You are on page 1of 19

Home

Search

Collections

Journals

About

Contact us

My IOPscience

Ductile Materials under Combined Stress

This article has been downloaded from IOPscience. Please scroll down to see the full text article. 1909 Proc. Phys. Soc. London 22 130 (http://iopscience.iop.org/1478-7814/22/1/311) View the table of contents for this issue, or go to the journal homepage for more

Download details: IP Address: 141.117.79.62 The article was downloaded on 27/05/2013 at 16:10

Please note that terms and conditions apply.

130

M R . W. A. SCOBLE O S DUCTILE

Theit necessaq- liiiiitations are also exctmined numerically. For a higll frequency, it is found that the change of self-inductance due to twisting of the wire tends to vmish, and that the change of resistance tends towards a value independent of the frequency.

DISCUSSION.
Dr. RUSSELT, stated that the Author's paper was most instructive and that he had done excellent pioneering mathematical work. This paper contained the first published attempt to get a solution of the very difficult problem of iinding the effective resistance and inductance of R helical coil when traversed by high frequency currents. Previously only cylindrical curwilt-sheets had been considered. H e pointed out that, in the particular case when the wire was very thin, an approximate value of the inductance could be found by counting the linkages of the magnetic lines of force with the helical current. Even when the coil was of finite length this presented no great difficulty. I n the Author's problem, howover, the wire WBS of finite thickness and so the difficulties to be overcome were much greater. He considered that the Author's solutions would prove very helpful to other workers in electromagnetic theory. congratulated the Author, and remarked that both Prof. C. H. LEES the functions occurring in Equation 39 had been tabulated by Nr. Savidge. Prof. Lees indicated the form of the graphs of these functions. The AUTHOR, in reply to Prof. Lees, said the tabulation of the functions referred to was most valuable.

x. Ductile Mateieials under Combined Stress.


[Plate 111.1

By

WALTER A. SCOBLE, A. 2. C.Sc., B.Sc., Whitworth Scholar".


l r t tvoduction.

THE theory of conibined stress, and the results recorded by earlier writers, were discussed in a previous communicationt, and the literature of this subject has since been very fully reviewed, from an engineering standpoint, by Mr. L. B. Turner $ and Mr. C. A. Smith Q. The most important experimental. resu1t.s obtained by other observers are given later. The object of this paper is to consider furt,her the results
* t
4
Read Norember 26,1909. Proc. Phys. Soc. London, vol. xx., and Phil. Mag. Dec. 1908. Engineering, Feb. 6, 190% 5 Engineering, Aug. 20, 1909.

SCOBLE.

Proc.

FIG. 1.

FIG. 2.

Pliys. Soc. Vol. X X I I . , P 1 .111. FIG. 3 .

MOMENT

MOMENT.

FIG.

6.

FIG.4.

'hi,. el

e lI I I I I l I l I I I
I I
Maximum

FIQ. 4.

& Minimum
~ p& p , ~ ~

Princiba I Srrsssas.
cuc6r.i

5000 M i n i m u m PrinciCal

Srraor. lbs/s,on. x . COPPER . Tusr . X U

100oO

U BRASS

. .

,M

SE

)I \TERI.\LS UYDEB COBIEIXED STRESS.

131

of the writers earlier experiments, and to record the data which were obtained from some later tests made on tubes. The results are also compared with those obtained by other observers, who employed different methods and combinations of loading.

Furtl~e Consideration ~ of the Earlier Tests. Specitneizs ami Apparatus.-The first test bars were of steel, f inch diameter, and 36 inches long. The ends were of square section, 2 inch side, to allow a torque to be applied to the bar. One squared end was held in a special clamp wliich prevented this end of the bar from i*otating under the torque, but allowed i t to take its natural slope as a supported benin under a bending load. lhe bar was also supported, a t 30 inches from the centre of the clamp, on V rollers, which formed the other support under the bending load, and offered no resistance to the torsion of the bar. A bar was bent as a beam 30 inches long, supported at its ends, by a dead load directly applied midway between the supporh ; it was twisted by means of a wooden pulley which fitted on the other squared end of the bar. Two flexible wire ropes were attaohed to the pulley, and exerted a eouple upon i t because of the weights which they were made to carry, Therefore it will be noticed that when a bar was twisted, it was subjected to a uniform torque all along its length, but the maximum bending moment acted on one section only of the bar. Within the elastic limit of the metal, the shear s t r e s produoed by the torque varied from zoro at the axis oE the biir, to a maximum all over the surface. Similarly the maximuni stresses rhue to bending were a compression a.t the top of the bar, and a tension only a t its lowest point ; and these were confined to the mid-section, which was under the maximum bending moment. This system of loading represented the most common example of combined stress in engineering practice, which is a shaft subjected to combined bending and torsion, and i t had the further advantage that the critical stresses were produced by comparatively small loads, The bending deflexion was measured by means of a scale which rested on the bar, and was guided in a slide provided
K2

182

MR.

w.

A.

wowx

DUCTILE

with a vernier. This simple arrangement was quite satisfactory On such a long beam. The torsion was measured by a pointer clamped t o the bar, which moved over a fixed, finely-divided circle. The twist wns also measured a t three points by clamping mirrors to the bnr ; fixed telescopes ancl vertical scales were used in connesion with these mirrors.

The Cviterion of StyengtJt. The theory of elasticity is based on Hookes law, that strain is proportional to stress. Beyond the elastic limit of a material this law is no longer strictly true, and therefore the usual forinulze for calculating the stresses cannot be applied. Consequently it is clear that the elastic limit is the correct criterion of strength. But it is nom the usual practice to adopt the yield-point. Guest * stated that Hoolies law holds to the yield-point, and he considered that the first deviation from proportionality of stress to strain was caused by local variations in the material, which altered the stress distribution, and caused local yielding. Thus he assumed that stress was proportional to strain for the main part of the material until the yield-point was reached, and therefore lie selected the yield-point as the criterion of strength. Undoubtedly the yield-point o a material is less affected by special treatment than the elabtic limit, and it is. more easily determined, consequently other observers, following Guest, have adopted the yield-poin t. Unfortnnately opinions cliff er concerning the exact location of the yield-point, and for comparative purposes, in order to have a well-defined point, the writer also neglected the intermediate state between perfect elasticity and complete rield, and obtained the critical loads from the intersections of the continuations of the lines which rcprescnt t8hese two limiting states on the stress-strain curves. It is probable that Guests assumption is partly true, because, with the loading adopted by the writer, with which tl,e stresses are unequally distrihted, the ratio of the strain to the stress is constant until it increases very rapidly just before the complete bre:tkclown of the sl)eciiiien. Tho slight
f

Prcc. Pliye. Soc.. TJoiidoii, vol. srii. Sept. 1900.

MATERIALS UNDER COMBtSED STRESS.

133

deviation from Hookes law, which is noticed between the elastic liinit and the yield-point in a simple tension test, is not observed. I t therefore appenrs that for this kind of test the eIaPtic limit and the yield-point practically coincide, and therefore the stresges are now taken, in all these tests, from the point where the strain ceases to be proportioiial to tlie stress.

T Ire Quantities TctbulutetE.

Ths strains have not been tabulated because the masimum strain theory is not supported hy engineers i n this country. The forinulce relating to combined stress which are given in the test-books are based on tlie maximum stress theory. Recently, however, the shearing stress theory, or the stress difference theory 1 s elasticinns prefer t o name it, has rapidly p i n e d favour with engineers. It is, therefore, clear that the maximuin stress and the maximuin shearing stress are iiiost important from a practical standpoint, but the maxiinuni strain hypothesis is indirectly considered later, when the deviations from the shear-stress law are discussed.
Culculution of the St)*esses.

The maximuin tensile stress, $3, in tlie material due to Leiding is calculated froni tho formula

in which 8 1 is tlie maxiinuiil bending inoinent ; I is the nionieiit of inertia of the area of the section about its neutral axis, in this case a diameter ; y is the greatest distance of a point i n the section froin the neutral line, and ecpals half the diameter of the section. The maximuin shear stress, S, caused 1)y tlie torque, is calculated from tlie foraiula

134

MIt, W. A . SCOBLE OX DUCTILE

in wbich T is the torque ; D is the diameter of the bar. F o r the tubes employed in the later tests the formule becomes
7r T=-SD 4 C E 4

lti

L ,

in which D is the external diameter of a tube ; (2 is its internal diameter.


Having found p and S,the maximuin and miuiinum principal stresses are represented by the expressions

one of which is positive, and the other is negative. Tne third principal stress is zero, because in the case of solid bars the stress due to the shearing force ou the section is zero where the bending stress is a maximum. Since the tubes are bent by couples, there is no direct shearing force on : L cross-sec tion. T h e stress difference is the difference between the masimum and minimum principal stresses, and therefore eqnals

It is twice the maximum shearing dress.

The maximuin strain is given by


P,--?;)P,+ Pel E in which

PI is the maximum principal stress ; P, and P, are the other principal stresses j
71

is Poissons Ratio ;

E is Youngs Modulus of Elasticity.

MATERIALS UNDER COMBISED STRESS.

135

I n these tests there are stresses in one plane only, and the expression becomes y* --rP,

'

I f the maximum strain be constant, then P,-qPa must be constant, so that this hypothesis comes between the maximuin stress law, that P1 is constant, and the stress difference or shear-stress theory, that P1-PI is constant, but it is iiearer the former. The Results front Steel Bays.

TABLE A.-Original

Tests

OE Solid Steel Bars.


Stress
Minimum Principal Stress. Iba./s. in.
Differeuce =2ce Maximum Shear Stress.

Tensile Shear Maximum Stress Number Bending l'wisting Stress Moment Noment due to due to Principal of Stsess. lbs. ins. Lbs. ins. Bending Torque. 1bs.h. Ber. in. bs./s. in bs./s. in

-111. ..

0 16220
fLP2OO

--96600 - 19040 - 14320 - 12860 - 5100 - 7050 - 4900 - 1050 - 2328

--_

IV... XI...

0 667.5
1160

IX. VIII. VII.

XII. .. VI. ..

v...,
*.

..

..

1331 20()0 2000 8020 2320 2420

2190 2130 2033 1985 1360 1630 1335 645

980

32350 48600 48600 49000 56,300 58750

26600 25880 24700 24100 16520 19800 16220 7840 11900

26600 35260 42620 45210 53700 55650 53900 57350 61075

53200
5 4 3 m 5fi840

58070 58800 62700 58800


58400 63400

Only one test was iiiade on each bar. The bending and twisting moments a t yield are plotted in fig. 1(Pl. I.). I f the maximum principal stress were constant at the yield-point, the twisting moment under pure torque would be double the bending moment required to cause the bar to yield. I f the maximum shearing stress were constant, the plotted point,s would lie 011 a circle described with the origin as centre. Fig. 1 confirms the writer's original conchision that the inaxinium shear stress is approximately constant at failure, but that an ellipse lies between the points better than a circle, and that the bending moment is greater than the torque. The maximum stress varies between 26,600 and 61,075 Ibs./s. in,, it certainly is not constant. The stress difference

136

MR. W. A, SCOBLE O S DUCTILE!

is not esnctly constant ; it wries froin 53,200 to 63,400 I hs./s. in,, but it increases steadily with tlie bending moment, consequently the deviation from the stress difference: or shear stress law, is still further away from a constant inasiinuin stress, so that the maxiinnin strain also varies considerably.

2 % Apparatus Jor the Tests on Tubes. The tests on tubes were made in a inachiae which is part of the equipment of the engineering laboratory a t the (:ity and Guilds Technical College, Finsbury. The machine \vas designed by Prof. E. G. Coker, nnd it has been reccntly described *. A tube to be tested was swe:itcd, nucl \\-lien it wbs necessary pinned, on two deel holders, or inniidrils, ench of which was provided with two keyways. One 1io:der fitted iiit,o a warmwheel so t,hat this end of the tube could be twisted, but the worm wheel casing was pivotted on roller bearings so t h t there was 110 resistance to betiding. A Iengtl~eni~ig piece was fitted o:i the holder to carrj- a11 o1'erhung load, wliich prodiiced a bencling ~noincnton the silecimen. The other holder was held in :L sl)eci:il fitting wliich was support,ed on a spindle of sinall dianleter, so that the resistance to torsi 111 was negligible. The' fitting was also arranged to carry an overhaegirig load at this end, ancl an arm projected at right angles to the specitiien to support the load which measured the torque. This latter arm was al~vvaj~s kept horizontal by turning tlie svorrnwheel as the twisting load w:ts increased. The loading in this inachine was also by dead weights. The torque was produced by the twisting load and the extra snpporting force which i t required at its support. The torque was uniform nlo11g the tube, The bending moment W R J c t ~ u s e r iby a couple at each end, coinposed of the suppoi*tin,aforce and the overhanging load ; it was constant along the specimen. There was 110 shearing force on n section due to the bending loadu. The twist of a tube was measured by Prof. (Joker's torsionmeter, and the bsiiding deflesion by an adaptation of the Ayrton-Perry twisted strip. The latter apparatus W:IS designed for use. during some beiidiiig tests in which the
+

Proc. Phys. SOO.London, vol. xsi., and Phil. IIag. t i p r i l 1 ~ ~ 9 .

BIITETiI.\LS VSDRR COJIB11ED $ T R E h S .

137

&flc~\ioiiTKIS cstimiely small, and it was tliert:k'ort. vrry miiitive for the purlme of the prcmit tehts. It w i s n e c e ~ ~ a r y t o nieasure the strains soparatel;-, because the writer has s1iovt.a that wlieii a ductile inaterial is under cornbincd loading, it does not alwvays yicld first in the way which is indicated 1)ythe increasing loatl. Tho first yield is probably deteriiiiiietl I y tlic loading which prodnces the greatest s1ie:wing stress.

The 72,ct~ 011 S t e e l 7ithe~. ziolicl clrnwii steel tubes v e r e teste& ant1 the yicltl (inring each test was kept very small so thit several espcrinicntu coal(! be made with e d i titbc. All tliese specitiicnr were cut froin tlie sa1110 length of tube. The results arc collected 1x107s.

TABLEB.-Tests of Solid D r a w Steel Tiihcs. Internal dinineter 0.818in. External tlinnietcr 0 . M in.
I Bending
Test.
~

j lhe. ins.

lioiiient

I
i
E l ......1 E 2 ...... E3 TC4 ...... D 1 ...... c 1 ......I B J ...I B4 A 1 ...... 9 2 ...... B 3 ......; E 3 ......
I

re~-'sile Sheiir Stress rwistin, Stress vloinent due to due t o ba. ins. lending Porqie. IS..;^. in. bs.is. in
940 1U30

laxini 11111 .'rincipd St i w a . Lbfi.'~. in.

-0
(J

. .i

0
16330 33700 43550 44100 59900

300

600 800 510


1100 1120 1m 1330 1400

1090 970
600

28050 ZB'i00 2G400

25c,oo

28600 28050 389T0

4iWO
48!180

..I

700 4oa

800
0 0

1400

- - --

0 so0

61000 GSlOO 724400 76200 76200

16330 19100 10900 2 t aoo 0

51250 61550
6SOOO

6'9100

0 0
8166

72-100 78900 77ooo

The mnsinlum stress has 23,600 and 77,000 Ibs./s. inch for its extreme values, and the stress difference varies froiri 51,200 to 77,800 Ibs./s. in. The bending and twisting moments are plotted in fig. 2, in vvhich the numbers of the tests are shown against the points. The letter refers to the tube, and the number. indicates the order in which the test was iiinde on that specimen,

138

MR. W. , 4 . SCOBLE ON DUU'L'LLE

The figure is interesting because it shows very clearly the effect of repeated loading. An ellipse is drawn to lie evenly between the points. Specimens cut from the same length of material are not exactly alike, and an error is possible in locating the yield-point ; but after allowance is made for these facts, it is evident that the yield-point was raised by a previous test. Nevertheless, the results clearly indicate that the maximum shear stress is more nearly constant than the principal stress, and that the bending moment is again greater than the torque, so that the shearing stress and the stress difference increase with tlie bending moment.

The Tests on Copper Tubes. Te& were also made on solid drawn copper tubes, of 0.79 inch internal and 0.881 inch external diameter. The specimens were given a set to correct for their defective elasticity, and to allow several tests to be made on each tube. As in the case of the steel tubes, the yield during each test was kept small. The data obtained from these tests are tabulated below.

TABLE C.-Tests of Solid Drawn Copper Tubes. External diameter 0*8512 inch. Internal diameter 0.790 inch.
Tension Bending ?wisting due to Moment bfoment Bending lbs. ins. lbs. ins. Ibs./s. in

Test.

Shear Stress to

,
~

[~~'$~~:,
24200 24800 23150 19140 15560 1 12620 6310 0

Uaximum Principal Stress. Ibs./s. in.

Minimuin Priuoipnl Stress. lbs./s. in.

Stress 3ifference =2ce Maximum Shear Stress.

.i R 3
~

A 4 ...... A3 c 3 ...... c 2 ...... 0 4 ......

......

500
500 800 l ( N 1 0
1 1 m . . . .

B 2 ...... B4

... --...._.

......

1300 1330

1150 1180 1100 910 730 600 300 0

0 12620 21040 33660 42080 46300 54700 56000

24200 31910 35920 42330 47080 49510 55450


56000

_-. -24200 - I Y290 - 14880 - 8670 - ROO0 - 3210

--48400 51200 50800 51000 52080 52720 56200 56000

750
0

Here again the stress difference is approximately constant, and the deviation from this law isopposed to a constant maximum stress, because the stress clifference increnoos

BlATEIlIbLS UNDER COJIBISEC STILESS.

139

stendily with the bending monient. aiid t,he torque are plotted in fig. 3.

The bending moment

Deviations from the Shear-Stress Lato.

The stress difference is given in the tables, because it is the difference between the maximum and minimum principal stresses, which i t follows. It is now more convenient to deal with the maximum shear stress, which is half the stress difference. The stress difference and maximum shear-stress laws are therefore practically alike, but the former does not indicate the existence of the shearing stress which appears to i:aiise the a c t d fracture of a ductile material. A ductile rnatcrial behaves like a viscous fluid after the yield-point, so that it would be expected that the flow is caused by U shearing stress. The stress difference theory indicates a result, but it does not indicate the behaviour of the material. Noreover, the shearing is the stress considered by engineers. By referring to the tables it will be seen that the maximum priacipal stress increases tremendously as the bending moment increases, whereas the maximum shear stress is approximately constant. If the niasimum stress were ccnstant, the maximum shear stress would decrease with increasing bending moment ; but it is found that the shear stress increases, so that the deviation from the shear-stress law is opposed to the maximum stress theory, and it disproves the inaxiinum strain hypothesis, which is in effect an intermediate law, as has been already explained. The results are similar for each series of tests. Even before there were any reliable experimental results available, certain elasticians, influenced possibly by the forms of fractures, were convinced that a material failed by shearing. But further, there v a s usually a stress perpendicular to the plane of greatest shear, and i t was ticliev3d that this force across the plane affected the valne of the shear stress a t the elastic failure of the material. The shear stress tended to make the opposite surfaces at a section slide upon each other, and therefore it appeared that a compression across the section mould iiitroduce a resistance to the shear analogous to friction. Since a material tended to flow or

140

MR. W. A . SCOBLE ON DUCTILE

fracture along an iiueven section, so that the surfaces fitted into each other to a certain estent, it was conceivable that a tension across a section assisted the shearing stress, and lowered its value a t the breakdown. Many of the esperimental results available were exainined i n the former paper to determine whether the friction hypothesis would esplain the deviation from the shear-stress law. The final conclusion was that :-" I t inust be concluded that the maximum shear stress determines when yield takes place, b17t this will vary slightly on account of the difference in the shearing resistance in various directions, and any idea of a force analogous to friction must be abandoned." Bnt the bending moment appears to be always greater than the torque, and it was suggested that when a shaft was under a torque, T, and a bending nioment, M, the eqnirulent torque, T,, should be calculated from th5 equation

i n qhich is the shear stress, ancl ,fi is the tensile stress in the material at yield under simple loading. This equation represents an ellipse which replaces the circle of the sliearstress law, and thus allows for the liriiiting bending niomeut being greater than the torque. The preseut esperimental results justify the above suggestion. The theories of failure under combined stress can be very simply expressed in terms of the three principal stresses, PI, Pp,and P,, of which P, is the greatest and P, is the least. The inaxiinuin stress hypothesis assumes thai PI=constant. If the maximum strain theory be correct, then, for stresses in two directions only, PI--P,=constant. The stress difference or shear-stress law states that PI -PPS=constant. Since all these equations are particular fornis of P, mP3=coiistant, it appears to be desirable to esainine the results to find whether the stress distribution at yield may be expressed in a siinilar form. The maximum and iniiiiinuin p r i n c i p l stresses for these tests are plotted in fig. 4 . The points for the copper tubes lie very close to a straight line 1Jecauae the original results

X A T E R l b L S USDER CONBINED STRESS.

141

were exceptionally good. Tlie other points would have been close to the straight lines if corrected values had been taken, but, if the reinarlrs against the points are considered, tLe lines represent the re!ations between the principal stresses fairly well. The letters ag'tinst the points in fig. 4 indicate the positions of the corresponding points in the former diagrams when compared to the mean curves. The letter indicates the error in the original result from all causes, the experimental error, difference in material, and in the case of the steel tubes, due to tlie effect of repeated loading. On the diagram, fig. 4,A is the inaximum and B is the iiiiniinuiii principal stress. Using tha original notation, the equations which express the relations bet ween the priiicipal stresses are PI- 1-57 Ps= 71000, PI -1.37 P3=62200, PI - 1-26 P3= 54500. The constant varies because it indicates the strength of the material, and is the tensile stress at yield. Tlie values for b b ?)a " are 1-57,1.37, and 1.26, but the materials are different. It is now evident ,that the shear-stress law is most nearly correct of those stated above, and it is iiitended to be applied to all ductile materials. Although the deviation from the shear-stress lam is sometimes considerable (under combined bending and twisting), if an equation be. adopted to espress the conditions at yield inore closely, i t inust contain a constant which depends upon the material considered. This is a yery serious complication, and eugineers will probably prefer to assume a constant shear stress at failure, although the results obtained mill be slightly inaccurate. Also conibined bending and torsion is not the only example of coiiibiiied loading, and it is desirable f o r practical purposes to adopt oiie rule t o apply to all cases, if that be possible. It is therefore proposed to further consider this matter by esamiiiing the results obtained by other observers, who eniployod different methods and coinbinations of loading. Guest c d l d tlicse principal stresses PI, P,, and P3, and cacli reprcsentoil tlie stress in a particular gcneral direction, irrc.pccti\-c of its relative inngnitudo. ' h e notation has

Results obtuhed by other Observers.

GuEsr. Proc. Phys. Soc. London, Sept. 1900, vol. xvii. Loading by Tension, Torsion, and Internal Pressure.
-

PC.
!I I I -

PA.

50300 51050

17900

-17900
9000 13500 14800 0 -17900

0 -4450 -640

30600 -30800 60500+ 0 -29500 1 29500

0 0 0 0 0 0
0 34400

-0
0 0

P , .

P , .

15650 -15660

33000 I

0
s TWBN XII.

0 0 0 -1050

BR

10630
10950 11500 10830

-1755 1-660 0

0 -

10240 13160 -8.50 0 0 0 0 3355 -740

/
I

COPPER

TUBE XI.

5 9 % 9925 12730 6070

11850 0 7740 -6070

-750 0
-500 0

9450 -9450 17400 16650 -?215 14685 -2315 20770 14500 19260 15360 20100 0 10350 -10350 17610 - 2940 19980 16800 21150 14500 18820 1040

19

0 -880 -900 0 0 0 -986 -650 -750

STEEL TUBE Iv.


Pa.

1
-l

STEEL TUBE V.
p,.

I
I-

PB.

PC.
0 0

22500 41200 86650 42900 37700 42700 39000 41200

-22500
0 -8360 24000 -l?oo

00

3%06 -4000 0 22500 -22500

1
I

-1050

1
1

33500 17500 30900

- 17500
-

- 2300

4100

0
0

E 1
~

B~rss TUBE X T I I .
19250 17750 8100

I I
15340
0

30500 j 30350 1 L

24000 5500 29800

- 1050

-1050

- 1360

-900
0

-8100

JIATERIALS UX'DER COJIBIKED STRESS.

1 4 : :

been changed to PA,P,, aud P,, because P,, P2, and P3 have been reserved to represent the principal stresses in descending order of magnitude.

C. A. SMITH.'Engineering,' /I

Aug. 20, 1909.


I'

/I STEEL. Combined Compression ,, Tension, Compression, and Torsion. and Torsioil.

TAULE A.--SOLILI STEEL.

I TABLE C.-SOLID MILD


STEEL. Go;npression and

--19800 261ov" 29500 32300 31900 36500


-

Max. ' Max. Shear Priucipal Stress. Stress. lbe./e. in. lbs./a. in.
~

/I
Mar.

Max.

1 9 . 5 0 0 // -20170 19500 : : ; 1 20000 1 -31570 -35620 20400 -38180 20400 j -20750 19900 26370

/ /I--

I
~

32300
36420

194BO 20420 21000 20500 1 17810 19190 /I 20610 ,) 2Q320 1 19710 // 20500 'I 20420 21300 , 18210
~

24900 47601)

23700 83800

20420

19050 20320

1'

The maximum and minimum stresses for Guest's steel tubes are plotted in fig. 5, and for his brass and copper tubes in fig. 6 (PI. 111.). Mean lines are drawn through the points, and their equations are :-

I. IT. 9, ,, 111. IV. V. Guest's Copper Tube X. 93 9, 3, XI. Guest's Brass Tube XIT. 9, ,) XIII.
Guest's Steel Tube
99
97

$9

99

99 9,

97

79

$9

99

99

144

MR. \V. A . SCOn1.E

Q S DUCTILE

Mr. Sniitlis results are not shown plotted lierc, Init they are found to give,Smiths Solid Steel. Series SC. Table A. P, - 1-03 P3=11000. 99 39 ,, Series AD. Table C. P1--0*775 P3=36300. 7, Hollow Steel. Series SE. Table D. Y1- 1-01 =47~00.

It is Rvideiit from the diagrains that the points are not a.lways very close to the straight lines, so that i t is difficult to assign an exact value to 1 7 ~ for a giveii series of tests. But when average values are taken for m, they are found to differ considerably for different tests, even when t,hose are made under very similar conditions. F o r steel tuber, Guests << 7 ) varies from about 0.9 to 1.9. The copper tubes give consistent results, but one brass tube gives m equal t o 1.03,and the other has p n equal to 1.34 ; the points are very irregular. I t is evidently not necessary to give a more elaborate method for finding ))a, since it varies so much. Mr. Smiths values for steel are 1-09, 1.01, and 0.775. An esainination of Guests results sliows that the third principal stress has no appreciable effect on the otlicr stresses at failure. Conclusion.-In most cases the dsviations froiii the sliearstress law are opposed to a coiist:int mnsiinum stress, and this is always so wit,h bending. But it is probable thnt the value of (7 n varies somewhat for ductile materials, because there are degrees of ductility. The writer has shown that cast-iron behaves quite differently to a ductile inaterial *, but it does not coilforin to any exact law. He hopes shortly to publish results which prove thnt a strictly brittle ni:iterial behaves differently to cast-iron ancl ductile materials. It is therefaro not surprising that tlie results from ductile inaterids vapy soinewliat, and it is desirable that a stressstrain diagram for a teasion test of the inaterid should he considered in order t o estimate its ductility. I t is possible th:it the behaviour of :ill inaterids might bc expressed in one form, PI +~izP,.=;coiistaiit, in which depeiids 011 the degree of ductility of the inaterial. But in the present state

* Pruc. P l y . Soc. T O 1

3 5 .

JIATERlALS U S D E R COMBINRD STRESS.

145

of our knowledge it may be fairly claiined t,hat the shearstress or stress-difference law expresses the average behaviour of ductile materials under combined stresses, aiid that the maximum stress and maximuin strain laws are not true for ductile materials.
-4BSTRACT.

The Author further considers the results from some earlier tests made on mild steel bars, 3/4 inch diameter, and 30 inches effective length, under combined bending and torsion. It is pointed out that the yieldpoint is usually selected ria the criterion of strength, because it is more easily determined than the elastic limit, it is less affected by special treatment of the material, and it is assumed that the failure of Hookes Law between the elastic limit and the yield-point is due to local yielding. The elastic limit is the correct point, and is used throughout because the intermediate state mentioned above does not appear in bending. The results of tests on steel and copper tubes under combined bending and torsion are also given. All the results indicate that the maximum stress and maximum strain laws do not apply to ductile materials. The stress difference or shear stress law is approximately true, but there is, in each case, a deviation from the law which is opposed to the other theories mentioned. The deviations from the shear stress law are considered. In an earlier paper the -4uthor suggested a formula for combined bending and torsion which allows for the fact that the bending momelit is always greater then the torque. The internal friction hypothesi:, was also shown to be untenable. The three laws are now expressed in terms of the principal stresses, P,, P,, and P,, of which P, is the greatest, and P, is the least. Guests experiments prored that P, does not appreciably affect the failure of a material. The maximum stress law states that !?,=constant; the maximum strain theory that P 1 - ~ P , constant, in which 7 is Poissons Ratio j and the stress direreuce or shear stress hypothevis is expressed in the form E , - P, = constant. In the general equation P,-mP, = constant, the value of LL * indiwtes wliich law is most nearly true for the material. The Authors tests appear t o be the only experiments in which bending was adopted, and for tliese the values of m are 157, 1.37, 1.26. The figures apply to different materials, but are all greater than unity. The results obtained by other observera, who employed different methods and combinations of loading, are also examined. For Guests steel tubes, () I I varies from 0.9 to 1.87 ; his copper tubes give 1.04 and 1-09, For brass tubes the values are 1.03 and 1.34. Smiths tests on steel, three series, show (m to vary from 0.776 t o 1.09. The shear stress law appears to state the average behaviour of ductile materials, but there are considerable deviations from the law, which are uJua!ly opposed t o the other theories. Other tests by t h e Author indicntt!
3,

VOL. X X I l .

146

D U C T I L E M A T E R l A L S U X D E R COMBINED STRESS,

that brittle materials obey the maximum stress law, and it is therefore suggested that the value of m depends chiefly on the degree of ductility of the material considered, and to a lesser extent on the system of loading.

DISCUSSION.
Prof. COKER congratnlated the Author on being nble t o present so many actual tests. Although many experiments had been carried out during the pest 10 years there was still difficulty i n stating the law which was followed when a ductile material was subjected to stress. An initial difKculty was to determine whether the elastic limit or the yieldpoint should be selected as the criterion of strength. Guest had shown that for tension both points yielded similar results, but for combined stress the results were very different. H e thought the elastic limit should be eelected. With reference to the repeated loading of the tubes he thought more consistent r e d t s would have been obtained if a new tube had been used for each test. Although the results obtained threw light on the various theories they did not favour any one of them. RIr. C.A. S m T H expressed his interest in the paper and said he did not think much further progress would result from experiments on tubes. H e enumerated the disadvantages connected with the use of tubes and said he thought the case for the use of solid rods was proved. H e did not think copper specimens were suitable for tests, as the sanie system of loading did not always give the same result. RIr. Smith pointed out that no reference had been made t o non-axial loading. I f the load ma3 applied acentrically errors were introduced which should be allowed for. H e thought i t was advisable t o work with as simple a formula as possible (,--Pa =cons.) and to introduce a suitable constant for each different material. Mr. SCOBLE, in reply to Prof. Coker, said that Guest used the yieldpoint as the criterion of strength. I n his paper he (the Author) had referred his stresses to the elastic limit, i. e. the point a t which the strain ceased to be proportional to the stress. H e had used solid bars of 2 inch thickness because engineers required results on elecimens as large as possible. H e appreciated the advantage of using a simple law, but the results of his experiments did not follow any simple lam.

You might also like