction itself. Tower succeeded in this way in carrying a load of 625 pounds per square inch of projected area at a velocity of 471 ft. It has been proved mathematically. the of oil under presfluid borne. run in an oil bath. tends to be crowded back from the side where the lubricant in is is carried in. as shown in an exag gerated manner pressure half the figure. It has been to oil. per minute. ^. and verified experimentally. by means of the molecular attraction between it and the lubricant. 15. Mr. conditions are as follows: the journal. rising to a maximum at the middle and falling to zero at the edges of the bearing. Perfect shown in the last article that any rotating journal will.106 MACHINE DESIGN of these relations for imperfect lubrication. little The consequently greatest at a point a more than way beyond the centre of loading where the distance be tween surfaces is least. being slightly smaller than the bore of the bearing. that the conditions which exist in a bearing running under these 625 lbs. velocity of about 150 ft. 33. the amount so drawn in depending on the velocity and pressure. in the is ment all form of a general law or mathematical expression. giving a wedging effect. 1883) found and bearing to above action was so marked as sure such that the load form a film arranged as in Fig. inch. If the journal be allowed Lubrication. With a load of about 330 lbs. combined with the viscosity of the lubricant. or is otherwise plentifully supplied with and the velocity be high enough for the pressure carried. . and the becomes only that due to the fluid friction of the lubricant Mr.nd a per minute. was completely The distri bution of the pressure in this film was found to be as indicated by the diagrams above the crosssections. and such expressions are misleading.
* where w is in pounds per square to inch. Sept. to serve as a general guide for average conditions. and show that the laws of friction for perfectly results of The lubricated surfaces. and seem check fairly well with con siderable other data. which accepted as accurate. values obtained by Tower with shows the simultaneous where frictional resistat least well ance was a minimum. would form considerably below the values given in In Moore's experiments. The values given by this expression arc plotted in 1. 16. formed.400 450 500 Feet perTVTinute Fig. The values obtained by Moore are on the safe side judged by Tower's work. 18. indicating that the film was 300 No. Fig. and pressures. as in Tower's. likely that exact limits can ever be set. are quite pressure. that a film Tower found Curve ture (2). 18. 2 No.v.47 y.GENERAL THEORY OF FRICTION The is it IO' exact relation which must exist between velocity and is not known nor Enough is known. and v in feet per minute. and conclusive. and probably do not indicate the very lowest point at which a film will form. F. to allow this pressure film to form. was constant Moore's experiments were on mineral Tower's experiments are very concordant oils. pressure. Curve is (3) shows similar values for mineral oil. . Curve (2). 1903. Curve No. Fig. ^~ ! SO 100 150 200 250 300 350 . very nearly.' 3 vzl ^100 my/*~^ JJOji. Moore found that minimum will just limiting values of pressure and velocity where the film form may be approximately expressed by the expression \ w— 7. the temperaat oo°. 18. r . the coefficient of friction varying as the square root velocity of the and inversely as the American Machinist. for ordinary speeds definite. however. for circular journals the Professor H. olive oil.
00093 .00227 •00367 . 262 ft.00243 .00503 . and the low values of the coefficient as compared with imperfectly lubricated surfaces.00239 . for any velocity.00107 .00122 •OOI95 . 573 520 4i5 363 258 i53 IOO . under more ideal conditions.io8 MACHINE DESIGN for olive oil the relation is expressed Thus P very closely by = y/ v .00213 •00334 .000 ft. but such low values must not be considered as attainable under ordinary practical working conditions. per Min. while there is no good reason why such coefficients as given below cannot be obtained in wellconstructed machinery. Table VIII will serve to show the remarkable regularity of the results. temperature the product of the frictional resistance load.00714 Tower's experiments have been amply verified and may be accepted as reliable for the range which they cover. Much lower values have since been attained in oil testing machines.00095 .00119 . 314 ft.00162 . That practically constant with to change of This was actually found be the case in the experiments. Their experiments show that for velocities between ft.000 ft.00663 . per Min. is and w will be a constant.OOIIO . that for any fixed velocity and is. per Min.00125 .00619 •00139 . 157 ft.00105 .00149 .00147 . for a given load and temperature.00118 . TABLE VIII OIL BATH OF RAPESEED Coefficients of Friction for Speeds as Below.00178 .00132 •OOI33 .00108 . Inch. 419 ft.00102 . This point is . X) have extended the range of knowledge on this point to velocities over 2.00084 .00277 .00300 •00576 .00140 . per sq. per Min.00142 .2 w . The ex periments of Stribeck and Lasche (see Chap. It follows from /* this.00200 •0O357 . varies as the 5th root of the velocity.00126 . per Min.00115 . is independent of the velocity.00107 . 500 and 2. Load in lbs.00423 .00158 •00155 .00139 .00396 . 366 ft.00148 .00162 . per Min.00267 .OOI30 . 47i ft. per Min.000 per minute the coefficient of friction. 209 ft. a variation of pressure per square inch from ioo to 500 not appreciably affecting the resistance. per minute.00134 . and beyond 2.00960 . 105 ft. per Min.
and some escapes by radiation without doing any work whatever on the piston. is A considerable portion of the heat so received lost by con densation of steam on the cooler cylinder walls. where 34. all has been pointed out that is the energy supplied to a machine that not transformed into useful work. Summary. and in a greater degree on the character and amount of the lubricant supplied.Efficiency. c re The friction of perfectly lubricated surfaces depends very little on the character of the rubbing surfaces. its X in connection with the design is of bearings. varies very nearly as the ft. square root of the velocity for velocities up to 500 per minute.000 ft. independent of the intensity of velocity'. per minute.GENERAL THEORY OF FRICTION discussed still IO9 further in Chap. The frictional resistance of perfectly lubricated surfaces within the ordinary limits. and a careful distinction must be made between certain of these ways in order to get a clear of it is some always and doing useless work. but depends mainly on the character of the lubricant. Thus the steam engine re ceives its supply of heat in the form of steam under pressure. and part is . per minute. but lost in overcoming frictional resistances There are many ways in which energy losses may occur in machines. and practically independent of the velocity for values 35. It above 2. definition of the term efficiency. for any given pressure and temperature. and must be kept very low where this supply stricted. Of the energy actually applied to the piston. (d) is. part is transformed lost in over into useful work at the driving belt. pressure and dependent only on the (e) The coefficient of friction of perfectly lubricated surfaces.000 ft. principal application found. (b) The load that can be successfully carried on an imperfectly of lubricant is lubricated surface will vary greatly with the amount supplied. statements (a) From may be made: the foregoing discussion the following The friction of imperfectly lubricated surfaces depends partly on the character of the surfaces themselves. approximately as the fifth root of the velocity for velocities beis tween 500 and 2.
the amount of work to be done is fixed. The first class of these energy losses called leakage losses. Efficiency has been defined (Art. and no machine can transform the energy supplied into useful work. Hydraulic and electric machinery have similar might be as losses elements of loss. or to that portion only of the total energy which the machine transforms into useful and useless work. in general. Thus. since. and trans forms 200 units into useful work. Thus the strength of a . frictional losses of the machine.20.000 thermal units.000 —= . is had to total energy supplied. and the source of energy must supply enough more energy than this to compensate for the many machines . a large part of the heat of combustion escaping to the jacket water or to the at mosphere by radiation.IIO MACHINE DESIGN frictional resistances just discussed at the various coming the constraining surfaces. for the design but the it does not need to be considered mechanical efficiency can seldom be neglected. The gas engine is subject to similar losses. and 50 units into the useless work of friction. 2) as the ratio of useful work ence and from the above it appears that a machine may have two efficiencies depending on whether referto energy supplied. and the mechanical efficiency lute efficiency is is — 200 = . its absolute efficiency is 200 — 1. and doing no work on the piston. The consideration of abso 250 of beyond the scope of this work. but must lose some in friction or other wasteful resistances. losses of the medium which machine itself used to transmit the The all losses in the are known as frictional and are common of it to all machines. as they are of the same character is by actual leakage energy. a gas engine supplied with 1. The mechanical efficiency of any train of all mechanism is the continued product of the efficiencies* of * It the several pairs of used in a number of ways may be noted in passing that the term efficiency is other than as the ratio of work done to energy expended. These is efficiencies are respectively known as the Absolute if Efficiency and the Mechanical Efficiency.80. while only a part of the energy actually applied to the piston reappears as useful work.
in when what really is meant is is its relative strength. divided by the total energy supplied. and their respective efficiencies be e. The sum of all the work. the trains are arranged in parallel is so that the total energy several trains of transmitted simultaneously through train transmitting only a por mechanism. If. however. evident that such efficiencies are of a different character from those discussed above and do not enter into the calculations of the efficiency of the machine. the ratio of the air actually discharged per stroke. the efficiency of the whole machine. to the wholecalled the volumetric efficiency. by reasoning similar is to that in the last paragraph. A machine may consist of several trains of mechanism. and the total K amount of energy available for transformation into either useful or useless work. Let let E be be the mechanical efficiency of the whole machine. the above reasoning for the efficiency of the whole machine does not hold. the efficiencies of the several constraining sur faces of a machine are known. till the amount of. each tion of the energy. e 3 .GENERAL THEORY OF FRICTION constraining occur. delivered by all the trains. e v e 2. efficiency of the therefore. the amount of energy supplied amount of work which it will deliver If the can be computed as above. . the continued product of the efficiencies of the several trains of mechanism. energy delivered by the last element (or the work done) is K (e X e X e ej. unpunched plate. in the manner indicated above. If these several trains are arranged in series so that the energy passes from one to another consecutively. an air compn >or. Again. amount It is raised to the required pressure per stroke. ly x K 2 But the mechanical efficiency of the train is work done energy supplied K X et (e X el X e2 e n) K X e2 = {e ej. will be the efficiency of the If. III losses let surfaces in the train at which frictional Let any machine have n pairs of such surfaces. whole machine. e n . and the amount which the second delivers to the third is Ke X e and so on. as a whole. to each train is known. e 4. is called the efficiency of the joint. Then. the amount of energy which is the first pair of constraining surfaces delivers to the second X e. the mechanical compared to the strength of the original riveted joint.
tions are constant. depending on the manner in which the load Thus. when all other condisame pair of constraining surfaces will have an entirely different efficiency for the same amount of is power transmitted. any machine element is. it is impossible to calculate precisely from the analysis will of a design if be. therefore. If the coefficient of friction for any constraining surface could be . the tem perature. is far new ones on For all efficiency tests If machines of more accurate to base made on those already tests standard machines such results have been made. particularly the what the mechanical efficiency mechanism is at all complicated. the revolutions remaining con In general. however. But when a matests are ef be designed. as a conse quence. the alignment of the surfaces. doubled. the design of in existence. and no recorded be had that will give any information as to the probable ficiency. belt on the periphery of the wheel. and the recorded sign of chine to new machines of a new type is of like to form a valuable basis for the decharacteristics. an estimate must often be made and the efficiency cal culated as outlined above. Furthermore. is the diameter of the wheel in a like ratio. however. the and the bearing pressure. In general. stant. the of a given amount of power will bring a certain definite frictional load on the bearings. for the same power transmitted. and. the velocity of rubbing. which comes ef In such cases a knowledge of the machine elements becomes necessary. If. the frictional resistance at the bearings will be reduced to onehalf the original value. consider a simple wheel and axle driven by a applied. be onehalf of the former value. and the making of such estimates only with experience. ficiencies of various a great aid to the development of that judgment in such matters. it possible. a close approximation is can be made. With a given diameter of transmission wheel. the belt speed is increased and the belt tension will.112 MACHINE DESIGN The mechanical efficiency of whole machine can be calculated. a variable quantity. is though a reasonable approximation a similar type have been built. for the will coefficient of friction of any pair of constraining surfaces its vary with the lubricant and method of application.
in the quantity varies with the velocity of rubbing. including bearings Bevel Gear Cast Teeth. including bearings 96 92 95 Bevel Gear Cut Teeth. as used on bicycles High Grade Transmission Chains 8 97~99 .: GENERAL THEORY OF FRICTION accurately determined. The dent is following are rough average values of the efficiencies of the most common elements. long lines of shafting 9698 95 Roller Bearing Ball Bearings 98 99 93 Spur Gear Cast Teeth. with changes bearing pressures. and such methods of computation are cumbersome and to be attempted only where a very is close estimate required. including bearings Worm Gear. to some degree But. as before noted. efficiency with it 113 calculate its would be possible of certainty.. necessarily etc. singly Bearing. see Art. varies with thread angle. For more accurate values the stu referred to the respective discussions of these various elements which follow Common Common Bearing. including bearings Spur Gear Cut Teeth. 54 Belting 9698 _ 95 97 Pinconnected Chains.
As a means of storing energy. IV. often given some one of the forms commonly Springs are in called springs.CHAPTER V SPRINGS 36. common I. for (as tion of the member were it zero. or as a secondary source of energy. etc. 24) the stress due to a suddenly applied load would be infinite if the corresponding distorquiescent load. For weighing as in spring balances. Every machine member is is. The Principal Applications of Springs. Springs are characterized by a considerable distortion under a moderate load. constitutes III. what Reuleaux has called "force closure. as in clocks. a spring. is To meet this last requirement member 37. An absolutely rigid material would be practically unfit for the construction of any member subject to other than a perfectly shown in Art. By proper selection bution of material possible to control (within wide limits) the degree of distortion under a given load. etc. II. This in maintaining contact between a cam and its follower. there are many load the cases in which considerable distortion under is moderate desirable or essential. which would otherwise be incompletely constrained for example. for no material and distri absolutely rigid and the application of a load always produces stress and accompanying it is strain." For absorbing energy due to the sudden application . For controlling the motions of members of a . in a sense. use: forces. Distinguishing Characteristic of Springs. While of is usually desirable to to restrict the distortions most machine parts very small magnitudes. dynamomemechanism ters. of a force (shock) as in the springs of railway cars. etc. 114 .
are not infrequently employed as springs. The characteristic distortion of the solid springs is a change in form rather than of volume. cisely defined. maximum load and distortion which often need not be very preis The use of springs for storing energy (as the term spring light is ordinarily understood) pieces of almost wholly confined to little mechanisms or to operate apparatus requiring but power them. The links. Springs are usually of metal. although other solid substances. or cushions. but brass (or some other alloy) is often used for lighter ones. move relative to the shaft with any variation of the above correspond to the indicating mechanism of ordinary weighing devices. it is fre quently only required that the shall lie within certain limits. Softrubber cushions. or other connections. In those governors which use springs to oppose the centrifugal. or other inertia actions. the springs automatically weigh forces which are functions of speed. is the most common material for heavy springs. or other compressible fluid. A in confined quantity of air. A high grade of steel. as wood. or buffers. action to irv. is used many important applications to perform the office of a spring. . of 38.SPRINGS "5 in An important to class of a mechanisms which springs arc used weigh forces is common type of governor for regulating the speed of engines or other motors. designated as spring steel. in and these are in some respects intermediate in their It is usually not necthe two classes mentioned above. these simple buffers. are sometimes used. while the inci fluid springs are characterized by a change of volume with dental change of form. usually the most exacting as to the relation between and the distortion of the spring throughout the range of In the second and third classes of application. to secure a very exact relation between the loads and the distortions under such loads. The airchamber of a pump with its inclosed air is a familiar example of what may be called a fluid spring used to reduce shock ("water hammer"). which forces. Materials Springs. 1 or of change of speed. The forces first is of the abovementioned applications —the weighing of — the load action.
or bend them. Forms of Solid Springs. Flat Springs are essentially beams. flexure These springs are subjected to and the resultant stresses are tension in certain portions of the material. the character of the form I. hence the following treatment Springs be limited to solid springs. either cantilevers. of the spring which depends both upon and upon the manner of applying the load. may be subjected to actions which extend.n6 The will MACHINE DESIGN discussion of the confined gases (fluid springs) is not within the scope of the present work. as in all beams. . producing '<^^^^>: *» j^ A >« <&*e stresses in the material. shorten. and compression in others. 39. the shear may the load is when applied. or with more than one support. twist. with a transverse shear.
and the may be computed sepasum of the deflections of the commonly used to resist separate springs of the II. either to clear other bodies. leaves. 30. In such cases. although is preferable that the form of such springs approximate those' of the "uniform strength" beams (Figs. See Fig. It will appear from the discussion of these laminated Fig. Helical. or to give the spring an advantageous form when it under normal load. with constants appropriate to the particular material and form of beam used. It is often desirable or practically necessary to build up these springs of several layers. . producing a laminated spring. or plates. as in the " ellipti Two or more springs may be compounded. Fig. springs that they may be properly treated as a modification of one form of "uniform strength" beam. These are sometimes improperly called spiral springs. of uniform cross section (Fig. or twist the spring relatively to its longitudinal axis. rately. or Coil Springs are most actions which extend. shorten. as a spring The neutral surface of the beam used is may be initially curved. 30 (a). each spring total deflection is the set.SPRINGS usually 117 be neglected in computations. 19 or 22). 23 or 24). Flat springs may be simple prismatic it strips. The ordinary flat beam formula? for strength and rigidity may be applied to springs. 20 or 21. 27. cal" springs or in the platform springs frequently used under carriages.
in it or may be considered as a modified helical spring in which the radii of the successive coils are not equal. and are usually employed when a very large angle of torsion between the two connections is necessary. IV. : sides of the neutral axis. 30). properly so called are those of the the familiar clock spring. the predominating stress stress. It is thus intermediate between the two preceding classes.Il8 MACHINE DESIGN The stress in the wire (or rod) of which a helical spring is made is somewhat complex. Spiral Springs. The form of spring represented spring may be looked upon as a spiral set. the stress in the material is that due to flexure or tensile and compressive stress on opposite III. is shortened by the load. the predominating is stress (with ordinary proportions) is a secondary tensile stress in the wire. The used on the ordinary disc valves of pumps are often of this . In these cases. with a secondary compressive is When that the helical spring subjected to an action which twists the spring (as a is whole) the principal stress in the wire (tension and compression in opposite fibres) due to flexure and the secondary stress is torsion. is In a "pull spring. and given a permanent the direction of its axis. This spring used only springs as a push spring. car trucks. In this form of spring. is This last form not usual in machine construction. HelicoSpiral Springs. In a "push spring. Helical springs are sometimes arranged in "nests. or both. (Fig. it may be undesirable employed otherwise is to to advantage where is difficult or constrain the spring against buckling." one which a torsion. This last arrangement is common practice in form of These are best adapted for a twist relative to the axis of the spiral." springs of smaller diameter being placed within those of larger diameter. to resist a compressive action." one is tor which sion. the different springs of a set are com puted separately. consisting of torsion combined with tension or compression. and there extended longitudinally under the load. though it has the advantage over the siderable lateral common and it helical spring of con resistance. by the common upholstery spring which has been elongated.
Extension. (minus). The springs. of course. Pull. following summary the kinds of loading to which they are subjected. p = intensity of stress in outer fibres. are the will forms of rectangular section beams. Twist. in this sense. Helical Spring. Torsion. and the predominating stresses resulting from the different loads. II CI << 11 Spiral " Flexure or Bending. Predominating Stress. 41. etc. with the edges of the strip parallel to the axis of the spring. Every line of shafting is gives the ordinary forms of solid necessarily a spring. and Compression. as in the following . (plus). shown by Figs. Occasionally straight rods. = free length of the spring. and Compression. These are comparatively stiff springs. usually of circular or rectangular crosssections. and torsional. Tension Torsion Torsion Tension Tension and Compression. Car springs are sometimes made of a flat strip or ribbon of steel wound in this general form. The These six of flexure. h = dimension of this section in plane of flexure. GENERAL SUMMARY OF SPRINGS Form of Spring. Computations of Simple Flat Springs. Torsion. b = dimension of this section perpendicular to plane E = modulus of elasticity of material. are employed to resist torsion relative to their longitudinal axis. Push. to 24. V. / = moment of inertia of most strained section. the stress is. Load Action. springs with rectangular P = load / applied to the spring. flat The following notation will be used in treating of crosssections. o = deflection of the spring. be designated Type I. Compression. 19 most important of those usedassimple flat springs. Flat Spring. as they will close up flat between the valve and guard.SPRINGS II 9 form. Twist. II..
MACHINE DESIGN which gives the constants to be substituted in the general formulae for computations relating to each type. with load at free end. The last equation (5) may be used for all computations as to is rigidity of flat springs (beams). provided the elastic limit not .\Pl = pbh 4 . A I /3 B 1 K 1 c 3 2 As per " " " " " Fig. 2 . PI = \pbh .120 table. 1 PI Or beams may be pbh' (2) the general formula for the strength of rectangular section written PI = A pbh2 In which the coefficient (3) A has the values given in the Table. 6^ (1) z For the rectangular section cantilevers. The ends and loaded at the middle (Types 2 III). for the types under consideration. TABLE IX Coefficients. 19 2 3 2 2 3 1 1 4"8 1 "32" 1 4 ~8 6 II III " " " " " 20 21 3 1 T 1 1 1 " " " " 24 1 3" 2" 3 1 6 IV 22 23 IT 1 4 6 8 2 3 I V VI 1 2 1 "6 6 6 " 24 2 4 3 3 1 theory of strength against flexure (equation J and tables and 2) gives: For rectangular section beams supported at the I. The theory of elasticity of beams gives PI or for rectangular crosssections 3 EI PP (4) (5) In which /? and B are as given in the Table. . II. Type.
5 X S 12 = X 27 — M+ n 2 . the spring not only takes a permanent but the required ratio of the load to the deflection will not be secured. stress under the eq. plate. A change could be assumed. but it would demand a wider spring for the required stiffness. .000 to be ^ = 1. assuming the modulus of elasticity. the this limit. is is P— 1. A more general method will now be com given. £ = 30.84X9 beyond the = II2 ' 5 °° Ibs  P £r S 1 mch  This steel. spring. by which it is possible to determine the proper spring for trial given requirements without the necessity of successive putations. for is the elastic stress set. This gives the width deflection to load. may be used for this To illustrate. passed. inches. provided the stress limit. expression 121 The is only constant for the material which enters this the ratio of stress to strain elastic modulus of elasticity which holds up Equation (E). the load at the middle load lbs. this is simply the but not beyond. to. (5) : = B _ PI 3 =rrr3 holt 1 1. hence any computation made by (3) formula should be checked for safety.5 inches. purpose. of spring for the required relation of the is.000. it is often important for is economy of material to use as light a spring as or. It is produced does not exceed the elastic if necessary to check the spring as found above. . A thinner plate would reduce the stress. On the other hand.000. other words. it that gives a spring of the required stiff ness.000. (3) : maximum 3 From P = PI = Ablt stress is it X 1000 X 30 X 64 " 2x2..000 4 30. From b eq. as and new computations made with the new data.— SPRINGS exceeded. in consistent with safety. elastic limit of any ordinary grade of hence is probable that some different form of spring should in the thickness of the be used.000 X = ~ X ~ 27. I) matic spring (Type assume that a rectangular section prishas a length between supports of 1 = 30". the deflection under this and the spring is made of a single Required the breadth (b) of the strip of steel H inch thick (h). it is important not to have too low a working load.000 x 1.
h and I 50. (3) : . per sq.000 X 900 rr PF = ~ mch. h will be somewhat less than 50. the the material. = . K 1 .000 lbs.: — — MACHINE DESIGN 122 From eq. and the material are given.000.J» and the span (/). : r c Pl .5 Taking h = ^ inch.* required the dimensions of the crosssection.000 x 25 to .000 X 1.000. (6) and (7) : Plh A p From eq. (5) Ap BPP BPl pP w bh3 = ^r pP w From eqs. or p : 50.000. J .X 2~2 =K ^— Ed 6 6 30.000 and the safe maxi mum working stress be From eq. (9) : =C ~Pl —— 9 pnr is 1000X30X1024 3 = = . applied at the middle inches. (9) The two signing a equations spring (8) are in convenient form for dedeflection (#). b. and a load of 1. in. If to cause a deflection of 1.2 inches.5 the modulus of elasticity be 30. 1 PI (9) h TJre.X 2 47. flat when load (P).000. to use a regular size of stock.. \ p = 47. 3Q. ° * If the spring provided with stops to prevent deflection beyond a certain amount. section prismatic flat Example is The span of a rectangular spring (Type I) is 30 inches. (8) : taken at 50.000 lbs.000 \\f% :. From b eq. p J . (3) :— PI Plh Ap From eq. stress due such deflection may be nearly equal to the elastic limit of that is A very small factor of safety is all necessary. .
y h (10) K ' 2 p P~— is («) simply proportional (/>). . /. ." page 250. v P EI __ = ~TT = EI M ' —. It appears from eq. It is evident that (/?). so that eq. and the [See eq. the outer fibres will undergo a definite strain which not dependent upon the width of the to beam (b). Springs. (b) is If the breadth of the beam increased. and modulus of elasticity are intensity or known. nor upon the force required produce this change in relative positions of the molecules. the unit strain multiplied it by the modulus of this stress elasticity equals the unit stress. . (8). (h) will With the value of the thickness frequently happen that a simple spring breadth (b) thus limited require exit will cessive to sustain the given load. in connection with E.] it working stress has been assigned. Laminated. but the intensity of stress is not affected by these changes alone. It was shown in the precedflat ing article that the fixed maximum thickness of a simple spring is when the span. rela This same conclusion may be reached from the following tion. (10) or (10) The span /. or Plate. (8) may be used. 42.(8). determine p. the force (P) required to produce the given de(<?) fection will be proportionately increased.* in which /> = the radius of curvature due to load. As h. either the load a P or the breadth of spring of given span (/).SPRINGS If this 123 width is inadmissible. and is often necessary to use a spring built up of several plates or * leai See Church's " Mechanii 3. deflection. a laminated or plate spring article. and the deflection 8. (n) that the stress (h) to the thickness and the radius of curvature for any given value of E. (n) may take the place of eq. if beam (flat spring) and thickness is caused to deflect a given amount is («).= pi 2 Eh . Equations and (11) are important in connection with the theory of laminated springs. It will See next be noted that equation does not directly involve b. follows that may be computed from and <? (which determine the strain).
000 ' X 900 30. Table IX h J rr P? = ~ X = K—p7T ho 60.8 inches. SS/i inches. 4 30. based is upon the " uniform strength" It much better for the ordinary conditions. . the preceding article becomes: .000 lbs. with load at the middle of the span (Type to I) meet the above requirements would have: h = K ±=rr =tX 6 h Tr pP 1 — I 5° 60.000.000 X 000 = . 124 MACHINE DESIGN Example: in. PI ^ —7= 3 = =G X 2 p h 2 00 7 60. then. and it will usually be better to modify the spring to permit using standard If a thickness of %" be assumed for the leaves or plates.0225 . dicated in Fig. per sq. inches. to the simple spring computed above.000.000 ^_ [__ 60. an which has just been developed might answer in some other form. and £ = 30. that the spring is to be of Type In the simple spring. may be developed example except as follows. The thickness of . = 2". each 33.22 e. "* . taking the same data as the preceding II.5" wide. / = 3o".. Type II. P = 1. Suppose be split into six strips of equal width. 25.000 lbs. A simple prismatic spring of /> rectangular section. 20. 26. beam (Type II). .000 X 3° = x .15 inch thick is and ^)4 evidently an impracticathis plate any ordinary case. the spring would be exactly equivalent* as to stiffness and intensity of stress. ble one for .000 1.0506 14.. laminated spring for the case under consideration may be in derived from this simple spring by imagining the lozengeshaped plate to be cut into strips which are piled one upon another as size of stock.. 2 c A  PJ_ 2 TT2 p h = 3 . °° This spring. .15 inch. (8) of the stress.000.225 inches does not corre spond stock. While the form of laminated spring cases.000 x —= 2 . .X x 2 X 30 w _^ = x . except for friction between the various strips. Fig.000.000 x . as found from eq. with a span of 30 inches.3^6 = 5. and that these strips are piled upon each other as in Fig. £ = 60. consisting of a plate inches wide. to a regular commercial however.
G. 27" is wide.6^5 = 5. 900 With // = TV. be replaced by 30" span.000 X 2 = 16 X ^0. 41. depending upon whether Is a combination of the triangular and and Type the spring I. remembering that b . mum plates allowable width of spring is a larger number if of may be each necessary. n = b. Thus. R. steel f\" thick might be used. 4 when p = X 3 X ^0. f\" thick. This construction makes the spring prismatic types (Type II IV. A.6" wide at the 11" 5 equivalent strips. not strictly applicable The when last of these formula?. In cases where the maxifixed. in the preceding it problem. the usual 27.SPRINGS I2 5 p r = hEo = A r / 4 x . These springs approximate uniform strength beams.000. is the ends of the plates are cut square across. Mr. Fig. but it may It is generally be used with sufficient accuracy. and p 3 2 = 50. provided the successive plates are regularly shortened by uniform amounts.000. wide (nearly ^ l see Fig. each 25.000 X 9 „ If this spring.700. (9). of plates b { Or. if is the number and b x the breadth of each plate in the laminated spring.000 X 2 = 66. construction that shown by in which the several plates tri have the ends cut square across instead of terminating in angles. quite common practice to have two or more of the plates extend the full length of the spring. and 25. S. a laminated spring of good form A"). middle. or a cantilever). is necessary to use 6 25. b = c " /> = /r 1. 900 If this stress is considered too great./ is the breadth of the equivalent simple spring.6^6 = 4. the spring width must be kept within 4>2"> plates. and practical dimensions will result. In actual springs.000. is or Type V and Type at the supported in ends.000. 26.000x30x256 = 50. . and may be computed by equations (8) and (9) of Art. eq.25 X ^0. Henderson discussing the cantilever form (Trans.
uncommon to is make the longest plate thicker than the others. The several plates of a laminated spring are usually secured by a band shrunk around them at the middle of the span. XVI). may be deducted from I the full span to give It is the effective span to be used as in the above formulae. hence the thicker plate is subjected to the greater stress. and onehalf the length of the Fig. See equation (11). 27 is initially curved (when free). full but one plate given the length of the spring. the deflection would be decreased approximately onefourth of the difference between 6P1 3 Enb By x h3 and ..h 3 so for onefourth of the leaves full length. • 11 PI 3 32 En b x h3 This may be the otherwise stated as follows of fulllength leaves is onefourth the total When number instead of number of leaves in the spring. but it must be remembered that . for a spring loaded at the middle and supported at the ends. as are subjected to the same change in radius of curvature. says: "For a spring with length we would have (see eq. Type V. The following formulas (derived from the preceding) flat may be used in computing springs. then reduced to a minimum.or 3 5. stiffens the This band spring at the middle.— 126 : MACHINE DESIGN E. — all the plates full API' Enb. 27) band (% not if I. use ^B (8) instead of B and \± K K in equations (5) and of the preceding article. which is common practice. 5) » M. shown in Fig. all This cannot of the plates be looked upon as desirable practice. with onefourth the plates extending the whole length of the spring. as the case may be. however. The best results are obtained by having the plates straight when the spring is under its normal full Type II The or spring load (if this is practicable) because the sliding of the plates upon is each other. —4PI — Enbh 3 .5P/ 3 " Enbh ' 3 similar reasoning. Vol. the values of B and K being those given for the triangular forms. with the vibrations.
3 3 (I) v J *. more than one plate has the full length of the spring. II. and IV. over them. but when it is important accurately determine the scale of the spring actual tests must be made. n bx h . EQUATIONS. which will be observed by These computations will be sufficiently to exact for many purposes. when onefourth of the total number of plates are full length. for the plate maximum thickness. the having the stress should be computed III. When If the plates have different thicknesses. below. II. an appropriate modification of the values of the coefficients B and K B should be made. (ratio of deflection to load). onehalf the length of the band to be sub tracted from the length of the spring to get the effective length of the spring.SPRINGS there is 127 always liability of considerable variation in the modulus to of elasticity. \\ B and \^ K should be used instead of I. When the several plates are secured by a band shrunk. Thus. and K (Type II or V) in equations III. hence such computations can only be expected give approximations to the deflections tests of actual springs. In using these formulae the follow ing rules should be observed. is or forced. V  <v» = i^C (V!) P= bt fL n (VII) = r^p = A p h ^1 p h n (VIII) . O = B _ E PP . I.b ^p <») '"£45 >*£ <»» <v> ¥.
In the practice of a prominent eastern railway company. a component This direct stress increases as the pitch of the coils in creases relative to their diameter.77. and two equal and opposite will forces. along the axis of the helix. If two equal and opposite + P and — lie P. a rod or wire be wound into a flat ring with the ends bent in to the centre. for due to a given load. for: Plates " \ inch T\ iV " thick. then b t deflection found by equation (VIII). be applied to these ends (perpendicular to the plane of the ring) as indicated. the values allowed for the maximum intensity of stress in flat steel springs are. etc. which the material can be subjected. in. rod be If a longer wound into forces. 43. " " " " " i p= 84. act on these ends.000 " " " The above values are satisfied by the equation p = 60. the rod be subjected to torsion. they induce a similar stress (torsion) in the rod. If for safety. for the greatest stress to is These values are the stops. ^ = 90.000 lbs. as when the spring deflected down against value The modulus of elasticity.000 p. The other formulae are useful in checking springs already constructed. sq. a helix. in which h the thickness of plate in inches.000 p= 75. new spring. there the rod. Fig.000 in the absence of more In designing a definite data. may vary considerably. the typical helical spring produced. 28. or the reverse.000 + is ^ *7 COO h — . E. Helical Springs.000 p= 80. but as the coils do not is in planes perpendicular to of direct stress along the line of the forces." 128 MACHINE DESIGN elastic limit Experience shows that thin plates have a higher than thick plates of similar grade of material. the value of is h is to be found from equation (III). with the two ends turned is in radially to the axis. but its may be assumed at about 30.000. + P and — P. but with ordinary proportions .
when the ex along the axis of the helix.. to center of wire. is appears that the in energy expended against the resilience (the length of wire affected the same both cases remaining constant). as the force (P) and the arm are equal. subjected to an axial load: P = the r (/ force acting along the axis. the torsion alone ternal forces lie 129 need be considered. (r) are the same in both conditions. the angle of twist being". 29. to Suppose a to under an axial load be cut across of this section the wire at any section.z / = the length of wire in the helix helical spring =2 « r n (approximately). /> = thc maximum intensity of stress in wire (torsion). s '> . it is straightened out. = the number of coils in the spring. demands force shall equal the stress moment (Pr) of the external couple. = the radius of the coils. Since the stress and strain are it the in this helix and the straight rod. £ = the transverse modulus of elasticity." page . or moment of resistance [ m6 — : 3 p d for circular section ) ' If this free portion of the helix lines in Fig. will appear that the moment Pr equals the moment same of resistance. Fig. Or. = the diameter of wire. as indicated its by the broken till direction is perpendicular still to the radial end. — p d 16 3 . 7 p = the polar moment of inertia of wire. then Pr = [See Church's 9 ^ 236]. and the portion on one side that the be considered as a free body. The following notation will be used in treating of helical springs of circular wire. = the "deflection" (elongation or shortening) of spring. 29. equilibrium Neglecting the direct stress.(in measure). "Mechanics. the distances stress through which (p) this force acts to If produce a given torsional / a straight rod of length is subjected to a torsional moment Pr. SPRINGS of springs.
is load gradually applied. for designing a spring in the load and deflection are given. or too low for economy. The most convenient form for this purpose is usually. a = r d r. or the energy expended in the is j4 P8 is > As pointed out above.X — X E —E — = dd 4 s  s =L 64 r 3 n (1) Equation to (1) may be used for finding the load corresponding an assigned deflection in a given spring. becomes necessary to check for strength. this energy y Pr 2 «. n These equations = ^T? good only within the is (3) for rigidity hold elastic limit of the material. that the stress either too The formula under torsion is for the strength of a solid circularsection rod . and it will is often be found. The equation can to be put in the following form for finding the deflection due given load S a = ^^ d4 E {) (2 ) which B Or the equation may be employed r. = b % Pb Ttd 4 .: 130 MACHINE DESIGN the rod is The energy expended on tiplied is the mean force applied mulIf the by the distance through which this force acts.Pr = oI R —V^ = . by assuming any two of the three quantities. latter d and n. the energy expended two cases the same. as strain within this limit.'. In the case of the corresponding helical spring. E s simply a ratio between stress and It therefore any of the above indicated computations high for safety. or y 2 Pro. after thus checking. the mean force (J4P) acts through a distance equal to the "deflection" of the spring (<*).
SPRINGS
131
~
Pr =
—pd r 16
pd?
3
.'.
' '
P
pd
3
'
16 r
(4)
r
= ^p'>
*=^¥
i6Pr
(4) is for safe strength,
It is to
be remembered that as equation
the
can
load (P) should be the maximum load to which the spring be subjected; but equation (3) may be used with any load
and the corresponding deflection. Example: The load on a helical spring
corresponding deflection
elasticity of material
is
is
1600
lbs.,
and the
to
be
4".
Transverse modulus of
= 11,000,000, and the maximum intensity of safe torsional stress = 60,000 lbs., wire of circular section. To design the spring, assume d=y%", and r = \y 2 "\ from eq. (3),
n
=
4
X
625
X
64
11,000,000
X
8
4,096
X
X
1,600
last
1.5
X
512
=
19.4.
27
(4),
Checking
for the stress
by the
equation in group
p
=
16
X
1,600
X
be
X
*X
found
it
=
is
So> 2 °o lbs.
125
safe,
This
stress is
to
but
considerably below the
limit assigned,
and
may be
desirable to
work up
to a
somewhat
can
higher stress.
Another computation can be made (with a smaller
d or larger
r),
and by a
series of trials, the desired spring
be found.
The
following order of procedure avoids this element
of uncertainty.
The
load being given, assume a diameter of wire
and value
of safe stress, then solve in eq. (4) for the radius of coil.
some convenient dimensions (not exceeding that computed if the assumed stress is considered the maximum safe value). Next substitute these values of d and r (with those Thus, given for P, d and E) in eq. (3) to find the number of coils.
this radius
Make
with the data of the preceding example, assuming
r
d=W
•
f
;
=
is
p d?
16
7
n P
X 60,000 X = ~2 H 16 X 1,600 X

12s
"
512
=
179
If the yi"
rod
wound on an arbor
3" diameter, the radius to the
centre of coils will be about 1.81";
and the corresponding
stress
:
:
:
I32
MACHINE DESIGN
lbs.
would be 60,500
this value is
per square inch.
it
This
is
so slightly in ex
cess of the assigned value that
may be
permitted, especially as
steel.
a moderate one for spring
Substituting in
eq. (3);
1%
=
$ d*
64
It
E = Pr
s
4
X
11,000,000
3
64
fix
x
1,600
x
593
X 62^ = X 4,096
T T T
may
be desirable to
upon the radius
of coil, rather than the
diameter of wire, in the
first
computation, in designing a spring.
3
From
eq. (4)
^
In other cases,
it
l6Pr
Ttp
1 •^
\P~r =I 72\—
v
,
• .
n
(5)
p
may
be desirable to assume the ratio of the
(4)
radius of coil to the diameter of wire, then from eq.
In either of the preceding conditions, a standard
should be chosen.
In checking a given spring,
it
size of wire
may
If
be required to determine
If the
either the safe load, or the safe deflection.
case, eq. (4)
former
is
the
may
is
be used
directly.
it is
required to find the
eq. (4) in eq. (2)
safe deflection, substitute the value of
P
from
and the
result
12.57
n^P
(7)
Ed
The weight
material
is
of a spring is a matter of
some importance,
as the
expensive.
The
following discussion shows that the
weight varies directly as the product of the load and the deflection, inversely as the
square of the intensity of stress in the wire,
and
directly as the transverse
modulus
of elasticity.
Hence
for
a given load and deflection, economy calls for a high working
stress
and a low modulus
of elasticity.
From
eq. (4)
P =
p
— p— 10
7T
d3
r
;
also for a
member under
torsion,
=
— X ^j1
[Church's " Mechanics," p. 235].
:
SPRINGS
133
—
(/
X
—X
r
d
2
d$E —— —
E
s
2  r
=
rs
ii
4 
r
11
4 ~
rnp dE.
4
(8)
5^ .\P*—^
s
(9)
But the volume
of the spring
is
v%*#l = yi*&Tn
(10)
.;P*=fK,.:v = 1fPt
The
weight
is
fr
v
2
E,
......
<„)
directly proportional to the
volume; hence for
given values of
of the load
E
s
and
p, the weight varies simply as the product
and the
if
deflection.
All possible helical springs (of
same weight for a given load same material and worked to the same stress. It can be shown that a helical spring of square wire must have 50 per cent greater volume than one of round wire, the stress and modulus of elasticity being the same in both. The
similar section of wire) have the
and
deflection,
of the
round section
is
generally admitted to be best for helical springs
under ordinary conditions.
A
limit
in
small wire of any given steel usually has a higher elastic
than a larger one, while there
is
not a corresponding change
the
modulus
of
elasticity
with change in diameter.
is
This
suggests the use of as light a wire as
consistent
with other
requirements.
An
Adams,
extensive set of tests of springs, conducted
by Mr. E. T.
to stress
"
in the Sibley College Laboratories, indicates that the steel
is
such as
used in governor springs
lbs.
may
be subjected
varying from about 60,000
per square inch with
y
A
wire to
80,000
lbs.
per square inch (or more) in wire yi" diameter.
The
following expression
springs
may
be used to find the safe stress in such
p
=
40,000
+
——
1 5
,000
.
(12)
Mr.
J.
Springs before the
W. Cloud presented Am. Society
a most valuable paper on Helical
of
Mechanical Engineers (Trans.,
134
MACHINE DESIGN
Vol. V, page 173), in which he shows that for rods used in rail
way
springs
lbs.
(W
to i T g" diam.), the stress
5
may
be as high as
80,000
per square inch, and that the transverse modulus of
elasticity is
about 12,600,000.
being subjected to the same
Two or more helical springs are often used in a concentric nest
(the smaller inside the larger);
all
deflection.
This
is
common
practice in railway trucks,
where
they
the springs are under compression when loaded.
If these springs
if
have the same "free" height (when not loaded), and
are of equal height
when
closed
down
"solid,"
Mr. Cloud shows
that the length of wire should be the
set for
is
same
in each spring of the
equal intensity of
n,
stress.
The
is
1
"solid" height of a spring
H=d
and the length of wire
=2
~ r n\
hence the num
bers of coils of the separate springs of the set are inversely as the
diameters of the wire and inversely as the radii of the
the ratio of
r to
coils;
or
same in each spring of the nest. This conclusion may be somewhat modified when it is remembered
d
is
the
that the wire of smaller diameter
may
usually be subjected to
somewhat higher working
helices;
stress
than the larger wire of the outer
and
also that the wire of these compression springs is
commonly
the seats.
flattened at the
end
to secure a better bearing against
See Fig. 30.
SUMMARY OF HELICAL SPRING FORMULA.
Pr = 7pd
16
r
3
.
.
(I)
S
E.d
3 64 r n
•
(VII)
=
r:pd3
16
P
•
(II)
(VIII)
d

>>*'<!
j
(HI)
d
_ ~
64
Pt%
d4
d
E
•
(ix)
'
s
'*>/&)
i6Pr
d*E
2
(IV)
U
a
3 64 Pr
(X)
' '
P
(V)
2A
.
(XI)
p
*P#
(VI)
.
SPRINGS
I
35
Two common
in Fig.
30
(a).
fit
One end
methods of attaching "pull" springs are shown of the spring shows a plug with a screw
This plug
is
thread to
slightly,
the wire of the spring.
usually tapered
and the
it
coils of the spring are
somewhat enlarged by
screwing
in.
The
is
other end of the spring shows the wire bent
lies
inward
to a
hook which
(I)
along the axis of the helix.
The
(VIII)
former method
usually preferable for heavy springs.
to
Formulae
to (VII), inclusive, relate
strength;
to (X), inclusive, relate to rigidity, or elasticity.
In the absence of more exact information as to the properties
of the material of
which a
steel helical spring is
made, the
fol
lowing values
may
be taken:
E —
a
12,000,000,
p
44. Spiral
=
40,000
Springs
+
——
15
,000

or
Helical
in
Torsion.
The
following
formulae for either true spiral or helical springs subjected to torsion are derived from
leaux.
<p 9
"The
Constructor," by Professor Reu
=
PRl EI
PR
'
P
In which
P = load applied to rotate axle, R = lever arm of this load, = angle through which axle turns,
<P
= length of effective coils, £ = modulus of elasticity (direct), / = moment of inertia of the section.
/
CHAPTER
45. General Considerations.
VI
RIVETED FASTENINGS
The
is
simplest form of fasten(Fig. 31), a straight
ing
b,
is
the rivet.
It consists of
c,
a head a
shank
and a second head
which
formed while hot and known
as a point.
gether,
is
When
it
is
desired to rivet two pieces
drilled as
MN
to
mating holes are punched or
a
larger in diameter.
shown, the rivet
is
heated white hot and pushed into the hole which
little
purposely
made
The head
is
held up firmly
by a heavy bar or sledge and the point may be formed with a hand hammer, or with the aid of a forming tool or set. In riveting on a large
against the plate
scale
this
operation
is
per
formed by hydraulic or pneu
^p
matic machines.
The
relative
merits of the two methods will
be more apparent after further
discussion.
The
rivet is a per
be
Fig. 31.
manent fastening and cannot removed without the destruction of either head or point.
It is largely
used in structures
such as bridges, the framing of buildings, ship work, boilers,
tanks, etc.
Fig. 32
shows various forms
at
of rivet
heads and points.
for small rivets
The
form shown
T
5
g
B
is
most commonly used
in.
diameter, which are driven without heating, for
up to such work
The form at C is much as light tank and smokestack work. used in ship work, or wherever smooth exterior surfaces are
desired.
In machine work, where great accuracy
is
required,
the holes are reamed, and
pletely to
fill
the rivet carefully fitted so as
com
the hole;
both heads in such cases are usually
countersunk and formed cold.
n6
RIVETED FASTENINGS
137
When
sile
a rivet
is
"driven" hot
it
shrinks in cooling, drawing
cold
it
the riveted parts firmly together.
stress
When
and
is
under a
tenit
due
little
to this shrinking,
for the
same reason
it
is
always a
smaller than the hole which
hot.
originally
com
pletely filled
effect
when
rivet,
The
tensile stress
it
due
to this cooling
cannot be accurately determined as
depends on the temit is
perature of the
and the manner
in
which
driven.
are
Rivets
are, for this reason, unreliable as tension
members and
(Fig. 31)
is
seldom
so used.
In most cases the parts
M and N M
have the load
P
applied as shown, and the tendency
relative sliding
to shear off the rivet
and produce
load, P',
of
to
between
and N.
The normal
due
to the tensile stress in the rivet, holding the surfaces
M and N firmly in contact, sets up a frictional resistance equal
!'.P'
which opposes the action of P.
From experiments made
,
Conical
Sutton auiion
Conoidal
Countersunk ante Co
VPan
j
5£fe
J
v
Fig.
3:
by Stoney *
at
it
appears that
lbs.
this frictional resistance
may
be taken
about 10,000
per square inch of rivet area.
Experiments
it
by Bach, and
evident that
stress
if
others,
show a much higher
resistance, but
is
the
normal pressure
of the rivet is such that a
is
equal to or greater than the elastic limit
induced, the
permancy of the resistance cannot be relied upon. In some French and German practice the design of the joint is based entirely upon the frictional resistance, but in England and America it is neglected, and the design based upon the tensile and shearing strength of the plates and rivets. Riveted joints are of many forms de46. Forms of Joints. pending on the character of the work to which they are applied.
In structural work, such as bridges, they are used simply to
*
"Strength of Riveted Joints," page
75.
I38
resist direct loads;
MACHINE DESIGN
but in boiler construction, and similar work,
they must not only resist direct loading but must also be tight
against fluid pressure.
This
last
requirement materially affects
of joints for
the proportions of the joint,
and makes the design
withstanding fluid pressure most important.
divided into two general forms.
(a)
Riveted joints are
Lap
Joints,
where the sheets
to
be joined are lapped on
each other and riveted as shown in Fig. 33 (a). (b) Butt Joints, where the edges of the sheets to be joined abut against each other, and have auxiliary butt straps or cover
plates riveted to the edge of each, as
A
lap joint
and these
Fig. 33
shown in Fig. 33 (e) and (f). more rows or seams of rivets, rows way be arranged in the form of " chain" rivet
may have one
,
or
ing, Fig. 33
( c ).
(b)
or in the form of zigzag or staggered riveting,
A
butt joint
side of the
may have one or more seams of rivets on each joint, and these may also be arranged in either chain
shown
that
in Fig. 33 (f)
or staggered form, as
and
(g).
are very numerand the student is referred to any treatise on boiler work for fuller information on this point. The distance between rivets along the seam is called the pitch
The combinations
may
be thus
made up
ous,
or spacing,
tion
and
will
be denoted by
joint
it
s,
Fig. 33 (b)
.
An
examina
of
any riveted
pattern as
will
show
that the arrangement of
rivets, or
may be
called, continually repeats itself as
joint, the repetition
the
"seam" extends
along the
occurring with
is
the greatest pitch, where the pitch of the various seams
as in Fig. 33 (h).
unequal
not
A
unit strip
is
equal in width to the pitch, the
the pitch of
is
maximum
the same.
pitch being taken
when
all
seams
is
The
transverse pitch
the distance between the
centre lines of adjacent seams Fig. 33 (b) and will be denoted The diagonal pitch is the distance between the centre of by s
. t
a rivet and that of the one nearest to
it
diagonally, in the next
row, and will be denoted by s d Fig. 33 (d). The margin is the distance from the edge of the plate to the center line of the nearest
row
of rivets, as e Fig. 33 (d).
It is
sometimes defined as the
distance from
the edge of the plate to the edge of the rivet hole.
RIVETED FASTENINGS
47. Stresses in
139
that exist in
Riveted Joints.
The
stresses
the various
members
of riveted joints are complex,
and do not
admit of refined calculation.
to the
Not only
are the plates subjected
apparent direct stresses of tension and compression, and
J
L
^
00
,
!(D
^V^V
U7
WR^
^etecfr
AA
(e )
.
s
A
(0
\ip^jrxptp7
*m^
Fig. 33.
(a, b, c, d, e,
f,
g, h.)
1885. But and hence some bending of the rivet and concentrated crushing on the plate must is result. 1882. 33 (a). . of Mechanical * Proceedings Institute Engineers. 1896. Watertown Arsenal Reports. even here the rivets do not completely the holes when cold. fill 34 cannot occur. tends to take the form shown in Fig.140 the rivets to shear MACHINE DESIGN and compression. entirely destroyed The when the conditions illustrated in Fig. 1895. as in the latter the plates are ini tially in line and the condition shown in Fig. the work manship not so easily controlled and may be very defective and yet not tests show on the exterior. while the quality of the material forming the joint may be well known or determined. 1885. 34. as that bending actions which are mathematically. and rivet. The above defects are more marked in the lap joint than in the double strapped butt joint. putting a bending action on the plate greater tensile stress on the rivet head. 34 exist. 1887. Again. The force applied tends to draw the plates into the same plane. 1888. 1881. 1891. 1886. a and a concentrated crushing frictional resistance is load on the corners of the sheets. and while there have been many * made to find the ultimate strength of riveted joints. when subjected to a load. but often there are also difficult to analyze and provide for lap joint. Thus a simple shown in Fig.
: RIVETED FASTENINGS such joint tests 141 show only the stress at which a certain element of the failed. 35 (d). and do not throw any light on the distribution and progress of the stresses in the various individual ing the test. (e). in designing. Where due on to the joint is complex in form. ficiency of the joint. 35 in Fig. Fig. shows one that has been similarly tested It is in double shear. while giving the only data available These experiments. 35 (b). members dur Such tests have usually been performed on joints made of straight plates while in practice these are often curved. since the very fact of making holes in it reduces ratio the crosssectional area in the line of the rivet holes. on Plate are reproduced from these failure reports. 35 (c). should be used with judgment For these reasons the theoretical formulae deduced for the design of riveted joints. therefore. relative to the ultimate strength. shows a rivet that (d) been tested to destruction in single shear. 35 (a). has been found that riveted joints may fail in following ways (a) Shearing of the rivet as in Fig. while Fig. obvious that no riveted joint can be as strong as the un perforated plate. and will be used in work. in Fig. The of the strength of the weakest element of the joint. Figures may be The Watertown Arsenal I. reports include cuts of ruptured joints which are very instructive this point. as a rule. ultimate failure one or more of the above causes. (a) and (b). desirable to reduce the strength of the plates as little as possible by perforation. and show very clearly all the (c) ways in which has may occur in the plate. (d) (e) (f) plate. in staggered riveting. take cognizance only of the apparent simple stresses and provide for the unknown by one of the means It of a factor of safety. or rivet as in Fig. to the strength of the imperforated plate. rivet holes Rupturing of the plate diagonally between by tension. and . (b) (c) Rupturing of the plate by tension as Tearing of the margin as Crushing the Shearing of the margin as in Fig. is called the relative strength or effirst The this expression It is is more suggestive if.
142 Plate I. .Ca.
It 49. where d It is the diameter of the rivet. however. at destroy the rivets. for equal strength. prevent tearing or shearing. 35 (c) has been found that. and 2. of the rivet. Transverse has been deter mined. as can be made depend on the diameter and Diagonal Pitch. an excess of strength in other directions.RIVETED FASTENINGS therefore. that the distance plate be from the center of the rivet to the edge of the made twice the diameter of the rivet. the correct relation 1 43 between the size of rivet is and cross is section of perforated plate. at the expense of The width of margin is independ ent of the proportions of the other elements. joint should by tearing of the sheet about the same load as will fail and the relative strength of a welldesigned be the ratio of the cross section of the perforated to that of the unperforated plate. has also been demonstrated mathematically (see Cathcart's " Machine Design. not a defect but good design. It is important. in order to insure excess strength enough to either shear the rivets or rupture the plate by tension. with the usual proportions. as in Fig. 48. if the margin be the rivet. If this equality does the relative strength of the joint can be made greater by strengthening the weaker of these elements the stronger. A Committee Master Steam Boiler Makers Association recently recommended." page 148) that in staggered riveting . as it insures that the full strength of the perforated plate will be in service before rupture can occur. Marginal Strength. in a similar way as above. that the margin be not as this excessive in boiler work makes it more difficult to steamtight joint by calking the edge of the plate. as a result of experiments. as marginal distance. A welldesigned joint should hence along the line of the rivet holes. make a The con sideration of the marginal strength can hence be omitted as far as its it influence on the relative strength of the joint to is concerned. it made equal to one and one half the diameter of of the will be safe against both shearing and tearing from rivet pressure. the shearing and crushing resist ance of the rivets being equal to the former. not exist. established.5 d is better. that in chain riveting the transverse pitch should not be less than twice the diameter of the rivet or 2d. and hence can be made and sufficient to It (e).
when in double shear. d as the minimum diagonal pitch in staggered 1. in single shear. Let. p\ = crushing strength of plates or rivets inch. b and e to be considered. can be omitted from the theoretical discussion of the strength of riveted joints. page 141. p = crushing strength of plates or rivets in pounds t c per sq. inch. inch. a). Since the margin 50.— 144 MACHINE DESIGN the transverse pitch should not be less than 0. . t (1) . in pounds per sq. rivet in inches. d and /. p s = shearing when strength of rivets in pounds per inch. Theoretical Strength of Riveted Joints. p = tensile strength of plates in pounds per sq. leaving a. and a greater distance 2 is recommended for safety. in order to avoid rupture along the diagonal pitch. Unwin d. d = diameter of 5 = pitch of rivets in inches. and transverse pitch can be assigned from the diameter of the rivet. three of the ways in which a joint may fail. known that the unit shearing is less resistance of a rivet is greater in single shear than in double shear. namely c.7 which would make the transverse pitch that and recommends strength. 33 is The tensile strength of the unperforated strip P = stp . somewhat greater distances be used for added An examination of the practice of several boilermakappears from the above that the trans ing and insurance concerns shows that these values check fairly well with practice. if rivets are in single shear. (page 123) gives riveting. p\= shearing It is strength of rivets in pounds per sq. Consider first a simple lap joint (see Fig. made to depend on the diameter of the though it is not a direct function of the rivet diameter. where 'rivets are in double shear. It verse pitch can also be rivet. the problem being so to proportion the rivet and the pitch along the seam as to give equal strength against failure in any of these three ways. while the unit crushing resistance in single shear than in double shear.4 times the pitch along the seam. inch.
s = (s + a s _. d2 4 —p pt B . a similar joint. . (4) For uniform strength against rupture and hence for greatest T = S Equating (3) =C c and (4) p 4 . . s = dtp 1.27 I and * s ———[Pt + d It is known from experiments on little it is indentation that the resistance to indentation its depends very area. but one rivet per unit strip and 5 = The it T^ (3) resistance to crushing of the rivet or the plate against is. .2*1 \ d = P. . on the form of the indenting body but mainly on projected Hence 10 customary to take the resistance of rivets to crushing as propor tional to their projected area. which bears C = dtp* relative strength.RIVETED FASTENINGS 145 along the seam of The tensile strength of the perforated strip rivets is T= its (sd)tp is t (2) In the simple lap joint there shearing strength is. The relative strength = T — P = — d) x t = —d s stp In DoubleRiveted Lap Joints. 1 Equating (2) and (3) (s—d)tp = t . may be developed for any other form of P — way equations Thus for double riveted lap joints d = = 1.
. that where only one cover plate of the is its thickness should not be less than that and in practice. is Where as in the outer cover plate narrower than the inner plate. and. single cover plates are made a little thicker than the main plate to insure an excess of strength. than half boilers where a lap joint has Double butt straps should not be made less the thickness of the main plate. ' d=.5 7 d p' 2 71 s + d —d s and the relative strength DoubleRiveted Butt Joints with either chain or staggered riveting and double cover plate. for the same seams of steam reason as above. it is not unusual to increase their thickness to about y$ t where the cover plates are the same width. joint of this kind is really equivalent to two lap joints. It is evident Thickness of Cover Plates. Here.146 MACHINE DESIGN relative strength The = s — d and will be greater than in the case of the singleriveted lap to d. *t* Note that the ative values of s relative strength = s —d . A butt joint with a single cover plate is shown at Fig.64 tf a s = = i. and compare the rel and d used in this case with those in single riveting. A main plate. joint since s is greater in proportion SingleRiveted Butt Joints with double cover plates. 33 (e). Here. They are used where a smooth surface the longitudinal is desired or in such places as sufficient strength.
.. maximum pitch being taken kt is of width equal to the pitch. the outer cover plate is 147 often of the X. strip in double shear per unit The strip is general expression for the net tensile strength of the unit T=(sd)l Pl * The in the unit strip is (1) general expression for resistance to shearing of the rivets m* n* S— —A+ — — f .X 4 nk. General Riveted Joints. cannot shear one row of rivets without shearyield in which case the joint would by shearing of the rivets and not by tearing. . =p in 5 n Pa ] k2 = unit strip single shear n = number of rivets per = and m number of rivets per unit p\?p\\ strip of joint. the outer row or that farthest from the edge of the sheet has the greatest pitch (see Fig. which gives the proper diameter of plate. for such width of unit strip if rows do not have the same Let pitch. eqs. 33 (h). it if it all by tearing. The funda mental equations for riveted joints form. (2) (4) and s (3) and solving 2 for d (5) d = . and ing all. to same thickness /. will row of for cannot tear along an inner row without shearing the outer row of rivets.RIVETED FASTENINGS Fig. yield along this outer It is evident that rivets. as before.p + np a — m k pj 7+ 2 m pi t . : a (2) 4 4 The general expression for resistance to crushing of the rivets in the unit strip is C = The nd t k l pa + m d t k 2 p' s .. the all may be put in a more general The unit strip. as the main plate and the inner one from Equations for T\ 51. . 33 h). . as the sheet yield at in some forms it of butt joints. in rivets for a given thickness of when the number of rivets in single shear and the number double shear and the corresponding shearing and crushing resistances are * known. do not all Where the rows of rivets have the same pitch... (3) tensile resistance of the solid strip is P = stp Equating 5 and C. d? 2 d? ...
)+dtp =P.^ 148 Equating MACHINE DESIGN T and S. t . hence any one form. stp = If the joint is *d — r (np 4 s +2mp. n kx p s + m 1 k2 p a If all rivets are in single shear. eqs. (1) and (2). . or dividing (2) of these three quantities divided by T= P gives the relative strength (£) of the ideal joint. eqs. (2) /A and d s ) + '' • ' (6) (3) d 4 2 {n p s + • 2 m p') = = . S = C. (1) 2 and (2) and solving . and solving for s '^ Equating S and C. eqs. p (90 t n If all the rivets are in k. 2rn p 2 s ) E= i= P 4 ~d 4 2 (w p t ( + mp ') s \ d t p t Substituting the value of d as given by eq. (7) 2 and dividing d2 numerator and denominator by 4 (n p + mp r ). s — 2 p:) p w) KU ) s Equating T t and S. p s double shear E= 1 . (8) designed for maximum relative strength. for any given by (8) \ np s + . t (n k t p + m f k 2 p' ) g #* Tz(P(np — — 4(nk 1 + 2 m p + mk . L —2 (9") mk p\ .
It useful in finding the limiting relative strength of joint for any form and materials. The forms due which eqs. the theoretical spacing must sometimes be modified in order that it may not be so great as to prevent the making of a steamtight joint. but can never give higher relative strength. It is to be especially noted that the proportions of riveted joints as given on equal strength of also that rivet by the foregoing equations are based and plate. to adopt regular diameters and pitches for a It given thickness of plate and form of joint. Barnard. apparent any variation in the strength of the material used would affect the proportions as given table of rivet by these equations. expressions for T. These general equations were for the originally due to Professor William N. the corresponding theoretical diameter of the rivet sometimes becomes too large to be easily driven. Moore. an advantage has also been however. especially in the case of simple lap joints. who also suggested the above expressions maximum in relative strength in the general case.RIVETED FASTENINGS Equation (9) 1 49 is applies to any form of riveted joint. F. In the case of boilers. and that a diameters and pitches would have to be very exIt is tensive to cover the entire range of practice. in practice. found that as the thickness of the plate increases. and that any variation It is therefrom will destroy this theoretical equality. strength and Wherever such variations are made the general S and C can always be used to check the in show what direction the joint may be strength . or wherever fluid pressure must be withstood. (9') and (9)" are now given are to Professor H. the actual proportions adopted may give a lower relative strength. (9). Practical Considerations Affecting Proportions of Riveted Joints. in practice for boiler The following are rough average values of the relative strength of joints as made work: 55 Single riveted lap joints Double riveted lap joints 70 65 > Single riveted butt joints Double riveted butt joints 75 Triple riveted butt joints 80 85 Quadruple riveted butt joints 52.
the rivets in such cases If the need only be to checked for shearing strength. If. the rivets have a shearing strength of 49.000 lbs. the spacing of the rivets for thick plates ness. are to be of steel and are . the rivet chosen the diameter. tensile strength. while the shearing resistance varies with the square of the diameter. the shearing resistance alone need be regarded .000 pounds per inch of length of steel of The be of 60. per square inch in double shear. plate. as in boiler work. The between diameter by average practice may be expressed by the equation. while if of the rivet be greater than the theoretical diameter. the bearing pressure only need be considered. Example. as shown in Fig. It may be noted that the bearing resistance of a rivet varies with therefore. diameter of the rivet be and the pitch so chosen as make the tensile strength of the perforated plate equal to the shearing strength of the rivet.150 MACHINE DESIGN making it ened. for ordinary thickness of be made to conform closely but when to the theoretical proportions for equal strength. 33 (h). with the fundamental object of strong enough in all other directions to insure full service out of the plate itself. determined by equation (10). As before pointed out. If the joint to be made does not have to withstand the fluid or gaseous pressure. design can. for a boiler shell in is which the force tending shell. as diameter of the already noted.2 (10) For plates above ¥&" thick this equation will give rivets smaller all in diameter than required for equal strength in directions. the maximum relative strength of joint possible with the rivet chosen will be obtained.000 to lbs. relation of rivet and thickness \/t of plate as fixed d = 1. must be less than the theoretical spacing to insure tightand in all cases as the plates increase in thickness. fluid or gaseous pressure must be withstood. and 42.000 lbs. is for practical reasons reduced in diameter from that required for equal strength. the joint apart plate is to required to design a riveted joint. the rivet. The factor of safety is to be 5. be smaller in the diameter diameter than would be given by the theoretical equations. and it will be found that the joints will It is be steamtight. to pull 6. per square inch in single shear.
The Making of Riveted Joints. Th s = = . If the pitch found as above should be considered too great. a smaller pitch could be less.000 the lbs.000 12.000 = V inch.+ 4 L_ 2mpn J x tp Here. The relative strength of the joint will not be less than and hence the thickness of the plate must be X . . hence the design is safe. 52. . n 1 and m = L 4.000 16 s = 8 inches nearly.80. used. If the joint were as strong as plate unperforated plate thickness of the would be 6.  lMt\ *>f**> + 4 54 (» > <4 x 42 . For struc joints in compression the student is referred to treatises on tural work. Hence. be observed in making .ooo)  I x 60.'.1. Equation (6) of this chapter gives the relafor tion between pitch and diameter of rivet equal strength against shearing of the rivets and tearing d of the plate.2 %/ _5_ ^ = — nearly. it either on account of very high steam pressure or because is desired to make the structure stiffer.000 = 12. *jrr np.RIVETED FASTENINGS 151 The the joint the allowable stress per inch of length of the shell outside is — 2 60. but the relative strength would be Where no will fluid pressure is to be withstood the above methods always give satisfactory results for joints in tension.80 5" The diameter The rivet size oj the ill of the rivets will = 1. 8 8 punched hole and the diameter of the driven will be f " . It is evident that the firstclass following precautions must riveted joints. The relative strength of the joint is o 15 —^^ = 88%.
some is to drill them. The mating must be "fair". serithis ously impairing the strength of the plate. and with thin plates. when the plates are brought together the mating holes are not The . The rivet holes in the plate may be either punched or They are generally made about tV inch larger than the Generally speaking it is cheaper to punch the holes than In structural however. (b) drilled. and hence in the cheaper kinds of work. or annealing the plate. to by a bolt placed in a hole near that in which be driven. When punch forced through a plate the amount of metal which of a removes in the form filled is is "plug" This hole. and that by reaming out the hole about xV inch all around.152 (a) MACHINE DESIGN The plates driven. action either is It is found that confined to a thin ring next to the hole. rivet. The process of punching is apt to make inaccurate work and. setting if up a stress in the material. and if the plates have been properly rolled and fit For heavy work. is will accomplish the result the holes have been accu rately spaced. fair. There the it are. before is is the rivet the rivet driven. but that during the process there of metal a flow from under the punch to the walls of the hole. they must be in perfect alignment to insure full cross section of the rivet at the junction of the plates. will not be weakened by poor workmanship. the riveter is sometimes provided with a powertill driven closing device which holds the plates up the rivet nips from cooling. With comparatively if thin plates this method well. and. work the holes are generally punched. is is not equal to the amount that originally the accounted for by the fact that punching not a pure shearing action. or that of the plate. punching is almost always resorted to. this weakening effect disappears. that is. its rivet should be carefully driven so that strength. the metal is at all hard. (c) (d) The The rivet must completely fill the hole. where for other reasons machine riveting necessary. serious objections to punching. (a) In hand riveting the plates are drawn up together. therefore. to prevent a fin must be in close contact before the rivet is from forming between them and thus holes making a (b) tight joint impossible.
drawing them into line by force the injury thus done to the plate being often very is now largely prohibited. small. as if it is upset heavy rivets. may seriously impair the strength of the head. . the hole an neighbor. for the rivet drives better and the tapering tendency to relieve the head of part of the tensile it is Since it necessary to have the hole a little larger than the rivet. on the other hand. 36. the holes are punched a little small. plates are annealed the work should be properly done for if the plate be overheated structural changes take place that materially weaken too rapid nor the holes The heating should not be temperature above a medium cherry red. The the metal. care must be exercised and in machine riveting that the pressure applied does not create such a flow in the rivet as unduly to strain the plate. by undue conmay be brought on its of great length. together as in Fig. and where they are used as they machine the rivet has a (c) somewhat tapering in come from the punching should be punched so that they will come are necessarily stress. If a rivet does not completely centrated bearing. (d) On the other hand. to the strength of the When . and reamed ing is to the proper size. or shearing stress.RIVETED FASTENINGS old practice of driving a taper drift pin into such holes 1 53 and If. A small countersink. and all firstclass all if Heavy plates are boiler work requires the holes to be drilled in place and burrs carefully removed. made by punching rivet holes form. in which the effect of punchpunched and used without reaming or annealing. also that the pres sure applied is not great enough to crush the plate directly. difficult its entire length in order that rivets may completely fill Large to should therefore be machine driven. Machine riveting. serious. is clear that the rivet when driven must be upset it throughout the hole. materially contributes rivet. is superior to hand work. always better if drilled. Practice allows about 80 tons per square inch of rivet area. as the burr. >2 inch thick Plates more than  should always be either annealed or have the holes reamed after punching. allowed to remain. are Thin plates. the difficulties and put together due to punching are largely overcome. especially fill hand. when well done. This last is important. however.
if and also because the is impact cold. as the method of hand bevelling after riveting.154 the plates being held MACHINE DESIGN up firmer. In either case rivet should be exercised that the point if is formed on the concentrically. the joint will be tight against ordinary pressures.000 per square inch. tight joints is Where This do not result the edge of the plate Fig. This is sometimes recognized in writing spacing of the rivet is If the correct and the is riveting well done. a tensile strength of not than 55. Inst. 36.ooo " " " " 49. is It is known shear. Fig." page 132. specifications. for if the tensile strength is too high and the metal is hard they are liable to crack while being worked. is from hand care riveting. page 150.000 100. as shown an exaggerated manner.000 72.2. is almost sure to result in some well injury to the lower plate." often done. as shown in in by means of a sharpnosed tool T which tucks the sharp bevelled edge of the plate under neath.000 42. Strength of Materials for Riveted Joints. at B. f Proceedings sign. is preferable. per square inch. as often done in practice. that the strength of rivets different in single and double The following may be used as average values. especially the rivet worked too liable to result in the breaking off of the head. and the method shown The plates should be bevelled before riveting. the dies are not properly set eccentric rivetre ing will occur.000 double Steel Plates for Boiler Work less are generally specified to have lbs. p s* 40. 52. Engineers. For iron plates an average value may be taken as 45.000 80. There is liability of injuring the lower plate in using a sharpnosed tool as T. " calked. and Unwin's "Machine De . 36. Mech.000 pounds per * Master Boiler Makers' Association Rules. 1885.000 lbs.000 pet 60. and not more than 65. For structural steel construction the student is referred to handbooks on structural work. In machine riveting the pressure should be till tained on the rivet cold enough to hold the plate firmly.000 Iron rivets single shear Iron Steel Steel " double single " 39.
In boiler work the factor of safety is taken at about 5 which. Barr's " Boilers and Furnaces. this factor should be increased. Hutton. See also Cathcart's "Machine Design. Practical Rules." W." Proceedings of Inst. S. increase being sometimes if as high as 20%." Wm. It is questionable. 1885. this should be taken account of in designing.RIVETED FASTENINGS square inch. Machine Design.3. be subjected to hydraulic to where heavy shocks may have It it be withstood. has already been noted that practi cal considerations make necessary to modify the theoretical equations for uniform strength. may more than offset 52. however. Rules of the Master Steam Boiler Makers' Association. of Mech. brings the working stress well If the joint is to below the pressures. Rules of the American Bureau of Shipping." " Steam Boiler Construction. as the operation of punching this peculiar increase. Board of Supervising Inspectors. 52.4. especially where the holes are punched. There are many sets of practical rules for designing riveted joints a number of which will be found in the references given below. Unwin's " . M. S. of course. elastic limit. Engineers. Factor of Safety. Rules of the Hartford Steam Boiler and Inspection Co. Rules of the U. It is 1 55 shown by experiment this that the metal between the rivet holes has a higher apparent tensile strength than that of the unperforated plate.
It cannot be cut with a for die. is standard thread of this form In Fig. Screws for transmitting power. (a) (b) Screw fastenings. Screws. Fig. 37 (b) easily cut and (c). The efficiency of this it of thread is a little less than the square thread but can be of a cut with a die and wear can be compensated for by longitudinally split nut. but this it a desirable quality in fastenings. 37 (e) the Acme shown. therefore. may be divided into two classes. it is difficult to compensate the half wear with this form For these reasons V thread. the efficiency and the strength all of the square thread of the V thread. depends upon the service required. 37 (d). by machine is They are inefficient for transmitting power. Fig. as shown in Fig. or modified forms of V threads. is most used. The underlying principles of screws are the same. Fig. and before dis cussing the various forms and classes in detail the fundamental equations relating to their action will be developed. this means it compensation making very de(f) sirable for such service as lead screws of lathes. 37 (a). however. as used in machines. For transmitting power the square thread. and the back face usually makes an angle This screw has. of 45 axis. Form of Screws. illustrates the buttress thread. with this The pressure face perpendicular to the axis of the screw. etc. since its efficiency is higher than that of any other form. 37 is (e). as reduces the liability of un screwing. and of thread. for screw fastenings. the full V as shown in Fig. 37 which is often used to exert pressure is in one direction only.CHAPTER VII SCREWS AND SCREW FASTENINGS 53. of the thread The form Thus. are most used because they are strong and dies. 156 . is often used for transmitting power when wear form an important factor.
and the work due to this frictional force will vary with the radius at which it acts. = radius of screw at bottom of thread. is to when computing the reIt be added to that value. . 38 (b). = coefficient of friction between load and collar. Friction 157 In and Efficiency of Square Threaded Screws. Fig. can therefore. d = diameter of screw at bottom of thread. can be considered as forming a resisting moment opposing the Full (a) V Sellers or U. n d = nominal or external diameter of screw.SCREWS AND SCREW FASTENINGS 54. let N represent a nut moving on a square thread. 37. for simplicity. be omitted temporarily from the discussion. acting radius of the thread. consider that is can move only in an axial direction. Let = coefficient of friction between thread and nut. This frictional force F x 38 b) may It may not act at the same radius as P. and (Fig. moment of P.5° Square Thread (d) bW Acme Standard Thread (e) Buttress Thread (f) Fig. is Let W or represent the load under which the nut it moved. hence.S. couple. Standard (b) Whitworth Standard (c) ^14. r — nominal or external radius of screw. so that there mean Let is this force P be applied by means of a no lateral pressure against the screw. turning quired turning moment due to P. hence there friction between N and W. ft l l r. at the under the action of a tangential force P.
since the thread a true helix. of thread = d + 2 dj radius of thread = r + rx a = = pitch. angle made by thread with a plane perpendicular to the axis of screw. body N sliding up an inclined s plane of length nd m and vertical height carrying a load ting tion equal to the pitch. If now is the thread be developed as in Fig. 38 (a). and W which is free to move vertically only. 2 normal reaction R. «. and the It is required to determine for any angle the value of incline. is it is seen. or angular advance of thread per turn. all P required to slide the body N 2 (turn the nut) i*R. Omitthe fric F F y temporarily. the driving force P. between the thread and nut. up the resolving The frictional resistance F = Hence. the forces acting are the load W. that the action of the thread and nut identical with that of a .i58 r MACHINE DESIGN = frictional radius of collar. dm r s = mean = mean diam. forces parallel to ac .
or is very small. and if r be the radius at which F moment of F around the axis of the screw = p Wr and when this resistance is considered the total moment of P c x acts. If H dm = s. the directions of and F 2 are reversed.SCREWS AND SCREW FASTENINGS 159 . . F x (6) being lowered.'.~] — L7T d ft sl ( J/ 5) F l . that is if —— = = /«. Since (Fig.. .. the x l c .R Resolving all „ W sin a P cos a — W sin = a — ft R = . equation (4) may be written p= The friction. the if moment there is of the re sistance at the thread will be zero. (2) forces perpendicular to ac R Substituting in (3) — P sin a = W cos « the value of (3) R obtained in (2) p_ rsinoj^co^l w lCOS a—n sin aJ _ .. as in the case of ballbearing thrusts. and in this case the turning that must be applied is In equation 6 r the first term of the righthand side of the of equation the  is the moment is of the resistance at the thread.=p 1 W. if it and no collar friction.. moment . P cos . 38 a) cos « = —  and sin « = — . . around the axis is Prm = If the load is Wrm P + ** dm~] + H Wr. while second term the moment tan « the collar friction.. the friction of the thread alone just sustaining the load. (j) a. and P l will be equal to . w[^lA. this will give a condition of equilibrium.
since s = * dm e tan = W s or p rm *dm P and inserting the value of P from equation work done energy expended t—r = W— s 2 n (5) = dm W d fs m tan a TT + pWnd. equate the righthand side of the equation 5 (6') to zero and solve for  d . in which case there no moment due to collar fric tan P as before. and if increased till its numerical value is equal to the moment of collar friction. whence tan m a. efficiency = «." the nut exerting a turnin the ing moment downward direction. or very small.=/*. MACHINE DESIGN . the entire righthand side of the equation becomes negative. and ft = p1 tan a = I 2 p « 2 ' ' ' * • • (8) and n a if l p be taken as. If the pitch s is made greater than p x dm the moment of the resistance at the thread becomes negative. To find the limiting value of « where the nut will not over haul.2 whence — = tt° ii To find the efficiency of the screw. been raised a distance equal then e and rc = rm . that /*. . = „ . the screw will " overhaul. 65) tan « =  . w (o) If the collar friction is zero.1 (see Art. . . as in the case of ballbearing thrust collars . If the pitch is still further increased. = o. rc Hf. or. . consider that the load has to the pitch. and the moment P rm must be applied in the direction of raising x the load.l6o zero. = o. the entire righthand side of the equation will be equal to zero and the load will just be sustained by the friction of the thread and collar combined. + — (1 arc dm n tan « — p tan + 2 fi «) tan a nearly J . = rm is — • (7) If rc tion. .
(i — ft tan it *) (10) tan a 4If. found that the maximum efficiency when collar friction is considered will occur when fi tan « If fect. «= Suppose the load repose. when maximum (10) for 1? /' efficiency = 1.jl tan a is or a — angle of to now a force P apthe plied.3 18 or = 5 2°49 In a similar way from equation tan maximum efficiency a=Vi + — . as in tan a for worm gearing. relieve If (Fig. ft may be as low as a . If the frictional resistance of raising were zero.23 where lubrication is imper tan a and oil a In the case of bath lubrication. 38) to be just is sustained by the frictional resistance of lowering. ft = = v 51 . a force 2 P must be applied. namely. just sufficient this frictional resistance.2 as before. and tan « as . and a brief reflection will show that in no case can the efficiency of a selfsustaining hoisting screw exceed 50 per cent.1 = as in transmission screws 1. before the is equal to that of lowering. the slightest addition to P would move the body up to plane. = tan a. in equation (9). that the force which must be applied to start the load is equal to the friction due to lowering plus the friction due to Hence the maximum mechanisms is 50 per cent.05 / . load will be sustained and the reaction R. . and conbody can be started up the plane.SCREWS AND SCREW FASTENINGS e l6l . the first it is differential of equation be taken and equated to zero.1. efficiency for such selfsustaining In designing screws for power transmission it is desirable to know the pitch angle that will give If maximum efficiency for the (9) conditions taken. ft be taken as . that . raising. which is twice the force required to balance the frictional resistance. and the efficiency would then be 50 per cent. A similar reasoning will apply to other hoisting devices which are barely selfsustaining on account of friction. e will equal 50 per cent (nearly). 2 + 4 fi — 2 — . But the fric by the force tional resistance of raising sequently.
. as in the Sellers system.itfisJ 5^ * c ° efficiency of a triangular threaded screw. Threaded Screws. it is The friction increases directly about 15 per cent greater in the 6o° angular thread than in the square thread.i$ p* d R+iiSM'O 's B Ltt d i. friction is neglected when P = d ^n w p+/' L d — sec 7T m s (p » . at more fully discussed in Arts. R =R f = 1.15^. the normal pressure for a square thread is indicated is by R f while <p..R sec <p be substituted fori?. (12) 9 = 3°° Pr = Wrm m The + i. a = 42 = 43 — 34' on the efficiency of the screw is The and is effect of the pitch angle of great importance in designing screws for power transmission. is as for square threaded screws. the normal pressure for a triangular thread R' =R sec in which <p = half the angle between the adjacent faces of a thread. 64 and 65. or the sec 30° With 6o° angular common V thread. or inequations (1) and (s). 38 (c) at the threads is greater. the the axis when raising the load may be a p 7tdm written ' Prm = If + *d Wr\^ •U^^j d — \~s 's = sec <p~~\ <p m {is sec <pi <p \+HWr.05. and its the radial crushing action on the thread of the equal and opposite reaction tends to burst the nut. as the If normal pressure.. following the same reasoning and /*! = /*. moments around (II) If the collar friction is considered. hence the friction In Fig. W s W *dm tan « *d„P 2*rP . a Whence and for /i = = . other things being equal. Friction and Efficiency of Triangular With triangular threaded screws the normal pressure threads is the greater than with square threads. R" represents screw. thread. taking rm = rc .1. l . then by collar a similar method of reasoning as in square threads. 55. 1 62 MACHINE DESIGN for n ..
A throughbolt. to prevent is an advantage in fastenings as it tends them from unscrewing. Fig.1 SCREWS AND SCREW FASTENINGS 163 W * dm tan a ' W*dm — Lr rs or e + ^m — /^sec^J nearly. 2. as G3 CD (d) WQ §g(g) p(e) EHj(f) K(h) Fig. 5. Machine Screws. /* findm sec9~] + /t w W 7T rf = tan a (i tan a — + ft tan « sec 9) ^ sec ^ + (i3 fl ) For the thread on a oneinch bolt in the Sellers system tan a = . of screw fastenings but all 1. 39 (a). The efficiency of the threads on standard bolts is hence seen to be very low and this. has been pointed out. 39.04.1. Screw fastenings are used for hold 56. "bolt" it is com monly called. roughly classified as follows: ThroughBolts. or for adjusting one part relatively to another. There is a great variety may be 3. e = 11%. the best form of screw fastening and should . TapBolts or and as CapScrews. 4. Set Screws. ThroughBolts. ing two machine parts together in permanent position. Studs. has a solid It is head on one end and a nul on the other. and takings = . Screw Fastenings.
Under the term "machine screws" are of the head. and when if once screwed in should not be taken out. 39 (d). Tapbolts and CapScrews. Machine Screws. is A stud. throughbolt cannot be used the stud It is the next best should be a tight fit in the tapped hole. and one end part that can be drilled through is slipped over the stud. Tapbolts. of the stud is screwed firmly into the hole. They are a standard article in the market and hence can be bought very cheaply. and often used. g . or "stud. A The hole is tapped in the part that cannot be drilled clear through. especially the hole is tapped into cast iron. Where a fastening. Capscrews are a little more ornamental and are much used in cheaper grades of machinery. in such cases a studbolt. have a solid head on one end and a thread on the other.164 MACHINE DESIGN always be used when the hole can be drilled completely through the two pieces to be held together. is The next only difference between tapbolts and capscrews in the size head (see and the capscrew for the same size of bolt having a smaller head slightly rounded on top. . The length of the tapped hole should be at least one and one half times the diameter of the stud.(c). Fig. f. The machine screws are shown in Fig. they should be avoided. Fig. a circular bar having a thread cut on each end. the tapbolt having a standard ." Fig. Studs. They are used under exactly the same circumstances as the stud but are not as good a fastening. and form article) included slotted many forms of of small screws usually provided with a head so that they most usual forms may be set up with a screwdriver. where a large head is desirable. as repeated removal wears out the threads. and especially if the hole is tapped into cast iron. and a nut on the outer end clamps the two parts firmly together. in order to secure ample frictional resistance against turning when the nut is unscrewed. Where they have to be frequently unscrewed. Tapbolts are much used in such work as securing patches on boilers. 39 (b). 39 and cap screws. Sometimes it is not possible or desirable to drill a hole entirely through both pieces which are to is be held together. as they necessarily must be unscrewed from the tapped hole whenever they are removed. (39 e.
Set screws are a form quently used to prevent relative rotation of two machine parts. Machine screws are designated. in inches. being used for sizes above that diameter. is number 000 the diameter of which is . of screwfastening fre recommendations can be found in Vol. the American Society of Mechanical Engineers appointed a committee to establish. at g a flat countersunk head. been unable to agree upon stand ard numbers of threads per inch. for a given diameter of machine machine screw may have 20 or 24 threads per inch. as given in Brown and Sharp's catalogue. Machine screws larger than number 16. a system of standards This committee has reported and their for machine screws. Thus the the larger numbers indicating the larger diameters. Manufacturers have. 40 round as is in Fig. the the screw is made of tool steel the hardening if may be done by The ordinary process of tempering.01316. by numbers. 40. the point hardened to enable If to cut into the shaft. When is the set screw made it in the form shown at Fig.03152. of the Transactions. 40 (a). The difference in diameter between consecutive numbers is . The head while the point .0578. and at h a round head. so that these screws are usually specified by naming the size number first. another standard. far. result made of wrought iron the same objection to may be obtained by case hardening. 1 65 At c is shown an oval fillister head. for convenience. so screw. thus increasing its holding power. . followed by the number of threads per inch. larger than this is given N N N which is about %" in diameter. by the screw c. so that the diameter of any number by the formula d = . 40 (b) or conical as at c inFig. Thus shaft b in Fig. of threads per inch n. Thus.SCREWS AND SCREW FASTENINGS and h). and the serial number of diameter where d is the is not to be confused with the number The number the screw. if possible. 28 Set Screws.0131 + . 40 the hub a set is prevented from revolving on the of the set screw is square (a). are not much used in machine work. may be cupshaped as in Fig. to be discussed later. The diameter of a number o screw is . at /a flat fillister head. smallest size. an 1820 machine screw means size 18 and 20 threads Thus a number 12 per inch.0578. Because of the great confusion now existing regarding this point.
of threads per inch usually corresponds to those of the Sellers system. without the flattened tops and bottoms also in common use. The is thread angle 6o°. Standard Screw Threads.. The first system of this kind was that introduced into England Fig. the top of the thread filled cut off and the bottom of the thread in as shown. Screw to X inch diameter are order to made according some standard system in secure interchangeability. U. obviate this a small conical depression shaft with the To sometimes made at Fig.i66 the cupshaped MACHINE DESIGN end is that it makes a burr on is the shaft which sometimes greatly interferes with the removal of the hub. but there are many is variations in this parstrictly Where the Sellers standard not adhered to . as is shown in Fig. in Fig. only when the load is light. 40 is by Sir Joseph Whitworth. 37 (a). The angle of such V thread is generally 6o° in machine bolts and the number ticular. This standard full not used exclusively in this country. or Franklin Institute thread. The form of the Whitworth thread shown in Fig. The thread angle is 55 and the top and bottom of thread are rounded off as shown. is the Sellers. 37 (c). S. Set screws are not reliable for heavy work and should be used U. but a V thread. in the end of a drill and the form shown 40 used. fastenings larger than Sellers or System. however. The this recognized standard screw thread in the United States S. is The form is of thread is shown 39 (b).
W c 4> • _. .050 2.300 3030 37I9 4. The Sellers screws full have much greater tensile strength screws with V threads of equal angles and pitch. to buy machine bolts of turer only or so specify as to insure interchangeability. Nuts and Heads..027 •045 . Thus the . U. 2 1 507 i \ 10 9 8 7 7 1 1 620 731 83 7 . 3* L' 5J a« 4t 4§ 6ft The area of a 1" full 6o° thread is .428 *A \ 2. ift 3ft 3ft 4 *A 2{ aft 3$ 3 7 3A 3tt 3t 4tV 4* 4 • i it 2 A JJt 2 *A 2V. 4 tl 25 ft ft ft 9 i A i 5 4 A TU ft TS 3 8 § ft TIT A  & II I iA i^t ift i4 ift 23V aft * I ft it I iT lA I* A 5 8 8 * 1% 4 V T 1 8 ft ft itV 13 ft itV it . £P2 0hC/2 £ 3cS £ II H IS 1 — SI'S H £ Inches. Is C In ches. rt ft P Inches. [Inches. OR FRANKLIN INSTITUTE STANDARD BOLTS Bolts and Threads. The Whitworth system the of threads differs from the Sellers in shape of the thread.302 . Sq. Inches.' ^ It It :! 1 1 III 2 2^ 2ft 3 3 lA 1 8 2# 2^ A !. 420 1 ii ii if A ^ i\ 6 6 Si 5 5 1 1 1 1 1 *l 2 A A l 4i A\ 4 4 • :• i 2 2 2 940 065 160 284 389 491 610 712 962 176 426 629 •550 692 .55 square inches.293 1. TABLE x SELLERS.29 :.482 square inches. therefore.2 <S2 ^ (LI §° H No.510 1. because the is thread of the former flattening at the tops only threefourths as deep owing to the and bottoms. as noted above. .'. £0 < Inches. or 14 per cent greater. 620 5.068 •093 126 162 202 . Inches. and the number of is threads per inch also different for some diameters. Inches. Hex Nuts and Heads. while is the area at the bottom of a Sellers thread . S.741 2. 2^ *A 2f 2ft 3* 34 3* 4i 4i 2f 2ft 14 1 1 S s : ift 9 IT .890 . >3 2 u 3 t/5 '0 2 "o 03 53 <" 5 . *A t 5 ift 2 2T ift It 2 ItV It i°57 1.1 A A SCREWS AND SCREW FASTENINGS it 167 one manufacthan is advisable. Inches. I 20 18 16 14 13 1 185 240 2Q 4 i ¥ A 4 \ 344 400 4 54 . 2ft 4ii 4ft 5ft 6 2* 2\ 3 .
1" to 2" pipe. page 29. A. 8 threads per inch. and the rounded shape easier to at the stronger than the root of the Sellers thread. yi" pipe. see Trans. 27 threads per inch. which are comparatively with 4 threads to the inch. In Germany. %" and %" and 2^2 /. to rare. Pipe Threads. make these larger sizes. Straining Action in Bolts due to External Load. When the action of the load oblique to the axis. and other European countries other sys tems are in use. For form of threads and other system. by the SelAbove this size the standard is not adhered to rigidly. to prevent the relative translation of is two or more when a shearing stress is produced in the bolts. France. is The Briggs system of pipe threads the established standard in the United States. some of which are based on metric 57. 58. n X threads per inch. gives the proportion of screws as fixed lers Table X standard for bolts up to 2^" . therefore. the stress in the bolt may be combined If tension and shearing. The numbers of threads per inch for the various sizes of pipe are given below yi" pipe. any screw is tightened up under load there is an initial direct stress (tension or compression) and usually a torsional stress . in the direction of the axis of the bolt. 18 threads per inch. much produce than the Whitworth. E. % n pipe. M. as the size and pitch of the screw becomes rather large for convenience. The load applied to the bolts is generally one which tends to separate the connected members. S. X" bolts while the Whitworth The Whitworth system size of bolt flat gives somewhat is stronger screws as the diameter at the root of the thread greater for the root is same is.: l68 Sellers MACHINE DESIGN system gives 13 threads per inch for gives 12. details as to Briggs and over. It is common. Thus a 6" bolt in the Sellers system will have 2}^ threads per inch. and this action is resisted by a tensile stress in the bolts . but bolts are sometimes used pieces. Vol. VIII. The Sellers thread however.. 14 threads per inch. units.
as in (a). it be taken equal to \ p be taken at l and /«.054 r in a X" screw. 59. the mini mum cross section sustains the greatest stress. If the bolt is used simply to hold two machine parts together. to i°5o' in a 3" screw. Initial Tension in Bolts due to Screwing up. the threads. will the stress that will be induced in the bolt be that to the due to the external load P. page 167. as in to With bolts or studs making a steam tight joint. Thus only a load P be applied an eyebolt. the fit shank should accurately the holes in the connected pieces. is through the bottoms of the to threads.p = 4r* n a i • • • • (14) Values of d are given in Table X. in common if bolts.32 . 41. This be treated more bolt fully later. if at least for the portions near the joint. therefore. This smallest cross section. For a given diameter and pitch of screw the area at the bottom of threads full would be considerably less with V threads. the initial tension in excess of that due screwing up may be much will due to the working load. in . Fig. If p be the tensile stress due load P.1. If. appears that P varies from . stress. and p s the shearing In a bolt subjected to a load which produces tension. equation (12). 39 and there to is no external load tending to separate the parts. or tan a to .35 W with a / 2 " screw to . and d x the diameter of the bolt at the bottom or root of P = 1 4 t d? P'. In the Sellers system the pitch angle of the thread (a) varies varies from about 3 from . as in Fig. and the torsional stress due to the frictional resistance at the thread. rm . . If the load applied to the bolt produces a shearing action. d the diameter of the bolt (shank). the stress in the bolt will be the resultant of the tensile due screwing up the nut. screwed up hard.032 in this same range. and P is the load per bolt.15. for the various sizes of Sellers screws.SCREWS AND SCREW FASTENINGS due to friction 1 69 between the threads of the screw and the nut.
One set of experiments was made with rough nuts and washers. Pr m due . a few years ago. and each had a helper on the 1" and \%" The sizes of wrenches used were 10" or 12" on the 2 " bolts sizes. }i". up to 18" and 22" on the \%" bolts. from 15 to 20 per cent greater than the direct stress alone. and the resulting load on the bolt was weighed. stress as that due to combined twisting and is and it can be shown (see Art. and another set with the nuts and their seats on the washers faced off. A series of experiments was made in the Sibley College Labosizes ratory. chanic using a wrench of ordinary proportions can easily rupture any of these small fastenings. so that the axial force upon it could be weighed after it was screwed up. is equal to Pr m minus the frictional moment at the nut or The resultant stress. to determine directly the probable load produced in standard bolts when making a of bolts used tight joint. is resisted frictional moment body of the nut or collar frictional and the frictional is moment at the thread. The results were rather dis y . determined by equation page 49. 170 MACHINE DESIGN a 3" screw. under these circumstances. Each man repeated the test three times for every size of bolt. so it W with more than The coefficients may be assumed of friction will vary much that for the ordinary range of screw fastenings W approximately P or the tension in the screw W = — P = . direct stress. this. therefore. in general. 1" and \%". Refined calculations regarding the resultant to screwing stress in bolts due up are.. to the wrench pull. Each of twelve experienced mechanics was asked to select his own wrench and then to screw up the nut as if making a steamtight joint. 67) that the resultant (1). especially Since a me in the case of fastenings less than J/&" in diameter. so that the bolt itself is sub jected to a twisting moment collar.33 •33 The by the turning moment. useless and misleading. it follows that the actual stress in such bolts depends entirely on the judgment of the mechanic. The were %" . This moment at the thread transmitted to the of the bolt. A bolt was placed in a testing machine.
and it is probably not in exwhich may reasonably be expected in making a load due to screwing up be divided by the cross sectional area of the bolt at the intensity of the tensile stress bottom of the threads. the loads in the different tests were rather more uniform. . The above experi ments indicate that inversely as the this intensity of stress varies. that a mehanic will graduate the pull (b) on the wrench in about that ratio.000 in pounds.SCREWS AND SCREW FASTENINGS 171 cordant. and that it may frequently equal or exceed p = —— 30. (d) nominal diameter of the bolt. per sq. approximately. as before stated. The general results indicate: (a) that the initial load due to screwing up for a tight joint varies is. or (15) W which IT is is = 16. with the faced nuts and washers. ^ (16) In addition to this tensile stress there considerable twisting action on the bolt. per square inch on a Kinch It also and size this result is substantiated by the fact that steel bolts of this were broken in the course of the experiments.000 per inch of diameter of bolt. If the initial maximum. in is. rf screw P=. and This it d the nominal (outside) diameter of the screw thread. as well as higher. about p = 1 —= 3° "  33lF 30 « 33 X = 3° . value of W is rather above the average for the tests. the is initial obtained. or 30 times the radius. about as the diameter of the bolt. but is considerably below the cess of the load tight joint. agrees with as common experience which forbids the use of screws as small Kinch for cases requiring the nuts to be screwed up hard. lbs.000 lbs. as should be expected.$$W the force applied at the wrench 16. of the screw. a (16) Equation would bolt. That the load produced may be in estimated at 16. give a stress of 60. hence. In these experiments.000 „  . lbs. the average effective lever arm of the from 15 times the diameter.000 d the initial load in pounds due to screwing up. if it be assumed as in the previous far wrench was not paragraph that the turning force acting is at the radius of the is.
by some that the resultant load on the bolt is simply Others contend the sum of the initial and the external loads. A. 60. cates that the stress One to due McBride indiscrewing up a 3^inch bolt was case cited by Mr. even with a very E. 59. that the application of the external load does not change the action of this initial tension It is stated stress in the bolt. as it reduces the tendency to "overhaul" or unscrew. as in some applications the nuts hard than in others. that the resultant load is equal to the initial load alone. sum of the initial load and however. was shown. whichever is the greater. the total load on the bolt would be equal to the the external load. XII. nearly onehalf the ultimate strength. or to the external load alone.172 MACHINE DESIGN discussion indicates that the factor of safety should The above be increased as the size of the screw decreases. unless this external load exceeds the initial load due to screwing up. The excessive friction makes the screw bolt a useful fastening. or useful. James McBride (Trans. as the frictional loss reduces the tension produced in the bolt by screwing up. Resultant Stress on Bolts due It to Combined in Initial Tension bolts and External Load. S. is entirely correct for conditions at They lies represent the extreme limiting cases and every actual case If the bolt itself between them. and of course this factor should be varied with the conditions of the case. safety. Vol. from being broken. that is. that may be subjected to a high tensile stress by screwing up the nuts. Art. Neither of these views tained in practice. could be absolutely rigid while the members it forced together in screwing up yielded under pressure. If.. are much more apt to be screwed up A set of experiments was made by Mr. moderate external load. page 781). or probably very near the elastic limit. His direct determinations of the efficiency of a standard 2inch screw bolt shows an average of only 10. which show that the factor of as bolts are frequently used.19 per It is probably this low efficiency which saves many screws cent. M. The question often arises as to the effect of the combined and the external. load. the members pressed together . is very low.
must equal the reaction Fig. Fig. 48. With such great of the connected differ ence between the rigidity of the bolt and bers. the load mem on the bolt becomes practically the sum of the . of the spring. 47. Assume 1 that an axial force pounds till will compress this spring inch. m Fig.000 lbs. Now. or 4. or the external the greater. is The first of the in above conditions approached by the arrange ment shown of 2. 46. whichever is load alone.01 but this further extension of the bolt would reduce the compression on the spring by a corresponding amount and thus slightly diminish the spring reaction. that the extension of the bolt under this screwingup action. 42. the total (resultant) initial load on the bolt would be the load alone. Assume. if an external axial load of say 2.000 lbs.. be applied to the eye at the bottom of B. Fig.02 inch. the load on the bolt. to increase the length of the bolt by about .000 lbs. or under the initial load of 4. this added load would tend further inch. then its if the nut screwed up the spring inches shorter than free length. J 73 only the bolt yielding. is .000 is Fig. due to screwing up. 43. also. Screwing up the nut compresses the spring interposed between A and is 2 B.SCREWS AND SCREW FASTENINGS were absolutely rigid.
the bolt will not elongate surfaces relieved of pressure. The arrangement shown in Fig. In all ordinary practical cases the difference in rigidity between is much less than in the exThe resultant load on a bolt the bolt and the connected members treme conditions considered above. and he should be governed accordingly.000 inches) . as in Fig. Suppose the bolt lbs. as in and this resultant approaches the minimum limit when the bolts are relatively yielding. This resultant load approaches the maximum when the bolts are rigid relative to the connected members Fig. load is greater than the initial load (say 6. 45. reduced by an amount nearly equal to this But.174 initial MACHINE DESIGN and the external loads. case the designer can tell In any particular which limit is the more nearly ap proached. . and with the above data the bolt would have to stretch an additional .000 initial lbs.000 screw ing the nut up two inches. 43 is one which approaches the other limiting case mentioned above. because the load on the bolt (spring) cannot change without changing the length of the bolt. is The 4. A and B equal to If an external be now applied to the eye in B.02 inch (equal to the initial yielding of the connected members) before the contact surfaces would be entirely load. because the latter entirely to relieve the pressure produced between the contact surfaces in screwing up. It therefore appears that the addition of an external load in this case does not materially affect the resultant tension on the bolt as long as this external load does not exceed the initial load.. the elongation of the bolt increases (to 3 the resultant load on the bolt will be simply the external is sufficient load alone. but the resultant load is necessarily somewhat less than this sum in any possible case. unless the external load exceeds the initial enough to separate these contact and entirely remove the pressure between them. the pressure between is the contact surfaces external load. 44 . may be anything between the sum of the initial and the external loads as a maximum. If the external lbs. and the greater of these two loads alone as a limit minimum. and that the corresponding yielding of the member B is .). is load on the bolt (which is the spring in this case) the contact surfaces of axial load and the pressure between it. in to be a spring which is subjected to an axial load of 4.02 inch.
con necting flanges which are separated by an elastic packing. load on the bolt due to screwing up W '. as indicated in Fig. yielding than the initial connected members. not be difficult to assign a sufficiently close value to even when the actual magnitudes of yielding are un known. call the initial Let be called y and y let = x. in fact. x the exter nal (useful) load W 2 . and a relation derived from which the following method of treatment has been developed: The application of this method depends simply upon the ratio of It the yield of the connected will usually this ratio members to the yield of the bolts. of x is at once found by the above relation of x and is If the yield of the connected mem bers is between 1 and 5 times that of the bolt. only a rough approximation this ratio to the value of this ratio is necessary. design for the external load alone. consider the resultant load . known. 1897) contains an excellent is 1 75 article on the resultant load on bolts. Then it can be shown that W If the yield ratio (y) is = W x + x W 2.5 to 0. 44. and the total (resultant) load W.SCREWS AND SCREW FASTENINGS The Locomotive (Nov. the resultant load equal to the initial load added to from 0. the applied load ation since the value of y the external load is an important considerIn some other cases relatively great. When the conditions are such that the nut is is not apt to be screwed up hard.. The following suggestions may serve as a guide in practical initial problems involving the resultant load on bolts when the load due to screwing up (a) is apt to be considerable. may be a minor consideration as affecting the strength of the bolt. the value y. that when the initial load may be safely neglected. If a bolt is manifestly very much more the greater. or with a metal contact at some distance from the center is line of the bolts. (b) load or whichever is If the probable yield of the bolt that of the connected is from onehalf to once members.8. design the bolt simply for the for the external load. the external load. If a tight joint is made with short rigid bolts or studs.
the initial load on the bolt It is also clear that must be greater than the external . Example through : The cylinder of the steam engine is is 1 2 inches in diameter.000 are often preferable. where overstraining allowed. what must the bolt lbs. For these reasons the ordinary apparent in most ings. 61. 8.000 to 8. per square inch should be allowed.000 lbs. localizes the greatest and cracks are very liable to start from the roots of the threads. Allowable Stress in Screw Fastenings.176 MACHINE DESIGN from onefourth to onehalf the external as the initial load plus load. or for large bolts. especially where the thread is of the full V form. (d) In case of extreme relative yielding of the connected memof the bers. depending on the conditions may be and quality 1 of material employed. where the initial stress may be large.000 ? it is In this case evident that the bolts are much more yielding than the parts which they hold together and the conditions are those of case a in the previous paragraph. the resultant load initial may be assumed at nearly the sum and external it is loads. From the fore going seen that small screw fastenings are very liable to be heavily overstrained by the initial load due to screwing nut. area. held in place by 10 steel pressure is The maximum steam 100 pounds per square inch.000 to 10. (c) If the yield of the connected members is probably four or initial five times that of the bolts. are If the contact surfaces of the is head and cylinder ground together so that no packing of the bolts so that the necessary. take the resultant load as the load plus about threefourths the external load. be the diameter per square inch maximum stress in necessary to insure a steamtight joint will not exceed 7. The reduced stress. not likely to occur. from 6. and if shocks are liable to occur. due to cutting the thread. and the cylinder head bolts. fibre stresses allowed machine members cannot be permitted in screw fastenis For ordinary purposes. for steel. stresses as low as 3.000 lbs. per square inch For such work as steam and hydraulic joints.000 to 4. up the While the body of the bolt is is well designed to resist heavy loads a source of weakness found in the threaded portion.
. per square inch of section sufficient to insure a tight joint.130 ° lbs. whence W if 1 = x 2 W = 2. Therefore d 1 = . Tk 9 " = and X 12' X 100 = 1. Whence d be the diameter at the root of the thread X 4 7.260.000 3.d{ hence 4 X and d x 7. As before.107 lbs. Then _ _.130) 3. .107 = .130 ° 4 X 10 W = 2 W = 2. if may be is in duced by the workman or inexperienced. in screwing up the he careless Example 2 : If in the above example of the studs steel studs are used and rubber packing y§" thick be placed between the contact surfaces. + %w = 2 2.SCREWS AND SCREW FASTENINGS 1 77 load due to the steam pressure in order to insure a steamtight joint.000 = 2.64 inch (at root of thread) It is to which corresponds closely to a $4" screw. fair If this initial load be taken at twice the external load a is margin of safety secured.260 + (X X = 1. l ~ 2 From (c) paragraph 60 the total load w = \\ = w. what must be the diameter ? Here the parts held together are more elastic than the studs and the conditions may be taken as corresponding to those of case c. a much greater stress nut. the initial load W\ may be taken at twice the external load. be noted that is while a total stress of 7.76 inch (at root of thread) to a J/& which corresponds closely inch screw.000 lbs.260. If W\ be the initial load and W 2 the external load per bolt then TFo " = 4 2 12 2 X 10 X 100 = 1.260 lbs.
30. since has been most generally used in long to shock. be done by upsetting the ends where the thread is to This be cut. rather than to machines. 24. there is tie However. since it maintains the full very greatly last fact increases the capacity of the bolt to resist shock. 62. The most apparent result of this practice is to economize material without sacrifice of strength (as the shank still has an area of crosssection equal to the threaded portion). is in duce in the stud by screwing up the nut p which = — — = ——— = 30. since here permanency of form is essential as in machine parts which transmit motion. but may be noted that even should not so the elastic limit be slightly exceeded the efficiency of the fast ening is not impaired. as appears from the com mon It application of such reduced shank bolts only to structures. or impulsive load. Resilience of Bolts with Impulsive Load.000 y% a 34. the extent of distortion of the member due As shown to a given intensity of stress.000 . member subjected to shock. but it struction . while. has been seen that the resistance of a tension is member under a static load in a determined solely by its weakest section.000 lbs. This has not been very generally recognized. and if the weld (when the ends are welded) is perfect. in case of bolts liable an even more important reason for such concan be shown that the reduced section not only strength under static load. bolts. approximately.178 MACHINE DESIGN stress The maximum which the workman may. the total resistance depends upon the in Art. impact. In bridge work and other cases requiring long it is very common may or to make the crosssection through the body of the bolt about equal to the section at the bottom of the threads. maximum stress with impulsive load is . It seems probable that rods. will be increased a little it by the external load. perhaps. the strength of the bolt is not reduced. This is close to the elastic limit. by welding on ends made from stock somewhat larger than that used for the main length of the bolt. this reason it is responsible for the original adoption of this practice.
is greater as becomes less relative to h.SO. produced in a W.—— SS . . but the elonga be confined largely to the very short reduced (threaded) be much less in the larger portion of the bolt.000 (taken as the elastic limit) 1 f nnnr"* Each inch of section A will elongate while each inch of full section A ( = . becomes less for a given is impulsive or the resistance to such action greater for a given value of the If maximum stress. the stress for a given load.78 square inches. The total elongation will then be ..00435 inches.0007 .000 on the .0035". and the To X two remainder of the bolt to be 5" long.0035 = . a stress of only 21. r c bup pose the elongation per inch of length at a stress of 30. r 11 per sq. when the stress on ^' = 30. if A .500 lbs. while the area A f at the bottom of threads 1U lbs. stress on the threaded portion (1") is about the mean of 30.000 lbs.000 and 21.00085 = = .78 sq. is .000 21.000.SCREWS AND SCREW FASTENINGS 1 79 W P (h ~ k d + d) A For a stress within the elastic limit 2W ( h + 8\ 2W ( h \ This shows clearly that or with impact.00085"..78 full sections. the stress produced action. is an ordinary bolt subjected to shock in a direction to pro duce tension. in. the the tion will sections. applied suddenly member of sectional area. or say 25. is only . . stress will be a maximum at the sections through bottom of the threads.= . per square inch. • Therefore a stress on A' of 30.000. in.78 sq. the bolt hence the stress will will elongate.55 sq.55 and .000 X = . with a corresponding elongation of Assume the thread to be 1" long. In a Sellers bolt of one inch diameter the area is A of the shank . while the other 5" (of area A) will elongate under this load 5 X 4 . Hence. as the mean secHence the tion is an average of . inches.) will have is o\o ". It will appear that the mean =0007". d is in creased. elongation for this threaded 1 inch. inches.
8s 8250 X 0.55. Fig. per sq.0416 ^11 = 344 lbs.553 = 457 lbs.00435 . and is also the strongest in torsion. per sq.000 * X ^ = . 46. If the bolt is strained beyond the elastic limit. This form is easy to make. + . 47 is somewhat more expensive. but fits the hole better.l8o MACHINE DESIGN ith = Ty. the larger portion (area A). This The bolt to A'. on the other hand. otherwise it is perhaps the best. by turning down the body of the bolt to about the diameter at the bottoms of the threads.= . than stress of With a load which would produce a 30. section of the shank of the bolt may be reduced as in Fig. more apt to contain a weak section than is a short specimen. relative to the stress. Fig. latter load is 33 per cent greater than the preceding. The collars a and a' may be left to form a fit in the hole. Hence. 48 is the form which gives the best fit. The .10435 + # X .000 lbs. Then 5 the elongation of this portion under the above load would be the total elongation .00085 . inch. X . in greater rate.005".001 = . much in.000 ) 30.785 will be = 43. Now.55 X 30.0035" and would be d = . instead of . suppose to the 5" shank of this bolt were reduced in section an area ^4' = . w = A'p 8250 o ^ —X^ •55 2 30. the sharp notching of the threads is quite liable to start a fracture at their roots. but does not fit the hole throughout its length. W = . preceding example shows that the elastic resilience of the was increased 33 per cent by reducing the body of the bolt Of course the gain would be still greater with a longer It may be well to remember that the "long specimen" is bolt.000 lbs.005 = .'. but. If very long it is difficult to make. and is somewhat stronger in torsion. the stress in the reduced portion (area A f X . the portion thus strained yields at a that given above. the is effect of a long section in resisting shock without rupture elastic much greater even than that shown for deformation only.00585.ooc. and it is weak in torsion.000 X —^.
while the drilled bolt elongated 2. half half were solid (ordinary form) bolts. and the other were of the form shown in Fig. down . one of each kind. and it broke through the shank. or 9 times as much. of these. may be entirely unsuited.SCREWS AND SCREW FASTENINGS These high l8l resilience bolts only increase the resistance to im pulsive load. thus greatly affecting its required size. For cylinder head bolts. is The upon conditions under which a fastening to therefore be carefully considered in order that forces acting of may fastening may it be provided or for. 49. for some bolts such as are used in the connecting rod of the Straight Line Engine. screw fastenings are generally intended to be tension and from the foregoing discussion in this appears that members only. Professor Sweet prepared. Drop tests showed similar results. and other cases where a tight the joint is main consideration. Through bolts only can be used in such a case If the as studs or tap bolts cannot be accurately fitted. It was shown in Art. that where a machine member must absorb considerable shock. Thus in Fig. even when used be used should all manner alone they are subjected to very high stresses. to insure that each bolt receives its full share of this shearing load. the net section of which was a trifle less than that at the bottom of the threads. this form of bolt tests. These tests indicate the superior ultimate resilience of the reduced shank bolts. They are good forms to use in such cases as the socalled "marine type" of connecting rod. if the bolts alone are depended upon to resist the downward force P. 24. 48. because of the greater elastic resilience of the weaker and more ductile material. This principle is of importance in designing fastenings which are subjected to shock where they sarily must neces work under high of stress. the location the may not be advantageous. Location Fastenings.25". not to dead load.25" for the solid bolt. 63. a rather weak yielding material might be safer than one which is stronger and stiffer. they must be carefully fitted. showed an elongation at rupture of . page 77. Tests of a pair of these bolts. which broke in the thread. where the bolts are subjected to considerable shock. Further. As it previously stated.
if the bolts fit the holes in the lugs tightly such a combination of stresses will be induced. in Fig. the threaded portion of the bolt will be subjected to both flexure and direct stress. In small work it is con make all bolts the same size. 50. and where screws are used in this manner they should be designed with a large factor of safety. the sum of their resisting moments being made equal to P I. Thus in Fig. often occurs that the bolts cannot be placed directly in line with the applied force but must be at a distance / (Fig. It evident that the lower bolt must be considerably larger than the upper bolt. ^ p Fig. both flexure and direct stress. such as flywheel rims and brake bands.182 MACHINE DESIGN force is ward resisted by a projecting ledge. be located as far away from A as venient to possible. The bracket now tends and the moment of the load P I must equal of the bolts the is sum moments round the same point. c 4^ 1 Fig. as shown in such a case the lugs be weak and is yielding. The threaded portion of the bolt particularly weak against flexural stress because cracks are easily started at the root of the threads. as at A. in any case. In large work the bolt at C is often made large enough to exert a moment equal to P I. 51. 5 Fig. fit which "is the preferable way. The upper bolt should. 50. In such parts as brake bands the connecting bolts are often used as a If means of adjustment against wear. 50) from The bolts in such cases may be subjected to its line of action. the bolts need not either studs or tap bolts can to rotate the holes closely and around of the A be used. and the bolts near A serve only to insure correct location. 4q In it many machine parts. . to be equally effective.
multiplied by the space through which it moves. to the load. with similar materials. since (neglecting friction). this soon results in a when compared to the cylinder is on which it is formed. decreased. is numerator plied of the fraction and subtracted from the denominator. in Art. the pitch increased the axial width of both thread if it and space are necessarily increased. If the depth of the space reduced. that the square thread is most used transmitting power because of its higher efficiency. 53. a greater load can be raised by a given force P. the general equa tions which have been deduced for the square thread may be used without great error. the bearing surface of the screw . 54. and the moments due thrust collar. In other words. to avoid reducing the diameter of the cylinder. the force applied. This can be seen in another way by considering the energy sup the force P be applied through a complete revolution. and that is when wear must be compensated serviceable (see Figs. It has been for pointed out. if 5 is and the work performed. and very heavy thread is desired to keep the section of the thread square in form. must be equal to the load multiplied by it the space through which is raised. Evidently. or a distance of . Equation (6). 37 d half for the half V thread most and e ).SCREWS AND SCREW FASTENINGS 64. the me chanical advantage of the screw can be varied by reducing the pitch angle. the thread arc and nut may be equally and space on the screw made equal As to each other and therefore equal to half the is pitch. Art. as in the Acme thread.d m the load will be raised a distance equal to the pitch s. If . Where the thread angle of the V thread is small. Screws for 1 83 the Transmission of Power. and it is evident that by reducing the pitch angle a small force applied at a long radius load. the load W which can be overcome desince s s. may be made to raise a great In order that the thread on the screw strong. expresses the relation which exists be tween the turning moment which must be applied to the screw. friction at the thread and at the An examination of this equation shows that for a given applied force P. and inadded to the creases with an increase in the value of the pitch creases with a decrease in the value of s.
give maximum efficiency and hence a more durable thread.184 MACHINE DESIGN is and nut reduced.000 to 10. arranged alternately round the screw. and at low velocities. three. efficiency of the screw alone is considered. has been experimentally determined * by Professor Albert Kingsbury. up to 15 or 20 attaining a maximum beand then decreasing with an increase in the . do not show clearly the effect upon the efficiency due to varying this 65. It is customary in axis of the such cases to divide the axial width of the thread and space into several equal parts. thus forming several parallel threads and spaces. Power Transand (10). In Fig. for reasons that will be presently discussed. 52 will be found useful in making trial assumptions for the efficiency of screws. An examination of this figure shows that for the value of n chosen the efficiency increases rapidly. tical The theory of such screws is evidently iden with that of the single threaded screw. . by parallel threads. hence. their difficult to lubricate rubbing surfaces are imperfectly coefficient lubricated (see Art.000 lbs. Friction of and Efficiency (9) Screws for mission. and that when the tween 40 . in general. ments it appears that. . 52 fi t these equations have been plotted for various constant values of and these curves show graphically the effect of varying the pitch angle. steep pitched threads. From his experifj. 28). becoming zero again near 90 It is to be noted that between 20 and 6o° the efficiency does not vary materially with change of angle. or more this depth of the space can. and 50 angle. Vol. The of friction for screws working under pressures ranging from 3. which is not desirable. while expressing the general relations which exist between efficiency and the pitch angle. wearing ample Such screws are called multiple threaded screws and may have two. The curves in Fig. Equations angle. per square inch. the value of may be * Transactions of American Society of Mechanical Engineers. Screws for transmitting power are usually freely. It is seldom feasible to use such as pitches in practice. The means. as the angle increases. from 30 to 50 pitch angle. 17. for these conditions. be greatly reduced and surface be provided.
£ Efficiency .
as in the case of jack screws. the total torque. . If essential.1 86 MACHINE DESIGN For pressures lower than 3. where the thrust power is applied. (17) The load per unit of projected area is same as the load per unit of true area. be assumed at . but where accuracy of form ft. and is for such service as it lead screws. .000 lbs. minus This moment induces a direct action of the load * This statement applies is the frictional moment at the collar. If the collar is not located at the end to which the power is applied. strictly to the located at that end of the screw to which the . and the axial or projected pressure is equal to the normal or true pressure multiplied by the same function. moment at the from equation equal to W r r$ \ •TT + — d — ft .. since the projected area equal to the true area multiplied by cos «. and wear not an important. torsional stress* in the is screw. 59 that the resisting (6). The to induce a tensile or compressive collar most usual case only. Then W = pnl[d 4 2 — d>] the is . should be PT = load carried. . It has been shown thread is. in Transmission Screws. n = number / = length x of nut in inches. taken at inch. is and where the velocity exceeds 50 per minute. diameter of thread. . if fair lubrication is maintained. p = intensity of pressure per unit of projected area of thread. Stresses in Art.1. where maintenance of form as low as possible.15. very heavy pressures may be carried. 66. d = inside diameter of thread. the value of /i may. of threads per inch. The may be bearing pressure per unit area of thread surface that carried on a screw thread. will vary greatly with the If conditions of service. the pressure per unit area should not exceed 200 lbs. the velocity is low. Pr m of equation (6) is transmitted through the body of the screw.d ~i ln or to the total turning 5 /il m moment applied. per minute. d = outside. important factor. per square and velocities above 50 ft.
An in spection of Fig.SCREWS AND SCREW FASTENINGS the screw equal to 1 87 stress in — (where is A is the area at the root of the six if thread) if the screw short. III. so as to obtain as low a coefficient of friction as possible. flicting may therefore be a requirements. is Thus the turning compromise between these conmoment which can be with hand applied to a screw limited (as is often the case power) a low pitch must be selected in order to attain mechanical advantage. The force required to is open or close a certain sublbs. 64 that the mechanical advantage of a screw It also shown in Art. increases as the pitch decreases. considering (1) the screw as a long column. may be taken. On was the other hand. . Design Screws for Power Transmission. 18. ap plicable to the design of such screws. and of their discussion in Art. If there is no tendency for the screw to overhaul. 67. for man netting the log over to the and the screws saw may be and are made with a very efficient pitch. the screw in compression. The if best pitch. If the screw is over times as long as is its least diameter d lf the compressive stress. In such a case it is obvious that care should be used in selecting the materials of the screw and nut. will be that due to W. steel Thus a bronze nut will (See Art. It is re merged sliding water gate estimated at 6.* maximum efficiency can be Example. the pitch of moment can be selected. that a selfsustaining screw could not have an efficiency of over 50 per cent. as plate planers. a more efficient pitch Again in such cases as the screws in certain machine tools.000 quired to design a singlethreaded steel screw such that one * In certain forms of sawmill carriages these conditions exist. for a given set of conditions. and would seem desirable for that reason to it keep the pitch as great as possible. and the necessary applied.) screw with imperfect lubrication. are therefore. and for perfect safety against overhauling it should be much less than this value. Equations and (2) of Chapter page 48. run well on a 28. 52 shows that screws of small pitch have very it low efficiency. was 54 pointed out in Art.
2 (l\* 77 ~^^ 9. or a diameter of 1 Ms inches.000 Y = . But during one revolution tance equal to the pitch s.000 lbs. is The effect of the thread in stiffening the screw total small and will is be neglected.000V. attached directly to the screw.19" or say 0. 6. which checks enough with the mean Assume the efficiency of the screw and collar at The energy which the operator can supply in one complete revo lution of the wheel = ttX 40 X 60 = 7. lbs.000 per square inch.000 lbs. 28/ assumed. 5 X r 6. is in compression in closing the gate at be designed as a long column square if both ends (case must 4). Hence .000 = i sq. of the screw the gate must move a against a force of 6.140 *.140 inch lbs. per square inch.000 2 mn E\p/ closely ^8V = stress 15 5>3°olbs. per square inch whence f : the trial area of the screw at the root of the thread P = — = . be 10.000 6. greatest unsupported length of the screw friction found Let the coefficient of = .000. at the periphery of a inches in diameter. to The be 4 ft.600 in. 4 Xtt X 30. s = = . the lbs.000 = 1.15 = 1. The maximum unit stress in the screw to and it is evident that the maximum compressive stress p will form the larger part of the total stress.000.600 X . and it so designed it will have surplus strength when in tension. is Since the thread square the outer diameter of the screw will .15. and the mean compressive stress p assumed at 6.2" so that the thread may be easily cut in a lathe.000 lbs. can operate the gate.1 88 MACHINE DESIGN hand wheel 40 is exerting a pull of 60 lbs. maximum working Since the screw tensile or compressive stress = 10. Hence the dis energy delivered at the nut will be 7. Check p ing this assumption by formula N page 94 / = 1T = I+ P. p may therefore be taken at 9.000. inch. the crush ing strength of the material = 30. and the coefficient of elasticity = 30. per cent.000 lbs. per square inch.000 lbs.
" twisting lbs. greater Where the screw is short.4*. 52) the efficiency of the screw about 13 per cent. p = ^ e 16 ^ T rr 3 = 16 * X : 570 r= = x „ 1.ooo 2 direct stress.5— . this increase is may be from it 15 to 20 per cent. is torsional stress due to this moment by equation E. .325. i.15 rm = X 6. Hence the torsional moment at the nut* T = The page 91. or where If the the screw only in tension and hence admits of a high tensile stress. or in order to use a standard tap may be taken as \]4" The corrected diameat the root of the thread will be 1. page 2 ] (48). which less than the assigned limit and the design therefore correct..000 X X .1 — 1.600 is lbs.SCREWS AND SCREW FASTENINGS be dA 1 89 + s = \ /i" l the outer diameter ter of the screw + .*. lbs. 60 20 630 = 570 in. so much is mean direct stress can be carried.4 . 16. . and the corrected mean diameter these a is will be 1.3" . and the original assumption of efficiency is sufficiently close.320 lbs. lbs. The ment that a is increase in the maximum stress due to the torsional mo here only about 6 per cent. by equation (I). The in. II ' moment applied to the screw = 60X20 = 1.3) . For to values tan a = —X— 5 .2 = 1. at the assumed that the collar is upper end where the power is applied.200 The frictional moment at the collar is approximately . screw made of cast iron should also be checked for shearing by equation * It is (2) of Art.32o) 2 ] An = A [p l + \ f f + 4 A = K [o>°oo + is + (4X = 9.2 = 1.7 = — 630 in. (1. From curve (Fig.046 which corresponds = 2° — 40'.5— .2 = 1. the maximum \/9.
the For (c) small loads. but occasionally they are is made of circular form. It is so called It more used than any other. generally made tion which are used primarily to prevent relative rota between shafts and the pulleys. as shown. of Keys. gears. keys. and this is taken into account in designing shafting since the metal must be removed is from the outer applied loads. but the keyway in the hub pulley or gear which in Fig. AND FORCE FITS 68. but so small that it is used only for light work. keyed to On account of the frictional resistance which they induce between the surface of the shaft and the member which is it. and hold ing power is much is greater than that of the saddle key. thus requires more metal flat to be cut away from the shaft than the key. 53. Keys are wedgeshaped etc. 55. and is should fit well in both shaft and hub. The sunk and is key.. A saddle key shown to Fig. Forms of steel. 56. the most secure form of key fastening. or as a safeguard pin as when hub is shrunk on. Fig. to be made fast is cut tapering as shown driven The sides of the key are parallel. because it is sunk into a keyway or groove cut in the shaft. 54 (b) and 54 show two other methods stitute for round taper pins as a subin Fig. which they carry. A flat key is shown This form its requires a small portion of the shaft to be cut away. 56 (b).CHAPTER VIII KEYS. they also often prevent relative sliding of the parts. its This form of key does not require the shaft holding power is be cut. pieces. 190 When the key . COTTERS. 54 (a) is often used. a round shown in Fig. fibre where it is most serviceable for resisting The keyway is cut in a shaft for a sunk key made of the parallel to the axis of the shaft. Keys are most in usually rectangular in crosssection. of applying Figs.
With system. plane is not perpendicular is Where great accuracy flanged couplings on shafting. Fig. or if the key fits Fig. thereby preventing largely the tilting action of the key In the Woodruff system (Fig. owing to this tendency. is a circular segment and the keyway may be will cut with a milling cutter. may be made on the shaft. the part its keyed on may required. bore of the hub is tapering. 57). 54 (a). as in be thrown out of alignment so that to the axis of the shaft. in. the hub should such occur. and the frictional resistance thus to prevent relative sliding of the parts lengthwise If the up helps of the shaft. the hub must . 54 (c). 57. AND FORCE FITS TQ1 the shaft and the shaft set hub arc drawn tightly together on the side of opposite to the key.KEYS. hence it not throw the keyed part out this of perpendicular alignment. the key Fig. Fig. This allows the key in the to adjust itself to the taper of the keyway hub. COTTERS. If the part keyed on does not have a tight or press fit be removed often. 54 (b). 55. Fig. more tightly at one end than at the other. the faces of the flange or part secured to the shaft after the should be faced in place to key is driven. 53. Fig.
sufficiently to allow the shaft proper. there is a tendency to throw the hub eccentric to the shaft. often provided with a head. Sunk Keys. as shown is in Fig. or the key may be required to transmit only a part of it. Under these circumstances there is a tendency for the hub to rock and work loose on the shaft. This makes it necessary to exkeyway along the shaft at least the length of the key. In such cases two keys set 90 apart make a much more secure fastening If as this gives three lines of contact and prevents rocking. it is keyway to be Where it is desirable called a cut without cutting into the to withdraw the key occasion shown in Fig. however. the part to be secured on the shaft position is placed in and the key driven in. such as standardiza * Where a draw key cannot be used it the point of the key is sometimes case hardened so that will not upset so readily in being driven out. and the hub is driven over the key into its correct position. the tend (except when the hub is at the end of the shaft) unless the diameter of the shaft is enlarged under the hub. ally. ma In general. . the fitting is greatly facilitated and the fastening almost as secure as with two sunk keys. in draw key. 53. This would indicate that keys of different sizes should be used with any given diameter of shaft. For practical reasons. in which case the keyway in the shaft is made the same length as the key. 56 (b). of these keys is one a saddle key. Much more force is necessary to drive the hub into place in this manner than to drive the key. chine tools. This torsional moment may be equal to the total torque transmitted by the shaft. as it is which case on the shaft. especially if the direction of motion be reversed.192 MACHINE DESIGN These keys are used largely in be forced on over the key. however. on account of the friction between When the hub is a sliding or an easy fit the shaft and the hub. it is not desirable to extend the keyway beyond the hub. depending on the load which the key must transmit.* Sometimes. Stresses in relative rotation. 69. it Since keys are designed to prevent is evident that every key must transmit a certain torsional moment or torque. and only one key is used.
is subjected to a force. Shafts are usually designed for torsional stiffness rather than torsional strength. and to a resisting force. may and allow it to roll in the keyway. depending on the manof fitting.KEYS. for best of the hub and is shaft against the top fit the key not a tight pressures. the key it is designed as indicated above the shaft is will also have excess strength. 13 Where great accuracy is required. the dimensions of the key. F 3 and F if will by the pressure and bottom of the key. tion AND FORCE FITS 1 93 of the shaft and interchangcability. It is must be very great which to hold the stress key in This may bring a severe bursting fit on the hub. 56 (c) is induced. that keys sidewise only. resisted This tend ency to rotate should be. therefore. section of the key at the outer surface of the shaft. as . All practical systems of keys. the key should be more carefully considered. Keys resisting a torsional moment manner are subjected to simple crushing. F lt due . well evident. it is desirable that the dimensions and key should bear a fixed relation to each other. results. ner of their application and The ordinary to the sunk key (Fig. which results in a shaft consider ably larger than necessary as far as strength is concerned. being such that its strength is equal to the torsional strength of the shaft. where the resisting forces F 3 F 4 have been moved inward and their moment arm made so short that their magnitude position. give a fixed size of key for each diameter of shaft. therefore. If on the top and bottom these resisting be concentrated near the corners. 56 a). these two forces is to set up a shearing stress along the middle pressure from the shaft. COTTERS. If a severe crushing the conditions of service require a continual reversal of motion a state similar to that shown and in Fig. or to crushing and shearing. under these circumstances. deforming both the to keyway and key and subjecting the key action rather than simple shear. They also form a couple which tends to rotate the key in the keyway. presumably. Where short and is designed for strength alone. cannot fit be depended on to carry as great a load as those which on the top and bottom. If. This be sufficient to crush the key at these concentrated pressure points. F 2 due to The effect of the reaction from the hub which it secures.
.. and the key is square for equal resistance to shearing and crushing. the hub is often made a force fit on the shaft and the key fitted only on the sides.. (1) moment d applied to the shaft must equal of the crushing load applied to the side of the key T = P.. 56 (a) Let " / = the length of the key or " thickness of the key hub t= b " " " = T= p= " breadth of the key " torsional moment applied to the shaft " force acting at the radius of the shaft so that P=T 2 d Then for shearing stress p and the since the torsional s p= moment p t n . p = — a .= 2 p rc l(— ) V2 2 4/ . so that it cannot throw the parts out of relative alignment by radial pressure. and hence from (4) for equal strength in ....a 2 orP. then from & (1) and c (3) p lb . (2) J v If F x be considered reduces to to act at the radius of the shaft (which can be done without serious error for keys as ordinarily proportioned) equation (2) r = PJ L 2 Equations (1)  . is (3) and (3) may be used to compute the stresses in any to equal the crushing sunk key... If the shearing resistance of the key resistance. Referring to Fig. such as is generally pc 2 s . = used in keys. If p t = b.8....= F. ..194 in MACHINE DESIGN machine tool construction.. For machinery steel. = p l~ or 2 P* t — = ' = A (4) V4..
In addition. of the shaft. An average value of the thickness may it is be taken at — b. shafts are most usually designed for stiff ness or angular distortion. COTTERS. should not be less than — 2 .as determined by practice. which necessary to give a good grip on the shaft.4 d. and minimum length  d. as has been pointed out. be increased by lengthening the key or by using a hard grade of steeL Keys designed since the friction as above usually have an excess of strength. or thickness t therefore. which are most common. b = d — d = . give keys of breadth b d = — 4 . FITS 195 = i. between the shaft and the hub materially decreases the load actually brought upon the key. but. and therefore are greater in diameter . p = p\ whence from (5) lb=~r 10 2 The minimum is d *d3 6 length of hub (7). . the moment of the shearing resistance of the resisting is to be equal to the torsional moment of the shaft. shearing and compression t AND FORCE If. For steel shafts and keys.=ra (5) where p\ is the shearing stress in the outer fibre of the shaft. Substituting this value of 3 I in equation (6) bd _ *jF 2 4 . The crushing resistance can. key in addition. the thickness is to avoid cutting away so much usually much less than that given above.6 b = . depth = j. however. 8 This gives a key considerably thinner than it wide and makes weakest in crushing. Keys as used in practice conform closely to these rules as far as length and breadth are concerned.nearly 12 4 16  (7) The above would.6b. then T= Pjb.' KEYS.
the crushing stress should be com may be necessary to use two keys. and the shaft is cut Another form of sunk key deeper than for the common form. 58. .* own •2 The may be taken as representing average practice is when the length of the key less it not less than  d. If the length must be puted. . is shown in Fig. it. 70. standards. of countersunk screws 59 b). 5 . . but constrained to is rotate with In such cases a feather or spline used. This key by compression or as a strut. and such keys seldom unless subjected to severe shock or extraordinary loads. Small splines are frequently dovetailed into the shaft (or hub). The sides of the spline are parallel rigidly to the shaft or it and it may be either fastened may move with the hub. while larger ones are often held in place by means (Fig. of Shaft d. Feathers or Splines." page 977. 59 (a). as than this value. . 3 3 2 T6 4 32 T6 5 I 7 16 I « H T6 13 I I ii The drives taper of sunk keys is usually about ]/%" per foot of length. There are no fixed standards various machine builders having following table for the their dimensions of keys. . . Sometimes it is the hub free to slide axially along the shaft. as shown in Fig.196 MACHINE DESIGN If the it than would be required for strength alone. it A common way of securing the feather so that will move * See Kent's "Mechanical Engineer's Handbook. . It has been used with great desirable to have success on very heavy work. 1 1* I* If 2 2i 5 3 31 1 8 4 5 6 7 8 9 10 Breadth of Key b 5 . the keys more difficult to fit. key is made fail must. proportional to the shaft diameter as above. have excess strength against rupture. 1 16 3 8 1 T6 9 a 4 2 8 1 1* if X4 2 a ii Thickness of Key /. The keyways are more difficult to cut. or rivets. TABLE XI DIMENSIONS OF FLAT KEYS IN INCHES Diam. therefore.
greater than that of sunk keys. loosely on the top and bottom. . feather is not rigidly secured to either the hub or the shaft. 1 I 1 1 i* if if if 2 a* a The length of feather keys in general. . Fig.KEYS. 60. therefore. This rolling action tends bring a concentrated crushing force at a and b (Fig. resistance than much sunk keys to the rolling action imto posed upon them (see Art. with The following table common practice: gives dimensions of feathers which agree TABLE XII DIMENSIONS OF FEATHER KEYS IN INCHES Diam. . feathers are usually given a greater radial depth than sunk keys. 69). and from their general It is proportions are often distinguished as is square keys. if the Fig. COTTERS. 58. in order to reduce the bear ing pressure and increase the wearing surface on the sides. with the hub is AND FORCE FITS 197 shown in Fig. and any to a crushing stress fitted Being on the sides in the same way as sunk keys. 59 a). For this reason. they do not produce offer friction less between the hub and the shaft and.  li li if 2 4 t 3 3h 4 5 6 7 8 9 10 Breadth of Feather b. of Shaft d. and in order also to provide ample wearing surfaces. . evident that the holding power of splines not equal to that of sunk keys. Splines are subjected to a shearing stress across the midsection at the radius of the shaft. for the same size of shaft. i T* 6 § A h I I 1* if ij if 2 ai Thickness of Feather/ 1 * A is. 59.
of this securing a piston rod to a piston by case the connection is means In permanent in character. Stresses in Cotters. should be sufficiently great to prevent crushing of the material. 72. 61. but of a shoulder or means rectangular in section. The exact distribution of the loading is indeterminate. The area of the surface of the cotter where also it bears on the rod. In light work a taper pin of circular Fig. is prevented from sliding into the boss by by the cotter alone. 62 section is often used as a cotter. ment of a gib is and cotter (commonly known is as a gib and key). is A at cotter of the a beam supported the ends. rod is repaired or re In other forms of cotter ed joints of this character the not tapered. Fig. The function of the gib This arrangement permits a small amount of adjustment between the strap and the connectingrod for taking up wear on the pin form shown in Fig. 60. and 61 brasses. to prevent spreading of the strap. however. Cotters. This indicates that the diam . 61 shows a method of a cotter. sliding MACHINE DESIGN A cotter is a form of key used to prevent relative Fig. the cotter being is removed only when the piston or piston rod newed. such as used on the ends of the connectingrod of steam engines. and on the hub. ciently accurate. 62. to consider the load as uniformly distributed. Fig. The cotter is usually but sometimes the edges are rounded so as to avoid sharp corners in the opening cut through the rod or to facilitate machining. shows an arrange Fig. as the to concentrate the load bending of the cotter tends It is suffi near the points of support. between two members.198 71.
in fact. though may be very great. work. neglecting the bending action. An auxiliary locking device is often . For a steel rod and steel cotter where p c = .y6d of cotters. t = . which the usual proportion. b = breadth of cotter. cotter must be driven in especially true enough to prevent its is backing This when the load a reversed one as in the case initial stress in the of the steamengine piston. in order that the net crosssection of the rod may be as Let d " ' / • strong in tension as the cotter and rod. the proportions would be different the materials were of steel or of steel from those employed if all and = the diameter of the rod where the cotter passes through. should be so small no danger of backing out and should not exceed K inch per foot of length. be twice that of the rod. is for similar materials. The to the it conditions. the out. Then.& p t . in pump and the hub or boss of steel. it is and in many cases sufficient to compute the is section for shear alone. 61. as The taper that there is shown in Fig. where they bear upon each other. over and above that due to the load P. The all initial stress due driving of the cotter cannot be accurately computed. = thickness of cotter. When is a cottered joint of this character tight made.^d . For this reason calculations of dimensions based on the maximum if applied load should be modified to suit the conditions of service and the materials of which the joint is made. are somewhat similar to those which exist in screwed fastenings (see Art. as is Thus the rod be of brass common cast iron. 60). The section of the cotter at the point of support should be great enough to prevent shearing. cotter This induces an and rod. (2) (3) Good practice gives b =4/ —i.AND FORCE eter of the FITS 1 99 hub should. are in crushing.
shrinkage fit. in the form of a powerful hydraulic press is to be had. depends largely on the If the parts can be forced together in the shop. the hubs of heavy and work in general which is to be subjected to shock or vibration. flywheels. with between up to three or four inches in diameter. General Considerations. and the shaft is forced cold into the hub or the hub is expanded by heating till the bore is slightly larger than the shaft. But if must be erected in the field. must be fastened to the shaft more securely than can be accomplished with a key.000 pounds per square inch as computed from the applied load. . diameter and the bore ture spiders is very small. this allowance may have be reduced on account of the difficulties of erection. and the second as a 73. tween shaft and hub depends largely on the shafts Thus. The student is referred to treatises on this form. An ex amination of successful practice shows an allowable pressure of 15. then The first slipped over the shaft and allowed to cool in place. 62. a difference the diameter of the shaft and the bore such that the parts may Such be driven together with a hand sledge a fit is is often satisfactory. the hub is made slightly smaller than the diameter of the shaft. 62. an allowance requiring a pressure of one hundred tons or more the parts to may be made. impulse water wheels. in which case the taper may be as great as 1 in 8. called a driving fit.200 MACHINE DESIGN used in arrangements such as shown in Fig. the allowance facilities for erection. The degree of tightness or "grip" required beservice. steamengine design for relative proportions of It is often necessary to allow a rather high bearing pressure on the cotter to avoid large and clumsy proportions. and the design based on the maximum applied load. where adequate means. driving the key due to may be disregarded. It is . and the difference between the shaft With such work as armafor the press fit and flywheel hubs. method is known as a force or pressure fit. the stress In the form of cotter shown in Fig. when In such cases the bore of the hub is a sliding fit on the shaft. FORCE AND SHRINKAGE FITS Crank discs.
to obtain a sufficiently tight grip on the shaft by fit a press without inducing undue stress in the parts. etc. E _ L or x . to r. COTTERS. 2 . are of importance 74. AND FORCE FITS 201 usually possible in the case of armature spiders. let A represent a hollow shaft on which has been of the contact surface forced or shrunk a hub or boss B. it built is is often necessary to insure as strong a grip possible without inducing up from separate upon the shaft as undue stress. If.. and the shaft A is in the condition of a thick cylinder subjected to an external pressure. means of Dependmeasure. In such work as crank shafts when parts. r2 +e 2. and E be the coefficient of elasticity of the material. 2. therefore. the radius and the inner radius being r2 . 63. e be the difference between the outer radius of the shaft and the inner radius of the hub. of the hub was r2 —e . flywheel hubs. Due Force Fits. ii (I) In Fig.KEYS. Before the operation of pressing. The hub B therefore. stress at the inside surface of the and the greatest compressive shaft. Stresses of the hub is and the parts are usually put and the bore In the latter cases the stresses induced and should be carefully considered. then e = e + e' 2 . est tensile stress will The great be found at the inside surface of the hub. before pressing. or contraction of the circumference The elongation or 2 ~ x . A greater difference between the shaft diameter fits then allowed than in forced together by shrinking. in the condition of a thick cylinder sub jected to an internal pressure. in a large placed upon the key in all such cases. If x be 2 the elongation or con traction of any radius then 2  x is the corresponding elongation r. the outer radius of the shaft was is.r If p be the then stress which would induce this change of length of circumference. contraction of the circumference per unit of length is 2 . ence for preventing relative rotation may be.
Let Let w 2 be the unit radial pressure between A and B. • • • • v (5) by (5) Pc a (6) A From (2) .p 202 MACHINE DESIGN . the unit tensile stress on the ^ ~2 A = Dividing (4) 2 \" r 2 ' 2 _ /. r x for r 2 . • and (6) ^ and p Fc = * _EeL_ r„ ( « + + w.. w 1 for inner surface of and w 2 2 the hub is w for w + ' 3 . (2) The is." 1906 edition.. and this the tensile or compressive stress at any radius equation to the shaft. ) „2^ = a w 2 • • • • (4) In a similar way substituting in the general equation r 2 for r 3 . — 2 f2 2 W + 2 / 4.. r = r 1 2.. Let p be the unit tensile stress in the hub at a radius r 2 Let p c be the unit compressive stress in the shaft at a radius t r2 . that adopted in this work. w = o. whence the compressive stress 2 at the surface of the shaft is _ w 2 (2 r 2 2 + 4 r. w 2 the external unit pressure r. rt be the internal radius of the shaft. page 396. Applying \o/ Where r x is the inner radius of the cylinder. r 2 the outer radius.. . The notation has been changed to agree with . p) — r2 (a . w p x the internal unit pressure. Let r3 be the external radius of the hub.* ^" O = ^2 ' 2 3 r3 / ) O n ..— general equation for the stress in thick cylinders of this kind 2 2 ft * 2 W.. 2 3 ('.. Then from (i) e = £r 2 +r 2 or p t + e = — 2) . . ^* y2 2 2 (W x W 3 ( r2 2  O 2\ . /?) (8) J^ Merriman's " Mechanics of The following treatment is from Professor Materials..
l it is found that the compressive stress at the inner surface of the shaft is lbs.000X5(^+2. per square inch. A hollow steel shaft 10 inches outside diameter and 2 inches inside diameter is to have a steel crank shrunk upon its end.23 From 3 = £e "«" « 30.000 5 r/( +1) X . r2 = 2 5. if evident that e be assumed.KEYS.75 6. per square inch ? What be the corresponding compressive stresses at the outer and inner surfaces of the shaft? Here/'. Take £ = 30. It appears. made in induce stresses which should be considered . K should be noted that p must be well within the elastic limit to prevent the hub yielding and relieving the pressure. What must be the difference between the diameter of the shaft and the bore of the crank so that the tensile stress at the inner surface of the will hub shall not exceed 20.000 t \\ hence a = 2 r2 2 ——+ rj 2 4 r? ^7 3(>2 —O _ = (2 X 2 3(5—i) 2N 5— + ) 2 (4 X 2 i ) 3 =  4 2 ' and . that the allowances practice for force fits.000.000 X . = 9 (2 and p — 20. 3 4 and substituting 18.000. r x in the above equation becomes zero and the equations are much simplified. as pointed out by Professor Merriman.0044 x X 2.900 lbs.700 From 14) w 2 = A  6 7 ' °° 8.23) fi 30. and w =o. It the resulting pressure and stresses can be computed.98 .0044 2. COTTERS. + 4 rj X 5 ) 2 + (4 2 X ) 2 9 ''"3W0^ (7) 3(9 5 _ ~~ ) 23 Then from p% r2 (a Efi + 20. AND FORCE FITS 203 When the shaft is solid. _ . The hub of the crank is 18 inches in diameter.000. making r = r. r3 = 1. Example. which is usually the case.500 It is this value in (3).000 lbs.000.
fits. generally speaking. where . and the lubricant applied. although greater fits skill is required in fits.0088 and the allowance per inch which is of diameter = . almost im . while giving the probable stresses and radial pressure resulting from a force or shrink fit made with an ance force e. The allow foregoing equations. There is is. are limited in their application to the practical fits. 65. but little is and it is known of the coefficient of friction in such work. A somewhat greater allowance Fig. Fig. the finish of the surface. tween the shaft and hub (w 2) The probable radial pressure bemay be found as above. given.001 " per inch of diameter often allowed. 10 close to average practice lor force is fits. the total allowance or difference between the diameter of the shaft and the bore of the hub would be would be X. In making force how amount of pressure that can be applied to the parts often a controlling factor. is generally made for shrinkage fits. in the example 2 other stresses are to act on the members. 75. 0044 = . . no difficulty making of in making shrink with any practical allowance. as far as getting the parts together concerned.0088.00088". handling shrink ever. the is than force fits. Practical Considerations in Force and Shrink Fits.204 if MACHINE DESIGN Thus. 64. evident that this quantity will vary greatly with the character of the material. as here the difficulty of forcing on the hub does not occur. hence it is Experimental data are lacking on this point.
Machine Design" by Forrest R. shrink fits are superior to force since their surfaces are very dry and unlubricated. Directly as the length of the Directly as the allowance e hub (b) (c) (d) As some function of the radial thickness of hub With the character of the materials and the finish the surfaces. Cathcart. compared to its ring. although the foregoing discussion indicates that this should not be much exceeded considering the stresses induced. Ill. and recourse must be had to successful practice. diameter. Jones. If the ring or or shrunk on to a member be thin. XXIV. of It is evident that a mathematical expression accurately expressing these relations would not be practicable. while those of a force fit are lubricated. band which is forced radially. Total dependence but a key is is. the assumption can be made. Vol. COTTERS. May. In general. Experience shows that the pressure required to fit make a force any given diameter. Machinery. 1897. "Machine Design" by W. Vol. and the stress in the band may be taken This as that is stretching it over an incompressible body. seldom placed on the will vary for (a) also used for safety. that the stress is uniform throughout the crosssection of the in the is The change due to of form member on which the band it is placed compression so small in such cases that may due be to neglected. practically . No. onehalf this allowance sufficient. For shrink fits a greater allowance is often made. or flywheels. 76. is For often armature spiders. fit itself. " Thin Bands or Hoops. An allowance of . without appreciable error. 9. crank pins. AND FORCE FITS 205 possible to estimate the resistance to slipping offered by force or fits shrink fits. and in general where a tight fit is required. therefore. is For further information and practical data the student referred to the following: Transactions of the American Society of Mechanical Engineers.KEYS. L.001 " per inch of diameter will represent average practice in this country for such work as crank shafts.
000 = 15. or / = 47. Then 48— / = total amount of elongation of band.000. if E. and Let I ——  .000. .000 ' E = = unit stress  ' unit strain 30.000 = 2 square inches which may be distributed in any convenient be shrunk proportions.001 inches elongation ( per inch of circumference. where w is the radial pressure between the band and the part on which it is shrunk. . per unit width of the band wd = 2 P. 30. I Whence. If the part on which the band is is to is circular in form.962 The steel tires of locomotive driving wheels are usually shrunk of . — per let the crosssection of the band be %" Then P = lbs. ./ . Example. ? = the length of band before shrinking. Therefore by Art. P is the total stress orw = —— Thus in the above prob lem and let the band be shrunk upon a circular hub x of diameter 4".000 —r— —— = — 48 — 7 .0s ^' yo ins. be taken as 30. where . but rigidly true for circular shapes only. A thin steel is to be shrunk on is to to a casting is whose external 48 inches.band. then.48—^ = unit elongation of1. 78. 206 MACHINE DESIGN applicable to any ordinary shape of band.000 lbs.001" per inch of diameter ) on with an allowance for shrinkage which gives .000.000 4 square inch. . linear dimension where the band be placed What must be the length of the inside face of the band so that the stress per unit area due to shrinking will be 30. . and w = —— a 2P = 2 X 15. the band in the condition of a thin cylinder subjected to an internal pressure w per unit area. the coefficient of elasticity. 60. Thin bands band of this character are usually shrunk into position. 4& tt q — 1.000 lbs. T~ The total area of the crosssection of the band will be A = 60. ? What will be the area of the crosssection of the band in order that the total stress in the band may be 60.000 30.000.
000. an amount equal to the tension induced P at this same point. by bands shrunk is The theory outlined in the preceding article clearly applicable to these members. no stress can come upon the frame itself from the force P.KEYS. without being themselves elastic limit beyond the when the load is applied. Fig. they are designed and fitted so as to put the at frame in compression by the working load strained A. and their dimensions should be carefully calculated so that they will not be overstrained by the shrinking alone. however. COTTERS. E A = Forms 30.000.001 = 30. becomes ineffective. 65. 77.000 X . 64. is be strained beyond the when an external load greater than the total shrinkage stress applied to the parts which they hold together. Other of Shrink Fits. but ordinarily this neglected. AND FORCE FITS tire 207 Taking £ = 30.000 lbs. they are elastic limit. Fig.000. Many machine steel links or parts such as flywheel rims are held together into place. the unit stress in the tire p =. then If. care should be exercised that the stress induced in the bolts by the initial load stress due to shrinking is so low that the additional initial stress due to the external load does not raise this beyond the elastic limit. their usefulness. In computing the dimensions of such links allowance must sometimes be made for the is compression of the small and parts held together. may be Occasionally a bolt or link is used to reinforce a castiron in member shown. Let A. the benefit derived from If them is questionable. screwed up till it just . place. If such members are so designed that they will elastic limit be stressed up to the liable to from shrinkage alone. be reinforced against tensile stress by the that the nut is Suppose. thus giving the bolts a permanent set and destroying crosssection bolt B. the bolts and frame are each to carry part of the load. is Thus open frames. against tensile stress. a large bolt sometimes placed on each side of the throat as bolts are usually put in hot These and allowed to cool in As ordinarily applied. and the link. represent a castiron member of uniform which is to first. taking a permanent set. is and considering the a thin band.
let the section AB be stressed by the P whose arm Let It is O be the location of the gravity axis of the section AB. 4. the casting will be relieved of external load it is W all stress. now an external tensile load 1S applied to the casting. is But the coefficient of elasticity of ca steel. casting compressed. deformation in For these reasons usually the compressive the casting be neglected without appreciable error. now an external tensile load W is applied to the the' to is tendency is to relieve the compressive stress in the casting and increase the tensile stress in the bolt. . and after the compressive stress in the casting addition to W is fully relieved any further still induces a tensile stress in the casting and fur ther increases the tension in the bolt. 59. will" be proportional to the area of their respective crosssections. If 2. that the nut sive load up till W is applied to the cast iron. desired to keep the stress at A not * See Art. also If amount o. the The bolt will be elongat of elongation r ed* and the members. compression depending on the crosssection of the respective' The unit stress induced in the bolt and casting bolt. 65 is /. designed on the basis of the external load alone. sufficient to elongate the bolt as ally When the load applied much as the casting was originIf the itself. is set Suppose. Usually the crosssectional area of the casting is very much greater than that of the bolt. however.) Example. since p=Et> the stress per unit area in the casting will only be onehalf th. is applied to the bolt through the casting is evident that practically the same result obtained.000 lbs. the bolt and casting will be elongate the iron same amount A.000 lbs.208 MACHINE DESIGN If bears firmly on the casting. and the bolt may be (See Art. in the steel. load In Fig. per unit area is all that can be thus obtainec a total compres"  in the bolt. Case a. 60. This would lead to unnecessarily large bolts. Furthermore the compressive initial stress induced in the casting by the load on the bolt is usually very small compared can to the tensile stress induced by the working load. st only about onehalf that of Hence. is the allowable unit stress in th e casting. compressed.
which corresponds closely to a \%" bolt.500.000 X 30 X io\ J = 4. That r portion is of the boss which immediately adjoins the throat subjected to an aver age tensile stress nearly equal to the fibre stress at the surface of the throat or 3..000. If the lbs. where most of the stretching takes place.000. Then from Ple\ + ~T~) Also (If). I = moment of ^ = 10 inches.000 lbs. AT \A P'l'e \ (F_ ^200 P'X 8 X io \ ' I ' 4>5°o But p — p must equal 3. and also of decreasing the frame at the point where the work is done. The be Let " " P = 60. 11.000 1 then the area of each bolt at the root of the thread l l ^ r 1 i • is — 57)°°° 2 = x 15P00 The i.. is Hence the working stress in the body of the bolt 2  = X 2.o sq. These rein forcing bolts serve the double purpose of increasing the factor by reducing the fibre stress. let the area of the section be 200 square inches. cast iron... area of the body is of a i^" bolt.4 square inches. inertia of section = 4. reduce this to 3. per square inch.. per square inch.880 lbs. 2. is applied. and . per r *i in. The upper and lower portions of the boss have little or no tensile stress induced in them. AND FORCE FLTS material 209 is to greater than 3.000 by reinforcing bolts. the tensile stress at page A ) due to P is p= (P I — = to ( 60. induced at A by P' is of safety deflection of the . It will be reasonable to estimate that . J ' Whence P' = 57. as a consideration of a section such as axis is XX whose gravity at O'.4 sq.000. This is when the full working load P at the root of the thread 1 the total tensile load on both bolts. COTTERS. Let Then the compressive stress these bolts be located 8" from O. therefore f ™ 4.KEYS.000. desired it is \ 200 4>5°° lbs.. 91.300 — ( I P' V200 + P' X 8\ ) = 450 / 3.000 I + 60. will 14 show.000 lbs. m... maximum stress be taken at 15.300 lbs.
000 _ . . is made. When the external load is again much the will if smaller than the total load P will strain the casting to point where the bolt becomes effective. 0046 The number of threads per inch on a 1^4" bolt is 5. the total length being 19". Hence after the nut has been set up snugly .000 of the coefficient elasticity initial twice that of cast iron).2 degrees.2IO the stress in the boss MACHINE DESIGN is equivalent to the full stress of 3.. allowing to cool in position.000 lbs.000. the bolts may be stressed beyond the set the first elastic and take permanent time the external load applied.0046 it should be given 5 = . (remembering that of steel is be 3.000 X 2X14/19 = 4.500 = 7. „ The be allowance for shrinkage necessary to give this initial stress will a  l ± = ^9A!^ h 30. strain the casting further than it The total load P did originally. Whence the stress in the bolt will be 11. by first marking up snug in a cold state.380. induced in the will when the stress in the boss is 3. and rotating the nut the desired amount. be especially noted that a very small shrinkage allowto ance is needed is induce a great stress in the bolt.880 — 4.023 of a turn. the stiffness of the frame is materially decreased.023=8. a force applied. This is most easily done in the case of large bolts set if the nut with reference to the bolt when it and then heating the body It is to of the bolt. necessary. the stress lbs. Neglecting the compressive deformation of the boss due bolt to the initial load from the bolts. If too great an allowance limit.500 lbs. and even the stresses induced are not sufficient to rupture the casting. or should be turned through 360 X . its per square inch through 14 inches of length.
d d CHAPTER IX TUBES. 66 represent onehalf of a por tion of a thin cylinder so far effect removed from the ends that their may be Let " neglected. and resolve forces perpendicular to the cutting plane. FLUES. or l = w Yp (3) In other words. a tensile stress is induced in the walls. CYLINDERS. unit internal pressure w = the r / d = the diameter of the cylinder " " " " /> = the radius of the cylinder = the thickness of the cylinder walls p = the unit tensile stress in the longitudinal section = the unit tensile stress in the transverse section t " / =the length of the part considered all Consider the half of the cylinder as a free body. 2 normal force must equal the Hence resisting stress p 1 1 = w I d (1). PIPES. Resistance of Thin Cylinders to Internal Pressure. whose walls are very thin compared to its diameter. is subjected to an internal bursting pressure. If a hollow circular cylinder. the unit longitudinal stress in the walls of a thin . Let Fig. wlr do / sin The do sin this total pressure to this plane I is o w I r 6 = wIrJ do = 2 w r = w I d. AND THIN PLATES 78. orp = — w w = ~j 2 pt ( 2 ). This tensile stress is reduced near the ends by the action of the ends themselves which tend to hold the walls together. For equilibrium in the two sides of the cylinder. The is component of this force perpendicular to the cutting plane 0. The normal is pressure on a longitudinal strip of length I and width rdo normal sin w I r dd.
66. when the longitudinal seams are double or triple riveted. For this very common make the circumferential seams of a boiler shell with a single riveted joint. a (5) or t = wd — 4p t (6) A comparison of reason it (1) and (4) shows the practice to stress in transverse sections be only onehalf of that in longitudinal is sections. . on the head. is and 4 must be equal to the intensity of the transverse duced multiplied by the area of the metal in such a stress pro section. Poisson's ratio. which tends . laterally as if If a tensile stress p t 7ip t is induced in a body. to this cause rupture along a transverse section. . 67." 1906 pi in the page 359. or. (Pw wd A = 4* to — . ~d2 w . divided by twice the thickness of the cylinder walls. 67) it will be seen that the total pressure be considered. the body contracts acted upon by a stress is of pt where a edition.) Therefore the true longitudinal stress is above case (since A equals 3 for steel) P\ 1 wd (1) 1 wd w d This gives a lower value than equation safetv. * In this discussion the mutual interaction of the longitudinal and transverse stresses is neglected. —— = "dtp. and hence the latter is on the side of . Fig.212 cylinder is MACHINE DESIGN equal to the product of the diameter into the unit inter nal pressure. (4) w = 4p Tt t . and is independent of the length of the cylinder. If a transverse section of the cylinder (Fig. Fig. acting at right angles to the line of action (See Merriman's " Mechanics of Materials.
is In like manner by the force the resultant force normal to any section (per unit of length of cylinder) equal to the intensity of pressure multiplied axis of that section. one or circular. for the portion marked I. In a similar way. Since all the meridian sections of a sphere are the same it as the transverse section of a cylinder of equal diameter. Therefore W l + W if 2 + W = w (m + l m + to nt 3 ) = wA axis. Then these resultant forces per unit of length of the cylinder are as follows: W . and a flexural stress tension. Suppose a cylinder to have a crosssection made up of the upper half as a free circular arcs as in Fig. the resultant wB less than uA. FLUES. heads. Resistance of NonCircular Thin Cylinders to Internal Pressure. they will be subjected to a maximum equal to the transverse stress in the shell. but produced in addition to the circular approaches nearer sure increases. is As B is less than A. is greater than the force tending to elongate the major If the tube were perfectly flexible. of the is same thickness to as the are placed on a cylinder which withstand internal stress pressure. 80. 68.f <p ) 7YI. the section is taken along the minor is the resultant force normal to this axis found be wB. AND THIN PLATES 213 Thin Spheres. (section along the Take axis). or the force tending to elongate the minor axis axis. — wr //y sui <f d <f = u> r ( — cos + cos ?') = wm 1 Wn  wR r w / sin S dO = wRi—cosO' + cos 6") =wm — IV 2 . A it rigid material offers resistance to is such change of to the direct torm. body major Let the resultants of the com ponents of pressure which are normal to the plane of the section be TFj.w W* iv 3 — J 1 sin o f/ <pd*p = wr ( — 3 cos  + 2 cos .— TUBES. under pressure. given by shell. in which axes are equal. respectively. circular cylinder The existence of this flexural form as the presstress in a nonof Figs. PIPES. CYLINDERS. II. W W 2i 3. 69 becomes apparent from a comparison . 79. its form of crosssection all would become. is evident that the stress in the walls of a sphere If spherical is (4). III.
(Fig. A similar course of reasoning could be applied to a for such a section cylinder of any noncircular crosssection. each of which could be treated (like the special case of Fig. 69 (circular section) the lines of normal pressure pass through a single point (the centre of the circle) the re. In Fig. the resultant of components of pressure acting normally to any other section (as b . . the pressure tends to reduce the cylinder to a circular crosssection. A direct inspection will also show that in any noncircular section cylinder. the pressures do not in themselves form a concurrent. and they are in equilibrium. grating between proper limits. bottom lower total coinciding with this sectionplane be secured to portion of the cylinder being entirely the removed. and that a it. The pressure on this bottom evidently balances the components of the pressure on the curved surface which lie normally to this flat bottom. 71) could be considered as made up of circular arcs. In a similar way. 71. nor parallel. . hence. Fig. subjected to internal pressure. or moments. and P 2) also passes through this same point. but there must be a moment. 71) equals . the resultant of these normal components of pressure equals w (a . sultant (P r ) of the tensions (P 1 hence they cannot be balanced by a single force (as the resultant P r ) . per unit of length of cylinder. of stress for equi librium.214 and all MACHINE DESIGN 70. system of forces. a) =w A. b. Suppose the original cylinder its (Fig. however. 68) by inte FlG. In Fig. hence these forces form a concurrent system. 70. Fig. 71) to flat be cut along the greatest axis of crosssection.
as well as tension. or homogeneous in character. 78 or p If the cylinder = wd . . This peculiarity is undoubtedly due influence of the heads placed in the ends. uniform in But tubes are never absolutely circular in thickness. the value of which will be given by formula (1) Art. is in Thus a circular If cylinder under internal pressure the section is "stable equilibrium. the smaller this ratio the greater being number of lobes. . circular section. circular cylinder ternal force. were perfectly cylindrical. It is apparent. CYLINDERS." is other than a true circle there a flexural stress. Theoretical Considerations. If a thin hollow cylinder of it is circular section is subjected to an external pressure. tends to correct this departure under internal pressure. that any cylinder under internal pressure tends to assume a circular crosssection. 66 and 68) without course to the calculus. PIPES. when under pressure. obvious that a course of reasoning similar to that in Art. but departing A cylinder of nominal to from the true form some extent. form. of uniform thickness. 1901 edition. AND THIN PLATES 215 w (b . where p is a compressive stress. there seems to be and of homogeneous no reason why failure should occur until the compressive stress reaches the yield point of the material. it or if a under internal pressure its is deformed by any ex tends to resume circular shape. b) =w B < w A . then. material. FLUES.TUBES. Fig. This direct method might have re been used in the preceding cases (Figs. collapse. 78 will that a compressive stress is show induced in the walls of the cylinder. 72 shows the form of crosssection that the of collapsed tubes. RESISTANCE OF THIN CYLINDERS TO EXTERNAL PRESSURE 81. A tube which fails under external pressure is said to and the forms of collapsed tubes are very characteristic. . For values of — / See " Elements of Machine Design. and Unwin* has shown the number of lobes depends on the to the ratio of length to diameter. and is hence failure occurs long before the compressive yield point reached." page 101.
72. greater than about 4 to c only the forms of collapse shown at and d. 72. . Fig. the stress. the force tending to increase the major axis will be seen to be greater than that tend ing to increase the minor axis. Fig. In other words. it may be considered as being in the same condition as a long column. cir cular section any departure from the ideal section will be increased by the external pressure. and the In dimensions of the tube have been developed on view of the fact that the theory of long columns satisfactory. hence the external pressure In a cylinder of nominal will is cause collapse. it is this basis. this pressure will tend still further to increase the deformation. unless the flexural rigidity of the material sufficient to prevent this action. Since the wall of an ideal thin tube is subjected to a uniform compressive stress. a cylinder under external pressure equilibrium. Or. if a cylinder of true circular section is deformed in any way while under external pressure. but the analogy between long compression jected to members and tubes subOther deductions while throwing some light external instructive. If the noncircular cylinders of either Fig. tubes. Fig.2l6 MACHINE DESIGN 6. appear. or flues withstand external fluid pressure. and theoretical equations expressing the relation between the external pressure. is itself most un not surprising that such equations do not accord with actual results. 73." is in " unstable As perfectly true circular sections and homogene ous materials are not attainable in practice the danger of collapse must be taken into consideration to in designing pipes. and they may be pressure is safely disregarded. based upon the theory of elasticity. 68 or 71 be con sidered as subjected to external pressure.
Fairbairn's experiments were conducted with tubes whose lengths were small compared to their diameters. PIPES.547.675. Pipes. hence make the allowable pressure vary inversely as some if function of the length.600— f (2) Many other equations Thus Professor Unwin have been deduced from the experiments of Fairbairn. 82.000 j^^ (3) For tubes with a longitudinal buttjoint w = 9.600— where (i) w is the unit external collapsing pressure in /.w = 9. length. and outside Fairbairn himself modified this diameter respectively in inches.000 7n jii6 like (4) For tubes with longitudinal and cross joints boiler flue £235 an ordinary w = 15. relations. who. to the form .363.614. are Long Tubes.: 7 TUBES. made a series of careful experiments on short tubes and deduced therefrom the following formula: W = 9. in 1858.000 ^^ is (5) Other writers have deduced similar equations from the same data. gives the following as the result of a careful resume of Fairbairn's work For tubes with a longitudinal lapjoint 2. equation. FLUES. AND THIN PLATES 21 on the form of rational equations expressing these not as yet applicable to practical problems. CYLINDERS. /. the effect of the supporting action of the head his equations In such tubes noticeable. etc.21 w = 7. square inch and pounds per and d are the thickness.675. Until very recently the only experimental results on the collapse of thin tubes were those due to Sir William Fairbairn. usually of the same form but with different exponents. for simplicity. Now it is reasonable to suppose that .
the intensity of walls should theoretically yield by crushing. and the collapsing pressure would be independent of the length. except near the ends.. and that beyond that ratio the collapsing pressure is independent of the length. XXVII.52095>520  2. he gives w= k — — (t c where k and c are constants to be determined experimentally and depending upon the material.. 1906.090 . Stewart. 17. 1906. Vol. experiments Carman published* at the the results of a set of made of Engineering Experiment Staconclusively tion of the University Illinois.3 6 5 steel ~ ~ d / 2 A74 (6) For seamless drawn cold w= For lap. He found that the results of his experi ments could not be well expressed by a devised two equations to cover the range pressing the relation which exists between . June.270  1. (7) w = t 83. Ill.welded steel t 95.2l8 MACHINE DESIGN the tube were long enough the head would have no effect. No. .f in an elaborate set of experiments * See Bulletin of the University of Illinois Engineering Experiment Station Vol. single equation. P. p = iv d • 2 t In 1906 Professor A. and the the compressive stress would be given by formula (i) or. which prove that Fairbairn's equations hold only for tubes whose lengths are from four to six times their diameters. but these equations ex w and — a t .. and t\ length greater than 4 to 6 times the diameter. For brass tubes w= 93. T. Thus for values of V025.025 (8) Professor R. f See Transactions of American Society of Mechanical Engineers. In a similar way if the tube were very short..
Professor Stewart found that for values of practically the — below / . or in which the length is at least four times the diameter. and for thin colddrawn seamless steel tubes 50.TUBES.<. The works of Carman deal entirely with tubes which are so long is that the supporting effect of the heads negligible.023 is (11) w for  = about 600 lbs.023.023.200. 83. which corresponds closely with the upper limiting value of less w obtained from (10). For values of  than .\ .386 (9) which corresponds closely with (8) of Professor Carman's work.02 are not much used for en work under external pressure. his results were expressed by . For values results of his of . while . in Furthermore.000 (1/11. found that for values of . whose value for thin brass tubes is 25. PIPES.023 the results of his work could be expressed by the following: w = 86. Stewart and Summary of Equations for Long Tubes.670  1.. or same limit as above.150.o2 5 Professor Carman found that the work could be expressed by an equation of the form where k' as before is an experimentally determined constant. and convenience therefore equation (10) will be adopted. w = The value of 1. . showing the accuracy of the experimental work.000.600^) . Their experiments. CYLINDERS. the corresponding values of (10) w y as found by either or (11). FLUES.^. tubes in gineering which. on lapwelded steel boiler AND THIN PLATES for the National 2ig tubes made Tube Company. do not differ materially.000. .
From d \wc the following usually more convenient formula can be derived: «3b) tube 4 inches outside Example.090 = 83. the equations are not in the most convenient form for use by the designer. supplement and corroborate each other. equation 6 becomes 1  where " for brass tubes .000 steel tubes. As given above. which covers practically the whole range of values of — t . —r~ £ .250. What must order to have a factor of safety of at least 6 ..150. . since usually /. is subjected to an external pressure of the thickness be in ? 300 pounds per square inch. Transposing these equations. A lapwelded steel boiler diameter and 10 feet long.023 and pressures greater than 600 lbs.025 The following approximate formula. . they may be written as follows: of les For values — ^ .200. 2 I w = where k" = 1.. d and w are known and t is required.000 tubes. colddrawn seamless tubes or lapwelded For values of —y K .. . <*3a) • for colddrawn seamless tubes and 1.023 and pressures less than 600 equation 10 becomes 3 t = d Vt ' and 50.000 for thin brass tubes. .270 and c = 1.000.220 MACHINE DESIGN conducted separately. <"> for thin where £ = 25..474 " seamless cold drawn " steel " lapwelded steel £ = 95.520 and c = 2. is suggested by Professor Car man as useful in making rough calculations. d(w + c) (I 3> £ = 93.000 for lapwelded steel k T // "(V this. therefore.365 and £ = 2.
Short Cylinders.000 when computed by the equation w= This . instead of through collapse in (buckling)..675. .800 lbs. of about 4. FLUES.= a / . internal pressure. 4 less In case this ratio should be than . as CL remembering that under external pressure If this equation gives a lower working pressure than (14) the flue designed by it will be safe against The rules of the Lloyd's Marine Register allow the collapse. a second solution.675.14 inch. etc. PIPES.023. should be made. The 2pt — — British flues Board of Trade rules allow a working stress in furnaces and . is the stress p compression. When the cylinder or flue is short. and Fairbairn's apis proximate equation applicable. ~.600 If a cylinder under external pressure could be depended upon to fail only by actual crushing. the effect of the heads should not be neg lected as in Carman's and Stewart's work.800+1. or t 2 w = or transposing 9. following pressure in boiler flues: w= 1.07^. 84. CYLINDERS. Flues.600^ (14) t= \_wl±_ Xl ( } 9. AND THIN PLATES is 221 Here the assumed collapsing pressure 300 X 6 = 1 .025) = 83. . per square inch.\4 to 6. k Here the ratio . Applying equation / (13) = d (w + c) = 4(1. i.14 = °35> an d hence equation (13) applies. using equation 12. then the formula w= — pt — 2  would apply.200 == 1/ 2 t . This is CL Fairbairn's equation with a factor of safety of 9. which will seldom occur.e.270 .TUBES.
S. In practice long flues of large diameter are reinforced at short intervals rings. t and d are all in inches. and (3) may be safely used. Various other authorities give similar equations with practically the same coefficients. (2).075. and if the boiler is to be insured in any company the specific rules prescribed by it should be consulted. It will be observed that this relation limits the use of these flues. Board of Supervising Inspectors. to which equations (1). Thus Lloyd's Register for 19067 gives 1. Hence for the same allowable pressure under these two 2 rules w= or If. much against collapse than plain cylindrical made and with proper dimensions of corrugations may be safely of any desired length. thus by heavy rings of rolled or other section. 85.4 t therefore.200 £ 2 d t Id = 134. .222 is MACHINE DESIGN little less a than that allowed by the U. Their peculiar shape also permits of expansion and contraction under the influence 1 of heat. When the corrugations are not less than % inches deep. and not more than 8 inches from centre to centre of corrugations. I w X .000 2 1.4 equations (1).075. (2). a when /) I. 73 are very flues. change from time to time. and (3) may be applied.200 T 2 t w = . I < 134. and plain . equations to comparatively short Thus a flue %" thick could only be 34" long to have these equations applicable. Corrugated Furnace Flues. stiffer Flues corrugated as in Fig. Various Insurance and Government inspection departments These rules give rules for proportioning flues and furnaces. I p d t 8. known as collapse making the flue consist virtually of a series of short flues. >i2ot and w = 50(300/ a — when / < 120/ where /.
in it can no longer be assumed that the stress the wall is is uniformly distributed over the crosssection. Miller. s perhaps best known. the U. but greater at the inner surface and decreases is is tensile. Of these. subjected to either external or but in a gun tube. " Steam Boilers. allows a working pressure of : " = 14. AND THIN PLATES 223 S.: TUBES. is * now much used and will be adopted in this work. and cylinder thickness. B. Board of Supervising In Rules and Regulations of the American Bureau of Shipping. stress. relatively the internal diameter. which was published in 1880. Rules and Regulations of U. deduced in 1833. Lloyd's Register for 19078 gives a ing various types of flues. . Many of formulae have been deduced to express the relations between pressure. S. for example. spectors. number of rules for design The tion following references contain valuable practical informathis subject on Lloyd's Register of British and Foreign Shipping. is Ordinarily the cylinder internal pressure alone. that Lame. When the wall of a cylinder.000 / ~^r~ Board of This is also the formula of the British Trade. CYLINDERS. PIPES." edition of 1906. which * The student is advised to read the discussion of thick cylinders given in Merri man's "Mechanics of Materials. Clavarino's* modification of Lame's formula. Seaton and Rounthwaite's Pocket Book. to a minimum at the outer surface whether the pressure is internal or external. is which thick is subjected to to internal or external fluid it pressure. When the pressure is internal the stress is and when the pressure external the stress compressive. THICK CYLINDERS 86." by Peabody and S. FLUES. portions at the ends do not exceed 9 inches. I. Rules of the British Board of Trade. Rules of the Bureau Veritas.
.000 — 4 X 1.. PRACTICAL CONSIDERATIONS 87. is 2. has a hoop shrunk upon the cylinder is the more general case occurs in which subjected to both internal and external pressure. Let " w = the internal unit pressure. and is ''"TLtf^J or fi w r r* + 4r2 2 \ (I7) _ . Castiron pipes are widely used for underground water on account or for pipes and to of their some extent also for gas pipes. (16) If the external pressure w 1 2 be zero.w )~] 2 . = the external " p = the unit stress at the inner surface.000. largely durability against corrosion.224 MACHINE DESIGN it. For steam.000 lbs. Hence substituting = rjLAJ^q i L3 A ~4 ^iJ it is = IO 3x4. which is the most usual case.000+1. thick must the wall be in order that the stress at the inner surface may r not exceed 4. castiron pipes are now seldom used . per square inch ? Here in (18) =10. to withstand How an internal pressure of 1. From (16) found that p 2 the ..t r^±ii* L3P1 — 4^i J . per square inch. " " " " outer p 2 = the 1 Then by Clavarino's equation the unit stress at any radius r is (w l . 000 J r or the cylinder walls must be 2. is A castiron cylinder 20 inches in internal diameter. the greatest tensile stress is at the inner surface.000 1 = i2g „ L 3 X 4.000 1 and p x = 4. w = 1. " w = the external " 1 2 " r 1 " r 2 " " = the internal radius of the cylinder.8" thick.000 lbs. stress in the cylinder walls at the outer fibre.620 lbs. (l8) Example. high pressures generally.
For large cylinders both cast iron is yet. For made of wrought iron or steel are ordinary purposes most used. strong. such as marine work. known as extra strong. much used and probably be for some time on account of the ease with which complicated iron castings In the case of steamengine cylinis can be made and machined. CYLINDERS." or similar works. PIPES. per square inch must be withstood. such as stiffness and securing good casting. An examination of current practice shows the average thickness of lowspeed engines to be given by the * The Sttldenl is referred to Kent'. " Engineer's Pocket sizes of pipes.3 15 inches. The proportions of steam cylinders as fixed by practice are the best guide. thicker pipes. still steam and hydraulic will service. »5 . Thus standard of 1 iinch gas pipe has a nominal internal diameter inch.95 inches. Sec also for full tables of standard flanges. etc. and an external diameter Socalled standard wroughtiron piping 100 lbs. For still higher pressures. where pressures up lbs. thus keeping the outside diameter and hence the screw threads for the flanges to one standard. copper and brass are preferred. is still thicker are known as double extra made by decreasing the used. because pipes of their AND THIN PLATES all 225 unreliability. FLUES. of 1. in it is known as "seamless drawn tubing. der. Book. For hydraulic work. such as are found in highclass steam plants. current trade catalogue. thousand piping.587 inches. although in special cases. to several are used. Thus an extra strong would have an internal diameter of of the and a double extra strong inches in for same nominal size would have an internal diameter diameter remaining 1. the latter being which case commonly used for the smaller diameters.* iinch pipe (nominal size) . piping may be "drawn" so that there is no seam. may be used for pressures up to with safety.^ the thickness of the walls fixed by considerations other than those of strength. Wroughtiron or while steel steel pipes may be either lapwelded or butt welded. the external all cases.TUBES.3 15 of . These heavy pipes internal diameter of the standard pipe." its Standard Piping is designated by nominal internal diameter.
For pressures pounds per square inch and pipes not over 12 inches in J. more it must be borne in mind that the thicker the cylinder walls. in common equations based on the theory of caution elasticity.. o$d + ." by tions A. In such with all cases equations (16) to (18) developed above.1. and insure sound castings. in order to get a theoretically lower stress. for instance. should be used with Further when cast iron is selected for the cylinder. 87. to carry a high work ing stress. per Kent's " Mechanical Engineer's Pocket Book" gives the following as representing current practice. Care should also be exercised in cylinders made of castings to avoid excessive thickness of metal at any point. although steel castings are better in general. XVIII. with a is working stress of 5.3. Flanges. where made of castings. as a rule. Cast iron is also much used for the cylinders of hydraulic machines. Transac M.* where t = thickness and d = diameter when the steam pressure does not exceed 125 lbs. 57) ends of the pipes flanges. A 3inch wall. inch. The most usual method for accomplishing this purpose has been to thread the (see Art. etc. If p be taken as 125 pounds this equation reduces to that given by Barr. £ = . or a pair of pipe All of these are in very up to 100 common use. sq. and castiron cylinders are often lined with brass or sound castings. . therefore. Methods for securing the ends of pipes together have become of greater importance as higher steam pressures have been employed. bronze liners to obviate this difficulty. than to design thick walls which are open to suspicion.000 pounds per square inch. a pipe union.226 following.0004^ + . H. S. where d = diameter in inches and p = pressure in pounds per square inch.000 pounds per square inch preferable to a 4inch wall with a working stress of 3. if subjected to high pressures.3 inch. in inches. Pipe Couplings. Barr. the more liable are they to be porous in the interior. t MACHINE DESIGN = . within safe limits. It is safer. * See "Current Practice in Engine Proportions. and secure them together with either a cylindrical pipe coupling. thus insuring Thick castings of any metal are very liable to give trouble by leaking on account of porosity. Vol. E.
73 (b) (see also Art. such as . Fig. the flanges are in sometimes shrunk on as shown In order to insure tightness. In the ordinary screwed cut the thread accurately. especially side the pipe not machined on the out before the flange is shrunk on. 73). 73 (b). is as in heavy steam mains. CYLINDERS. Fig. 73 (c). The gasket usually covers up the joint between the pipe and the This form of coupling if is not well suited for highpressure is work. and permitting the packing or gasket (P) to cover up the screwed joint. the Fig. FLUES. will ensue. sure to work in the flange These flanges are sometimes fitted with into which a strip of soft metal. R (Fig. To obviate the difficulties of the screwed joint on pipe of Fig. so as to make Even a smooth surface. PIPES. however. Fig. and larger diameters they are not so is used on small pipes only. This can be remedied to some extent by making the threaded portion of the pipe enough to prevent long enough to project through the flange slightly. 73 (a). diameter these AND THIN PLATES 227 may be used with success. and then facing off pipe and flange at B. and secure a firmer grip on the usually expanded into the flange. larger diameter.TUBES. 73 b). but for higher pressures satisfactory. flange. The union difficult to fitting of large size it is and to screw the fitting on tight leakage at A. if however. as shown if 73 (a). 73 (a). is liable to leak the workmanship poor or the flanges do not align properly. to When subjected to the heavy straining action incident expansion and contraction. is this joint. the pipe and leaking a recess. 73 (b). Fig. end of the pipe is as shown in Fig. flange.
E. efforts have been made standard use dimensions for pipe flanges. be considered satisfactory in highgrade work. Fischer. 73 c). and the joint made between The flanges F. referred to standard handbooks.* to establish Although several in this country. wroughtsteel flanges are making the pipe is and its flanges one piece. Feb. XXI.. S. when subjected to load pressures. it F This last feature is a very useful one. Stone joints see articles by 1909. one of the most uncertain portions of the mechanics of materials. can be caulked to check small leaks. 73 (b) In the socalled Van Stone joint (Fig. the ends of the pipes themselves are flanged over the flanges so formed. and March .. M. S. M. as joint greatly facilitates erection. prevent the packing from blowing out. THIN PLATES The theory of the stresses induced in is thin plates. This construction. Power. as it possible in steam makes it difficult to renew the packing. and the catalogues of various manufacturers for details of these several systems. F.. but should be avoided lines. Vol. 88. See also Transactions A. pipe and the highest pressures. 23. General Theory. while expensive. of work. Fig. and that known as the Manufacturers' Standard. become clamps for and may be loose on the pipe. of the A. due in part to * For a fuller description of Van 2. The The first is and the second student is used for pressures up to 125 pounds per square inch for pressures up to 250 pounds per square inch. holding the flanges proper together. Fig. This form of has been used with success in high pressure work. 228 MACHINE DESIGN but this can hardly copper. almost essential for large To (b). as if in hydraulic work. several systems are in common The most important of these systems are the standards adopted by the Flange Standardization Committee. 1909. E. A by somewhat rolling the better grip of the flange is sometimes obtained pipe into a groove in the flange. For the highest grade welded to the pipe. as shown at E. W. the flanges are fitted sometimes with a recess and tongue as shown at is H. 73 This construction essential for very high pressures.
. and others..TUBES.. page 409. / of circular plate in inches. Benjamin. angular plate per unit area r = the radius = length and b = breadth of rectin inches. For flat circular plates supported but not fixed at the edges. the authorities differ mainly in the co general form of the equations being in most cases similar. Those due Merriman* will be used in this treatise safe. and P = concentrated load in lbs. as they are simple. the complexity of the problem. Bach. generally. The subject has been investi gated mathematically by Grashof. and others in a measure some of their conclusions. verities and the experimental work of Bach. CYLINDERS.. . (3) and For t = r \ I 3 w for cast iron (4) Mi' flat circular plates supported but not fixed at the edges. .. and give values as Let t = the thickness of the plate in inches. carrying a distributed load t = 2 r \l w for wrought iron or steel . AND THIN PLATES 229 and in part to the scarcity of corroborative experimental data.+ . carrying a distributed load t — = r — for \ > p * \w wrought iron or steel . as any others. is carrying a concentrated load P. w = load in lbs. r { 2 w o t = = I 1— 1 r v' + 2 / g K — r r 1 w — ° for steel or wrought iron (5) and / r y / f o I . Unwin." 1907 edition. p = maximum tensile stress. . PIPES. easy to apply. The mathematical by various to results obtained efficients. which applied to the centre of the plate over a small circle of radius ro so that P=T . FLUES. 1 (6) See Merriman's Mechanics of Materials. (1) and For / r \ 9 w — — for cast iron o p (2) flat circular plates fixed at the edges (encastre). Merriman.log " K w„ — — for cast iron . Then ..
r — 10. it if simply supported. 9 1.3 whence from equation 1 (6) V[l Rectangular Plates. in order that the tensile strength in exceed 6. W = n P n r 7. How thick must the plate be. 2 = 4...e. — 6. and I g e — r = / g e 10 — = 2. ./of —"I (5).000 n 2 = i I. (7) — (8) for cast iron .000 and 2b are the length and breadth supported but not per square inch fixed. . (6). shall not per square inch? Here P = . 4.26S ^ lbs.000..  t = lb^ W 2(?Tb r 3 2 )P .+ . Example.000 lbs. i. of a rectangular plate. ro = X 1 and p . then for plates and for a uniformly distributed load w •«**\ and for fixed edges /i..~m. a load of 4. the load being applied by a bolt whose head 2" in diameter. (8) In equations and w o is the pressure per unit area on the small circle whose radius is r .+ log O e — — for ° o steel or wrought iron I— O — (7) • £ and t r J T. which applied to the centre of the plate over a small circle of radius r o t = = ro y \ . encastre carrying a concentrated load P.000.000 A is circular castiron plate 20" in diameter supports lbs.240 „ 4 If 2Z J 6.230 For MACHINE DESIGN flat circular plates is . at its centre. . P w = o 7Z V 2 The value of w o should not exceed the elastic strength of the material.
but will serve as (I2) equations. in. PIPES. per sq. and these can be safely Thus the U. ( x 3) w = pressure in lbs. s = greatest thick pitch of stays in inches. with stays having a nut inside and outside. American Boiler Makers' Association rules give for steel plates w = 7Where Cxf t .5 . One most important cases flat of flat plates occurs in boiler work. .. Flat Stayed Surfaces. The various Inspection Bureaus and Insurance Companies give practical formulae for the design of such plates. where large areas are held against pressure by stays at regular intervals over the surface. These stays are usually screwed into the plate and the projecting end is slightly riveted over to insure steam tightness. as before stated. Board of Supervising Inspectors* and the used. is perhaps the best guide. S. S. FLUES. — thickness of plate in sixteenths of an inch. . This is is particularly true in cast materials where heavy ribbing of the used and where trained judgment 89.5 as thick as the plate and of a diameter at least * See General Rules the greatest pitch. CYLINDERS. are not to be relied on approximate guides only. and C= a constant as below given C = 1 1 2 for C = i2o " C = 140 " C=i6o " plates T y " and under. and Regulations of U. If / AND THIN PLATES 231 = b equation (9) reduces to ' = w / \lf 8 ^ p and equation (10) reduces to ''^J The above implicitly.TUBES. also Rules of American Bureau of Shipping. Supervising Inspectors. " over T thick. " " y " washers .
surfaces separated It is therefore necessary to keep rubbing by a thin film of some kind of lubricant. is and the whole subject of the design of constraining surfaces closely connected with the theory of lubrication. that when bath or is forced lubrication surfaces is maintained. Experience has that like metals usually do not rub together well. they take on hard glazed surfaces which will run tions except where the velocity is low and the pressure well together. rolling. correct position relative motion of a pair of constrained members may be its that of sliding. are light. machine must move with retained in definite relative motion. and Dry metallic surfaces. will not slide over each other without abrasion. they As the various members of a must be by constraining surfaces. This is well illustrated in slide valves and pistons of * See Chapter IV. under any appreciable load. or a combination of some of these as illustrated in certain forms of cams. as in the case of a shaft journal and its bearing. If two rubbing surfaces of cast iron can be run together for some time without cutting. and by collars which prevent motion endwise. General Considerations. steel Thus steel steel (except when hardened). Thus a shaft is held in position by bearings which locate its axis The of rotation. as in roller and ball bearings. rotation.CHAPTER X CONSTRAINING SURFACES 90. 232 . where both sliding rolling exist.* It has been pointed out in Chapter IV. even when smoothly machined. but when shown on the surfaces are "imper fectly" lubricated the frictional resistance depends somewhat on the metals used. or cast iron on cast iron. the friction between two rubbing independent of the character of the material of which the surfaces are composed. as in the case of an engine crosshead and guide. poor combina on wrought iron.
both run well on hardened steel at and hardened Steel steel mayif be run on hardened very high pressures and velocities. care should be exercised that the particular alloy selected to flow hard enough so as not under the applied pressure. SLIDING SURFACES 91. iron will run very well tin.CONSTRAINING SURFACES steam engines. for the constrainment of rectilinear motion.. is usually called the guide.* Innumerable alloys of this kind market under different names. which have relative sliding motion. influence Other materials. as the ram will of a shaping machine. lubricated will 233 when Care must be exercised that the surfaces are well Soft steel and wrought iron first put in service. run extremely well with iron the or wroughtare of journals. lead. or the crosshead of an engine. (b) Journals and bearings. and the practical considerations governing their application. They can be made upon any degree of hardness. the surfaces are ground true. such as wood. depending largely upon the proportion of antimony used. Very hard alloys of this kind are sometimes known is as white brass. The stationary member of a pair of surfaces. for the constrainment of motion of rotation. The general term sliding * Sec Kent's member be used here to denote the moving " Mechanical Engineer's Pocket Book" for detailed analysis and of the best properties of some known alloys. In using babbitt metal for heavy pres sures. . and wrought on brass or bronze. The most common forms translation of motion in machines are rectilinear and rotation. commonly known as antifriction steel or babbitt metals. the table of a planing machine. The alloys of copper. zinc. will be more fully discussed in connection with the several forms of constraining surfaces. antimony. steel. etc. Forms of Sliding Pairs. while the moving part has various names depending on the service. and polished. therefore the most important forms of constraining surfaces are (a) Sliding surfaces. The conditions which the selection of materials for rubbing surfaces. are sometimes used for rubbing surfaces.
Thus in Fig. the tendency of the platen to raise being resisted Fig. lateral constrainment also a steamengine This form of surface may be considered as a special form of the Z5~ Fig.234 member. guide. Fig. 76 (a). In Fig. Thus Fig. Figs. or circular. In each case the lateral constrainment is only partial. angular. In either case the rubbing surfaces of the guide and sliding member may be either square. Fig. 75 (a) shows a form of angular guide much used on planing ma by gravity. If the circular is . 74 (b) shows a set of square guides for a similar purpose. and this Partial lateral constrainment. while in the case of the the angular one that in the square guide two sets of adjustments must be made to compensate angular guide one set only is needed. shows the crosshead of a steam engine with an angular is complete. may be (b) Complete lateral constrainment. 75 (b) is crosshead with circular guiding surfaces. angular type. obvious that rotation cannot and 76 is (b) show square and angular guides where constrainment complete. Sometimes D is made tapering and provided with a for wear must be compensated . while lateral wear is taken up by the set screws C which press against the wearing strip or gib B. Here. If the surfaces have different centres as take place. 76 (a) O x 2. 74 (b). 76 (b) lateral and vertical wear are both compensated for by the set screws C which press upon the gib D. 74 (a) chines while Fig. vertical by lowering the piece A. The characteristic is which distinguishes the square guide from for wear. guiding surfaces have a common centre at O. (a) MACHINE DESIGN Sliding pairs may be classified by the degree of lateral constrainment afforded the slider by the guides. the crosshead prevented from rotating around O only by the connectingrod and as long as is it is it is so held from rotating the lateral constrainment complete. 74 (a).
There are many places. tion of the bed. in this especially in large tools. slides backward and forward upon another member B. as a rule. such as the cross slides of lathe carriages. In places such as lathe commonly used have the advantage of autoV matically taking up lost motion. There is a tendency flat among manufacturers to discard the V guide in favor of is surfaces. to the relative merits of it square and angular guiding sursquare surfaces are easier to may be fit said. . however. that machine and angular guide than the angular ones. 75. uniform over the whole surface of B. it is evident that. no matter how badly they are is beds the worn. guides But. Fig. although in as plate planers this condition is follows that some machines such Usually.CONSTRAINING SURFACES screw adjustment so that pensating for wear. 77. howapproximated. As faces. the frictional resistance and consequent wear would be practically in practice. English practice. thus comset In such cases the screws C are omitted. in advance of American practice A combination of Principles. the velocity cannot be uniform. particular. it •35 can be moved endwise. Fig. where the much more convenient. difficult to attain and seldom occur These conditions are Since A must it be stopped and started at each end of the stroke. General If short block. carrying a fixed load P. V a and flat guides is also often used. 76. the bearing surfaces of such V guides are very small and wear soon begins to be apparent. in general. if the material in A and B were homogeneous and the velocity were uniform throughout the stroke. 92. especially as the wear from the carriage is usually concentrated on a short por C 1> T7^XD rm / (b)  C ^E A ll r P1M (a) Fig.
however. shorter than the guide. experience that when the sliding block surfaces. and in other machines longer. of the stroke. and would be small compared to the unit normal pressure attained when the crosshead is short. in some machines. crosshead could be made the same length of the stroke. when both It has been found by and guide are made the same length. the sliding member is. the unit bearing pressure. etc. or sliding member. the of shaping machines. the wear. than on another. shaping machines. It is seldom possible. as the guide. and more uniform over the entire contact velocity and normal pressure are zero. In most cases of this kind the wear is liable to be greater on one part of the guide. Thus in the case of a shaping machine the ram seldom operates at full stroke. wear takes place on the forward end ram. and the guides of the are readjusted to compensate for the same. at the middle would be practically uniform over the whole surface. For positions of the crosshead near midstroke the wear would be approximately equal over the whole surface. In the ordinary case the greatest frictional resistance* and wear will occur near midstroke. at The velocity of the crosshead to a also varies from zero each end of the stroke maximum near midstroke. and the wear on the back end the same length Thus in lathe carriages. even under varying load and velocity. to make the sliding member rams and the tables of planing machines. the load Thus in the steam engine. 32. the result being that of the when appreciable back end all. because both velocity and normal pressure between the bearing surfaces are If the greatest at this position. as the guide. the normal pressure its P may vary P be tween the crosshead and guide is zero at each end of the stroke and a maximum near midstroke. but still theoretically greater than at the end positions. tfie velocity varies beginning to a maximum somewhere near the middle greatly. of the ram is very small. and much less than when the crosshead is very short. is very small. .236 MACHINE DESIGN from zero at the ever. as in the case of engine crossheads. the ram will not pass through the guides at hence the * See Art. Again.
etc. below the point of intersection for the wearing repeating this operation to the end of the surface. 77. the guide only need be relieved. 7! —4 (a) Fig." which repay reading. draw the diagonal From a draw the line ac. comparatively long. guide. He has suggested of laying out the wearing strips hand end as here shown. comes on the guide. a^ d S// Sl sT^ Fig. leaving the wearing p \ I ^^F Fig. 1 S and S 2. making any convenient angle with the horizontal. and a similar Professor Sweet* difficulty. as the case may be. Similar wearing strips should if it be cut in the opposite direction on the other member. 79. where the tendency to the following convenient method on the surface of a sliding member. b across the surface to be relieved. Bearing Pressures in is but where a short block slides in a long on Sliding Surfaces. In other machines the excessive local wear result occurs. Lay off ce equal to the width of the face x. a. where little wear occurs on the back or right wear is least. Draw de parallel to ac and take strips S. in certain machines which he has by reducing the wearing surface on that portion of the sliding member or guide. the vertical distance above the point of intersection of ab and de for the first gap.CONSTRAINING SURFACES adjustment 2 37 must be somewhat slack. . The shaded portions represent the parts of the surface which have been relieved. 93. To lay off the surface. It is pointed out Article 32 that the tendency of a loaded flat surface to expel is the lubricant resisted to a certain degree by the viscosity of * Professor Sweet has embodied some of that are Usually his experience in this line in a little will well book called "Things Wrong. and the corresponding first vertical distance strip. Fig. 78 shows a sliding surface such as is found on the ram of a shaping machine. and accurate work cannot be done. has corrected this built.
that the allowable bearing pressures as fixed by practice vary greatly. in general. power is much less marked in sliding surfaces than It is rotating surfaces. care must be exercised in the manner in which the lubricant is supplied. If possible. as here the motion is intermittent. therefore. E. marine engines 10 " 50 " " 30 " " 75 " 94. MACHINE DESIGN This in and its power to adhere to the stationary member. * See Trans. designed for rigidity.* stationary slowspeed engines 30 high " " " 50 lbs. The be and also the guides themselves. the lubricant should be supplied near the middle of the guide. it is much less than can be borne by rotating journals. Owing it to the difficulties of lubrication and compensation if for wear. resisting difficult therefore to lubricate sliding surfaces as efficiently as rotating surfaces. bath lubrication commonly obtained by enclosing the running parts. not surprising. that the bearing pressure must be kept so low that wear is inappreciable. M. in view of these considerations. sliding part.: 238 the lubricant. but the guides should be so It is stiff that localized pressure will not occur. The different following are average values of bearing pressures for forms of sliding surfaces. as a general principle. to Crossheads. as fixed by practice lbs. also The oil grooves in the moving part should be given careful consideration. as pointed out above. in fact considerations of strength seldom need to be considered. From the theory of lubrica753. even with similar classes of work. In highspeed engines. should. difficult to obtain initially true sliding surfaces and. and. very difficult to maintain their accuracy under service. Where this cannot be done. when of the sliding is the guide is horizontal. page . A. XVIII.. Vol. may be stated. S. Lubrication of Sliding Surfaces. therefore. they must be considered as " imperfectly " lubricated surfaces. Further. accurate sur faces are to be maintained. and allowing them to run in what practically amounts to an oil bath. The unit bearing pressure that can be sustained by sliding surfaces is. difficult to lubricate efficiently Sliding surfaces are very on account of the " wiping" action member.
conical. as in the case of thrust bearings. 79 (b) also oil is correct. to insure alignment. the system of grooves shown in Care should be taken grooves. in order that the alignment affected may not be seriously by to deflection. 79 (a) and not as In either case the grooves should be stopped some distance from the edge of the surface so as not to facilitate the escape of the tion oil. 95. JOURNALS AND BEARINGS BEARINGS Forms of Bearings. Where the guiding surfaces are very long. or rotating member. 105.— CONSTRAINING SURFACES tion it is 239 evident that the oil channels on all constraining surfaces should be at right angles to the direction of motion. wherever the velocity If is great enough to draw lubricant between is to the surfaces. to made otherwise their effect relieve any tendency form a pressure film. (b) (c) Strength. oiling devices such as rollers dipping in an oil pocket. to resist rupture under the greatest loads. When the load is so heavy that forced lubricaFig. and other sliding members. and the edges of the are chamfered so as to assist the entrance of the lubricant. the oil being forced in at O. or even (See more complex Art. be made as in Fig. but their profile may be cylindrical. are very effective.) in form. Adjustment. The grooves in crossheads. the edges are square and sharp their scraping effect may seriously impair the lubrication. placed at intervals along the guides. The part other member. member which engages with the bearing all known as a Journals are necessarily circular in crosssections. oil film. must be used. is known as a bearing. (d) (ej Formation and maintenance of an Automatic adjustment. as in planing machines. machine frame. such That portion of the of a is rotating journal. If that the outer edges of the slider. compensate for wear. One (a) or more of the following considerations affect the design of the bearing proper: Rigidity. . spherical. which constrains a as a shaft. in 79 (b). should. therefore.
It is is impossible to adjust the cap bolts so as to be sure that the load uniformly distributed among them. be checked should be such that that the cap stiffness is secured. Usually the question of strength does not enter into the design of the main part of the bearing. in general. wear If this will take place only on the bottom half of the bearing. hence adjustment for wear should also be along this line. The bending moment and Table 1. deflection for such beams are given in Case IX. Thus. first c Consider also that the bearing should be parted if at right angles to the line of greatest pressure. with and the required strength and rigidity may be secured by a proper distribution of metal in accordance with the general principles discussed in Chapter III. however. If. If the shaft must occupy a fixed position relative to the frame of the machine.240 (a its MACHINE DESIGN and b) . affect the form of the bearing. The inside diameter. If the problem more difficult. and its design The exact distribution of the pressure over a bearing is not known. wear is so small as not to if interfere with the alignment of the shaft. is should be called upon to carry the load. the load on the shaft be a simple vertical load P. It follows The last three items. Fig. c. d and e. the cap A. is for strength. For this reason the cap bolts should only two be designed load. as in Fig. to carry more than the apparent bolts are used each should be designed for twothirds of the total if four are used each should be able to carry onethird of the load with an apparent stress of not more than 6. or all the bearings on the shaft wear uniformly. its di mensions should. 80. and also length are fixed by the dimensions of the journal which engages it. alignment must be maintained by . of bearings. or bore of the bearing. and the un certainty of the initial stress due to screwing up the nuts makes load. adjustment may be made by lowering the cap A. 8c. and d. It is evident that the metal of the bearing will wear away most rapidly in the line of greatest pressure. as far as strength and which apply to all forms and stiffness are concerned. per square inch. as often the case. but the assumption a beam loaded and at the centre and of a length equal to the distance between the cap bolts strength will give dimensions on the greatest safe side for deflection.000 lbs.
. It is The greatest pressure will be found near this line of action. 81 is correct for If motion of rotation in either direction. 80.' and very defective as far as lubrication is concerned. load. while the wearing part may be raised to compensate for the wear. however. and is subjected at the same time to a heavy horizontal belt pull P. 80 fulfills both these requirements for vertical load either upward or downward. 81). as the surface is broken near the point the resultant of x P and P 2 be P . 24I lower wearing surface is usually Where this is desirable the made separate from the pillow block. Fig.CONSTRAINING SURFACES raising the lower bearing surface. 80. the arrangement of parts The resultant of these forces is P 2. . P\ :i be reversed in direction and the arrangement is not correct for adjustment against wear. that the journal carries a heavy vertical load P (Fig. will auto was shown in Articles 32 and 33 that a journal matically tend to form a film of lubricant between bearing. and shown will in Fig. be in an upward direction. necessary adjustment It may be made by means of the cap. as in Fig. but is unsuited for lateral pressure from the standpoints both of ad justment for wear and lubrication. thus allowing the bearing to remain fixed in position. Suppose. If the conditions fluid itself is and the supplied under which the lubricant are correct. all If the load P. 82. if. evident that the bearing shown in Fig. pressure may thus be created between the is journal and bearing provided the surface of the bearing con tinuous for some distance on each side of the line of action of the Fig.
82. adjustment is continually reversed is direction. first under one cheek. and in such cases the force P l due to the in steam pressure on the piston. by means of the wedges D. as shown in Fig. will form on the lower shoe. hardly allows time for the formation of pressure films on the cheeks. but the continual reversing of the lateral pressure P. nut A. Occasionally the journal . the the frame C. depending on the oil supply. while lateral wear can be taken up by setting out the "cheek pieces" B." or underneath it. and hence the Pressure films more Fig. best conditions for lubrication do not exist. allow the lubricant to be by the shaft. by this arrangement. 83. "brass" can be raised up by introducing thin "shims. For horizontal pressures in either direction the resultant close to the point at P 3 passes which the bearing is parted. no amount of taking up can in this shape. however. 8 or less perfect. in Fig. Bearings of this form are often used in steamengine work. It is obvious that once the bore of the rectify worn eccentric. and then under the effectively. These carried reversals in pressure. thus lubricating them Sometimes a bearing consists of a conical bushing split at some convenient place. for taking up wear in tions all direc and keeping the shaft accurately aligned and located. and screwing up on B. in fact taking up wear fit manner tends to destroy the of the journal in the bearing. other. Provision is thus made.242 MACHINE DESIGN where the greatest film pressure should exist. Another for a similar case shown Here the shoe or bottom liners. thus closing the bore of the bushing By releasing the bushing may be forced into slightly and compensating bushing its is for wear.
This surface engages with a similar surface bored in the outer casing B. The sleeve may swivel in any direction. but the fundamental principles outlined above regarding the point where the bearing should be parted apply the location to all forms. Practical Construction of Bearings. It is many bearings must be obvious that the fundamental principles regarding for adjustment wear and maintenance of the also. sult which would heating. depending on and service. The sleeve A has a spherical surface turned upon the outside. and various other means of compensating for wear are used. selfadjusting bearings have been found almost Fig. The bearings are sometimes split into three pieces. and adjustment for wear made by moving is the shaft endwise. is last item (e. in In running such as machinery. dy namos. Similar devices are within the range of motion of the sleeves. where used. 84.CONSTRAINING SURFACES itself is 2 43 is made conical. the centre of the surface being at O. this kind it When a shaft has only two bearings of is evident that perfect alignment can be secured. is where neces perfect alignment sary. Machine bearings are made in many forms. Fig. It was shown in . countershafts. The application of such bearings limited to short shafts. Consider the lines of shafting. and motors. automatic adjustment). refastalso. apply to all bearings of this form 96. order to avoid localized pressure. In long it which tend rapidly to get out of adjustment. oil film. used in the case of long shafting. desirable that the bearing be so constructed as to adjust itself automatically to the changing in position of the shaft. but the centre line of the shaft must always pass through O. such as machinetool spindles. 84 shows a bearing of this kind as used in dynamo and motor bearings. essential.
in order that abrasion film may film. be noted that even in the case of perfect lubrication. customary. must be exercised is in the selection of the material for the bearing surface. in that it is very desirable to have the journal maintain it is against wear. till it again fills the recesses. poured in a molten The necessary shrinkage due to cooling. the entire bearing member were made of brass and economy prohibits the use It is and bronze for the entire bearing. piece. which a soft surface than it its form more likely to do when rubbing against would against one harder than itself. Rigidity. where the character of the rubbing surfaces care is less important once the oil film is established. bronze. 80 is shown a to the main castings or may be removable.244 MACHINE DESIGN and the white It is to Article 88 that metal such as brass. so as to insure proper alignment of the shaft. and are prevented from This parting by the parting piece B. usually overcome by hammering the babbitt. or also permits taking lip wear by reducing its thickness as occasion requires. could not ordinarily be attained of the white alloys. Fig. make bearing of the type babbittmetal lining commonly used is for heavy shafts when the rigidly attached by means of dovetail is shaped recesses. therefore. alloys make excellent on account of bearing surfaces for wroughtiron or steel journals. and then boring the babbitt to size. In Fig. the steel casting). 81 shows removable linings of brass or bronze which are circular in section. when cold. to main body of the bearing of cast iron (or sometimes a and to fit into it wearing surfaces of the softer These wearing surfaces may be either rigidly attached metals. The bearing itself should be rigid. Fig. their antifriction qualities. but for all good work the bore of the lining is cast small enough to allow of hammering or peening. not occur before the formed. or in case of failure of the There is a further advantage in having the bearing surface softer than the journal. For cheap work the lining is often cast to size on a metal mandrel and no further work put upon it. 82 shows an arrangement of wearing surfaces common on horizontal steamin place when " liner." . against even if moderate pressure. into which the babbitt state. which would is leave the lining loose in the recesses. and then boring to a smooth turning surface.
The considerations affecting the design of any journal are one or more of the following: . and large stationaryengine important bearings. When tion for wear is important. it through pounding by the shaft. A detailed is description of the many arrangements bearing surfaces beyond the scope of this treatise. For less accurate the main casting itself. or of cast iron there Where danger of the boxes breaking. installations. lined with babbitt.. and of sufficient in Fig. and the lower box or shoe A. The of surrounding water furnishes the only lubricant necessary in such cases. 245 The cap C is is babbitted with some form of it. cheap metal since there either no wear upon all the pressure being downward or sidewise. may be allowed to run over the outside accomplishing the same purpose. water of the bearing. may be of brass or bronze. Theoretical Design of Journals. W here the bearing must work under water. T it is "lapped" out with emery and oil to the required size. Hardened steel bearing surfaces are obtained by making circular shells or bushings. When the bearing must work under trying conditions. These bushings are forced into openings in the main casting and no provision for taking up wear is made. thus carrying off the heat and maintaining the lubrication. The "quarter boxes" is B. are often equipped with a complete system of water circulation on the most JOURNALS 97. they is desired to use a and babbittand compensalined. and there is some question as to whether the heat of friction will be dissipated by radiation. 81." of the required internal diameter. the bearing is cast hollow so that water may be circulated around it. k ' thickness to insure strength. If the forcing operation closes the bore of the bushing.CONSTRAINING SURFACES engine bearings. as on shipboard or in a heated room. work the bearing surface is part of machined to the required size. a lining of or other hard wood is often used. it is customary to make the wearing surfaces removable as indicated of brass or bronze may be made castiron wearing surfaces are used. Highgrade marine work. as in the case of a propeller lignum vitae shaft or the lower bearing of a vertical turbine water wheel. and where babbitt metal. as in the case of machine tools. In an emergency.
The raising of the temperature of the bearing has a tendency to lower the viscosity of the lubricant. The consideration of the proper radiation of the heat . the lubricant it if the bearing becomes becomes so thin that the pressure squeezes out completely. are closely correlated. (d) (e) Maintenance of lubrication. therefore. yielding. and is radiated to the surrounding air by the metallic surfaces till of the bearing. in order to reduce the wear to an inappreciable amount. in such cases. It was shown in Articles 32 and 33 that if the unit bearing pressure on the journal is not too great. must be so great.246 (a) (b) (c) MACHINE DESIGN Strength to resist rupture. thereby reducing the frictional resistance. the same lubricant being used each case. where the special probstrength down in lems in connection with shafts are discussed. This frictional resistance can never be reduced to zero even with perfect lubrication. where the accuracy of the product depends on the accuUsually. and fail absolutely under others. and Chapter XI. The energy thus absorbed appears as heat. because of its viscosity. Radiation of the heat due to first frictional resistance. The third consideration (c) particularly affects such journals as those on the spindles of grinding machines and machine tools generally. material dictates that Economy of theminimum diameter of shaft be consistent with the applied bending and twisting moments. In welldesigned ma chinery the temperature of the bearing should not exceed 150 F. may be drawn in between the journal and the bearing. the wearing surface racy of the journals. or stiffness. The are two considerations. Rigidity. The considerations. (d) and (e). and failure of the bearing by abrasion occurs. covered by the general principles laid more fully considered in and rigidity. that the consideration of strength does not enter into the computations. the temperature of which rises that at which heat is the rate of radiation equals being generated. It is evident. to prevent undue Maintenance of form against wear. are Chapter III. and too hot. that a journal of given dimensions may in carry a given load very satisfactorily under certain conditions. the lubricant.
The number which ing air. without is serious error. transformed into English units. true for only approximately as will bearings with perfect lubrication be seen presently. It follows therefore that friction per unit any desired bearing temperature the work of of projected area of bearing must not exceed the jr rate of radiation per unit of projected area. the bearing be lengthened. statement. It may be assumed. while with the same total load as before. however. instead of per square inch of actual bearing surface. Traction and Transmission. at any given difference in temperature is between the bearing and the surround a fixed quantity for any is set of conditions. January. t 247 herefore. be determined experimentally. w the load in pounds per unit of projected area. that the rate of radiation of heat proportional to of heat units the projected area of the bearing. 57. or fiwV=K where a* or wV = — H (1) is the coefficient of friction. and if the heat of friction per unit area at the desired greater than can be radiated bearing temperature. of heat which will be radiated from a bearing has been experimentally determined by Lasche. if f* It is to be especially noted that be considered as constant. and foot K the rate of radiation per unit of projected area in pounds per minute.* The curves 84 (a) are those shown in his Fig. will be radiated from a unit of surface. the radiating surface is increased and the work of friction remains unchanged. increasing the diameter of a journal (the number of revolutions affect and the bing total load remaining constant) does not materially the development or dissipation of heat. most important. V to the velocity of rubbing in feet per minute. e . same ratio as radiating surface is in creased. 1903. page 52. and with the scale of radiation further modified so as to read in foot pounds per square inch of projected area per second. since the velocity of rubis increased in the If. This is last true for imperfectly lubricated surfaces. The amount shown in Fig. the temperature of the bearing equilibrium is must for rise till obtained.CONSTRAINING SURFACES generated is.
AREA. 8 10 12 IN. or which are ventilated by air currents. with bearings of still air. from a very thin bearing or sleeve and indicates that radiation is more effective as the bearing becomes thicker. as might be expected. 84 (a). PROJ. the usual proportions. for metal is a better contions. in Curve 2 is for bearings which are connected to large iron masses. Fig. 14 16 K RADIATION PER SEC PER SQ. and hence the thick bearing more easily .248 Curve i MACHINE DESIGN represents actual experimental results. Curve 3 was calculated from theoretical considera 4 6 IN FT. It. LBS. gives the radiation ductor of heat than air.
Articles 32 surfaces. 84 (a) give the relations which must and exist between the velocity and pressure in order that the safe bearing temperature may not be exceeded. the value of (1) fi can be deter mined. equation (1) may be made to serve as a useful <>. 84 (a) indicates whether radiation must be assisted by water circulation or currents of artificial means. but the regularity of the supply (over which the designer has little control) affects it pressure. bearings running under the same nominal load and velocity give widely different values of frictional resistance and temperature rise. M. may therefore be used K in Lasche points out that though these experiments represent only a limited variety of conditions. much more seriously.000. and formulae of the form wV = — = C. Notwithstanding this. even for the neous values of velocity and pressure. The for values obtained from these curves equation (1). If the rise of temperature in the bearing be taken as 75 and I 1 be taken as . Vol. Fred much used. Thus if w be V in expressed in pounds per square inch of projected area and feet per W. is ordinarily a safe Transactions A. for imperfectly lubricated a very variable quantity. oil same simultaNot only does it vary with and temperature.CONSTRAINING SURFACES carries the heat 249 away to a greater radiating surface. equation and Fig. Imperfectly Lubricated Journals. depending on whether the load is constant or intermittent. check in doubtful cases by assuma constant ing a safe value of The assumption is sometimes made that jr fi is . or whether the motion etc. which S. safe side and will serve at least as a very useful check in designing. . or if the pressure velocity are fixed by other circumstances. is It has been shown in and 33 that the value of p. Mr. such as 98. Taylor* is gives for mill work C= 24. in designing a journal. E.015.000 not safe for castiron bearings with ordinary lubrication. velocity. air. Fig. where C is a constant that has been determined from practice.. XXVII. is steady or vibratory. they are probably on the If. are minute. and says that C = 12. Further.
and for which . This is more evident when the wide variation of the value of such constants as determined by practice is considered. P.000 50.000 to 15.000 for bearing pressures not higher than pounds per square inch. then MACHINE DESIGN from curve 1.000 to 375. gives. may be checked for heating by equation the bearing pressure of is care should be exercised that kept within the limits which will admit good lubrication. form. due to material.000 200. Unwin. K= 384 whence for ventilated bearings C =  —= 25. Reist Company on generator bearings. Taylor's limits better than would be expected. While. K= 222.000 Railway Carriage Axles 75. as far as imperfectly lubricated journals are concerned.000 to 75. therefore. and also in view of the difference in lubrication and in radiating capacities of bearings. a limiting value of C = 50.000 Crank Pins 50.000 to 78.000.000 for bearing 30 to 80 pounds per square inch. H. The allowable bearing pressures as fixed by practice for various classes of machines are given in the following table. H. as the practice of the General Electric Thus Mr.250 value.u is apparently constant. Fig. Been gives the practice of one of the largest Corliss engine builders as C= 140 60. These values agree with Mr. page 249. All formulae of this empirical form must be considered. whence C =— = 222 . and it may be noted that these are more accurately . and location. which corresponds to the following values of C: Locomotive Crank Pins Locomotive Axles Marine Engine Crank Pins Stationary Engine 250. pressures from Mr.ooo to 100. G.600.000 7>5°° to Crank Shaft Bearings The great variation in these values of C is no more than might be expected in view of the foregoing.015 = 1=5. and while doubtful cases (i). gives values of a similar constant.000 20. /?. From curve 2. 84 (a). these coefficients may form a guide. as applying only to the conditions and range for which they have been found true.
it is evident that the length of the journal must is be such that the bearing pressure It within the allowable limit.. however. . or it may be that the type of machine will not allow space enough for Weight of shaft. (high speed) I overhung crank. 25J or the values of the coefficient of Economy in the use of material and the importance of mini mizing the work of friction suggest that the diameter of the journal shall be as small as ness. flywheels. may be. Bearings for very slow speed as in turntables in bridge work Bearings for slow speed and intermittent load as in 7000 to 0000 punch presses Locomotive Wrist Pins Locomotive Crank Pins Locomotive Driving Journals Railway Car Axles Marine Engine Main Bearings { ggS^SSto (high speed) { [ Marine Engine Crank Pins Stationary Engine 3000 to 4000 3000 to 4000 1500 to 1700 190 to 220 300 to 325 275 to 400 400 to 500 400 to 500 60 to ISO tO 120 2sO Main Bearings for for dead load* steam load . the values of . per Square Inch. . Class of Bearing and Condition of Operation. that the length of the journal thus deterwill mined be so great that localized pressure may result. is consistent with strength and stiff With the diameter of the journal determined by these TABLE XIII BEARING PRESSURES FOR VARIOUS CLASSES OF BEARINGS Allowable Bearing Pressure lbs. Main Bearings Gas Engines. Wrist Pins Heavy Line Shaft Brass or Babbitt Lining Light Line Shaft Cast Iron Bearing Surfaces to Generator and Dynamo Bearings to to to to to 15 to 30 to 400 1300 1500 700 1800 2000 150 25 80 considerations. centre crank [ Stationary' Engine Wrist Pins (high speed) (slow speed) for 900 400 1000 80 200 800 1000 500 1500 1500 100 1500 600 to 1800 to to to to Stationary Engine Main Bearings { [ for dead load.«. etc. Crank Pins Gas Engines.* steam load 140 Stationary Engine Crank Pins (slow speed) Stationary Engine Wrist Pins (slow speed) Gas Engines. CONSTRAINING SURFACES known than radiation K.
3 to 3 to 4 3 From the foregoing the following statements 99. 2 . that. they have an excess of strength while barely satisfying the conditions as to bearing pressure. . but (d) reduced by increasing the length. Values of y a 1 to to I 1 1 . (c) is work and consequent The tendency is of the bearing to heat. the less is the liability to heating. These considerations may require a larger liable to bearing than the previous requirements alone would demand. . other things equal. . of friction The smaller the diameter of the journal for a given co efficient of friction. it is found that the ratio larger its of the length of the journal to diameter ( — ) is fairly well defined for any given class of machinery. ifcto 2. . 5 1 . Summary. While practice shows wide variations. and the ratio of length to diameter must is not be so great that severe localization of bearing pressure result. projected area of the journal must be such that the The bearing pressure will be kept within the allowable limits for the particular conditions. It often occurs. there when journals are designed with the ratio as fixed by practice.: 5 252 MACHINE DESIGN such a length of bearing. may be made regarding imperfectly lubricated journals (a) The minimum diameter of a journal is fixed by the con: siderations of strength (b) and stiffness under the loads applied. Ordinary Shafting with Selfadjusting Bearings Generator Bearings .5 1 to 1 . not materially affected by changing the diameter of the jour nal. . The practice following are average values of — / as found in good TABLE XIV TYPE OF BEARING Marine Engine Main Bearings Marine Engine Crank Pins Stationary Engine Main Journals Stationary Engine Crank Pins Stationary Engine Crosshead Pins Ordinary Heavy Shafting with Fixed Bearings. fore. made In such cases the diameter must be and the length may be correspondingly decreased.
for any given pressure. for the allowable bearing temperature. which one of the several given in Tower's report. The frictional resistance also seen to vary very nearly as the square root of the velocity. per minute. Table VIII. for any given velocity oil is. per minute. and not unstable. but does depend largely on the character of the lubricant. must not exceed the rate of radiation. approximately as the fifth root of the velocity for velocities between 500 is and 2. a definite journal velocity will permit the carrying of a definite load per unit area upon the journal. shows clearly the truth of the second statement. and once the relation is established between the load. for velocities up to 500 ft. It was shown in Article 33 that if a journal is supplied with sufficient lubricant. CONSTRAINING SURFACES (e) 253 The work of friction. Article which was deduced from Table XV. the journal itself may draw in the lubricant till a film is formed under such a pressure that the load will be en tirely fluidborne. for the frictional resistance is seen to remain practically constant with is all loads at any fixed velocity. and practically independent of the velocity for values above 2. Perfectly Lubricated Journals. is constant.000 ft. per minute. of proper viscosity. as in the case of imperfectly lubricated surfaces. following table. The The first of these statements has is been abundantly verified by experiment and discussed more is fully in Article 90. (that w— a constant) (c) The coefficient of friction of perfectly lubricated surfaces. and perfect lubrication. within the limits of pressure under which the may be maintained. With any given set of conditions. (b) The film ft frictional resistance of perfectly lubricated journals is. it and coefficient of friction. therefore. independent of the pressure. varies very nearly as the square root of velocity. per unit area. (a) The friction of perfectly lubricated surfaces for a given little velocity depends very on the materials which form the rubbing surfaces.000 ft. velocity. It was further shown that the following statements are true regarding perfectly lubricated surfaces. shows .. 100. $$. per unit area.
the bearing must attain a tempera .So 3b 1 a u 1 05 ft. from which Table XV is taken. Fig. w.306 •357 . were conducted at 90 F. velocity. between pressure.306 .582 . and those obtained at temperatures as high as are usually allowed in practice.626 . appears that to radiate this amount of energy. and It is to frictional resistance for which fortunately covers the most usual conditions be noted that at the greatest pressure in practice.2 54 MACHINE DESIGN the coefficients of friction for the range of pressures and velocities given. The work of friction at 471 ft.648 •539 •444 •503 •459 •576 •758 . be taken as representing fairly well the rela tion existing this range.619 •794 Seized .62 . 209 ft. Tables VIII and XV may. if the bearing is to radiate the heat of friction.663 277 248 277 •357 . and without artificial means of cooling the bearing.408 •5°3 Tower's experiments ties.721 •546 •445 •35 .619 •532 •583 . 471 ft 573 520 4i5 363 258 i53 100 583 . 262 ft. per minute. especially since those at 90 are on the safe side.416 •3 6 4 •423 •459 .401 .605 .771X471=365 ft. lbs. as far as designing is concerned.561 . if a perfect oil film be maintained. per minute with a load of 520 lbs.678 •597 •495 . per minute. 157 ft. is seen to be . From curve I. for velocities in feet per minute as below. or that with this greatest load a lower velocity must be assigned. 84 (a). per second.691 . lbs. His principal experi ments.496 .386 . n. at different temperatures show that the veloci coefficient of friction. 366 ft. 314 ft. Temperature = 90° F.488 •547 .510 . for the above range of pressure and decreases as the temperature increases. therefore. The differ ence between the coefficients of friction obtained at 90 F. can be neglected. 419 ft. TABLE xv BATH LUBRICATION Frictional resistance in pounds per square inch of projected area of bearing surface = j«.714 735 . the latter not exceeding 500 ft. and highest velocity. indicating that with such velocity a lower pressure is to must be assigned.561 .771 •655 . the bearing seized. or about 6 it ft.
for a practically constant and independent of the (See Table XV. have per minute with perfect been quite fully determined by Wm. and coefficient of friction. w be the bearing pressure in fiw or (t He found pounds per square inch. As far as designing is con cerned. January. form.000 lubrication. Lasche found that high velocities the coefficient of friction was practically independent of the velocity. but varied inversely with the pressure as in the Tower experi ment^ and that if also varied inversely with the temperature. the results very con and should be carefully read by designers of highspeed machinery.000 efficient of friction varies about as the fifth root of the velocity. or a total temperature of at least i8o°F. * See Traction and Transmission. as shown by the experiments of Stribeck. may be neglected.) The frictional resistance. and t = 51. pressure. and Lasche's equation may be applied. without serious error."• wV the frictional loss per unit of projected area in foot minute. the difference between the coefficients for this range and those found by Lasche for the higher velocities. Since .CONSTRAINING SURFACES ture of over 255 no° F.* The experimental work was clusive. very extensive. ings running at velocities of over 2. per minute. . It is to be especially noted that within the limits of pressure oil film will is where a perfect given velocity. but a few of the most important results will at these be considered. the frictional resistance. to all velocities above 500 (3) ft. equation (4) may be used to pounds per which a heat compute the 1903. A discussion of these experiments is beyond the scope of this treatise. Equation may be written "™ V where V is is = ( 51 2 V (4) 7Z7^ ft. above that of the surrounding atmosphere. w = ('32) ft. the rubbing velocity in per minute. (v 6 per minute the co For velocities between 500 and 2. Lasche. then  32) ~ /'. O.2 (2) the temperature of the bearing in Fahrenheit degrees. for bearft.
Vol.968 ft. Since the specific heat of both water and oil are known the supply of either necessary to carry off the excess heat of friction can be calculated. and let it be required to keep the temperature 150 F.800 inch.990 ft. * Transactions A. Fig. per minute. M. whereas the bearing alone. XXVII. Then by equation (4) the frictional work is." by Frank Foster. E. 80 to 86 pounds per square inch were repeatedly carried at velocities up to 1. values given by these experiments were obtained on experimental machines and ings gives velocities ranging may be looked upon as limiting Successful practice in the design of steam turbine bear from 1. per second. per minute. connected to a heavy ft. radiate only 6. f See " Steam Turbines.000 ft. per second. per minute or 14. thus materially assisting the radia tion. at these high velocities. and gives evident that with values agreeing with those just quoted. The limiting values of the pressure under which a perfect oil film can be maintained. 84(a).4 lbs.000 and w = 75 in accordance with the empirical rule just at given.000 is much used for this class of It is work. will. lbs.256 MACHINE DESIGN perfectlylubricated highspeed journal can radiate per square inch of projected area. iron frame. the bearing it was artificially cooled. with pressures inversely as the velocity ranging from 80 to 50 pounds per square inch. The empirical equation wV = = 150. Where the pressure is as high as 90 pounds per square kept below 1. Thus t let V 2. from Curve 2.000 ft. page 425. the oil per square inch of projected area was carried at a velocity of 1. if c ft. . page 181. above the atmosphere. and not rise above a temperature t. The values. these high velocites the radiation must be assisted. ft. either circulation.800 to 3. or a temperature of say 75 F. 2 X 2. S. lbs.. 51 wV= — 11 — = 867 (150 — 32) . have not been fully determined. In Lasche's experiments a load of 213 pounds In Kingsbury's* experiments loads from per minute. In both Kingsbury's and or Lasche's work self.f it is found that the velocity must be per minute.
concentrated at the middle of will This assumption on the safe side. of course.. and the determination of the stresses acting upon them I is a part of the solution of the xl. the coefficient of friction is cor respondingly increased is . in the absence to strength of exact data. It is desirable. sometimes give shaft diameters excessively large as is far as strength concerned. (a. ]* M >TFig. 7 It is and c) show the most important assumed in each case that the bear . Journals generally form an integral part of a shaft or spindle. stresses in the shaft itself. as a rule. to point out some of the special features of journal design. 10 1. hence the total frictional resistance n W also increased. 85. but there is fairly is every reason to believe that the distribution uniform. in the is direction of the axis. however.. wear quite uniformly over their entire length. 86. It is customary. b The ] following examples cases of journal design. Thus bearings.CONSTRAINING SURFACES It is to 257 ft be noted that with perfect lubrication the product is w for any velocity a constant quantity. to rigidity assume for computations as journal is is and and that its the load on the length. where fair alignment is maintained. Fig. Examples of Journal Design. is For if the unit bearing pressure decreased. Care should. The actual distribution of pressure over a journal. be exercised in any is case that the heat of friction properly carried away. either by increasing the diam eter or length of the bearing. not known. It follows therefore that for any given total load W the unit bearing pressure should be it kept as high as possible provided does not exceed the max imum allowable value for the given conditions.
'. K 2 applies M e =[x + Vx + 2 i]T = [2 + 2 V2 + 2 i]4. inches (see Table the bearing 5 pressure F be — 7\ + T = *% X 7 500 + 300 = 50 lbs. by the location which the must occupy. or say 7 2 If the length of the bearing be taken at will M1 .000 (500 p. = 4. the most common con but the application of the theory to perfectly lubricated is journals obvious. and the journal may be considered as subjected to a combined bending ing and twisting moment. d — 8 n . torsional stress due to the twisting moment (T — T 1 d 2) — . there The bending moment The twisting M = (T (7\ . moment T = 2 — T . so a. . and shear due to the direct pull 7\ + T 2 The last is small and is usually neglected (see Art. 26).. page 94. Formula K 2) 2 or K 3 (page 49). 85 tre of the bearing fixed at O.= taking . It is required to determine the dimensions of the journal.. Example This case is is illustrated in Fig.ooo 3 . is 500 and the pull on the slack 300 lbs. .000 Hence M = 8. the pull on the tight side of the belt lbs. 1575 which is a safe value. a = 10 inches. 300) 20 .480 From equation J. which dition. 1 +T 2) a =(500 + 300) 10 = 8. it is found from r Figure 9 that equation .480 _ x 10. flexural stress due to the bend. . M 3« c = pTld pi — = or . 71 p n X 8.. that the pulley overhangs the of the pulley d is bearing by the distance The diameter side is 40 inches. fore. applies.'. (a) . 258 MACHINE DESIGN is ings are imperfectly lubricated.000 = — I 4. The stresses induced in the journal are.000 2 = x and — = p . . The centre line of the pulley belt M is also fixed at XX. moment (7\ + T 2) a.000 2%" XIV) tl d = 2}i inches (nearly). rfi _ 3i^. = 8. Here the cenby the construction of the machine.
25 / the length. and be taken as . . as in the case of the steamengine in Fig. total maximum crank pin be 25. X . inches. d be the diameter of the is pin and 1. From Table XIII safely carried it is seen that 900 lbs. it is seen that to radiate this amount of energy the temperature of the bearing will rise air. Let the line of action of the load pass through the centre line of the journal.2 5 = 78. and = 5X1.2 lbs. Fig. number of revolutions be 300 per minute.CONSTRAINING SURFACES If the 259 /'. pressure on the and the pounds. = 22.015.7 or say 1 inches. and hence are usually strong enough and enough if designed for a bearing pressure low enough to prevent overheating.25 =6.000 X 6. per min.000 or d 2 " . . than to as concentrated at the middle. 3. then the projected area of the pin lxd = dXd = i. case the maximum bending moment M = Wl = 25.000.2 5 ^ = 4. Let — be / taken as 1. 84 (a).125 inch pounds . lbs.25. about 50 the design above the surrounding is satisfactory. Let the length of the crank be 18 inches. per sec. Whence from Table I.25^X900 = 25. The pin may. In a short pin of dis kind more accurate assume the load uniformly assume it tributed along the pin.. therefore. ter are short it is seen that journals of this charac compared stiff to their diameter. From Curve 1. the work of friction per unit of projected area will be 50 ft.25 to The this pin it may now is be checked for strength. this type of pin.015 X 300 X ' —= 133 ft.2$d Whence 1.00c What should be the dimensions of the crank pin in order to be safe against rupture and overheating? Referring to Table XIV. This is a safe value and Example crank pin (b). be considered as a cantilever uniformly loaded with a load W = 25. 86. or 2. per square inch If may be on 2 .
000 X 21 }i = 537. From Table XIV. * . Thus is in Fig. 86 neither the length of the bearing B. the twisting moment moment . which a safe value. then the bending inch pounds. M = pi — or p = Me M — .000 M = 25. if by means (c) . into account (see next chapter). may be checked for deflection. and in such case a tentative method must be adopted. which unknown. it is In the case of the steamengine shaft. 6. (b) The data taken in example correspond to a cylinder diameter of about 18 inches. page 94. can be definitely decided upon diameter of the journal. till something known about the The diameter must therefore be asto the sumed. The projected area of the is 9X20 = 180 square inches. for example. may is be taken as yi" in height and since the from case (b).500 as before. Usually a close estimate can be made from is existing ma chines of similar type. of case 3. . and then checked by the equations which apply case. and from Table IoO XIII this it is seen that this is a safe value as far as the load due to steam a heavy flywheel pressure is concerned.400 pounds nearly. In a similar way the pin desired. which gives a bearing pressure of — — = 140 pounds per square 25.000  inch. nor the thickness of the crank hub t.= — 32 is = or ^> = ^— = 6. I. the length of the journal may be taken as 20 inches. that the known main journal frequently about onehalf the diameter of the cylinder. The induced in the journal are of the same character as in case Taking the length of crank Z = i8 inches. The boss under the pin pin.25 inches long. If the shaft also carries must be taken stresses (a). for this di ameter. the total distance from the centre of the crank pin to the centre of the shaft journal may be assumed as 21 % inches. Table Example Sometimes the location of the bearing is dependis ent on the diameter of the shaft. and the journal diameter may therefore be assumed as 9 inches. and the pressure on the pin = 25. The length of the hub should be at least 8 inches.260 MACHINE DESIGN from equation 7.
Common drop sight feed cup.000 inch pounds. and is thus forced in between the surfaces. 9 apply to the case. 9 a safe value and the design satisfactory. M 9 = I [. 25. Imperfect Lubrication Compression grease cup. the oil is supplied to the bearing under a low pressure is which insures that the journal application.v + vV + 1] r =   [1 . 102. as in always flooded at the point of it bath lubrication. Lubrication of Journals.500.'. no reliance being placed on the tendency of the journal to draw in the lubricant.K) + 2 1] 45°> 000 = 616. From J which is t/ (as in example ix a) p = 32 TZ —^e = = 32 X 7T 616. [ Perfect Lubrication i Bath lubrication. equation K 2 is found by Fig. Common wick or siphon feed cup. whence M — = to 1. Forced lubrication. The point of application of the lubricant is of utmost importance. while supplying the lubricant under slight pressure. Oily pad against journal. .19 = and taking p — s . In forced lubrication the supplied at a pressure in excess of the film pressure at the point of application. but does not force the oil is lubricant between the surfaces. The most common methods of feeding lubricants to rubbing surfaces as given in Article 30 apply fully to journals and may be classified as follows: ' Common oil hole. I In flooded lubrication (sometimes erroneously called forced lubrication).1 CONSTRAINING SURFACES 26 T = x. Ring or chain feed.000 X 18 = = 450.soo d is X 3 = 8. gives only imperfect lubrica . Flooded lubrication. and the method of supplying affects the lubricant to the journal sometimes materially the design of the bearing. The compression grease cup. 19 + Vi. Centrifugal oiler.8.600.
If the velocity of the journal is so low as to draw in . to While the decreased evident. In such a case the lubricant should be supplied if from underneath.) of The location and character of the oil grooves deserve special attention. Thus. (See Art. A pump in continually circulates the oil to the various bearings. a ring device necessary. is while again. therefore. lubricant can be supplied at in either direction. and the hydraulic pressure under which the is oil supplied should be greater than the maximum bearing pressure. If the H for motion oil pressure were upward. the very complete apparatus for supplying flooded lubrication will be found. except the compression grease cup and forced lubrication. The bearings are constructed so as to catch all as it leaves the journal. with greater speeds. it friction due effort perfect lubrication is does not follow that an should be made to design every bearing so as to secure this advantage. care should be exercised that the point of application is at.262 • MACHINE DESIGN is tion as the supply of lubricant not copious as in the case of forced lubrication. will suffice. In forced lubrication the point of application should be the point of greatest bearing pressure. In many modern power instal with either steam turbines or reciprocating engines. or wise an oil the direction of rotation were anticlockat hole as shown / would be good design. The same results are obtained and cooled by flooded is as with bath lubrication. or near. and the best installations the oil is filtered during the lubrication circuit. if the pres always downward. in others a constant supply from a wick feed oiling lations. as any tendency for a pressure film to form would be destroyed by relief of the pressure at the hole. oil. In some places a simple oil hole is sufficient. and pipes convey it to a central receiver. an hole at H would not only be useless for supplying lubricant but would be fatal to good lubrication. In applying any of these methods of lubrication. in Fig. Forced lubrication resorted to only where the bearing pressures are excessive and beyond those which can be supported by the natural action the film formed by rotation of the journal. 80. 33. a point of lowest pressure and at the place where the journal will naturally sure is draw in the lubricant.
The sharp edges of the bearings themselves should also be filed or away for the same reason. especially where the lubricant is But where the velocity minute (see Fig. If the oil is delivered at and the pressure is either downward or upward. thus relieving of a pressure film any tendency will to the formation it and the lubrication not be as good as would be if no grooves were present. A bearing which binds sidewise will not lubricate properly. 80. are allowable so long as they terminate at a little distance from the edge of the bearing. a When a shaft is subjected to heavy end thrust. so as to If cut distribute the oil along the entire length of the bearing. the surfaces should be relieved for distance from the parting line to help the wedging action of the oil to and to insure the journal against side pressure springing of the bearing under the load. /. kind running from and the pressure is downward. itself.CONSTRAINING SURFACES little 263 lubricant the oil grooves should be so cut as to allow the lubricant to flow in near the points of greatest pressure. Fig. care should be is used that no grooves are cut that will tend to prevent the If the lubricant is delivered at oil formation of the pressure film. known as thrust bearings. THRUST BEARINGS 103. the simple collars which prevent end thrust in ordinary shafting will not and bearings of special form. have been found helpful in drawing in the lubricant under such circumstances. on the journal heavy. is it must be provided. as in Fig. If the bearing is taken on the end of the shaft designed so that the thrust is called a stepbearing or . 80. General Considerations. the grooves should be cut at right angles to the direction of motion. power transmitted. H. some due little Where one bearing surface encircles nearly onehalf of the shaft. which will distribute the grooves of any over the surface H oil of the journal. of all oil grooves should be carefully oil The sharp edges to facilitate the removed scraped passage of the under the journal. oil and for above 25 feet per ordinary pressures. diagonally they will extend under the journal toward the point of greatest oil pressure. either from the weight of the parts carried or on account of the are used to suffice. 16). or scores Grooves.
and the bearing relative velocity they have the effect of reducing the between adjacent surfaces. follows that the surface will wear unevenly. is If the motion of rotation very slow. is if and if the rotative velocity of the pivot high. to Fig. and thus keep the diameter of the pivot small. theoretically. velocity of rubbing. or pivot. they are very useful as a safeguard against cutting. as is called a collar bearing. remove the wearing surface near the center. This will bring a concentrated pressure at other points. meet the requirements. often better. If a number of discs are placed (Fig. the greater wear taking place at the outer edge. that the unit pressure between the faces of a newly stepbearing is uniform at all points. always advisable in heavy work. and. is a maximum at the outer edge. where the motion is slowest. Fig. 88. is effective within limits. since the area increases as the square of the diameter while the velocity of rubbing increases directly as the diameter. It may be assumed. it is however. and where eventually the greatest concentration of pressure is likely to be produced (see Fig. Decreasing the bearing pressure by increasing the surface. even pressure per unit area is when the low. will If. and hence increases the lost energy. between the step. Since the wear it is pro portional to the product of pressure and velocity. 87). the velocity high. as if shown is in Fig. 88). increases the average Increasing the of radius. to carry a higher bearing pressure. for this reason. however. The without great fitted error. this simple arrangement will not give good results. however. moment arm the It is frictional resistance. therefore. for abrasion should begin between any pair .264 footstepbearing. MACHINE DESIGN If the thrust bearing must be placed at some it is distance from the end of the shaft 104. 87. 87. StepBearings. a simple cast iron step. even is the pressure heavy. and heating and It is cutting may result. the case in swinging cranes and similar work. zero at the geometric centre of the pivot.
or some other metal. 89. 90. as shown in Fig. the centrifugal action will set oil. 89. makes an effective bearing. and radial grooves cut across the faces permit a flow of oil between the became effective again. . up a con tinuous circulation of the making is the lubrication effective. of For very light work the shaft sometimes rests on a pair hardened steel buttons. it is found that the tendency to wear in an axial direction * See Church's " Mechanics. but the relative motion between the surfaces decreased and the wear thus reduced. An oil hole passes through the centre of the washers. 91. The shaft unit pressure between washers the same as between the combination and the is first washer. centrifugal action assisting the lubrication. as shown at N (Fig. order to allow the shaft automatically to adjust Sometimes the its washers are made lenticular in shape. in alignment. If the top of the bearing is connected to the bottom by an oil passage. 90). 265 the lubrication motion will cease at that point till These washers are usually made alternately of steel and brass. which often capped with bronze. rests is on a lignum ing water. Fig. vitae step and lubrication effected by the surround If the outline of a stepbearing be made that of a tractrix* (Fig.CONSTRAINING SURFACES of discs. Fig. and the upper and lower washers are fastened to the shaft and bearing respectively. alternating with brass or bronze washers. or a hardened steel ball which runs between hardened In the submerged stepbearings of is steel surfaces is introduced. of hardened and ground steel washers. water turbines the shaft. surfaces. A Fig. 88)." page 181.
is The lu brication thus obtained not to be confused with that obtained on horizontal rotating bearings discussed force to formerly. as horizontal bearings. While oil centrifugal does drive the from the centre action the outside. for all stepbearings. The foot step was and rested directly on the bearing. This is. page in. as perfect lubrication although giv ing excellent of ments The experiBeaucamp Tower* on a results. and pressures up fully to 160 pounds per square inch were success carried at 128 revolutions per minute. on account of its Viscosity. steel foot step. This is a misnomer. 1891. oil was found that a single diametral groove was better than more. If under heavy loads the maintenance of lubrication is im * Transactions of the Institute of Mechanical Engineers. theoretically. there is little on the to part of the surfaces themselves tending. At 240 pounds per square inch the freely lubricated. no washers being interposed. the correct shape. three inches in eter. are rotated together they tend to as has been proven however. gives diamIt considerable information on this subject. after the discoverer of the above property. for the friction of a tractrixshaped step is much higher than that of a plain pivot. uniform in fact if two homogeneous flat surfaces wear into the form of a tractrix by experiment.266 is MACHINE DESIGN at all points. it is seldom The tractrix has been called Schiele's Antifriction Curve used. but on account of the difficulty and expense of machining the surfaces. draw in the lubricant between them. Such lubri cation cannot therefore be looked on THRUST BEARING OF CURTISS VERTICAL STEAM TURBINE Fig. . therefore. It is evident that the rubbing surfaces of all the stepbearings which have been discussed can be submerged in an oil bath. 92. bearing seized.
The outer diameter of the ring. vertical shaft The A. and prevented from rocking on E by means of by the screws F. provided with collars integral with itself. carried on the disc B which rotates with The lower is disc is C can be adjusted vertically. Occa sionally a series of washers. due to the increased velocity. passes from G upward through the guide bearing escaping 105. is usually. the screw E. 92 shows a recent form of the stepbearing used on the Curtiss steam turbine. and the difference between and at its outer the rubbing velocities of the ring at the shaft diameter results in unequal wear. a single collar is sometimes used. load is Oil forced between the discs through the central pipe Enforcing the discs apart and escaping into the cavity G.CONSTRAINING SURFACES 267 portant. the shaft be placed is at some distance from the end shown in Fig. Collar Thrust Bearings. which bear against the resisting surfaces as 93. or collar. In cheap work. it. 93. which illustrates a thrust bearing as used for marine work. Holes Fig. Fig. When the thrust bearing must of the shaft. 88.Oil lubrication maintained. not more than one and one . as in Fig. or where the load the collar is small. is The objection to the single collar bearing for to heavy loads that the large diameter necessary obtain a practical bearing pressure increases the work of friction. are interposed between and the bearing ring. The is thus completely fluidborne and perfect . which supports the heavy rotating parts of both is turbine and generator. The oil atH. therefore. the lubricant should be supplied at the centre of the step bearing under a pressure such that the metallic surfaces are forced apart and the load is fluidborne.
^qpP 12 rx N . so that the thrust is The most modern practice in marine work make the bearing rings horseshoeshaped. In work each horseshoe has its own independent water circulation." page 180.P 3 ^—h) . (2) where N is the number of revolutions per minute. the bearing surfaces of the thrust block are sometimes made integral with the bearing proper.. of the bearing or Occasionally the horseshoe collars are adjustable along more easily brought to a proper bearing. and These rings must be scraped its until each collar on the shaft bears properly against uniformly distributed. 91. 3 49MPiV(jjji=^) * Church's " Mechanics. is to mating ring. • • • (4) .. width of face of the collar even in large shafts to a few inches. so that each ring can be with drawn without disturbing any other portion shaft. and inside radius then Y **\ M = (Y ^ /j. as in Fig. then the frictional moment resisting rotation is * 1 M = hPy 3 If r x (1) be in inches and is P be in pounds then the energy lost per minute in foot pounds E = fiP 3 2 ri X 271 N = . MACHINE DESIGN which limits the half times the diameter of the shaft. so that local heating may be prevented.268 . and the necessary area is obtained by using a number of rings. flat If p. Thus the main casting of the block may be of cast iron and the bearing rings of brass are inserted held in place by radial grooves cut in the block. Friction total load and Efficiency of Thrust Bearings. This oil bath also has a water circulation for cooling the 106. P be the on a circular pivot of radius r t and be the co efficient of friction. 3 (3) 3 andfi. and the lower the shaft so as to be firstclass part of the bearing constitutes an dip.. oil bath into which the collars oil. In small or cheap work. but usually they are made detachable. . if In a similar manner side radius r lf the thrust be taken on a collar of outr 2.
M.0102 .0083 .0045 .M. 128 R.P.0080 . where In the case of thrust collars.M. as the case moment or the moment or the energy may be. 107.P. per Unit Area. such in as shown in Fig. the frictional resistance of the step has do with the power transmitted. may be carried to an flat extreme. but at flooded.P. the surfaces themsimilar to that of the selves tend to draw in lubricant in a way ordinary journal.0062 . the decrease of centrifugal (the bearing being oiled from the This would seem rubbing velocpivot indicate that devices for reducing relative such as multiple washers (Art. 50 R.0196 .. 194 R.0052 . a little to Thus in the case of a shaft carrying heavy load of gears.0093 . the frictional energy loss must be added to the turning supplied. In the case of the thrust bearing of a steamship the frictional moment and energy force P. however.M.P. 91.0054 At 50 and 128 R.0107 . is Where the velocity of rubbing very low and wear not important. very heavy unit loads may be put .0167 . The following coefficients of friction are taken from Tower's experiments TABLE XVI Minute as given below. 290 R.0061 . 20 40 80 120 140 .M.0063. as in the case of swinging cranes.0054 . Pressures in lbs.M.0178 .P..P.0147 . The coefficients of friction for this class of thrust should therefore be as low at least as those given above.0048 .0096 . .0181 . Bearing Pressures is on Thrust Bearings. causing more friction than a plain centrifugal action is effective. loss are directly proportional to the driving In either case. running an oil bath.0046 . 102).0063 •°°53 . the the other velocites the bearing coefficient oil supply was restricted. which force was probably due to ity. R.M. 353 R.0064 .P.0221 .: CONSTRAINING SURFACES 269 The vertical efficiency of a thrust bearing cannot always be expressed as a function of the power transmitted. was In all cases the increased at revolutions to below 40 the centre).
H. is and shaft only about twothirds of the indicated power thrust block. X 33. and the indicated horsepower be denoted by H.000 to 3. not bear simultaneously.. per Min. as found in practice: TABLE XVII Mean Velocity in ft. and wear im much lower pressures must be carried with imperfect lubrication.000 . propeller. per hour) 15 Indicated horsepower of one engine Inside diameter of thrust collars 5>o°o 14" Outside diameter of thrust collars Allowable pressure per sq. tained. in an effort to keep the bearing pressure down to a low value. for thrust bearings. The following are average values of bearing pressures.080 ft. of surface 21" 40 120 lbs.P. they rotate in an oil bath. or even moderate. 75 to 50 Thrust Bearing and Bath Lubrication as in Fig. in. then Xl. 2 /. 92. Character of Lubrication. if P be the 5 the speed of the ship in knots per hour. Revolutions of the shaft per minute Owing to frictional losses in the engine.000 Fig.270 MACHINE DESIGN bearings. Example. and estimate the frictional loss in the thrust (1 knot= 6. there is danger that all of the collars will If. too many collars are used on a collar thrust bearing.P. per Square Inch. than velocity. bearing. as in the stepbearing shown it in Fig. 87 Fig. or thrust. Bearing Pressure in lbs. 88 2. 88 Fig. Up to Bath Bath Bath Bath Bath as as as as as in in in in in Fig. 88 Fig. 88 200 i5° 100 5°. 93. Speed in knots Design the thrust journal for a steamship having the following data. delivered to the The pressure against the thrust block multiplied by the distance through which the ship moves per minute must equal the energy delivered to the block per minute. is high. Very slow as 50 125 200 500 in to to to to hand cranes 50 125 200 500 800 ft. especially velocity is if upon pivot Where the portant. on ordinary bearings running at the same With forced lubrication. is evident that very heavy pressure may be mainon the other hand.
There fore the total allowable pressure on each collar =192 X 40 = and the 7.349 X X 72.680 ' If the number of collars oil = 72. The area of each thrust collar = 4 (21 2 — 14 2 ) = 192 sq.H.080 QJ n _ 2 X I.349 Therefore from (4) E = f* PN . per min. the coefficient of friction will not be more than under the worst ordinary conditions. bearings of this character can be obtained will which run well under very severe conditions. At present. ROLLER AND BALL BEARINGS 108.300.H. of great antiquity.300 X 120 rio.080 X 2 I. The application of this principle to very heavy loads as in the case of and low speeds. is moving heavy bodies en rollers. General Consideration of Rolling. in.000 6.P.P.z J 10. and the intensity of the load sustained.1 CONSTRAINING SURFACES 27 60 __ _ PxSx 60 6. lbs.5 3 7 3  U10. X 217 i X 5 X 5 1 *7 Hence in the above example r OOO X P = — = 72. the surfaces of both light. .P. for a given load.000 ft. the amount of deformation depending on the character and hardness of the materials a curved surface rolls When forming the surfaces. X ^ 3 33. upon any other surface with which it theoretically makes line or point contact. and the load is very deformation negligible and true rolling can be practi . y D or say bearing runs in an . in Article 29 that the resistance less It due to rolling friction was pointed out was much than that due to sliding friction. however. but only in recent years have mechanics been able to produce surfaces of such a character as could carry even very light loads at high speeds on either roller or ball bearings. the two surfaces tend mutually to deform each other.5 a —7 I 1 = 405.01 [ %" r \j = .3 H. the If members is are very hard.680 o. or 12.300 —— = 7 .01 bath.
it theories. which is very complex. and of uniform hardness. S. appears from experiment and experience that carry fairly bearings of this character can be constructed to heavy loads at high speeds for a long period of time provided is the intensity of pressure not too great. which elastic properties of materials. borne on a small a combination is This occurs even when the surfaces are very hard.. XXVIII. rollers or balls. is misleading and useless as far as the design of such bearings It concerned. touch each other. . of rolling and sliding. is Undoubtedly the work of most value in the design of roller or ball bearings.* The It true theory of this action. Vol. In the case of roller or ball bearings this may result in the destruc tion of the surfaces either by flaking off locally. It is obvious from the foregoing that the materials used in such bearings must be homogeneous. on which much of this research work has been done. permanent deformation will occur. any appreciable load and the load is is to be carried the mutual deformation of the surfaces destroys the theoretical line or point contact surface. modern ball and roller bearing has been The made success of the possible by im proved materials and workmanship rather than by new Referring to Fig. the directions of motion of the common points of contact are in opposite directions. is evident that when two adjacent A and B. has not been fully demonstrated and beyond the it scope of this treatise. or by simply this crushing out of shape. can readily be seen that is closely connected with the character. by rolling a The student may demonstrate this action round lead pencil on a piece of soft rubber under pressure. Henry Hess. and is the action instead of being that of pure rolling. character is In either case continued action of destructive to the bearing. MACHINE DESIGN When. however. is that of Professor Stribeck whose masterly report has to this been translated into English by Mr. M. It is often stated that this results in * considerable frictional loss.272 cally attained. 94. elastic limit of If the intensity of pressure be such that the the materials is exceeded. and translation reference will be made hereafter. f See Transactions A. E. Experiments on either balls or rollers to determine the ultimate crushing load are. therefore.
94. normal to the and these have no component tending to force the ad jacent rollers or balls together. shown A shell of hardened it. friction of roller this is a very and ball bearings while at rest very and important point in the design of heavy. A brief reflection will show that very little pressure can possibly exist between A and B. in wellbuilt roller bearings. slowmoving machinery where. do not wear appreciably.CONSTRAINING SURFACES and sometimes small intermediate balls. to obviate the supposed Such mediate balls or rollers must be kept in place by a cage such as £. The only pressures that can be exerted by the guiding surfaces balls or rollers are in a radial direction or upon the surfaces. The is fric from this source is The small. or rollers. it often takes a much greater effort to start the machinery full from rest than to maintain motion at speed. A form of construction of in Fig. roller bearings for radial loading steel. according to the manner in which the load typical is is sustained. Roller bearings. which secured to the bearing proper. Both rollers and shells are usually made of high carbon steel hardened and ground. or of mild steel case hardened. and this cage will give rise to a greater frictional loss than that which it is expected to remedy. 95. 94. Fig. . Sometimes the . not borne out by experience. as these cages. 273 used as inter shown at C. undoubtedly very small. Forms of Bearings. 95) and if any appreciable pressure could exist between adjacent rollers or balls the same would necessarily This theory tional loss is exist between them and this guiding cage. E. are classified as radial or thrust bearings. with ordinary sliding bearings. in common with the ordinary bearing. B. and against an outer shell D. and 18 . surrounds against the shaft this shell A and is secured firmly to The rollers C bear is B. ROLLER BEARINGS 109. thus replacing the theoretical line contact with point contact. are loss. rollers or balls are separated by a guiding cage (see Fig. The rollers are held parallel with the axis of the shaft by means of a cage F which is made of brass or other soft material. Some form of cage is necessary in all roller bearings on account of the tendency of the rollers to twist out of line with the shaft. Fig.
it it cannot make line contact with the shaft unless assumes a spiral form. thus making a hollow flexible roller. or the roller is roller shown in Fig. This end thrust is sometimes provided for by putting a If tendency to small ball at each end of the roller to act as a thrust bearing. 96. the axis of the roller is not parallel to that of the shaft. If the rollers are if hardened this may result in fracturing them.274 also causing MACHINE DESIGN an end pressure and cramping on the rollers. steel strip spirally upon a mandrel. It is to be especially noted that neither of these methods For continuous will compensate for inaccurate workmanship. Fig. 94. as shown made flexible as illustrated by the This roller is made by winding . 96. as it roller are accurately made. the roller cannot get out of parallelism with the shaft without being bent into a spiral form. trouble the rollers are sometimes at H. especially they are relatively long. 95. and the clearance is should be. If the surfaces which confine the very small. 97. in Fig. Fig. To obviate this Fig. Hyatt made in short lengths.
A carries a thrust collar B and the thrust is taken on the frame of the ring machine by a corresponding collar C. A hardened steel D is attached to B and rotates with it. theoretically. If the direction of the load to be carried roller thrust bearings of the form shown in Fig. the apex oj the cone lying in its tlie centre line of the shajt. is the velocity of any point in periphery proportional to its distance from will the axis o! the shaft and. and the rollers shells must be guided so as to remain perfectly parallel to to the shaft. . These conditions are service. and almost impossible to maintain with great accuracy under continuous The 97. which to obtain where the is axial. A thrust ring prevents the rollers from moving radially outward. carrying with E is The conical rollers G them the cage F. Fig. difficult obtain initially. cylindrical. true rolling be obtained. in form.CONSTRAINING SURFACES line contact the outer 2 75 and inner great accuracy. placed in must be machined with very accurate alignment. The apex angle of the roller should not exceed 15 and in most cases is kept down to 6° or to prevent serious end pressure against 7 this retaining ring. rollers in bearings for radial loading may be adjustroller is cylindrical or they in Fig. rings. The shaft PLAN OF CAGE Fig. 98. 98 are often used. 99. may be conical as in the Grant bearing shown The construction here is difficult shown permits of ment for wear. while a similar ring fastened to the stationary part C. It is evident that where the roller is conical roll between these H .
N= W= w = d total number of rollers. (1) (2) and W = k I N d— for From Stribeck's* experiments rollers k has a value of 550 for unhardened and bearing surfaces and 1. 444. Allowable Bearing Pressures. S. The outer rollers rotate faster than the inner rollers. M. XXVII. multirollers in plied by onefifth the total number of the bearing. as shown in Fig. for cylindrical bearings. 276 MACHINE DESIGN Bearings of this character with conical rollers are expensive to make in an accurate manner. many forms of bearings on the market. according to Stribeck. bearing surface beyond the little but theoretical considerations are of service in the actual designing of such bearings. E.. no. lbs. is considered and diameter of a single roller. and a simpler form. in lbs. but their fundamental principles and the student is referred to current trade cata logues for variations in methods of construction. hardened surfaces. k = a constant to be determined Then w = k d I I experimentally. is sometimes used. Here the rollers are cylindrical in form and are made in short lengths so as to reduce relative slipping. page . In roller bearings under radial pressure the load is con sidered as carried on onefifth of the total number of rollers. total load on bearing load on one roller in = diameter of roller in inches (mean diameter for conical rollers) = length of roller in inches. if Thus.000 * See Transactions A. 99.. It is evident that the bearing pressure in roller bearings must not be great enough to stress the material of either roller or elastic limit. and the quantity equivalent to the projected area of the ordinary bearing. roller Space does not permit of discussion of the are the same. of the rollers are such that and the lengths and arrangement ridges are not worn in the seat. The most reliable ex perimental data bearing on the subject are the results of Stribeck's work. as far as carrying capacity as the product of length is concerned. Vol.
rolling contact with B. the discs must be so placed that the centre of the ball. 100 (a). BALL BEARINGS in. along the circular path * B. Fig. 99 have been constructed to carry a load of 156. at right angles to B. that the ball can around only one diameter at a time. roll Let the ball A. with pure rolling motion. B be parallel it rolls to the plane CD. and these discs must have a common axis of rotation O k perpendicular to their planes and passing through Further. and make contact with it B along the edges of such a disc as would be cut from by a plane passing through the point of contact b perpendicular to the diameter around which the ball rotates. making point contact with the path. 100 (b). as an axis.CONSTRAINING SURFACES In the case of thrust bearings the load distributed over the total 277 considered as may be number of rollers. Let the path plane. If the ball (b) contact as B and E. and roll along the edges of the disc b however. 100 * The guiding surfaces of ball bearings are almost invariably circular in form. Theoretical Considerations. 100 a) respectively. .000 pounds at 250 revolutions per minute. around Ok as an It is clear. and suppose from this also that the ball as remains a if fixed distance same Then it it is evident that A rolls with pure rolling motion along B. Fig. 100 (b). will rotate around P jL r ^— /A / Q 6 a ci e Nj^\ / fe b > Fig. Bearings of the type shown in Fig. axis. and preserve true has two concentric paths of whose points of contact with the ball are b and e (Fig. some one will of its diameters. Fig. then it must roll along two discs bi and el. Thus rotate the ball may rotate i.
It is not possible to have more than two points of its contact between the ball and one of guiding surfaces. The tional force rotating hence. for loads acting at right angles to the shaft. If the load carried is so small that no distortion of the surfaces takes place. Spinning. Forms of Bearings. for then p e Yl = rb YV may be or the circumferences of the rolling discs are proportional to the circumferences of the paths of contact. (b) Thrust bearings. this frictional force will act tangent to the outer circumference of the disc of contact and be parallel to its plane. passing through the common centre of B and E. as the proportionality given above is not true for any other points on the line b except those given. is and the fric carried on a small area instead of a point. for taking loads both perpendicular and parallel to the axis of the shaft. the friction between the moving member and the ball causing the latter to roll. and true point contact exists. .: 278 the lines i I MACHINE DESIGN and b e intersect on the line m. The above with principles are fundamental circular guiding surfaces. Such load theoretical conditions never exist in practice. as the sur faces of contact are deformed. This action is known as spinning and is necessarily accompanied by friction. with pure rolling. for loads acting parallel to the axis of (c) Angular bearings. and true rolling attained. the shaft. Ball bearings are divided into three types. and apply to all ball bearings 112. even under light loads. It is clear that inaccurate workmanship will give the same result. Usually one of the guiding members is fixed and the other rotates. according to the character of the load and the way it is sustained by the bearing (a) Radial bearings. the ball is. 113. indeterminate and in general has components which tend to rotate the ball about other axes than the one which will give pure rolling motion.
often act as reservoirs for lubricant. each race may be made filling in one solid piece. ioi (b). threepoint or four point bearing. on account of the break in the surface of the race. These separat ors. If both the shaft A and load is hub C rotate this cannot be accomplished. also. Fig. In order to place the balls in the raceway an opening race is often cut in the side of one of the races. Either A or C may be the at E> If the rotating part. and the full brought upon this thus decreasing the capacity of the bearing to sustain a load. filling piece with reference to the load. while the race F secured to the other member C.CONSTRAINING SURFACES Each of by the these types 279 maybe either a twopoint. If the velocity of the rotating is member is high this break in continuity of the race If fill destructive to the bearing. depending on the ball number of points of contact made on the guiding race surfaces. as filling shown and the opening then closed with a piece as shown. in and then Separators of elastic material are then pushed to between the balls maintain correct spacing. is The B is secured to the shaft A. loaded side and no wear brought upon If B is stationary the in locating the opening must be cut in it. Figure 1 01 (a) shows a twopoint radial oearing. Radial Bearings. ioi (a). and the same care used Fig. F is stationary this filling piece can be located on the unit. They may be of . 114. distributing them. of balls necessary completely to of about half the is total number is the race used. filling piece. In such cases the bearing assembled by moving the inner race the balls over eccentrically to the outer race.
D and the ball races. surfaces C and D are sometimes made both flat and It is difficult. of less than one thousandth of an inch will cause concen trated loading of the balls such bearings should be seated on spherical surfaces. so long as the velocity of rotation uniform. In Fig. 101 (b). as shown at D. The parallel. between C. Thrust Bearings. 102 the angles it is this is not necessary as and $' are equal. however. The carrying capacity of radial ball bearings. without interfering with the pair shown which may roll on A. 115. is within reasonable limits. to obtain absolute parallelism. 102 illustrates a fourpoint thrust difficulty in filling in the balls <j> bearing. either from poor workmanship or deflection under on one to side. therefore. Fig. load. . Here there is no when the races are solid. according to is Stribeck's experiments. The lessened number compensated by using balls of larger diameter and hence greater carrying capacity. 102. and much more difficult An error to maintain this parallelism under running conditions. If possible. not affected materially by velocity. initially. This construction of balls is is showninFig.2 8o MACHINE DESIGN or such soft material or felt may be made for in the form of a helical spring. but evident that any line drawn through Fig. O which and intersecting the ball circle will locate a pair of rolling discs will roll on B. in alignment. but sharp variations of velocity at high speed reduce the capacity. thus allowing the races adjust themselves correctly.
A simple form of ball thrust bearing balls is shown in Fig. If possible. all evident that the arrangements shown in Fig. and adjustment cannot for this. radial loads should be supported by radial bearings. axial loads by thrust bearings.500 revolutions per minute as a max imum. (c). and some soft alloy. 104 . inserting the ball and then closing down the upper edge a little with a set as shown at e. three. Henry Hess states that speed is 28l in an important factor such bearings and he gives 1. A and B. Angular Bearings. 104 (a). ball bearings Properly designed if do not wear appreciably. This to last feature. and angular bearings should be avoided.CONSTRAINING SURFACES Mr. and wear does take place it will occur on the loaded side only. Fig. and fourpoint angular bearings. forms of angular bearings. justable. Fig. bearing at the mercy of an unskilled person. (b). are kept in position by a cage C made of 116. compensate It is but only hastens the failure of the bearing. while sometimes necessary and often really a detriment as it claimed be an advantage. 103 (b). Here the run against flat hardened surfaces. For light loads the angular bearing will sustain pressure in either of these directions. The races can be made continuous is in all cases. The cage may be made to retain the ball loosely by drilling the openings for the balls almost through as shown in Fig. 103. There are innumerable and (d) may be and are often adputs the taken as typical of two. Radial bearings should not sustain heavy axial loads and thrust bearings should not be loaded axially. 103 (b).
ball. 104 (b) and 104 it (c) the point a may. (b) Theoretically. and often repeated. a result to be expected. 104 (b) should not be less than about cally.or fourpoint contact so that the load is uniformly dis . in order to prevent wedging not be greater than threequarters the diameter of the the For same reason the angle 2 . 109. Casehardened materials are of doubtful value for severe service. can be demonstrated mathe and is evident on reflection. of the ball races should. For most trying circumstances special steels and alloys will no doubt be much used. that a greater bearing surface will be formed for a given distortion of ball and ball race the more closely the crosssection of the ball race corresponds to the crosssection of the ball. as in the other forms. put upon a ball bearing will depend on the following (a) (b) (c) The character of the materials forming the The shape of the raceways. as c. Allowable Load. 104. Ball bearings balls. but (a) raceways or If the stress practically the unavoidable distortion of the material increases It the point to a small surface. theoretiroll be anywhere. It is almost impossible to machine and adjust ball bearings of three. 117. rect On the other hand. the surfaces will flake off and failure will occur. nearly to equalize the loads at b and The allowable load which may be races. Fig. Experiments on the crushing strength of balls or races are useless and misleading as the life of the bearing depends on the elastic and not the crushing strength. For this reason the radius of the ball. <j> in Fig. Evidently none but hard materials can be used for appreciable loads and these must be homogeneous in texture. The path of the ball is not so definitely determined at a. In Fig. should be so placed. matically.5 : 282 fulfill MACHINE DESIGN the requirements for pure rolling contact as outlined in Art. as long as It lies between the discs which on the outer raceway. found that the approaches the friction increases as the crosssection of the races crosssection of the ball. however. fail balls and by over stressing the material of the induced is far beyond the elastic limit. a ball supports the load on a point. and as a diit is consequence of this increase of surface. The diameter of the balls.
three. for best results. It is 283 borne out by experiment and it is well known that twopoint bearings can carry heavier loads. 2 radius of curvature equals /i With more perfect materials Stribeck states that these values fifty may be increased per cent. . The bearing must be kept free from acid and rust and provision made for excluding dust and grit and for retaining a supply of lubricant. It is 118.or four point contact. statements have been proven experimentally by who found that the carrying capacity of a ball could be expressed by w = kd where 2 . the surfaces must be highly polished and free from scratches. clear that in order to insure an even distribution of load. It is also found that. is Stribeck assumes that the total load total carried on onefifth of the number of balls. the function of the lubricant being largely to prevent rusting. therefore. (1) w = k greatest load on one ball in pounds. = = a constant depending on the material and shape of ball races. (c) The allowable load which a ball can carry varies with the square of the diameter. the workmanship in order to on both balls and races must be very accurate. and maintain this distribution the material must be uniform in quality and hardness throughout. d diameter of ball in inches. These Stribeck.CONSTRAINING SURFACES tributed at the various points of contact. k =1. than any other form for a given diameter of ball.500 twopoint contact and raceways whose d. Practical Considerations. W be the total N the total number of N N = kd? — W = w— If. initially. pounds on one row of balls. (2) 5 5 For hardened k steel races to made of good quality of steel = 450 750 for for flat or conical races. load in and balls.
any two balls in contact as i. XXVII and Vol. if possible. draw a line connecting the centres of G and H. M. Vol. S. to some form by the load ard sizes. 101 (a). Where this cannot be done special provision should be balls carry its made to insure that each of the several rows proportionate load. . Then and r = RsinO e = 180 R . 180 This value must be increased clearance. number and diameter diameter of the circle number of balls a be made to fix the proper Knowing these. full In designing a bearing for a tentative computation must generally of the balls. XXVIII. in stand The minimum diameter of the shaft carried. the exact passing through the centre of the balls can be determined as follows Referring to Fig. that of applying ball bearings are so no attempt can be made here is to illustrate the student this point referred to the following sources of numerous and them. of equalizingdevice which complicates the design.. it As before one row of of has been found better to carry the load on balls. and information on Transactions of A. Trade publications generally. This usually leads is fixed. and draw the radii O G. commercially. E. O H and O as shown. Also let r be the radius of the ball and R the radius of the ball circle.: : 284 MACHINE DESIGN stated. and balls are made. sufficiently to allow for the necessary The methods varied. approximately.
which directly carry a tool for actually doing work. but not subjected to any torsional stress . The (a). (a) Axles. the most important considera When in the spindle there is designed with these latter require ments view. The discussions given in Art. and which as a consequence must have (c) Spindles. loaded transversely and subjected principally to bending. is 105) be an axle which carries the loads Pn P 2 and P 3. or to the predominating stress (see Art. or the shafts of steam (b) . subjected to torsion or combined torsion and bending. SHAFTS. torsion or bending. while lathe and millingmachine Considerations of strength seldom enter into the design of spindles. usually exists as in the case of loaded beams. engines are good examples of spindles illustrate (c). according 26). axles of railway freight cars are good examples of case transmission shafting in factories. in the bearings are. and spindle are applied 119. particular purpose for which they are intended. (b) Shafts. AND SHAFT COUPLINGS The terms axle. In these members is torsional stiffness and accuracy of form tions. or short shafts accurate motion. Axles. shaft. or to combinations of torsion and bending. usually an excess of strength against rupture. usually.CHAPTER XI AXLES. and it will not be considered further here. Shear. 120. somewhat indiscriminately to machine members which are so constrained by journals and bearings as to admit of motion of These rotating members may be subjected to simple rotation. Let A (Fig. 12 apply in this case. also. Rotating members to the may be classified roughly as follows. General.
so far as the strength of the shaft is concerned. shaft. on the safe side. as the slightest deflection of the shaft tends to concentrate the re indeterminate. to be supported by the bearings is N and N. identical with Case XIV of Table I. to make a solution of the general case given above. it is but from the general theory of beams (Art. would be structive. this latter If. and the loads (which is P 1 and P 3 were equal and symmetrically placed the most usual condition.286 MACHINE DESIGN Suppose the axle except that due to negligible bearing friction. usually. The bending. be where the bending The dangerous section of the shaft moment is a maximum. 14). The distribution of the bearing reactions and the assumption is This assumption is the middle of the bearings. however. If the load P 2 were zero. 95. the case space diagram Fig. be treated as a simple beam (Art. the shaft and consequently need not be large to withstand the applied bending moment. 105. may is be neglected in the body of the short. the section of the bearing at XX should be checked for shearing will it is stress. as in car axles). therefore. The axle can. as indicated. usually made that they are concentrated at action at the inner edge of the bearing. 14) known that. principal stress to which the axle is subjected is simple Shear also exists in every section. It will be in however. as explained in Art. which also . and hence necessary to determine this maximum moment.
.l P ! + P l + PJ *t x l x 2 3 x x 2 2 3 j and R =P +P + 2 x 2 P. in the and draw OA. .. A at graphical solution once where the is much more convenient. Join the intersection of oa and ea with the intersection of od and de. assumed scale of the force diagram. draw oa and respectively to OA and and be OB diagram. parallel to oe. the corresponding force 2. etc. . It is to be noted that these forces are drawn consecutively downward. SHAFTS. must equal the sum of the reactions. to the Then in the force diagram DE = R 2. Draw OE locating E. making to AB = P BC = P and CD = P 3.R x The bending moment at any section is the algebraic sum of all Thus the the moments at either side of the plane considered. ab. 105. or to determine the cross section for an assumed stress. pole. In Fig. the perpendicular distance of the reciprocal of the scale of the space diagram. be. and their sum. RJ = P . involves reactions the AND SHAFT COUPLINGS 287 determination of the unknown reactions. From any point on ob. O from AD.AXLES. + PJ + P. The to the vertical ordinates of the space polygon are proportional bending moments at the points considered. any convenient scale. is The numerical value of any bending moment the continued product of the length of the ordinate. in the force OC Take any convenient and OD. as it shows maximum bending moment is located. AD. OB. thus locating the closing string oe.. The may be determined mathematically by taking moments .. and draw x . etc. P 2. parallel to OC. cd. parallel space diagram. — — — = = (I l P (l l and R bending moment at P 2) 2 M 2 x x x 2) this value may be used in equation J of Table VI ( M 2 = — J to determine the stress for a given cross section. reciprocal and the . or vertical forces. and EA = P. thus. ab. since they act in that direction. as O. diagram as shown. denote the forces P. From the intersection of ob draw oc. and in similar manner draw^ od. around R 2 Then.
as well as to belt pull. be uniform and interchangeable the driving point for convenience and economy. as etc. should. and for such cases or where other stresses. it and weight of pulleys which supports. it is important that shafting. such as those due to bending. and 1" = 5. as bending almost always present to the due to the weight of the shaft itself. This is especially true if care is exercised to place the pulleys as close to the bearings as possible. If the shaft is of considerable length. is usually based on the formula for torsional strength. This would be undesirable. Shafts Subjected Principally to Torsion. at effect is difficult to is compute. and from this moment the diameter of the shaft may be computed. are negligible. The fundais 121. and all is given off in small increments at short interto^ vals is along the shaft.288 MACHINE DESIGN Thus if of the scale of the force diagram. the space scale be 1%" 2]/ 2 to 1 ft.000 inch lbs. on the force diagram. where the torsional stress is where the secondary bending in long factory shafting.. mental relations existing in a shaft which subjected to torsion only have been fully discussed in Article 12. the ordinate at some point be 2" long. In nearly all cases. therefore. Shafts subjected to pure torsion is rarely occur in practice.000 = 200. how power is delivered in varying quantities all along the shaft. and stiffness are made and may be of service. or yi size. The design of shafts subjected principally to torsion. and such computations are not only difficult to make but would its indicate that the diameter of the shaft should vary at different parts of length. Thus where the power supplied to the shaft one point.. however. power is applied at one end of the shaft. as far as possible. Article 12 is applicable.000 lbs. and cases. . it and the practice of reducing the diameter of the shaft as is extends from confined to larger shafting (say over 3" in diameter). then the bending moment at the point considered is M= 2 X X 8 X 5. There are many predominant. computations for both strength easily ever. and taken off at the other end.. etc. the is angular distortion of importance. the bending due the pull of the belts small. and it may often occur that a shaft having sufficient torsional strength will If the not have proper torsional stiffness. the pole distance be 2X". hangers.
The following factors of safety are indicated by successful practice: For head shafts " line shafts carrying pulleys . 6. N Substituting this value of T in (1) above. and the service is which the shaft intended.P. were to be considered.P. it is customary to assume a lower stress (or higher factor of safety).000 lbs. .500 respectively. ps might be taken as high per square inch. Article 12. whence . as the failure the shaft. AND SHAFT COUPLINGS 289 modified by practical coefficients which experience has shown will provide for stiffness against torsion and bending. for depending on the material used.024 H. and to provide for the indeterminate bending in line shafts. countershafts. then the horse power transmitted will be H. Referring to equation E.000 2TnN 33. For steel shafting. '^'J?* If • • • • w and P be the equivalent force applied at the periphery of the shaft. N be the number of revolutions of the shaft per minute.— AXLES. the allowable stress for the above factors would be about '9 4. 27CN 63. The larger and more important stress. 15 . where r is the radius of the shaft in inches.000. ing stress alone as 9. so that if T = Pr.10 etc. A where £ is X p s N \ N If shear a constant depending on the stress assigned. and 8. 7 " small short shafts.P.000 X 12 X H. SHAFTS.000 x 12 X 12 or T = 33. In order to secure stiffness.000. for steel shafting. the lower should be the of a working is head shaft or shaft of a prime mover accompanied by great inconvenience and expense. = 2PmN 33.
the theoretical diameter of the shaft at that section * See Kent's "Mechanical Engineer's Pocket Book.290 For head shafts.. d = . In engine head shafts driven by heavy belts. etc. Shafts Subjected to Torsion shafts. . be less than that due to bending." page 869. 9). 25 owing 26). (5) It must. in addition. From that if Article 16 (equations K. the For example. and many others. be borne in mind that a universal rule class of shafting. the twisting moment high. it appears for the bending and twisting moments can be determined any section. 122. the torsional stress is not predominant and may. <Z = 375^^ 33^?Lp (4) For small short shafts. The student is referred on the size of handbooks* for tabulated data transmission shafts for various purposes. however. . countershafts. If. . and Bending. span in the span of shafting where power is applied by a large belt the bending action this particular excessive. eter to and than the remainder of the may be may have to be of larger diamshaft. varies in magnitude. . cannot be laid down for any and cases v/ill always arise which need further consideration than given by the above equations. It is to be especially noted that a shaft carrying a transverse load. MACHINE DESIGN For line shafting carrying pulleys. in fact. . the stress factor of safety. is subjected to a reversed stress as the shaft rotates. A full discussion of the relations which it exist in this case has been given in Article 16 and remains to show the application of this discussion to actual cases of design. to complete or partial reversal of and must be and this accounts for the low stresses allowable with such shafts. which applies a bending moment to the shaft. K x and Fig. (see Arts.
s. that. as indicated. would not soon be safe as a general principle. A shaft typical case of combined twisting and bending 106 Art. at that position. In line shafting. can be found. the shaft wears is estimated at 22. One of the greatest sources of uncertainty the location of the reactions at the bearings. The weight a of this spider. is where the distance between the centres of bearings or five diameters. . the error from this source of such shafting is small. and that with the probable magnetic pull which little. may occur when lbs. as already pointed out. the crank 25. (c). 5. downward lbs.000 This force is a maximum when it about vertical. or less. due to the steam pressure. is When the shaft is of appreciable length (15 or 20 diameters).000 The maximum is is pressure (P) on the crank pin.1 AXLES. (a). to and at first sight the conditions would appear approximate those of a con tinuous beam. is of short duration. without AND SHAFT COUPLINGS 29 Usually the twisting but the bending difficulty.>i<lrr is neglected. and also a bending moment on the shaft hub of the * The reinforcing effect of the . the shaft itself. and sometimes the designer must be content with an approximation. and heavy of generator spider at Y. even under best conditions. it only four is is evident that the assumption in the direction of excessive safety. SHAFTS. the error small. the data taken being those of the example Case Here the shaft is supported by the carries a bearings at the points X and X'. and. particularly with the usual swivel bearings. therefore. Usually. and bending It stresses appear as a result of lack thereof. While such an assumption might be safely made it when the shafting has been put in perfect alignment. when computing bending is moments. the safest appears. but in such cases as the crank shafts of multiplecylinder engines. procedure would be to treat each span as if disconnected at the bearing. in this case. as perfect alignment. moment can be determined moment is often difficult to is determine. exerts a twist ing moment on is the shaft from the crank to the point F* where in a power delivered. the safe assumption is made that these reactions are concentrated at the centre line of the bearing. the engine shown in in Fig.
direction.. MACHINE DESIGN The weight of the moment in a downward generator.292 horizontal direction. 106 (b) shows. and point of application of the various forces and . isometrically. exerts a simple bending and at right angles to that induced by P. the direction Fig. etc.
SHAFTS. f Reduced to onehalf size in cut. W. The scale of the space dia gram for is Y^" to or yV size. as Z. and it. thus fixing the distance of its centre line from the centre line of the crank pin at 21 K" Other data fix the distance between bearings as 7' — 9". diagram. for convenience. * See also Article 120. and the order of procedure (a) (b) (c) will be as follows: to the Find the bending moment due Find the bending moment due steam pressure P.t (c) To combine S the bending moments distance at any it section. the space diagram. here taken as 8. take the intercept the to T. on P. and lay T' off is as S' V on diagram the MN The U proportional combined bending moments and may be used as an ordinate VV in the diagram of combined bending moments DGFE. taking O f on a horizontal of the space later line through A'. is Graphical Analysis here very convenient. and the corresponding space diagram force W. AND SHAFT COUPLINGS maximum how 293 equivalent and it is required to find the shaft. force for the T J I. to the dead load W.— AXLES. and the reactions due to a convenient scale. the Consider. thus making the closing string also diagram horizontal. Draw 1 ft. MNP which is convenient for P. It often occurs that the shaft carries a heavy flywheel at Y. it. reactions. work. H I J. and for force the reactions due to as shown.000 lbs. (c). bending moment on the It was shown in Example for the Article 99. . Draw the force * to it. making the pole distance = A O taken here \ H as 3 ". that the force P. O' B' A' C for force P. per inch. (b) In a similar manner construct the force diagram r f A BC O. reactions due to so that have been rotated into the plane of the paper P is represented as acting vertically. Combine these bending moments to find the maximum all resultant bending (a) moment. a tentative solu tion could be made diameter and length of the main jour nal.
the direction of this force be vertical. It is evident that the resultant force due to the of the wheel and the pull of the belt. In general. iT yX3X V6 X — is = 485. less than 90 with the direction of the force P. In such a case the moments may be combined by the angle <£.294 MACHINE DESIGN may also instead of a generator.. This diagram can be used for for either simple bending or twisting moments. for a given diameter of shaft. and p the 3 . A to to convenient diagram is shown in Fig. The methods outlined above are clearly applicable to any shaft which has not more than two points of support since in such cases the reactions can be readily found. of solid circular crosssection.000 lbs. can be determined. subjected any moment. is the continued product re of the ordinate which . Its lbs. bending and twisting actions. triangle of forces taking into consideration the The numerical value of any moment it. but will make an angle. and the reciprocal of the scale of the force diagram. which somewhat facilitates the numerical work of this computation. <p. twisting moment seen by inspection to be uniform it over the whole length of the shaft which as before.400 inch pounds. and a heavy belt run on the weight wheel. of the cle 16. which occurs at Y = The value is. 25.000 inch and these two moments may be combined. most strained fibres. T = —pd n Therefore. determine the safe diameter main part of the shaft according to the methods of ArtiA graphical method will be given later. Thus the maximum bending moment. both in magnitude and will not direction. fibre stress from zero per sq. 107 for determining the diameter of a shaft. the ciprocal of the scale of the space diagram.represents the pole distance.000 affects. or Its use in combined first. numerical X 18 = to 450. T . connection with probwill lems involving simple twisting moments If be discussed T is the twisting moment. d the diameter intensity of stress in the of the solid circular shaft. and with any intensity of 15. inch.
.
d = 4".000 lbs. d = f In a similar way. The cance of these designations will be explained by further tions. consider scale U A" as representing the moment in 100 inch lbs. 57^.4.4 outer end.000.000 inch lbs. pass along the horizontal through 9 on this scale to the vertical.000.000 lbs.58 — 3. or T ^ of 90.000 v p n > aJ = „ . to the point . X 64P = 12.296 is MACHINE DESIGN Thus. inch." and the signifi other diagonals each bear three separate figures. f/fm .000 per sq. V^o" To use the diagram when T = and when T = 90.000.726". if directly proportional to p. d = 4".. T = 125.. viz.000. moments and "A" of 500 inch lbs.000.000.000) on scale "A.732 — 1. Vio = 3 — 1. represent intensity of stress (to the upper scale.37 900. the as the intensity of stress..: example bears three ". moment (or of 90. p T = = 10.56 10 since d varies as the cube root of T.196 lbs. As the moment varies directly of shaft. per sq." lies The intersection of this horizon at its below the diagonal marked 3. or the shaft should be about 3. pass along the horizontal through the point marked "9" T = 90. inch to each division). = . abscissas lbs. 337 — . The oblique line nearest to the point located in the last figures. d = 3. per sq. through u b" and take the first 12 of scale "B.000." of 1. and the T = inch scale . for relations shaft) a. inch.000 a * 90.000 lbs. any given diameter between corresponding values of be represented by the straight origin O. T and p (for a 4" will line through the point and the In a similar manner straight lines through for other shaft diameters. and p = 12. j d = 16 a y 12. the origin are drawn To inch determine the diameter of shaft for a with a fibre stress of 12." as before. 107 ordinates represent to each division). or 9. "5.37" (or 3H") a little diameter to give a stress of 12. d 3 = 64. and p = 12." (or tal to the vertical line through the point marked " 12" p = 12.700. the point a corresponds to 125.000) on and vertical (b) scale "J3. if If p be taken as 10. If illustra T = TV T = —16 7: of 90. 337"if T = 900.700 (to if In Fig.
be cause the diameter was found to be about . The horizontal through 9 of intersect "A" and the vertical through 12 of scale is as before. with a given diameter and numerically equal fibre stress.000. For finding the diameter appropriate to a combined bending and twisting moment.} in 100. 900.732) 297 by the nearest diagonal or. as given above. and read the middle is on be the nearest diagonal (1.AXLES. Therefore.726 for a of 900. multiply M by 2 to get and with this value of T. as the approximate diameter of the shaft.000 inch lbs. 32 while the expression for a twisting moment is. greater than 130. To determine d for given the equivalent T.000 inch lbs. find the diameter consider scale = T = 9.56". figure borne AND SHAFT COUPLINGS (. "A" as representfigure ing the moment in 1.000 3 inch lbs.57 = 15. and it etc. The expression for the bending moment in of using scale a shaft of solid circular section is M = —pd". If the or. It thus appears that the diagram covers it all moments. 1." The diagram can also be used for simple bending moments. The required diameter about 26". the diameter found to moment 12.7".000). T = M + VM 2 + r . '(= io X 9. the diameter is 10 X 1. values of p and M. withif it out being of such impracticable size as would be were not for the peculiar designation of the oblique lines and the method "A.000. SHAFTS.000 with a moment of 9. proceed as in the former examples. the equivalent twisting moment.726". p = 12.000.58) as the required approximate diameter of the shaft.000. by is interpolation.000 the and scale u at p = consider scale "A" representing moment b . the diagram if is quite as applicable as for smaller moments. For p = 12. T is numerically equal to 2M. moment must be 10 times as great for a moment of io 3 X 900 = 900. If by interpolation.. Thus as T= "5" 7. .000.000.
8 in this "A" (at point 8) then o. consider the moment units as 100.000 lbs.000. and T at 4 "C" with the other point of the dividers. the simple bending (T) are given.14".. equation ing moment is readily determined This equivalent twist4 from the diagram by the use of . A shaft 3 T diameter would be proper for this case.. 3. K scale "C" at the bottom of Fig." Since the moments correspond to units of 10. moment (M) and Suppose Consider scales the simple twisting when moment 40. The diagram or the stress.776 (7.000 inch lbs. it is required to find the twisting to an intensity of point slightly stress of 9. then . of the diagram in connection with equations and y The 3 use K K of Table VI is obvious from the above. pass vertically moment corresponding on a shaft i%" downward from "9" on scale U B" to diagonal a above the marked "1. proceed as in case of a simple twisting moment. Swing the dividers on scale A" .000 moment on Thus. . if a given shaft corresponding to any intensity of a 7^" shaft is subjected to moment inch lbs.49". and then vertically up ward If to scale "2?. per sq. of Fig. pass to a point slightly horizontally from 10 on scale "A" below the diagonal marked . The point between 3. Example: M M on 2 = 30." where the stress is read as about 11.+^ e. as a centre until the other point reaches scale 1. on scale "A" the largest figures of the diagonals are to be read in determining the diameter. "A" and at 3 scale C" measure with moments in 10. Take on scale "i" one point of the dividers. per sq. 1 6. e T found way. inch diameter. of 1.0" and as 3.15".76" diameter).298 is to .2'''. inch. The intersection of the horizontal through 8 (T e ) and the vertical through 13 (p) is at point "c. then the distance between 3 on scale "A" and 4 on scale "C" about the point o at 3 represents \/ U M + T 2 .8 on scale "A " = = M + VM + T = T With the value of 2 2 . machine design be determined see Art.000. 107 is equally convenient for finding the intensity of stress in a given shaft under a known moment. U T = to and p — 13.000.000.000 lbs. 107 and a pair of dividers. "c" therefore indicates a diameter of by interpolation the diameter is taken the diameter is By computation found to be 3.000 inch lbs.000 inch lbs.
.000 — 5.49 is the middle lbs. 123. centrifugal force. . acting on the eccentric mass of the shaft.900 inch lbs. the required turning is amount power transmitted. applied to a shaft. If a slightly deflected shaft is rotated." page 869.. For the average range of velocities found in practice the following formulae* can be used for ordinary small shafting. L = where L If 5 v 2 for shaft carrying pulleys . Torsional Stiffness and Deflection of Shafting. T be the twisting moment in foot lbs. moment. however.000 inch therefore T = 5. plane. then the power transmitted at N revolutions per minute \s2TN. as does the construction of the framework to which the bearings are fastened. number on the diagonal.9 X 1. the matter of torsional stiffness is important. to insure proper contact at the bearings. At low speeds the action will and the deflecting force hold the shaft deflected in its As the effect of centrifugal force increases with the velocity. (2) = distance between hangers in feet and d = diameter of shaft in inches. the moment units are 1. do not enter so largely into the spacing of bearings of line shafting. A rule. common in practice. that provision is made for fastening the hangers close enough together to avoid excessive deflection. SHAFTS. it is clear that as the speed is increased the centri * See also Kent's " Mechanical Engineer's Pocket Book. Theoretical considerations. Care should be exercised in laying out such structures. for a given from which the smaller of it appears that the greater the velocity of the shaft.9 on scale AND SHAFT COUPLINGS 299 "A" As 1. L = 7 \/ d d 2 for shaft without pulleys (1) . one plane and to whirl the shaft whole around the axis of rotation. The lateral deflection of the shaft should not exceed yJV per foot of length. horizontally to 5. tends to equalize the forces in which hold the shaft deflected as a of centrifugal force is small. When a shaft has considerable length. is to limit the twist in the shaft to one degree for every 20 diameters in length. to 0. while the effect of the deflecting force is constant.075 degree for every foot Another rule limits the twisting in length.AXLES.
as the cutting relieves. d = diameter of shaft in inches. the span." page 549. at flecting force. balance the effect of the deand the shaft will become unstable. t See Article 12. If keyways are cut the shaft must.300 fugal force will. The speed of shafting in practice critical speed. considerably below the Shafting 124. If L = distance between bearings in feet.. it Unless the interior is "fed" during will be dense than the outer portions and shrinkage cavities are apt to * See Rankine's "Milhvork. etc. less less in the exterior walls than it is in the hotter interior this period. locally. . This equation the refers to the bare shaft only (3) and it determines maximum safe span. is much mass. the skin tension due to the cold rolling f thus causing the shaft to warp. diameter of shaft there is one definite speed within which it will maintain a stable condition with a given deflection. Where pulleys are carried at some of the distance from the bearings. in this country.. be carefully straight ened afterward. Beyond this For a given speed the shaft will whirl about the central axis. Hollow Shafting... up to 3" in diameter is. must be less than the value given by equation pulleys. during complete cooling. it is made) very greatly stress. MACHINE DESIGN some speed. almost always. The subsequent contraction. (3) on account of the added mass and the great liability of the latter to is. L. Practical Considerations. increases its reliability under repeated application of Shortly after a steel ingot the exterior solidifies is and still becomes comparatively cool while the internal portion fluid. in general. made of coldrolled steel. but the removal of the metal from the core of a steel shaft (or of the ingot from which is cast. then for the speed * ^75^1 . Such shafting is true and straight and needs no turning whatever. Larger sizes of shafting are forged and machined. The use of hollow shafts not only reduces the weight for a given strength... critical and N = the revolutions per minute. be unbalanced.
is also involves much all expense. the defect not removed. This flaw is more or less irregular or ragged. Numerous expedients among which is "fluid overcome compression. hence fracture. The work and forging may squeeze the it sides of the cavity together so that is is not easily detected at any section. favorable to starting finally a under variations of to which may effectually extend enough cause rupture. 'jreater. far form stress. but as this a temperature is done at much below its that corresponding to is welding. to Gas The result of these two actions cut off is form what is called the "pipe. This leaves a shrinkage liberated during cooling cavity at the upper end of the ingot. the pipe is is removed. as the walls of large ingots become too rigid to yield to the pressure before the interior is entirely solidified." after which the internal shrinkage is made up by metal flowing from the upper portion toward the bottom as long as any of it remains fluid. SHAFTS. and the shrinkage shaft effect takes place in the latter. In fact. The external walls "freeze. is a made from a hollow ingot worse than the solid shaft. AND SHAFT COUPLINGS 301 be present near the centre of the ingot. and It will the metal remaining superior to that of a solid shaft. The difficulty is not entirely by such means. is if be evident that casting a hollow ingot boring out one which was cast solid." or subjecting the ingot to heavy pressure immediately after it is poured. for not the equivalent of the ingot is cast hollow the outer and inner walls cool before the intermediate mass does. ." which frequently extends is to a considerable depth. in the respect that the former has the defective material nearer the outer fibres where the stress i. If the portion cut off sufficient to remove of the pipe. have been adopted to reduce this evil. The top end of the ingot all and and not remelted. collects in this cavity also. If the ingot is bored out. a piece rolled or forged is from the ingot contains a flaw near the centre which into a long crack rolling if drawn out the ingot is worked into a long piece.AXLES. however. but this does not insure removal of it of the pipe.
and discussion of * See Transactions A. service clutches For this this making use is of friction are much used. the coupling of a different construction and generally known as a clutch. . 108 illustrates a type of coupling known as a splitmuff coupling.3° 2 MACHINE DESIGN COUPLINGS AND CLUTCHES 125. Permanent Couplings. Couplings are machine members of which fasten together the ends of is two shafts. but they are much used for engaging and disengaging pulleys at will. to bring Couplings should be placed near a bearing. 108. the Such couplings are desired to disengage couplings are generally constructed so that they must be partially or wholly dismantled to separate the shafts. to be broken only at rare intervals. Fig. how Fig. so as the joint in the shaft near a supported point. M. known is as permanent couplings.. 109. for a full description various forms of clutches. in Figures 108. Fig. E. and in are used. and particular type discussed in Chapter XIII. and should be placed on the side of the bearing farthest is away from is the point where power applied. is When The it is the shafts at will. as in Where the connection making of repairs. couplings such as are shown no. Where the axes of the two shafts be connected are parallel and coincident. so that rotary motion one causes rotary motion of the other. S. ever. General Description. 1908. to supported near the end. is so that when the shaft disconnected the running part 126. in connection with the shafts on which they are placed.* use of clutches is not. 109. confined to securing together the ends of shafting.
iio. Let D= = n = r = p = d % diameter of the shaft in inches diameter of the bolt in inches the number of bolts radius of bolt circle in inches allowable shearing stress per square inch. Couplings such as shown in Figures 10S and no are regularly Fig. and the student is referred to the trade catalogues of manufacturers for dimensions and capa such couplings. but the bolts should be designed so that their combined resistance and the to a torsional moment. Fig. 109. thereFor heavy work a bolts C. for steel. The general proportions are usually designed empirically. AND SHAFT COUPLINGS 0^0 The fore. drawn are they tighter the hence wedges are split as shown at D. key as shown provided. SHAFTS. Here the circular The tapered wedges B. hi. . are drawn inward by the bolts C. bolts should be accurately so as to distribute the load evenly among them. Fig. B. manufactured cities of in standard sizes. Sellers Fig. inward the more firmly they clasp the shaft. will be at least as great as the torsional strength of the shaft itself. Coupling.AXLES. no shows the Muff Coupling. but in lighter shafting friction alone may suffice to prevent relative rotation. parts A and B is are separated be clamped to the shaft by the by a small space and can. For light work no key is necessary. effective is The Flange and also one of the most common one of the most forms of permanent couplings. but for the full capacity of the shaft keys are advisable. around the fitted axis of the shaft.
may be be modified for convenience in spacing. and reamed after the flanges are placed together. The disc has a tongue running diametrically across each face. 112.3°4 MACHINE DESIGN Then  nD — 16 3 — p = tz <P n 4 r whence d w  D 3 (i) Good practice gives 3 + D — . these tongues . in all This should be done marine good work. thus insuring a perfect fit for the bolts. the flange sometimes forged solid with the shaft. thus protecting workmen from becoming entangled. and reliability are desired. and also facilitating their withdrawal. ii2. and an intermediate disc C. as in work. thus Fig. its full share of the it The projecting outer flange is an important feature as covers the revolving bolt heads. Here the bolt holes are sometimes bored tapering. each keyed respective shaft. great strength is When Fig. to It consists of its own often fast two heavy flanges (A and B). or the axes of the two shafts are parallel. but this number etc. For best results the flanges should be pressed on to the shaft and the faces trued up in place. The bolts should carefully fitted to insure that each one carries load. insuring greater accuracy of alignment. Fig. as in in. used. Oldham's Coupling. but not coin when there is danger of parallel and coincident axes is wearing out of coincidence. When cident.
113 Fig. the angular velocity the same. and hence If another shaft C be coupled to B so that A and C make the same angle with B. or. intersect and make an by means called. rough work the simpler design is adopted. G equal to Positive clutches are much 198. then the angular velocity of C will be identical with that of A and vice versa. The is construction required to usually in make the axes of the pins intersect 113. If the axes of the two shafts. is with each other they may be coupled together it is Hook's Coupling or Universal Joint. 113. SHAFTS. Positive Clutches. is very common. With this coupling the rate of rotation of the driven shaft is identical with that of the driver. used for * Sec " Kinematics of Machinery. page . Empirical practice makes the diameter of the pin onehalf the diameter of the shaft. D which pinconnected to an intermediate intermediate member The holes in this member for receiving the pins this G are at right angles to each other. in other is words. Barr. Fig. With arrangement the angular velo city of the driven shaft is not the same at all points of the revolu tion as that of the driver. AND SHAFT COUPLINGS 305 being placed at right angles to each other and fitting into grooves cut in the flanges. angle of a A and B. as is fitted often In this coupling each shaft with a jaw F. but is the difference between the angular velocity of the driven shaft is less of the driver struction and that when the con such that the axes of the pins G intersect. if also the pins G." J. G in B are parallel to each other and all three shafts lie in the same plane. The coupling is often used on the propeller shafts of small power boats.AXLES.* The construction shown in Fig. more complicated than that shown in Fig. H. 113. 128.
114. for light work. fitting the groove G loosely in a radial direction. shafts. but two feathers are. In order to facilitate the engaging of the jaws they are often made as in Fig. in general. Fig. while part A. faces and the area of the jaw is must be sufficient to prevent crushing. 114 illustrates which reference the most common form of disengaging coupling for fast to the shaft to heavy work. The part is B is made be driven. is connected by the pins P to an operating lever Fig. 115. must work work. These members are much steam engine. to has already been made. Vol. 129. The total crosssectional area of the jaws must be such that they will not shear off under the load.306 starting MACHINE DESIGN and stopping such machines as punch presses which intermittently. or bearing pressure. can be moved axially along the driving shaft. from lack of accurate alignment of the two . 116. A will drive B positively in either direction. E. the scope of this is A very full description of many forms Transactions of the A. but in this case the driving can be in one direction only. A ring R. better. In the con used for connecting rapidly revolving machines to prime movers. XXX. Fig.. They are made in so many forms given in the that a description of them would be beyond S. only one feather used. flexibility in Where it is desirable to have a small amount of is a shaft. which com pelled to rotate by the feather F. is employed. as in the case of a dynamo directly coupled to a the object being to prevent undue stress. such as shown in Fig. a flexible coupling. 115. M. both on account of the driving effort and for ease of operation. Frequently. which is not shown. Flexible Couplings. When the part A is moved forward till the jaws / engage.
struction shown.4 AND SHAFT COUPLINGS and 3° 7 B arc fitted with heavy flanges. . P.AXLES. which carry pins. or for a small lack of coincidence in the axes. The pins in one disc are sometimes placed on a smaller diameter than those on the other. elastic material F. the shafts . so that in case of failure of the links the pins will not strike and cause breakage. SHAFTS. there being as many links as there are pins in each flange. This arrangement Fig. Links of leather or other connect pins on one flange with pins on the other. 116. allows for a slight angle between the axes of the shafts.
flexible elastic connectors are is much used. round lately belts of leather are much Chain drives. while for longer distances steel ropes have certain advantages. heavy and quadruple belts. which are virtually flexible connectors running on toothed wheels. or four thicknesses are needed they are known respectively as double. and completely covered by a rubber composition. or riveting together strips of leather cut from oaktanned oxhides. positive velocity ratio They are very efficient. Where only one thickness is used they are known as single leather belts. in such a when such two must be constant. or ropes of cotton most common. flat belts. XII AND CHAIN TRANSMISSION When power is 130. ratio manner that the velocity ratio of the some form of toothed gearing is usually it is When. three. to obtain a where two. with a special composition between each this country. sewed together. ROPE. used. are less). have come into extended use for transmitting power over commaintain paratively short distances. sewing. shafts. or are built up of several layers of canvas. CHAPTER BELT. fold. belt. triple. Rubber belts are They are very little used in made of several layers of canvas. For small amounts of power. They are very effective in wet places. is When the distance through which power to be transmitted comparatively short (50 feet or or manila. not necessary that the velocity remain constant. Cotton belts are made either by weaving in a loom. Belts of raw held together with. General Considerations. Leather belts are made by cementing.. hide are also used to some extent. 308 . employed. between the two is and can be used when the distance between shafts too great for convenient use of gears. especially are not far apart. however. to be transshafts mitted from one shaft to another.
BELT. this frictional resistance increase the tension in the lower part of the belt. and this pressure will induce a frictional resistance opposing relative sliding between the belt and the pulley. ROPE. and which Fig. Let this tension on each side of the belt be called T z . Other kinds of joints reduce the strength of the belt from 60 to 75 per cent. . It is evident that this initial tension will cause the belt to exert a pressure upon the If pulley. whose centre is at Or When no turning moment is applied to the driving pulley A. makes the joint practically and gives a smooth surface which it runs better than any joint. connected by a belt as shown to the pulley B. or by some kind of special fastening of which there are many on the market. except possibly for friction of the bearings. is In Fig. tension with which the belt total initial is and is that due to the initial placed upon the pulleys. now to will a turning B. Theoretical Consideration of Belts 117. 117. to continuous. preferable where it can be applied. as as strong as the rest of the belt. 131. and decrease the tension tensions be called the upper Let It is these new total T ] and T 2 respectively. and belt a resisting moment the pull upon the in due to part. let and Ropes. ful as it this initial loss of efficient strength is not as waste at first appears. evident that the tendency of the belt to slip around the pulley. but inasmuch as the lacing can be replaced and the belt itself has its life prolonged by reduced load. or by making a permanent The latter method is much joint by cementing and riveting. AND CHAIN TRANSMISSION make them 309 The ends either of all belts are joined. moment is applied to A. the tensions in the two parts of the belt are the same. by lacing or sewing. is A represent a pulley whose centre at O.
that the surface speed of the driving pulley should not exceed that of the driven pulley this value to by more than 3%. shrinks in length accordingly. and is unavoidable. there fore. will small the slipping of the belt in practice. = the tension per square inch of belt section at any point on the . and its surface slower than that of the belt which moves over results in of is. The total loss speed due to both slip and creep should not exceed 3%. subjected to what is all belts are known as creep. rotation will take place. as it moves over the pulley. In a similar way the pulley it B its receives a lesser length of belt than velocity surface. on the pulley occur. for the In addition to the slipping action noted above. When the total slip approaches off of is 20%. there is danger of the belt sliding the pulley entirely. It has been found to be better practice to use smaller belts and allow the belt to slip somewhat. between T and is T 2 is necessary to to overcome the frictional resisting resistance compared between the pulley and belt. This creeping of the belt.3io owing is MACHINE DESIGN to the difference in tension on the two parts of the belt. 117 consider a piece of the belt of unit length moving on to the pulley under a tension As this piece of belt. Referring again to Fig. no moment. is delivers. power transmitted. resisted by the frictional resistance between the belt and pulley. Since the pulling power of a belt proportional to the differ ence between T x and T 2. of unit length. owing to its Tv M elasticity. some loss of power. moves to N. To obtain this result would necessitate the use of very large belts. )r ± The difference in tensions when the turning moment resisting moment applied to If the difference tends to rotate the pulley B. Good practice limits about 2%. it is necessary to know the relation which Let t exists between these quantities. and (7\ — T 1 2 becomes equal which to the B. continually receives a greater length of belt than it delivers. the tension to which it is around with the pulley from subjected decreases from 7\ to T 2 and the piece. The is pulley A. and the velocity of the surface of the pulley faster than that of the belt which moves over it. pulley. relatively.
a = = the angle of in degrees. — / 2 ). /= = w = q = v effective pull of belt per square inch of crosssection = (/. the reaction of pulley against one linear inch of belt of the width con sidered. in pounds. in pounds. belt contact the angle of belt contact in radians of = of . following forces: (a) (b) (c) — when slipping impending.BELT. the tension per square inch of belt section on the slack side in pounds. The tensions / and t f dt. friction = = the coefficient of between r belt and pulley. c = the centrifugal force of one cubic inch of belt in pounds ft at the given speed. held in equilibrium.0175 a belt will The c centrifugal force one cubic inch be = 12 wir gr 1 hence the centrifugal force of one linear inch of 1 2 belt having ww square inch of cross section will be Let the crosssectional area of the belt be one square inch and consider an elemental portion of It is its length as is shown in Fig. by the (d) The centrifugal force = c ds The radial reaction of the pulley The frictional force = jt q ds. AND CHAIN TRANSMISSION 311 = = the tension per square inch of belt section on the tight side in pounds. the weight of one cubic inch of belt in pounds. 118. the velocity of the belt in feet per second. against the belt = q ds. /. . p = the in maximum allowable tension per square inch of belt pounds. ROPE. /_. the radius of the pulley in inches.
without appreciable error. Hence from (2) q ds = dO —z d — z) dO .434 /* —% = k 0. * . . (i) Here d d is o so small that sin 2 may be taken as equal to of d do — in radians. rh h J / t dt t —z j L log = ti I <* H. cds = 12 WV gr 2 ds = t 12 WV g a = = (t z for convenience. .dt = p. .312 Resolving all MACHINE DESIGN forces vertically q 2 ds + c ds = do t sin —+ 2 (t + dt) sin do — 2 . (6) . Hence may be written qds but + cds = . (3) From equality of moments around t O fi + dt = t + q ds (4) (3) . a =10 for convenience . (t — z) d d /• u r do J .439 11 'l =10 * =10 0.'.qds Substituting in (4) the value of q ds obtained from d t = p. : = f e (5) and common S 0. tdO 7 (2) c = 12 wv* g r and ds 2 — d r do . and the product t and sin do — may be (i) neglected.0076 ju.
Thus the belt is to run in a horizontal position.. Experiments made by Professor Died er/< richs in the laboratories of Sibley College gave the values of . force tends to The tity. the initial tension of the belt. = [t 1 z]— the C i) . including creep. and the rate of slip. — 132. with the slack side on top. and some following quantities a x and z. the full theoretical value of a may be taken.. ROPE. the and that. not in excess of about good practice. as before stated. . where centrifugal lessen the arc of contact. (8) If (8) be multiplied through by it will express the horse power (h. a maximum 3 per cent is rate of slip._ io k i = I h r where C ° k ~C 1 + ' ' (7) 10 k— and /  [tl z] f^] v =[t z]C ± . ^ is an exceedingly variable quanchanging with the character and the condition of the surfaces coefficient of friction of contact. a. . can be taken from the drawing of the drive in question. In above the must be known or assumed before /* a solution for t or / can be made. the slack side must be on the bottom (an arrangement which if should be avoided position. It has been found by experiment that. Practical / 1) =h.p. allowance should be if made for the conditions of operation. equations (9) Coefficients. AND CHAIN TRANSMISSION tx 313 value of Now t2 / = t x — t2 * . within reasonable coefficient increases with the slip limits. If. This belts running at high speed. t2 = — f and substituting this in (6) and reducing ' . possible) or if the belt is to be run in a vertical the theoretical also applies to some reduction must often be made in value of a to allow for sagging of the belt. however. .BELT.p.) which a belt of one square inch crosssectional area will transmit or.. The angle of contact.
E.03 to 0.04. Table XVIII has been calculated with a value of w = 0.. VII.035 pounds. quantity z is proportional to the weight of the belt per is cubic inch. For pulleys made of wood. .22 o .0 64. For pulleys made of cast iron. per sec. or V= •0 ft per minute. while Table TABLE Values of XVIII for v = ft.035.coo 6.46 0.000 3.9 sq. the belt velocity is when 2.61 X 40 20 . S. 0. Assume = 0.'.30. M.6 220. the crosssection required = —= 1. Allowing for the difference between conditions in the laboratory in practice. fa" thick and 8" wide. qJ =79 55° .2 2555 Example.600 4.200 4.30 Values considerably above these were found for paper pulleys of special construction.P.9 3 2 5 47. capacity. in. ft. which is equivalent to a belt * Transactions A.800 8. w = 80 o35 3° 40 5° 60 70 90 100 no 120 130 140 V z 1. For ordinary leather (which most commonly w may be taken from 0.2 83 4 io 55 130 5 1576 187. o 20 . l equation the horsepower transmitted by a belt is having a crosssectional area of one square inch for these con * = u \t 1 — z\ 1 CV — = 55° r 1 200 — .314 MACHINE DESIGN first shown in the column of the following table. per minute. The used).600 7.200 7. a = 180 and (9) tt = 200 From ditions: h.800 2. For pulleys made of pulp. the value shown in and those found the second column may be used in designing leather belts. XIX is abbreviated from "Transmission of Power by Belting"* by Wilfred Lewis. p.400 lbs. page 579.31 o . fi Design a belt to operate a dynamo of 15 H.400 3. an average value being 0.400 n75 20. // fi ft = = = o 29 . Vol.800 5>4°o 6.
270 2 3° . Hence T = 9 T — P = x 380 — 206 = 174 pounds. experimentally. * See Transactions A. TABLE XIX Values of C *— — (Nagle) Degrees of Contact /" = a 90 100 1 10 120 130 140 ISO 160 170 180 T 5 .624 •673 •715 •785 40 •45 . It was formerly supposed that the sum of T and equal to iT. E. . The total tension (T 2 ) to in the slack side pull.822 507 •578 544 . The ratio of stress to strain in leather and rubber increases with the strain instead of being proportional to it as in ductile metals.270 •319 •381 288 3 64 •307 325 . Wilfred Lewis* has shown.520 •575 544 .467 .652 . page 566. be this value minus the required effective P.386 .400 206. VII.617 .408 .502 536 579 . Vol.624 .610 .9 will total tension (7\) in the tight side of the belt will X 200 = 380 lbs.667 715 •757 . S.408 45 7 432 457 .667 3° 438 •489 .480 •544 •342 •359 376 . which is found by dividing the foot pounds of work velocity of the belt or.520 •567 .428 •5°3 •567 448 •524 •590 . but they do not indicate tension the relation between them and the initial x T 3.684 •739 763 Equations (7) and (8) involve the relations which exist between T x and T 2 for a given set of conditions.649 .805 55 .600 . When is a belt transmits power the tension side till is increased on the tight side and decreased on the slack the difference in tension equal to the required driving force.. M. ROPE.646 •695 25 325 376 .610 640 7i3 737 . 354 . be done by the P 15 X 33P°° = 2. AND CHAIN TRANSMISSION 315 be The 1.BELT.^ calculations.467 494 548 597 35 423 467 .210 . and this relation may still was constant and be used for very rough 2 T Mr.20 295 342 •407 . that this is not true.692 .
and thus the tension on the slack side tends to remain nearly constant. It is found that the sum of the tensions on the two sides. VII. Place a rubber band over the fingers of the two hands stretch it and moderately . may exceed the sum belt) is largely and in by the strength of the belt. and decreases adhesion between the belt and the pulley. of the relation stress and strain noted above. however. E. a given amount. page 569. but does not increase the loads on the shafts which produce pressure at the bearings and flexure of of the initial tensions in vertical belts.. but * See Transactions A. S. as to is left initial Ordinarily. belts should therefore be put on with care. the tight and slack sides of the belt. In addition to the causes discussed. and. because the tension on the slack side (with a proper initial tension in the due to the sag of the belt from its own weight.6 31 MACHINE DESIGN is This accomplished by what virtually amounts to shortening the belt on the tight side. when driving. at a given speed. the increase of tension on the tight side. Because. the amount Sugges sum of the two ten sions tion: is increased* as the effective pull is increased. is greater than the decrease of tension on the slack side due to an equal of lengthening. In the case of a long horizontal belt the increase in the of the tensions is still sum further augmented in driving. as a consequence. due to this amount of shortening. by transferring this amount between to the slack side. the tension on both parts of the belt are increased by the centrifugal action due to the mass of that portion of the belt which is rotating round This latter cause increases the stresses on both the pulley axis. then twist one of the hands in either of force tending to bring the direction and the increase hands together will be apparent. the initial tension it to trained judg would seem that the more advanced practice of splicing the belt under a known initial tension will add to the life of large and important belts. by about 33 per cent horizontal belts the increase may be limited only the shafts. ment. Large tension. Vol. . while the tension on the tight side increases with the power transmitted. M.
In equation (8) when belt 2 = t 1} f= o and the If belt will exert no turning force. Hence for single leather. the centrifugal force relieving all frictional resistance /j between the and pulley. Strength of 317 of Belting. .500 to 6. The thickness of leather belting from T\ ^ inch for single leather.013 v X 0. If equation it (8) be multiplied through by v.035 this will occur when = 400 or when 6 = 400 whence v = 175 it. enters as a factor in belt design. will express the rate at which energy being de livered. per second or 10. lower stresses increase the 134.500 feet per minute. He also found that fourply rubber belting had a tensile strength of from 840 to 930 pounds The ultimate strength of belting seldom per inch of width. as the real strength of the belt is in the joint. be taken as high as 400 pounds. the velocity of is the belt. a = 2 ] 180." by Benjamin. Velocity of Belting. 2 and w = .000 pounds per square Professor Benjamin * gives the strength of cotton belting as about the same as good leather.035. Lower stresses than these are often advocated. The ultimate strength good leather belting will vary inch. a maximum working found to stress of be good practice." a working stress of 400 pounds per square inch varies may be used. which are average conditions.6 t x v — . w = v[t 1 . and from y to ]/ 2 inch for double leather.3. from 3. Where the ends of the belt are laced together. p = 50 p to 75 pounds per inch of width belts = 100 pounds per inch of width for cemented For double leather p may be taken life at twice these values. page 186.6 = 0. inch is and where the belt is cemented to"endless. and undoubtedly of the belt. gether. thus making to it 200 to 300 pounds per sq. belts. the equation becomes jv = — . or fv = v [t .z] C = v\t t x C If now a = .And chain transmission 133.0078 v* * See "Machine Design. for laced belts.
v x = 102 feet per second or 6. should is better..1 \/ .6^ — . When t x = t 400. but not economical to exceed these limits.0234 v 2 = o or v = tx 5. mitted. or 5.0000216 v 2 ] vC A . Common as 305 Rules. x (10) which gives the relation between v and for maximum power.032 and h. a speed of 3.0000216 v h.000 to 4. P. . not exceed 3 per cent. horsepower trans = [ . A speed of a mile per minute may be taken as about the economithis is also cal maximum limit. equation be multiplied by will express the total A . The slip losses of power in belt transto mission consist of the loss due to and creep. negligible although the effect on the of thick belting running on small pulleys considerable is if important. due to belt pull. The 2 first two.. w be taken as 0. It is often necessary to run belts at it is much lower speeds than these. or H. For durability feet combined with minute efficiency. and the frictional losses at the shaft bearings. and per cent The loss due to bending the belt life is. 135. .. 136. though practical limitations fix belt such as speed of shafting and diameter of pulleys often velocities at much lower values. and absolutely nec should be avoided except where belt. Efficiency of Belting. the belt The losses at the bearings may be laced on under great initial tension in order to this condition must be carry the load.1*00 feet per minute. it is essary to use a short A welldesigned belt transmission it should have an efficiency at least as high as 95 per cent. slip and creep. and it so happens that about the limit of safety for ordinary castiron pulley rims. vC . usually.55 — 0. (12) . If in equation (9). Other Equations.120 feet per minute and when = 275 pounds.t 3l8 • MACHINE DESIGN and equating Differentiating the righthand side with respect to v to zero 0. the area of the belt crosssection.000 per may be taken as a fair value. v = 85 feet per second. and may be as high as 97 per cent including bearing losses. that due bend ing the belt over the pulley.P tx pounds the equation reduces 2 ] to = [ 55 — 0. (11) where C = — as before If the it and p = horsepower per square inch of belt area.
care life should be taken that it not carried to the extreme where the of the belt may be shortened by ex cessive bending.* and commonly known by his name. given in Table XIX. and gives many valuable and practical suggestions it in his paper. Mechanical Engineer's Book. in Vol. literature practical rules. Taylor advocates thick narrow belts rather than thin wide He sums up his investigation in 36 "Conclusions. He also presents scientific methods for measuring the tension in belting. AND CHAIN TRANSMISSION is 319 iden Professor Diederichs has pointed out that equation (12) tical with that reported by Mr. or 30 pounds for other types of leather belts and 6. II. 1 inch wide. A satisfactory abstract of is not possible here. S. Taylor. Mr. In general these formulae neglect to belt centrifugal action and are hence applicable only per minute.to 7ply rubber belts. for nine years. Carl Barth presents a more ex tended mathematical treatment of the driving capacity of belts. 1909. Nagle to the A." page 877. of XV of the Transac American Society Mechanical Engineers. 137. Practical One the of the most valuable is contributions to of subject " Notes on Mr." pounds "The number * Vol. E. page 91. M. Many other formulae of a strictly empirical charauthorities acter are given by different and last some of them speeds that a are very convenient.BELT. gives a number of these socalled Considerations. t is of lineal feet of double belting. Taylor kept an accurate record of measurements and observations on belts in use at the Midvale facts Steel Co.f Mr. Values of C have already been In the transactions of the American Society of Mechanical Engineers.500 single feet Thus wide Kent's a common rule is leather will belt one inch 1 traveling " 1. tions of the M. the F. below 2. ROPE. Belting." among which are: "A of 35 double leather belt having an arc of 180 will give an belt effective pull on the face of the pulley per inch of width of for oaktanned and fulled leather.P.'s works. January.000 feet per minute Pocket transmit H." by Mr. While in general this conclusion is justifiable. . belts.
square inch of section.500 or 3. This would. this it is very small. produce less Suddenly applied loads. Taylor says is safe to run double leather belts on pulleys 12 inches in diameter. In laying out belt drives.000 to 4. 200 to 225 pounds per This corresponds to an effective pulling of 30 power effect pounds per inch of width. required to transmit one horsepower is 950 feet for oaktanned and fulled leather belt.e. till it passes 2.100 feet for other types of leather belts." should be especially noted that Mr. and 6. Mr. action to which the belt is is The it constant bending subjected as runs around the pulley is a great source of wear. Taylor advocates a belt tension of about onehalf that ordinarily used. tension is applied to The longer the belt the greater be the total stretch for a given load. of course. total length of the belt or distance The when will between shaft centres also deserves attention. therefore. it which is more easily more soft and pliable. and where the pulley to the thickness of the belt. compared For this may be excessive. i. acts like a spring it." little "The on speed at which belting runs has comparatively its life. however. reason also probably better to run the hair side of is the belt next to the face of the pulley as this side cracked by bending. A belt being elastic.000 feet per minute. it Whether his practice is followed indicates the true aspect of the problem.. and is a step in advance. stress in long belts than in short ones ." "The most economical is average total load for double belting. and 1.to 7ply rubber belts. but will transmit that on the size of belt which horsepower for a given time with minimum wear and loss of time due to breakage or taking up to restore tension. first maximum materially. than the flesh side. are not based on the minimum simply transmit a given horsepower. or not. care should be taken to keep the diameters of pulleys reasonably large. 65 to 73 pounds per inch of width.500 feet per minute. increase the size of belt required to cost of the installation His values." "The It belt speed for maximum economy should be from 4.320 MACHINE DESIGN passing around a pulley per minute.
(see Art. A. VII. as a consequence. AND CHAIN TRANSMISSION is 32 too however. 136. and the distance between shaft centres must not be too great. page 549. page Book. Mr. . F. For transmitting power is comparatively great distances. When the amount is of power be transmitted large. the shafting must be kept in perfect parallel alignment. S. Vol. of For drives moderate length. though fibrous ropes are also used for comparatively long transmissions. II. even when the belt made very thick. fibrous ropes of cotton. A number Engineers. hemp al or manila fibre are chiefly employed. In all longdistance transmission the rope must be supported at intervals by idler pulleys. run badly. Nagle. 40 to 150 feet. as in the case of "quarterturn" drives. page 347. 24). wire rope more common. ROPE. XX. the distance between centres great. XV. the belt is liable to flap and run unevenly on the pulleys. page 204. Abstracts of these as well as other valuable data are given in Kent's " Mechanical Engineer's Pocket Aldrich. especially where a belt would have to be of considerable width and would. General to FIBROUS ROPE DRIVES Considerations." pages 876 to 887. Vol. W. compared to the size of the belt. 138. Taylor. satisfactory large. VII. Lanza. See the following papers in the transactions of the Mr. page 91. In all fibrous rope drives the surfaces of the pulleys or "sheaves" are provided with wedgeshaped grooves to receive the rope and thereby give the rope a better grip on the sheave. Vol. the width of belt required is may be run wide excessive. If. Professor W. To belts successfully. narrow belts a maximum distance of 15 feet good practice. is For small. Mr. Wilfred Lewis. while for heavier belts 25 feet is found satisfactory. Vol. For these reasons rope drives have been found very where the amount of power to be transmitted is and the distance of transmission relatively great. F. Vol.1 BELT. They are also particularly serviceable for connecting shafts which are not parallel. Society by: of important to investigations of belt transmission have been reported the American Society of Mechanical Professor G.
application very limited. but since its costs about six times as much as vegetable fibre rope. quite extensively used. it Rawhide is especially useful damp places. cotton and manila fibre are by far the most common for transis missions of any considerable size. but hemp. . while in England and on the Continent cotton rope It is is also much employed. In this country manila fibre used almost exclusively. obvious that as a twisted rope of any fibrous material bends while passing over the sheave. Leather belts or ropes of square f or wedgeshaped section have also been Fig. Round ropes of leather." by H. In certain is localities in Great Britain. J. Spooner.322 Fig. which is a local product. 119. Materials for Fibrous Ropes. there must be a certain Reproduced by permission from " The Blue Book of Rope Transmission." For a fuller discussion of such ropes see " Machine Design. 139. are used to a limited extent. used to a limited extent. when the amount of is power in to be transmitted is small. MACHINE DESIGN 119* shows a typical rope drive where the is line shafting of each floor of a mill driven by its own rope drive from the main shaft of the engine. or rawhide.
Its disadvantages are: loss of Greater first and possibly a greater power due any case to pulling the ungreased rope out of the groove is —in this usually small with speeds over 2. same as Let 1 it/ = the weight of a piece of rope 1 inch in diameter and inch long. ROPE. then. g_ * " Rope Driving. broken open it is found to be with powdered due is to the internal chafing. 131. or some such lubricant. graphite. In these equations the unit mass of belt was taken as one cubic inch. result of this action very noticeable in any old manila rope which has been used without lubrication. of Art. makes the following comparison between cotton and manila rope: " As compared with manila. the advantages of cotton ropes of the same diameter are: Greater flexibility. which were deduced for flat belts hold also for round ropes the proper notation be substituted. such a rope fibre. greater elasticity. on the other hand. Flather. therefore. and the use of smaller pulleys for a Professor Flather* given diameter of rope.BELT. usually laid up dry into rope. (8). page 81. which twisted naturally rough. and (9). Cotton fibre is not as strong as manila. on the outside. With the following exceptions. less internal wear and loss of power due to bending of the fibres. soapstone. Let z' = 1 2 w' v 1 where w' has the value above. cost. while being into with tallow. With ropes in length it is more convenient to take a piece of rope one inch and one inch in diameter. J. rope.'' by J. are smoother and hence give to less internal friction.000 feet per minute. thus This dressing also excludes moisoils ture and retains the natural in the interior fibres. Cotton rise fibres. For this reason manila cated. lesser strength. is usually lubriparaffine. filled When fibre. if The general equations (7). . AND CHAIN TRANSMISSION The is 323 is amount of internal friction. Theoretical Considerations. They from are. the notation used here will be the that used in Art. 131. therefore." 140. a dressing or lubricant being applied to the rising exterior to prevent small fibres starting the rope to fraying.
where q is the In a grooved is pulley the pressure between the pulley and the rope greater than the radial pressure in the ratio of cosec is — to unity. = In equations pulley (8) [t\  Cv ' z'\ 55o (14) and (9) the frictional force between the p.24 It is obvious that if <j> be used instead of /* in Table XIX. Professor is Flather* after reviewing what experimental data there subject. 40 45° 5o° 55° 6o° .12 cosec e — Angle of groove. Then equations and (9) become (13) f = [t\z']C> mdh. the corresponding values of * " C in Table XIX will be Driving.46 . = the tension in a rope of i inch diameter on the Let C = f a new (8) coefficient = C modified on account of wedging effect of groove.40 •35 •3 1 .12 cosec — (p = TABLE XX coefficient of friction = 3°° 35° 0. and the belt for a flat belt is taken as q radial pressure between the pulley and the belt. concludes that 0. ' the new constant Rope page 112. The p.26 . .p. cosec — = 0.12 is on the a fair value and computes the following values of </> = /j.324 MACHINE DESIGN Let t\ tight side. The value of // for rope sheaves has not been determined with any degree of accuracy. where the angle between the sides of the groove. frictional resistance between the rope and sheave is therefore q cosec 6 — If pi cosec — $ be substituted for fi in the quantity C (equations 8 and 9) the result (13) C may f be used as indicated in equations and (14) for rope drives.28 .
One of the most important contributions Mr. = total tension in rope on slack side. AND CHAIN TRANSMISSION = . he assumes a his equation. W. is satisfactory . effective turning force P = = 7\ — T 2 Then 7\ = T + and T = T + 2 K K + P in a groove Mr. = weight of one foot of rope of diameter d. = distance between pulleys in feet. = tension necessary to give the rope adhesion. for ropes are scarce. = total tension in rope on tight side. Hunt (see Transactions A.. with = 45 (15) and a = 180 h.. = the total tension applied to each side of the rope due to centrifugal force. XII). E. C. S. For such sheaves.61 about.6i[f 1 As before stated. hence the pulling capacity creased.p^. = weight of one inch of rope oj oneinch diameter. adhesion when 7\ T = 2 an angle of 45 there is However." . (/> If also a = 180 . = sag of rope in inches. but the it power is loss and wear of rope due to drawing out of the grooves greater. on the coefficient of friction and designing engineers have approached to the subject is that of the problem of rope drives without regard to this coefficient. ROPE. Thus HO = and 45 .. Let d L 11/ W T T T x 2 K = diameter of the rope in inches. is in the wedging action. Vol. The angle 45 has been found to be the most If the angle 6 be less most commonly used. Hunt's article has been changed somewhat to correspond with that used in this text. reliable data z'}~ .. for somewhat different ratio in the development of which he assumes "that the tension on the slack side necessary for giving adhesion is equal to onehalf the force doing useful work on the driving side of the rope. C from Table XIX = than 45 .BELT. M. The notation of Mr. Hunt says that "when a rope runs sides are inclined sufficient whose toward each other j at 2. 325 C.31.
.326 MACHINE DESIGN Or T = . be noted that the values of z' .. applicable to manila and if multiplied by — they may be used for cotton ropes. is the .P. w' = the weight oj one inch of rope oj one inch diameter = . (l8) horsepower transmitted by a rope one inch in is diameter.and T t = T° 2 +K + P = +K 2 + P = ^P+K 2 and P T = T + K = .028 for manila rope and . the values given in Table XVIII are multiplied by . therefore. Hunt's formula therefore may be written M_ip W] j_^w] where h. Hunt's assumptions give by using the value 0.. (15) This differs identical in it form with the theoretical equation and from only by a negligible amount in the value of the coefficient.p... . If. 2 and a rope of one inch diameter is which is iden o with the constant z' in equation (14).022 for cotton rope. . The 180 tical ( I7 ) tension K on each side of the rope for 1 an arc of contact of w' v 2 . w= the weight of one cubic inch oj leather = In equation (18).+ 2 o K by assumption.12 z results very close to those obtained for ^ as recommended by It is to Professor Flather.they are ropes. given in Table XVIII may be used in computing values of for the weight w'. . Mr.P = ^[T K] 1 (16) If equation (16) be multiplied through by — V it will express the total horsepower transmitted or H.035. It would seem therefore that Mr.l^^ . The quantities are the same except In Table XVIII.
Co. . z." by American Mfg.1 BELT. for the given velocity = 64 nearly. transmit 25 H. The same factor of safety would give but 130 as the allowable working tension for cotton ropes. and w' = . giving the total horsepower transmitted by ropes of various sizes for 7\ = 20od 2 . The ultimate strength of manila transmission ropes for cotton may be taken as about j.oood2 and rope as about 4. having an angle of 45 \ z' = From Table XVIII.section required = —— 12. .*.P.028. I h.z']£ = [200  51]^ = = 12. 120* shows curves based on equation (17). From equation (18) the horsepower which a rope is one inch in diameter will deliver under these conditions h. Velocity of Fibrous Ropes. 141. Hunt recommends that the working tension (T t ) be not over 200 d 2 dr . Strength of Fibrous Ropes. Fig. and will be found convenient for making calculations. AND CHAIN TRANSMISSION 327 What diameter of manila rope is necessary to when running 4. be safely taken at a rather higher value.p. The centrifugal z' force produces a tension in a rope of one inch diameter of or in a rope of diameter d the centrifugal force = d 1) 12 w r v1 = T 2 IV . For manila rope Mr. The 12 centrifugal force will w' d? equal the allowable tensile stress when the v2 = 200 d 2 or * From "The Blue Book of Rope Transmission.p. since cotton ropes are somewhat less affected by internal chafing the working tension may. perhaps. . ROPE. .000 feet per minute. 64 X— — 5 A 51. in grooves Take t\ = 200 pounds.600^* where d stress = diameter of rope in less inches. Example. the cross. . 142. = [/'. twice the area of a one inch rope which corresponds to a rope \yi" in diameter. The g allowable stress in the rope is 200 d 2 . The working must be taken very much life than these values or otherwise the of the rope is much shortened.
to for a rope feet of one inch in diameter. ^^ * S*B ape 3 /<i" 5 I / Rope k& ?> i m ^ ] ?eet per MiE ute \\\ \0 ^ ^ X^N cm 00 Fig. effective pull for this allowable working equation (18) be differentiated and the differential be equated to zero as in Art. 120. at when v = 140 becomes zero If which speed the stress. found be about 4. Since the centrifugal force. of the velocity where the work done This is is a maximum.900 per minute.328 MACHINE DESIGN feet per second. 134. and the total working this limiting velocity both vary as the area of the rope . stress. the resultant equation will give the value 55 2" Roge \ 50 \ 50 \ 45 \ 1? i"R ope 45 \ \ \ 40 \ 35 1 i (A 30 a 1 1 / IVfi"B )pe \ \ \ \ L 35 I B OQ g OS \ 25 Ph i 1 / / 1*1 [~R ope \ \ u '""R : \ \ \ I O ft 25 g O 20 Pm 2 15 1 /1 // 1 /' z "/ /' '/ '/> / "Re>pe \\ VA \\ \ \ \ K 20 15 10 1 I/.
or driving pulley. by means of an idler. Systems of RopeDriving. or guiding sheave. T = x 200 (P but the reduction in speed increases the life of the rope. There are two methods of In the Multiple. It has been found. 143.000. or decreased to 3. ROPE. or English placing fibrous ropes on the sheaves. 121). that the most economical speed for ropes is from 4.400 feet per minute with Fig.000 to 5. 121. so that the rope is as to [eaves the last groove of the driven sheave first returned the groove of the driver. it This idler is usually arranged a suitable tension may be put upon the rope . is For a fixed value of T = x 200 <P the first cost of a rope a mini mum at about 4. system. when the velocity vivcr is The first cost is increased 50 per cent reduced to 2. In the Continuous or American system one rope only used.BELT. is and this first cost is greater by 10 per cent the velocity increased to 6. it and it is then spliced. in practice. and running constantly in its own particular groove on each is pulley. each rope forming a closed circuit in exactly the same manner as a flat belt. the wear on the rope excessive. applies to all sizes AND CHAIN TRANSMISSION which is 329 of ropes. the till rope being carried continuously from one pulley to the other all the grooves are filled.000 feet per minute.700 feet per minute. is If speeds greater than this are used.900 if feet as above. several separate ropes run side by side. 120. a conclusion borne out by the curves of Fig. so that through (see Fig.
considerable security against the loss as it is and that it also provides of time due to breakdowns. leads to unequal stress. This form of drive of is. In this system the individual ropes must be respliced occasionally to take up the sag in the rope due to stretching. thereby changing tendency effective The velocity ratio of the two sheaves can. farther into the grooves. on account of the broken rope in one of may cause becoming entangled the rope sheaves and winding up upon it before the machinery can be stopped. however. failure of a rope does occur. A much hence brought on the driving span of rope next to . and slipping and conthan sequent wear are sure to occur. may be run much principal used for vertical and quarterturn drives. for. of course. however. and hence its fits d~wn radius. when the smaller the tendency is the driving sheave to throw more of the load on the larger and new ropes. and. this will tendency to throw more load on the old ropes. When a load is suddenly applied to the continuous system all the spans on the slack side become is slacker except that which runs over the idler and which greater load is kept at a fixed tension. The continuous system due is more flexible in its application the multiple system. The velocity ratio transmitted by a new rope will be different from that transmitted by an old one which has worn smaller. the breaking of a rope in the multiple system great delay. The unequal speed ropes. The objections to the system are the danger of loss of time due to a breakdown. the rope safely at any angle. owing to the limited sag in the ropes to the action of the weighted idler. and. where the transmission is a complicated nature. have but one value. therefore. at a When may be removed and repaired more convenient opportunity. result in a When is the driving sheave is the larger. therefore.330 MACHINE DESIGN it Regarding the merits of the two systems the multiple system is may be said that the simpler. allowing Occasionally. the will be for either the old or the new ropes to carry the whoie load. and the unequal straining of the various spans of the rope particularly with a varying load or inequality of grooves. generally. the broken rope not likely that more than one rope will break at a time. the other ropes to carry the load temporarily.
by making the angle of the groove in the small sheave somewhat sharper than that in the larger.BELT. position. XII. Fig. 122 (b). are the principal points of difference between the The particular conditions of the installation must be considered in making a choice between them. that the form of the grooves. S. the effective diameters are the same for all grooves of the same sheave and the surfaces should be accurately finished and well polished. so that the product of the arc of contact and the cosecant of half the groove angle are equal. T. much experimentation two forms (a) of grooves as As the result shown in Fig. thus making the tendency to slip equal. Mr. 122 (a). the sides of the groove are straight while in 122 (a) the sides are curved. the idler AND CHAIN TRANSMISSION 331 and some time must elapse before this load can be equalized over all the spans. M. The sheaves over which ropes are to run deserve special attention. Vol. In Fig.slip bottom and hence when the rope has been it lies in diameter from wear lower in the groove and will a little more This readily than is when it is new and occupies a higher article. . Sheaves for Fibrous Ropes. page 243. This curving of the sides makes the angle of the groove flatter at the somewhat reduced . and. Care should be taken Fig. 122 (c). as any roughness or unevenness seriously of affects the life of the rope. where the sheaves are of different diameters. Spencer Miller* has pointed out that the general tendency to unequal straining may be some what obviated. of importance in relieving the old rope of a also said to assist the rope to roll in the tendency to pull harder as indicated in the preceding is The curved outline * Transactions A. Fig. E. 144. The above two systems. ROPE. 122 (b) 122 and 122 (b) have become most common.
is Where the span between the pulleys considerable the amount of deflection is sometimes of import ance. The wear internal of fibrous ropes is both internal and external. Since the deflection varies with the distance between the size and speed it of the rope and the difference in elevation of the pulleys. A 2W 4 W . for excessive before resplicing can be perMr. rotates as In the continuous system the rope necessarily first passes round the idler to the The The grooves the stretch angle of the groove. L the span in feet and A the is deflection in feet. should not have a diameter less than forty diameters of the rope. ample allowance must be made on the slack side. son sheaves should be large as possible. pulleys. usually 45 . .332 MACHINE DESIGN it groove. T is the total tension on either the slack or tight side deit is pending on the side for which desired to compute the deflec W the to weight of rope per foot. where uniform tension is maintained by a tension weight. and the deflection formed. however. as new ropes stretch very rapidly. a very desirable feature since the rope. For this rea and. Deflection or Sag. existing is impossible to express the relation between them in a single formula.zij^r^. As the . is multiple system. of idler pulleys for simply supporting the rope when is great are not made vshaped but as shown in Fig. 145. distributes the wear on in the The curved it groove is therefore much used groove. in general. 2 (I9 ) 8 Where tion. In the multiple system. For the simple case of the horizontal drive the side approximate deflection on the driving for the may be determined both continuous and multiple systems and also the deflection of the slack side of the continuous system. as before stated. Where will the 2 . 122 (c). Hunt gives the following equation (transformed). tension on the driving side assumed be equal to 200 d regardless of speed. . the to wear being due largely as chafing of the fibres on each other in bending the rope over the sheaves. computing the deflection in horizontal drives: may become A=. the deflection on the driving side be constant for a given span.
0 57 7:8 10.9 25 32 4.9 35 43 6.6 8. (19) to find the deflection.Soo 5. 3.8 6. Frederick Green * gives the following approximate formula for computing the deflection A = Where that W XD ~sV~ as in equation (19). AND CHAIN TRANSMISSION 333 tension in the rope due to centrifugal action increases as the square total tension T 2 on the slack and hence the deflection on the slack side decreases with the velocity.000 5. the span remaining constant. Distance Sag on Driving Side.0 1 1 33 59 92 3 7 •3° •53 .1 1.4 i7 2 .19 •34 •45 80 I I •39 . TABLE XXI Sag on Slack Side.64 . Feet. ROPE. Co.76 1 2 8 4 .1 : 1 BELT. The introduction of electrical transfield as far as still mission has.1 3 2 4 5 3° 41 54 7 2 9 9 12 9 2. may be computed and substituted in equation of the velocity. . General. there is an increasing side for a fixed value of 7\. (2o) the symbols are the same and from which he has calculated the following table on the assumption T — x 200 (P.4 1.2 85 11 .7 33 9 5 4.6 WIREROPE TRANSMISSION 146. greatly curtailed the transmission * See is power concerned.2. however. 2 3 3 3 7 5 3 7 2 2. 1. The value of T. Feet per Minute.6 2 . Velocity. between Pulleys.1 1 •36 .2 1.000 4.000 4. Ropes made of iron or steel wire have been used to a considerable extent for transmitting power over comparatively great distances. Feet. although wire ropes are much "The Blue Book of Elope Transmission.500 3° 40 5° 60 70 80 90 100 120 140 160 . Mr. All Speeds." by American Mfg.0 2 i7 .69 1 .2 S3 .84 1 .
it is difficult to obtain from a sample of rope in more than 90 per cent of the aggregate strength This is due to the difficulty of getting a perfect grip on the rope so that all the wires will carry their full share of the load. in. in. Materials for Wire Ropes. Wire ropes for power transmission are usually made of iron or soft steel and are laid up with a soft core of hemp in order to give greater flexibility. severe is work especially strong crucible steel known as plough steel For a few special cases.000 to 350. iron. the radius . used for conveying materials such as coal. Plough Steel lbs. In wirerope sheaves.000 to 130.000 lbs.000 130. wrought used.000 190. per sq. 000 to 100. of all the wires. unlike the sheaves for fibrous ropes. of buckets attached at intervals along the rope. 148. or crucible steel. Power Transmission by Wire Rope. Wire ropes are also much used for hoisting work such as elevator and mine work and for carrying static loads masts and suspension bridges. Great care must be taken that the rope does not chafe and. and also because the inner wires of a strand are shorter than the outer wires and are therefore more quickly overloaded. 45. on account of the twisted construction. thus rendering them more liable to fracture On account of this latter action ropes made with a short twist break at a lower percentage of their full strength than those of a longer twist. publications give the tensile strength of various fol The John Swedish Iron lowing values for the kinds of wire. rock. by means rope in The such installations moves at very low velocities and constitutes a different problem from that of power transmission. of as in supporting smokestacks.. per sq. in. Roebling's Sons Co. copper and bronze are employed.000 to 190.000 lbs. per sq. They cannot be run on metallic surfaces and the sheaves must be lined at the bottom with soft rubber or similar yielding material. mutually cut into each other. They also state that a testing machine. the grooves in sheaves used for wire rope are so formed that the sides of the groove do not at the compress the ropes. The wires.334 MACHINE DESIGN etc. also tend to under heavy loads. 147. A. open hearth Wire ropes are usually made For very steel. per sq. lbs. Open Hearth Crucible Steel Steel 50. in.
simply through the friction on the bottom of the groove.P. one hundred rope diameters being often taken as the minimum diameter of the sheave. is seen that at the 1%" diameter transmits only 30 H. transmits only 32 H. By rea manila rope of ferring to Fig. The is general theory and equations developed for fibrous rope It hold also for wire rope. Roebling's Sons Co. proper constants being substituted. is taken from a circular John A. evident from this discussion that wire ropes can safely trans mit a greater amount of power than fibrous ropes of the same diameter. irregular. wood.m. AND CHAIN TRANSMISSION 335 bottom itself of the groove is always greater than that of the rope flat so that wireropes drive.m. of course. or at about twice the velocity in the above instance.. some other comparatively soft material) gives increased friction as well as less wear of the rope. most economical velocity. while the same rope transmits 64 H.p. ROPE. This consists of is six hemp core. which are. a strands around a more flexible rope is used.. diameter on the bending For example : a y%" rope on an eightfoot sheave running 100 r. like belts.. shows the power that may be transmitted by iron ropes of various sizes with sheaves of These values are for a different diameters and rotative speeds. the rope may be bent at a sharp .p.BELT. foot sheave at 80 r. be comes through wear. due to the tension of the rope.P. The sheaves should be as large possible to as minimize the bending effect on the rope. in The lining of the bottoms of the grooves If it the sheaves should be maintained in good repair.. of smaller diameter than those used for corresponding sizes of sevenwire strands. or at the 120 it when running on a tensame linear velocity. For hoisting and for transmission. but does consider the influence of the sheave stress. but each strand made up of 19 wires. each strand The table on the following page. rope made with six strands around a hemp core.P.. for the This table does not make allowances to the change of stress it due change of centrifugal force at various speeds. because of the much higher allowable tensile stress. or The lining of the bottom of the groove (leather. which of the consisting of seven wires. must be much smaller than those given if the sheave diameters in the preceding table.
of Trenton. Roebling's Sons Company. 11 16 84 64 68 80 85 i 1 18 f H T6" ! j Q6 102 112 119 6 6 6 6 7 80 IOO 1 1 1 14 17 ji9 ( 18 i7 \ f 21 21 80 IOO 120 (18 1 \ 1 1 j 120 140 20 23 (18 ( 21 1 9 1 17 18 i7 \ T6 ¥ 3 3 1 11 16 \ 4 1 93 99 116 124 140 149 173 141 80 IOO 120 20 16 20 25 120 16 1 I f 7 20 20 * _9_ 14 14 80 IOO 7 16 30 P {? I* 1 I I* 1 148 176 185 f * Taken from a J.i 33& MACHINE DESIGN TABLE XXII. .fl 8'g Vi s ^ . u 'o V 1 Ph 4)— <up4 3£ 3 3 3 3 £o H 8o IOO 120 I40 23 B C O IS Q i to 1h 5* 3 3£ 3 7 O w 35 8 I40 20 19 19 A f f 1 23 23 23 23 1 1 I § 3* 4 8 8 8 8 9 80 IOO I20 26 32 4* 4 5 19 19 t 39 45 f I 4 80 IOO 120 I40 f 4 4 4 5 23 1 80 IOO 120 140 j I 20 i (10 20 19 23 23 22 22 1 1 _7_ 6 7 9 9 9 1 ( 140 20 19 } I A A A 9 f I 47 48 58 60 69 73 82 f 1 5 80 IOO 120 j "i 20 19 ! 16 9 11 16 8 ! 5 5 T% 10 10 10 10 12 12 12 12 80 IOO 120 140 ii9 1 18 ( f 5 xi 1 22 22 21 TV 16 *3 1S 5 140 ji9 1i8 i9 i 8 \ f. TABLE OF TRANSMISSION OF POWER BY WIRE ROPES * u <u to Ml § Optn fe. O Ph § s << <uP4 . The above table gives the power transmitted by Patent Rubberlined Wheels and Wire Belt Ropes. N. at various speeds. Horse powers given in this table are calculated with a liberal margin for any temporary increase of work.S u S. publication of the John A.
is not equivalent. and quarry work. E.. by experience. to run ning over a correspondingly smaller sheave. little or no help in hoisting and recourse must be had to successful practice on which. if and clumsiness 1 of Thus. chafing are. and It indicates the most economical size of rope for a given load. for every portion of each wire is bent around each sheave once during every circuit while it of the rope. should run over a sheave hoisting it at least 40 inches in diameter but to if might be required run over a block sheave 12 inches or even 8 inches in diameter. The following from a paper presented by Mr. This last action. crane. table. fortunately. In power transmission it is usually possible to install sheaves large enough to prevent the bending action from seriously affecting the in hoisting size life of the rope. S. with a resultant increase in the stress of the wires. but in hoisting tackle. is greatly shortened. Hunt. on account of lower speed. W. M. Theoretical considerations are of installations. . but work this is not always possible. however. gives the results of a long series of observations. The speeds indicated in the table are defined as follows "Slow" — Derrick. Fibrous Ropes for Hoisting. The internal friction life and external of the rope. is not likely that the same portion of the rope will frequently come in contact with any single irregularity in the lining. a manila used for rope of inch diameter. ROPE. has been found. AND CHAIN TRANSMISSION 337 angle in passing over the high spots of the lining. C. there are considerable data. 50 to 100 feet per minute. in such cases. used for power transmission. on account of the the resulting tackle. before the A. very great and the even when working at a lower stress. so far as the life of the rope is concerned. ROPES AND CABLES FOR HOISTING 149. the frequency with which any portion of the is rope passes over the sheaves much less than is ordinarily the case in power transmission. that ropes larger or smaller than those recommended in the table are shorterlived under the load indicated.: BELT.
It will be noticed that these 5.000 13. mate Medium. Slow. as here is usually possible to install sheaves or drums of large diameter. Pounds. a factor as high as 10 or 15 some Table XXIV.000 11. and mine work. Chains for conveying purposes. (a) (b) (c) CHAINS AND CHAIN TRANSMISSION Chains may be conveniently divided into three Chains for raising and supporting loads.800 18. For openhearth steel the strength as given for iron rope times desirable. gives data on standard hoisting ropes. Roebling's Sons Co. TABLE XXIII WORKING LOAD FOR MANILA ROPE A.8oo 200 250 300 380 45° 530 620 400 500 600 750 goo 1. especially made and is of great safety is desired.33% MACHINE DESIGN "Medium" —Wharf and minute. F. or similar work. G. deep mine work or wherever great strength necessary. feet per minute.250 1. in installing such ropes and if in general. it made of crucible or plough Great care should be exercised is well.000 1.250 1. D.200 2. Minimum Diameter of Sheaves Medium.900 2. to obtain the advice of the manufacturers before selecting any rope crucible steel. Rapid.500 1. Inches. B. A factor of for safety of at least 5 should be used in ordinary work. Slow.000 40 45 12 13 8 5° 55 14 15 9 10 11 12 60 65 70 16 17 18 13 14 150. tables are based on a factor of safety of 151. Diameter of Ulti Working Load Rapid. cargo work. Rope.IOO 9. may be increased 25 per cent. classes: . On overhead travelling cranes. taken from a publication of the John A.800 2I. iron or steel cables are used almost it exclusively. elevator. Chains for powertransmission purposes. 1 I* I* 1* I* if if 7. H. 150 to 300 feet per "Rapid" —400 to 800 C. in Pounds. Strength. is For rough service.400 15.. elevators Wire Hoisting Ropes. E.600 3. these ropes are sometimes steel.100 1.
5 c fe DO a ooC 60 t p.2 0. The links themselves should be as small as possible to the pull of to minimize the collapsing action or bending due the adjacent links. ROPE.20 8* 7i 7 If If 415 355 3.76 1 1 2i If Ii Ii I 0.48 § i The strength of a chain link in tension is less than twice the is strength of a bar of the iron from which the chain made on account of the bending action due to the manner in which the load is applied.oo 6. Chains for Hoists. are Chains of character made with ellipticalshaped links and should be manufactured of the best is wrought iron to insure perfect welding where the link joined.2 24. and d = the diameter in inches of the .8 19 .00 .88 0.20 A A 1 i i J f 1 1 58 . 1 10 39 4 4 3 4 2 1 1 0. crane chains.20 Ii If 16. and also on account of the weld.68 0.62 13 5i 4i 4 1 1 0.2 9i 8i 8 7i 6 . 034 .72 2 4* 4 3i 3 tV .00 ii 1 0.8 14.2.30 48 5 415 355 3.30 4.89 0.4 13 6 11 8.50 34 26 19.45 2 .95 » c to P O J) 91J .15 . Such chains are sometimes called shortlink.40 7 .50 97 6 8 5 5 .4 10 .23 u < ii .94 36 . •J S .8 52 3.62 .88 2 . TABLE XXIV STRENGTH OF IRON AND STEEL HOISTING ROPES Swedish Iron.30 .85 22.68 0.8 18.50 3i 2 »i 2 t 1 0. If W = the breaking load in pounds.20 3 40 2 .58 . BELT. Crucible Steel.00 245 2 1 228 190 156 124 96 84 72 62 45 6 10 379 31.0 5.00 2. 10 J 1 10 3 4 2 4 0.9 15.89 .00 1 48 42 36 3i 25 21 17 10 8.30 .00 6.15 .g <U > D il 5 114 95 78 62 <D £3* < s 2f 2* ai 2 If If I* If I* 98 5 8.22 8 8 6 8 5 1.20 .4 6.00 4 .60 12 .24 A A 1 J o39 0.4 12.60 1 1 6i 6 ^ ii 1 6* 5! 54 5 50 42 ii i .60 16 15 13 12 2f 2i 2 2 n95 98 5 8. and also that due to winding the chain circular close. 22 5 7 . v . or upon a drum.20 36 . AND CHAIN TRANSMISSION first 339 In the class are such chains as are this used on cranes and hoisting appliances.40 9 .
The " proof " usually applied is onehalf the ulti Where chains are used for hoisting work. as overwinding brings severe stresses on the parts eter of the wound upon The diamis better. and the lengths are joined together by a link made of special welding steel. page 452. Drums on which crane chains are to wind should be carefully grooved so that alternate links lie flat on the surface of the drum. have caused them to be largely replaced. The uncertainty regardit ing the exact condition of a chain in service. The chain made lengths of from 60 to 90 feet. and this should be done periodically with all such chains. likely to for slings. The state more easily determined by inspecshape in Weldless* steel chain rolled from a bar of special is has lately come into use to some extent. but severe bending stresses are also induced. on such appliances as overhead cranes." by H. Whenever not a direct pull. J. is steel rope. Annealing by heating allows a readjustment of the structure of the iron. as in chain "slings" for handling heavy iron castings. the chain should have great excess of strength.000 d (1) The working value or W= load (W) 2 should not be more than onethird this (2) 18. sufficient capacity to receive the chain in one layer.000 d many cases a lower stress than indicated the load is by (2) should be adopted. then the following empirical may be used for iron crane chains. drum should in no case be less than twenty times the this diameter of the chain used. Chain Drums and Sheaves. particularly chains used mate load. and thirty times * See diameter "Machine Design. They are said to be much stronger than iron chains. 2 W In = 54. . 152. Spooner. and should have the drum. by of the strength of the latter tion. This also affords an opportunity to thoroughly inspect chains which are greased in operation. and the fact that gives no warning of weakness.340 MACHINE DESIGN is bar from which the link equation made. but may break at a load below the normal working load. Chains should be carefully inspected and tested or " proved" before using. they are become badly strained.
the stress p" being . HoistingHooks. pockets around the periphery into which the links links are prevented The the wheel. then XY.BELT. The hooks used for raising heavy They are usually made of Fig. 123 (a). steel or iron some employed to hook can be kept low the use of steel castings may be justified. the most dangerous section. forgings although steel castings are If the stress in the extent. weights deserve special attention. Fig. These wheels are made with fit. pocket chain wheels are often used. is apparently acted upon by a direct stress p' flexural stress p" P due (where to the A is the area of the section) and by a moment Pa. to avoid clumsy proportions. 123 (b). ROPE. 123 (a) be subjected to a vertical load P. 153. from coming out by a guide over a portion of and hence cannot slip on the sheave. of certain forms of chain blocks for raising weights. Anchor chains. AND CHAIN TRANSMISSION 341 Where it is not possible to have the chain wind upon a drum. Let the hook in Fig. but where the load is great and the fibre stress in the hook necessarily high. the hook should be forged from ductile material. and the chains run over such sheaves.
S. " t See On a Theory of the Stresses in Crane and Coupling Hooks. . but located nearer the tension side." by Slocum and Hancock. however. apparently. With Ex perimental Comparisons with Existing Theory. but the larger the hook the less frequently will it be fully loaded and a working stress as high as * See "Strength of Materials. The however. Hooks for small cranes and hoists are much more likely to be applicationf of the is. as the true tensile stress is is therefore greater than that given by equation M. E. That this is true is fail in borne out by the fact men tioned above. The theory of article 19. full full capacity than hooks for raising thus a hook on a fiveton crane may be loaded to its capacity several times every day. The stresses in small hooks must therefore be kept low. while the hook of a twentyton crane would be thus loaded at rare intervals. in a Recent investigations* have shown that in this curved beam loaded manner. almost invariably yield to rupture on the tension side. thus decreasing the tensile stress and increasing the compressive stress. is apparent compressive stress considerably above the tensile although the elastic limit for either stress about the same. and the section is made as shown. as in straight beams. more accurate theory M loaded frequently to their large loads. and equation M (Table VI) may be used to design the section.342 tensile at MACHINE DESIGN X and compressive at F. as computed by the above equation. Dulan & Co. the neutral axis does is not coincide with the gravity axis. and fortunately this can be done without making the hook clumsy. somemay be what complicated and it is believed that equation safely used if due care is taken in assigning the limits of stress. and the above theory defective. the stresses must necessarily be creased to avoid clumsiness. As the size of the in hook increases. even when whose tensile and compressive usually elastic limits are about the same. the gravity axis being located nearer the load. Messrs. or p= made of materials p' + P" = P Pae + A I r this Experience shows that members of kind. therefore applies. Andrews. that hooks tension when designed with an stress." by Professor Karl Pearson and Mr.
" as a result of both Fig. or more.5 a + 1. (See Art.75 D = 1.33 A K = 1. the designer has no that hooks are subjected to much abuse and assurance that they will always be loaded with a true axial load. In the following in tons of 2. Thus a twentyton hook would require a shank . Henry R.25 H I .05 A made from following gives the capacity of the hooks various sizes of bar stock: TABLE XXV Capacity of hook Size of bar in tons i 1 1 \ 1 I* 1 IT 2 3 4 2 5 6 8 IO A in inches n * T 1 1 I* If if 2* 2h 2 3l It is to be noticed that the stresses allowed by Mr. ROPE. The following measure ments are then expressed in inches: D = G= The . for improper arrangement of the sling often throws the load more is toward the point of the hook and the member carry a bending called upon to moment greatly in excess of that for which it is intended. to provide is the nominal capacity of the hook The dimension ample room D is assumed arbitrarily but so as for the slings. 123 (a) and the mathematical and experimental work. shank of only 3. The basis for each size is a commercial size of round iron or dimension A.13A L = 1. allowing for finishing.000 should be borne in mind. however.000 formula A pounds. the dimension tensile stress in the It B may be taken as 3". hooks larger then those covered by Mr.000 pounds per square inch. Thus in a tenton hook the dimension A is 3^4" or.000 pounds per square inch would be in a tenton hook. Towne in his ''Treatise on Cranes.0SA = 1. and these proportions have been much used with uniform success. AND CHAIN TRANSMISSION 343 15. however. following formulae give the most important dimensions of a hook according to this work. 25.BELT.) The most valuable data on crane hooks is that given by Mr. is as safe in a fiftyton hook as 10. Towne's work are to be designed his proportions lead to clumsy dimensions. Towne's proportions are very low. When. which would give a pounds per square inch.
style of This forms. such as grain. In such cases the chains must be run at low speeds. 123 (b) shows a twentyton hook of Norway iron which has practice. 154. they are made in many and the usually selection of the particular form for a given problem made in conference with the manufacturer or taken from trade catalogues giving the desired information. but of late a demand has arisen for chains which may be run at high speeds for the purpose of transmitting power. 4X" in Fig. 124 (a). 124 (b). coal. The chains heretofore 155. necessarily. 124 (c). ranging from 10. Fig. are usually made of Fig. Examination of current practice and measurements taken from a number of large hooks therefore stressed to about 6.000 lbs. Chains for conveying and elevating materials.000 in successful service indicate an allowable tensile stress at X. to 15. On account of the diverse purposes to which they are applied. as they become noisy and unreliable even at moderate velocities.. ashes.000 lbs. at low velocities. Such chains are used when a . malleable iron. etc. for the transmission of power.344 MACHINE DESIGN diameter and a fiftyton shank would be 6}4" in diameter. such as cultural implements. Towne's dimensions. been successfully used in $yi" in diameter is The threaded shank being pounds per square inch but yet is only as large as the shank of a tenton hook as given by Mr. per square inch in tenton hooks. Conveyor Chains. the links hooking together in some manner. This form of agri chain is also extensively used in rough machinery. Fig. as computed by equation M of Table VI. per square inch in fiftyton hooks. is chain is known as link belt. Chains for Power Transmission. discussed move.
125. the pin also bears over the whole thickness of the block roller is necessarily omitted. Fig. 126. socalled silent chains. ROPE. namely. 125 and 126. chain in which the pin A is riveted fast in the outer roller and rotates in the inner links. there is D. but in a very short time the chain stretches . roller Fig. In this form of chain the wear between is the pin and the inner link excessive. In the block chain. 124 (a) illustrates the simplest form of links. Here the bushis B is pressed into the inner links. Roller chains may be used for velocities up about 800 feet per minute. for this reason it is now little used for power transmission. The 124 roller (c). Fig. block chains. The and R lessens the friction against the tooth. it fits When the chain down on the wheel shown in Fig. positive velocity ratio shafts. R rotates on the bushing.BELT. roller chains. and has the same pitch as as the wheel. AND CHAIN TRANSMISSION 345 must be maintained between the connected and where the distance between shafts is so great as to make and tooth gearing inconvenient. new. Fig. The form shown ing in Fig. Of this class there are at present three principal types. it is then known used for very low velocities only. 124 (b) is most common. 125. The defect in the operation of the roller or block chain may is be seen by referring to Figs. and block chains up to about 500 feet per minute. which riveted fast to the outer links. and the pin. bears over the whole length of the bushing. but since the to more friction against the tooth. In this form It is sometimes made is without the roller and with several inside links and as stud chain.
thus limiting the speed at which the chain may be run. on the other hand. 127. the increased pitch causing the rollers to ride higher and higher on the back move refund the The roller A coming down to its seat. pitch of the driven sprocket wheel is and the condition when new As the chain stretches. Here the pitch of the Fig. the driving being done on the last tooth P. It is evident . As a consequence the chain does not run quietly and smoothly and the wear is excessive. thus increasing the pitch of the links. The wheel. so that the driving of the chain being is done by the last tooth such that the incoming roller M just clears the the root as at is back of the first tooth and seats itself close to it at N. chain when new is made a is little less than the pitch of the driving L. MACHINE DESIGN due to wear of the joints. and this action takes place when A and B are carrying full load. 127.346 slightly. the rollers move gradually backward away from the driving faces of the to that of the chain. and clearance allowed between the roller and the tooth. 126. sprocket. a condition like that in Fig. not change the pitch. may wear but The operation of the chain is then of the tooth as they is this does as shown in Fig. This difficulty is sometimes met by the arrangement shown in Fig. tooth. the faces of the teeth. is The 127. 125 and riding commences. made equal that shown in Fig. the pitch sprocket. As the chain stretches. the rollers till move backward toward reached. shown Before fully seated while B is just B can become fully seated A must rise.
contact with hence the chain move ahead till (5) is in full is M. Fig. will creep as it may be is show that the tooth outlines of the struck from the centre of the roller when and that when the chain is stretched a little it wound upon the driven sprocket. This outline is also necessary for the back of the than tooth in order to allow the incoming roller to swing in without striking. Brief reflection will driven sprocket rooted in place. 128. 128 is obvious that if the outline of the tooth M be an arc of a circle struck from the centre of the roller (2). When this construction is used. must be fully rooted against the next tooth. is When I\ roller the roller (4). is But before it roller (3) can take the load. AND CHAIN TRANSMISSION 347 that this construction extends the time preceding the condition shown in Fig. a little less the theoretical velocity on account of this continual slipping back ward. Referring to Fig. a small distance now separates the two. of the roller. Therefore as (3) to whereas (from Fig. this roller will swing from its position i' rolling on the face of the tooth. therefore. the form of it the tooth must be slightly modified. (5) is about to roll up the face of not in contact with will M (wear having begun). 127. 128). The velocity of the chain is. from a point a inside the pitch polygon so as to give a curve tangent to the first and last positions Fig. ROPE. which about to give up. ex The load is greatest defect in this construction the fact that the carried entirely on one tooth and hence the wear is . 126. The tooth outline little is therefore struck (as shown on if). (2) rolls up the curve of the tooth it should allow slowly settle back in place. and this is the (2) usual outline.BELT.
348 cessive. In the Renold chain of this type. its the sprocket. While thus keeping contact along a fixed line they rock on each other as the C and D move relatively to each other. Morse chain and B. is difficulties are overcome in the socalled silent In these chains the inevitable stretching of the links for in a peculiar is compensated manner. the shouldered ends of fitted to the links which are riveted into a washer W. The hardened A fitted respectively to the sets of links. are links this sliding is eliminated by an ingenious form steel parts. Fig. D and C. Fig. 127. polygon to Each link therefore remains in constant contact with own tooth. MACHINE DESIGN This may be so great that the chain creeps forward on the driven wheel so as to cause the incoming roller to strike the tooth S. from the time of Fig. 129. but the relative In the of motion of the pin to the bush is still a sliding motion. The is links seat themselves without sliding action and the operation nearly noiseless. and sliding is . 129. The true theory of the is action of these chains as follows: to take very complex. In a later form half bushings of bronze are so that the pin has a bearing over its full length. the links move relative to each other on a round pin P. the links continually tend up a position farther and farther away from the centre of thus increasing the length of the sides of the pitch suit the elongation of the link. 130. but the general action —as the chain stretches. The above chains. engagement till release takes place. thus holding the chain together. rocker joint shown in Fig.
. and in connection desirable. the makers Morse chain claiming an efficiency of nearly 99 per cent. the speeds of which are limited by the cant.BELT. produce a certain driving force on is and hence the pressure on the bearings much reduced. Such chains are particularly useful too far apart for gearing. of the Fig. this to be especially noted that tension form of transmission requires no definite to on the slack side of the chain the tight side. the parts A and B are in contact on flat shown little at E. hence the chain maybe run at higher speeds than others requiring lubrication. ROPE. places where positive for connecting shafts far is is which are and not It enough for a belt. 130. This construction has the advantage of requiring or no lubrication. eliminated. velocity at which centrifugal action throws off the lubri The Morse chains also work well in dusty places. as in motors driving heavy machine tools. for a given effective pull on the wheel rim. The efficiency of both of these chains is very high. AND CHAIN TRANSMISSION simple tension 349 When transmitting between the surfaces as sprockets.
In Fig. and the wheels are pressed together 35° at the . as in feed mechanisms. or contact exists.CHAPTER XIII APPLICATIONS OF FRICTION Friction Wheels for 156. friction wheels have been found very useful. A may be pressed up against B no with a force P. Power Transmission When and it is required to drive driving is a rotating member intermittently. They is are particularly applicable where the amount of power comparatively small. is and let B be the driven wheel which required to be driven intermittently. of friction wheels For continuous driving the transverse sections B \ S^^S  : m Fig. and this form. The it shaft of A is so mounted that. 133. IM used in must be practice. circular in crosssection. 131 let A be the driving wheel which rotates continuously. an equal and opposite force up in the bearing of B. only. 132. is Figures 131 and 132 show common forms of friction wheels. General Considerations. If can be moved slightly away from B until now the force P is applied to the bearing is set (which should be close to A). by means of a lever attached to the bearing. Fig. the rate of not necessarily positive. but they may also be used for heavy work when properly constructed.
108. A will cause B to A and B will roll together with pure rolling motion. and many combinations illustrates of friction wheels are used. the resisting force at the surface of B. the wheels here having the is form of tion rolling cones. depending on for whether it is pressed against the surface m or the surface n. Fig. will } 35* The ft resistance to slipping at the line of contact be fiP where is the coefficient of friction of the materials if t of which the wheels are made. Since. 133 a friction wheel arranged so that the driver A Fig. or greater than. Fig. as even with very hard materials the wheels (See Art. rotate. If A were infinitely thin. can rotate the driven wheel B in either direction. The at driver A may be moved along the shaft C at will. however. and since the velocity of it is B varies with the radius. the rotative moved across the centre of B to the other side. it roll upon B with pure rolling motion. the direction of rotation of B is reversed. for best results. 135 (a). 132 illustrates the application of friction wheels to shafts which are not parallel to each other. Fig. For reason the thickness of A must. be kept small compared to the radius of B. theoretically. . evident that there must be this some sliding at the line of contact. and Theoretically. uP is equal to. is When A' the angular velocity of velocity of B B is a minimum. Obviously the principle of wide applicaFig. 134 shows a form of friction mechanism much used imparting variable speed to the driven shaft. but practically this cannot be attained. 135 (b).) flatten slightly at the line of contact. When A it moved inward. would. must have an appreciable width of face.APPLICATIONS OF FRICTION line of contact. As A is increases. 134 Fig.
roller type and. does not tend to flat wear out locally. within narrow was practically independent of the pressure of contact. leather. '. per cent. Driven wheels therefore. . The tangential force F. paper. rubber or of the most and driving wheels some composition of various forms of these. E. f Vol. less however. Safe Working Pressures per Inch of Contact. rapidly wear spots on the driven wheel. of wood. exerted by A upon B. to He found these values values up to . The bearings. is most comprehensive investigation of these relations that The made by Professor Goss. common being leather and paper. for the reason that the driven wheel any time to be held stationary by the load. Pressure. Materials for Friction Wheels. He recommends the following pressures. \±.352 MACHINE DESIGN 157. Material. power than the ordinary XXIX. This action. The driven wheels of friction devices should always be made of a harder material than is likely at the driver. range for different combinations from low In these experiments the friction due to the bearings was neglected.* whose experiments cover a variety of materials. probably. were of the absorbed M. while the driving wheel revolves against it under pressure. while severe it on the driver.515. 158. Practical Coefficients. therefore. Straw fibre 150 Leather fibre Tarred fibre 240 240 150 100 to 150 all Leather Woodf Professor Goss found that the coefficient of friction for the wheels tested approached a maximum 2 value when the slip be tween the two wheels was about limits. and. while it does are. Fig. * See Transactions A. which are about onefifth of the ultimate crushing strength of the respective materials. is dependent on the pressure necessary to P and the the coefficient of friction It is. 131. S. The value for wood is not from Professor Goss's paper. almost universally made of iron. know allowable pressure per unit of length along the contact elements and also the value of /jl for the particular materials used.
m. face. = fiwl 12 X nd N  = x 33P°° 0.22 025 025 If 159.15 0. laboratory conditions and found Goss recommends the following approximate values the various combinations. Power Transmitted by Friction Wheels. Coefficient of Friction. . 0. w the allowable load per inch of face.p. V be the velocity of the surface of the friction wheels in feet per minute. of revolutions per minute. 28). is and N the number .power can be transmitted by a run strawfibre friction pulley of 8" diameter and 6" when ning at 500 r. the horsepower H . which power transmitted in foot pounds per minute is and the horsepower HP of if =  nPV (1) / 33. the driven wheel :o be of cast iron? *The 23 coefficient for wood is not from Professor Goss' paper.14 0.. Friction. (2) Example.000 d be the diameter of the driver in inches.31 o 30 .27 0. and a the coefficient is of friction.APPLICATIONS OF FRICTION bearing. Working Values of Coefficient of Materials. then since F = /iP. How many horse.000008 uwldN. P the total normal pressure in pounds. in the case where starting the driven wheel under load (see Art. Straw fibre 0. that allowance coefficient * for In this connection it is to be noted must be made the for a decrease in the value of this when the linear velocity of driver is the driver is great. the is rate at fiP V. P.18 0. F the resulting tangential force. 353 between Professor of // Making due allowance those for the in difference practice. the length of face in inches. and cast iron and aluminum Leather fibre and cast iron Leather fibre and aluminum Tarred fibre and cast iron Tarred fibre and aluminum Leather and cast iron Leather and aluminum Leather and typemetal Wood and metal Straw fibre o 26 .
000 feet. I = 6". 135 surface. let P be the radial force applied to the wedged F the tangential force transmitted. P. . . MACHINE DESIGN = 8". It is to be noted that the number of wedges does not it affect this ratio. the width of the driving wheels should be kept as narrow as possible for best If results.000008 0.354 Here d ' . a short distance from the centre. 134. N X = 500. the may be Where the driver must. 132) the component of the applied force R P presses the wheels together and R = P The velocity of the mean circumference of the driver may be taken as the velocity of transmission. is the velocity of the outer edge of the driving wheel not more than 4 per cent above drive coefficients at greater than that of used. and are then known as wedgeof this The faces of a pair of metal friction wheels are sometimes in Fig. and corroborates the empirical face is rule often used that the friction same width of necessary for a wheel as for a belt. The object construction is to secure a greater resistance to slipping. Barr. In Fig. to transmit a given horsepower at the given speed.26. In the case of bevel wheels (see Fig. formed as shown faced friction wheels. This last item is important.* points slipping or a grinding action occurs and wear must The teeth therefore should not be very long.5 It may be a little noted that the horsepower per inch of width of face is more than unity. = . In face friction driving as in Fig. as it is easily seen that the contact surfaces of the driver and the driven wheel can have the same velocity at one point only. for a surface speed of 1. fx = X 0. as in the above example. and that at all other result. This corresponds closely to the empiri cal rule given for belts in Art. at times.26 X 150 6 X X 500 = 7. w = 8 150 H." by John H. — r> the reaction on 10 * See "Kinematics of Machinery. 135 (a). with a given radial pressure. p^ge . lower values of the coefficient of friction must be taken. but decreases the wear by distributing over several sur faces. the inner edge. 136. (b).
then for rotation in either direction P (a + b) = Rb or since R = F — Fb . F p 2 ' P = or  F sin I' —+F ll cos 6 . Considering body and taking moments around O. 136. 136. . and as dynamometers for measuring energy. 161. for The student is referred to treatises on power measurement is a discussion of dynamometers. tangential frictional resistance B be rotating. the reactions due to the — and the frictional resistances a forces — 2 wedging action. and dissipating it as heat. Equating vertical P = 2 I (R — \ s i n 6 ' f — cos 11 \ 0) or since / F = ff l 2 ( R\ ) or 2 2 v R = . Brakes used in heavy work. Block Brakes. then the wedge in equilibrium is 355 held by the force P. Friction brakes are used for controlling machinery by absorbing energy through frictional resistance and stopping from some moving part. the line of action of F A O the centre of the fulcrum of A. FRICTION BRAKES 160. Then will between the wheel and the block be R.R APPLICATIONS OF FRICTION each face and 26 the angle of the wedge. lever A. must often be fitted with water circulation to carry away the heat. . (1) F = p —r sin + a 1 (2) cos 6 less To avoid sticking the angle 2 should not be than 30 . With the arrangement shown passes through as a free in Fig. a its aR = F oppose motion. acting on the Let the reaction if C against the wheel B. presses the block Here the force P. The simplest form of brake the block brake as shown in Fig.
opposed by an equal force bearing near the effect is. In the calculations above. Obviously such proportions should be avoided. b that is. the braking due to friction of the journal is neglected. and for is reason this form of brake not well adapted to heavy work. 137. for the arrangement shown. . or when b — a c. tion in the opposite direction. In a similar manner for Fig. 137 the line of action of effect F does not pass through O A and its therefore in writing the equation for the equilibrium of must be considered. 138 P = ^U+*] R R r (3) the plus sign referring to clockwise rotation. the brake is selfacting and it if put in contact the moment of the frictional force will apply with everincreasing pressure. whence Fb p = a r~i b en • • • • + u±»] (2) be used for rotation in a clockwise direction. P = o. and the plus sign for rotasign is The minus to Fig. It is to be especially noted that for clockwise rotation when — = — fi . and the minus sign to rotation in the opposite direction. as its lever arm usually. In this class of is brakes the pressure of the brake at the against the wheel wheel. small. 138. this It cannot be neglected in designing the bearing.35^ MACHINE DESIGN In Fig. Fig. for the arrangement shown.
causes the brake beams of the wheel. the frictional force is is exerted on the wheel that F = 2 ti T. a toggle obtained and the tension intherods^. wooden blocks 131). Strap Brakes. B and B to press equally on opposite sides and causes no pressure on the bearings of the drum. line of the effect is rod A passes through the centre of the pin P. it If the pin O is so located when the load W applied moves to O'. ally lined with "locked" This in position may be removed. and the centre Fig. last feature is often Brakes of as shown.APPLICATIONS OF FRICTION 357 Fig. 139 shows an arrangement of brake beams for heavy work such as is used in mining machinery. in If T — be the tension 2 each of the rods A and A. acting on the system of levers. and^4may be in fact with made any desirable value. lected (see If the effect of centrifugal force is neg Art. such an arrangement care must be exercised in adjusting the brake that such pressures are not brought on the pins as operating force will cause failure by shearing. is When O and the a moves flown to O' the brake valuable quality in a brake. which may be applied by a steam cylinder. The force W. 139. this character arc gener 162. and the total tensions in the band (7\ .
which is subjected to the greatest tension T 1 is anchored.358 and MACHINE DESIGN T 2) be taken instead of the tensions per inch of width. and 142 show the most usual In Fig. 140 the end of the strap Fig. F = T T. If. equa tion (6) of that article reduces to irrr 10 Where also. 140. Fig. it could be anchored to any other convenient part of the frame. arrangement of band brakes. 141. a being the arc of contact in degrees. for con venience.0076 fi a. i i k (I) / \ k is = 0. 140. 141. 142. at the pin which serves as a fulcrum for the operating lever L. From (1) and (2). i (2) It is obvious that these equations are applicable to the discussion Figs. T = 2 F 10 — 1 Taking moments around O Pa = T or Fb 2 b 10 P = Fb a(io K i) * * * \ (3) ' which expresses the relation between the applied force frictional resistance applied to the wheel. Fig. F the total frictional force exerted by the band upon the wheel. of band brakes. P and the .
164. should be "selfsustained.J APPLICATIONS OF FRICTION In Fig. designed clutch should start its A load quickly but without shock. 141 the end under greatest tension lever is 359 attached to the and the end of least tension is anchored. For this case in a similar manner as above K^] o'S. that brake will is if any force itself applied to the lever. that is. Radially Expanding. P = is o. 139." when the clutch is driving. so that the force which it exerts assists the operating force P. Conical. its up with ever This form tendency to brake is exceedingly dangerous on account of if jjl "grab." especially is materially increased through a change in the character of the friction surfaces. 143 shows the elements of a conical clutch which shaft selfsustained. In be lined work they may engage with a castiron surface or may lined with leather. This is known as a differential brake. the continue to set increasing force of till motion ceases or rupture occurs. Disc. it necessary to hold the contact surfaces together. 142 the end under greatest tension at a shorter is anchored to the lever radius than the end of least tension. It is to « bl be especially noted that if io k = b2 . It and should disengage quickly. hence for this case io k P = „ Fb r a — Lio — 1 4 1 In Fig. Strap brakes are usually light made of wrought iron or steel. Friction clutches though made in a great variety of forms well can. in a large measure. it. and Band. is often necessary that the clutch should "lock" in place. namely. and the band will brake automatically. be classified under four principal types. The cone F is fast to the S and rotates with The pulley H rotates upon F and . FRICTION CLUTCHES AND FRICTION PULLEYS 163. no external force should be In addition. after the manner of the brake in Fig. Conical Clutches. is Fig. but in very heavy work they should be with wood.
For clutches that do not operate frequently.. 159. When the sleeve E is forced inward by the ring G. which fits the inner surface of C. The circular segment B. The loose ring G is operated axially by a forked lever fitting on the pins P. same wedge gearing in Art. (6) angle should not be less than io°. Fig. The clutch body A is keyed to the shaft. Heavy the springs at is G (not shown) throw the surfaces apart when thimble withdrawn. desired. shows the ele ments of a radially expanding. 144. Radially usually lined with wood. one surface 165. the levers E force the cone surfaces in contact. cork.. the links D force the segments B outward against C. selfsustained clutch. while the pulley C rotates loosely upon the shaft.. Fig. 143. In the arrangement shown the links have a toggle effect and can exert enormous pressure against .3 6° carries with it MACHINE DESIGN the levers E. 144 Expanding Clutches. When the thimble B is forced under the rollers C. metal is is surfaces are often used. The is relation between the transas that of the mitted frictional force F and the force the P applied to the cone. or F = ^ir stn $ + The is 11 P fi cos n . or leather.. can be moved radially upon A. but where the operation of clutching frequent. unless some mechanism provided for separating the cone surfaces. positively. in a direction parallel to the axis. when Fig.
145. Usually the sleeve has motion enough to carry the inner end of the link slightly past the centre position shown. 36 hence adjustment must be carefully performed. D adjust by means of turnbuckles or similar which also provide a means of compensating for wear. The It is discs often run in an oil bath to prevent " grabbing. Obviously any number of pairs of discs may be used. thus locking the clutch in place. alternate discs of steel and brass are employed. P. making the length of the link devices. disc clutch as Fig. A heavy as helical F (sometimes made of rectangular section shown) presses the two sets of discs together with a the clutch T . the total frictional force transmitted is proportional to the number surfaces n or F = /inP If the (7) mean friction radius of the discs is be r. C are carried on the part B and compelled spring to rotate with it by the keys G. 166. above form of clutch For small work. known load P. For large work the discs are sometimes made of iron and wood (or woodfaced). the frictional mo ment transmitted Fr = is jinPr. In Fig. when The sleeve B while compelled by the feather 5 to rotate with the transmission shaft A can be moved axially by means of the grooved collar / and the I being ring J. n = 7. shaft A being fast to the engine and B to the transmission shaft. which presses each pair of of pairs of contact contact surfaces together is the same. made fast to B but built separately from is it for constructive purposes only.1 APPLICATIONS OF FRICTION B. C. the cover of the case. is in and the shafts connected. Disc Clutches. The known as the Weston clutch. The by part A carries a number of discs. When B moved to the right the spring is compressed and the pressure on the discs relieved. which fit loosely in a radial direction but are prevented from rotating relatively to hold L. This is usually accomplished by able. in place." readily seen that while the force. 145 in shows the elements of amultipleautomobile work for connecting sometimes used the engine to the transmission shaft. placed alternately between the discs A bolts E which also A second set of discs D. and the discs run in Many pairs of contact surfaces are then used .
145. 146 illustrates the elements of a band clutch. Occasionally the band the wheel to be driven. though occasionally on self very large diameters they are attached to the wheel so that they may be turned true in place. but a special case of a . but in heavy work. the iron band C is tightened and clutches the rim of the driven wheel B. Band Clutches. is may When be the fast to one thimble F (which slides on the shaft) forced under the lever E. 105). 140 of Art." The width of the wearing faces of the discs should be made small to prevent undue wear toward the It is outer edges of the discs D. 146. Fig. such as mine hoisting. These clutches are made locking by arranging for a toggle effect in some one of the operating levers. better to use face. Fig. or other soft wood.3 62 oil to MACHINE DESIGN prevent " grabbing. 162. Fig. are used. Obviously the principles involved are identical with those of the strap brake. is made to expand inside is of the rim of It is to be noted that this case is not the same as the one just discussed. as in a thrust block (Art. blocks of bass wood. Fig. lining is The wood usually made fast to the strap. For light work the band may be lined with leather. shaft) The clutch wheel A (which carries the woodlined band C. more discs of a smaller diameter than a few of great 167.
78. Wooden should not be loaded beyond 20 to 25 pounds per square inch. July. regarding transmissive power.— APPLICATIONS OF FRICTION radially 363 expanding clutch. These are most In Evidently the general principles above.25 to 0. 1908. lower values must be assumed than for lower speeds (see Art.35. M. as the result of considerable experience in designing clutches. the compressive stress at any section being that due to the pressure applied by the operating lever. The pressure per unit area of surface is also if an important this is feature in the design of friction machinery. C. XXX. and Clutches. cork on iron. excessive wear will result. Inst. and released by magnetic action. leather. and iron with iron. W. A number of clutches have recently appeared which are operated magnetically. generally of the disc type. multipledisc type. and for wood on It For iron on iron >i may be taken as 0. Hunt gives the following values of jj. for taken too high. Thus in disc clutches the pressure is usually taken at not more than 25 to 30 pounds per surfaces square inch and lower values are desirable. Mr. 0. Vol. 168. namely: iron 0. apply also to these clutches.2. thus insuring safety against accident should the electric service Practical Coefficients for Brakes fail. The most In the usual combinations of friction surfaces for brakes and clutches are wood. Transactions S. ample surface mut be provided References: to properly radiate the heat generated. magnetic brakes.. 28).3. Mechanical Engineers. E. brass or bronze on iron or steel are sometimes used. the load is usually applied by a spring.3. If the clutch or brake is to operate frequently. Transactions A. or cork with iron. Magnetic Clutches. should be remembered that if the friction surfaces are to be engaged at high velocity. . considering the band as a thin cylinder under compression. leather on iron. 1903. or weight. 0. The outward force exerted by the band may be computed by the theory of Art.
General Principles. Barr. It is which will roll together with pure rolling motion. surface thus outlined shall the angle which the shafts friction wheels. will not suffice where any considerable is amount of power to be transmitted. and the relation of the axes. thus: Kind. Whether the elements of the be parallel or otherwise will depend on with each other. bevel. evident that friction wheels. Relation of Axes. Hyperboloids Any of the above None. may and works on mechanism the methods of drawing the sections of such surfaces for various conditions and velocity ratios. page no. In such cases the peripheral surfaces of evident that any pair of surfaces the transmission wheels are provided with teeth. strictly The most important and a few special forms of twisted and screw gears. so that the motion shall be positive. as discussed in the preceding chapter.CHAPTER XIV TOOTHED GEARING 169. The motion is transmitted by a pair of properly designed toothed gears identical with that If r x of the base curves or surfaces rolling together. as in the case of classified make and tooth gearing may be * according to the character of the pitch surfaces. 3 64 . and r2 be * See "Kinematics of Machinery. so as to give the serve as a basis for the design of a treat fully of required velocity ratio. Spur Bevel Parallel Intersecting Cylinders Cones Cylinders Screw Skew Twisted Face Not in one plane Not in one plane Any Any of these are spur. When it is necessary that rotation of one it is shaft shall produce definite and positive rotation of another. Pitch Surfaces." by John H. pair of toothed gears.
wheel. 147. is the driver. Contact between two teeth to the point of con has just begun at and the common normal O h passes through the pitch point O. and the Fig.TOOTHED GEARING 365 the instantaneous radii of such a pair of surfaces at the point of contact. The upper tact a M. represented in crosssection C on and D. commonly employed. namely. must that fulfil in order that the velocity ratio may be constant. must always pass through the point of langency of the rolling There are many curves which can be used for tooth outlines. then of both shafts is In the most common case the angular velocity constant and hence r x and r 2 are constant. Fig. is tlie common normal to the tooth outlines at the point of contact circles. but in practice the involute only two are cycloid. It If the teeth are properly proportioned the motion trans mitted will be identical with that produced by the rolling of C can be shown that the condition which such tooth outlines Fig. 148. and <" 2 be their instantaneous angular velocities. D. and the rolling surfaces are circular in crosssection. As the wheels rotate the [joint of contact will move along the the line aOb till contact to the ceases at b. a. and w. Hence in involute system the normal . 147 shows a portion of two gears whose rolling surfaces are a pair of circles circular cylinders. 147 illustrates a portion of two gears with involute teeth. Thus by the Fig. and which would fulfil the condition.
providing the thickness of teeth them to mesh. work that any gear of a given pitch pitch. 148 MACHINE DESIGN makes a fixed angle with the common tangent to shows a portion of two gears with cycloidal teeth. In order to obtain involute outlines of series of gears with fixed sufficient length. therefore. H. and any gear with involute teeth will run with any other gear having similar will allow teeth. R. for all gears of the series. Barr. by keeping the diameter of describing circle constant for gears of the series. ratio will whether the pitch circles are tangent or not. J. has practically superseded the A is fuller treatment of the theory of this geartooth outlines. The involute it gear." part In the systems in also. by F. " "Kinematics of Machinery. and a 9. page in. contact ceasing at The normal to the first point of contact is drawn. is a normal to the common tangent maximum at this point. Jones. Any involute tooth outline will run properly with any other similar outline.366 point of contact the pitch circles. and continually varies in direction though always passing through the point O. Contact is just beginning at a. Machine . as far as the length of the involute outlines will permit. will be found in treatises on mechanism. and as the gears rotate the point of contact will b. and it is clear that the inclination of the of the pitch circles. must be placed In the cycloidal upon the form and dimensions system interchangeability may be all accomplished. but for the transmission of constant velocity ratio with cycloidal gearing the pitch circles must remain tangent. Fig." by i. shall run properly with any other gear of the same In order that this may be so. certain limitations of the tooth. has a decided advantage for general use and cycloidal for most work.* 170. nominal pitch the angle Fig. and be constant Design. 147. Interchangeable Systems of Gearing: It is desirable in practical Standard Forms. within the limits of action. as far as the the tooth outlines are concerned. circles. It can be shown that in the involute system the angular velocity remain constant. move along the curved path a O b. made by the line of action with the common * See tangent to the pitch circles must have a proper value. which beyond the scope of work.
but the metal tooth may be thinner than . the greater will be the arc of contact.. In the case of cut / where no clearance of t^ears. is called the circular and will be noted by s. the diameter of the describing the radius of the fifteentooth gear of the series. . pitch . extending the arc of contact away from the pitch point. The tooth also becomes weaker as it is lengthened. i. and in some cycloidal systems the radius of a twelvetooth gear of the required pitch of the describing circle. increases the sliding between teeth. contact. 151. In some form> such as shown in Fig. but on the other hand. so nearly parallel that they may be cut with a rotary cutter. taken as the diameter this will re For a twelvetooth gear sult in radial lines for the tooth outlines below the pitch circle. It is evident from Figs. Fig. and also the velocity with which the teeth approach each other. In the practice of the circle Brown is & Sharpe Mfg. a compromise between conflicting conditions. the thickness remaining the same. This gives spaces between the flanks of the teeth on the twelvetooth. the tooth will have radial flanks. Co. = . 150. The length of tooth adopted in practice therefore. where a metal pinion engages with a gear having wooden teeth. or smallest gear. the pitch may not be equally divided.TOOTHED GEARING though there most common use this angle is 14K dency in modern work toward a greater angle. and for these reasons a practical limit is placed on the length of is. teeth. 147 and 148 that the tooth outlines of any system till may be extended both above and below the pitch line the longer the teeth the earlier will they meet.. 367 is a ten It is found undesirable in practice to make is gears with less than twelve teeth. is allowed between teeth. The distance along the pitch line from any point on a tooth to a corresponding point on the next tooth. The by /.e. thickness of the tooth along the pitch line will be denoted gears. which experience has shown will give good results. It is also clear that will they engage with each other. of teeth and the greater be the number of teeth continually in The is distribution of the load over a number of pairs in itself conducive to smooth running.
or of diameter. the clearance between top of tooth and bottom of space gears are in mesh. portions are given for cut teeth. Thus a gear 24" in diameter and 3 diametral pitch would have 24 X 3 = 72 teeth.' S = . the circular pitch. " N h the number of teeth in gear. the diameter of a circle through bottom of space. the addendum = when height of tooth above pitch line. " / the thickness of tooth on pitch line = width of space on pitch line in cut teeth. Co. then evidently Ns = nD. " " " = D = D = S = = 5 a = = c Z>] 2 the pitch diameter of the gear. number will of teeth N be divided by the diameter. diametral pitch is.05 inches. and which Let " " " used very extensively in America. the most convenient for use.. Then N = DS = 5 2 7T — 71 2 5 20 c = / 10 S . is the teeth per inch called the diametral pitch and s be denoted by . the diametral pitch. the quotient. and the circular pitch would be — = 1. 368 the MACHINE DESIGN wooden tooth. ordinarily. or depth of space below pitch line. the outside diameter of the gear. In the sys tem of teeth is adopted by the Brown & Sharpe Mfg. Since S = The — D and s = — N . S. If N be the number of teeth and If the D the diameter. total " d = = = = the dedendum. and in this country practically all interchangeable systems are based upon the diametral pitch.and s s = — 71 . " the total height of tooth. 151. S X = 71 .. the following proSee Fig.
54 5 to 0. Brown is & Sharpe Mfg.47 5 for small gears. and permanent patterns should be made of metal. greater clearance must be allowed in cast gears than in cut gears. for even with the impossible to obtain true spacing.53 s.48 5. or be slightly tapering to facilitate drawing. and consequently the cast tooth must also be tapering. Care should be taken in assembling such gears. solid Metallic gear wheels cast or forged solid. very important that the pattern itself be very accurately greatest care in moulding. and on account of the pattern in the sand. are either cast from a pattern.TOOTHED GEARING I 369 a d h x = s = a + c = 2a + c B = ^±^ and D =D2 2 (a + c) In the case of rough gear teeth. If the pattern for a spur gear is withdrawn from the sand with a movement parallel to the length of the tooth. the tooth pattern must have draft. For machinemoulded gears 5 and the corresponding space would be 0. that the tapers in i ( * The difference between the ommonlv ailed " backlash. to 0." ( thickness of the tooth and the width of the space 24 . patternmoulded gears good practice gives gears. on account of their tendency to warp and shrink.45 5 for large and the corresponding width the space / = 0. Wooden patterns are very unreliable for such work.55 s to 0. as already noted. metal by rotary or reciproit is Where the gear teeth are cast.* and the clearance in cut teeth. / = 0. difficulty of obtaining smooth surfaces.465 to 0. the difference need not be quite so great as in patternmoulded gears. If at the bottom of the space must be greater than For of the gears are machinemoulded. Co. would be 0. it is made. Methods Making Gear Teeth.52 Table in XXVI gives of dimensions of gear teeth for cut spur gears. or the rim and the teeth are cut from the cating cutters. accordance with the standards of the 171. the thickness of the tooth must be less than the width of the space. cast from a wooden pattern. on account of shrinkage and displacement due to " rapping" the For this reason.
0588 •o555 .0625 .0769 .2500 . particular defect of spur gears The due to draft does not exist in bevel gearing. t 5=°a 1 *& + c Depth **\'ooth.0500 "57 1052 0964 0890 0826 0771 0723 0681 0643 0609 °579 1079 mounted on an "indexed. The pattern then withdrawn radially and rotated to the next succeeding . are used only for rough or large work. Circ :ular Pi tch. Thick ness of Addendum Dep Space Pitch To oth.0666 .1666 . come together. or In forming the mould the segmental pattern is is placed in position and sand is rammed around it. In gearmoulding machines the pattern consists of a segment of the gear pattern.0909 •0833 .6666 •5714 .2000 .4000 •3636 3333 •2857 . carrying several teeth.0526 .5000 •4444 .1000 .0714 . 6 7 4488 39 2 7 349 3 J 42 2618 2244 1963 J 745 J57 1 1428 130Q 1208 1122 1047 0982 9257 7714 66l2 5785 5*43 4628 4208 3857 33° 6 2893 2314 1928 l6 53 I I I 1571 7257 438i 2326 0785 9587 8628 7844 7190 6163 8 9 IO ii 12 mi 1446 1286 5393 43 x 4 3595 3082 2696 2397 2i57 1961 1798 1659 i54i !437 1348 1269 1 198 11 35 13 14 15 2856 2618 2417 2244 2094 1963 1848 1745 1653 1571 16 17 l8 19 20 0924 0873 0827 0785 .8000 . 2 S z s a a + c I I* I* If 2 ^ I a* a* 2 3 i 3i 4 5 1416 5133 0944 795 2 S7o8 3963 2566 1424 0472 8976 7854 6283 5 236 1 1 1 5708 2566 0472 8976 7854 6981 6283 5712 523 6 4488 3927 3 J 42 1 .1429 .1 37° the MACHINE DESIGN two gears are reversed sets of teeth to avoid having the thick ends of both at one end." axis in such a manner that it can be rotated accurately through any portion of a complete revolution. and not for high speed. The pattern is TABLE XXVI PROPORTIONS OF GEAR TEETH Diametral Pitch. of the kind described above.1250 . 0000 1 1571 2 I . thus concentrating the pressure Rough cast gears.
150. Relieving the points of the teeth. The mould for this last the hub and arms then completed. however. all cast gears are much more inaccurate than Metallic gearing. especially any appreciable " backare unavoidof exists. because modern methods of gearcutting produce metallic gears of such accurate form that they may be run in places where mortise gears In were formerly considered indispensable. Usually the large gear. is the weaker of .TOOTHED GEARING position (the 371 indexing device insuring accurate spacing). They much used as formerly. on account of the shape of teeth. 149. is when run at a peripheral speed of more if than 1. reduces the tendency to produce noise.200 lash" feet per minute. slightly. teeth. Wheels with wooden are not as teeth known as mortise wheels. Where high speeds is able the teeth of one of the mating gears sometimes made are wood or rawhide. P. K. only. the pinion is made with wooden This is its or " mortise" rational since being made of metal. the pinion. in large work If being often accomplished by means of cores. when accurately cut and aligned. even inclined to be very noisy cut gears. the till operation being repeated the whole circumference is is moulded. The teeth proper are dressed to correct form with hand tools or by special machines using a fine circular saw for a cutter. Fig. Fig. making mortise is fitted wheels the wooden teeth are roughed out and the shank into openings cast in the rim of the wheel. and the latter are preferable where high speeds and smoothness of action are required. and gears may be made that can be run moderately high speeds. is machine at moulding well done the results are far superior to those obtained by pattern moulding. The teeth are held in place by the keys. 149 and 150. as shown in Figs. as shown. or pins. Obviously.
. There are many methods operation. and hence suffer greater wear.000 feet per minute being a not unusual velocity. for * See " GearCutting Machinery. number methods speeds. MACHINE DESIGN and also because the teeth of the pinion come into contact more frequently. Formerly it was cheaper to cast gear teeth.Flanders. that the cutter can be used.and the wooden gear teeth of thickness greater than to equalize strength. theoretically. See Fig. used that they are not subjected to extreme moisture nor run in too dry an atmosphere. 150.372 the two. held together by passing through the blank. found in practice. In recent years gears made of rawhide have been The blanks for rawhide gears prepared tallic rawhide discs much used for high speeds. assist the rawhide teeth in retaining their form. and have also so greatly reduced the cost of production that for high and where smoothness of action is necessary. made by cementing specially together under great pressure. a different cutter required for every different diameter of gear in a series of the same pitch. cut gears have largely superseded cast gears even in large work. Meare rivets discs. same manner gearing the that metallic is teeth are In of using rawhide pinion almost always made rawhide and the larger gear of Such a combination may be run at a very high rate of speed. how ever. on each side of the blank. Rawhide gears are almost noiseless in operation but care must be cast iron or brass. . the pitch or thickness of is tooth remaining constant. and. 3. the most of cutting gear teeth in practical of cutting spur gears being common method by the use with the of a rotating cutter. The teeth are cut in the blank in the cut. . of gears with of Modern gearcutting produce teeth of great accuracy. less the metal gear frequently has teeth of thickness than . without serious error. but the development of gearcutting machinery has changed the situation where a large small teeth are to be made.* The outlines of gear teeth vary number of teeth in the gear. To meet same this requirement would lead to an excessive It is number of cutters for each pitch." by Ralph E. In such com binations.
. Forces Acting on Spur Gears. 151 let the gear circle of let A be drive the gear B. It is to be noted. very accurate work may all be accomplished. generate correct of teeth for all gears of a series. as given below. " 35 " 54 " 26 " 34 " 21 " 25 " 20 17 " " 16 14 " " 12 " 13 When the teeth gearcutting is carefully done. are not theoretically correct. and only 8 cutters for each pitch in the involute system. 3 4 5 6 7 8 " " " " " 55 to 134 teeth. in the first The letters and numbers column are the manufacturer's designations. may be considerable. " " 23 24 " " 2^ 26 Y X W 100 150 2 so 49 59 74 " 99 " 149 " 249 or more Rack. i cuts 2 " " " " " " " from 134 teeth " " to rack. that the form of when cut with a set of cutters. Let V a be the velocity of the pitch circle of A. 172. Co. and let V b h be the velocity of the pitch B.* and even itself. for pur poses of ordering cutters. In Fig. and W * be the equivalent resisting force acting at the pitch circle forms There are gearcutting machines which. theoretically. however. only 24 cutters are used for each pitch in the cycloidal system. in best practice the error in the gearcutting machine cutters coupled with that due to dullness of hardness in the and deviation due to different degrees of metal.TOOTHED GEARING several sizes of gears of a given pitch. TABLE XXVII CUTTERS FOR CYCLOIDAL TEETH itter A B C " " " " u " " " D E F CUtS 12 teeth. " " *3 " " 14 a " r5 " " Cutter " " « U " " " " « " " N O p M 16 G H I " " " " " " " it 17 18 19 R S Q T U cuts " " u U " " " " " " " 27 to 29 teeth. In the system adopted by the Brown & Sharpe Mfg. TABLE XXVIII CUTTERS FOR INVOLUTE TEETH Cutter No. 20 a " " 3° 34 38 43 5° 60 75 " " " 33 37 42 " " " J K " L 21 to 22 teeth. as above. Also W & the equivalent driving force acting at the pitch circle of A. .
i^ system of gearing. The the shaft distribution of the reaction at the bearings due to the pressure between teeth (W lf Fig. 151). in Fig. in this respect. Therefore. and its bending effect on which supports B. be to As the obliquity of the line of is increased. rolled The . = 1 W a sec 0.i 374 of B. x hence sec is also increased. ' . the angle 6 (Fig. a V a =W h b V h b Therefore W a = W h b tangential driving force exerted by one gear upon another any correct therefore. independent of the angle of pressure. If MACHINE DESIGN now the tooth outlines are properly constructed. but the latter will. the line of action of the actual driving force W x will always pass through the pitch point and the angular velocity ratio of A to B will be constant. since 1 W. in any case. 151) is increased and W . will depend upon the relative positions of the gear directly proportional action CD and bearings. the same as a pair of teeth were continually in action at the pitch point. action of the pitch circles will be as though they their linear velocity will upon each other and h be the same or V = V & From the principle of work W The is. and the action if is.
three times the circular pitch 153). at intended to carry heavy and important loads. be neglected. is usually 14X is or 15 . 152. With it well is supported bearings. the influence of friction has In the above discussion been neglected. 148. sec = 1. that certain degree. during the arc of recess. at the end of contact. of recess it During the arc acts in the opposite direction and deforce is creases the obliquity. . maximum it at when the contact point lies increases and. In cycloidal gearing the obliquity varies from a the beginning of the contact to zero in the line of centres. more than one pair of teeth shall be owing to the unavoidable inaccuracy of form and spacing previously noted. 151) the line of W t such a way as to increase the effective obliquity. or the normal pressure 3^ per cent greater than the tangential force. and the angle Fig. The for influence of this frictional its small and may. about 8 per cent greater than the tangential The obliquity is constant throughout the arc of action in 0.08.035. not unreasonable to consider the load as fairly well distributed across if the face of the gear. the wellknown fact gears run more smoothly during It is recess than during approach. and machinemoulded or cut gears. it is not safe to depend upon a distribution of the load between two or more teeth of a gear. such gears should be carefully inspected and corrected. or the When equals 22 is sec maximum normal pressure rotative force. as indicated in Fig. The obliquity of the line of pressure gives rise to a crushing . It usually intended that in action at all times. Fig. usually. this load may be concentrated and if all one end of the tooth. is about 22 . but. on a In the rougher classes of work. the face does not exceed in width about (see Fig. with usual forms of cycloidal equals 1. is safest to provide sufficient strength for carrying the entire load single. When = 15 . but to a action accounts. During the arc deflects of approach the action of frictional in force F (Fig. but the torque on the driven shaft remains unchanged. involute gears. The maximum value of the angle teeth. 147. to a maximum 0.TOOTHED GEARING the pressure 375 on the bearings is increased with an increase on the obliquity of the line of action.tooth.
flexure. in determining their strength. teeth. at the 154 shows four gear teeth which have the same thickness Fig. its see Fig. the margin of safety assumed usually it unnecessary to consider 173. seems to gearing.376 MACHINE DESIGN component of the normal which results from the component. while in the case of steel. or composition gears. in January. teeth From the The 12tooth involute pinion it is weakened by a correction for interference. Mr. . the to first to Wm. made on a when acting on the point * of a tooth was drawn may have in. Wilfred Lewis. may be makes often neglected in gears made of that metal. is The tooth marked (a) is one of an involute rack. does not exceed 10 per cent of the normal pressure. This crushing ordinary proportions of teeth. 1893. The assumption made that the teeth of spur gears can be considered as rectangular cantilevers. of one of a cycloidal pinion of 12 teeth. far stronger in compression than in tension. an<^ n ^ s method of investigation was as follows : Accurate drawings of gear teeth were line of action of the large scale. Fig.. Gear Teeth. 155. is not satisfactory. Strength of Spur this component. is Its effect is to reduce the tensile stress due to Since cast iron this and to increase the compressive stress. es pecially when treating pinions having a small number of teeth. but usually better to correct the points of the mating gear. Sellers & Co. have been investigate the strength of gear teeth with due regard the actual forms used in modern His work was published originally in the Proceedings of the Engineers' Club of Philadelphia. * (c) is one of an involute pinion having 12 (d) is one of a cycloidal gear having 30 teeth. 154 pitch line and the same (b) height. and the normal force. with the action on the teeth (due to the radial force). in addition to the flexural stress tangential component.
(2) For the 20 involute system. thus beam uniform strength equal to the effective strength of the tooth (see Article 15). depending on the shape of the tooth. Modern practice sometimes makes the radius of the 15tooth pinion the diameter of the describing circle. W1 Lewis are found that its value is practically s. Lewis deduced the ing equations in which N = the number of teeth in the gear. N be found in Kent's " Mechanical Engineers' r \ Pocketbook. locate the weakest section of the tooth. a. (x) Where stress. The factor y a variable. From these tabufollow Fig. 155 Mr." page 901.124 r— ) N . The points of tangency a. . but depend \ 1 // \ '> ent gear. a parabola was locating a drawn tangent of to the sides of the tooth. For the 15 tooth pinion. Mr. lated "values. The .TOOTHED GEARING 377 intersection of this line of action with the centre line of the tooth. hr 6 = y bph 3 2 = bps(y ) 3 orir = bfs(y z —) = = bps(y) in inches. to this section is W e I. and the bending moment applied Then from equation J. b = the breadth of the tooth s p = is the tensile and the circular pitch. (3) Mr. which gives somewhat weaker teeth than the first system. involute system and the cycloidal system using of the 12 a generating circle whose diameter equals the radius / W=bpS o. page 94. Lewis' investigations on cycloidal gears were made on a system using the radius of the 12tooth pinion as the diameter of the describing circle. . l> i independent of the pitch proportional to (since h and S 1/ / the pitch). mainly on the number of teeth in the Tabulated values of this coefficient may ak.684\ [0.
'^) = b X ^ X ^(. is to is known. S. circular pitch will also be transformed to diametral pitch.. however. and shafts when the face of the gear is common problem and as follows: their velocity ratio is A is known or assumed. illustrate. where the number of teeth (N) problem in design known. and the velocity ratio fix the diameter of the gears. b. the diametral pitch = — N=DXS Therefore W= bsp (. the load per inch of face. gears to connect these shafts for a given load and working stress on the teeth. equation (2) will be adopted or p. (6) ' The pitch can be found from equation p. A. 194 V + J \ * ° .* To accord with modern practice. s. To M. compared to the variation in the little assumed for small stress. for any values of w.g _^1^\ p (6) D . . The centre distance of the shafts.. and since cycloidal teeth are now used and moderatesized is gears. but a very common (s). Barr. in this work for standard gears. be governed by the space available. or designer The face of the gears may it may be assumed by the upon other considerations. Trans.124^) (4) or since w = W = f b and therefore S w *(. The Lewis' formula convenient for determining is W. page 766..124 Then sb or s =— . suppose * See a discussion by John H. Let " " D = w = S = the pitch diameter.378 difference is MACHINE DESIGN small. p. E. Vol. D. XVIII. determine the pitch when the pitch is diameter of the gear given and the number of the teeth to this last stated un The formula may be adapted problem as follows. The distance between two known required the pitch of spur .
o?8 °  2.000 ( V 1.000 lbs. and then moving horizontally at the top of the is to the lefthand scale B. as scale C. let (5) A curve 6. hence. in Figs. D= 10 w = 983. S = 1.. drawn for each pitch.000 per square inch may be found by passing vertically upward from the given diameter on scale A. pounds the allowable load per inch of face for a stress of 6.500 lbs.000 lbs.500 V + \1 ~ .sOo\ / ^ 5 = . to illustrate. 156 and 157 are plotted from That in Fig. Assume also that the be taken as 6 inches. and that the be 40 inches diameter. and a 45 diagonal the lower righthand corner of the . 6 face of the gear face is may f The load per inch of w = 15. then. etc. is smaller gear to p = 8.389 6. hence when D= w= o. w = 1.15 D X 2.000x40 X )= Li or say 1 diametral pitch.7.>.5. Plotting these corresponding values of D and w as abscissas and ordinates.2s) ) w 3. The diagrams shown equation (5). any other is other stress could have been taken for plotting the diagrams. which gives the required load per inch of Scale B is re produced marked 6. for a stress of 6.000 pounds per square inch. the curve is for a diametral pitch = 1% drawn through the points.000 = 2.1c. 157 covers the range from 5 to 1 diametral The abscissas (Scale A) represent pitch diameters of gears in inches.ooo/ (. to the curve corresponding to the pitch.. per square inch. and the ordinates (Scale B) the load in pounds per inch Any of width offace. let p = Substituting these values in W = or .270. respectively.TOOTHED GEARING 379 W = 15.194 2. face. 156 covers the range from 12 to 6 and Fig.000. 2. the diameter of the gear known. diametral pitch pitch. If. The other curves are constructed is in a similar manner. D = 20. and may be used in solving problems by them.5 — 2. ° J 8.000 drawn from diagram.
instead of moving the horizontally from the pitch curve to scale B Scale on the left. the marked 6. If.3 8o MACHINE DESIGN to scale C. movement be horizontally to the right (or left) to the diagonal and then vertically upward to the scale C.ooo same reading will ? be obtained on C as originally found on B. . diagram then.
*1/ TO § 7 / A 1\ / / 7 / A / / 4 4 5 V / 4> r > g 0. » »S 1 CO t* $ H 00 N £ •£ N» M OS of face for stresses corresponding to these respective points this on scale. / / 7 71 Ty </ ^\ # ///// '///// ay ^^ / } ^r " 8 Si oov ^cT  . if 381 other diagonals be drawn.' TOOTHED GEARING Furthermore.r 1 / / 3 a 1 s /E^// / / _77z // ?$/Y/ / / / // r / / / j£ 77 1 / <v & F^  te / * // IV// /fe ////. //JK / y / . they may B be used to read loads per inch Scale ^N b*L \ w ^L ! */£" 7* \ ^U f £ i 1 / /^ 7< / 4 A /4> / »^ s u \vr\^% \ \ »i l\\\ \i £ DO \\\\ I \ \Y ^ /// / / / / / \ / / X/ ' / /# A? / > X \ ! / f V / / / / / / / . cr / \ / / /I \ W /// K // K 7^" /If /i j / / / ^K / /» ' / f / p^ II l« = ti^ // / . S Dfaunetnl Pitch Circular Tit. since from equation (4) it appears that the stress varies directly with the load. from various points on scale C. These stresses are indicated along . 8 L ///// V/// /// // / y . as shown.
on scale A. beyond this point will not give additional strength.. Fig. with an assigned start at the point on scale (A). that the unit load taken cannot be carried with the stress and diameter assumed. because the length of tooth will be increased. on scale A. which The pitch curve nearest this point pitch.382 the several diagonals. pass upward from the diameter reading the adjoining diagonals. then horizon tally to a point on the vertical rising from D = 40". and the flanks will be undercut to an extent which more than compensates for the added thickness of tooth. The i. pitch. upward to thence horizontally (right or thence left as the case to scale may (C). be). The various curves have not been extended far beyond this point as in that case they intersect. for 1 diametral pitch would be the required If it is required to find the load per inch of face for a gear of stress.000. points of tangency correspond to the diameters of gear at which cycloidal teeth have radial flanks for their respective pitches.000.500 on scale (C). a 12tooth gear. (See Fig.e. by any possible pitch. MACHINE DESIGN Thus to find the pitch when w= 2. . It may be noticed that the different pitch curves in either Fig. 157. The data may be such 157. D = 40. The only recourse for the If the common tangent through O. it is evident that the remainder of the curve is of no practical importance. and mar the clearness of the diagram. p = 8. Since this intersection occurs after is the diameter of the 12tooth gear reached. 156 or 157. 156 or 157 have a common tangent through the origin O. to the proper pitch curve. and the stress may be found by interpolating between If the diameter. upward where the unit load may be read off. pass the given pitch curve . i. given diameter and pitch. thence horizontally to a point under the unit load on scale C. will lie to that a point corresponding to a.. to the proper stress diagonal. or a less number of teeth for a given diameter.) and unit load are known quantities. Fig. all the left and above the pitch curves. pass verti cally downward to the diagonal marked p = is 8.e.500. from 2. corresponding to the diameter. increase of pitch. In fact. above same data were substituted This means in equation (4) an imaginary quantity would result.
is caused by the rotative effort of the driving gear. have just come into contact as shown is 147. nor. then the driving force applied to the point of the driven tooth by the root of the driver. a pair of bevel gear at a. will The tooth of the driven wheel be deflected a certain amount. Hence. the magnitude . If different material. 174. of the given diameter and total load face. Strength of Bevel teeth. where the thickness of of and spaces are unequal. Fig. they come in also. necessarily. This does not apply teeth teeth to certain forms of reinforced or " shrouded" discussed later. nor where the mating gears are Gear Teeth. while the deflection of the drivSince the deflection of the driven tooth ing tooth will be negligible. 158. is to increase the and thus reduce w } or to use a material which permits a higher intensity of stress. conis tact more frequently. in general.TOOTHED GEARING a gear. if the small gear properly designed the larger gear will have sufficient strength. 158. 333 W. Fig. Fig. It should be noted that the teeth of the smaller gear of a mating pair are is weaker in form than those of the larger. greater on the teeth of the smaller gear since The wear.
384
of this deflection at
MACHINE DESIGN
any point on the
line of contact of the
two
teeth will be proportional to the
movement
its
of the corresponding
contact point of the driver or to
distance from the axis of
rotation of the driver; hence, from similar triangles, will also be
proportional to the distance from
its
own
axis of rotation.
all
Now
points,
the crosssectional outlines of the tooth are similar at
and
it
can be ^hown that in the case of simple cantilevers of
deflection.
It
form the load applied is proportional to has just been shown, however, that the deflection
similar
of the tooth at
any point
is
proportional to the distance of that point from the
axis of rotation.
Hence
the load
also be proportional to the distance
on the tooth at any point must from the axis, being least at
the small end,
and greatest
at the large end, the
mean value
be
ing at the middle of the tooth.
the
Therefore a spur gear which has
same width of face, and teeth of the same form and pitch mean section, will have, theoretically, the same strength as It can also be shown that in simple cantilevers the bevel gear. of equal breadth and similar outline, the stresses induced at coras the
responding points on
applied
is
the
cantilevers
are equal,
if
the load
the
proportional to the linear dimensions.
stresses are the
Hence
maximum
It is
same
at all sections of the tooth.
evident that the relation established above between the
mean section of a bevel gear and a spur gear with similar teeth, may be (and often is) used as a means of designing bevel gears. It is much more convenient, however, to deal with the teeth at
the outer or large end.
of the
If, also,
the pitch radii are used instead
addendum
radii, the error will
not be great.
Let r 1
"
"
11
"
11
= r2 = r = b = w = w=
2
the pitch radius at the small end of the tooth.
the pitch radius at the large end of the tooth.
the
mean
pitch radius of the tooth.
the width of the face of the tooth along the elements.
the load per inch of face at the radius
r.
.
the load per inch of face at the radius r2
" "
at
W=
w=
e
the resultant load on the tooth
= w
b.
the equivalent load per inch of face which,
r2,
if
acting
a radius
would produce the same
rotative effect as the
actual load.
TOOTHED GEARING
Since the load on the tooth varies as the radii, the total
sul tii nt load will act at a radius
385
re
R
— =  — 3 ('V — >V)
2
(r
3
r
3
;
)
^.
and the torsional
moment due
to the resultant force,
W,
will
be
3
(r2
2
''i
J
Xow
by
definition
2
w
e
b r2
(r 9
3
=
r
W
3
)
i?.
Therefore,
w =  w 7—=
3
^— (V — >YV2
(7)
Also, since the load varies with the radius,
71:
w
r
r2
w
+
rx r2
2
w
+
>\
r2
2wr r + r
2
(8)
t
And from
(7)
2 '
and
'2
"
.
>'l
(8)
2
— = ™
U>*
t
r2
+
W j~i
will
3
— W—
(r
3
r
3 )
; '2
:r
=
3 (V 2
— —
rx )
r2
;
'l
2
''I')
(V
will
~
^
—
*
=
k
(o)
)
The
actual load,
w2
,
always be greater than
therefore,
w
in the ratio
shown above.
carrying the
A
bevel gear,
be more heavily however,
(5)
loaded at the large end than a spur gear of the same diameter, and
same
torque, in the ratio
shown above.
r, is
If,
w
t
is
known,
w
2
can be computed, and used in equations
Usually
and
(6) of Art.
173 instead of w.

made
not less than
r 2
3
.
When r,= r 2
3
,

=
k
=
1.4,
and
this value
can be used
in
It
computing
w
2
unless the face of the gear
is
excessively long.
should be especially noted that in solving problems in bevel
gearing, either
by the diagrams
that
f
of
Figs.
156 and 157 or by
(([nations '^) and (6) of Art. 173, the diameter
D, which must
as the
be substituted therein,
circle,
is
corresponding to the formative
sec
0,
whose radius the tooth is fixed by
is
R =
r2
Fig.
158,
form
.
of
this radius
and not by the radius r2
The
* See also Mr. Lewis's article,
Jan., 1893.
Proceedings Engineers' Club of Philadelphia,
25
386
MACHINE DESIGN
computations should be made for the smaller of the two gears, as
in the case of spur gears.
Example.
the
with a velocity ratio of 3 to 2
Design a pair of bevel gears to transmit 50 H.P. the gears to be of cast iron, and
;
maximum
fibre stress to
be 4,000 pounds per square inch.
The
revolutions per minute of the shafts are to be 300 and 200
respectively.
Lay
to 2.
off the
that corresponding radii of
Then it is
and OT, Fig. 158, and draw OU so and are in the proportion of 3 found that = 34 and 0' = 56 Assume tentaaxes
OV
N
M
.
tively that r 2 for the large gear
=
15" and r 2 for the small gear
4".
=
10",
and take a
trial
width of face of
The
feet
velocity at the radius r 2
=
2
X
7T
X
IO
X
300
=
1,575
per minute.
Hence
the equivalent total load at a radius r 2
1,050,
=
W
Tir
=
50
X
2
—
1.2
33,000
=
k
or
w =
=
i,575
—— =
1,050
262 pounds.
4
e
Therefore
face.
w = w X
X = =
=
262
X
1.4
Since sec 6
sec 34
=
1.2, the
367 pounds per inch of diameter of the formative
Fig. 157, or
circle is
20
24"
and from the diagram,
that for
from
24", 4,000 pounds, (6), it is found and w = 367 pounds, the diametral pitch is very nearly 3X? which may therefore be selected. This would give 70 teeth for The width of face is a the small gear and 105 for the large gear. little more than three times the circular pitch and is therefore in accordance with good practice. Gear teeth of all kinds are likely 175. Stresses in Gear Teeth.
equation
D=
p =
to
be subjected
to shock, unless
running
at a very
low
velocity,
and the danger from shock increases as the velocity increases; hence the allowable stress must be reduced as the velocity is increased. Reliable experimental data on the allowable stress
in gear teeth are lacking, although
be found in
treatises
on the
subject.
many empirical rules are to The values given by Mr.
The
following
Lewis, in the paper already quoted, are probably as reliable as
any, for teeth that bear along their entire length.
equations have been deduced from Lewis's work:
TOOTHED GEARING
387
\
(600 7?)
;
600
+ V)
•
•
(10)
for steel r p
=
20,000 f
V600
+ V)
j
.
.
(11)
where
V =
the velocity at pitch line in feet per minute.
stress
It
would
be probably safe to take the
for
bronze as r p
=
is
12,000
(—
\6oo

+ V)
)
.
.
(12)
since the resilience of bronze
greater than that of cast iron.
is
An
old empirical rule for rough cast teeth
W
where W, as before,
is
=
200
X
s
X
b
(13)
the total load, 5 the circular pitch,
and
b
the width of face of tooth.
made
cast
of
beech or
The strength of wooden mortise teeth, maple, may be taken as about onehalf that of
and the strength
of
iron,
under the same circumstances;
good rawhide gears
may
be taken as equal to that of similar gears be noted that a rawhide gear
will en
made
of cast iron.
It is to
dure considerable more shock than one
made
of cast iron.
While rough cast teeth are more
only,
likely to bear
on one corner
they
are
stronger
in a
than
cut
teeth of
the
same
pitch,
which compensates
there
is,
measure
for this defect;
furthermore,
usually, an excess of strength, to allow for wear, in all
new
gears,
and the subsequent wear tends
to correct the initial
unequal bearing, along the elements.
On
there
safely
April,
is
account of the increased
liability to shock,
with increase
of speed,
and
also because of the noise of operation at high 'speeds,
a limit to the speed at which any form of gear
may be
which
and conveniently operated.
Mr. A. Fowler,
in Engineering,
at
1889, gives the following as
maximum
values
gearing
may
be successfully operated:
Ft. per
Min.
Ordinary castiron gears
Helical castiron gears
1,800
2,400
Mortise wheel and
<
astiron pinion
2,400 2,400
3, 000
Ordinary
caststeel gears
Helical caststeel gears
Special castiron machinecut gears
3,000
388
Although higher
MACHINE DESIGN
velocities are occasionally
found in practice,
these are undoubtedly
maximum
average values and, in general,
more than twothirds the values given Rawhide gearing, which operates almost noiselessly, may be run satisfactorily up to 3,000
the velocity should not be
above, on account of noise and wear.
feet per minute.
176.
Width
of
Face of Gears.
Equation
the load per inch of face that
may be
of the
(5) of Art. 173 gives applied to a tooth of the
given form and pitch, the total load depending on the width of
face as
shown by equation
with increase
of
(4)
same
article.
The
of
dur
ability of
a tooth for a given load
face.
is,
therefore,
difficulty
theoretically
increased
The
is
securing
in
uniform distribution of the load along the contact element
creases, however, as the
width of face
increased,
and
this imposes
a practical limit to the width of the face.
intensity of the load on the tooth
is
On the other hand, if the
too great, excessive wear
may
stress
result.
The
equations given above do not take wear into
account, the allowable load being fixed with reference to the
alone.
On
this basis
a large tooth
may
The
carry a
much
higher load per inch of face, but the wear will be proportionally greater,
the velocity being the same.
empirical rule
given in equation (13) of Art. 175 assigns a load of 200 pounds per inch of face, per inch of circular pitch. For a tooth of 1"
circular pitch this load will give,
of only 2,000
by the Lewis equation, a stress pounds per square inch, for moderate sized gears. This is a very low stress, for ordinary speeds, so that this rule would give more durable teeth than the Lewis equation, as
ordinarily applied.
Experimental data on the durability of teeth are lacking.
is
It
evident, however, that the allowable load will
depend
largely
on the character of the service, velocity of rubbing, lubrication, and the material used. Thus, for ordinary cut castiron teeth under constant service, the value given above (200 lbs.) is probably conservative; while with teeth of highgrade steel
greater loads
of over 2,000
much
may be
carried.
Cases are on record where loads
of face
pounds per inch
were successfully carried,
with a peripheral velocity of over 2,000 feet per minute, the
TOOTHED GEARING
389
pinion being of forged steel and the gear a steel casting, 4.92"
circular pitch.
to
Wellmade gears
of
rawhide
may be
loaded up
150 pounds per inch of face, per inch of circular pitch; but in
no case should the load exceed 250 pounds per inch of face.* In the case of machines such as punchingmachines which
work
intermittently,
and whose operation extends over a short
is
space of time, the element of wear
not so important in the de
sign of the teeth; but in such gears as those connecting street
railway or automobile motors with the driving axles, where the
work
is
both continuous and severe, wearing qualities
may be
fully as
important as strength; and gears
made
of steel or other
hard materials
may have
makes
;
to
be used solely on
this account.
Good
practice
the face of the tooth about three times
the circular pitch
but in fixing the pitch and width of face, in
points
extreme cases,
sidered.
177. Other
to the
the
discussed
above
should
be
con
Forms
of
Gear Teeth.
Gear teeth made according
dis
Brown
is
&
Sharpe standard, on which the foregoing
cussion
based, have been found very satisfactory for average
conditions,
and are
in
most common use
it
in this country.
For
extreme conditions, however,
has been found necessary to
reinforce such teeth, or to use teeth of a different form.
A
very
common way
of reinforcing teeth of cast gears is
by
shrouding, which consists in casting an annular ring of metal on
one or both ends of the
the gear tooth
teeth, as
shown
in Fig. 159.
cast as an integral part of the gear casting,
This ring is and hence strengthens
are shrouded to the
their outline,
by
practically twice the shearing strength of the
crosssection of the tooth,
top.
when both ends
from
The
teeth of the pinion are,
always
is
weaker than those of the gear, and the wear on them
greatest.
also
The shrouding
should, therefore, be put on the pinion;
and
if
carried to the top of the tooth
on both ends
it
will give
them an excess
widths of
in
of strength over those of the gear, with usual
If the gears to
face.
be reinforced do not
differ greatly
diameter, the teeth of both
may be shrouded
the
half
way
up.
Private communication from
New
Process
Raw Hide
Co.
390
Shrouding
is
MACHINE DESIGN
used mostly on rough cast gears, the shroud practi
cally prohibiting the cutting of the teeth
If the gears are to
by the usual methods.
run in one direction only, and where very
buttress tooth,
is
heavy pressures are to be withstood, a form of tooth as shown in
Fig.
1 60,
and known as a
may
be, but
seldom
is,
employed.
The
driving face, A,
made
of correct theoretical
outline, while the
give the required strength,
back face B may be of any outline* that will and clear the teeth of the mating gear.
form
The
front face should be of standard cycloidal or involute
obliquity of generator than
and the backs are preferably involute forms, with a much greater would be permissible in driving.
For some time past there has been a marked tendency f on the part of the designers of gearing for extremely trying service
to depart
from the Brown
&
Sharpe standard, and
to use teeth
by that standard. In some instances the same angle of pressure has been retained, while in Mr. C. W. Hunt reported others this angle has been increased. to the A. S. M. E. in 1897 (Vol. XVIII) the results of the adoption of such a system and gives full information for their design.
shorter than those given
somewhat
A
few other manufacturers have adopted similar systems.
The
need of a small gear of great strength, in automobile work, has
increased the
demand for a stronger form of tooth, and it would seem that the old standard must be modified or a second standard adopted for extreme service. The most prominent form of these
socalled
" stub teeth," at present,
is
that
advocated by the
Fellows Gear
Shaper Co.
In
is
this
system an involute tooth
with a pressure angle of 20
c.8 as high as that of the
used, the
addendum being about
Sharpe standard.
Brown
&
gives a tooth nearly twice as strong as the old standard.
This Some
times these stub teeth are given the height of a standard tooth of
smaller pitch;
thus a 6pitch stub tooth
may have
the length of
is
a standard 8pitch tooth, in which case the gear
described as a 68 gear.
sometimes
Notwithstanding the fact that the arc
is,
of contact in stubtooth gears
generally speaking, less than in
* See " Kinematics of Machinery," by John H. Barr, page 131.
f See a paper by R. E. Flanders, Trans. A. other systems.
S.
M.
E., Vol.
XXX,
resume of
TOOTHED GEARING
the old standard, they run well and will undoubtedly be
391
more
used in the future.
178. Strength of
Gear Rims and Arms.
The rim
of the gear
wheel must not only be strong enough to
resist the forces
brought
upon
teeth
it,
but
stiff
enough
also to prevent improper, action of the
due
to springing of the rim.
A
section of
rim between two
arms may be considered
as a
beam
fixed at the ends
is
and carrying
0,
a load at the middle, the value of which
W
t
sin
Fig. 151. 1.25
t,
Good
where
practice
/
makes
the thickness of the
rim
at least
is
the thickness of the tooth on the pitch line.
For small
In
gears this proportion gives ample stiffness, but for very large
gears stiffening ribs are also sometimes necessary.
many
cases the thickness should be sufficient to allow of dovetail
Shroud
Fig. 159.
Fig. 160.
Fig. 161.
ing a tooth into the rim, in case of accidental breakage of one
or
more
teeth.
Gear wheels are seldom run
induce
at
peripheral
velocities
which
dangerous centrifugal
stresses.
The
principles governing the design of such wheels are discussed,
however, in Chap.
XV.
wheels
The arms
ing that each
of gear
may be
treated as cantilevers, assum
arm
carries a load
W — n
,
where n
is
the
number
of
arms, and
of either
the tangential load. Computations for strength arms or rims must, however, be considered as giving minimum dimensions, stiffness being the prime requirement, and due regard must be paid to proportions of rim, arms, and hub,
to minimize shrinkage stresses
W
due
to cooling.
392
MACHINE DESIGN
179. Efficiency of Spur Gearing.
The
experimental data on
the efficiency of spur gearing are very meagre.
available data are those obtained
of
Probably the best
for details
by Mr. Wilfred Lewis,
VII.
of
12
which see Trans. A.
S.
M.
E., Vol.
His investigation
was made with a cut spur pinion
of 39 teeth.
teeth
meshing with a gear
The
circular pitch
was
\%
inches and the face
3^
inches.
tooth,
The
load varied from 430 pounds to 2,500 pounds per
and the peripheral speed ranged from 3 feet to 200 feet The measurements included the friction at the teeth, and the friction at the bearings. The efficiency, as observed, varied from 90 per cent at a velocity of 3 feet per minute
per minute.
to over 98 per cent at
200
feet per
minute.
It
appears that the
friction at the teeth is a small part of the loss with
good cut gears,
the greater portion of the loss being at the journals.
The
effi
ciency of bevel gears
is
somewhat
less
than that of spur gears,
friction
on account of the axial thrust, which induces
the
between
hub
of the gear
and the
collar at the supporting bearing.
HELICAL OR TWISTED GEARING
180. General Principles.
Suppose a spur gear to be cut into
n small
sections
If
by a
series of planes perpendicular to the axis of
rotation.
each section be then placed a proper distance ahead
(a), it is
is
or behind the adjacent section, Fig. 161
evident that they
just
may be
of
it is
so arranged that
its
some one
section
th
coming into
contact with
mating section when the n
section in advance
in contact at the pitch point.
some
section will always be in contact near the pitch point,
With such an arrangement and
n points of contact with the mating gear between the pitch point and the point which marks
there will always be approximately the beginning of tooth action.
Since the action of gear teeth
is
smoothest when contact
of gearing runs
is
near the pitch point, this arrangement
more quietly and smoothly than ordinary spur gearing, and it was at one time used in machine tool and similar work where smooth action is very desirable. As the number of sections is increased, the total width of
the gear remaining the same, the spacing of these sections being
161 then the total load normal to the face is W l = W cosec 6. load which (c). Gears of this type are also called herringbone gears. made form of helical. and may follow any desired law. is for an ob vious reason. as shown in desired to use cut teeth the wheel is Fig. Since the pressure. When the number becomes infinite become helical in form. or the wheel may be made in one piece and the two sets of teeth staggered so as to times allow them to be cut. teeth to the surface. otherwise the full load will be thrown 181. and the breadth of the gear Hence the load per inch of face on . one tooth of a twisted gear transmits to the length of the tooth be mate be W. If denoted by by b. Strength of Twisted Gears. If the effective its on onehalf the gear and the object of the double gear defeated. is that since the relative position of adjacent laminae arbitrary. within the arc of contact. this form being more practical of gearing is also known as twisted gearing. Fig. teeth on each gear. With the arrangement shown in Fig. the form of the stepped tooth approaches that the teeth shown in Fig. 162. 162. but may This have any desired shape. It is however. 162. 161 (c). causing end thrust on the This can be obviated by making two sets of helical shaft collars. between mating component.TOOTHED GEARING 393 kept uniform as before. and contact is continuous along is that portion of the face which evident. then / = b cosec 0. and should not be confused with that of screw gearing. but in both of these constructions there is some loss of strength due to the absence of the reinforcing action of teeth cast solid as in Fig. /. must be normal which tends to move the gear in an axial direction. A screw gear must have regular or unia twisted gear does not necessarily form helical have teeth. When it is some made in two parts and fastened together. care accurate. The action of such gears identical with that common spur gearing. the outline of is the tooth in an axial direction not necessarily helical. Fig. while this limitation. 161 (c). although these teeth are most usually to cut. one righthand and one lefthand. or end play of must be used that the alignment in an axial direction is must be provided so that the middle plane both gears coincide. also twisted though certain limiting forms of the latter are gears. there is a W.
the twist be If made equal It is clear. diagonally across the tooth face. 162. 16 Fig. perhaps. The and however. a difference of 25 per cent being. If the twist is so great that when the end in advance is going out of contact the other end is just coming into contact. 164. a twisted tooth WcosccO W. MACHINE DESIGN . the line of contact will run diagonally across the tooth from point to flank. Fig. the arm of the force which amount of decrease depending Fig. assumed that the load per inch of face in twisted gears is the same as that of a spur gear of equal width and equally loaded. tooth action is continuous at every point of the arc of action and this proportion is the one most used. in axis of ro This is twisted gears. On account of continuous tooth action and consequent smoother operation in twisted gears. =— = — / TT or the b same as in a b cosec spur gear of face b. to the pitch. may be the tooth to face. This would be strictly true if all points in the liae of contact were at the tation as in a spur gear. pitch is however. the line always extending. the effect of shock is lessened some . is small. that the assumption often made same that twisted teeth are twice as strong as spur teeth of the not true for teeth of usual proportions. on the amount of twist in the tooth. and the average arm of the driving force will be about onehalf the height of the tooth. of contact same distance from the never so. .394 . and on the side of safety. decreases the lever tends break the tooth. This diagonal distribution of the load across error it due to this. as much as can safely be assumed.
name While the teeth of such gears resemble those of helical twisted gears. t A highly successful form of this arrangement is the wormandrack drives on planing machines. This last arrangement is commonly known as a worm and wormwheel. as in spur gears. making a very strong tooth. Halsey. and are very rarely used. Co. is right angles. 395 Twisted gears have been used with success on heavy windrough cast and ing and hoisting engines. and on these surfaces helical teeth may be constructed which will transmit the desired motion. Fig. Fig. and on the line cut worm is at the proper angle is The contact in these cases point worm wheel tooth confined to points in a from the working surface of the tooth by a plane passing spiral gears.TOOTHED GEARING what. pitch cylinders may be described on the axes. 165 illustrates such a worm and wormwheel. If They are difficult to construct." by Brown & Sharpe Mfg. * For a full discussion of the . Sellers & Co. see a "Practical Treatise methods of laying out and producing socalled on Gearing." by F. their theory and action are quite different. When it is the axis of two shafts are not parallel and do not intersect. The same sometimes obtained by using a plain spur gear. for. the teeth being often both gear and pinion half shrouded. the load can be carried on point contact. and also "Worm and Spiral Gearing. there is also a sliding and sliding component along action of screw at the elements between contact surfaces. The gearing is very smooth. The special case where the axes are helical teeth. possible to lay out contact may be constructed which will give Gears of this kind are known as skewbevel gears.f contact. the latter Such gears are known as screw or being really a misnomer. SCREW GEARING 182. A. in addition to the conjugate rolling action.. surfaces on which gear teeth line contact. first used by W'm. the teeth on the worm wheel being truly helical in form and cut result is at an angle to suit the worm thread or helix. and where a large wheel having many helical teeth meshes with a small one having a very few that an im portant one on account of the great reduction in velocity ratio may thus be obtained. and setting the axis of the with the plane of the gear. 163. spiral * gears. Forms of Screw Gears.
The sides of the involute rack face are at right angles to the line of contact. it will cut a trapezoid from the worm somewhat different from that cut by the median plane. if and. It is evident also that any other form cf worm thread may be similarly treated. in general. in fact. however. aO b. the the teeth of the worm itself is moved axially it will engage worm wheel in the same manner as a rack does with a gear. Fig. is 2 0. to construct a worm wheel in such a Referring manner shown is as to secure line contact. as in spur gearing. Fig. it to Fig. like all spur gears. and wheel tooth in the plane rectly it is M N may be so made as to wormmesh cor with this new trapezoidal section. can be seen that when the singlethreaded . page 125. be similar to those clear that the shape of the on the median plane. worm wheel and that rotation of the to a translation of these sections worm. worm moved to a rotated through 360 any median section as A is forward an amount equal to the pitch of the position B. 147. The with action is equivalent to translating a rack of similar proportions. and hence and in = 29 If other planes such as 2 N be worm and worm wheel parallel to the median plane X X.* in and this property is usually taken advantage of making worm gearing. 164. since a worm thread of such a crossis section easily machined. that the teeth can be cut with a rotary cutter. 147. however." by John H. It is evident that if enough such sections be taken. the inclination of the sides to each other the standard system . Barr. The rackpassed through the like action of these trapezoids M would.396 MACHINE DESIGN through the axis of the worm at right angles to the axis of the worm area. to construct. and suggests a method by which * See " Kinematics of Machinery. In practice the point of contact becomes a limited of this The advantage is form of worm wheel. . wheel. is equivalent backward or forward. In the involute system of gear teeth the rack has straight sides. a complete tooth outline may be formed that will give line contact with a worm across its full face. The preceding discussion demonstrates the possibility of line contact in screw gearing. 164. Fig. and patterns for rough cast teeth are comparatively easy It is possible.
mounted in a gear cutter. little but with the axes of the wheels a final distance. In heavy work the roughed out or " gashed" before hobbing.TOOTHED GEARING the teeth of such gearing could be drawn. from the cutter spindle by means teeth of the wheel are of positive gearing. 1 is made it is into a cutter is by cutting as a hob. Fig. and when hardened and The wheel blank. thus cutting the teeth in the worm wheel in a very Sometimes the wheel is accurate manner. hob is also mounted in correct relation to the wheel. . which is has been turned to correspond to the outside of the teeth. and the Fig. greater distance apart than the required toward the is The hob is then rotated and at the same time fed worm wheel till the proper distance between the axes reached. or a special hobbing machine. with the proper velocity ratio. caused to rotate simply results are obtained if by the action of the hob. wheel the desired This worm Fig. but process. flutes across the face as in 68. actually making such drawings. 166. 165. teeth having this property of line contact are automatically pro duced by what of tool steel is is known as the hobbing made of the exact form of A worm worm. 397 There is no practical value in and hence constructed. but much better it is driven positively. This known tempered used as a milling cutter.
where a cutter is fed radially inward toward the axis of the worm wheel. The lead pitch. 166. 164 the lead of the singlethreaded worm to the axis. . A. 165. shown is the distance. . wheel. and is equal to the circumferential If the worm were doublethreaded the pitch of the worm wheel. * Figs." t See "Worm and Spiral Gearing. thus obtaining contact on several teeth at the same time. In the case of a doublethreaded twice the pitch. a worm would and so on. and in general worm if the tooth would be moved N be the number of teeth in the worm wheel.t 183. to a corresponding point on the section B. 167* shows a used where the wheel is sometimes rotated by hand or when the projecting teeth are undesirable. Velocity Ratio of Worm is Gearing. angular 01 wormwheel n . fixes the radius of the worm wheel for worm may then number of teeth. be three times the makes one revolution. . In the Hindley worm the pitch line of the worm is curved to coincide with the pitch line of the wheel. but are usually cut by the approximate method shown in Fig. If a singlethreaded worm tooth of the worm wheel is moved a of the triplethreaded distance equal to the pitch. then. and 167 are reproduced from Browne & Sharpe's "Treatise on Gearing. . or equal to the distance between corresponding points on A and C. producing what is known as a dropcut wheel. per turn of the worm thread called the lead. : . Fig.  vei. worm wheel which Such wheels may be hobbed. Halsey. is It is to be especially noted that the angular velocity ratio independent of the diameter of the worm." by F. 169. . and n the number angular velocity of worm N — = — of threads on the worm. MACHINE DESIGN shows a has been hobbed. lead would be twice this amount. which must be decided upon by The pitch of the worm a given consideration of the strength of the teeth. and would then be twice the pitch of the worm wheel. 166 . and its form of wheel occasionally mating worm. Evidently a very great velocity ratio tively small is possible with a compara wormwheel. but the radius of the be varied to suit other conditions. parallel from any point on the tooth section A. The axial advance Thus in Fig.— 398 Fig.
a socalled angular thread. Fig. 54. M. equation 13 strictly that Article would apply. is However. Efficiency of 399 general expressions Worm Gearing. Lewis' curves for to be noted. small. 52. This dotted curve was plotted * Trans. apply also to wormgearing. page 297.d Fig. a value of // = 0. summarized also of his They show clearly increase of efficiency with increase of thread angle at all velocities..05. thread is. 54 of Chap. Fig. S. E. assuming that the coefficient of friction at the thrust collar is the same as at the tooth. 167. to this subject is the One of the most valuable contributions experimental work of Mr. deduced Since the in Art. and of equations the square thread. =. 168. worm (a) of usually. the inclination and (10) of that article. the coefficient of friction. VII. VII.* Fig. Wilfred Lewis. A. is of the face of worm threads so small that the error introduced in using the simpler (9) which were deduced from These equations show that the efficiency of all screw gears is a function of the angle which the thread makes with a plane perpendicular to the axis. . 170 have The the full lines in been plotted from the diagram on which he has results. Vol.TOOTHED GEARING 184. They (2) of show a remarkable agreement with the theoretical equations Art. 169. The for the efficiency of screws. The its dotted curve is reproduced from curve is Fig. and close agreement with Mr.
Lewis' value of p. Lewis' calculated average value velocity of 20 feet per a minute (5) is 0. Halsey \ has examined the design * Velocity here of a number of success* between worm means velocity of rubbing at the point of contact and worm wheel.z ^^ ^^ ^ ^"^ s7 ''^^^ ? M/ ? 6 y £ ^ 3 / # ^Z J ? y i 40 . 52 may.7 ^ «y ^ "\T ^ ^ ^"i^/ ^^ 7 J! I §60 4r 50 ^. Mr. _3tt5o 90 ^•~~ 1^.' / j§£. .^ J1 ' ^ <s^ ^^ ^^ ' ***~'z5r \c\— ~~z~"~'~' /.— MACHINE DESIGN of this coefficient for 400 Mr. and may be used.014 (which would be obtained only at 1. Mr. for these velocities * (200 feet per minute) ranged from 0.— M'^tZr" ' — Z ' " 7^^^^^^4W^^J'^^ =:S5 iZ . as they were intended. screws. therefore. 54. will coincide very closely with curve aoo onft Fig. his average value being 0. with a value of \i = 0.""'^" "**" •"'' **'*' ^"^Cg= ~\j «'"'* 1 ""'^ — — *~" — — "~~~ —— = 70 .^ Z ^^ Z ^^ ^ ^ / 2 v ^ .^  ^. high speeds).074." page 38. 170.f / 30 10 15 20 25 30 35 40 45 Lead Angle in Degrees Fig. 170.015. > ' ^ .026 to 0. Art.059 an(^ ^ or IO ^ eet P er m hiute 0. This coincidence is closer than might be expected from the nature of the problem and the assumptions on which equation (9) is based.^ Jtf ^. f See "Worm and Spiral Gearing.02. be taken as supplementary to those in Fig. 170. Curves (4) and in Fig. for designing slowmoving and poorly lubricated A theoretical curve plotted from equation (9).
gearing. with the small amount of surface contact obtainable in screw gearing. be classified as imperfectly lubricated surfaces. limit. the thrust at being taken worm by a loose brass steel washer running between two hardened and ground (see Art. on the other hand. what may per be expected with sliding surfaces: experimental results going to at show that the lowest coefficient was obtained about 200 fixes feet minute. James Christie. It is which still further corroborates the general theory given. while the side thrust on the wheel is it is increased. although running in an oil bath.7 TOOTHED GEARING ful 401 and unsuccessful worms used for transmitting power and found that every worm among those examined whose lead angle was greater than 12 — 30' was successful. to create a true oil film so that the load would It be fluidborne (Art. is little to be gained in using a pitch angle above 30 not to be understood that the increase in efficiency being is very small. oil. minute as the point of maximum efficiency of worm which is accord with the general theory of lubrication. In Mr. in a general all way. Limiting Pressures and Velocities in Worm is Gearing. 200 feet per gearing. and every worm whose lead angle was less than o° was unsuccessful. many cases. effect The and of the velocity of rubbing on the coefficient of friction of imperfectly lubricated surfaces. Mr. 32. The surfaces of worm must. was noted in Art. Lewis' experiments the the end of the worms ran shaft in a bath of and the efficiencies given include journal friction. who has had considerable experience with this form of gearing. and where the velocity in which worms of less pitch are not only effective but neces sary. was stated in the last two articles that the best results are obtained from worm 26 gearing when the rubbing velocity about 200 feet . with a lead angle less than 9 . washers 104). from the nature of the contact. Lewis. An increase of velocity may. 185. Fig. it is up to a certain decrease the coefficient of friction. as giving 1 — 15' as the lower limit for successful design. 33). 17 of that article indicates. not is primarily for power transmission. as the in perfect result of his work. to be noted. but not possible at any speed. and quotes Mr. It never proper to design a for there are worm low. that there .
a constant to be determined by experiment (see also Art. however.500.700 feet per minute. worm made concave & of this much form due. and necessitate a much higher velocity than 200 feet at the pitch line. i. 64) . It is not always possible. to keep the design within these limits.e. however. it Thus in order to obtain mechanical advantage (see Art.000 . (see Art. Mechanical Engineers (British).720. the law undoubtedly being complex varies. corresponding to K= This great gain is * " Machine Design" page 301. therefore.000. It would appear. which give an average allowBach and Roser experimenting with able value of 690. may be necessary kinematic worm requirements may to use a with a very small lead angle. he has succeeded in carrying 25 tons at a velocity of 120 feet per minute. 32). Smith and Marx* quote corresponding pressures and velocities. Mr.500 feet per minute. V = the velocity of rubbing in feet per minute. The allowable axial load that may be applied to a worm under experi varying velocities has not been very accurately determined. or the law may be roughly expressed as W V = K. to show that the pressure approximately. softsteel worms and bronze worm wheels. succeeded in carrying a pressure of 800 pounds at a velocity of 1. to take . . made on castiron worms and oil was found that the limiting value of K. where cutting began. it . greater load may be With improved threads 6.000 for K. WV The above shown that if = 750. (14) discussion has reference to worms as ordinarily con structed with straightsided threads. t Proceedings of Institution of the year 1906. running in an bath. In Lewis' experiments. Robert Bruce f has are the sides of the carried.360. . where W = the axial load worm on the worm.. for velocities worm it will be up to 1. Enough mental work has been done.000. and K= 98) . attributed to Stribeck. was about 1. .000. obtained with hardened steel worm and bronze worm wheel running in an oil bath. that for average conditions and bath lubrication of the safe. inversely with the velocity.402 MACHINE DESIGN per minute and the lead angle not less than 12 — 30'. wheels. page 57 of . which gives K = 1.
to design the wheel teeth alone. The velocity ratio the worm shaft . to be 20 to is Design a worm gear to connect two shafts which are 11 inches apart. The pitch diameter of the worm. obtain a high lead angle with a singlethreaded worm without making a very large thread. In the case of roughcast. circumferential pitch be taken as one inch. the lead of the If the worm thread will be three inches. but in hobbed it wheel. which is the It greatest on the wormwheel teeth. between the mating convex and concave surfaces of the teeth. carried by a single assume that the load is distributed between teeth. The be will solution of problems in worm gearing must. is 1. hence the tangent of the lead angle = = n X 2. double. to 1.M. will be 2. The corrected circumference of the wheel will then be 60". therefore. depending on the number of teeth in the must be assumed tooth. the wear is made of a harder material. angle not to be less than 15 is to make 320 R. If the velocity ratio is to be 20 the wormwheel have 20.9 0.33. tentative. 186. gearing it is safe to two. therefore a trial assumption will be triplethreaded worm. or even three.. the In general.P. is is and where the worm usual case. 173.11". and can therefore be easily cut in a lathe. generally. . which is an efficient angle.TOOTHED GEARING 403 without doubt. the strength of worm exceeds the strength of the teeth in the wormwheel. and the circumference will therefore be 63 inches approximately. with the given distance between centres. depending on whether the It is difficult to worm is single. and to transmit 7^ H. corresponding to a pitch diameter of 19. or 60 teeth. or triplethreaded. and 60 trial teeth in the wheel.P. Example.9". 40. made with a Twenty inches trial pitch may be taken as a diameter for the wheel. considering them as simple spur gear teeth as in Art. Design of Worm Gearing. that the entire load is or dropcut teeth. usually sufficient. or the lead angle is 18 — 15'. the lead and the worm wheel is to be cut with a hob. to the improved lubrication obtained by what practically amounts to surface contact.
170. per minute. — 15") x of velocity = 12 7T X 2. 255 X a safe value. therefore the .000 From pounds. at the root at least and each tooth have a face or length equal to the pitch of the worm (see Fig.000 pounds per square inch with inch circular pitch. The torque T. 75 From the diagram. The depth below circular pitch is. 175.P. the allowable stress is 7. or 866 footpounds per revolu worm.404 MACHINE DESIGN of revolutions per The number be 320 minute of the worm wheel will =16.200 tion of the footpounds per minute. the pitch line of a standard tooth of one inch from Table 0. it is found that the efficiency is about 90 per cent. or V= 71 7 (cos 18' 2.3857 inches.k J H. 164).90 = 277. = 1. Hence the velocity of the worm wheel at the pitch line = 12 = worm will 80 feet per minute.0X320 =25$ JJ 0. From curve (1). 158.000 which by equation (14) high.650 inchpounds.95x12 — ft. The total axial thrust on the city of be equals — 00 = 3.100 and axial pressure is on the worm = 790. however. r The product 3. hence the horsepower which must be supplied to furnish j. hence the tooth has an excess of strength to provide against the wear. which falls heaviest on the worm wheel.4H. Fig.P.0X320 X ~ —. 0. The velo rubbing the length of one turn of the worm thread multiplied by the number of revolutions per minute. Fig. Hence the load per inch of face of tooth = — 2 X2 = .100 pounds. 560 pounds.75". it is seen that this load corresponds 1 to a fibre stress of about 5. or say 2. although = somewhat The load may be will considered as distributed between two teeth. seen that for the velocity. 80 feet. will be T = XXV. which must be applied to the wormwheel shaft. at the wormwheel shaft will be ' 7 • 5 = 8. it is equation (10) of Art.
TOOTHED GEARING diameter of the worm at the root of the 405 thread = 2. 16 X 16 T — = — = = 850 pounds per square inch.3857) p = s . be considered satisfactory worm is to be cut integral with the shaft.9 (2 = 2. . 104 The type of bearing much used. applies in this case. further calculation as to the strength of the shaft which 187. The general discussion in Art. An important frictional worm gearing occurs in the thrust bearing. Thrust Bearings for loss in may be fitted is necessary. Worms. and of late ball bearings have shown in Fig. page 1. X 0. which therefore deserves special attention. 88 is met with considerable success in such places. it fitted is be bored out and over the shaft.13*. If.650 rj 94. therefore. however. the torsional stress . and from equation E. which is very if low. the to The design may.
tends to absorb any thus having is its velocity increased. the variation from normal speed may be limited to onehalf of one per cent. When the work to be done in excess of the energy supply. with a resulting reduction of velocity. therefore. those redistribute energy.CHAPTER XV FLYWHEELS AND PULLEYS 1 88. the necessary weight of the body may be found. whose continuously. 4. —There are two distinct types of flywheels. or less. Thus in engines driving electric generators. as noted in Articles which also act absorb and and 6. When a flywheel is attached to a train of mechanit ism in which the supply of energy varies excess energy. If. then. the wheel tends to furnish the deficiency at the expense of its kinetic energy. the variation as twenty per cent. depending on whether its velocity is decreased or increased. to be must vary in velocity. If the velocity of the is body be changed from v to v 2 the change in kinetic energy work which the body will do. = 32. the work to be done or the energy to be absorbed with a given change in velocity is known. namely. and K 2 be the kinetic energy at a velocity v 2 then the energy delivered or absorbed during a change of velocity is 406 . Capacity of Flywheels. If may be as great W be the weight in pounds of a body moving with a velocity is of v feet per second. while in such machines as punching machines.2. then the kinetic energy in footpounds which the body possesses K = Wv x 2 where g .2. or the energy it will absorb. for if K x be the kinetic energy of the body the when moving with a . effective Flywheels. the allowable amount of variation depending on the conditions of the case. and those as a pulley or band wheel and transmit power sole function is to 2. velocity v t .
. See Church's "Mechanics.2 2) .. (4) (1).K.K = j. = 1 (<» 2 t  » 22 ) .o this case the kinetic energy of the body is W — 2 2 p o> 2 g where is the radius of gyration. the velocities of differ ent points in the axis." Art. In the case of a wheel with arms whose sides it is are parallel. to be noted that the square of the 2 radius of gyration of the arms or p is very nearly equal to Hence total for this case. in general.. and the rim only need be considered. or heavy disc wheels./) .= X Wv? ". (1) If the body is rotating around a fixed axis. and in nearly all cases the effect of the hub may be arms is so small as to be negligible. Hence for rotating bodies equation (1) may be written E=K K X 2 p ^ W 2 («>*  . and equation becomes identical with equation (1) since. ( 2) or since W— — o 2 = /. (3) cases of flywheel design the effect of the neglected. . is A to punching machine is make 30 strokes per minute and * solid punch holes inertia of yi" in diameter in steel plate y 2 " The student should distinguish clearly between the moment of inertia of a body and the moment of an ami. 6 E= In all K . sufficiently accurate to take the When radius such is the case it is mean (2) of the rim R m as the radius of gyration. . as in direct driving of electric generators. page 91. In the case of wheels with it is many heavy arms. .g FLYWHEELS AND PULLEYS 407 *) E = K K. equation (2) or equation (3) is applicable. a weight of the arms EExample K . R «> = v. In the case of a circular disc / = WR 2 . and where inertia effect of the desirable to compute the wheel closely. and w the angular velocity.Wv? 2 = 2 g 2g W (v*v — 2 2 . if W be the weight of the rim and %R W the 2 . or nearly so. 86. the moment t of inertia * of the body. body vary as the distance of these points from the For .(W + K WJ o t 2 K to 2  ..
represent the path of the pin F. at every stroke of the punch. A.M. therefore. M 171 (a). The circumference of the driving * See "Kinematics of Machinery. moves through an angle of 30 while the work of punching is being performed. or of doing the equivalent amount feet per minute. 171. it is known that the efficiency It is required to find the cross section of the flywheel rim. it should be capable of punching a of hole. lower face of the plate the when the pin is at c. and emerges from the When the pin is at d punch is at its lowest position. 171 (a). Barr. The preliminary layout also shows that the diameter of the driving pulley should not exceed 18". work in shearing. The pin. the pin is at b the punch enters the plate. N of the flywheel rim should not exceed 42". and the mean diameter y .* so that the punch moves with harmonic motion. in the crosshead." by John H. The belt speed is to be about 600 type.P. Fig. and. Let the diagram. Let Fig. from existing will not machines of the same exceed 85 per cent. 171 represent the machine under discussion. and has entered the die A " at e the punch is withdrawn from the plate and at /is at its highest position. hence the driving shaft will make 30 X 6 = 180 R. page 184.408 thick. Fig. continuously. When Fig. MACHINE DESIGN Since the machine may be used for shearing also. A preliminary estimate also fixes the ratio of the diameter of the pinion the diameter of the gear B to C as 1 to 6 . The mechanism in the head. . is a slotted crosshead.
every energy cycle. therefore. 4) every stroke of the punch.FLYWHEELS AND PULLEYS pulley 409 = belt speed = 3— = IoO 600 180 3. the diam eter of the driving pulley will be 13* which is well within the (see limit set. 600 feet per minute. and will be equal to the area of the metal in shear multiplied by the shearing resistance. and the time of the cycle ^ of a minute. —T 2) 20 2 The effect centrifugal (8). While the hole being punched the flywheel is giving up energy to assist the belt.85 1740 ft. the belt moves a distance = 1740 or = Afip. be neglected at 0.lbs. would have to be of double As the punch passes through the plate the shearing resistance decreases. The will greatest pressure which the punch must exert on the plate be at the beginning of the punching operation. 131). . hence equation gives/ = an arc of contact of 180 is and = The shearing strength of a plate in punching about equal to its tensile strength.X 11 —= loss. The machine makes an energy is cycle Art. If the had to exert this effort unaided. total work performed in punching 1 . belt orP = 60.= 20 feet T .34 feet and.800 is X . x per cycle. Hence of 1 (7 . The energy delivered by the belt belt per cycle is the 1 product of the difference of the belt Since the speed of the belt is tensions (T — T 2) multiplied by the distance through which the is moves (see Art. small and The work of withdrawing the punch from may be considered as part of the frictional machine is the sheet Since the efficiency of the 85 per cent the belt must supply = 0.T = 87 lbs. pressure may therefore be taken as half the maximum. hence the need of a flywheel.000* = 70. and during the remainder of the cycle sheet. ji force Art.800 XX~X 4 2 it lbs. and the leather 15 inches wide. or every six revolutions of the driving shaft. the belt withdraws the punch from the and restores the wheel to normal speed. 70. * may 131. 1480 is ft. The average until it becomes zero as the punch passes out.6/ for this belt speed.lbs.
the time consumed during Since the pin. during leaving 1750 — = 1600 ft. since revolutions per second and. = 2. the weight of the rim is from equation (1) = W = ^ < ~ v? = 2 2 gE 2 X 32. the 2 mean 7T radius of the wheel =21 inches.2 2 X 1. per inch of width. The mean cir cumference of the rim crosssection = n X 42 = . to be supplied by the flywheel.600 33 ~ 3° i lbs. Fig. = 545 ^s. In the more general case this cannot be done so readily. or 3 The driving shaft makes 180 R. hence v 2 = 33 X 0. therefore.. during the operation of punching. the The 150 belt. 410 0. of machinery where excessive slipping of the belt / is In this class sure to occur.lbs. 24 lbs. should be taken at not more than 40 lbs. The allowable variation in velocity may be taken as 10 per cent.lbs. 132 inches.26 hence the number of cubic inches in the rim will be 0.. neglecting the effect of the hub and arms. .M..6X40 = inches. One cubic inch of cast iron weighs 0. and the width = — = 3K 24 moves through 30 or ^ of a revolution.90 30 feet per second.P. 171 (a). vt = X X 21 X 3 = 33 feet per second.100. MACHINE DESIGN where/ = the effective pull per inch of width of belt and t = tension per inch of width of belt on the tight side.26 = 2.7 X ^q = 1. and methods such as those outlined in Articles 5 and 6 of Chapter of energy to be redistributed. therefore the .. and since the punch makes 30 strokes per minute. or a section 3H inches wide by 5 inches deep be selected. Therefore. the operation of punching will be 12 —X — = —^ of 30 360 ft. a minute. moves 600 operation and supplies 87 X 1..3.7 = 150 ft. .100 132 =16 may square inches. . of the belt whence 7=0. therefore. quite easy to In the above example it is compute the amount from the conditions of the problem..
2 (d) the flywheel redistributes energy on both strokes. or in other cases may be based on perience.000 feet per minute. but the represented by maximum Again in excess of effort is K and not by engines internalcombustion v with K such machines as governors. then it is sufficiently accurate for most work to take v = ]/ (v^ + v 2 ). i. where the diagram representing work to be is superimposed upon that representing the energy supplied. 2 but the true relation between these quantities depends on the manner in which the velocity changes. * See a paper by The I. and the sides of belt speed is to be about 4. 0. to limit the variation in velocity to a definite it is desirable and below the mean velocity. The wheel to also to act as a belt wheel. as t amount both above mean velocity. J. v and in parallel operation of alternating generators.M. M.P. 2$ J feet. XXII. A. Find the weight of a castiron flywheel neces sary to limit the speed of the engine discussed in Art. Vol. E. Therefore the diameter of the wheel = 8 feet.. such as steam engines. . 5 to a total variation of not more than is 0. to take this into account.FLYWHEELS AND PULLEYS II 4II done in must be employed. which of the may be taken without great error as a mean diameter will wheel rim.e.005 above or below the mean the speed. For very exact work. Since the engine the wheel makes 160 R. "hitandmiss" giving a very variable energy supply... page 972. the circumference of = 4.01. Astrom. S.000 — —= 160 . such a manner that the excess and deficiency may be measured. Trans. may be based on the greatest Thus in Fig. If v be the v 2 the maximum and minimum velocities respectively.* it may be necessary Example (2). be parallel and their effect is to The arms are be considered. Care should also be exercised that the solution covers a complete energy cycle in order that the solution excess or deficiency. empirical constants which are the result of ex In some machinery. the design of the flywheel may have to be based on a hypothetical performance of the engine covering a number of successive strokes. since the thickness of the rim face of the wheel should be at least 16 not be great.
2 inches or say 2 X A inches. (W + & X^ a) = &(l*a. E = 4.88 .and 60 R = the Then «> radius = n 4 feet. problem.020 = 3. to obtain the total weight of the Since the rim is to to be 16 inches wide and the mean will be diameter be 06 inches the thickness — 2 720 0. is W = 2. tabulated values of this ratio are deceptive unless they refer to specific conditions. Thus in the punch ing machine the flywheel absorbs and redistributes nearly the entire energy supply per cycle.005^ of o> the <d = = 1. It is evident that the ratio of the energy to be absorbed.*\ 1. to the total energy supplied per energy cycle. each weighing 170 pounds. weight of reciprocating It is to parts. 0. It is readily seen that in the steam engine this ratio will vary with the point of cutoff.620 footpounds. Practical Coefficients. 2 "2sr 2 ) Therefore W + JFa = 2X32.740 lbs. ) = 6. = —. / 189. = 2 71 = a> 2 n X 160 7— = ^ 16.7 — A 412 MACHINE DESIGN preliminary inches wide to give the necessary width of belt. while in the steamengine example the amount absorbed is about onethird the total.720. 88 T 16.020 3 to = 2gE „ # 2 K .995 w = 2 x and hence from the conditions 1. and also in the same machine under different conditions. ?/ which must be added the weight of the hub wheel. steam pressure. or 1..020..1 3 = 2.26x^X96x16 = 2. from Art. and therefore.005 X 2 I0^8 = 2 16. layout gives 6 arms.1 From Equation or Ty (4). and 2 = = I ^7 Therefore (e^ 2  o> 2 2 ) = (16.8 radians. The latter . will vary in different machines.720 + 1. W & = 170 X6= Let n = the revolutions per second 5. etc. 60 Now.620 7 7 16x6.2X4. in general. the weight of the flywheel is be noted that directly proportional to the energy to be absorbed and inversely proportional to (v* — v 2 2 ).
etc engines driving pumps. which is undesirable because and also because heavy wheels bring great loads on losses. 5. saw mills engines driving machine tools. Stresses in Flywheels. from the nature of their requirements.FLYWHEELS AND PULLEYS is 413 is usually a small quantity and. causing frictional For this reason it is always desirable so be absorbed.016 0.02 o . be treated with reference to the allowable variation per pole. TABLE XXIX Values of v 0.025 0.01 o . if E large the weight of the flywheel of the cost.20 0.05 0. crushers. in general. therefore. and experience has shown what the limits in variation of to the character of the work to be done.0003 190..003 to o ..03 0. (or increased) K may be greatly decreased positions of the crank pins. The 2 following limiting represent average values of the proportionate variation practice. For engines driving spinning mills for fine thread For engines driving single dynamos For engines driving alternators in parallel 0. therefore.007 to to 0.025 to 0. — The particular case of direct driving of alternating generators in parallel must. the number of poles is great the total allowable variation is corre spondingly small. the bearings. the desirability of obtaining .. weaving and paper mills 0. very high. It is classes of velocity may be for successful operation. to arrange the sequence of events in the energy supply and work to be done as to minimize the excess energy to This is illustrated in Art.03 to For engines driving spinning mills for coarse thread . 6. is wheel to act as a band wheel. may be excessive. where the area by changing the relative is This procedure closely restricted. and when. Fig.10 to For For For For punching machines and similar machines engines driving stamps. is of great import ance to avoid wheels of great weight in large steam engines when variation in velocity must be The allowable variation in velocity fixed with reference evident that some work require much more constant velocity than others. The velocity of the rims of all flywheels If the is.15 o .
the results are usually very disastrous. the tion. 78) as in a boiler shell acted .000 feet per minute. however. the centrifugal action being balanced by the stress in the arms. other hand. makes high speed very desirable. to understand fully the general character of the stresses even their when no accurate computations can be made is as to magnitude. mathematical analysis of the stresses in fly wheels and pulleys is not satisfactory or conclusive. tant. Great care should be used in the design of such wheels.may exist. Fig. Unfortunately. In the case stresses of small wheels cast in one piece. at normal speed is exceedingly dangerous to life and and when such wheels "explode" or break from overspeed ing. alone. free to Consider that the rim of the wheel in Fig. capacity of a given wheel velocity and. 172 radially. on by an internal pressure tensile stress the rim expanding until the induced in any section A A. stress will exist in the rim. which render useless any refined elastic theory of calculations. in exactly the same manner (see Art. the presence of joints vitiates any calculations based on the and when the parts are of cast materials and of large sectional area. In large wheels built up of sections. on the hub and rim. tendency of the wheel to separate along that section. is centrifugal pull it due to their is clear that own mass and that they may be placed so and practically no close together that the rim cannot exuand. of great unknown shrinkage magnitude . 134) brings the peripheral velocity up to It has been shown that the 4. . 172. for a flywheel which breaks limb. there is no assurance that the character of the material is uniform throughout. under the action of the of the rim. or the limiting of the external dimensions. balances the If. therefore. inelastic that their stretch.414 MACHINE DESIGN high belt speed (Art. It is importhe strength of materials. is proportional to the square of the wheel is its when to act as a flywheel economy in the use of material. If the is expand arms exerting no restraining force wheel be rotated on its in a radial direc axis the action of centrifugal force such as to cause an outward pressure on every part of the rim. the arms are rigidly attached to both and are so negligible.000 or 5.
depend upon the amount which the arms stretch. Here. and a bending moment of half the maximum occurring at the centre p^ig. so that a condition re shown. of the span. 172. no bending action while if they are so inelastic as completely to restrain the rim. The If relative values of the hoop tension and bending evidently. the somewhat than that in the free ring. so that the rim expands will freely. but in the most usual case the arms stretch a certain amount and are not placed sults similar to that close together. the greatest bending moment being at the arm. The maximum tensile stress will be the sum of the hoop tension and the tensile stress due to the bending action. With any intermediate amount arms the rim will be held in equilibrium. do not and. The hoop tension section of rim between each pair of arms is so long that it becomes a beam fixed at the ends and loaded uniformly by the unbalanced centrifugal action.FLYWHEELS AND PULLEYS 415 Flywheels approximating both of these conditions are sometimes built. Fig. 173. but the full centrifugal force be applied to of stretch of the bend the rim. stress will. will occur. though somewhat. partly by the hoop tenion and partly by the restraining action of the . in an exaggerated manner. in Fig. stretching 173. no hoop tension will be induced. stretch enough to allow the is rim to expand less freely. therefore. the arms. they should stretch enough owing to their own centrifugal force.
k) °' * Trans. If the rim is of a wide thin section. Consider a section of the rim one inch wide on the face. D = 12 wv g 2 = v — 2 . A. therefore. the length of the rim between arms in inches. the arm. . mean radius of the rim in feet. hoop tension for instance. and the bending be proportional to threequarters of the centrifugal exists stress will force of the rim. 78. the bending stress may be very Professor Lanza* has shown that. 173. p = x 12 w t v 2 ztRg =. no bending exists. and has a on the arms as at B. M. and will of course vary with the relative and shape of the rim and the arms. be noted that the wheel is to act as a wide thin rim. S.6 — MACHINE DESIGN 41 arms. page 208. and increases the bending on the forward D = R = Let / = Let v — Let w = = Let Let Let I the the mean diameter of the rim in feet. ordinarily used. the hoop tension will be onequarter that due to free expansion. with the proportions serious.. and the unit stress in the section. further distorts the rim side. centrifugal force per unit of length this section is c (1"). . by Art. the arms stretch one quarter the amount necessary for free expansion. the bending action still band wheel. Vol. of w ti? = — Rg — and. the weight of the material per cubic inch. the total load ' which tends to separate such a ring along a diameter if is wtv — 2 X is 1 2D. E. and of the is bending stress caused thereby. Since the expansion of the rim the directly proportional to the stretch of arms to it is clear that the If. therefore. and the arms are few. for iron wheels y ' . XVI. is also directly propor tional the stretch. Fig. the latter being a measure of the unbalanced centrifugal force of the rim. the thickness of the rim in inches. theoretically. 10' nearly. stretches about threeIt is also to quarters the amount necessary if for free expansion. the velocity of the rim in feet per second. The mathematical relation which between these stresses size is complex. The circumferentially.
now the stretch of the arms be taken as threequarters that necessary for free expansion of the rim. The stress due to this bending moment when no hoop tension exists is therefore. and n = 6. but implicit faith must not be for the reasons given in the first paragraph of similar this and all results obtained from this or formula? should be checked by successful practice wherever a doubt arises. Professor Lanza finds the stress due hoop tension = 575 and the stress due to bending 5. The above placed upon article.060 or p the total stress (8) 1. the total unit tension in the rim will be #** + *#.635.81 which agrees quite closely. it equation may be used for checking.FLYWHEELS AND PULLEYS The maximum bending moment arms. The equation does.600. I is tit should be noted that this the entire crosssection. *where If e is Me — ~S7 cPe If the (6) the distance to the outer fibre. 2 For the same 4. however. in 417 the rim occurs at the and its value is M = cl2 * 12 considering ° the rim as a straight beam.. * Sec Table case 17. for a unit (1") of width of rim and not for 27 . show clearly that in wheels having I. (^ + ^7) if (iv 1 cPe\ 12 • • • (7) n be the number of arms in the wheel. to D = 4 ft. v = &&. and moment of inertia of the crosssection of the elementary ring. = 5. roughly. n and if the crosssection of the rim be rectangular (7)  = 5 whence equation reduces to For / = — 16 = = . or a total data equation stress gives j^Pi = 581 and %p = p 5. the allowable stress in flywheel rims. * / = nD .
and that part of the rim which ently supports. that this speed so. Example. is It is to safe only because experience has is shown it to be and not. I. per sq. then v = 6. and care should be exercised at high speed. flywheel is is being accelerated from off. arms may be fixed at upon to carry the full torque load.000 (a factor of safety is = 20). cordingly (5) when such wheels must run Equation often taken as a basis for the design of flywheels. the bending stress much greater than ac that due is to hoop tension.. when the energy supply suddenly cut called as it may be in a steam engine. and since the bending is stress in the simple cantilever twice that of a cantilever with the free end . If therewith.8 41 MACHINE DESIGN is thin rims. based on equation (5) . . beam If the one end.2. arms stretch threequarters the amount necessary for free expansion 2.000 feet per minute.000 66. Compute the stress in the rim of the castiron assuming that the flywheel discussed in example (2) of Art. the arm is sub jected to a tensile stress due to the centrifugal action of its it own is weight. in equation (5) p l be taken as 1. so that appar equation to M (Table VI) applies.025 = 1. speed for ordinary cast iron wheels. as will be seen. and this found to be a safe peripheral be noted. or few arms. however. or square inch. thin and flexible.025) \ J = (8 The stress. from (8). Here n = v2 6. would be 444 pounds per rest. free but guided and rying a concentrated load at the free end (see Table case rim is. arm of a wheel with a very rim approximates a cantilever at the other. The direct stress difficult compute. p = 3 (— + 2 0. 188. in. of the rim. t = D= 8 and v = — —= 60 4. because the stress necessarily as low as 1.6.000 pounds. a large factor of safety to cover uncertainties. the arms approximate a simple cantilever loaded at the free end. stiff Each car7).668 lbs. if — + 2 0. When the a. using. In addition. however.
it 419 is considered sufficiently accurate to compute the arm as a simple cantilever Let Let P = a = = and neglect the direct stress. and which have burst have been successfully re placed with wheels having rims Difficulties made of wood. is misleading as. the greatest force due to the belt pull arms.64 a ™ A w The allowable is. in general. Small flywheels and band wheels are usually cast in one piece. Construction of Wheels. page 618. may have great excess of torsional strength. minute. or and in extreme cases wheels made of steel wirewound wheels have been constructed. cast in two parts to and the diameter of wheels about twenty feet. and a lower value is sometimes desirable. plates. 191.. or * Trans. stress p. Wheels from about sixare usually teen feet in diameter upward made in several sections. for cast iron. the length of the arm. E. may be The allowed should not exceed 2. strength divided by the weight per cubic unit therefore. XIII. a measure of the value of the material for castiron wheels this purpose.: FLYWHEELS AND PULLEYS guided. For higher velocities steel castings are used. M. The shaft of a steam engine may have to be very large to avoid excessive deflection and. are usually below 5. or the stress ( 9) moment of inertia I. shafts are designed for stiffness. at the rim. S. on account of the uncertainties of the case.000 pounds per square inch. and not for torsional strength. . for velocities Flywheels and band wheels. the number of Then from J (Table VI) Let n Pae ^~nf from which the determined. A. The statement sometimes made that the arms should be as strong against bending stress as the shaft is against torsional stress. as a consequence.000 of feet per made of cast iron on account low cost. Vol. For this reason some woods are superior to cast iron for wheel rims. written v Equation unit tensile (5) may be = 1.* the diameter of in transportation limit wheels cast in one piece to about ten feet.
particularly balance weights are attached to the rim between the arms.000 feet per minute are necessary. and the spokes are of steel. If. Here the rim and hub are of cast iron. with perfect success. properly proportioned. Band wheels to of wroughtsteel construction feet in can now be obtained up strong. latter cases unknown shrinkage most probably These shrinkage stresses are sometimes relieved by casting the hub in several pieces. each piece being cast integral with one or more arms. fortunately. If joints exist in the rim. is practically The spokes are placed close together so that there no bending of the rim. and the rim is also prevented from expanding freely. The openings between the parts are afterward filled with lead. under heavy service. Where speeds above 6. and the metal poured around them.000 feet per minute which. 174 are sometimes built. they are light and and are rapidly coming into favor. stresses will In either of the exist. force. Fig. so that on cooling they are gripped very firmly. thus increasing the centrifugal bending Fig. Wheels of this construction are used for large band saws at velocities above 10. their relative strength must be considered. The spokes are placed in the mould. wheels such as shown in Fig. each cast in one piece. the wheel has a very wide thin rim if it cannot be considered safe at this speed. 174. are safe up to 6. 175.000 feet per minute. about 4 diameter. when in place. however. and rings are shrunk upon the hub to hold the parts Experience shows that solid castiron wheels. is also about the limit of efficient belt speed. .420 MACHINE DESIGN in made two parts for convenience in erecting.
. of the flanges which add to the strength. but leave it free to expand. Fig. Fig. 176. The joints in the rim are simple flange joints. length of the span it Where the joint is should be about onequarter the the arm. the arm being bolted to each segment. The arms are secured in the hub by means on one side may be movable axially so as more firmly to clamp the arms. S. where. 176. XX. The best location is at mass without contributing to the the arm. 174 and 175 lies wheels which correspond closely to the limiting types discussed in Art. that shrinkage only. ter not a factor are those due to centrifugal force and the arms are simple illustrate Wheels of this charac have been used with success in rollingmill work. built thus. Fig. placed midway between This is the most dangerous location possible. Fig. on the arms. away from by the theory of elasticity. A. account of the added bending effect due to the centrifugal force between these types. 176 illustrates a the Fig. The stresses in the is rim when cast in several pieces so cantilevers. and many wheels are well. Vol. Occasionally links are also * Trans. E.1 FLYWHEELS AND PULLEYS 42 In Fig. The arms do not constrain the rim radially. The hub may be solid or the flange construction is very common. page 944.* Figs. 175 the rim is cast separately in one or more pieces. This of fitted bolts.. the bending moment is zero. and the seg ments themselves bolted together as placed between the arms. The construction of most wheels band wheel with arms and hub cast in one piece and the rim in sections. 190. M. 177. The segments are held together at the rim by means of links of rectangular crosssection shrunk in place. 177 shows a heavy flywheel with an arm and a segment of the rim cast together. as at A.
. 179. E. A much is better arrangement is shown It in Fig. A. are fastened to The segments is the arms. K. 178 are joined bolts. Wheels with thin. Mansfield has pointed out (Trans. Fig. 180. 63). so that the lever arm. where an arm placed on each side of the joint. may be noted that thin rims are often stiffened trie by light circumferential ribs at outer edges. Fig. 180. sometimes shrunk in place. 177. simple and the machining easier than with flanged connections. S. 181. Art. a shall be as great as possible compared to the arm b (see C= Hib' m Fig. . This is par ticularly applicable to wheels cast in two parts. A. and the bolts should be placed as near the rim as possible." at held together by Theaded links. 178. When such joints are used they should be well ribbed for stiffness. shrunk into recesses on the outer face of the wheel. which is them by through hub is This construction that in Fig. Mr. wide sections are almost always joined by Fig. The construction of the similar to Fig. Fig.422 MACHINE DESIGN In Fig. XX) that these ribs maybe a source of weakness. 182. the segments are called "prisoners. as indicated. It is evident that the wheels is most important. 181. M. Vol. flanges as shown in manner of joining segments in builtup Wheels seldom fail at the hub.
. Fig. midway between arms end of arms 25 joint. the relative strength of the joint com pared to the solid rim will vary with the exact proportions selected. If the rim be made Ishaped. 177. at Linked Linked joint as in Fig. 178 the slot can extend entirely across the section and the link can be made as wide as it is the rim itself. 177 limited. A S. * See Trans. 178 accurately. A rim having such ribs is not necessarily as strong against bending in this direction. Feb. latter type velocities.00 . ribs are used the section The prisoner link modulus should be calculated. obviously. which is difficult do with the construction in Fig. This permits of greater accuracy in induced in the link by shrinking 77. While. page 944. E. . the initial stress importance of which has been noted in Art. a wide rim unknown shrinkage strain may not be as safe as a narrow if deep rim of the same sectional area joint. Furthermore. as one and when of rectangular crosssection having the same area . The rim is made of laminations held to the spider by dovetails. the links can be so pro portioned that the joint will be as strong as the rim proper or even stronger. M. 4. wheels built up of steel plates. made of are now plates fastened to a central spider made used. 178 has certain advantages over It is the link shown in Fig. held together by a good or For extreme wheels with rims of steel castings. bolted.* as in Fig. 177 50 60 65 1 1 joint as in Fig. 178 Fig. shown in Fig. 179 shows a flywheel of the used in rollingmill work (see Power. while in Fig. average practice gives the following apparent values: Flanged Flanged joint. XX. possible to machine both wheel to and link in Fig. the above that a thin solid rim with necessarily the best. it computing the in place. is evident that the depth of the recesses in Fig. 00 It must is not be inferred from as. 1908). Vol. 177. 182 Linked joint as in Solid rim .FLYWHEELS AND PULLEYS The are 423 greatest bending moment is near the arm where these ribs on the tension side of the beam. evidently. bolted. 182.
" by C. but do not add to the strength of the rim. they are very valuable as indicat ing the manner in which various types of wheels fail. 176. The best experimental data upon the strength of flywheels are from tests conducted S. flange joints. teeth of gear wheels constitute such a load.* While made to determine the bursting speed of small castiron wheels only. Due al lowance should be made in such cases. is The large. of various joints The strength was found to be about as tabulated in Art. Balance weights. to and large castings These experiments go show that solid cast wheels will burst at a peripheral velocity somewhere near 400 feet per second. H. and other periodicals. Vols. Fig. due allowance must be made for the difference in quality between the metal of small in estimating probable bursting stresses. Being conducted on small wheels.424 as MACHINE DESIGN shown. steam turbine rotors. the laminations being assembled with overlapping joints. the magazine Power. XX and XXIII. the wheel and the peripheral speed high. The rotors of some forms of electric generators. placed between the arms. especially this should be considered. E. the velocity of the wheel rim is per second. 250 feet by means of a of through bolts. by Professor Benjamin and reported to the A. *See Trans. and such wheels are safe only Rim joints at a velocity of not more than 100 feet per second. Such loads add to the centrifugal force acting on the rim. be found in the Transactions of the A. and similar revolving members are often loaded as shown at W. common the arms.. Experiments on Flywheels. See also "Machine Design. 191. M. Wheels for great speed have also been constructed by winding the rim with many turns of steel wire. Benjamin. . and throw no light on the increase of stress with an increase of speed. particularly or loads are placed near a joint as if the load shown and in Fig. 192.. if 176. when the rim is thin and the velocity high. particularly the midway between were found to reduce the strength materially. Heavy outside plates clamp the whole structure together In the particular case noted above. should be carefully considered. these experiments were M. to Descriptions number of examples of such wheels are S. E.
corroborat and not altogether of these stresses is satisfactory. (J J 3) 4) and * Sec = 0. are complicated Experimental data. on account of the added bending If the radial 193. at induced are a tangential tension. the difference being analogous to that existing between thick and thin cylinders.000003 55 p' wN 2 [ (3 4 /•) r2 (3 2  (1 X) ( + r2 2 3 . p = the radial stress at any radius r. When a radial a disc of uniform thickness is its axis. the preceding articles do not apply. and every point in the disc. r = the radius at any point.ooooo 3 5 5 T^Y [(3 + A)(r 2 2 Hr 2 1 + ^)(i+3/)r](ii) 2 1 . and only enough will be inserted to show the general character of the rapidly rotated on problem.00000355 solid vN 2 [(3 + /) (r + r 2 2 ^L r ) J 2 (12) For a disc p = 0. pages 192 and 204.00000355 wN2 [ + ~ y2 ) 1 • ( "Theory of the Steam Turbine. having a hole at the centre of radius r lt /> = o. were found effect. Rotating Discs. Jude. = Poisson's ratio = ]4." by A. p = the tangential stress at any radius r. . The notation and units have been changed to correspond with those used in this text. for steel and Y± for cast * N = revolutions per minute. are also lacking. to be very dangerous. Then it can be shown* that for a flat disc of uniform thickness. Mathematical with those analysis of the stresses in a rotating disc.FLYWHEELS AND PULLEYS 425 Extra loads. A full mathematical treatment beyond the scope of this treatise. and p' = 0. the principal stresses stress. = the inner radius of the disc in inches. w — weight of one cubic inch of the material. Let " " " " " " " = the outer radius of the disc in inches. depth of a wheel rim be equations deduced in the great compared to its axial width.) ' 2 l • . such as balance weights located between the arms. in existing in thick cylinders common under internal pressure. ' s iron. ing the theories. r2 r.
appears that the tangential stress p = 2 X 0. however. r^ the stress 2 ) p = 0. A Determine the tangential r2 also at the hole. N= in (11) 500. thus indicating the shape for which discs should have uniform strength.643 lb s P er m  The foregoing equations. to increase the stresses greatly.426 It is to MACHINE DESIGN be noted that the radial stress is less at any point than of at the corresponding tangential stress.28. in. Example. and must also be applied with caution to brittle materials.00000355 (1 X 0. = 40. are deduced on the is hypothesis that the material It is clear that perfectly elastic and homogeneous. and rx = 2. by making r '— in equation The even a very small hole at the centre of a rotating disc therefore. and an examination is equation (n) shows that this tangential stress a maximum and a minimum At the surface of the bore or where r = the surface of the bore at the outer periphery. w= r 0. (11) to (14).M. They are of great value. for a brief reflection will show that such discs must be thickened at the centre to reduce the stress at that . they cannot be intelligently applied to builtup wheels of the disc type. per sq.00000355 w N 2 [ (3 + A) r 2 \ which (13). A circular steel saw % inch in thickness and 80 inches in diameter has a hole 4 inches in diameter in the centre stress at and runs at the rim and Here rate of 500 R.P. in showing the general character of these stresses and the location of the greatest stress.00000355 (1 X + 0. 2 [ lbs. is just twice that obtained effect of is. = %. Whence rim is making = r2 = 500 40 the tangential stress at the p = 0.28 2 ] X 500 (3  + K) SCL (4o 2 + 2 2 + 40 2 )  + 1) 2 = 2.28 1) X 2 ] 2 [ (3 + K) 2 (40 + 2 2 + 2 2 )  40 = = 535 2.00000355 rx oj N 2 [ (3 + A) ( 2 r 2 + 2 r  (1 + 3 it A) r 2 ] If now be taken so small that r is is negligible. and p at the hole making r = rx = 0.
Stodola.: FLYWHEELS AND PULLEYS point. French." by Dr. 427 shapes. Jude. forged down from much thicker ingots give the safest construction. and working the material are not Rolled sheets good for is very high speeds on account of their seamy Discs structure. "Steam Turbines." by A. "The Steam Turbine. References of the Steam Turbine. and cast ma brittle structure must be of firstclass quality. terials of which conducive to incipient cracks. reference turbine. A." by L. It is For complete mathematical analysis of discs of different may be made to the various works on the steam evident that great care should be used in selecting for highspeed discs. "The Theory .
the stresses If the character. by applying those fundamental fit formulae of Table VI. very members are. must be governed largely by judgment and experience. Since machine frames must. the predominating requirement For these reasons for the frame is stiffness and not strength. as closely as possible. If it is essential that the moving members be held in accurate alignment. as in the case of machine tools. of every force acting in the section can be static determined by applying the fundamental requirements for equilibrium of the section. the stresses in certain important sections. sum sum of all vertical component forces must = (c) The algebraic of all the 428 moments must = o . but sometimes necessary to check. and in determining the forms of the various sections. in general. even in cases matical analysis is possible being rare. in general. and line of action upon a given section are known. receive the reactions from the forces applied to the various moving members by the energy transmitted. page 94. Stresses in Machine Frames.— CHAPTER XVI MACHINE FRAMES AND ATTACHMENTS 194. in most complex and beyond mathematical analysis. which apparently stances. value. where judgment must be the guide. it is obvious that the stresses induced in frame cases. algebraic o. the cases where complete matheHowever. what may be termed a ^qualitative is analysis" of the frame very desirable as a guide in properly distributing the material. the design of machine frames. the circum In all cases. namely (a) : The The algebraic sum of all horizontal component forces must (b) = o. it is not only helpful.
a mathematical analysis can be made with some degree of completeness. will be the algebraic stresses acting in that direction. one of the few types where Fig. outlined as Suppose the frame be shown so that the dimensions of the crosssection at any place maybe assigned. DE. at the sections Evidently. due punching. In the case of a punchingmachine frame as illustrated in Fig. Fig. in general. but the principles will be illustrated It is to be noted that it is by applying them to typical cases. found by applying classification of and (c). but. and the . whose gravity axis is at 0„ consider the portion of the frame above BC as a free body. and FG. 183.section BC. 183 illustrates a type of frame which It is is quite common and known as an open frame. at any point. great stiffness is not essential and the design may to be based on the strength required. Fig. and then checked for strength or stiffness. if the stresses are checked BC. 184. seldom possible to find the required dimensions of a section. In the . by solving the particular equations from Table VI which apply.MACHINE FRAMES AND ATTACHMENTS The stress. It is impossible to of make a machine frames that would be any service. the strength of the frame will be fully determined. sum of all the in 429 any direction. to It is in equilibrium under the action of the exterior force P. as (a). directly. {b). the section must be assumed from the conditions given. at that point. 183.
19.43° internal forces exerted MACHINE DESIGN upon it by the lower half of the frame. The vertical force P must be balanced by an equal and opposite force at the section BC. Table VI). There are no horizontal forces. This therefore. may be combined with the direct stress Pj. is Consider next the section DE. whose gravity axis at 2. or p2 = P ae — — where is e is the distance from O t to the outer fibre and 7 X axis Oi. causing a resisting resisting tension at B. which tends to rotate the upper part of the frame around 1} the gravity axis of the section. to find the maximum tensile or tensile and compression to this compression stress. The resultant of this tensile pull. = —+ . as determined by equation J. Table VI. a case of combined flexure and which is fully discussed in Art. direct stress. There are no horizontal forces and the DE to be a free vertical force P must be balanced by a vertical tensile pull upon the upper part of the frame by the lower part. the section The due only that due. Chapter III. may This acting at the centre of gravity. which induces a tensile stress uniformly distributed over the section. as before. is The maximum intensity of these flexural stresses given by the fundamental equation for flexure in beams (see J. where A is is the area of the section. The only moment acting on the part Pa. due to the action of P. to P. The moment. the intensity of which is p1 = P — pounds per square j± inch. which is distributed uniformly over the whole section. which is equation M of Table VI. will the moment of inertia of the crosssection The greatest tension will therefore be at B around the and its value be p = p1 + p2 is. and suppose the part of the frame at the left of body. and producing a and is tensile stress at right angles to the section P = 2 2 H parallel to DE and producing a shearing stress along the section. The shearing stress is . stress moment acting upon whose moment arm is a 2 . and a compression at C. be represented by 2 K force may be resolved into the components HK =P x perpen dicular to BE. as in the section BC.
riveter. will result (see Art. and unless the jaws are properly designed they fail to may spring so much that the dies will align properly. 185 illustrates an open frame as applied to a power The of the rivet which is to be "driven" is is placed between the dies D and D v and pressure applied to the movable die D. of balanced as before by the resisting stress moment and the resulting may be computed by equa Table VI. The intensity of this shearing stress A. and that which may result from the elongation of the bolts which hold these members together. — P . there are no hori zontal forces. Evidently these general principles may be applied to any section. and faulty work is. Fig. the essential factor in stiff the design. and the stake C. and the consequent openwill ing up between the frame and the stake. When the riveting pressure P O. Consider last the section As before. be neglected except near the (see Art. the stretching of the bolts. 53). advisable to compute its The moment Pa 3 of the section tion J. 14. . is The yielding which most affects the alignment that due to the bending of the frame B. but the vertical force vertical resisting force P must be balanced by a stress at which induces a shearing is the section. where is A3 is the area of the section. in strong enough. the parts are enough they will be. general. the beams B and C tend to rotate around the point bolts. be applied to the bolts by the force If the P will P — t  —7 nuts on the bolts are set up so that a combined total initial tension somewhat greater than P x is induced in the bolts. smaller than at DE is or BC. Stiffness and not strength for if therefore. this tendency being resisted by the tension in the The be load which may . The pressure which is applied may be very great (150 tons or more). 2 usually small and may FG. ends of the beam as in the section FG Chapter III). is applied. Since the area of the section it is much value. be negligible.1 MACHINE FRAMES AND ATTACHMENTS pa This is 43 = Po p where A A = 2 the area of the section DE. by means power cylinder R.
B. which carries the crosshead It of cast iron. This tendency is by P'. the rotation of the engine to be taken in a clockwise direction as indicated. and may therefore be Fig. approxiuniform strength of length (See Art. 186 illustrates a closed frame as applied to a vertical steam engine. The stake. C. 186.) The intensity of stress in the bolts should not exceed 6. and Case 1 of Table which occurs at D may. the combined thrust on all the columns. be computed and the maximum stress which occurs at E F may be checked by Equation J of Table VI. guide is is The back column.) Fig. It may be reasonably assumed that the back column the cylinder . B. 77. the vertical component of which must equal P. 43 and also Art. required to check the stresses in these columns when the piston is ascending and also when it is descending. the steam pressure tends to and bed closer together. J. When draw resisted the piston is ascending.000 pounds per square inch. C. are of steel. The upper II.) part of the frame. therefore. while the front columns. the total steam pressure on the piston. approximates a cantilever of a. Table I. and Equation Table VI.43 2 (See Art. treated in a similar manner. 15 The maximum deflection mates a cantilever of uniform crosssection. 60 MACHINE DESIGN and Fig. (See Case 1.
The tension or compression in the pistonrod and connectingnormal to rod. or. C. outward and induce a bending therefore. the long column due to P' — must be added to the flexural stress due to R'. R' however.) then it be desired to check the central stress section I V of the column. as long columns (see equation The columns. may This reaction. The P' thrust of the back column. either ascending or descending. R will then be balanced by an equal and opposite reaction at the feet of the front in the columns. the bolts are loose in the holes. also bends the column B locally. (See Case 18. be secured to the The column if should. Table VI). that is as a beam If encastre at is S and N. better still. only be checked . the effect of R being greatest when the crosshead near half stroke. is made up of light construction this effect '. By similar reasoning each front column is subjected to a com pressive load P' — and the total horizontal component R is balanced by that of the back column through the bed. MACHINE FRAMES AND ATTACHMENTS carries onehalf of the total thrust. the feet. however. in a lefthand direction. bending being resisted by the fastenings at the cylinder. 433 and that each of the front columns carries onequarter. thus setting up a negligible tension bed and leaving a compressive force only on the column. constitute a very The columns and and except stiff structure. The horizontal component R tends to spread the foot of the column stress in it. and hence C also. the foot should fit against a ledge cast on the bed plate. or. Table I.. — 2 may be resolved into components perpendicular and parallel to the face of the foot. need N. have a resultant R' to the guide. the foot should be well dowelled to the bed. which may have a large value where the connecting rod is short compared to the crank. where the frame be neglected. This resultant tends bend B. The sum of these stresses should not. of course be greater than the allowable stress for the material used. The vertical component will equal p — . bed by fitted bolts.
AB. of course. BC. axis at right angles to this plane. proportional to the respective areas . The most dangerous stress will section in this case will be under and the be that due to — plus the P' 2 tensile stress due to the bending effect of R f . etc. 187 case. in which to it is required to in the check the stresses when the force P. is C be the centre of gravity of the section.434 MACHINE DESIGN the piston is When descending the steam pressure tends to cylinder. due position shown. the stress at M being The tensile when direction of upward to the plane of the paper. Thus may represent the crosssection of the column of a radial drilling machine. the horizontal components inducing negligible compression in the bed. centre of gravity. parallel lines ab. necessary only to find the axis at right angles to as follows: at UV. the tendency to rotate will be around the axis X' X' at right angles to PC. lay in a plane of symmetry of the section. by finding the intersection of any pair If the section of gravity axes. and the resistance of the section against inertia of the tensile such rotation will be measured by the moment of section with reference to this axis. is and com when its direction downward. preferably at right angles to the lay off known axis In Fig. In the foregoing examples the lines of action of all forces acting on the section considered. If drilling. 187. cd. the centre of shown by dotted lines gravity of each area draw xxin UV. separate the bed and the The reactions at M and N R f are reversed in direction and the columns are put in tension. Fig. as From be. The maximum and the compressive stresses will occur at the fibre farthest removed from X' X' or pressive at M P is and N. the arm of the force P. This is most readily done graphically Divide the Fig. be sufficiently strong in tension to resist the force tending to separate the cylinder and bed. occasionally the force or forces acting are not in a plane of symmetry. The fastenings of the columns to the cylinder and to the bed plate must. 187.. C. 187 (a). and the section tended to rotate around a gravity While this is the most usual Fig. may be located readily. it is has an axis of symmetry. as UV. section into small areas.
The accuracy of the work may be checked with a planimeter. The moment around X X r r may be lines. 187. From the to AO. areas . as follows: Draw Fig. etc. plus its area into the square of the distance between these of the axes.MACHINE FRAMES AND ATTACHMENTS 435 whose gravity axes are ab. 188 (b). will The sum be the required moments of inertia of all such small moment of inertia of the section. From any point on ab. 187 by the dotted line L. Divide lines this equivalent figure into approxiparallel to mate rectangles. on each side of CB as ordinates of an equivalent section. ao and od locates the gravity axis evident that this XX (see also Art. X' X and plot the intercepts made by it on the given section. etc. draw od parallel to OD. Take any pole O and draw AO. draw ao indefinitely. be its moment of inertia round its own gravity axis parallel to X' X f . as it is as X" X" . BO. It is method may be applied when both axes are unof inertia of the section known. as shown at Then the moment of inertia of r around the axis X' X' will r. Fig. be. parallel to f . evident that the area of the transformed section will be equal to that of the original. shown in Fig. most readily found by transforming the area of the figure into an equivalent figure with RP as a base. 120). by drawn X' X' . intersection of ob and be draw co parallel to CO. parallel From the same point draw ob parallel to OB. and from its The intersection of intersection with cd.
188 (a) allowable. therefore. 187. 19 (Fig. and. in general. however. if. torsional strength flexure is must be withstood. or a combination of these. due to shrinkage. 187. The most trying stresses to which a frame may It be subjected are torsion. If (in then the pre dominating stress in a frame is simple flexure is a given plane). In addition to the stresses induced in the frame by the energy transmitted by the machine. while steel castings are rap coming into use for severe work. on account of their complicated shapes. as in Fig. since (a). also. 7) is resisting torsion. by massing the metal on the tension side as shown in 188 (a) thus making the tensile and compressive stresses more . its Economy in the use of metal. or the plane of may change. 12 that the hollow section (Fig. in Art. as the construction of the pattern simpler and the shrinkage stresses less serious than in such sections as shown in Fig. or other metal whose tensile strength is much is less than its compressive strength. the general prim ciples governing the forms of sections must be kept in mind. and the problem must quently be is the judgment of the designer. 188 (b) it better design. or the settling of a part of the foundation. 187 and better to 188 Sometimes it is make the section so large that the flexural stress can be safely withstood by a wall of uniform is thickness. as the thick sections of Fig. a section similar to that shown in Fig. Distribution of Metal in Frames. 188 are very likely to have a porous interior. flexure. Castiron parts more . if has been noted effective for most this 7. shall demands that distribution throughout the frame be in accord with the best analysis possible. It shown in Fig. Both these classes erally of stresses are.436 MACHINE DESIGN 195. combines the merits of both Figs. especially if stiffness a large factor. Machine frames are usually made of castings. but if in addition. in proportion to the strength of the material. very complex and genfre beyond mathematical left to analysis. are correct. 188. and. 10) that in the case of cast iron. was also noted in Art. be the predominating or modified sections as stress sections such as are shown in Fig. it may also be subjected to severe accidental stresses due to such causes as shrinkage. The metal in the walls will be much sounder. cast iron being the material idly most used. a great saving of material effected Fig. a section like that shown in Fig.
as is unequal shrinkage and porosity are sure edge of the rib to result. Attachments and Supports.MACHINE FRAMES AND ATTACHMENTS than four or this five 437 inches thick are almost sure to be defective in walls of such sections as manner. Fig. is The general appearance of more by the outline of the main frame than by that of any other member. of course. 183 is shown the outline of a frame in which the various a machine affected sections have been proportioned in accordance with the loads brought upon them. 191). as they seldom cast well. 188 should taper uniformly from the thick part to the thin parts. be governed is by it required to render and the manner in which If the outline of loaded and supported. little as possible (see Fig. and also making it more difficult to judge of the relative strength of various appear as attachments to the N sections of the frame. thus starting rupture across the entire section. In Fig. to minimize as far possible the concentration of shrinkage stresses. Thin This wide flanges or webs should not be cast integral with thick heavy parts. therefore. This outline should. absorb the vibrations call This may for removing cores should be placed near the gravity axis so as to reduce the strength as 196. that even strong and it enough may not have mass enough rapidly. 184 illustrates the same machine with the attachments merged into the main member. as the is liable to crack through shrinkage. since the greatest is strength of cast iron in the outer skin. especially true of thin ribs cast on the tension side of large sections. and not obliterated at places by the various attachments which restrain the moving parts or support the frame. and all corners should be well rounded. thereby destroying the character of the design. Small brackets or other attachments of thin sections should never be cast large frame. and the various bosses and the support S main member. The form the service is it of an attachment will. The shown in Fig. and as filleted. the attachment is . dictated set up when running more more metal in the frame than is Openings for supporting or by other requirements. be clearly shown. It should also be stiff membered to do the to when a frame is both required work at low speeds. on a A section of moderate re thickness is often stronger than a thicker one.
438 MACHINE DESIGN based on theoretical considerations. ance of stability. in the This construction designed to not correct.) frame rests directly on the floor its outlines should be carried down to the floor in such a manner as will give an appear Fig. is obvious. Fig. Thus of parabolic outlines are given to an attachment. by the bending 192. U. so far concerned. which should be stiff it. 191 shows the outline . which fre quently neglected. carried 190 shows the same machine with the is outline straight to the floor as stability and the improvement in appearance. such as the housings the H for supporting the tool in Fig. 190. enough withstand such stress thus brought upon Any settling of the foundation might affect the alignment of . vertical outline of the back of the frame is undercut. as U is an attachment to the bed. to all resist the force of itself the cut and to transfer it the bed. Thus Fig. 189. 189 shows such a machine frame on which the Fig.the upright. If the (See also Article 15. the upper end housing must be modified from the theoretical parabolic effect of the force outline indicated P. of a planing machine in which form of a leg. is carried to the is floor at V. care should be exercised that if all the modifying influences are duly considered. Fig. so as to is provide for the shearing effect at the upper end.
the supports themselves being so stiff as materially to assist the to socalled frame and practically reducing the support support. to obtain threepoint support. in general. carried on eight points. by settling of the foundation. of support throughout.MACHINE FRAMES AND ATTACHMENTS 439 U and hence the arrangement shown in Fig. sides 189. In smaller machines the frame is carried on (a) cabinet or box shown in Fig. the twopoint Small machines can often be supported on a single overhanging parts of the frame having a parabolic in outline as suggested height if Fig. and should have sufficient stiffness to resist distortion due to the settling of the foundation. 192. The choice of support will. which pillar may be of two general types. 193. since the latter is very difficult to avoid. In large machines the frame usually rests directly on the foundation. depend on the type and In any case the number of points of size of the machines. 192 is more nearly correct. (b) supports (Fig.. and not one or more of each.e. preferable to use one form pillars or all all box legs. If the box pillar is of considerable the top. boxpillar. where the frame is taken from an Fig. . Fig. 192 shows the same frame properly carried on box supports. 191 (which actual design). i. 192). of course. but it is seldom necessary is to place supports as close together as in Fig. and legs as supported on three points affected it is evident that the frame cannot be It is difficult. If the machine can be supports. for the should taper slightly toward at made parallel the pillar will It is appear wider the top than at the bottom. support should be as few as possible.
and by using the simplest and most direct design with smooth It is transition curves between straight lines which intersect. 193 lengthens the distance between the reactions. but where curves are necessary they should be as simple as possible and in general the best results can be obtained by using arcs of circles or parabolas. The of Fig. 196. a proverb in design that " what is right looks right. unless it is absolutely essential in order to obtain stability. R. 195. and. and adds to the cost of production. It is not always possible or desirable to make machine frames and supports with simple straightline outlines. 193. R. as shown in Fig. Fig. 194 is better legs should and much outline easier to make. as it really detracts from the appearHarance of the machine. less but expensive. as in Fig. Fig. therefore. 194. The be so placed that the L forms a continuation of the principal vertical outline U of the frame. 195. these should not curve outward as in Fig. Fig. When Spreading the legs as in Fig. ." 440 MACHINE DESIGN the frame must be supported on legs. The leg shown in profile in Fig. mony of of design can be attained by making the various members correct proportions to withstand the loads brought upon them. 193. 195 and 196. 195 are not only use complex curves and ornate features Fig. . Ornamentation of a fanciful nature is not permissible anywhere. 194. increases the bending effect on the bed and legs as a whole. the end view of the legs as The same remarks apply to shown in Figs.
168. 26 Air reservoir. coefficient of friction of. 263 Belt. 251 ball. no Belts.INDEX Absolute efficiency. of. 163. Morse. 170 allowable load on. 271. allowable pressure on. 178 of. 239 355 for. example of design tubes. 164 on sliding surfaces. stud. allowable stress in. of. 169.. 168 radiation of heat from. theory of. 271. 252 Cap screws. 264 243 363 construction differential. 23$ Ball bearings. 308 Chains. 169 Barnard. P. 2^ strap. 277 efficiency of. 318 proof test of. 164 tap. of. efficiency of. 264 Bearings. 154 Bolts. (riveted fasten location of. 138. 49 Bevel gears. of uniform strength. ings). roller. 154 strength of. 207 initial of. 146 273 table of proportions thrust. 253 Briggs' system of pipe threads. 149 W. 163. Brakes. 319 practical rules for. 317 Bending moment. 238 Bearing. 345 construction of. 247 Butt joints in plates. 344 creep of. strength rivets. 163 Prof. 355 89 Boiler plate. 176 Babbitt metal. Apparent factor of Axles. of. 310 practical consideration of. 163. Belts. 233 velocity of. 318 314 Chain drums and sheaves. collar. 282 Bands. 308 weight of. table 251 resultant stress in. block. A. 338 conveyor. 181 Professor Sweet's experiments on. 344 for power transmission. general theory 40 181 resilience of. through. Carman. 285 safety. 359 forms metals of. 172 straining action in. 29 Antifriction metals. block. hydraulic. 89 Belting. slip of. 205 tension in. ^8^ Block brakes. 310 Accumulator. 172 stud. 348 313 roller. 357 perfectly lubricated. 355 coefficient of friction for. equivalent or ideal. 164 277 of. N. 29 Air compressor. Beams. 340 Renold. 340 44i . 172 experiments on the strength for reinforcing castings. friction. thin. experiments on 314 transmission. 218 Carrying strength. step. Prof. 42 Bearing pressures on journals.
226 Flather. silent. of. of. rotating. 200 practical considerations in. 6 in air compressor. friction. 353 of friction of pivots. 36 Constraining surfaces.. 363 radially expanding. 301. INDEX 345 Efficiency of belting. 73 Fairbairn. 36 Conservation of energy. 99. 424 215 406 413 Deflection of ropes. 211 Flywheel rim 422 412 301 Flywheels. 197 232 Flanges. 329 Cotters. 198 on rope drives. 363 Energy problems. 324 Flexure and direct Coupling. pipe. Collar bearings. 6 no 109 coefficient of. Elasticity. Sir Wm. allowable pressure on. 359 disc. 223 thin. 88 on boiler work. flues.442 Chains. 419 of. in machine elements. Flues. 61 of. 98 mechanical. coefficient of. experiments on the strength general theory stresses in. 104. materials of. applications of. experiments on 217 Compression and torsion. definition of. stresses in. Fatigue of materials. 96 general theory laws of. 172 strength of. thick. 184 Coefficients of friction for brakes clutches. 303 Hook's. eccentric loading or long struts. 40 joints. 105 general theory of. 196 Feathers. Crankeffort diagram. 362 coefficient of friction for. 77 361 359 magnetic. 264. 141 Clavarino's formula. 62 of friction for friction wheels. 363 of screws. flange shaft. 9.. 359 31 Discs. Continuous system of ropedriving. 201 6. 43 stress. table of dimensions of. 157 of square. 77 stresses due to. Differential brake. no of circular surfaces. 57 in machine elements. work Force fits. absolute. Elastic limit. 26 in steam engine. 340 weldless. 162 Efficiencies of band. 223 Clutches. Oldham. table of. 406 coefficients of fluctuation. 155 Factors of safety. 35. 425 350 Efficiency. 332 table of. 340 of riveted fastenings. construction of. 6 redistribution of. 323. 157 of triangularthreaded screws. flexible shaft. 204 of. 363 machine elements. 267 91 Columns. 16 305 Coefficient of elasticity. 360 shaft. 34 of friction for screws. 365 Cylinders. 211. 29 and Euler's formula for columns. 34 Energy cycle. combined. clutches. 304 58 torsion combined. Friction.threaded screws. 34 resilience. Prof. 113 conical. 20 Cycloidal gear teeth. 269 Factor of safety. 359 Forces acting on machines. 82 Feather keys. 306 shaft. 304 Couplings. 97. 333 Deformation. 97 . 318 of bolts.
39c Involute gear teeth. 162 Helical of screws. experiments 255 strength of twisted.. 386 Inertia effects in general. 341 strength of. 85 Lewis'. draw. table of proportions of. 192 Gearing. 369 proportions of. 390 involute. 205 Hunt. 19 cycloidal. Hook's coupling. 393 worm. cut. 395 spiral. efficiency of spur. 6 Lame's formula. of. 390 Impact. 350 for. 392 Hindley worm. 251 368. 388 width of face perfectly lubricated.. C. sunk. 343 materials 352 by. 393 interchangeable systems screw. 157 of rolling. 190 flat. 387 effect of. 366 Lasche. formula for columns. 376 390 of. 29 Inertia forces in steam engines. W. 389 Gordon's formula for columns. B. 395 skewbevel. rawhide.. rawhide. 365 Johnson's. allowable pressures on. 191 366 Kinematics. 353 power transmitted wedgefaced. 365 Hunt system. 102 work Furnace of. allowable load on. H. 398 Hobs and hobbing. 305 Hooks. bearing pressure on. 249 of. 29 369 redistribution of. 389 design of. 253 388 wheels. 245. 397 100 of general laws of. allowable stresses in. formula for columns. 370 shrouding strength stub. 190 stresses in. of. 109 Hoisting mechanism. 373 mortise. gearing. 9. 372 strength of rims and arms. 247. herringbone. 391 364 saddle. of. 192 190 forms of. 82 teeth. Woodruff. joints in plates. 223 395 standard form Lap of. 257 examples of design wear on. 65 Journals. forces acting on.INDEX Friction of dry surfaces. 392 general principles helical or twisted. T. 395 Gears. 98 of flat 443 surfaces. 97 Gears. 78 Imperfect lubrication. allowable speed bevel. 325 system of gear teeth. 145 of. . shock. 371 Keys. 196 392 of. 29 summary wheels. 67 of lubricated surfaces. 365 Fellows system of stub. 97 corrugated. 383 Launhart's formula. hoisting. 99 of triangular threads. machine moulded. 353 341 forms of. of. 352 coefficients of friction for. Gear teeth. 138. 99 static. 190 table of dimensions of sunk. 354 Hoops. Wilfred. 66 J. 257 imperfectly lubricated. formula for gear Live load. on rope driving. 369 methods of making. 222 flues.
226 threads. 136 general equations for. 322 materials for fibrous. 138. 138. definition of. 323 (by wire). 436 stresses in. of bolts. thin. 143 Morse chain. butt 146 106 chain riveting. 102 77 methods of. 327 wire hoisting ropes. 262 Ropedriving. 2 Micro flaws. 429 making of. 30 Plates. 106 transverse pitch of. experiments on ciency of bolts. dead. sheaves for fibrous. 428 432 forms of joints. 276 grooves. 139 Mechanism. 269 Planing machine. 153 Rivets. 144 Riveting. 339 velocity of fibrous.. Riveted fastenings. 141 design. 76 elastic. 149 theoretical strength of. 211. 395 . 106 Rope transmission. 154 stresses in. machine. 217 hemp. 309. 67 Rotating discs. 143 Roller bearings. 331 systems of fibrous. definition of. definition Pulleys. 106 in perfect lubrication. cotton. screws. deflection of fibrous. 348 pitch of. 163 Renold chain. James.. 14* Mechanical advantage. 338 Rankine's equation 141 for columns. theory of. theory of. of sliding surfaces. 329 Oldham Perfect coupling. 228 Power. 334 strength of fibrous. 68 joints. 327 Pivots. 7 strength of fibrous hoisting. 322 Piping. 338 406 Punching machine. 138 Multiple system of ropedriving. 163. of. 138 efficiency of. suddenly applied. 172 effi practical consideration of. steady. F. 149 practical rules for.444 INDEX Resilience. 322 materials for wire. Screw fastenings. no staggered riveting. 348 gearing. 138. 151 of. 60 strength of wire hoisting. H. 273 allowable load on. experiments of. 338 Ropes. 168 Pipes. 428 distribution of metal in. 10. 83 Moore. 28 efficiency. 137 general considerations. 138 strength of materials foi. 178 of journals. 141 Machine attachments. methods 261 Ritter's formula for columns. 304 lubrication. 155 relative strength of. 1 frames. 261 of. stresses in open. 100. Prof. 332 fibrous hoisting. etc. 238 perfect. 147 lap joints. 144 stresses in closed. 155 failure of. 31 Lubrication. 425 Relative strength of riveted fastenings. leather. 107 on riveted fastenings. 271. diagonal pitch of. 437 factor of safety in. Load. 322 Ropes. ^^y Pipe couplings and flanges. 138. 164 marginal strength 143 supports. 437 McBride. practical considerations 224 Manila. coefficient of friction of. of. 329 Oil film. imperfect.
238 lubrication of. 303 standard screws. 75 stresses Sliding surfaces. Tap 389 bolts. 157 machine. 166 Stayed surfaces. S. k>6 Tension in machine elements. 237 Shear in machine elements. 165 Storage battery. 183 in. Spur gear 156 376 forms gearing. of. Prof. 233 bearing pressures on.INDEX Screw and screw fastenings. 392 gears. 78 200. 163. 255. 33 subjected to torsion and bending. coefficient of expansion. 303 couplings. 35 Stribeck. allowable speed of. 114 characteristics of. or Sellers standard. M. design of. 32 table of formulae for. 213 Splines. 33. 319 Temperature. 207 fits. 40 strain diagram. Prof. 186 cap. experiments on tubes. 164 coefficient of friction of. 301 coupling. 186 U. 94 allowable span of. energy distribution in. design of. 272 of. 268 Springs. 119 forms of. whirling 299 Stub gear Stud teeth. 75 Spheres. 36 Shock Shrink in machine members. 301 Shafts. T. 32 subjected to torsion. 238 due to. Springs. 187 efficiency of. for power transmission. efficiency 184 of. 390 164 Shaping machine. 290 torsional stiffness of. 184 allowable stress of. Taylor.. 388 stresses in transmission. flange. 300 definition of. 263 allowable pressures on. 164 mechanical advantage multiplethreaded. laminated or plate. 116 128 helical. allowable deflection of. 264 Stewart. 405 Thrust bearings.. 164 practical considerations in. energy distribution Step bearing. 117 design of. 387 friction of. strength of. 386 machinemoulded.. angular velocity of. of. Prof. 288 predominating or primary. 365 . 116 Toothed gearing. 114 flat. design materials spiral. 218 Set screws. 299 Strap brakes. in. 289 compound. 166 Sellers shaft coupling.. 157 118 of. method of relieving sliding surfaces. 299 Straining action. 163. bolts. 156 Screws. 231 Steam engine. 16 Whit worth standard. 163. 163. 40 hollow. 369 width of face of. 29 Strain. 10 Sweet. 298 working. 184 445 flat. 135 spiral. 115 118 teeth. 93 factors of safety for. nature of. F. 166 R. 35 Thrust bearing for worms. rules for belting. experiments of. bearing pressure on. table Stress. 183 power transmission. efficiency of. 270 efficiency of. 123 for for power transmission. of. 32 Shaft clutches. 204 Shrouding of gear teeth. 357 Strength of materials. definition of. applications of. springs in torsion.
Beaucamp. 166 Wire rope transmission. 84 Towne.. 226 Hindley. 43 in machine elements. 392 399 limiting pressures on. INDEX 364 of. 405 . Tubes. experiments Tower's experiments. experiments on hooks. ^^^ theory of. 398 Unions. and flexure. R. 334 of. standard screws. 336 Wohler's experiments of. strength. Whitworth standard screws. 106 334 ropes. 211. combined.446 Toothed gearing. 403 efficiency of. materials for. H. 217 Twisted gears. 34 velocity ratio of. efficiency friction of. 401 Ultimate US. 227 formula. pipe. 398 of. 36 Tower. 254 power transmitted by. of. Van Stone pipe flanges. 401 limiting velocities of. 366 Torsion 57 classification of. 86 interchangeable systems Weyrauch's and compression. 162 162 Working stress. Work of deformation. 35 Worm and worm wheel. 76 343 Triangular threads. definition of. 395 gearing. combined. table 167 thrust bearing. design of.
.
.
* Herrick's Denatured or Industrial Alcohol 8vo. 12mo. 12mo. 8vo.) 12mo. Limited ARRANGED UNDER SUBJECTS Descriptive circulars sent on application. In Kemp and Preparation) * McKay and Larsen's Principles and Practice of Buttermaking 8vo. Maynard's Landscape Gardening as Applied to Home Decoration 12mo. Rewritten. Green's Principles of American Forestry (Woll. 8vo. 1 25 Winton's Microscopy of Vegetable Foods Woll's Handbook for Farmers and Dairymen 2 50 7 50 1 16mo. Systematic Pomology Elliott's Engineering for Land Drainage (Second Edition. (In Preparation) .. Armsby's Principles of Animal Nutrition Budd and Hansen's American Horticultural Manual: Part I. $4 00 Part II. Propagation. 8vo. Culture. 2 50 5 00 3 50 2 00 3 00 3 50 3 00 . 12mo. Books marked with an asterisk (*) are All books are bound in cloth unless otherwise stated. Rewritten) Practical Farm Drainage. AGRICULTURE—HORTICULTURE—FORESTRY. 8vo. 8vo. 8vo. sold at net prices only. Shorttitle Catalogue OF THE PUBLICATIONS or JOHN WILEY & SONS New York London: CHAPMAN & HALL. 50 50 1 50 1 50 4 00 1 50 2 00 4 00 1 1 1 1 1 50 50 50 * Schwarz's Longleaf Pine in Virgin Forests Stockbridge's Rocks and Soils 12mo. Graves's Forest Mensuration 12mo. 8vo. Grotenfelt's Principles of Modern Dairy Practice. (New Edition. 50 ARCHITECTURE. 4to. 12mo. Baldwin's Steam Heating for Buildings Berg's Buildings and Structures of American Railroads Birkmire's Architectural Iron and Steel Compound Riveted Girders as Applied in Buildings Planning and Construction of American Theatres Planning and Construction of High Office Buildings Skeleton Construction in Buildings 1 12mo. 8vo. Waugh's Landscape Gardening. 12mo. and Improvement 8vo. Sanderson's Insects Injurious to Staple Crops Sanderson and Headlee's Insects Injurious to Garden Crops.
. Law of Contracts Law of Operations tecture 8 vo. 8vo. mor. 8vo. 2 50 Chase's Art of Pattern Making 12mo. 8vo. 00 16mo. 12mo'. 8vo. Nitrocellulose.and Brickmasons' Edition Sabin's House Painting Siebert and Biggin's Modern Stonecutting and Masonry Snow's Principal Species of Wood 8vo. Azimuth Crehore and Squier's Polarizing Photochronograph * Davis's 8vo. mor. vised and Enlarged) * Modern Baths and Bath Houses Sanitation of Public Buildings Theatre Fires and Panics Johnson's Statics by Algebraic and Graphic Methods 8vo. 00 4 00 3 1 12mo. 16mo. 12mo. . Durand's Resistance and Propulsion of Ships * Dyer's Handbook of Light Artillery 12mo. * Stone. mor. Elements of Law 8vo. mor. . 25 3 50 2 50 1 1 1 Entirely Re12mo. $4 in Construction. 16mo. * Treatise on the Military Law of United States 8vo. Explosives High 8vo. 8vo. mor. 5 50 Conditioning Wilson's Air 12mo. Preliminary to Construction in Engineering and Archi 3 3 6 6 3 00 00 50 50 00 00 50 8vo. Hamilton and Bond's The Gunner's Catechism * Hoff 's Elementary Naval Tactics 8vo. Ingalls's Handbook of Problems in Direct Fire * Lissak's Ordnance and Gunnery 8vo. mor. 2 50 Screw Propellers and Marine Propulsion 8vo. SteamFitters'. Bernadou's Smokeless Powder. 1 50 Worcester and Atkinson's Small Hospitals. mor. Gerhard'r Guide to Sanitary Inspections. 8vo. mor. 8vo. Carpenter's Heating and Ventilating of Buildings * Corthell's Allowable Pressure on Deep Foundations Fieitag's Architectural Engineering Fireproofing of Steel Buildings (Fourth Edition. 50 50 50 00 3 00 12mo. 1 25 ARMY AND NAVY. 18mo. Suggestions for Hospital Architecture. Richey's Handbook for Superintendents of Construction Building Foreman's Pocket Book and Ready Reference. 16mo. 8vo. 8vo. 24mo. . 3 00 * Cloke's Enlisted Specialists' Examiner Gunner's Examiner Craig's 8vo. Sheep. with Plans for a Small Hospital. 1 1 1 1 Towne's Locks and Builders' Hardware Wait's Engineering and Architectural Jurisprudence 18mo. 8vo.) * Dudley's Military Law and the Procedure of Courtsmartial. * Building Mechanics' Ready Reference Series: * Carpenters' and Woodworkers' Edition * Cement Workers' and Plasterers' Edition * Plumbers'. 16mo. mor. 16mo. DeBrack's Cavalry Outpost Duties. and the Theory of the Cellulose Molecule 12mo. (Carr. 8vo. 5 00 Sheep. 12mo. 4to.Large 12mo. Kellaway's How to Lay Out Suburban Home Grounds Kidder's Architects' and Builders' Pocketbook Merrill's Stones for Building and Decoration Monckton's Stairbuilding Patton's Practical Treatise on Foundations Peabody's Naval Architecture Rice's Concreteblock Manufacture. 16mo.Briggs's Modern American School Buildings Byrne's Inspection of Materials and Wormanship Employed 8vo. Establishment and Maintenance. and Tinners' Edition. 8vo. 2 00 1 50 3 50 3 00 2 50 7 00 2 00 2 50 5 00 3 00 4 00 2 00 1 1 00 50 4 00 6 00 2 . 8vo. 4to. 16mo. 8vo. Modern Eissler's * Fiebeger's Textbook on Field Fortification Large 12mo. . 2 2 5 5 4 5 7 2 4 5 1 1 00 00 00 00 00 50 00 00 00 50 50 50 50 .
8vo.8vo. (Sixth vised and Enlarged) 8vo. Jamestown Meeting. 8vo. Rust's Exmeridian Altitude. Hartford 8vo. * Weaver's Military Explosives Woodhull's Notes on Military Hygiene leather.* Ludlow's Logarithmic and Trigonometric Tables * Lyons's Treatise on Electromagnetic Phenomena. 1 00 00 50 00 00 00 00 50 50 ASTRONOMY. 8vo. 4 00 1 Pardoe's Manual of Practical Assaying. (Matthews. Putnam's Nautical Charts 8vo. (Hall and Defren) 8vo. (Mercur. mor. Treatment of Gold Ores Ricketts and Miller's Notes on Assaying Robine and Lenglen's Cyanide Industry. mor. Azimuth and Star. 8vo. 8vo. 1906 8vo. 8vo. Miller's Cyanide Process 12mo. half mor. Comstock's Field Astronomy for Engineers Craig's Azimuth Crandall's Textbook on Geodesy and Least Squares Doolittle's Treatise on Pracical Astronomy Hayford's Textbook of Geodetic Astronomy 8vo. Arnold's Compendium of Chemistry. 8vo. 16mo. 50 50 00 00 00 1 00 2 50 3 00 5 00 2 00 2 3 3 4 3 CHEMISTRY. Edition. Merriman's Elements of Precise Surveying and Geodesy 8vo. mor. Alexeyeff's General Principles of Organic Syntheses. . 8vo. Re 50 3 00 3 00 3 00 1 1 Manual of Assaying 12mo. Betts's Lead Refining by Electrolysis 8vo. 12mo. 50 2 00 4 00 1 Nixon's Adjutants' Manual 24mo. Association of State and National Food and Dairy Departments.). Azimuth and StarFinding Tables * White's Elements of Theoretical and Descriptive Astronomy 12mo. Fletcher's Practical Instructions in Quantitative Assaying with the Blowpipe. 8vo. 1907 * Abderhalden's 5 00 1 25 3 00 3 00 4 00 3 50 3 00 3 00 3 . * Abegg's Theory of Electrolytic Dissociation.) 12mo. Physiological Chemistry in Thirty Lectures.) Ulke's Modern Electrolytic Copper Refining Wilson's Chlorination Process Cyanide Processes O'Driscoll's Notes on the (Waldo.mor.) Large 12mo. 4to. . 8vo. Armsby's Principles of Animal Nutrition (Mandel. 8vo. Peabody's Naval Architecture * Phelps's Practical Marine Surveying 8vo. (von Ende. Hosmer's Azimuth 16mo. 50 3 00 1 50 ASSAYING.12mo.) 8vo. 8vo. 8vo. 8vo. $1 00 6 00 7 50 1 Manual for Courtsmartial * Mercur's Attack of Fortified Places * Elements of the Art of War 16mo. 8vo. * Mahan's Permanent Fortifications. 8vo. and II.) I. Rust's Exmeridian Altitude. its Minet's Production of Aluminum and Industrial Use. Furman and 16mo. 7 2 2 5 1 00 50 50 00 00 50 24mo. 12mo. Vols.. Low's Technical Methods of Ore Analysis 8vo. * Michie and Harlow's Practical Astronomy 8vo.Finding Tables Sharpe's Art of Subsisting Armies in War 18mo.each. 2 2 3 4 3 1 12mo. Lodge's Notes on Assaying and Metallurgical Laboratory Experiments.. * Tupes and Poole's Manual of Bayonet Exercises and Musketry Fencing. Allen's Tables for Iron Analysis 8vo. Meeting. (Le Clerc.
12mo. Color Pigments. Laboratory Methods of Inorganic Chemistry. Laboratory in . 1 50 Fowler's Sewage Works Analyses 12mo. — Grotenfelt's Principles of Modern Dairy Practice. 8vo. (Wells. 12 50 When Sold Separately. Vol. 2 00 Fresenius's Manual of Qualitative Chemical Analysis.) * McKay and Larsen's Principles and Practice of Buttermaking 8vo. 1 25 Dannerth's Methods of Textile Chemistry 12mo. Descriptive. (Burgess. and Varnishes. SteelLarge 12mo. 1 50 Furman and Pardoe's Manual of Practical Assaying. 2 00 Fletcher's Practical Instructions in Quantitative Assaying with the Blowpipe. (Mandel.) 8vo.) 8vo.) 8vo.) Landauer's Spectrum Analysis. Low's Technical Method of Ore Analysis 12mo. )8vo.) 12mo. Revised and Enlarged. (Hall and Phelan). 1 25 2 00 1 25 4 00 5 00 2 00 2 50 4 00 3 00 1 00 * Haskins and Macleod's Organic Chemistry 12mo. 8vo. (Boltwood. Textbook of Organic Chemistry.) Groth's Introduction to Chemical Crystallography (Marshall) Hammarsten's Textbook of Physiological Chemistry. 2 00 Duhem's Thermodynamics and Chemistry. Organic Chemistry. Leach's Inspection and Analysis of Food with Special Reference to State Control 8vo. (Cohn.8vo. 12mo. Bernadou's Smokeless Powder. * Holley's Lead and Zinc Pigments Large 12mo. and Theory of the Cellulose Molecule 12mo. 12mo. 3 00 Quantitative Chemical Analysis. and Natural Waters LassarCohn's Application of Some General Reactions to Investigations in (Tingle. mor. $8. making Alloys and Graphite 8vo. * Laboratory Manual for Students 12mo. Holley and Ladd's Analysis of Mixed Paints. (Lorenz. 3 00 * Blanchard's Synthetic Inorganic Chemistry 12mo.Austen's Notes for Chemical Students 12mo.8vo.) 8vo. Lob's Electrochemistry of Organic Compounds. Textbook of Inorganic Chemistry. Directions for Jackson's Johnson's Rapid Methods for the Chemical Analysis of Special Steels.) 8vo. 2 00 Gill's Gas and Fuel Analysis for Engineers 12mo. 1 50 * Claassen's Beetsugar Manufacture. 1 00 * Browning's Introduction to the Rarer Elements 8vo. (Hall and Blanchard). 8vo. Hinds's Inorganic Chemistry 8vo. II. $1 50 (In Preparation. 5 00 Manual of Qualitative Chemical Analysis. Vol. (In Press). (Tingle. (Walker and Mott. Lunge's Technochemical Analysis. 2 50 Biltz's Introduction to Inorganic Chemistry. (Cooper. Hering's Ready Reference Tables (Conversion Factors) 16mo. $6..) 12mo. Part I. 1 00 2 50 2 50 3 00 2 50 3 00 1 Large 12mo. (Hall and Rolfe. mor. (Wells. 2 00 Tests and Reagents 8vo.. 4 00 Erdmann's Introduction to Chemical Preparations. (Walker. 3 00 Classen's Quantitative Chemical Analysis by Electrolysis. Fuertes's Water and Public Health 12mo. 3 00 Cohn's Indicators and Testpapers 12mo.) 8vo. * Langworthy and Austen's Occurrence of Aluminum in Vegetable Prod8vo.) 8vo. Baskerville's Chemical Elements. 3 00 * Danneel's Electrochemistry. 8vo.) 12mo.) Hanausek's Microscopy of Technical Products. (Merriam. (Winton. 4 00 Effront's Enzymes and their Applications. (Sixth Edition. Maire's Modern Pigments and their Vehicles 25 3 00 3 00 2 00 1 00 50 00 00 00 1 00 1 50 2 00 7 3 3 3 4 . (Prescott.) Large 12mo. 3 00 Eissler's Modern High Explosives 8vo. I. 1 25 * Gooch and Browning's Outlines of Qualitative Chemical Analysis. * Holleman's Laboratory Manual of Organic Chemistry for Beginners. 8vo. * Herrick's Denatured or Industrial Alcohol 8vo. Animal Products. 8vo. (Woll. 1 25 * Fischer's Physiology of Alimentation Large 12mo. Hopkins's Oilchemists' Handbook Work Physiological Chemistry.) 2 vols 8vo. 12mo.) 12mo. Nitrocellulose. (Cohn.) 8vo. ucts.). (Dunlap. 3 00 * Getman's Exercises in Physical Chemistry 12mo. Lodge's Notes on Assaying and Metallurgical Laboratory Experiments.
8vo. mor. Richards and Woodman's Air. Vol. . * Reisig's Guide to PieceDyeing 8vo. * Elementary Lessons in Heat 8vo.) . Owen and Standage's Dyeing and Cleaning of Textile Fabrics 12mo. Sewage and the Bacterial Purification of Sewage 8vo.) 8vo. Treadwell's Qualitative Analysis. (In Press). Rideal's Disinfection and the Preservation of Food 8vo. 1 8vo. * Tillman's Descriptive General Chemistry 8vo. 8vo. 25 25 50 00 00 50 50 00 50 00 00 00 50 00 5 00 1 * Qualitative Chemical Analysis Smith's Lecture Notes on Chemistry for Dental Students Spencer's Handbook for Cane Sugar Manufacturers 16mo. III. Handbook for Chemists of Beetsugar Houses 16mo. 8vo. Mulliken's General Method for the Identification of Pure Organic Compounds. (Chemical and Bacteriological. II. 16mo. Stockbridge's Rocks and Soils 8vo. (Biddle. * Palmer's Practical Test Book of Chemistry 12mo. (Fischer.) 12mo. (Hall.) 12mo. Vol.) 8vo.11 8vo. Compounds of Carbon with Hydrogen and Oxygen. Robine and Lenglen's Cyanide Industry.) 4 00 1 00 3 50 4 00 1 1 Third Edition 12mo. (Mathewson). (Tingle. Rewritten Textile Fibres. (Hall. (In Press. Part Two. (Le Clerc. Rewritten) 8vo. 12mo. Vol. Large 8vo. 4 00 4 00 1 25 4 00 2 00 1 00 3 00 2 50 1 12mo. Ward and Whipple's Freshwater Biology.) Sabin's Industrial and Artistic Technology of Paint and Varnish Salkowski's Physiological and Pathological Chemistry. (In Preparation). with Special Reference to Sanitary Water Analysis 12mo.). * Meyer's Determination of Radicles in Carbon Compounds. .Mandel's Handbook for Biochemical Laboratory 12mo. . 8vo. * Pauli's Physical Chemistry in the Service of Medicine. Morgan's Elements of Physical Chemistry 12mo. 8vo. I. 1 2 1 3 1 1 1 4 25 00 00 50 50 00 00 50 50 00 5 00 The Commercial Dyestuffs. Vol. (Turnbull. Poole's Calorific Power of Fuels 8vo.) Ware's Beetsugar Manufacture and Refining. 12mo. * Mathewson's First Principles of Chemical Theory Matthews's Laboratory Manual of Dyeing and Textile Chemistry 2d Edition.) 12mo. Penfield's Tables of Minerals. (Fifth Edition. 2 3 3 2 3 3 1 3 4 5 1 3 4 00 5 00 5 . 8vo. Minet's Production of Aluminum and its Industrial Use. and Food from a Sanitary Standpoint 8vo. Vol. (Boltwood. Mixter's Elementary Textbook of Chemistry 12mo. Stone's Practical Testing of Gas and Gas Meters 8vo. mor. 8vo. Ricketts and Miller's Notes on Assaying 8vo. 8vo. Mason's Examination of Water.) Pictet's Alkaloids and their Chemical Constitution. 12mo. Including the Use of Minerals and Statistics of Domestic Production 8vo. Miller's Cyanide Process 12mo. mor.) 8vo. (In Press. (Ramsey. Venable's Methods and Devices for Bacterial Treatment of Sewage 8vo. O'Driscoll's Notes on the Treatment of Gold Ores Ostwald's Conversations on Chemistry. 2 00 1 Part One.) 12mo. (Orndorff. 8vo. 1 8vo.12mo. Van Deventer's Physical Chemistry for Beginners. Nitrogenous Compounds. Morse's Calculations used in Canesugar Factories * Muir's History of Chemical Theories and Laws 8vo. (Considered Principally from a Sanitary Standpoint.12mo. Outline of the Theory of Solutions and its Results * Physical Chemistry for Electrical Engineers 12mo. 50 2 00 2 00 1 1 00 25 1 00 5 00 3 00 1 15 50 00 2 00 3 00 Ruddiman's Incompatibilities in Prescriptions Whys in Pharmacy Ruer's Elements of Metallography. Prescott and Winslow's Elements of Water Bacteriology. 8vo.) Schimpf 's Essentials of Volumetric Analysis Manual of Volumetric Analysis. Water. Quantitative Analysis.) Watersupply. $1 50 60 25 * Martin's Laboratory Guide to Qualitative Analysis with the Blowpipe 12mo. Turneaure and Russell's Public Watersupplies 8vo. Rigg's Elementary Manual for the Chemical Laboratory 8vo. Manual of Assaying (Waldo.
HYDRAULICS. Folwell's Sewerage. * Burr's Ancient and Modern Engineering and the Isthmian Canal 8vo.) Statics by Algebraic and Graphic Methods (In Preparation). half leather.) Smith's Manual of Topographical Drawing. Whipple's Microscopy of Drinkingwater 8vo... Textbook of Chemical Arithmetic 12mo. Howe' Retaining Walls for Earth * Ives's Adjustments of the Engineer's Transit and Level 16mo. I. $2 00 g vo 3 00 8vo! 1 50 ' Short Course in Inorganic Qualitative Chemical Analysis for Engineering Students 12mo. mor. 4to. Winslow and Pratt's Purification of Sewage. Crandall's Textbook on Geodesy and Least Squares 8vo. B. 16mo. II. 3 2 1 1 00 50 50 50 25 00 00 50 50 00 00 00 50 50 25 00 50 00 25 25 4 00 2 00 1 50 Merriman's Elements of Precise Surveying and Geodesy Merriman and Brooks's Handbook for Surveyors Nugent's Plane Surveying Ogden's Sewer Construction Sewer Design Parsons's Disposal of Municipal Refuse Patton's Treatise on Civil Engineering Reed's Topographical Drawing and Sketching Rideal's 8vo. Higher Surveying Vol.and Ventilation of Subways * Tracy's Exercises in Surveying Tracy's Plane Surveying * Trautwine's Civil Engineer's Pocketbook Venable's Garbage Crematories in America Methods and Devices for Bacterial Treatment of Sewage 12mo. Goodhue's Municipal Improvements 12mo. 1 1 50 25 50 50 3 50 1 1 Cyanide Processes Winton's Microscopy of Vegetables Food Zsigmondy's Colloids and the Ultramicroscope. 12mo. 8vo. (Corning and Siebert and Biggin's Modern Stonecutting and Masonry (McMillan. Flemer's Photographic Methods and Instruments 8vo. Breed and Hosmer's Principles and Practice of Surveying. mor. J. 8vo. * Mahan's Descriptive Geometry 8vo.8vo. mor. bds. 8vo. 25 3 2 3 2 1 3 1 1 1 5 5 3 3 1 1 . 16mo. * Hauch and Rice's Tables of Quantities for Preliminary Estimates. Hering's Ready Reference Tables (Conversion Factors) 16mo. Wilson's Chlorination Process 12mo. BRIDGES AND ROOFS. Elementary Surveying 8vo. 2 50 2 00 3 50 12mo. Soper's Air .) Theory and Practice of Surveying 8vo. * Fiebeger's Treatise on Civil Engineering 8vo. 3 00 Baker's Engineers' Surveying Instruments 12mo. Vol. 12mo. mor. 16mo. mor. 3 2 2 7 5 Sewage and the Bacterial Purification of Sewage 4 3 1 Riemer's Shaftsinking under Difficult Conditions. Peele. 8vo. Large 12mo. 8vo. Comstock's Field Astronomy for Engineers 8vo. .Washington's Manual of the Chemical Analysis of Rocks * Weaver's Military Explosives Wells's Laboratory Guide in Qualitative Chemical Analysis 8vo. Johnson's (L. (Alexander). Bixby's Graphical Computing Table Paper 19i X24J inches. Hayford's Textbook of Geodetic Astronomy 16mo. MATERIALS OF ENGINEER RAILWAY ENGINEERING. 8vo. Davis's Elevation and Stadia Tables 8vo.) 12mo.). 7 50 3 00 CIVIL ENGINEERING. 2 2 1 3 5 2 3 00 00 00 00 50 00 00 00 50 50 50 00 00 00 00 6 . Kinnicutt. (Second Edition Rewritten. 12mo. 8vo. ING. (Designing and Maintenance. Freitag's Architectural Engineering 8vo. 8vo.) 8vo. 8vo. mor. Elliott's Engineering for Land Drainage 12mo. 8vo. 8vo. Hosmer's Azimuth 12mo. Johnson's (J. Large 12mo. * Corthell's Allowable Pressure on Deep Foundations 12mo. 8vo. Practical Farm Drainage.
3 50 HYDRAULICS. Roof Trusses 8vo. Heller's Stresses in Structures and the Accompanying Deformations. mor. 6 50 3 00 Architecture 8vo. 8vo. Oblong Hydraulic Motors Coffin's Graphical Solution of Hydraulic Problems Flather's Dynamometers.) 12mo.) Drawspans. 3 00 2 00 5 00 1 from 8vo. 8vo. 10 00 Part I. 2 50 3 00 4 00 5 00 1 50 2 50 4 00 Water in 8vo. (In Preparation. 2 50 Part IV. 8vo. paper. Bridge Design 8vo. Bryan and Turneaure's Theory and Practice 8vo. Iron. Sheep. Stresses in Simple Trusses 8vo. Graphic Statics 8vo. 12mo. Howe's Design of Simple Roof. 16mo. and the Measurement of Folwell's Watersupply Engineering Frizell's 4to. Warren's Stereotomy Problems in Stonecutting * Waterbury's VestPocket Handbook of Mathematics for Engineers. Pocketbook for Bridge Engineers 16mo. 7 . 8vo. . 8vo. parts Two in one volume Wright's Designing of 8vo. 12mo. Webb's Problem's in the Use and Adjustment of Engineering Instruments.8vo. Fowler's Ordinary Fopndations 8vo. Higher Structures 8vo. 2 00 Power 16mo. 5 00 5 50 2 50 1 1 00 25 3 50 BRIDGES AND ROOFS. (Hering and Trautwine. Bridge Trusses 8vo. $6 00 Sheep. ) Bovey 's Treatise on Hydraulics Church's Diagrams of Mean Velocity of Water in Open Channels. 2 00 Waddell's De Pontibus. mor. Foster's Treatise on Wooden Trestle Bridges 4to. mor. 8vo. in the 2 00 5 00 5 00 3 00 10 00 5 00 3 50 2 50 2 50 1 25 2 50 3 00 2 00 2 50 4 00 Modern Framed Structures Merriman and Jacoby's Textbook on Roofs and Bridges: Designing of Small 4to. Grimm's Secondary Stresses in Bridge Trusses 8vo. 8vo. and Arches 8vo. 50 Waddell and Harringtoon's Bridge Engineering. and Stone 8vo. 2X5 inches. Waterpower Fuertes's Water and Public Health Waterfiltration Works Ganguillet and Kutter's General Formula for the Uniform Flow of Rivers and Other Channels. of Contracts of Operations Preliminary to Construction in Engineering and 8vo. Oblong paper. Vol.. 2 00 * Specifications for Steel Bridges 12mo. 8vo. mor. Wilson's Topographic Surveying — 8vo. Boiler's Practical Treatise on the Construction of Iron * Thames River Bridge Highway Bridges. m Symmetrical Masonry Arches Treatise on Arches Johnson. 2 50 Part III. II Small 4to. Greene's Arches in Wood.trusses Wood and Steel 8vo. Barnes's Ice Formation Bazin's Experiments upon the Contraction of the Liquid Vein Issuing an Orifice. 2 50 Part II. 2 50 Morison's Memphis Bridge Oblong 4to. .. with Applications to Trusses.Wait's Engineering and Architectural Jurisprudence Law Law 8vo. 50 8vo. (Trautwine. 10 00 Sondericker's Graphic Statics. Burr and Falk's Design and Construction of Metallic Bridges Influence Lines for Bridge and Roof Computations Du Bois's Mechanics of Engineering. 8vo. Beams.
Vol. Treatise on Masonry Construction 8vo. 5 00 Merriman's Mechanics of Materials 8vo. Color Pigments and Varnishes. Conduits 2 00 8vo.. 3 00 Holley and Ladd's Analysis of Mixed Paints. 2 00 Stresses in of Flexures Framed Structures. 6 00 WaterSupply of the City of New York from 1658 to 1895 4to. Kinematics. 7 50 The Theory Small 4to. 5 00 Rice's Concrete Block Manufacture 8vo. 5th Ed. $1 50 8vo. 2 50 Lanza's Applied Mechanics 8vo. 6 00 Stone and Clay Products used in Engineering. Second Edition. and Plasters 8vo. 8vo. Church's Mechanics of Engineering Du Bois's Mechanics of Engineering. Large 12mo. 10 00 Whipple's Value of Pure Water Large 12mo. 6 00 Keep's Cast Iron 8vo.) Materials of Construction Large 8vo. 3 50 * Greene's Structural Mechanics 8vo. 5 00 * Strength of Materials 12mo. Black's United States Public Works Oblong 4to. 2 50 * Holley's Lead and Zinc Pigments Large 12mo. 3 00 Johnson's (J. (In Preparation. 4 00 * Lyndon's Development and Electrical Distribution of Water Power. Revised and Enlarged Large 8vo. 8vo. 5 00 Wegmann's Design and Construction of Dams. Inspection of the Materials and Workmanship Employed in Construction. Limes.) 2 vols. 3 00 Mason's Watersupply. 4 00 Wood's Turbines 8vo. 10 00 * Eckel's Cements.. Baker's Roads and Pavements 8vo. A Manual for Steelusers 12mo. 3 00 8vo. 6 00 * Thomas and Watt's Improvement of Rivers 4to... 5 00 5 00 5 00 7 50 7 50 5 00 3 00 6 00 16mo. Kinetics Vol. Waterpower. 2 00 Martens's Handbook on Testing Materials. Experiments on Metal Herschel's 115 8vo. (In Press. Merriman's Treatise on Hydraulics 5 00 * Molitor's Hydraulics of Rivers. 1 00 Metcalf 's Steel.Large 12mo. 8vo. Bleininger's Manufacture of Hydraulic Cement. I. 6 00 Turneaure and Russell's Public Watersupplies 8vo. and Domestic Watersupply. 4 00 8vo.) Blanchard's Bituminous Roads. Steelmaking Alloys and Graphite Large 12mo. 8vo. 2 50 the Carrying Capacity of Large. M.. Burr's Elasticity and Resistance of the Materials of Engineering 8vo. 7 50 Maurer's Technical Mechanics 8vo. (Considered Principally from a Sanitary Standpoint.. 4 00 Merrill's Stones for Building and Decoration 8vo. Weirs and Sluices 8vo. II. 2 50 Hazen's Clean Water and How to Get It Filtration of Public Watersupplies Hazelhurst's Towers and Tanks for Waterworks MATERIALS OF ENGINEERING.) 8vo. Strength of Materials and 8 . 2 50 Johnson's (C. Construction Highway Byrne's 8vo. (Henning. 1 50 Wilson's Irrigation Engineering 8vo. Riveted. 7 50 Maire's Modern Pigments and their Vehicles 12mo. Small 4to. 2 00 * Richards's Laboratory Notes on Industrial Water Analysis 8vo.) Rapid Methods for the Chemical Analysis of Special Steels. enlarged 4to. 2 50 Patton's Practical Treatise on Foundations 8vo. 1 00 Williams and Hazen's Hydraulic Tables 8vo. 2 00 Hoyt and Grover's River Discharge Hubbard and Kiersted's Waterworks Management and Maintenance. 50 Schuyler's Reservoirs for Irrigation.) * Bovey 's Strength of Materials and Theory of Structures 8vo. Statics.) Fowler's Ordinary Foundations 8vo. . (In Preparation. 2 00 Morrison's Highway Engineering 8vo. B.
and an Appendix on the Preservation of Timber 8vo. Taylor and Thompson's Treatise on Concrete. mor. III. Revised and Enlarged 8vo. Philbrick's Field Manual for Engineers 16mo. $3 00 Richardson's Modern Asphalt Pavements 5 00 Richey's Building Foreman's Pocket Book and Ready Reference. * Trautwine's Field Practice of Laying Out Circular Curves for Railroads. Properties. 16mo.8vo. A Treatise on Brasses. Railroad Construction Wellington's Economic Theory of the Location of Railways Wilson's Elements of RailroadTrack and Construction Large 12mo. Bronzes. 4 00 * Stone and Brick Masons' Edition (Building Mechanics' Ready Reference Series) 16mo. Sabin's Industrial and Artistic Technology of Paint and Varnish 3 00 Smith's Strength of Material 12mo. Railroad Engineer's Field Book. mor. mor. 2 SO 2 00 2 50 5 00 5 00 2 00 Method of Calculating the Cubic Contents bankments by the Aid of Diagrams Webb's Economics of Railroad Construction * of Excavations and Em8vo. . 16mo.) Vol. 4to. 3 volumes. and Uses 8vo. 16mo. mor. 2 00 Spalding's Hydraulic Cement Textbook on Roads and Pavements 12mo. 2 50 8vo. mor. 5 00 * Ries and Leighton's History of the Clayworking Industry of the United States 8vo. 8vo. .mor. 8vo. 16mo. mor. mor. II. Vol.) I. 50 00 3 00 3 00 3 00 1 3 50 3 00 1 1 50 50 12mo. 5 00 1 50 2 50 1 50 1 50 * 50 5 00 25 2 50 1 1 bankments Ives and Hilts's Problems in Surveying. 12mo. mor. Railroad Spiral 16mo. * Cement Workers' and Plasterers' Edition (Building Mechanics' Ready Reference Series) 16mo. 16mo. Vol. 2 00 Wood's (M.) Rustless Coatings: Corrosion and Electrolysis of Iron and Steel 8vo. 2 00 8vo. 1 25 16mo. Plain and Reinforced 5 00 In Three Parts. and Other Alloys and their Constituents 8vo. 3 50 Part III. 8vo. 16mo. Nonmetallic Materials of Engineering and Metallurgy. . Raymond's Railroad Engineering. mor.) Treatise on the Resistance of Materials. (In Preparation. Iron and Steel 8vo. 8vo. Railroad Surveying and Geodesy 8vo. Railroad Field Geometry. Papei Cardboard. 8 00 Part I. mor. 3 50 Waterbury's Cement Laboratory Manual 12mo. 16mo. (In Preparation. 1 50 * Ries's Clays Their Occurrence. Tillson's Street Pavements and Paving Materials 4 00 Turneaure and Maurer's Principles of Reinforced Concrete Construction. Andrews's Handbook for Street Railway Engineers Berg's Buildings and Structures of American Railroads Brooks's Handbook of Street Railroad Location Butts's Civil Engineer's Fieldbook Crandall's Railway and Other Earthwork Tables Transition Curve * Crockett's Methods for Earthwork Computations Dredge's History of the Pennsylvania Railroad. Large 12mo. 1 50 Handbook for Superintendents of Construction 16mo. 16mo. 00 Molitor and Beard's Manual for Resident Engineers Nagle's Field Manual for Railroad Engineers * Orrock's Railroad Structures and Estimates 8vo. Taylor's Prismoidal Formulse and Earthwork 8vo. Second Edition. mor. 1 00 Wood's (De V. Elements of Railroad Engineering 8vo. 2 50 8vo. RAILWAY ENGINEERING. mor. mor. 4 00 : . (1879) Fisher's Table of Cubic Yards 3X5 inches. P. Hudson's Tables for Calculating the Cubic Contents of Excavations and Em.8vo.. 2 00 Part II. Godwin's Railroad Engineers' Fieldbook and Explorers' Guide. Thurston's Materials of Engineering. Searles's Field Engineering 16mo. mor. Snow's Principal Species of Wood 3 50 12mo. mor.
4 2 2 2 2 00 50 50 00 00 2 50 1 50 3 00 Kimball and Barr's Machine Design.. and Perspective 8vo. Elements of Plane and Solid Freehand Geometrical Drawing. $2 50 8vo. Kinematics of Machinery 8vo. (Thompson. W. Barr's Kinematics of Machinery * Bartl&tt's Mechanical Drawing 8vo.) MacCord's Elements of Descritpive Geometry Kinematics. 3 00 5 00 8vo. and Perspective 8vo. Form.) and Marx's Machine Design 8vo. 8vo.) . Part II. M. Theorems.. 4to. T. 4 00 1 50 1 50 1 50 3 50 2 00 5 00 2 00 3 00 3 00 3 00 3 00 2 50 1 25 1 25 3 50 7 50 1 00 3 00 Elementary Projection Drawing 12mo.DRAWING. Emch's Introduction to Projective Geometry and its Application 8vo. Schwamb and Smith Merrill's Elements of Mechanism 8vo. S. Durley's Kinematics of Machines 8vo. Manual of Drawing Coolidge and Freeman's Elements of General Drafting for Mechanical Engineers Oblong 4to. 8vo. 8vo. 8vo. 8vo. Jones's Machine Design: Part I. and Proportions of Parts 8vo. French and Ives' Stereotomy 8vo. 12mo. Anthony and Brackett's Textbook of Anthony and Ball's Lecturenotes on the Theory of Electrical Measure12mo. Svo. Manual of 00 50 1 25 2 50 1 1 Klein. (In Press. Shadows. Elements of Mechanical Drawing 8vo. Physics.) Moyer's Descriptive Geometry Reed's Topographical Drawing and Sketching 4to. Problems. . Elements of Descriptive Geometry. Hill's Textbook on Shades and Shadows. Practical Mechanism Mechanical Drawing Velocity Diagrams McLeod's Descriptive Geometry * Mahan's Descriptive Geometry and Stonecutting Industrial Drawing. Jamison's Advanced Mechanical Drawing Svo. Textbook of Mechanical Drawing and Elementary Machine Design. * Titsworth's Elements of Mechanical Drawing Oblong Svo. * Abegg's Theory of Electrolytic Dissociation. and Examples in Descriptive Geometry (Hermann and Weisbach's Kinematics and Power of Transmission. 8vo. Warren's Drafting Instruments and Operations 12mo. (von Ende. (Magie.. General Problems of Shades and Shadows 8vo.Large . 1 00 2 50 * Coolidge's ' " " Abridged Ed . ments 8vo. Large 12mo. Large 8vo. mor. Elements of Machine Construction and Drawing 8vo. 8vo. Smith's (R. (A. Strength. 8vo. Andrews's Handbook for Street Railway Engineering 3X5 inches. 8vo. . 5 00 3 50 1 50 1 00 2 50 3 00 ELECTRICITY AND PHYSICS. 1 50 8vo. Manual of Elementary Problems in the Linear Perspective of Forms and Shadow 12mo. paper. (McMillan) 8vo.) Descriptive Geometry Freehand Lettering Freehand Perspective Woolf's Elementary Course in Descriptive Geometry 8vo.) Manual of Topographical Drawing. 3 00 8vo. . or. 3 00 3 00 10 . Reid's Course in Mechanical Drawing 8vo. Plane ProbVems in Elementary Geometry 12mo. ) Wilson's (H. 12mo. 8vo.) 12mo. Robinson's Principles of Mechanism 8vo. Benjamin's History of Electricity Voltaic Cell 25 25 3 00 1 1 1 00 8vo.) Topographic Surveying * Wilson's (V.
25 5 00 2 50 4 00 3 00 Flather's Duhem's Thermodynamics and Chemistry. mor. 8vo.. * Karapetoff 's Experimental Electrical Engineering 8vo. . Holman's Precision of Measurements Telescopic Mirrorscale Method. 8vo.). Baker's Elliptic Functions (Bocher) Briggs's Elements of Plane Analytic Geometry. Vols. 8vo Law of Contracts Law of Operations Preliminary to Construction in Engineering and Architecture 8vo. 1 Morgan's Outline of the Theory of Solution and its Results * Physical Chemistry for Electrical Engineers 12mo. 2 00 3 00 1 12mo. .8vo. (von Ende.) Dawson's "Engineering" and Electric Traction Pocketbook. 3 (Lorenz) 8vo. I .Large 8vo. 1 1 50 00 00 11 . (Merriam. 3 * Rosenberg's Electrical Engineering. (Burgess. 3 * Lyons's Treatise on Electromagnetic Phenomena. 1 12mo.) 12mo. mor. 8vo. 3 00 12mo. 16mo. * Treatise on the Military Law of United States 8vo. New Edition. . 6 8vo.. Vol. half mor. Wait's Engineering and Architectural Jurisprudence Sheep.) . 3 . 8vo. Kinzbrunner's Testing of Continuouscurrent Machines 2 (Tingle. Large 12mo. 8vo. * Brennan's Handbook of Useful Legal Information for Business Men. Adjustments. * Collins's Manual of Wireless Telegraphy and Telephony 8vo. and Electric. 00 8vo. 5 00 2 50 7 00 2 50 1 50 6 00 6 50 3 00 5 00 5 50 MATHEMATICS. 12 Reagan's Locomotives: Simple. (Haldane Gee Kinzbrunner. I 2 12mo. Ryan. 3 Le Chatelier's Hightemperature Measurements. 1 * Norris's Introduction to the Study of Electrical Engineering 8vo. * Parshall and Hobart's Electric Machine Design 4to. each. Norris. mor. * Lyndon's Development and Electrical Distribution of Water Power. LAW. . Compound. )12mo.8vo. 8vo. Manual for 8vo. * Mor. 16mo. Hobart and Ellis's Highspeed Dynamo Electric Machinery 8vo. 3 Lob's Electrochemistry of Organic Compounds. * Hanchett's Alternating Currents 12mo. 2 50 6 00 2 00 — 75 00 00 00 00 00 00 00 00 00 50 50 50 50 00 50 00 50 00 00 — . 2 Tory and Pitcher's Manual of Laboratory Physics Ulke's Modern Electrolytic Copper Refining 8vo. 1 50 Crehore and Squier's Polarizing Photochronograph * Danneel's Electrochemistry.) Landauer's Spectrum Analysis.. Large 12mo. Dolezalek's Theory of the Lead Accumulator (Storage Battery). 16mo. 12mo.and II. .) Dynamometers. Betts's Lead Refining and Electrolysis Classen's Quantitative Chemical Analysis by Electrolysis. 6 * Michie's Elements of Wave Motion Relating to Sound and Light 8vo. 4 12mo. * Dudley's Military Law and the Procedure of Courtsmartial.•••• 8vo. Schapper's Laboratory Guide for Students in Physical Chemistry 1 * Tillman's Elementary Lessons in Heat 8vo. Courtsmartial 16mo. mor. and Hoxie's Electrical Machinery. . 1 Large 12mo. * Davis's Elements of Law . * Buchanan's Plane and Spherical Trigonometry 8vo. and Tests. (Boudouard Burgess. 2 Norris and Denmson's Course of Problems on the Electrical Characteristics of (In Press. Skeep. $4 00 (Boltwood.) Circuits and Machines. 2 8vo. and the Measurement of Power Getman's Introduction to Physical Science (Mottelay) Gilbert's De Magnete.8vo. 2 50 1 Hering's * Ready Reference Tables (Conversion Factors) 12mo.
No. Maurer's Technical Mechanics Merriman's Method of Least Squares Solution of Equations Rice and Johnson's Differential and Integral Calculus. (Truscott and Emory. 8vo. 50 50 1 25 2 50 1 00 1 50 1 75 1 50 1 00 1 1 100 copies. No. Curve Tracing in Cartesian Coordinates 12mo. 8vo. Large 12mo. : Hyde's Grassmann's Space Analysis Johnson's (J. W. inches. 12mo. 8vo. Smith's History of Modern Mathematics * Veblen and Lennes's Introduction to the Real Infinitesimal Analysis of One Variable 8vo.) Abridged Editions of Differential and Integral Calculus. No.) Threeplace Logarithmic Tables * Vestpocket size. Theory of Errors and the Method of Least Squares 12mo. Fiske. Weld's Determinants 8vo. and Theory of Errors. McMahon's Hyperbolic Functions 8vo. Elementary Treatise on Differential Calculus Large 12mo. * Trigonometry and Tables published separately Each. 11. B. Series 25 00 Woodward by Thomas S. * Ludlow's Logarithmic and Trigonometric Tables 8vo. Edward Hyde.Large 12mo. by James McMahon. No. 4. 8vo. by Robert S. by William Woolsey Johnson. by George Bruce Halsted. 8vo. . 8 X 10 10 copies. by William E. 4 00 2 00 1 00 1 1 2 vols. No. by Laenas Gifford Weld. Harmonic Functions. Grassmann's Space Analysis. Determinants. 12mo. No. 8vo. * Waterbury's Vest Pocket Handbook of Mathematics for Engineers. 3. * Theoretical Mechanics 12mo. 2JX5 8vo. Wood's Elements of Coordinate Geometry 8vo. * Ludlow and Bass's Elements of Trigonometry and Logarithmic and Other Tables 8vo.) 12mo. Macfarlane's Vector Analysis and Quaternions 8vo. 5. Elementary Treatise on the Differential Calculus 8vo. by Mansfield Merriman. 15 5 00 25 2 00 2 50 1 1 1 1 3 1 3 3 3 00 00 50 50 00 50 00 00 50 (In Preparation. 8. 12mo. S. Large 12mo. in one. Functions of a Complex Variable. Treatise on the Integral Calculus Large 12mo. paper. Treatise on Ordinary and Partial Differential Equations. Synthetic Projective Geometry. Large 12mo. Byerly. Woodward's Probability and Theory of Errors 3 00 00 2 00 00 00 2 00 1 00 1 1 12 . Karapetoff's Engineering Applications of Higher Mathematics. 8vo. 2 00 2 50 Elements of Geometry * Rational Geometry 12mo. 10. Manning's Irrational Numbers and their Representation by Sequences and . mor. No. Elementary Treatise on the Integral Calculus Large 12mo. 8vo. each No. inches. No. 7. No. Hyperbolic Functions. .. Differential Equations. by David Eugene Smith. Treatise on Differential Calculus Large 12mo. $1 00 Compton's Manual of Logarithmic Computations * Dickson's College Algebra * Introduction to the Theory of Algebraic Equations Emch's Introduction to Projective Geometry and its Application Fiske's Functions of a Complex Variable Halsted's Elementary Synthetic Geometry 12mo. 2 00 3 00 2 00 1 1 1 1 1 00 00 00 Mathematical Monographs. 2.) Laplace's Philosophical Essay on Probabilities. Johnson's (W. Vector Analysis and Quaternions. The Solution of Equations. History of Modern Mathematics. Woodward. 50 Large 12mo. 6. Large 12mo. 1 vol. Edited by Mansfield Merriman and Robert Octavo. Mounted on heavy cardboard. No. Probability by W. Differential Equations 8vo. 1. by Alexander Macfarlane.Byerley's Harmonic Functions Chandler's Elements of the Infinitesimal Calculus * Coffin's Vector Analysis gvo. 9.
3 00 Preston. Flather's Dynamometers and the Measurement of Power 12mo. 8vo. .. 3 00 16mo. Gearcutting Machinery Large Flanders's 12mo. 8vo. 5 00 Kent's Mechanical Engineer's PocketBook 8vo. mor.) Levin's Gas Engine. 8vo. 2 50 8vo. Dingey's Machinery Pattern Making Durley's Kinematics of Machines 8vo.) Pressworking of Metals 8vo. 8vo. 5 00 Mechanical Drawing 4to.) 8vo. Kerr's Power and Power Transmission 2 00 Kimball and Barr's Machine Design. W. 8vo. paper. 3 50 Mehrtens's Gas Engine Theory and Design Large 12mo. or. 2 50 Oberg's Handbook of Small Tools Large 12mo. 8vo. 1 50 Mahan's Industrial Drawing. 12mo. 8vo. 2 00 Textbook of Mechanical Drawing and Elementary Machine Design.) 8vo. (Thompson.) (In Press. 3 00 Schwamb and Merrill's Elements of Mechanism 8vo. 3 00 (Woodward and Sorel's Carbureting and Combustion in Alcohol Engines. MATERIALS OF ENGINEERING. Small 4to. 8vo. 4 00 * Lorenz's Modern Refrigerating Machinery. 12mo.MECHANICAL ENGINEERING. 8vo. 8vo. 3 00 Poole's Calorific Power of Fuels 8vo. $1 50 12mo. 8vo. 4 00 Velocity Diagrams 8vo. Haven.) Greene's Pumping Machinery. and Proportions of Parts 8vo. 8vo. (In Press. Strength. 3 00 Richards's Compressed Air 12mo. . 3 00 8vo. 1 50 MacFarland's Standard Reduction Factors for Gases 8vo. Coolidge's Manual of Drawing •Coolidge and Freeman's Elements of Geenral Drafting for Mechanical EnOblong 4to. STEAMENGINES AND BOILERS. 3 00 Smith (A. Large 12mo. (Pope. half leather. Hering's Ready Reference Tables (Conversion Factors) * Hobart and Ellis's High Speed Dynamo Electric Machinery 16mo. Bacon's Forge Practice Baldwin's Steam Heating for Buildings Barr's Kinematics of Machinery * Bartlett's Mechanical Drawing " " " * Abridged Ed * Burr's Ancient and Modern Engineering and the Isthmian Canal Experimental Engineering Carpenter's Heating and Ventilating Buildings (New edition in press. 2 50 8vo. 8vo. 3 50 Stone's Practical Testing of Gas and Gas Meters 13 . (In Preparation.) and Marx's Machine Design 8vo. and Dean). 4 00 MacCord's Kinematics. 3 00 * Porter's Engineering Reminiscences. 8vo. Cromwell's Treatise on Belts and Pulleys 12mo. 3 00 * Parshall and Hobart's Electric Machine Design. 12 50 Peele's Compressed Air Plant for Mines 8vo. Leonard's Machine Shop Tools and Methods. Practical Mechanism 8vo. 150 3 50 6 00 4 00 1 1 1 12mo. mor. 8vo. 50 50 00 2 50 1 1 50 50 2 00 4 00 3 00 3 00 2 00 1 25 2 00 Rope Driving Gas and Fuel Analysis Goss's Locomotive Sparks Gill's for Engineers 12mo. Treatise on Toothed Gearing 12mo. 1855 to 1882 8vo. 1 50 Part II. 3 00 Smith's (O. Form.) Clerk's Gas and Oil Engine. Hutton's Gas Engine Jamison's Advanced Mechanical Drawing Elements of Mechanical Drawing Jones's Gas Engine 50 00 00 00 50 4 00 2 6 5 2 2 Machine Design: Kinematics of Machinery Part I. Compton's First Lessons in Metal Working Compton and De Groodt's Speed Lathe 12mo. 3 00 Reid's Course in Mechanical Drawing 8vo. gineers 12mo. 1 50 Robinson's Principles of Mechanism 8vo.
12mo. Maire's Modern Pigments and their Vehicles 12mo. Part I. 2 00 1 50 2 50 5 00 5 00 1 00 6 00 5 00 2 00 5 00 5 00 4 00 14 . and Varnishes. and Other Alloys and their 8vo. Bovey's Strength of Materials and Theory of Structures Burr's Elasticity and Resistance of the Materials of Engineering * Church's Mechanics of Engineering * Greene's Structural Mechanics * Holley's Lead and Zinc Pigments Large 12mo Holley and Ladd's Analysis of Mixed Paints. Hutton's Heat and Heatengines . P. 8vo. 8vo. Merriman's Mechanics of Materials 8vo. Wood's (De V. (Henning. 8vo. Kent's Steam boiler Economy . 50 50 6 00 2 50 3 00 7 7 2 50 3 00 6 00 2 50 7 50 2 00 7 50 4 00 5 00 1 00 2 00 3 00 1 00 8 00 2 00 3 50 2 50 3 00 2 00 Steel 8vo.). 3 00 16mo.) Machinery of Transmission inches.) Strength of Material 12mo. Ford's Boiler Making for Boiler Makers * Gebhardt's Steam Power Plant Engineering 8vo. 4 00 STEAMENGINES AND BOILERS. Treatise on Friction and Lost Work in Machinery * Tillson's Complete Automobile Instructor * Titsworth's Elements of Mechanical Drawing and Mill Work. 8vo. * Strength of Materials 12mo. Klein. Bronzes. 8vo.. $1 00 .) Martens's Handbook on Testing Materials. Smith's (H.8vo. Wood's (M.) Rapid Methods for the Chemical Analysis of Special Steels. Nonmetallic Materials of Engineering 8vo. 1 50 Oblong 8vo. Berry's Temperatureentropy Diagram (Thurston.8vo. 7 50 . 5 00 5 00 2 50 MATERIALS OF ENGINEERING. Johnson's (J.) Elements of Analytical Mechanics Treatise on the Resistance of Materials and an Appendix on the Preservation of Timber 8vo. 8vo Maurer's Techincal Mechanics 8vo. 12mo. Part II. 12mo. Iron and Steel Part III. . Carnot's Reflections on the Motive Power of Heat.) Rustless Coatings: Corrosion and Electrolysis of Iron and . Sabin's Industrial and Artistic Technology of Paint and Varnish Smith's ((A. Klein. Dawson's "Engineering" and Electric Traction Pocketbook. 1 00 (Herrmann and Governors. 8vo. A Treatise on Brasses. . 18mo. 8vo. M. SteelMaking Alloys and Graphite Large 12mo. Mechanical Engineering of Power Plants 8vo. 16mo. Constituents 8vo. 8vo. Keep's Cast Iron 8vo. Thurston's Materials of Engineering 3 vols. E.) 12mo. 12mo. Color Pigments. Warren's Elements of Machine Construction and Drawing * Waterbury's Vest Pocket Handbook of Mathematics for Engineers. (Hermann — Wood's Turbines 8vo. mor.— Thurston's Animal as a Machine and Prime Motor. A Manual for Steelusers Metcalf's Steel.8vo. W. 2JX5 Weisbach's Kinematics and the Power of Transmission.. and the Laws of Energetics. Johnson's (C. 8vo. Hemenway's Indicator Practice and Steamengine Economy 8vo. B. mor.) Materials of Machines 12mo. 1 25 8vo. Lanza's Applied Mechanics 8vo. Large 12mo. Goss's Locomotive Performance 12mo.) Materials of Construction 8vo. Chase's Art of Pattern Making Creighton's Steamengine and other Heat Motors 8vo.
12mo. their Designs. Peabody's Manual of the Steamengine Indicator 12mo. Design. Wood's Thermodynamics. Whitham's Steamengine Design 8vo. Structure. History. 8vo. Moyer's Steam Turbine 8vo. 8vo. Elementary Mechanics 12mo. 6 00 2 00 1 50 Mechanics of Engineering. 25 50 Maurer's Technical Mechanics * Merriman's Elements of Mechanics Mechanics of Materials * Michie's Elements of Analytical Mechanics Robinson's Principles of Mechanism Sanborn's Mechanics Problems 4 00 1 00 5 00 4 00 3 00 1 50 3 00 3 00 1 25 12mo. 8vo 8vo. 8vo. 6 8vo. and Kinetics. (Patterson). Dana's Textbook of Elementary Mechanics for Colleges and Schools . W. Valvegears for Steamengines 8vo. Part I. and Marshall's Elements of Steamengineering 8vo. 12mo. Vol.) 8vo. Schwamb and Merrill's Wood's Elements Principles of Elements of Mechanism of Analytical Mechanics Large 12mo.. I. 2vols. Church's Mechanics of Engineering 8vo.12mo. Large 12mo. Snow's Steamboiler Practice 8vo. 2 50 2 00 3 00 7 50 1 1 Large 12mo. 6 1 12mo. of the Steamengine and Operation 8vo. and the Use Handy Manual Manual 8vo. Du Bois's Elementary Principles of Mechanics: Vol. and Operation Steamboiler Explosions in Theory and in Practice Wehrenfenning's Analysis and Softening of Boiler Feedwater. Reagan's Locomotives: Simple. Statics. 8vo Thomas's Steamturbines Thurston's Handbook Tables of cator and the Engine and Boiler Prony Brake Trials. 15 . and Refrigerating Machines. II. Peabody and Miller's Steamboilers 8vo.iss's Practice and Theory of the Injector 8vo* MacCord's Slidevalves g vo Meyer's Modern Locomotive Construction 4to. 8vo. Kinematics 12mo. 50 00 00 00 00 50 4 00 5 00 5 00 4 00 MECHANICS PURE AND APPLIED. Heat Motors. Kinematics 8vo. 8vo. * Johnson's (W. Thermodynamics of the Steamengine and Other Heatengines. $1 50 2 00 10 00 4 00 1 50 1 00 5 00 2 50 4 00 1 25 Valvegears Spangler. Smart's Handbook of Engineering Laboratory Practice 12mo. Sinclair's Locomotive Engine Running and Management 12mo. II. Steam. Construction. 8vo. Compound. . 4 00 Small 4to.) 12mo. Statics * Vol. 5 10 8vo. . . II * Greene's Structural Mechanics James's Kinematics of a Point and the Rational Mechanics of a Particle. Lanza's Applied Mechanics * Martin's Text Book on Mechanics. . 8vo. Tables of the Properties of Steam and Other Vapors and TemperatureEntropy Table 8vo. 8vo. 7 50 Small 4to! 10 00 8vo. 50 00 50 00 1 00 2 50 3 00 4 00 3 2 2 3 5 00 1 of Steamboilers. I. (Osterberg. and Electric. Notes and Examples in Mechanics 8vo. Construction. and Steamengines. 8vo. Vol. Weisbach's Heat. 8vo. and Theory Part II. Voi. Greene.) Theoretical Mechanics 3 50 8vo. Spangler's Notes on Thermodynamics 12mo. (Du Bois. of the Indi8vo. New Edition. I Vol. Pupin's Thermodynamics of Reversible Cycles in Gases and Saturated Vapors. .
de Fursac's Manual of Psychiatry. mor. Metcalf 's Steel. Thurston's Materials of Engineering.Large 12mo. 1 50 tions 8vo. 8vo. (Fischer. Part II. Keep's Cast Iron 8vo. . page 9. A Manual for Steelusers 12mo. Ruer's Elements of Metallography. (Orndorff. * Personal Hygiene 12mo. Nonmetallic Materials of Engineering. Whys in Pharmacy 12mo. 8vo.) 1 25 Jackson's Directions for Laboratory Work in Physiological Chemistry . Ulke's Modern Electrolytic Copper Refining 8vo. Bordet's Studies in Immunity.) LassarCohn's Practical Urinary Analysis. (Mandel. (In Press.) Ehrlich's Collected Studies on Immunity. mor. Part I. * Fischer's Physiology of Alimentation Large 12mo. 4 00 3 00 2 50 2 50 1 Foundry Terms used 12mo. 2 00 2 50 (Rosanoff and Collins.) . Iron Founder Supplement Goesel's Minerals and Metals: * Iles's Leadsmelting Douglas's Untechnical Addresses on Technical Subjects A Reference Book 00 3 00 2 50 3 00 2 50 3 00 2 00 2 50 1 Johnson's Rapid Methods for the Chemical Analysis of Special Steels. Betts's Lead Refining by Electrolysis. and Other Alloys and their Constituents 8vo. 1 00 12mo. Steelmaking Alloys and Graphite Large 12mo.) (Gay). * Satter lee's Outlines of Human Embryology 12mo. 1 50 Bolduan's Immune Sera 8vo. 4 00 Hammarsten's Textbook on Physiological Chemistry. A Treatise on Brasses. (Cohn. 2 50 1 25 2 50 3 00 1 1 50 00 1 25 METALLURGY.8vo.). (Bolduan.MEDICAL. Iron and Steel 8vo.). Worcester and Atkinson's Small Hospitals Establishment and Maintenance. Smith's Lecture Notes on Chemistry for Dental Students 8vo.). 12mo. * Whipple's Tyhpoid Fever Large 12mo. 1 00 Salkowski's Physiological and Pathological Chemistry. (Waldo. 2 00 Ruddiman's Incompatibilities in Prescriptions * Abderhalden's Physiological Chemistry . 12mo. (Hall and in Thirty Lectures.) .) 8vo. 6 00 (Bolduan. $5 00 Defren. with Plans for a Small Hospital 12mo.) 12mo. Part III. (Mathewson) Smith's Materials of Machines 12mo. 8vo. 1 50 Mandel's Handbook for the BioChemical Laboratory * Pauli's Physical Chemistry in the Service of Medicine. Bronzes.. . Woodhull's Notes on Military Hygiene 16mo.. 00 Tate and Stone's Foundry Practice 12mo. Bolland's Encyclopedia of Founding and Dictionary of in the Practice of Moulding 8vo. . 12mo. 12mo.. 1 00 Rostoski's Serum Diagnosis.12mo. 12mo.) 12mo. Moulders' Text Book 12mo. 1 00 von Behring's Suppression of Tuberculosis. (Boudouard Burgess.) 12mo. Minet's Production of Aluminum and its Industrial Use. (Bolduan. . 1 25 * PozziEscot's Toxins and Venoms and their Antibodies. 12mo. 2 00 8 00 3 50 2 50 3 00 2 50 2 50 16 . 16mo. Le Chatelier's Hightemperature Measurements. — . Davenport's Statistical Methods with Special Reference to Biological Varia16mo. 12mo. West's American Foundry Practice 12mo. 12mo. . In Three Parts 8vo. and Suggestions for Hospital Architecture. 1 00 (Lorenz. see Civil Engineering.8vo.
3 2 5 4 5 1 Douglas's Untechnical Addresses on Technical Subjects Modern High Explosives Goesel's Minerals and Metals: A Reference Book Eissler's 12mo. Iddings's Igneous Rocks 8vo. How to Study Them Large 8vo. * Penfield's Notes on Determinative Mineralogy and Record of Mineral Tests. 60 4 00 5 00 50 8vo..8vo. 8vo. . 1 00 Dana's Second Appendix to Dana's New "System of Mineralogy. for Mines Difficult Conditions. Baskerville's Chemical Elements. 8vo. 12mo.. Ries and Leighton's History of the Clayworking Industry of the United 8vo.MINERALOGY. Pocketbook form. 8vo. paper. (In Preparation. mor. * * * * Tables of Minerals. 16mo. 8vo.) 8vo. Treatise on Practical and Theoretical Mine Ventilation 12mo. 12 50 12mo. pipe Nonmetallic Minerals: Their Occurrence and Uses 8vo. 1 50 Mineralogy. System of Mineralogy Textbook of Mineralogy Technical Subjects Addresses on Untechnical Douglas's Minerals and 12mo. Pirsson's Rocks and Rock Minerals 12mo. Eckel's Stone and Clay Products Used in Engineering. Properties and Uses 8vo. Johannsen's 4 00 1 00 1 25 3 00 1 1 25 50 5 00 5 00 With Thumb Index * Martin's Laboratory Merrill's 5 00 Guide to Qualitative Analysis with the '. Richards's Synopsis of Mineral Characters 12mo. (Penfield. 3 00 Butler's Pocket Handbook of Minerals 8vo." Large 8vo. rewritten Wilson's Hydraulic and Placer Mining. 8vo. * Weaver's Military Explosives 8vo. 8vo. * 8vo. Pocketbook form. 8vo. 2d edition. 8vo. 8vo. 5 00 Dana's First Appendix to Dana's New " System of Mineralogy" Large 8vo. 2 50 2 00 2 oo MINING. mor. mor. 1 25 * Crane's Gold and Silver 8vo.. 8vo. 1 00 Chester's Catalogue of Minerals Cloth. Compressed Air Plant 00 00 00 00 00 00 00 00 00 50 00 00 00 50 25 17 . Groth's Introduction to Chemical Crystallography (Marshall). mor. Goesel's Minerals and Metals: A Reference Book 12mo. . 8vo. 00 2 50 1 25 5 00 1 States * Tillman's Textbook of Important Minerals and Rocks Washington's Manual of the Chemical Analysis of Rocks. Riemer's Shaft Sinking Under (Corning and Peele). * Hayes's Handbook for Field Geologists 16mo.. 1 50 half leather. mor.. Ries's Clays: Their Occurrence. Including the Use of Minerals and Statistics of Domestic Production 8vo. 2 00 Manual of Mineralogy and Petrography . Blow12mo. Ihlseng's Manual of Mining * Iles's Lead Smelting Peele's 4 3 5 2 3 3 3 2 1 12mo.). mor. Rock Minerals of Rockforming Minerals in Thin Determination Sections. . Eakle's Mineral Tables (In Preparation). . 12mo.Large 12mo. * Beard's Mine Gases and Explosions Boyd's Map of Southwest Virginia * Crane's Gold and Silver * Index of Mining Engineering Literature . 4 00 Brush's Manual of Determinative 16mo. 16mo. $2 00 Boyd's Map of Southwest Virginia * Browning's Introduction to the Rarer Elements 8vo. Stones for Building and Decoration 8vo. paper.
3 00 Value of Pure Water Winslow's Systematic Relationship of the Coccaceae Large 12mo.) 8vo. 2 00 Parsons's Disposal of Municipal Refuse 8vo.) Whipple's Microscopy of Drinkingwater 8vo. 1 25 12mo. Large 12mo.SANITARY SCIENCE. 4 00 Soper's Air and Ventilation of Subways 12mo. 8vo. (Chemical and Bacteriological) Watersupply. 4 00 * Merriman's Elements of Sanitary Enigneering 8vo. 3 00 Sanitation of Public Buildings 12mo. Emmons's Geological Guidebook of the Rocky International Congress of Geologists Ferrel's Popular Treatise on the Winds Fitzgerald's Boston Machinist Gannett's Statistical Abstract of the World Mountain Excursion of the Large 8vo. Folwell's Sewerage. 1 00 2 50 MISCELLANEOUS. 2 00 Fuertes's Waterfiltration Works 12mo. 5 00 Venable's Garbage Crematories in America 8vo. Association of State and National Food and Dairy Departments. Water. 3 00 8vo. 8vo. 2 50 Water and Public Health 12mo. 2 00 Method and Devices for Bacterial Treatment of Sewage 8vo. 1 50 Gerhard's Guide to Sanitary Inspections 12mo. 2 00 * Richey's Plumbers'. A Study in Dietaries 12mo. Hartford 8vo. (In Press. 4 00 Watersupply Engineering Fowler's Sewage Works Analyses 12mo. and Food from a Sanitary Standpoint 8vo. 1906 8vo. 2 50 Turneaure and Russell's Public Watersupplies 8vo. 24mo. Leach's Inspection and Analysis of Food with Special Reference to State Control 8vo. 2 00 Prescott and Winslow 's Elements of Water Bacteriology. Winslow and Pratt's Purification of Sewage. 1 50 Richards and Woodman's Air. 3 00 (In Preparation. $3 00 Meeting. 3 00 Jamestown Meeting. with Special Reference to Sanitary Water Analysis 12mo. Construction. 7 50 12mo. Steamfitters'. 1 00 * Richards and Williams's Dietary Computer 8vo. 8vo. Sewage and Bacterial Purification of Sewage 8vo. Hanusek's The Microscopy of Technical Products. 1907 * Bashore's Outlines of Practical Sanitation 12mo. 1 50 * Price's Handbook on Sanitation 12mo. 1 50 Large 12mo. 1 00 (Designing. (Considered principally from a Sanitary Standpoint). Haines's American Railway * Management (Winton) 18mo. 2 00 Ogden's Sewer Design 12mo. 1 50 Rideal's Disinfection and the Preservation of Food 4 00 8vo. 1 50 Hazen's Clean Water and How to Get It Filtration of Public Watersupplies 8vo. mor. 1 00 Cost of Shelter 12mo. 1 00 Cost of Food. and Maintenance. 1 25 Mason's Examination of Water. 1 50 4 00 1 00 75 2 50 5 00 18 . 1 50 * Modern Baths and Bath Houses 8vo. 3 50 * Typhoid Fever Large 12mo.) Kinnicut. 12mo. 1 50 Richards's Cost of Cleanness 12mo. and Tinners' Edition (Building Mechanics' Ready Reference Series) 16mo. 1 00 Cost of Living as Modified by Sanitary Science 12mo. 1 00 Sanitation of a Country House Sanitation of Recreation Camps and Parks 12mo. 3 00 Ward and Whipple's Freshwater Biology.
and the Administration of Workshops. 3 00 Rotherham's Emphasised New Testament Large 8vo. 8vo. 8vo. half mor. Large 12mo. 5 00 1 25 Hebrew and Chaldee Lexicon to the (Tregelles. Glass. 2 00 Thome's Structural and Physiological Botany. $3 50 Metcalfe's Cost of Manufactures. ganized Industrial Efficiency 8vo. 1 50 HEBREW AND CHALDEE Gesenius's TEXTBOOOKS. 2 00 8vo. etc 12mo.A Collection of Published Papers on OrJacobs's Betterment Briefs. 12mo. (Bennett) 16mo. Small 4to..) Green's Elementary Hebrew Grammar 19 . Metal. 2 25 (Schneider) Westermaier's Compendium of General Botany. 5 00 Putnam's Nautical Charts 8vo. Azimuth and Starfinding Tables 5 00 Standage's Decoration of Wood. 2 00 Winslow's Elements of Applied Microscopy 12mo. Old Testament Scriptures. Rust's ExMeridian Altitude. 2 00 Ricketts's History of Rensselaer Polytechnic Institute 18241894.
.
.
.
.
.
"+ A * 8 <* ^ ^ <> ^ V> P > ^ d? ^.<< ^ \°<^ ^ rv .\V ^ * 8 .
LIBRARY OF CONGRESS H9MB .