On the existence of an element of Strength in .
Beams subjected to Transverse Strain , arising from the Lateral Action of the fibres or particles on each other , and named by the author the 'Resistance of Flexure .
' By WILLIAM HENRY BARLOW , Esq. , F.R.S. Received February 23 , -Read March 29 , 1855 .
IT has been long known , that under the existing theory of beams , which recognizes only two elements of strength , namely , the resistances to direct compression and extension , the strength of a bar of cast iron subjected to transverse strain cannot be reconciled with the results obtained from experiments on direct tension , if the neutral axis is in the centre of the bar .
The experiments made both on the transverse and on the direct tensile strength of this material have been so numerous and so carefully conducted , as to admit of no doubt of their accuracy ; and it results from them , either that the neutral axis must be at , or above , the top of the beam , or there must be some other cause for the strength exhibited by the beam when subjected to transverse strain .
In entering upon this question , it became necessary to establish clearly the position of the neutral axis , and the following experiments were commenced with that object ; but they have led to others , which are also described herein , and which establish the existence of a third , and a very important element of strength in beams .
I was desirous that the experiments for determining the position of the neutral axis should be made on such a scale and in such a manner as to place this question beyond doubt ; and with this object the following means were adopted : Two beams were cast , 7 feet long , 6 inches deep , and 2 inches in thickness ; on each of which were cast small vertical ribs at intervals of 12 inches : these ribs were one-fourth of an inch wide , and projected one-fourth of an inch from the beam .
In each rib nine small holes were drilled to the depth of the surface of the beam , for the purpose of inserting pins attached to a delicate measuring instrument ; the intention being to ascertain the position of the neutral axis by measuring the distance of the holes in the vertical ribs when the beam was placed under different strains .
The measuring instrument consisted of a bar of box-wood , in which was firmly inserted , at one end , a piece of brass , carrying a steel pin ; and at the other end a similar piece of brass carrying the socket of an adjusting screw .
The adjusting screw moved a brass slide , in the manner shown in Plate XII .
which carried another pin similar to that inserted in the box-wood bar , at the other end of the instrument .
The instrument was first made entirely of brass ; but the effects of expansion from the heat of the hand were so sensible , that the wooden bar was substituted .
The pins on the instrument fitted loosely into holes in the beam ; and the mode of using the instrument was , to bring the pins up by means of the screw against the side of the holes with a certain degree of pressure , which , with a little practice in using the instrument , was attained with considerable accuracy .
rTuwo beams were employed in order to avoid erirors which might arise from accidental irregularities in the metal .
The head of the adjusting screw was graduated to 100 divisions , and the screw had 43'9 threads to the inch , so that one division was equal to 4-3---th of an inch .
The measurements were , in all cases , taken by the outsides of the pins.of the neasuring instrument ; and when the instrument read zero , the -actual distance of the outer sides of the two pins was 513661 inches , so that the constant number 51661 being added to the micrometer readings gives , in each case , the total distance in terms of T-3%-Lth of an inch .
The form and dimensions of these beams .
are given in Plate XIII .
The measurements were taken four times in each position of the beam , and the error of measurement did not generally exceed from one to two divisions ; but if in the four observations an error amounting to more than four was found , it was corrected by remeasurement .
The numbers given in the following Tables are the micrometer readings , and the means of four observations in each case .
In these experiments more than 3000 ineasur'ements were taken ; but to avoid unnecessary figures , only the more prmominent results are given .
Table No. I. contains the measurements of the centre division of the first beatm under eight different conditions .
Table No. II .
contains similar measurements of the second beam .
In the first experiment it was found that , when the beam was inverted , the measuring instrument appeared to bear upon a different part of the holes , so that a direct comparison between the distances , in the beam erect and inverted , cannot be made with the same accuracy as the comparisons of different strains upon the beam when in the same position .
The first beam had been subjected to strain for the purpose of testing the measuring instrument previous to these experiments being made ; but the second beam had not ; and it will be seen that the effect of the strains in the latter case caused a permanent lengthening of the beam .
The same strain was frequently applied afterwards , but I could not observe any increase of this effect .
There was certainly a further apparent lengthening of both beams ; but I ascertained that this arose from a slight wearing of the working parts of the measuring instrument , from the great number of measurements taken .
In both experiments the beam was measured , first , in an erect position ; and secondly , inverted ; but in the Tables , the measurements of the same parts of the beam are placed opposite each other , so that they may be compared throughout with greater facility .
Determination of the Neutral Axis .
Measurements of the First Beam .
Beam erect .
Beam inverted .
No. 1 .
No. 2 .
No. 3 .
No. 4 .
No. 5 .
No. 6 .
No. 7 .
No. 8 .
At rest Strained with Weight taken Beamreversed , Strain of Strain of Strain of .
Weight taken previous Difference .
7373 lbs. on Difference .
off , condition bearing its Difference .
2893 lbs. on Difference .
5133 lbs. on Difference .
7373 lbs. on Difference .
off , condition uI to being the end , equal the same as own weight the end , equal the end , equal the end , equal the same as strained .
to 14,746 lbs. No. 1 .
on the centre .
to 5786 lbs. on to 10,066lbs .
to 14,706 lbs. No. 4 .
on the centre .
the centre .
on the centre .
on the centre .
Micrometer Micrometer Micrometer Micrometer Micrometer Micrometer Micrometer Micrometer readings .
readings .
readings .
readings .
readings .
readings .
Teadings .
readings .
2 2208 +70 2278 -67 2211 2210 -33 2177 -25 2152 -26 2126 + 71 2197 2186 +55 2241 -53 2188 2187 -25 2162 16 2146 -21 2125 +53 2178 2095 +36 2131 -33 2098 2103 -20 !
2083 -12 2071 -13 2058 +36 2094 2127 +14 2141 -13 2128 2129 -10 owl 19 8 211 15 2106 +21 2127 2110 5 2105 +5 2110 2117 2 2115 1 2114 +2 2116 2116 M 2052 -21 2031 + 13 2054 2060 +5 2065 +6 2071 + 12 2083 -20 2063 2095 -39 2056 +42 2098 2101 +15 2116 +14 2130 +19 2149 -37 2112 2052 -58 1994 +68 2052 2056 +21 2077 +21 2098 +29 2127 -60 2067 2101 -73 2028 +76 2104 2111 +28 2139 +26 2165 + 34 2199 -73 2126 Note.-The extensions are marked + ; the compressions are marked - .
Determination of the N7eutral Axis .
Measurements of the Second Beam .
Beam erect .
Beam inverted .
No. 1. .
No. 2. .
No. 3 .
N4 No. 5. .
No. 6 .
O^XEX S~~oX 4 , X. .
okNo .
4 ; 2 No. 7 .
At rest ~ Strain of Strain of Strain of Strain of previous to 8000h Weight bl~tl~esstOin g lbs. | 16,000 lbih 6r4 8000 lbs. | 16,000 lbs. Weight strained .
~ on centre .
a on centre .
are on centre .
on centre .
A removed .
Micrometer Micrometer Micrometer Micrometer Micrometer Micrometer Micrometer readings .
readings .
readings .
readings .
readings .
readings .
readings .
1633 + 37 1670 + 65 1735 -89 1646 -44 1602 -.56 1546 + 97 1633 1525 + 28 1553 +47 1600 -63 1537 -24 1513 -46 1467 -t67 1534 1481 +921 1502 +34 1536 -44 1492 -19 1473 -928 1445 + 42 1487 1442 + 11 1453 + 21 1474 -923 1451 -10 1441 12 1429 +922 1451 1392 +2 1394 +7 1401 1 1400 +1 1401 1401 +4 1405 1375 -10 1365 9 1356 +18 1374 +17 1391 +11 1402 -17 1385 1338 -18 1320 -24 1296 + 44 1340 +920 1360 + 27 1387 -35 1352 1257 27 19230 -37 1193 + 64 19257 +31 1.88 +43 1331 -57 1274 1248 -42 1206 -46 1160 +85 1245 + 44 1289 + 57 1346 -78 1268 Note.-The extensions are marked + ; the compressions are marked- .
Considering the very minute quantities which had to be measured , and the numerous causes of disturbance to which observations of so much delicacy were liable , such as changes of temperature or want of perfect uniformity in the dimensions or texture of the beams , the results , as shown by the column of differences , exhibit more regularity than could have been expected ; and they point out the position of the neutral axis , as the centre of the beam , in a manner so decided , as to remove all further doubt upon this subject , not only in the smaller strains , but in the larger ones also ; which , in the case of the second beam , were carried to about three-fourths of the breaking Weight .
It will be observed also that the extensions and compressions increase in an arithmetical ratio from the centre to the extreme upper and lower sides of the beam .
These experiments having established the fact that the neutral axis is in the centre of a rectangular beam , and that its position is not sensibly altered by variations in the amount of strain applied , it becomes evident that if there were no other elements of strength than the resistances to direct extension and compression , the well-known formula 2adf W-3i should give the breaking weight whenf is equal to the smaller of these two resistances , which in cast iron is the tensile resistance .
But the weight so calculated is less than half the actual strength of the beam .
In considering this question , I was forcibly struck by the circumstance , that , in applying the law of " ut tensio sic vise " to contiguous fibres , under different degrees of tension and compression , the efect of lateral adhesion is omitted , and each fibre is supposed to be capable of talking up the same degree of extension and compression from the same force as if it acted separately , and independently of the adjoining fibres .
But it is well known as a practical fact , that there is a powerful lateral action which tends to modify the effect of unequal strains .
If , for example , a bar , abed , have a strain applied at efdb , the portion A-d-7 , defb will not be extended so much as it would be if separated from acef , unless an equal strain is applied to the portion acef And if a portion of a bar cannot be extended in proportion to the force applied to it , unless the contiguous part is equally strained it follows that the outer portions of a beam subjected to transverse strain will not be extended in proportion to the force applied , because the part nearer the neutral axis I= is not equally strained .
The measurements made for obtaining the posti.tion of the neutral axis afford direct evidence on this point .
In the first beam , a strain of 5786 lbs. caused an extension of twentyeight divisions of the micrometer ; the points measured were +-1-ths of the depth of the beam .
The extension at the outer fibres was therefore 28 X -12=30 divisions .
The micrometer reading before the strain was applied was 211 1 , and the total distance of the points measured was 2111 +51661 =53772 .
The effect of the strain caused therefore an extension of 53072 17 ?
24 of the length .
The beamn was 7 feet 4 inches long , 6 inches deep , and 2 inches thick ; and as W 2adf 31W 2n ad 3 x88 x5786 orf= 2x12 x6 10,608 lbs. ; so that , with a strain of 10,608 lbs. at the outer fibres , the extension produced was 17924 of the length .
But in referring to the experiments made by Mr. IHODGKINSON , it will be seen that a force of 10,538 , applied by direct tensile strain , extends cast iron -1--th of its length , being nearly double that exhibited by the beam .
In the second beam , a weight of 8000 lbs. ( from the mean of two results ) produced an extension of forty divisions , which at the extreme fibres will be 401-1 =44 divisions .
The mean reading of the micrometer , previous to the strain being applied , was 1439 ; therefore the extension was 44 1 51661 +1439 1207 The strain at the outer fibres produced by this weight was 14,666 lbs. ; so that 14,666 lbs. to the inch caused an extension of 1-2%th of the length .
But referring again to HODGKINSON 's experiments on direct tensile strain , a weight of 14,793 Ibs .
produced an extension of -th of the length ; which is again nearly double that produced by the same strain when excited by a weight applied transversely .
From these and other considerations I was led to think it probable that the effect of the lateral action of the fibres or particles of a beam , tending to modify the effect of the unequal strains and opposite forces , and thus diminishing the amount of extension and compression which would otherwise arise , constituted in effect a resistance tojiexure ; and itwill be found that the following experiments fully confirm the existence of this resistance as an additional element of strength in beams , ; and that it explains the apparent anomaly in the amount of tensile resistance when excited by direct and by transverse strains .
Assuming the probability of a resistance , acting independently of , or in addition to , the resistance of direct tension and compression , and varying with the flex ur , it occurred to me that it might be exhibited experimentally by casting open girders of the forms shown figs. 2 , 3 &amp; 4 , having the same sectional area in the upper and lower ribs ; the same number of vertical ribs , but the distance between the horizontal ribs , and consequently the deflections of the girders , different .
In these girders the neutral axis would necessarily be ( like that of the solid beam ) in the centre , and the sectional area of the ribs subjected to tension and compression being the same in each , the circumstances under which rupture would ensue would be similar , except in the amount of flexure .
The formula for the strength of a girder of this form is as follows : Let a=the united area contained in the upper and lower ribs ; a'=the intervening space ; d=the total depth ; c=the distance between the upper and lower ribs ; I=the length of bearing ; W=the breaking weight ; I and F=the force required to produce rupture in the extreme fibres or particles .
Then a+a'=the total area of the rectangle m , n , o , p , 2dF , 2cal cF W= F(al+ a ) dX d } oll W= 2F ( a'+a)d d } ' &amp; W= 2Fa d+c+ C , The formula may also be obtained by calculating the moments in the usual way .
Using the same letters as before , we have , for the distance of the centres of com pression and extension , 2 ( d--+2)The force acting when F is the strain which breaks the outer fibre , will be 22 Hence W =2d+ C2 )(+daF or W= 31'(d+c+w ) .
The value of W being obtained by experiment in each case , we have from the formula F.31W 2a(d+c+ ?
) and if the strength depended only on the direct tensile power of the material , 1 ' should in each case be constants and equal to the direct tensile resistance ; but if , in addition to this , there existed another element of strength in the resistance occasioned by the lateral adhesion and varying with the flexure , the value of F would be found , in every case , greater than the tensile resistance , and to increase when the flexure increased .
Four beams were cast of each form , of which the details , the exact dimensions deflections , and breaking weights are given in the Appendix .
The results were as follows , obtained from the mean of four experiments on each form of girtder : _ " 1 , g _.~~~~~~~eflection with , Description of beam .
Total depth of Sectional area Distance between nine-tenths of Breaking beam .
of the two ribs .
the ribs .
breaking weight .
weight .
Form No. 2 ... 2 1.98 54-510 we468 in .
in .
in .
in .
lbs. Form NO .
2 ... ... ... ,2.51 1-98 * 54 In connexion with the above-described experiments , I made four others on solid beams having the same sectional area and length as the open girders ; and the mean of the four gave a breaking weight of 1888 lbs. Obtaining the value of F from these experiments , we have , Deflection with nine-tenths Value of F. of breaking weight .
*670 41709 lbs. which again exhibits an increase in the value of F , with an increase in the deflection .
The foregoing experiments having shown that in girders containing the same depth of metal , the resistance arising from the lateral action of the particles depended on the amount of the flexure , I thought it desirable to make other experiments to ascertain how this resistance varied in girders having the same total depth , and consequently nearly the same deflection , but with different depths of metal in the girder .
For this purpose beams were cast of the forms Nos. 5 , 6 and 7 , each 4 inches deep , and with the upper and lower ribs 11 inch by -a inch , the ribs being placed as shown in the figures , so that the depth of the metal in No. 5 was twice as great as in Nos. 6 and 7 .
Four beams were cast of each form the exact dimensions and breaking weights are given in the Appendix , and the mean results were as follows : Description of beam .
Depth of beam .
Depth of metal .
Sectional area .
Deflection .
Breaking weight .
Form No. 5 ... ... 4 04 3-01 2-320 322 5141 Form No. 64 04 1-48 2-230 *310 5147 Form No. 7 ... ... 4 07 1.56 2-380 -262 6000 Obtaining the value of F from these experiments , and comparing them with beam No. 4 , which had the same total depth , we haveDeflection .
Depth of Value of F. Form No. 5 ... ... ... ... ... ... ... ... ... -322 3-01 37408 Form No. 4 ... ... ... ... ... ... ... .- ... .
301 1-97 28032 Form No. 7 ... ... ... ... ... ... ... ... ... -262 1-56 27908 Form No. 6 ... ... ... ... ... ... ... ... ... -310 1-48 25271 These experiments did not afford so complete a comparison as the former series , because the intervals between the vertical ribs were not equal , nor in the same proportion to the depth of metal , the effect of which would be to vary to some extent the form of the curve of deflection .
Nevertheless , they show in an equally decided manner , that when the deflection is the same the resistance increases when the depth of metal in the beam is increased .
The foregoing experiments have therefore elicited three facts as regards beams formed of two parallel bars separated at givren intervals by vertical ribs : First , that in every case the resistance , or the value of F , is greater than that due to the tensile resistance of the metal .
Secondly , that with the same depth of metal in the beam , and the same distance of bearing , the resistance is greater when the deflection is greater .
Thirdly , that with the same deflection and the same length of bearing , the resistance is greater when the depth of metal in the beam is greater .
And it follows from these results , that there is an element of strength depending on the amount of deflection in connexion with the depth of metal in the beam , or in other words , dependent upon the degree of flexure to which the metal forming the beam is subjected .
The existence of an element of strength in addition to the resistances to direct tension and compression being clearly proved by these experiments , it becomes interesting to ascertain the law under which it varies , in the form of beams experimented upon .
Now if from the value of F , the tensile strength of the metal is deducted , it will be found that the remainder maintains nearly a constant ratio in each case to the depth of the metal in the beam multiplied by its deflection .
It would appear , therefore , that the total resistance , or the value of F , is composed of two quantities ; one being constant and limited by the resistance to direct tension , and the other varying directly as the degree of flexure to which the metal foretming the beam is subjected .
The applicability of this simple law may be tested by the results of the experiments , as follows : Let L=the resistance to flexure in the solid beam at the time of rupture and let D=the depth , z=the deflection , f=tensile resistance , and F=total resistance .
Then in the solid beam f+ -F ; and let F ' , D ' and i ' , represent the total resistance , depth of metal , and deflection of any other of the beams ; then , the lengths being equal , if the resistance arising from the lateral action varies as the depth of metal into the deflection , F't=f++ .
The value of p may be determined from this equation , applied to each of the experiments , in two ways ; first , by supposingf to be a constant quantity ; and secondly , by supposingf and p to have a constant ratio .
By the first mode , the whole of the errors of observation and irregularities of the strength of the metal would be accumulated in p. By the second method , these irregularities will be divided between the values off and A. Adopting therefore the second method , let 1 to m represent the ratio off to @ : then f=mp , and mnp+cp Ft ; Ft of PD'18 m+ Da which ought to be a constant quantity in all the experiments .
We cannot obtain the deflections at the line of rupture , but they may be assumed to be proportional to the deflections with 1-9ths of the breaking weights in each case .
Now the value of F in the solid beam was found to be 41,709 lbs. ; and the value off , from the experiments on direct tension , was 18,750 lbs. : and as in the solid beam f+ =F p will be 22,959 lbs. , and the ratio of p tof will be as I to ^81 .
For the purpose of comparison , I have deduced the value off and p , in solid beams , from the experiments of Mr. HODGKINSON on ten different descriptions of metal ; the results of which are given in the following Table Transverse Value of p from strength of bar Value off+@ the formula Description of iron .
I inch square Tensile strength from the formula w 2ad(f+o ) and 54 inches per square inch .
2ad(f+p ) .
31 between the 31 31 31w supports .
2ad lbs. lbs. lbs. lbs. Carron iron No. 2 , cold blast ... ... ... ... 476 16,683 38,5.56 21,873 Carron iron No. 2 , hot blast ... ... ... ... ... 463 13,505 37,503 23,998 Carron iron No. 3 , cold blast ... ... ... 446 14,200 36,126 21,926 Carron iron No. 3 , hot blast ... ... ... .
527 17,755 42,687 24,932 Devon iron No. 3 , hot blast ... ... ... ... 537 21,907 43,497 21,590 Buffery iron No. 1 , cold blast ... ... ... ... 463 17,466 37,503 20,037 Buffery iron No. 1 , hot blast ... ... ... .
436 13,434 35,316 21,882 Coed-Talon iron No. 2 , cold blast ... ... 413 18,855 33,453 14,598 Coed-Talon iron No. 2 , hot blast ... ... ... 416 16,676 33,696 17,020 Low Moor iron No. 3 , cold blast ... ... ... 467 14,535 37,827 23,292 Means ... ... ... ... ... 464 i6,502 37 , i61 21,114 The mean ratio of p tof in these metals appears to be as 1 to *78 .
The metal used in my experiments was a mixture consisting of two-thirds of South Staffordshire No. 3 , hot blast pig , and one-third old metal recast .
As compared with Mr. HODGKINSON'S experiments , its strength accorded nearly with that of the Carron iron No. 3 , hot blast .
The mean ratio of p tof , obtained from Mr. HODGKINSON 's experiments , being as I to *78 , and from the experiments herein detailed being as 1 to *81 , we may considerf to be four.-fifths of p ; and therefore rt=hr 8o Using this ratio , the values of p and f , derived from the formula F @= IY &amp; ' and as applied to each of the experiments , are given below : No. 1 .
p 21709 =23,171 lbs. , f= 18,537 lbs. 2'012 x '670 No. 2 .
=8 510=22,904 lbs. , f= 18,323 lbs. 1P348 No. 3~ 31977 'No .
3 .
p= 2'01 x 401=22,890 lbs. , 1=18)312 lbs. '8+ 1'348 No. 4 .
A= 28032 =22,606 lbs. , f= 18,085 lbs. 31*9 x1221.8+ 1'348 No. 5 .
, P= 37408 .
2224,626 lbs.)f= i9.5o ilbs .
No. 6 .
p= 25270 31=22,167 lbs. , f=17,734 lbs. 1'348 27908 No. 7 .
'= + 56 x 262 =25,302 lbs. , f=20,242 lbs. P1348 These results , though not exhibiting complete regularity , are sufficiently uniform to indicate that the assumed law of the variation of this resistance is a close approximation to the truth .
It will be observed also , that Nos. 2,3 , 4 and 6 , give a smaller value of p than Nos. 1 , 5 and 7 , which probably arises from the difference in the proportion which the distance between the vertical ribs bears to the depth of the metal ; a circumstance which would affect , to some extent , the form of the curve of deflection .
In the formula P= Fh DIa represents the ratio of the depth of metal in each meDa beam multiplied by its deflection , to the depth of metal in the solid beam multiplied by its deflection .
But the deflections , as might have been expected from known laws , were nearly in the inverse ratio of the total depths of each girder ; therefore the degree of flexure , and consequently the resistance to flexure in each , will be nearly as the depth of metal divided by the total depth of the girder , and we are thus enabled MDcccLv .
2 K to obtain a formula for computing , approximately , the breaking weights of these girders , without first ascertaining their deflection .
Using the same letters as before , we have , for the resistance due to tension , 3(d+c+ ) f ; and for the resistance to flexure , 3(d+c+ d ; and consequently , for the united effect of the two resistances , W=c , ( d+c+ 4(f+t &gt; I shall therefore conclude these observations by comparing the breaking weights computed for tensile resistance alone , and those obtained from the formula which includes the resistance to flexure , with the actual breaking weights obtained by the experiments , taking the value off= 18,750 lbs. , and p=23,000 lbs. Breaking weight if the Breaking weight cor-Breaking weight Description o beam resistance depended on puted by the formula , as obtained by the or girder .
direct tensile strength .
th resia e experiments .
to flexure .
lbs. lbs. lbs. No. I ... ... ... ... . .
849 1890 1888 No. 2.1308 2567 2468 No. 3. .
1808 3287 3084 No. 4 ... ... ... ... . .
2912 4659 4353 No. 5 ... ... ... ... 2578 4935 5141 No. 6 .
3819 5533 5147 No. 7 ... ... ... .
4031 5919 6000 The accordance exhibited by the computed and the actual breaking weights , evinces the general accuracy of the formula , as applied to this form of beam ; while these results , compared with those computed for direct tensile force alone , show how large a proportion of the strength of cast iron , when subjected to transverse strain , is due to the resistance arising from the lateral action .
It will also be seen that comparisons of the relative strengths of different forms of section , calculated , as has been customary , on the assumption that the resistances are constant forces , or governed by a constant coefficient , , must be entirely fallacious .
It was my intention to have included in this paper a similar investigation as to the position of the neutral axis , and the amount of the resistance arising from lateral action of the fibres in wrought iron ; but as the experiments will take some time to complete , and as the facts elicited in reference to cast iron are of sufficient importance to render it desirable that they should be made known , I will reserve the examination of wrought iron for the subject of another communication .
Girder No. 1 .
LExperiment No. 1 .
Experiment No. 2 .
Experiment No. 3 .
Experiment No. 4 .
inches .
inches .
inches .
inches .
Depth .
2-015 202 2-073 2-040 Thickness. .
*975 '98 1'030 '990 Area of section ... ... ... ... ... ... 1'965 1'98 2'135 2'020 Weight applied , lbs. Deflection .
Deflection .
Deflection .
Deflection .
40 *015 '013 '014 014 376 '145 *115 6oo *203 712 '280 '233 '264 '244 936 '330 1160 *490 '420 '397 '414 1608.725 *625.579 -614 1664 Broke'755 *655 1720.680 '629 '659 1832 ... ... '737 -679 * 734 1888 ... ... Broke '699.764 1916 -. .
A ... ... Broke 1944 ... ... .734 2000 ... ... '762 2028.** ... ..774 2056 ... ... .
' ... . .
789 2084 ... ... ... ... Broke Breaking weight , lbs ... ... ... .
1664 1888 2084 1916 Deflection with nine-tenths 3 643 *667 699 *670 of breaking weight , inches 666 '67 Girder No. 2 .
Experiment No. 1 .
Experiment No. 2 .
Experiment No. 3 .
Experiment No. 4 .
inches .
inches .
inches .
inches .
Total depth .
2-54 2-53 2'49 2'50 Depth between upper and '56 Girder No. 3 .
Experiment No. 1 .
Experiment No. 2 .
Experiment No. 3 .
Experiment No. 4 .
inches .
inches .
inches .
inches .
Total depth ... ... ... ... ... ... ... 3'02 3'00 3'00 3'00 Depth between upper and 1.00 1'01 101 lower ribs ... ... .
Area of top rib .
103 1'02.97 1J01 Area of bottom rib.99 '98 1-01 97 Weight applied , lbs. Deflection .
Deflection .
Deflection .
Deflection .
40 *006 005 '005 *005 712 '085 '085 '085 844 '113 1516 '216 '185 *197 *195 1740 '248 2188 *328 '297 *293 2300 -295 2524 '388 .
2636 *418 '363.375 2748 '433 '377 2972 *483 '410 *423 Broke 3028 Broke '425 ' *438 3084 ... ... '435 * 452 3112.437 Broke 3224 ... ... Broke Breaking weight , lbs ... ... ... .
3028 3224 3112 2972 Deflection with nine-tenths * 435 402 397 3,1 of breaking weight , inches Girder No. 4 .
Experiment No. 1 .
Experiment No. 2 .
Experiment No. 3 .
Experiment No. 4 .
inches .
inches .
inches .
inches .
Total depth ... ... ... ... ... ... ... 3.99 400 3'99 4-01 Depth between upper and 1 2.00 2'03 2-05 2-04 lower ribs ... ... ... ... ... ... Area of top rib ... ... ... ... ... ... 1'00 '97 '98 '98 Area of bottom rib ... ... ... ... ... 1'00 '99 '98 1'01 Weight applied , lbs. Deflection .
Deflection .
Deflection .
Deflection .
40 '002 '002 '002 '003 712 '047 '040 '048 '058 1516 '104 097 '102 '108 1964 134 2188 '161 '155 '155 '148 2636 '199 '197 '1 85 '183 3084 '227 '227 '223 '218 3420 -259 3532 '269 '267 '255 '253 3756 '299 '282 '285 3980 '317 '312 *300 '303 4092 '329 '320 '307 4148 '336 '322 '313 4204 Broke '327 Broke 4260 ... ... Broke ... ... '333 4316 ... ... ... ... ... 4400 ... ... ... . .
'343 4428 ... ... ... ... 4745 ... ... ... ... ... ... Broke Breaking weight , lbs ... ... 4204 4260 4204 4745 Deflection with nine-tenths 3 297 *93 Girder No. 5 .
Experiment No. 1 .
Experiment No. 2 .
Experiment No. 3 .
Experiment No. 4 .
inches .
inches .
inches .
inches .
Total depth ... ... ... ... ... ... . .
4-02 4 05 4 05 4 04 Depth between upper and 1*04 1*04 1-04 1 00 lower ribs ... ... ... ... ... .
Area of top rib ... ... ... ... ... ... 19125 1-16 1-14 1.22 Area of bottom rib ... ... ... ... ... 11662 1 13 115 1.20 Weight applied , lbs. Deflection .
Deflection .
Deflection .
Deflection .
712-1516 2188 2290 .
*133 *148 *138 2636 2885 2 *173 *182 *178 3084 3445 *2213 *221 , 223 3532 3980 4005 9268.9270 259 4428 .
4565 t ?
*313 *320 *308 4652 4705 *323 Girder No. 6 .
Experiment No. 1 .
Experiment No. 2 .
Experiment No. 3 .
Experiment No. 4 .
inches .
inches .
inches .
inches .
Total depth ... ... ... ... ... ... ... 4'02 4 05 4 03 4'06 Depth between upper and 2522'55 2'56 2'61 lower ribs ... . .
5 Area of top rib. .
113 1 18 1'08 110 Area of bottomrib 1'13 1.09 1'11 1P10 Weight applied , lbs. Deflection .
Deflection .
Deflection .
Deflection .
7 12 _ 1516 2188 2290 '130 '138 '138 2636 -4 2885 X *168 * 186 '175 3084 3445 E r-1 *205 220 *222 3532 3980 At 4005 3*251 TRANSVERtSE STRAIN .
241 SummInary of the Experiments on Transverse Strength , giving the mean results .
ScinlDistance Deflection with Depth .
Sectional between Breaking nine-tenths of area .
the ribs .
weight .
breaking weight .
in .
sq .
in .
in .
lbs in .
F 2015 1i965 ... ... 1664 *643 Form of beam No. 1 ... ... ... .
2'020 |1950. .
1888 '667 2'073 2'135 ... .
2084 699 L 2'040 2 020 ... ... 1916 '670 Mean ... ... ... ... ... ... 2'012 2'025 ... ... 1888 *670 { 254 2'01 '56 2188 '489 Form of beam No. 2 ... ... ... .
2-53 2'00 *55 2748 *532 2'49 1 96 *51 2412 '482 2'50 1'95 *55 2524 *516 Mean ... ... ... ... ... . .
' .
251 1'98 * 54 2468 '510 302 2'02.98 3028 -435 Form of beam No. 3 .
3'00 200 1'00 3224 *402 3.00 1P98 1'01 3112 '397 3'00 1'98 1'01 2972 '371 Mean ... ... ... ... ... 3'01 2'00 1P00 3084 -401 3,99 2'00 2'00 4204 *297 ) 400 196 2'03 4260 '293 Form of beam No. 4 ... ... ... ... ... ... . .
3.99 1-96 2'05 4204 '282 4-01 1'99 2'04 4745 '331 Mean ... ... ... ... .
400 1'98 2'03 4353 '301 F 4'02 2-287 1'04 5050 ... .
o4 05 2'290 1'04 5125 *321 Form of beam No. 5 .
v@@**e ' 4'05 2'290 1'04 4985 '313 L 4'04 2'420 1'00 5405 '331 Mean ... ... ... ... ... .
4'04 2'322 1'03 5141 Experiments on Direct Tension .
Number of Sectional area at the Last weight supWeight with which Remarks .
experiment .
place of fracture .
ported .
the bar broke .
inches .
lbs. lbs. 1 .
1-0506 18,560 18,840 A small air-bubble .
2 .
1-0557 19,680 19,960 A small air-bubble .
3 .
1-0100 21,360 21,500 A small air-bubble at corner , very small .
4 .
1-0364 16,320 1 6,320 Honey-combed .
5 .
1-0301 17,440 17,440 Sound .
6 .
1-0403 16,3 0 17,440 A small air-bubble .
7 .
1-0150 21,640 21,920 Sound .
8 .
1-0200 22,200 22,470 Sound .
Mean ... ... 1-0323 19,190 19,486 Mean greatest weight supported , per inch - ... 1 8,590 lbs. Mean weight which broke the bar , per inch ... 18,876 lbs. Considering the actual breaking weight to be between these two , and rather nearer the latter , when due allowance is made for the small air-bubbles , the mean breaking weight may be taken at 18,750 lbs. per square inch .
] ?
eam/ ~mpl~d/ u deteinuis the , Po9 dww bf tk &amp; itera , atci .
1~~~~~~~~~~~~~~~~~~~~~~~~~t -F 7 Li ' Lr u Li Lf~~~~~~~~~~7 ~ J~~at~~ur ~~~ --I------um----L~ ~~~_ii m ; wanu l W/ / 0 W Dg/ -6id 7c1 --j __ _____ ___ ______ 11 &gt; &lt; ____ __ __-ETT L FT71 ETTT E7 EZTTT7 __________________________________-A [ 7777 [ 7777 [ 77 7 [ 7777 E77777 77777 7777 r77Z [ 77777 _ __ _1 r-K I s. 77 777 [ 77 [ 777 L [ 777 [ 7711 77 F7 F77 [ 7 7( .
1 rIII L___ ii j __ - .
-__iV ? ?
7 FFF[ 1IH1iLLLLLLLLLLiLLLLLL[LLLLLKiLLrHH ] ---------- ; / / -0 '

