}. 

University of Texas Bulletin 

No. 1815: March 10, 1918 
PHYSICAL PROPERTIES OF DENSE 
CONCRETE AS DETERMINED BY 
THE RELATIVE QUANTITY 
OF CEMENT 

By 
F. E. Giesecke and S. P. Finch 
BUREAU OF ECONOMIC GEOLOGY AND TECHNOLOGY 
J. A. Udden, Director 
DIVISION OF ENGINEERING 
F. E. Giesecke, Head of the Division 


Published by the University six times a month and entered u 
second-class matter at the postoffice at 
AUSTIN, TEXAS 

Publications of the University . of Texas 
Publications Committee: 
F. W. GBA.JT. R. H. GRIFFITH 
J.M. BRYANT Jl. L. HENDERSON D. B. CASTEEL I. P. Hil.J)JrnRAND FREDEnJO DUNCALF E. J. MATHEWS 
The University publishes bulletins six times a. month, so num­bered that the :fint two digits of the number show the year of iaJUe, the last two the position in the yearly series. (For ex­ample, No. 1701 is the :first bulletin of the year 1917.) These eompril'J6 the official publicatioru1 of the University, publications Ol!l humanistic and scientifie subjects, bulletins prepared by the Department of Extension and by the Bureau of Municipal Re1eareh and Reference, and other bulletins of general educa­tional interest. With the ~ception of special numbers, any bulletin will be sent to a eitizen of Texas free on request. .All eommunicatiom1 about University publications should lie ad­dressed to the Chairman of the Publications Committee, Uni­nrsity of Texas, Austin. 
B303-918-2m-8366 
University of Texas Bulletin 
No. 1815 : March 10, 1918 
PHYSICAL PROPERTIES OF DENSE 
CONCRETE AS DETERMINED BY 
THE RELATIVE QUANTITY 
OF CEMENT 

By 
F. E. Giesecke and S. P. Finch 
BUREAU OF ECONOMIC GEOLOGY AND TECHNOLOGY 
J. A. Udden, Director 
DIVTSTON OF ENGINEERING 
F. E. Giesecke, Head of the Division 

Published by the University six times a month and entered as 
secoud-elass matter at the postoffice at 
.AUSTIN, TEXAS 

The benefits of education and of useful knowledge, generally diffused through a community, are essential to the preservation of a free govern­ment. 
Sam Houston 
Cultivated mind is the guardian genius of democracy. • • • It is the only dictator that freemen acknowl­edge and the oD.ly security that free­men desire. 
Mirabeau B. Lamar 
CONTENTS Page Acknowledgement 5 llJitroduction ......................... , . . . . . . . . . . . . . . 5 Materials and Proportions. . . . . . . . . ... . . . . . . . . . . . . . . . . . . . 7 Mixing, Moulding, Storing. . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Compression Tests. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Elasticity Tests. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 T,ensi·on Tests. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Transverse Tests ........................... , . . . . . . . . . . 51 Bond Tests. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 Permeability Tests. . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . . . . . . . 76 Abrasion Tests... ·. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 

PHYSICAL PROPER'l'IES OF DENSE CONCRETE AS 
DETERMINED BY THE RELATIVE QUANTITY 
OF CEMENT 

ACKNOWLEDGMENT 
This investigation was undertaken as a part of the work of the Engineering Division of the Bureau of Economic Geology and Technology of the Univiersity of Texas and was conducted by Prof. F. E. Giesecke and Prof. S. P. Finch, assisted by Mr. E. 
F. Ries. Mr. J. P. Nash, Testing :Engineer for the Bureau of Econ-0mic Geology and Technology, also gave much assistanc·e, being in charge of the work 1of making and testing the' specimens of Series II and III. Mr. G. A. Parkinson acted as laboratory assistant and displayed great care in making the specimens to be tested. Miss Nellie Jefferson, laboratory assistant, prepared the diagrams for this bulletin. 
The authors are much indebted to Mr. H. R. Thomas, C. E., 1912, University of Texas and M. S., 1914, University of Illinois, f-0r valuable suggestions in regard to the best type of compression moulds and for making the University of Illinois compressometer to be used in the elasticity tests. 
INTRODUCTION 
Object of the investigation. It has long been the custom in the use of concrete to arbitrarily specify the proportions of ce­ment, fine, and CJoarse aggregate, as 1 cement, 2 sand, 4 stone, ehoosing the proportions which the results of tests and standard practice seemed to indicate would give the necessary str·ength or other properties desired. Prompted by the belief that it was much better and more economical to proportion the fine and coarse aggregate so· as to produce a dense concrete and to incor­porate the amount of cement necessary to obtain the physical properties desired, this investigation was undertaken so that these properties might be known for different percentages of ce­ment. 
Scope of investigation. The attempt was made to determine the variation in the compressive strength, modulus :of elasticity, 
Univel'sity of Texas Bulletin 
tensile strength, modulus of rupture, bond strength, pervious­ness, and resistance to abrasion, for conc.nte approximating as closely as possible a maximum density mix and containing vary­ing percentages of cement. 
Tests were made on both broken stone and gravel C;'.lncrete with six difforent percentages of cement varying from about 4% cement by weight, or 1.61 sacks of cement per cubic yard of con­crete, up to abotlt 28% cement, or 11.55 sacks of cement per cu­bic yard. 
History of the investigation, the three series of tests. All of the tests, as 1outlined above, were begun in the spring of 1916 and extended throughout the summer of that yEar with the ex­ception of the test for modulus of rupture, which was not in­cluded in the original plan. To make the results more complete, it was later decided to add this tESt, including compressive tests also, which would serve to ' 'tie in'' these new tests with the ear­lier ones of 1916. As the results of the bond tests seemed some­what erratic, it was thought best to have these tests duplicated and extended, making additional bond tests in order that corrob­orative data might be obtained. More tension tests were alsu in­cluded. 
A new set of experiments was accordingly begun in. the spring of 1917 in which were made the four kinds of tests just men­ti;oned. It was the intention to use the machine mixer as in the tests of 1916, and keep the materials and method of mixing, etc. as much alike as possible to those of the preceding. tests. How­ev0r, after a few batches of this concrete had bee~·'made up, it was noticed that the fine material of the mix stuck in consider­able quantity to the mixer and that the concrete therefore dirl not have the proportiions intended. It was a.ccordingly decided to mix the materials by hand and at the same time to use a some­what wetter consistency than had been used in the former set of tE·StS. 
All of the results of the several sets of tests are reported in this bulletin. They are grouped under three different serirn: 
Series I includes the tests of 1916. The concrete is machine mixed and due to the fact that some of the finer material re­mained in the mixer, it is !Of proportions somewhat different from those intended and recorded, the fine aggregate being less 
Physical Properties of Dense Concrete 
than indicated by the, mechanical analysis curves and the cement falling· short of the amount reported by probably 10% to l2%­
Series II includes the machine mixed concrete of 1917, alike in. all particulars to that of Series I. Only tests on the gravel con­crete fall in this se·ries. Although there are but few specimens. in this group; it was thought best to report the results obtained; for, by comparison with those of Series I, they would at least help to indicate the variation which might be expected in sep­arate·-batches of the same kind of concrete and would give addi­tional values for determining final averages. 
Series III includes all of the tests of 1917 which were hand mixed and in which larger percentages 1of water were used than in the preceding series. 
In all of the concrete, essentially the same mate·rials, coarse aggregate, fine aggregate, and cement, were used. 
MATERIALS AND PROPORTIONS 
Cement. The 0ement used in the tests of Series I was a blend 
of four brands of cement on the local market: Alamo, El Toro, 
Lone Star, and Trinity. One sack of each kind was placed in a. 
Smith concrete mixer of four cubic feet capacity and thoroughly 
mixed. This was then stored in a large can with cover. A 
second batch was mixed in the s.ame V\"aY and the can, which had 
a capacity of about eight cubic feet, was thus filled. When this 
amount of cement was used up, enough to fill the can was again 
prepared. In the mean time, the cement, in sacks, wa~stored 
in the dry air of the laboratory until mixed to form the blend. 
During the progress of the work in making up the specimens 
of Series I, the cement was tested as follows: 
A sample was takm fr.om the first two bags of each brand 
used and the standard tests made. A complete set of tests was 
also made of the first batch of blended cement. Thereafter, dur­
ing the making of the compr-essive speci:m,€ns, only strength 
tests were made of the blend as new quantities of cement were 
mixed. As the cement used in making the bond, tension, perme­
ability, and abrasion specimens was taken from later shipments, 
two or three months after the cement for the compressive spei­
cimens had been used, tests were made 1of the new cement similar 
University of Texas Bulletin 
to those already outlined, i. e., complete tests of each brand and of the blend, a11id for each subsequent blend, strength tests only. The results of the tests on the cement used in this series are· 
1
given in Tables1 and 2: 
'l'ABLE 1 

RESULTS OF 'l'ESTS OF CEMEN'l' USED I N MAKING COlllI'RESSIVE TESTS, 
SERrnS I 

Kind of Cement 
No. 1 No. 2 
Kind of Test 
.S:? ·.o 
·-oo CJ N 
778 658 
2~..9 r eat, 7 Days ------------· 
797 740
8 .e e 28 Days ------------­
~~~ 
272 205
S~g 1 :3 Mortar, 7 Days -------------­
350 297
~~ ~ 28 Days -----------­
p
"' 0. 
3.11 3.17 
Specific Gravity -----------------------­
1.59 2.35
F ineness, % on JOO --------------------­
21.64 19.03
% on 200 --------------------­
0. K. 0. K.
Soundness, Normal --------------------­
0. K. o. ]{.
Acccleratcd ---------------­
2b-25m 2b-25m 
~rime of set, Initial -------------------­
4h-5m 4h-10m
Fioal ---------------------­
21%% 22%
Normal Consistency --------------------­
No. 3 
686 
662 
255 
328 
3.12 
5.63 
19.51 
o. K. 0. K. lb-15m 
2h-40m 
221h% 
No. 4 
Blend 
714 733 
715 
768 230 
261 312 
30-2 
3.15 
1.95 
2.77 
23.00 
20.51 
0. K. 
o. K. 
0. K. 
o. K. 
lh-55m 3b-30m 
2h-50JU 5h-::5m 
22%% 
22% 
(Eacb value of strength test of single brand the average of three briquHtes. 
Each from  value three  of sep strength arate mix test es of  of blended' oemeat threethe cement,  the average briquettes fr of om  n ine each  briquettes mix.)  made  
TABLE  2  

RESUL'l'S OF 'l'ES'l'S OF CEMENT USED JN IVIAKI\'f.l AT,L 0 1' 'rliE 'l'ES'l'S 
EXCEPT THE COMPRESSIVE TEW!'S, SlRIE::l T. 

Kiod of Cement 
Kind of Test 
~ .;-e
§;S..9 Neat, 7 Days 
8 00> 
28 Days
~·:: ~ 
~~ g 
3 Mortar, 7 Days ------------­
S gfoo §~~ 28 Days ------------­
~w"' 
Specific Gravity___----------------------­Fineness , % on JOO sieve --------------­% on 200 sieve --------------­
Sounclness, Normal ------··-------------­
Accelerated ----------------­
rrime of set, Initial -------------------­
Final ---------------------­
Norm al Cons!ste:1cy--------------------­
No. I 
No. 2. 
658 
642 
837 690 
218 217 
283 292 
3.11 
3.12 
2.1 
5.4 
23.1 
23.6 
0. K. 
0. K. 
O. K. Jh-lOm 
lh-Om 3b-30m 
2h-15m 23%% 
23%% 
No. 3 
598 785 
257 
347 
3.14 
1.24 
0. K. 
Jh-15m 3h-5m 24% 
,.__
No. 4 
Blend 
646 
572 
677 
778 
232 
262 
295 
319 
3.12 
3.14 
3.4. 
2.9 
23 .2 
22 .1 
0. K. 
0. K. 
0. K. 
0. K. 
Oh-40m 
lh-20m 
2h-JOm 
2h-50m 
25% 
231h% 
(Each value of strength test of single branrl the average of three briquettes. :Each value of strength test of blended cement the average of six briquettes made from two separate mixes of the cement. three briquettes from each mix.) 
Physic-al. Properties of Dense Concrete 
It is well to note that the results given in Table 1 ave deter­mined from tests which.were made in March and April, while those in Table 2 are from tests made on the new shipments in August by a di:fforent operator. In general, the tests agree as closely as eo-uld be expected. Due probably to the difference ir the time of year, the normal consistency results of Table 2 are higher and the times of set shorter than those of Table 1. 
The cement used in Series II and III was a blend of the four brands of cement already named. Enough of these cements was mixed at the time the new series were begun to make all of the specimens, the blend of cement being kept stored in a closed box lined with paper during the progress of the tests. The results of the tests on this cement are not included. 
Sand. The sand was obtained from the bed of the Colorado River near Austin. It was a clean sharp sand with an average specific gravity of 2.64; composed of about 50% quartz and flint, the remainder being limestone with some feldspar. This ma­terial was screened into the various sizes needed as described later, the coarser sand being separated from the bank run gravel, while the fine sand was sifted from material gathered from the top of sand beds especially chosen for the :fineness of the material. 
Broken Stone. The broken stone came from Comal County 
near New Braunfels, Texas. It was fairly hard limestone with 
an average speiei:fic gravity .•of 2.60 and an ultimate crushing . strength of 6,000 pounds per square inch. It consisted of ma­terial passing through a 1% sieve and held on a sieve of% mesh. This was screened into the various sizes neede·d as described later. 
Gravel. The gravel was obtained from the bed of the Colo­
rado River near Austin. It was a smooth water worn gravel 
with an average specific gravity of 2.62, 0onsisting of quartz, 
flint, and gneiss together with a large percentage of lime stone 
and containing many flat pieces. · This material was screened 
into the various sizes needed as shown later. 
Proportioning the, material. The attempt was made to pro­
portion the dry material in such a manner that the resulting 
mechanical analysis curves for the combination of cement, sand 
University of Texas Bulletin 
and 00-arse aggregate should be as near as practicable to the­maximum density curves as determined by Fuller's tests. 
It was planned to have mixtures of this kind containing vary­ing percentages of cement approximating 2, 4, 6, 8, 10, and 12 sa-cks of cement respectively per cubic yard of concrete for both gravel and broken stone. 
To illustrate the method of determining the ratio -0f the ce­ment to be used to the total dry material by we•ight, the per­centage of cement for the ''four sack'' concrete of Series I will be calculated. Assuming concrete to weigh 150 pounds per cubic foot and a sack of cement to weigh 94 pounds, the approximate percentage of cement (neglecting the presence of the water) 
4x94 equals equals 9.28%. The exact number of sacks ·of 
27xl50 cement p·«r cubic yard was determined later for each mix by weighing the. total amount of material used, including the water, and finding the weight of each specimen of the concrete imme~ diately after it was placed in the mould. The total weight of cement divided by the total weight of dry material and water and multiplied by the average weight of a specimen gives the weight of cement per specimen. This divided by 94 and by the volume of the specimen in cubic yards gives the number of sacks of cement per cubic yard 1of concrete. The actual number of sacks for each m:ix in Series I determined in this way, was 4% to 6% lower than the number used as shown above in computing the percentage. of cement, except for the "2 sack" gravel and brokm stone concrete, in which, by mistake, 4.16% of cement was used instead of 4.6{%, which resulted in a concrete with 
1.66 and 1.63 sacks per cubic yard respectively; except for the "6 sack" c-0nCTete of both kinds in which 14.4% of cement was used instead \Of 13.92%, giving a gravel concNte of 5.96 and a broken stone concrete of 5.9 sacks per cu. yd.; and except for the "8 sack" gravel concrete, where 20.3% of cement was used in place of 18 6%, giving 8.45 sacks per cubic yard. 
In Series III, the difference between the assumed number of sacks of cement per cubic yard and the actual number differed for the richer mixes by as much as 10% to 12%. This was prob­ably due to the larger percentages of water used in this serirn. 
Physioa.Z Prop-erties of Dense Concrete 

C> 
University of Texas Bitlletin 
The method of determining the proper proportions of the sand and coarse aggregate can be shown by ,explaining in detail what was done to find out the percentages of these different materials to be used for the gravel concrete. 
A plot of the: Fuller "practical" curve for sand and gravel was first made, taking 11,4" as the maximum size of the material. This curve is a combination i0f a straight line and an ellipse, 
b2 (y-7)2--(2ax~·x2), beginning at a point of tangency to the 
a2 Y axis at y = 7 and extending to a point of tangency with the straight line where x = one tmth the maximum size of the ma­
terial, or in this case .125". This straight line extends to a point where x is the max. size of material which is 114". For this material, b=28.6, a=.164 multiplied by the maximum size of materia1=.164x1 .25=.205". 
Fig. 1 shows a plot of this curve for sand and gravel, while Fig. 2 shows the curve for sand and broken stone. The point "A" is the point of tangency with the Y axis. The ellipse ex­tends from A" to the point of tangency with the straight line, "'B-C," at "B." 
To secure a combination which would give a curve like the one shown, it was decided to use two classes ;Qf coarse aggregate and three classes of fine material in addition to the cement. The coarse aggregate was therefore divided into that which would pass a 11,4" opening and be held on a screen of %" opening, and that which would pass a %" opening and be held on a screen of %" opening. By referring to the curve, Fig. 1, it can be seen that 29.75% of the first size and 29.5% of the latter size would be needed. The fine material was to be divided into that pass­ing a %" opening and held on a No. 10 sieve with opening .073", that passing this latter sieve and held on a No. 50 siE've with .011" opening, and that passing this sieve and including the ce­ment. The percentages of these different sizes needed, as read from the cune, are 12% for the first, 11.25% for the second, and 17.5% for the third. 
Enough material ;0f each of the above sizes was prepared by screening through ordinary comme,rcial sieves. As a sieve with .073" opening was not available, a No. 12 was used. 
Physical Properties of Dense Concrete 

University of Texas Bulletin 
Below are given the numbers of the sieves used and the ap­proximate size of opening, together with the. name to be applied describing each material by giving the, sieve through which it passed and the sieve on which it waS' retained. 
No.  of Sieve 11,4  Size of Opening 11,4 II  D·ascription of Material 11,4-%  
%  %, II  %,-1,4  
1,4  1,4 II  1,4-12  
12  .06111  12-50  
50  . 01111  50­and below.  

The percentages of the different sizes as used in making the "1.66 sack" gravel concrete of Series I are shown in Table 
5. The total pc-rcentage of material which should pa!'s the number 50 sieve is 17.5%. li..s it takes 4.16% cement to make this concrete, then 17.5% minus 4.16% or 13.34% of fine sand, 
i. e. '' 50-and below, '' must be used. F-0r making the ''11.5 sack" concrete, about 27.9% of cement is needed and so no sand passing a number 50 was used and only .85% of "12-50" sand, as the cement supplied a percentage of fine material practically eqnal to the part needed of that passing a number 12 sieve, which is 28.75%. 
Curves for the material used in the ''1.66 sack'' and the "11.50 sack" gravel concrete of Series 1, together with the Ful­ler maximum density curve, are given in: Fig. 3. The corres­ponding curves for the later series are shown in Fig. 4. Fig. 5 
· shows the samie set of curves for the material used in the broken stone concrete of Series I, while Fig. 6 gives those for Series III of this concrete. In the lower right hand corner, the parts 1of the curves showing the percentages of fine material are plotted to a larger scale than the entire curves. Although the curves of Series I are based on the materials used in only part of that series, they are typical of what was used in the whole. Tables 3 and 4 give the information on which all of these curves are based. These tables were ciomputed by using the mechanical analyses of the materials of each size, together with the per. centages of each which were required. 
Tables 5 to 8 show the percentages of the materials of the dif­ferent sizes, together with those of the cement and water, used in all of the series of tests. 
100 
90 
00 

~ 
«::: 
"' 
uJ 70 
[
> 
lLl 
;n 
~ 
;;;
\.] 60 I 
r. 
~ 
-~
if) 
..,
V1 
11. so 
"' ~· 
lLl 
\.] 
~ 
<( 
I-40 
l::i
z: 
~
lLl u fr: lLl 
~ 
"' ~ 
D-30 C"::l 
c 
...,
20 
"' ~ 
~ 
~ 
-
10 
0 0 .05 .10 .1 5 .zo .25 .50 .75 1.00 1.25 
I-'
DIAMETER Of PARTICLE IN INCHE:::. 
C)l 
Fig. 3. Mechani cal Analysis Curves for Gravel Concrete, Seri es I. 
100 
kJ 70
i:'i 
if'i 
\.'.) 
~ 
;r; 
n 
it 50 
20I 
10 
0 

,_. 
J) 
~ 
~­"' ~ 
.,.,.
"'· 
'.-! 
~ 
........ ""3 
8
"' 
~ ~ 
"'.,.,.
*' ~ 
)!'i~. 4. Mechfl,nical Analysis C11rve;;; !or Grfl,vel Concrete, Series II anq Ill, 
100 
90 
80 
w 70 
G':i 
<?) 
~ 60 
if) 
if) 
if_ 50 
w 
'.ii.
!c 40 
w 
ii! 
LJ 
"-30 
20 
l(J 
0 

~ 
<.;:::: 
"' c>· 
,.,..... 
l ·"' ..,,
.,.... 
~­
"' 
~ 
......., 
N 
.~ 
"" c;; 
Q 
.C 
;:! 
s, 
ci
.,.... 
"" 
1.2.~ 
....... 

-:i 

00
inn 
~ 
~ 
t..J 
"'· 
~
7i 
"' ~ 
~ 
""·
....
?i 
~.:e: 
vi 
ct 
<:>
......., w 
~ 
<..:J 
~
;S 
~ 
r. 40 
t..J 
Q:j 
~ 
~ 
t..J 
.,.....
CL 30 
..... 
"' 
~­
20 
10 
0 
0 	.05 .10 .15 .20 .25 .50 .75 1.00 1.1.5 PIAMETER Of PARTICLE. IN INCHE.S 
Fig. 6. Mechanical Analysis Cur ves for Broken 'Stone Concr ete, Series III. 
Physical Properties of.Dense Concrete 
TAB.LE 3 tuECHA~ICAL ANAL\:SES OF )1A1'ERIAL USED 
Size Sieve 
of open­
ing in 
inches 
-----1 ----­
1\4 
1\4 l 
1 % 
% 
'h 'h 
14. 
14. 
Vs 1/s
10 .065 20 .0328 28 .0"23'2 35 .0164 48 .0116 65 .0082 
100 .0058 200 .0029 
Series I 
Porccot pass-Percent pass­ing for 1.66 ing for ll.50 sacks of ce-sacks of ce­
ment per cubic mcnt per cubic 
ya"d yard 
98. 4  93. 4  
83.5  83.5  
70.3  iO.:J  
55 .3  55.3  
41.6  41.6  
36.1  30 .1  
31.1  31.l  
26 .3  28.·3  
23.0  28.3  
19.9  28.l  
17.4  ~7.9  
8 .5  27 .9  
4.7  27.1  
I  3.3  21.8  
'  

IN GRAVEL COSCRETE 
Series II and III 
------------· 
Percent pass­
Pel'ccnt pass-ing for 11.5:> 
ing for 1.64 a 1d 10.86 
sacks of ce-sacks of ce­
ment per cubic mcnt per cu bic 
yard yard 
~ ~-;. I :5 .1 
78.3 78.3 
07 .3 67.3 
54.l 54.1 
40.2 40.2 
35.0 35.0 
30.6 30.6 
25.7 28.5 
22.3 28.3 
19.6 28.l 
15.7 27 .9 
7.7 2i.9 
4.8 27. I 
3.5 21.8 
'fhe values given above for Series I represent the results obtained for this series by combining the mechanical analyses of all of the materials, including the cement, used in making all of the tests except the compressive tests. The values given for Series II and III were obtained by combining the mechanical analyses of all the materials, including t.he cement, used in mak­ing all of the trsts of these serirs. 
TABLE 4 
MECHAI\!OAL ANAL\"SES OF MATERIAL US.ED IN BROKEN S'l'ONE CONCRETE 
SiC\"C 
-----·­
1~ 
% 
'h 
\4 
Vs 
:o 
20 28 35 48 ().5 
JOO 
200 
Size 
of open­
!ng in 
~nches 
1\4 1 % 
'h 
\4 
'h 
.065 .0328 .0'232 .0164 
.om; 
.0082 .0058 .0029 
Series IIISeries 1 
Percent pass-· Percent pass-Percent pass-Percent pass­
ing for 1.63 ing for 11.30 
ing for 1.61 ing for 10.53 aeks of ce-sacks of ce­
sacks of ce-sacks of ce­
ment per cubic ment per cubic ment per cubic ment per cubic 
yard yard yard yard 
------------11------·I-----­
97.8  97.8  96.0  96.0  
80.2  80.2  83.1  8'2.9  
70.6  70.6  72 .3  72.0  
49.0  49.0  63.2  62.9  
43.0  43.0  H .2  43 .9  
38.2  38.2  39.8  39.i  
33.3  33.3  34.0  33.7  
28.3  30.3  28.0  30.4  
24 .6  29.4  24. l  29 .2  
21.2  28.6  21.1  28.5  
18.3  28.0  16.8  28 .0  
8 .S  27 .IJ  7.9  27.9  
4. 7  27.J  4.8  27. l  
3.3  21.8  3.5  21.8  

'l'he values given above for Series I represent the results ob­
University of Texas B11lletin 
tained for this sel'ies by combining the mechanical analyses of all the materials, including the cement, used in making all of the ksts .except the compressive tests. The values given . for Series III were obtained by coml:)ining the mechanical analyses of all the materials, including the cement, used in making all of the tests of this series. 
*'l'ABLE 3 
:11AT.i£HJAL FOR GRAVEL OONORE'l'E, SERIES I AND Il 
Prcportion of each material expreEsecl in percent by weight 
Sacks of 
Ccmm1t J~r 
Cu . I'd. of 
Cc~11cnt 
Concrete 
----.--­
1.66  4.16  
3.8  9 .28  
*5.54  I  l4.40  
15.82  14.40  
5.96  14.40  
8.45  ~0.30  
9.67  '.03.20  
11 .50  27 .90  

Sand 
50­
---·­
13 .34 
8.22 
3.10 
3.10 
3.10 
0.0 
0.0 
0.0 
of total dry rnaterial · 
Saud 
12-50 
11 .25 
11 .25 
11.25 
11.25 
11.25 
9 .95 
5.55 
0.85 
Sand 
14-12 
12 .00 

12 .00 


12.00 12 .00 12.0() 
11. 74 
12 .00 
12.00 
-
Gravel 
%-1!. 
29.50 
29.50 
29 .50 
29 .50 
29.50 
28.91 
29 .50 
29.50 
Gravel 
11!.-% 
29 .75 

29 .75 

29 .75 

29 .75 


29 .75 

29 .10 


29.75 
29.75 
'\"\ ater 
7.71 
6.73 . lJ.57 
8 .20 

6 .78 


6.85 
7.19 
j .70 
*'l'hc "very \Yet" mixture. tThe "wet" mixture. 
'l'ABLE 6 lllA'l'ERIAL FOR GRAVEL CONCH.E'l'E, SERIES IlI 
Sacks of 
Cement J)er 
Ou. Yd. of 
Concrete 
l .64 

3 .76 5.88_ 

9 .18 

10 .86 


8.20 
ProJ)ortion of each material expressed in oercent by weight o! total dry material 
·-­-­ ·­-----­---­ 
Cement  Sand  Sand  Sand  Gravel  Gravel  ·\Yater  
50­ 12-50  1.4-12  %-V.  11!.-%  
4.16  12.34  11.25  12 .00  29 .50  W.75  8.12  
9.28  8.22  11.25  12.00  29 .50  29 .75  'i .77  
. 14 .4l  3!10  ll.25  12 .00  29.50  29.75  7.40  
20.30  n.o  9.%  11 .74  28 .91  29.10  8 .31  
23 .20  C' .0  5.55  12.00  29.50  29.75  9.20  
27.90  0.()  ().85  12 .00  29.50  29 .75  10.26  

*'l'ABJ.. _E 7 
)lA'l'EHIAJJ FOR B110KEX STONE COXCRE'I'E, SERlES I 
PrcJ)Ortion of each material expressed in percent by weight 
Sacks of 
Cement per 
Ou. Yd. of 
Cement Sand 
5()­
Concrete 
-----1 ------· ­
4.JC, 14 .3-1
J. .68 
9.28 9.22
3.75 
5.90 
14.40 4.10 
7.63 
18.60 0. 00 
9.45 
23.20 0.00 
27 .90 0.00 

11 .30 


of total dry material 
Sand 
12-50 
12.5-.'1 
12.50 
12.50 
12.40 
7.80 
3 .10 
Sand Stone I Stone Water 
\/,,-12 %-'4 11;.-:y. 
11.75 28.50 28.75 8.S5
I' 
11.75 28.50 28.75 <.9"2 
11.75 28.50 28.75 7.50 
ll.75 28.50 I 28.75 7.6Q
ll .75 ~S .50 28.75 8.55 
11.75 28.50 28.75 8.65 
-~------------------­
*Sec page 6 for possible variations in proportions. 
Physical Properties of' Dense Conc1'ete 
" l '.lll LE 8 
:.UA'l'ERIAL FOH, BROJrnX STO:\'E OOSOHE'l'E, SERIES lll 

I Prcportion of ench material expressed in percent by weight Jc~{,l~~to~erl --------of total dry material ·-----~· 
Cu . . Ld. of Cement Sand Sand Snnd Stone Stone \·\"atcr 
Concrete 50-12-50 1;,,-12 %.-'4 1\4-%. 
--· ·-­
J .61  4 'IG  l -l. 3J  12 .50  11.75  28 .50  28 .75  9.58  
3.63  9.28  9 .22  l~ . 50  11 .75  28.50  28 .75  9 .10  
5.64  l J .40  U()  12 .50  ll .75  28 .50  28 .75  9.23  
7.38  18.60  0.00  12 . ~0  11 . 75  28.50  28.75  9. 76  
9 .0-2 10.53  I  23 .20 27.SO  0.00 0.00  i .80 3.10  11 .75 11. 75  28 .50 28 .50  28. 75 28 .75  10.J5 11.85  

*The perce·ntages of the dry materials arc recorded the same as for Series I, although in some case.c: tho actual percentage dif­fered very slightly from the recorded one. 
To bring out as clearly as possible the sizes and intenclation of the component materials and to make evident the drnsity of the finished concrete, photographs were taken of polished section" madr frnm some cf the broken tensile spc·cimcns of the "six-· sack" and the "twelve-sack" concretes, both ?ravel and broken stone. Prom these photographs, which wore made to full siz:', typi<·al po1tions \\'ere selceted and nsed to make the Cll'ts shown -011 pp. 22 and 23. 
Consd:ency. For Series I, the consistency adopted nftm· mak­ing a tl'ial batch of conc1·Et e, was such that the conerctc \\'Ould not fiow bllt could be readily talllped to c0111plctcly fill the moulds. The ami:;uut of water neccc4t-:ai·y for each inix \\'<IS de­termined by noting tho appem·ance of the concrete in the mix.cr and adding enough water to make tho corn.:istcncy tho same as the standard. This, of course, required diffe.rcnt percentages of water with the changes in the percentag-~s of cement and fine . aggregate. For Series II, the; same pi.TCcntagcs of water were used as in Series I, but tho eonsisteney seemed a trifle dryer. 
'l'o find out the effect of using ti wetter consistency, parallcI tests were run with a "six: sack" (actually 5.82 sacks per cubic yard) gravel concrete in Series I in which 8.2% of wntcr was used instead d 6.78%, which was the standard for this mixtmc. This concrete was wet enough to flow into tho moulds and take its place with a little jarring or spading. In Sei·ic'> II, a "very wet" "six sack" mixture was also made up in which 1L57% of water was used. 
For Series III, a wetter consistency than tho standard of Se­
University of Texas Bulletin 
"6 Sack" Broken Stone Concrete. (Full size) 
"6 Sack" Gravel Concrete. (Full size) 
Physioa,i Properties of Dense Concrete 
"12 ·sack" Broken Stone Concrete. (Full size) 
"12 Sack" Gravel Concrete. (Fuil size) 
University ol Te:ras Bulletin 
ries I and II was adopte.d, the concrete being so wet that, as mixed on the floor, it soon flowed and the mass flattened out, the p2rcentrtges of water being from 1110 to 1/3 greater than in the preceding tests. 
MIXING, MOULDING, STORING 
l'l1ixing and monlding. The concrete of Series I and II was machine mixed. 'l'he proportions having been detel'mincd for the various concretes, the materials were carefully weighed out and placed in the Smith four cubic foot mixer, the interior of \\'hiC'h had been previously wet. ·water was then added and the mixer allowed t o. run for five minutes at 19.4 r. p. m., after which the concrete was dumped into a wheel barrow and then placed in the moulds. In Series I, the concrete for each per­centage of cement for the tension, bond, permeability, and abra­sion tests was made on the same day, the mixing being done in two batches, and part of the specimens for each kind of test being made from one. batch and part from the other, except for the; permeability test specimen which was made from the fin;t batch. The concrete for the compressive specimc-rni, while mixed in two bntches for e&ch percentage of cement, was made at an· earlier date-. When the tests were made, some samples from each batch were used. 
In Series II, the specimens for all of the tests for each per­cent.age of cement were; ma.de at one time. 
'l'he method of mixing in Series III differed from that in Series I and II, first, in that the materials were hand mixed: second, in that three: batches were made up on as many different days, for each percentage of cement, one specimen being made from each bat12h fol' each of the different kinds of tests. The materials were thomughly mixed dry on a cement floor, water was then added and the concrete turned until homogeneous and of uniform consistency. At the time of testing, one specimen from each of the three batches was used. 
Storage. After twenty-four hours in air, the specimens were removed from the moulds, except as later specified, and placed under water for eleven days. They were then remnved and in most cases some were immediately tested, while the remainder 
Physical Properties of Dense Concrete 
were kept in the dry air of the laboratory for testing at tht: various ages shown later. 
COMPRESSION TESTS 
S.cope vf the investigation. For the tests of Series I, one hundred and forty-four specimens, Eight inches in diameter ancl sixteen inc.hes Long, were made of the standard consistency. Onc­half of these were of broken stone concrete, the other half of gravel. For. each different percentage of cement, there were twE'lve specimens. Three of each kind were broken at age:-; oi 12 days, 28 days, and 3 months, and the remaining three were held for testing at the end of one year. In addition, twelve spe­cimens of "six sack" gravel concrete, the "wet" mixture, were made and tested as outlined above. 
For the tests of Series II, twelve specimens were made of 
gravel concrete, there being tvvo specimens for each different 
percentage of cement. One of each kind was broken at <lges 
of 28 days and 3 months. In addition, four specimens of ' 'six 
sack' ' gravel concrete, two 0£ the ''wet' ' mixture and tw.o of 
the ''very wet'' mixture, were made and tested. 
For the tests of Series III, twenty-four specimens were made 
of the gravel concrete and thirty-six of the broken stone con­
crete. For i'ach different percentage of cement, there were four 
specimens of gravel and six of broken stone concrete, one-half 
of which were tested at the age of 28 days, the remainder at th'.) 
age of 3 months. 
Moulds. The moulds used for the compressive specimens 
were made by cutting an eight inch wrought iron pipe into 
lengths of sixteen inches and carefully facing the ends. A cut 
was "then taken along a longitudinal section through one wall 
of each length of pipe and extending from top to bottom. The 
edges of this cut were planed until they failed to meet by ahout 
one-quarter of an inch. Tw·O· small angles, one on each side of 
the opening near the top, and two similar ones near the bottom 
were riveted to the pipe. Through the outstanding legs of ~hese 
angles were passed two and one-half inch bolts by which the 
edges of the pipe could be drawn together. A small bent plate 
was riveted on the outside. near the top of the mould and oppo­
University of Texas Bitlletin 
site the part where the cut was made. The mould could be lifted by using this plate and the top ~ngles as handles. A smooth cast iron plate lO"xlO", resting on three short legs, sup­ported the mould. 
Mixing, placing, etc. The concrete for each percentage of ce­ment was mixed as already described and placed in the moulds in layers of about two to four inches and thoroughly tamped with a tamper four inches in diameter and weighing 13 pounds. 
D1ometer-when c.lo.:i e d -e.· Heogh! -16" 
Fig. 7. Mould for Compression Specimen. 
The specimens were smoothed off on top and weighed in the moulds which had been previously weighed empty. After set­ting until the next day, they were removed from the moulds and stored as already described until the time for breaking. 
Testing. 'l'he specimens of Series I, broken at each p•eriod, were selected so that part were taken from the first batch mixed and part from the second batch. In Series II and III one spe,.. cimen from each batch mixed was broken at the end of each period. Several hours before testing, the top of each specimen was covel'e.d with a thin layer of plaster of Paris to give a smooth bearing for applying the load. 
A Riehle 400,000# universal machine was used in making the compressive tests. The specimen was carefully centered on the w·eighing ta.ble and a hemispherical compression block placed between it and the movable cross-head. The load was then ap­plied by motor, the cr.oss-head being run down at the rate• of 
Physioal Properties of Dense Concrete 
.05" per minute. The load was continuously applied until rup­ture in the case of the 12 .day and 28 day tests; while in the three months and one year tests, since the compressometer was used to take deformation readings, the load was applied by in­crements, varying in size with the richness of the mixture, with short time intervals betwHm the appliGations of the increments. This method was followed up to a point near the rupture of the specimen, when, on the removal of the compressometer, the load was applied continuously to the1 breaking point. 
The concretes containing 6 sacks or mo;re of cement per cubic ya:r;d, at ages of 28 days or over, br.oke suddenly with a loud re-· port, the pieces flying out some distance from the machine. The: specimens from the mixes leaner than 6 sacks per cubic yard merely cracked and gradually crushed. 
Results. Tables 9 to 11 give the results of the tests on the 
gravel concrete, while Tables 12 and 13 give those for the broken 
stone concrete. The values for >each series are the averages de­
termined by combining the individual results ;obtained from test­
ing the specimens of that series. 
Figs. 8 and 9 give the curves for the gravel concrete, while Figs. 
10 and 11 give those for the broken stone concrete. These curves 
were determined by first plotting the individual r>esult of the 
test of each specimen for a given age for all series. The lowest 
points for a given series weve then joined in ;order with light 
straight lines. The next lowest points were then connected, and 
so on. This was done for -each series. The heavy curve was ob­
tainoe.d by averaging the unit breaking loads determined from 
the light curves for 1, 2, 3, -----11 sacks of cement per 
cubic yard ~.espectively. It accordingly represents average 
values as determined by combining the results of all thoe. series 
for a concrete .of given coarse aggregate and of a certain age. 
Particular attention is called to the values indicated by • 
e · o in Figs. 8 and 9. The first shows the results of the 
tests on the ''wet'' mixture ''six sack'' gravel concrete of Series 
I, the second, the same thing for Series II, while the last showR 
the results for the "very wet" mixture "six sack" concrete of 
this series. 
The curves of Fig. 12 show average values for Series I for the; 
different ages. These are based on the d·ata giwn in Tables !} 
and 12. 
10000 
r 
~ 9000 
t.J 
cc
:3 5000 
~ 
~ 7000 
().. 
if) 
D
Z:. 6000 
:::> 
t2 
z:_ 5000 
r t; 4 000 
z:_ 
OJ 
(l:'.
tn 3000 
OJ 
> 
if) 2000 
if) tLJ 
ex: 
CL.
I: IOOO 0 
u 0 

NJ 
~ 
"' ~· 
.,;;; _ 
<:::"'" a
-....,., ""'l 
,., "' 
"' 
"' 
~ 
"' ~ 
""" 
~· 
10000 
5 9000 
z 
~ 8000 
~ 
Cf 
<fl 7000 
er: 
lU 
£L 
<fl 6000 
0 
z 
:::> 2 5000 
~ 
; 4000 <.:J 
z 
w 
::: 3000 
<fl 
w 
>
Ji 2000 
"'w 
"' 
~ 1000 0 
u 
0 

~ 
<c:: 
"' ~· 
"'
-

'ti 
i-.... 
.... 
"' ~· 
<::>
-b 
;:S
"'" 
"' 
"' 
Q 
<::> 
;:S 

" ;_;; "' ­
N) 
'.z> 

10000 
I 
u 9000 
~ 
i...J 
~ 8000 
:J 
CJ 
tf) 
.	er: 7000 i...J 
CL 
2 6000 
r:. 
:::i 
~ 
r. 5000 
I 
t; 4000 
r. 
i...J 
er::
tn 3000 
(JJ 
> 
if\ 2000 
tf) 
(JJ 
C!'. 
~ IOOO 
0 
v 
0 

<:.¢ 
~ 
~­
""... 
-~ ~ ~ <::, 
;.., 
"" ~ 
~ 
tt1 
~ 
.,.....
"" 
~· 
10000 I 
u ~ 9000 lJ 
er: 
<(
58000 

~ 
~
<fl 
.."'. 
Cr 
IU 7000 
~ 
CL 
~ 
</) 
~ 6000 
~ 
:J 
"'< 
""
.,....
~ 
5000 
~ 
"' ~­
~ 
I 
l::::J
t; 4000 
z 
"" 
lJ 
D:: 
"" ~ 
<'.':)
tii 
300 0 
c 
;:l
lJ 
0 
"'<
>
ii) 2000 
.,....
"" 
cf\ 
lJ 
"" 
D:: 
~ \000 
0 
u 
0 0 2 3 4 s 6 1 e 9 10 II 12 
C>:I
SACKS Or CEMENT PE:R CUBIC YA RD OF CONCRETE. f--' 
Fig. 11. Ul-timate Compressive Strength of Broken Stone Concrete, Age 3 Months. 
v~ 
8000 

·~ 
5 7000 
e= 
(,_] ll:: 
<1: 
:::> 6000 
0 
tfl 
c-:: 
er: 
--~.
t.J 
<-;
D-5000 
V) 
·% 
D 
~­
r: 
'.::.::: 0 D-4000 
::J 
2 
~ 
'"-3 
r:: 
"' 
~
t? 
..:.·:
r: 3000 
(,_] 
t;j
cc 
r-
v, 
.,_ 
"' ~· 
(,_] 
'-'""-~ 
2: 1000 
tfl 
<{) 
UJ cc "­
6 1000 
u 
0 
0 3 4 5 6 7 8 9 10 II 11. 
.SACKS OF CEMENT PER CUBIC YARD OF CONCRLTE:. 

Fig. 12. Ultimate Compressive ·strength of Gravel and Broken Stone Concrete at Different Ages, Series I. 
Physical Properties of Dense Concrete 
· TABLE  '  
'COMPRESSIVE  S'l'R ENGTII  OF  GRAV.EL  CO:'<CRE'l'E,  SERIES  I  
(Each  Result  tbe Average of •res ts  of 'l'hree Cylinders  8"xl6")  


1.66 
3.80 
*5.82 
5.96 
8.45 
9.67 
11.50 

1.63 
3.75 
5.90 
7.63 
9.45 
11.30 
Weight per cubic foot Ultimate unit stress In 
in pounds , pounds per square inch 
Age 
12 cla. 
149 .0 152 .0 152.0 15'1.5 
155 .3 
156.1 
154 .9 
Age 12 da. 
147 . 7 
151. 7 
153.9 
154 .2 

153 .3 


153.2 
Age Age Age Age 28 cla. 3 mo. 1 yr. 12 da. 
143.5 141.9 141.3 282 
150.0 147.9 147 .5 1418 
150.4 148.1 147 .9 2347 
153.2 
152.3 150.9 3185 

154.5 
153.7 152.3 4547 


154.8 15'1.0 102.9 4662 
154.3 153.0 152.2 4377 
*Tho "wet" mixture. 
tAge 31 days. 
TABL1" 10 

COMPRESSIVE S'l'RENG'l'H OF GRAVEL 
00.'W RE'l'E, SERIES Ii 
(Each Result from the •r est of One CyJin-

Sacks of ce­
ment per cu . 
yd . r f con­
crete 
1.66 
3.80 
<IE!) ,f -J 
t5 .86 (5.82) 5.9;; (5.S6) 
8.45 
9 .67 11.55(11. 50) 
dcr 8"xl6") 
Ultimate un it stress in pounds oer square inch 
Age 28 days 
Age 3 mo. 







,______, 
381 
289 
2060 
2037 
149) 
206) 
3332 
3809 
J7JO 
5785 
5350 
6880 
6120 
5660 
5500 
6210 
Age · Age 
Age 
l yr.
28 cla. 3 mo. 
574
445 604 
1948 2565 
2401 
4180
t3310 4367 
4379 5237 
<778 
5820 6733 
6656 
5802 6273 
6640 
5447 5803 
6180 
'l'ABLE 11 
COMPRESSIVE S'l'RENG'l'H OF GR..\.VEL 
CONCRE'l'E, SERIES III 

(Eacb Result the Average of Tests of 
'l'wo Cylinders 8"xl6") 

Sacks of ce­
Ultimate unit stress in 
ment per cu . 
pounds per square inch 
I
yd. of con­crete 
Age 28 clays 
1.64 
351 
2180,
3.76 
5.88 
4535 
S.20 
4550 
9.18 
4533 
10 .86 
4330 
*One specimen 
Age 3 mo. 
*541 2216 4257 5357 4810 4985 
*The "very wet" 1nixture. tThe "wet" mixture. 
'l'ABLE 12 
COMPRESSIVE STRE:\'G'l'H OF BROKE:\' S'l'ONE CONCRE'l'E, SERIES 
(Each Result the Average of Tests of Three Cylinders 8"xl6") 
Weight per cubic foot Ultimate unit stress in 
in pounds 
Age 28 da. 
141. 7 
149. 2 
151.9 
152.2 
152.0 
152.9 
Age 
3 mo. 

146.() 
150.4 
151.2 
150.9 
151.6 
pounds per square inch· 
Age 1 yr. 
145. 7 
148 .9 
150.2 
149.8 
150.9 
Age 
12 cla. 

28-1, 1278 2871 4397 4858 4950 
Age Age Age 
1 yr_. 
28 da. 
3 mo . 
461  598  590  
1718  2"227  23HI  
3965  4772  4905  
5153  58-13  6272  
5787  6450  6360  .  
6263  6597  7103  

·-------------­
University of Texas Biilletin 
TABLE 13 
COMPRESSlVE S'l'RENG'l'H OF BROKEN S'l'ONE CONCRETE, SERIES III (l' ach li~sult the Average of Tests of 'l'hree Cylinders 8"xl6") 
Sacks of ce­ment per cu .  Ultimate unit stress in pounds per square inch  Sacks of ce­mcnt per cu .  Ultimate unit stress in pounds per square inch  
yd. of COil· crete  A1.te 28 days  Age 3 mo.  yd. of con­crete  Age 28 days  Age3 mo.  
--­ ---­- 
l.61 3 .63 5.6-1  279 1174 2297  449 1499 3192  7 .38 9.02 10.53  3237 4388 4286  4123 5050 4777  
--­--·--­- 

EL,\STICITY TESTS 
Scope of tfi,e investigation. In Series I, a test was made on three of the compression specimens of each kind at the age of three months and two of each kind (in some cases three) at the :ago of one year. All of the one• year speci~ns used and two ·f)f the three mcmths specimens for each percentage of cement werei tested so as to det ermine the modulus of elasticity, while in the case of the third three months specimen, the load was :applied and removed several times and the pe:rmanent set ob­tained, as wdl .as the stress-strain relation for both loading and lmloading. 
No tests for the modulus of elasticity were made on the speci­mms of Series II. 
In Series III, two of the compression specimens of each kind 
wel'e tested at the age of three months. In those cases where 
results of the two tests did not agree closely, a third specimen 
was tested. 
Testing. A wire wound dial compressometer was used to measure the de.formation taking place in the test of these speci­mens. This apparatus is shown in the sketch, Fig. 13. It con­sists of two yokes attached to the specimen by four pointed thumb screws, two in each yoke, and placed ten inches apart by two removable spacing bars. On one side the yokes are main­tained a constant distance apart by a bar connected at the top and bottom by ball and socket joints closely held by small springs, while on the other side the movement is measured by· means of a pointer moving over a dial fastened to the upper yoke and graduated to read to two ten thousandths of an inch. This pointer is connected to a small cylinder, free to turn, around 
Physioa·l Properties of Dense Concrete 
Fig. 13. Compr;essometer. 
which passes a fine wire, one end of which is attached to the lower y.oke, a small weight being connected to the other end. As the two yokes are in effect pivoted on the side where the constant distance bar is, the other side, to which is attached the dial, moves twice the distance that the two screw points move, provided these have been correctly placed half way between the bar and dial. This movement is transmitted by the wire to the cylinder which is made. to turn and the amount of shortming is thus determined, readings of actual deformation as close as one ten thousandth of an inch being obtained. 
The apparatus was carefully attached to the specimen to be tested about midway between the top and bottom, the screw points being centered as near as could be done by eye. The specimen was then placed on the machine, the spacing bars re­movied, and the pointer set to read zero. The load was applied by increments of from 5000 to 20,000 pounds, depending on the estimated ultimate strength of the concrete. As soon as the load set on the b2am of the trnting machine had been reached, a dial 
University of Te:ras Bidleti1i 
reading was taken. After an intermission of about one-half minute, a second increment was applied and so on up to the point at which it was thought best to remove the compresso­me·ter. It was not eonsidered safe to take readings up to rup­ture as in testing many of the richer mixes, the concrete, in breakii1g, popped out with great violence. 
Tests similar to the ones just outlined wel'e. made on the1 one year specimens and on two of each set of three specimens tested at the age of three ni;onths. Thei third three months specimen of Series I was tested in a slightly different manner, as shown by the following : 
The load was applied in increments similar to those already mentioned up to a point equal to about one third of the maximum load to be expected, then the load was .released by decrements, compressometer readings being taken at the same load points as when the load was being increased. This was done to see to what extent the specimen return~d to its original length at the different loads and to find out the permanent set. A second series of readings was made in the same way extending to a point approximately equal to a load two thirds of the maximum and the load taken off as before. The load was then applied with the usual increments, with approximat•ely one half minute1 intervals of rest, up to the point at which the compressometer was re­moved. 
Results. To obtain the values of the modulus of elasticity, a curve for each mix was plotted based on the average unit de­formations and their corresponding loads forthe specimens tested without the load being repeated. Curves for gravel are shown in Fig. 14, and those for broken stone in Fig. 15. Fig 15a shows the same curves grouped to make comparison easier. These are biased on the data of Series I, the three months test, and are typical of the curves for the one year test of this series and of the three .months test of Series III. From stress-strain curves like these, the secant modulus of elasticity for a load equal to about one third of the total load could be obtained by using the deformation between the point located by the1 first load incre­ment and the one third load point. The former was chosen rather than the origin; for in nearly every case, there SE1emed 
6000 
5000
:c 
u 
r:. 
UJ 
():'. 
<( 
::i 4000 
a 
.n 
·er: 
UJ 
Cl.. 
(Q 3000 
r:. 
::i 
2 
r:. 
D 2000 
<( 
s 
f­
z:: 
::i 
IOOO 

'"ti 
~ 
~ 

"' 
[ 
'"ti
.., 
a 
'lj 
~ 
""
.,...
""· 

-"" "' a 
;;;: 
"" ~ 
"' 
"" 
Cl 
a 
;;;: 
" 
.,...
"" ~ 
"" 
.;.,:) 
-J 
G
r: 
v 
r. 
"-' 
D:'. 
<( 4000 
:::> 
0 
VI 
(>: 
"-' 
Q._ 
Vl 3000 
D 
r. 
:::> 
f2 
~ 
D
' <( 
g 
~ 
r. 
:::> 

,Fi~. 15. Avera~e Stress-strain Cµryes for J3r9!1;en St9ne ·C9ncrete1 
00 
C'.:i 
02 
"' ~· 
-; 
...

.,...
"' 
<::: <::>
-..,.. "'3 
"' 
f.-l .,:;
"' 
t:r;:i 
.: 
~ 
"'.,... 
~· 
Age 3 Months, Serie:;; I, 
:c 5000 
u 
f ~
ri-1-+-+-+-_J_wLf1A~7/
i..J 
:::> 4000 
~ 1iv
5\ 
er: ,#_W
lU ()_ 
;Q 3000 
r:. )(/ ,,.,, 
:::> 
~ n --t-+--hv/a_-10.~ ....,;;; 
r:. 
<( ---t-+-J.--?J/~r:___J/I~
D 2000 
g . , ,.;W,W" 
I­$[~'" f 
:::> 
1000 
/,w 
~~ .......

_ l 
0 
" -A 
0 3 4 5 6 0 GRAVEL CONCRETE 
6000 I 
/ 
'"ti
;;.­
1$ 
I/ 
""'..."' 
<">
)/ """ .~v i "" 
.~ 
JL?;Y 
07
'""_/~/

,,;P)Y ,,,,,,,,..,~ 

Y,I/ 
// 
IC ,., ~"·' 
~ I
10 e IO 
DROKEN STONE CONCRETE. 
~ 
Cb 
"i
.....
.,.., 
"' 
"' 
-
c t:I 
;Z 
"' 
"' 
"' 
-c 
c 
;Z 
<">
"i 
..... 
"' 
"' 
UNIT DEFORMATIONS IN .OGO\ 
i:o
"'° 
Fig. 15a, Average Stress-straill C11rvei; for Gravel aP.<t Broken Stone Concrete, Age 3 l\1op.ths, Series I, 
.University of Texas Bulletin 
to be some initial error in the compressometer reading, as can be seen by referring to many of the curves. . 
Tables 14, 15, 16, 17 give for each mix the ultimate unit stress obtained by averaging the results from testing the specimens used in finding the modulus of elasticity, and the. modulus of elasticity determined from the average curve described abo:ve. The variation of the modulus with sacks of cement per cubic yard of concrete is shown for the gravel concrete in Fig. 16, while in Fig. 17 is shown this variation·for broken stone con­crete. The curves drawn in light lines were obtained by plot­ting the data givm in the several tables. The heavy line curves 
wer~ obtained from these light line curves by averaging the · moduli determined from them for 1, 2, 3 ----11 sacks of ce­ment per cubic yard respectively. 
The data taken for determining the permanent set for some of the specimens may be: considered almost too erratic to warrant p1esentation here. However, in Figs. 18 and 19 are given the curves based on this test of the third thn•r; months specimen of each kind for Series I. 
TABLE 14 
MODULUS OF F.LAS'l'ICJ'l'T OF GRAVET, CONCRETE, SERIES (Eacb Result tbe Average of 'l'ests of Two Specimens) 
Sacks of cement 
per cu. yd. of 
concrete 
J.66 
3.80 
*5. 2 
5.96 
8 .45 
Q.67 
11.50 
*r.rhe "wet" mixture. tThrce specimens. 
Age 3 montbs 
Modulus of IUltimate unit 
elasticity in compressive 
potinds per stress in pounds square inch quare incb
Iper 
2,800,000 4 ,400,000 4 ,200,000 t5,400,000 
5,800,000 
5,600,()()() 5,400,()()() 
617 264() 4400 5235 6760 6390 59'20 
TABLE 15 Age 1 year 
Modulus of 
elasticity in 
p ou nds per 
square incb 
1 ,300,000 t4,400,000 4,100,()()() 
5,700,()()() 5,700,000 t5,200,()()() 
Ultimate uni 
compressive stress in poun els per square inCb 
-
585 2400 4350 
665() 6650 6180 
MODUJ,US OF F.J,AS'l'ICI'rY OF GRAVEL CO~CRF.TE, SERIES III 
(Eacb Result the Average of Tests of 'l'wo Specimens) 
Age Tbree 111ontbs 

Sacks of cement per cu. yd. of concrete 
1.64 
3.76 
5.88 
8.2Q 
9.18 
10.86 
Modulus· of Elasticity in 
pounds per square inch 
1,700,000 3,900,000 5.J00,000 5,000,000 4,900,()()() 5,000,000 
----·-----"------­
Ultimate unit compressive stress in pounds per sq. in. 
49.'i (approx.) 2215 4200 536() 4810 4985 
:r:: 
u 
;: w 
~ 
::i 
a 
<I) 
D:: 
w 
D... 
:lf 
0 
C> C> 
g 
~ 
>­
r 
Ci 
~ 
<:( 
'--' 
li...l 
l:5 
if) 
3 
:::J 
C\ 
0 
2:: 
!000 900 
800 

'i:; 
~ 
"' 
700 
g· 
..... 
600 
1
<'> 
.... 
;·""" 
500 
. "' -2.. 
400 
t::; 
~ 
<'>
300 Q 
;s ~ 
(":>
200 
;;; 
~ 
100 0 
. >;'". 
IOOO 
r 
u 
~ 900 
U-1 
Cl:: 
§ 600 0 
If) 
c:r:: 700 
"-.I 
CL 
600I 0
'"'0 
0 
0 

500 
z: 
>--400
t: 
u 
[;; 300 
<t: 
_J 
lU 
"--200 
0 
if) 
3 100 
::i 
D 
0 
L: O 

~ 
~ 
;· 
~ 
...~
.,.,. 
<.::: 
<::>
-'"S 
<I> 
~ 
~ 
~ 

...... 
~ 
~· 

6000 
' ! 
]Y 
' '' ~v :i:: 5000 u GRAVEL CONCRETE A9e 3 mo V,/ v '"tj
f l---t-----1 -f1r~l Appl1ca11on /JL_V-1---.J---ri--+--+-t------i-r-----r---1 ;:.<
(._____)_-,L_
,,,
!.J 
~ Second Appl oca1oon /V/ / ""' § Third Apphcot1on / / [
~ 4000 
Cr: 
/V J/ :? 
ll.-~ 
(t>
..,
J) .,.. 
lW // ~/ / ~­
~ 3000 
::;, "' 
~ ~,/; /' // c 
~ -t;
1 
;jl 2000 wl0'l v 1 J~, ""' 
(t> 
<'.".:) 
g t: ~~rlivo1 ./ 01 ~ ~vv~?10,L+.--/b7---+---l-1+--+------t---t------i "' ;Z c 
"L ;Z 
::i <:>
~~f;( !11 /ft~ry~ ~vv · J .., 
IOOO (t>
.,..
JG<-§ W /t/ j/ ,~o't,j;; I //;V ~/'/'. /
VJ/'?,;~): _,"fr/ /~ /~/~ / (t> 
VY l~f' 171 v §/[;tV _/ ~~~ 
UNIT l/E:FORMATION'.:i JN ,0001 
FiS', 18, Stress-strfl,in Curvei;; for Repeated Loads, Gravel Concrete, Series I, 
~ 
II'­
*" 
6000 
G5000 
r_ 
"-'
~ 

c:::: 
;:5 
::::> 
~·
0 4000 
ii;
VJ 
.... 
cc 
,,...
"'
'""· 
<.::
~ 
<::>
.....,, 
C2 3000 
'"3
r. 
::i 
"'R
0 
"'­
~ 
?::. 
t'1 
~ 2000 I 
~ 
"',,...
c3 
_J 
~­
t:: 
r_ 
::::> 
UNIT DffORMATION'.':> IN .0001 
Fig. 19. Stress-strain Cu rves for Repeated Loads, Broken Stone Concrete, ·series I. 
Physical Properties of Dense· Concrete 
TABLE 16 
MODULUS OF ELAS'l'IOITY OF BROKEN S'.rONE CONCRE'l'E, SERIES (Each Hesult the Average of Tests of 'l'wo Specimens) 
Sacks of cement 
per cu. yd. of 
concrete 
1. 63 
3.75 
5.11 
7.63 
9.45 
11.3 Age 3 m onths 
Modulus of elasticity in pounds per 
square inch 
1,500,000 4,000,000 5,500,000 5,500 ,000 5,600 ,000 5,100,0CO 
Ultimate unit 
compressive stress in pounds 
per square inch 
602 2215 4930 5690 6515 6510 
Age 1 year 
Ultimate unit
Modulus of 
elasticity in 
compressive
stress in pounds 
pounds per 
per square inch 
square Jnch 
*l ,000,000 
590 
3,600,000 
2345 
4965
4,900,000 
5,200,000 
6040 
*4,900,000 
6365 
6970
5,400,000 
*'I1hrce specimens. 
TABLE 17 
MODULUS OF l~LAS'l'ICI'l'Y OF BROKEN STONE CONCRETE, SERIES III (Each Result the Average of 'l'ests of 'l'hree Specimens Except as Specified) Age 'l'hree Months 
Sacks of cement per cu . ycl. of concrete 
*1.61 
3.63 
5.64 
7.::s 
9.02 
10.53 
*'l\wo specimens. 
Modulus of E lasticity in 
pounds per square inch 
800,000 3,000,000 4,700,000 4,6)(},0Gi'I 4,700,000 4,500,000 
TENSION TESTS 
Ultimate unit compressive 
stress in pounds per sq. in. 
412 1500 3190 4125 5050 4775 
Scope of the ·investigation. For the tests of Series I, one hun­dred and eight specimens of standard consistency we:re made of such shape that they could be broken in tension. One half of these were of gravel concrete, the other half of broken storn:~. ·For each d1fferent percentage of cement, there wer e, nine speei­mens. Three of each kind were broken at ages of l 2 days, 28 days, and 3 months. In addition, nine specimens of"six sack" gr.av el concrete, the ''wet'' mixture, were made and tested. 
For the t ests of Se1ries II, eighteen specimens were made of the gravel concrete, there being three specimens for each differ­ent percentage of cement. One of -each kind was broken at ages of 12 days, 28 days, .and 3 months. In addition, six specimens of ''six sack'' gravel concrete, three of the ''wet'' mixture and three of the ''viery wet'' mixture, were made and tested. 
For the tests of Seri<es III, thirty six specimens were made of 
University of Texas Bulletin 
the gravel concrete and fifty four of the broken stone concrete. For each different percentage of cement, the.re were six speci­mens of gravel and nine specimens of broken stone concrete, one third of which were testEd at the age of 12 days, one third at the :age of 28 days, and the remainder after 3 months. 
Moulds. For making the tension specimens, special moulds were made similar in general outline to those used for making standard cement briquettes, but large enough to give -a breaking 

0 " 0 
---5~ -l 
~ 

0 " 0 
Fig. 20. Mould for Tension Specimen. 
section of twenty :five square inches. Fig. 20 shows the dimen­sions of these moulds. They were made of cast iron and finished smooth inside and on top and bottom edges. The pins shown in sketch enab1e one to quickly fit· the1 two parts together, while the bolts hold them in place. ,
0 
Mixing, placing, etc. The moulds, well oiled, were placed on greased galvanized iron plates. The concreta for each percent­age of cement was mixed as already described and carefully tamped into place in layers of about two inches. Afte;r one day in air, the1 moulds were removed and the specimens were stored until broken as in the previous tests. 
Testing. The specimens of Series I, broken at each period, were selected so that part were take-n from the first batch mixed and part from the second batch. In Series II and III, one speci­men from each batch mixed was brofoen at the end of each pe-riod. 
Physioal Prop'3rties of Dense Concrete 
'l'he 100,000# Olsen machine and special holders were used to break the specimens. 
The apparatus used in making the tension txost is shown in 
Fig. 21. Apparatus for Testing Tensile Specimens. 
Fig. 21. This differed somewi1a:t from that used in some of the earlier tests but worked on the same gmeral principle. It was made up of two holders so constructed as to grasp the enlarged parts of the briquettes. One was fastened to the fixed cross-head of the machine -and the other to the movable cross-head. The holder ·consisted of a roller about 1 %" in -diameter and 6" long, from which was suspended by two eye bolts the grip used to clamp the specimen. This grip was made -0f a cylindrical piece of steel 1%" in diameter, from each end -of which were suspooded two flat steel hangers with bolts and spacing bars as shown in figure. These were free to turn in one plane about the points of support and could be spread to a posi­tion such that, together with the eye bolts, they took a shape similar to that of the letter "Y" (inverted for upper cross­head). Through the extreme ends of these flat pieces constitut­ing the legs of the "Y" and parallel to the longitudinal axis of the cylinder to which they were fastened, passed two %" bolts, <me on >each side, connecting the four hangers in sets of two anci 
University of Texas Bulletin 
serving to transmit the pull to the specimen to be tested. These two bolts were held the proper distance apart by two spacing bars. When the test was to be made, the roller was placed on the fixed cross-head (for upper holder) and connected by the eye bolts, which were passed through the opening in the cross­head, to the cylindrical piece from which the hangers were sus­pended. This cylinder was set so that its axis was at right angles to the axis of the roller. This arrangement, together with the connection made by the two eye bolts, permitted the appar­atus to adjust itself to any inaccuracy in centering, etc. 
Each specimen was carefully centered and the load applied 
continuously by motor causing the cross-head to move down at 'the rate of about .05 of an inch pe<r minute. This was equiva­lent to applying the load at the rate of 40 pounds to 48 pounds per square inch per minute. The specimens broke at or very near the center, giving a breaking area equal to or but little larger than that of the least section. Tests were made at the 
end of 12 days, 28 days, and 3 months. 
Results. TablES 18 to 20 give the results of the tests on the 
gravel concrete, while Tables 21 and 22 give those for the broken 
stone concrete. The values given in these tables are represented 
graphically in the curves of Figs. 22 to 27. The method of plot­
ting these curves and determining the average curve for each 
aggregate at a given 'age was the sa~ as that used in plotting 
the results of the compressive tests. 
Particular attention is called to the values indicated by ·~ 
of Figs. 22 to 24, and by ~ and o of these figures. The 
first shows the results of the "six sack" "wet" mix of Series 
I, the second, those of the ''wet'' mix of Series II, and the third, 
those of the ''very wet'' mix of this series. 
The curves of Fig. 28 show average values for Series I for the 
different ages. These are based on the data given in Table8 18: 
and 21. 
:r: 
u 
~ 
tu 
~ 
::) 
0 
Vl 
et: 
lU 
p._ 
VJ 
D 

r:. 
::) 
0 
p._ 
~ 
:r: 
I­
<..'.) 
r_ 
w 
Cr: 
t:i 
w 
...J ;;:) 
r_ 
IU 
I­
i"ooo 
900 800 700 600 500 4 00 300 200 
\00 
0 

"'ti
;:::.­
«:! 
"' 
[ 
~ 
ci 
~ 
"'l
.... 
~· 
"' 
c 
........ "" t;:, 
~ 
"" 
c 
c 
;:l 
~ 
"'l 
....
"" 
"" 
~ 
1000 
l: 
u 
r. 
900 
lU 
~ 
0 800 VI 
(\:'. 
~ 700 
~ 
5 600 
E: 
z:: 500 
r t; 400 
r_ 
UJ 
er: 
f-::-300
\fl 
UJ _J 200 
;/) 
r_ 
UJ I-100 
0 

&i 
~ 
;· 
~ 
.,,,;
... 

<-<­
~ 
<:>
-'"'3 
"" 
""
~ 
ti:: 
~ 
.,....
.,.... 
.,..,.

""... 
;:> 
IOOOJ 
I 
CJ 
900) 
~ 
l.J 
er 
I
800
<t 
:J 
0 
.fl 
700 

?2·
n:: 
.,....
l.J 
Cl. 
"tl 
600 
If\ 
D 
~ 
z 
"i
.,....
:J 
0 
)
500
0.. 
"' ~­
<::> 
~ 
400I 
-t:::i
I 
I­
;;> 
"" 
z 
" 
"' 
UJ 
"" 
300I
er 
C':: 
c 
;;>
lri 
\:'; UJ 200 
~ 
.,....
_J 
V'i 
"" 
z 
li...J IOO 
f­
0 
'l 
10 
12 , 
C)1
SACKS OF CE:ME:NT PE:R CUBIC YARD Or CONCRETE 
I-' 
Fig. 24. Ultimate Tensile Stren gth of Gravel Concrete, Age 3 Months. 
1000 
:r: 900 
u 
~ 
w llOO
cr: 
<( 
:::i 
~ 700 
er: 
lU 
D.. 600 
if) 
D 
z: 
:::i 500
i2 
?:. 
-400 
:r:: 
1-­
\.'.} 
6 300 
Cc: 
1-­
<f) 
200 
ILJ 
_J 
if) 
z: 
100 1-­
IW 
0 

C!l 
N 

~ 
~· 
"" 
..."' ~ 
.,... 
«::! <:>
-... 
"'3 
1:'l
"" 
~ 
~ 
~ 
..... 
.,...
"" 
~· 
IOOO 
r 
u 
~ 900 
w 
er: 
<( 
BOO
:J 
0 
tf) 
cc 700 
LU 
CL 
V} 
bOO
D 
r. 
:::i 
2 500 ~ :r: 400 
1­
CJ 
r. 
LU 
300
IX: 
ti) 
!W 200 _j 
if) 
z: 
100
1W 
1­
0 

:;: 

"' 
[ 
'"ti 
~ 
~ 
"'I 
"' 
<-+­
-
"' ~· <:> 
~ 
"' 
i; 
"' 
a 
§ <:> ~ 
<-+­
"' 
o:> 

Cil
1000 
.... 
r 
u 900 
z 
~ 800 
<1: 
~ 
0 V\ 700 
er: 
~ 
;·
iW 
Ci.. 
600 
"" ~ 
V) 
.,>.
.,.._
0 
z 
""
::i 500 
~
6'. 
'""'3 
z 
..,
""' 
400 
z;
I 
l;;lj
1­
~
\'.J 
.,...,
.,...,
z 300 
w 
.,.._
er 
""' 
~
.,.. ·
t;; 
200 
w 
_J 
;:;) 
i5 100 
I­
0 0 2 3 4 5 6 7 () 9 10 II 12 SACKS OF CEMENT PER CUBIC YARD OF CONCRE:.TE 
Fil?. 27. Ultimate Tensile 'Strength of Broken Stone Concrete, Age 3 Months. 

r 
u 
~ 500 
uJ 
()::'. 
§ 
0 
<[) 
()::'. 400 UJ 
D­
~ 
r_ 
::> 
~ 300 
r_ 
:r:: 
r­
c:i 
r_ UJ 1.00 
()::'. 
~ 
':] Ji 
r_ 
w 100 
r­
,.,,, 
~ 
"' 
b'
.,,
.,... 
,.,,, 
ci 
~ 
"'l
.,....
.... 
"' 
"' 
c 
....... 

"' ~ "" ~ 
Q 
~ 
5 
"'l 
"".,.. 
"' 
<:.l1 
<:,Tl 
University of Texas Bitlletin 
'l.'ABLE 18 
'l'E:\'SILE S'l'REJ\GTJ:I OF GRAVEL 00:\'0RE'l'E, SERIES 

(Each Result the Average of 'l'hrco Specimens 5"x5") 
------. ------------------------­
Ultimate unit stress in pounds per square incb 
Racks of cement per cu . yd . of concrete Age 12 days 
J. 66 
39 t54 
3.80 
193 236 
I 
247 275 
*5.8"2 
296 33'25.93 
8.45 
330 335 
9.67 
324 349 
11.50 32-3 316 11 
•The "wet" mix. 
1Two specimens. 
'l.'ABLE 19 
TE:\"SILE S'l'RE:\'G'l'J:I OF GRAVEL '00:\'0RE'l'E, SERIES II 

(Each Result from '11est of One St)ecimen 5"x5") 
--------·-----------·-----­
Ultimate unit stress in pounds per square inch Sack s of cement per 
cu . :rd. of concrete 
1.66 
3.80 
*5.54­
t5. (5.82) 

5.93 (5.96) 
S.45 
9.67 
11 .55 (11 .50) 
Age 12 days  Age 28 days  Age 3 montbs  
48 212 176 200 312 371 ·121 372  80 332 231 324 378 357 404 304  116 358 273 387 484 386 46~ 320  

.i,rl'hc "very wet" mfxture. 
tThc "wet" mixture. 

'l'ABLE 20 
'fEJ\SILE STRENGTH OF GRAVEJ, 00:\'0RE'l'E, SERIES III 
(Each Result the Average of 'Tests of ,.l'wo Specimens 5"x5'') 
Age 3 montbs 
71 
299 317 432 416 414 429 
Ultimate unit stress in pounds per square inch 
Sacks of cement per cu . yd. of concrete 
1.64 
3.76 
5. 
8 .20 9.18 10.81\ 
Age 12 days 
38 173 246 
287 
256 275 
Age 28 days 
70 253 308 281 238 281 
Age 3 montbs 
72 2H 350 366 335 294 
'l'ABLE 21 
TE:\'S[LE S'l'RENG'l'J:I OF BROKEN STO:s'E OONORE'l'E, SERIES 
(Each Resu lt th e A ,·crngc of 'l'ests of 'l'hree Specimens f/'xfl'J 
Ul timate unit stress in pounds per square inch 
Sacks of cement per 
cu . ycl. of concrete 
1.63 
3.75 
5.90 
7.63 
9.45 
11.30 Age 12 days Age 28 days 
38 171 
277 
352 
339 
364 
~-10 
211 
254 346 2ll4 
363 
Age 3 months 
70 265 406 463 444 388 
*Two specimens. 
Physioal Properties of Dense Concrete 
T ABLE 22 'l'E~SILE S'l'RENG'l'H OF BROKEN S'l'ONE OONORE'l'E, SERIES III 
(Each Result the Average of 'l'ests of '.rhree Specimens 5"x5") 
Sacks of cement per 
cu . Yd . of concrete 
1.61 
3 .63 

5 .64 


7 .3cl 9.Cf2 
10 .53 
Ultimate unit stress in pounds per square incb 
Age 12 days 
36 116 223 294 336 337 
Age 28 days 
48 
164 260 28'2 282 285 
Age 3 montbs 
62 184 345 374 372 317 
TRANSVERSE TESTS 
Scope of the investigation. For the tests of Series II (no transverse tests were made in Series I), eighteen specimens were made of the gravel concrete, theire being three specimens for each different percentage of cement. ' One of each kind was broken at ages of 12 days, 28 days, and 3 months. In addition, six specimens of "six sack" gravel concrete., three of the "wet" mixture and three of the "very wet" mixture, were made and tested. 
For the tests of Series III, thirty six spwimens were made of the gravel con&ete and fifty four of the broken stone concrete. For each different percentage of cemE·nt, there were six speci­mens of gravel and nine of broken stone, concrete, one third of which were tested at the age of 12 days, one third at the age of 28 days, and the remainder after 3 months. 
Monlds. For making the be.ams, moulds which had been used in some former tests were adapted to the! size required for this experiment. These were l'ectangular in cross-section, about six feet long, and were made of two-inch plnllEd lumber. Rest­ing on the bottom, which consisted of a single board 10 inches wide, >v·ere the side pieces, 8 inches high, which ·were held six inches apart by two end pieces. These were kept in position by cleats nailed to the sides. The various pa.rts were held to­gether and the sides pre!vented from spreading by yokes drawn up tightly about the mould with long bolts. On the removal of the bolts, the side and end pieces could l'eadily be removed. As it was desired to have speocimens of 6"x6"x3' -0", it ·was neces­sary to place on the bottom of the mould a piece two inches 
University of Texas Bulletin 
thick and to insert at th21 middle a galv;anized iron plate which divided the mould so that two beams three fee:t long could be made. 
Jrlixing, placing, etc. The concrete for each percentage of cem2nt was mixed as already described and placed in the moulds in layers of about two inches and thoroughly tamped. The specimens were smoothed off on top iand the next day they were carefully ranoved from the moulds and stol'ed as in the pre­ceding tests until the time of breaking. 
Testing. One specimen for each batch mixed was broken at the end of each period. 
'l'he beams wi:,re tested on an Olsen three-screw 100,000# uni­versal testing machine. The method of applying the load, to­gether with the essential dimensions, is shown in the sketch, Fig. 
29. An 8" I beam was first laid flat 9n the weighing table of the 

1------------->•"-----------j 
Fig. 2 9. Sketch 'Showing Method of Testing Concrete Beams. 
machine. On this were placed the two supports consisting of wood blocks, turned to a radius of about three inches on the under side ·with a semi-circular groove on the top side, in which rested a 1 114" iron roller. These supports were spaced the proper distance apart and the beam placed directly on the: rol" lers, no plates being inserted between the two. On the top side were usEd two blocks with rollers similar to those used on the under side, except that the blocks were inverted as can be seen on the drawing. By means of 1a short wooden beam under which the1 specimien was carefully centered by using the plumb 
Physioal Prop.erties of Dense Concrete 
bob shown in the sketch, the load was transferred from the movable cross-head to the two rollers on the top of the concrete beam. The above arrangement pe•rmitted of an even bearing at both load and support points, avoiding the bad effects of any twist or inequality in the beam to be tested. During the test, the machine was operated slowly and continuously by motor up to rupture. 
Results. Table 23 gives the results of the tests of Series II, the gravel concrete. Table 24 gives the results of the tests of the gravel concrete of Series III, while Table 25 gives those of the broken stone concrete• of this series. These values were deter­mined by calculating the moment at the center due to the weight of the beam and the ultimate load and applying the formula for flexure, Mc=SI. 
Figs. 30 to 32 show the curvrn for the gravel concrrte and Figs. 33 to 35 give those for the broken stone concrete. On these curve she.ets will be found a point for the test of each beam. The light curves were determined by joining in order the points showing the results of individual tests for each series, while the heavy curves were obtained as already described for the com­pressive tests and represent average values of all series for a concrete of a given coarse aggregate and of a certain age. 
Particular attention is called to the results of the tests Oll the "wet" mixture and the "very wet" mixture of Series II. These are shown in Figs. 30 to 32. 
'l'ABLE 23 
MODULUS OF RUPTURE OF GRAVEL OONORE'l'E, SERJES II 
(Each llilsult from Test of One Beam) 

Sncks of cement per 
cu . yd. of concrete 
l.66 
3 .SO 
*5 .54 
t5.8& (5 .82) 
5.93 (5.96) 
8. 45 
9 .6'i 
11.55 (11.50) 
Modulus of rupture in pounds per square inch 
Age 3 montbs
Age 12 days Age 28 days 
-----------1--------1------­95 132 296 436 
547 366 401 
464 464 563 
690 519 435 
602 544 520 
470 
5q()
fi51 .-:.,27 
011 
581 
*The "verv wet" mi:'\. tThe "wet" 1n ix. 
:i:: 
u 
~ 
w 
r£ 
<( 
:::> 
a 
Vl 
cc 
w 
C>.­
~ 
r. 
::i 
~ 
r:. 
t...J 
~ 
::i I­
D.. 
::i 
cc 
li­
0 
If) 
::i 
_J 
::i D 
0 
L: 
1000 
900 000 700 600 500 400 300 200 100 0 

CS> 
c::j ;> 
~· 
~ 
~ ~ 
<::: 
~ ~ 
~ 
!-1 
~ 
ti;;; 
;;:: 
~ 
~ 
.,.... 
~­

"'::;
;::­
~ 
"' ~· 
i::.. 
'}j 
~ 
~ 
.,.;.. 
"'· 
-"' 
"' 
c 

"' ;z ~ 
"' 
"' 
Q 
c 
~ 
(":) 
"i 
.,....
"' 
"' 
O> 
I-' 
0 2 
Fig. 
1000 
I 
u 
900
~ 
w 
cr
:§ 800 Cf 
<fl 
700
er 
L.J 
Q. 
8 600 
z 
::> 
0 
Cl. 
500 
~ 
L.J 400 cr 
::> 
f-­
~ "00 
ll::: 
i... 0 200 
Ul ::> 
_J 
::> /00 
Q 
0 
l: 

CJ':> 
N) 
C::::! 
;2 
~­
"' 
~ 
.... 
..... 
~ 
c
-.., 
8 
"' ~ 
\;:> 
"" 
O::I 
~ 
..... 
"' 
~­
1000 
:c 
v ~ 900 
L..J
ex: 
~ 800 ~ 
cc 
w 700 CL 
~ 
z: 600 '.) 
0 
CL 500
z: 
UJ ~ 400 
f-
D... 
::J 
ex: 300 u... 
0 
if) 200 ::J 
_J 
:::i
§ 100 
2: 
0 

"ti 
;:s"' 
<.: 
....
"' 
2 
~ 
"ti ci 
~ 
..,
"' 
.,... 
"' ~· 
·~ 
b 
~ 
;;; 
"' 
\'.".) 
<::> 
;Z C>
.., 
"'.... 
"' 
2 3 4 5 6 7 5 9 10 II cro 0 -;,: 
1000 
I 
u 
~ 900 
w 
cr
:§ 800 0 
if) 
700 w 
er 
Cl.. 
8 600 
z 
::::> 
0 Cl. 500 
~ 
w 400 
():'. 
::::> I­
~ 300 
cr 
!.... 
0 200 
if) 
::::> 
_J 
100
::::> 
a 
0 
L 

CS> 
N> 
c:::j 
;l
""·
<:: 
<.'> 
~ 
,,...
""· 
<:;;: 
<::>
-., 
""3 
<.'> 1:-l .::. 
"' 
~ 
~ 
<.'>
,,... 
~­
1000 :r: 
I..) 
~ 900 
t...J 
~ 
~ 800 
cc 
w 700 
0... 
~ 
r. 600 :::> 
0 
0... 500
r. 
UJ ~ 400 
f­
(L 
:J 
p::: 300 
lL 
0 <{) 200 :::i 
__J 
:::i

g 100 
L 

"'1j ;:so <..: 
"'
"'· 
2 
"""' 
"'1j 
.... 
C> 
~ 
;l. 
«>"'· 
"' 
C>
....... 

b 
~ 
«> 
"' 
(°'::) 
C> ;::! 
~ 
'"' 
'"' ~ 
~ 
°' 
IOOO 
:r: 
900 
\J 
~ 
"-1 
~ eoo 
:::i 
a 
Vl 
700 
Ct: 
t.J 
0... 
If) 600 
C> 
~ 
:::i 
2 500 
~ 
~ 400 
:::::> 
I-
Cl­::J 300 Cl'.: 
"-­
0 
200 
U) :::i 
_J :::i 
8 100 
L: 

ii:­
°" 
~ 
<;· 
"" 
~ 
""·
<-+­
<:<::: 
~ 
"".3 
"" ~ "' ~ 
~ ~ 
...... 
<-+­
"" 
~­
IOOO 
:r 
u 
~ 900 
LJ Ck'. ·<{ 
:::> 800 
c 
oil 
e:; 700 
a_ 
:::> 
~ 500 z 
LJ 400 Ck'. 
::::>
I-. ~ 300 
er 
i.... 0 200 
:::>
"' 
6_j . 100 0 
:L 
0 0 2 3 4 '.:> 6 7 5 9 JO II SACKS OF C[M[NT PE'R CUBIC YARD OF CONCRETE 

Fig. 35. Modulus of Rupture of Broken Stone Concrete, Age 3 Months. 
'i:j
:;:.­
<:::: 
"'
()· 
~ 
'ti
..., 
c 
~ 
e.;.
"'· 
"' 
"" 
c 
'-..,., 
b 
"' 
;:l "' 
"' 
Q <:' 
;:l 
..,
"' 
.,.,..
"' 
"' 
CJ• 
°' 
University of Texas B idletin 
'.PABLE 24 
MODULUS OF lWP'l'URE OP GR.AVEL CONCRE'.PE, SERIES III 
(Each Result the Average from '.rests of '.rwo Beams) 

,, ____ _ odulus ____
Mof rupture in pounds per square inc.h 
Sacks of cement per cu. yd. of concrete Age 28 daysAge 12 days
___ ____,____ _ 
*96
*70
1.64 
528301 
3.76 
3864605.88 
449
4~3
8.20 
404
5209.18 
370492 
10.86 
Age 3 months 
*151 
430 556 562 *510 434 
*One specimen . 
TABLE 25 
MODUT, US OF RUP'l'URE OF BROKE:'.-! STO::-i'E CONCRETE, SERIES III 
(Each Result the Average from 'l'ests of Three Beams) 

Modulus of rupture in pounds per square inch 
Sacks of cement per 
cu . . yd. of concrete 
1.61 
3.63 
5.64 
7.38 
9.02 
10.53 
Age 12 days 
*274 
400 
007 
598 
586 
Age 28 days 
81 252 
478 t484 
515 547 
·---­
Age 3 months 
*71 328 531 
563 
610 
545 
*One specimen . 
tTwo specimens. 

BOND TESTS 
Scope of the investigation. For the tests of Series I , seventy two specimens of standard consistency were made, each a con­crete cylinder about six inches in d~ameter and seven inches long, embedded in which was a %" round steel rod. One half of these were of broken stone1 concrete, the other half of gravel. For each different percentage nf cement, there were six speci­mens. Three of each kind were broken at ages of 12 days and 28 days to obtain the unit bond stress at first slip of steel, taken at a slip of .0002 of an inch, at a slip of .001 of an inch, and at failure. In addition, six . specimens of ''six sack'' gravel con­crete, the ''wet'' mixture, were ·made and tested as outlined above. 
For the tests of Series II, eighteen specimens were made of the gravel concrete, there being three specimens for each dif­ferent percentage of cement. One of each kind was broken iat ages of 12 days, 28 days, and 3 months. In addition, six speci­
Physioal Pn1perties of Dense Concrete 
mens of "six sack" gravel concrete, three of the "wet" mixture 'and three of the "very wet" mixture, were made1 and tested. 
For the tests of Series III, thirty six specimens were made of the gravel concrete and fifty four of the broken stone concrete. For each dilferent percentage of cement, there were six speci­mens of gravel and nine of broken stone: concrete, one third of which were tested •at the age of 12 days, one third at the age of 28 days, and the remainder after 3 months. 
Moulds. For making the bond tests, some1 effiilltY tin cans about six inches in diameter and seven inches long, were used as moulds. The to·p of the can was removed and a hole made in the centEir of the bottom through which could be passed the rod to be tested. This can was set on a wooden frame holding a l"x8" laid flat and containing holes slightly larger than the diameter of the rods. About fourteen inches beneath this piEce was a second l"x8" parallel to the first and with holes vertically below those in the top piece. Immediately beneath was a sup­port for the rods. Through the opening in the bottom of the can and the two holes in the boards, the rod was passed and in this way held in position during the placing of the concrete. The rod was allowed to project above the mould about two inches. 
Steel. The steel was %" round stock material, free from rust but not polished or rubbed smooth. 
Mixing, placing, etc. The concrete for each percentage of 
cement was mixed as already described and placed in the moulds 
in layers of two to three inches, being thoroughly tamped. 
After setting until the next day, the specimens were placed 
under water, without having the tin cover removed, and stored 
as already described until the time for breaking. 
Testing. The specimens of Series I broken at the end of each 
period were sclected so that part were ti;iken from the first batch 
mixed and part from the second batch. In Series II and III, 
one specimen from each batch mixed was broken at the end of 
each period. Several hours before breaking, that part of the 
specimen against which the force would be exerted in the trnt 
was covered with a thin layer of plaster of Paris on the outside 
of the, can to give a smooth bearing. 
An Olsen 100,000 pound universal machine was used to make 
the tests. In order that the load at first slip and the corres­
University of T exas Bulle.t'in 
ponding slips for subsequent loads might be obtained, a special attachm,ent was required to measure the movement of that part 
Fig. 36. Sketch 	Showing Apparatus Attached to Bond Specimen for Determining the 'Slip of Bar. 
of the rod projecting through the top of the concrete. This ap­paratus (See Fig. 36) consisted of two steel uprights fastened to the, specimen by four thumb-screws and kept from spreading at the lower ends by a circular band enclosing the can and at the top ends by a bar passing about 6 or 7 inches over the specimen. To this bar was attached the dial used in the compressive tests. A fine >vire passed .around the cylinder connected to this dial and was fastened through a clamp to the end of the rod. By means of this apparatus, the movement of the bar could be measured to 115,000 of an inch. 
In the tests of Series I, the specimen was placed on a steel plate with a leather. pad on top of the fixed cross-head of the machine with the rod sticking· down through the opening. After carefully cente,ring the specimen, the rod was connected to the movahle cross-head with the wedges. In the tests of SeriES II and III, two plates with holes, through which the rods would p.ass, were used. One was placed on top of the fixed cross-head, resting on it, and one was placed on the under side of the speci­men to be tested. The projecting rod was gripped by the mova­
Physioal Properties of Dense Concrete 
ble cross-head which was moved down until there was not more than about 1/s of an inch between the plate on the fixed cross­head and the other one when held tightly against the specimen. Between these two plates, four small wedges were then inserted and lightly driven into place so that the specimen had an even bearing. 
The specimen being properly placed, the dial and frame were attached and the zero reading of the pointer taken. The load was then slowly applied, the cross-head moving down at the rate of about .05 of an inch per minute. One observer kept the beam balanced and took the load while a second took slip read­ings on the dial. These observations were continued until the highest load vvas obtained. 
In all of the 28-day tests and some of the 12-day tests of Series I, the bars were completely pulled out of the specimm resisting the greatest pull and the one standing the; least pull in each set of speicimens, and the cylinders of concrete remaining were then tested in conpression after removing the tin cans. 
Resitlts. Tables 26 to 28 show the unit bond stresses for gravel concrete at first slip, or slip of .0002 of an inch, at slip of .001 of an inch, and at failure. The total loads by which the first set of values was obtained were read off the beam of the testing machine when the wire wound dial showed a movement of the bar equal to .0002 of an inch. The1 intermediate values are based on the load-slip curves for the different specimens. Tables 29 and 30 show the results for the broken stone concrete determined as just outlined for the gravel concrete. 
The curves for the 12-day and 28-day results are shown in Figs. 37 and 38 for the gravel concrete, which in Figs. 39 and 40 are shown the corresponding curves for the broken stone con­crete. Particular attention is cal~ed to the values indicated by 
• 
Figs. 37 and 38 and by ~ and o of these figures. The first shows the results of the ''six sack'' ''wet'' mix of Series I, the second those of the ''wet'' mix of Series II, and the third those of the "very wet." mix of th.is series The method of plotting these curves and of determining t'he av­erage curve for eHch kind of concrrte at a given age is the same as that already described under the compress.ion tests. 

-.:i 
~ 
;:;· 
"" 

""l 
;:;:
"' 
<.:: 
<;:,
-""3 
~ "'"' ~ 
E: 
<-!­
°" 
~· 
12 Days, at Slip of Steel of .0002 and .001 of an Inch. 
:r: 
v 
r:. 
w 
~ 
::i 
$'{ 
CL 
tr 
II"\ 
p.
z:.
6 . 500 
i:... 
z:. 
;;:; 400 
If) 
L...I 
~ 300 
C>. 
L.
2 200 
•) I-§ 
1000 
goo 
800 
700 
600 
100 0 
1000 
r: u 900 
~ 
tJ 
er:: 800 
4: 
:J 
a 
<() 
700 
(l:: 
tJ 
Cl.­
600
<() 
0 
z 
:::> 
0 500 
0.. 
z 
If) 400 
"1 

tJ 
I>'.
lri 300 
0 
z 
as 200 
t: 
z 100 

:::> 
0 I 2 3 4 5 6 7 b q 10 I I 12. SACKS OF CEMENT PER CUBIC '<ARO Or CONCRE.TE. 

~ 
~ 
"' 
i 
"'c; 
~ 
.... 
<S. 
"' ~­
0
-
t::I "" 
~ 
"" 
Q
0
;z 
n 
"' ~
-

"' 
-:i 
....... 


IOOO :i:: 
v 
r. 
-900 
t.J 
~ 
::::> 800 
a 
VI 
!:j 700 
().._ 
~ 600 
-r_ 
::i 
i!2 500 
r_ 
{) 400 
{) 
l.J 
flC
ln 300 
D6 200 IQ 
I­
100 
z: 
:::> 0 
1:-:l 
c:::: 
E­
~ ~­
<:::: 
;::: 
~ 
"'-3 
;;..... 
"" 
;:; 
"' 
Oj .,..... 
"" ~ 
~· 
IOOO 
:r:
i! 900 
LU 
~ 

"ti
:::> 
;::.­
CJ 
<:::: 
if\ 
...
"' 
{r. 700 
g
....
~ 
"ti
..,
If\ 600 
C>
0. 
~
r. 
.., 
;t:
6 soo I I 
P­
"' 
-
"' 
C>

r. 
400 I 
\:::1
If\ 
If\ 
"' ;::;:
!LI 
tx:: 
"' 
t-300 I 
"' 
cf) 
<:":) 
2 
c::;
...,
~ 
""' 
<-+. 
"' 
in 
"' 
~ 
::> 
3 4 5 6 TB 9 10 II 5ACK::i Of CEMENT PER CUBIC YARD Or CONCRETE -"!
c: 
Fig. 40. Bend Stress for Broken Stc·n~ Concrete, Age 28 Days, at Slip of Steel of .0 002 and .001 cf an Inch. 
Univei·sity of Texas Bulletin 
The results of the· compressive tests of Series I made -on the bond specimens from which the rods had been pulled are not given here; however, it can be said in general for any mix that that specimen testing highest in bond also showed higher unit compressive strength than the one, testing lowest in bond. This however was not always the case. Furthermore, the average unit crushing strengths obtained by compressing the; concrete used in the bond tests were much lower than those determined from the special compression tests of similar mixes, the ratio running from 6110 to 9/10. 
TABLE 26 
BO:\D STRENGTH OF GRAVEL 00:\0HETE, SERIES I 
(Each Result the Average of 'l'ests of ,.l'hree Specimens. Diameter of Bar Was %"" 
with Depth of J-:mbedment of 7" . Plain Round Rods Used) 
Sacks of 
cement per 
cu. yd. of 
concrete 
1.66 
3.80 
*5 .82 
5.96 
8 .45 
9 .67 
11.50 
Unit bond stress in 
Unit bond stress in 
Ultimate unit bond 
pounds per square inch 
pounds per square inch 
stress in pounds 
at slip of .000-2 of an inch 
at slip of .001 of au inch 
per square inch 
-_ _ ______ ___,______ _ _ 
Age 12 days \Age 28 days 
..\ge 12 dayslAge 28 days 
Age 12 days\Age 28 days 
--t3f~ ---:---~~r---350 ---tm 1·---:­
200 335 
295 380 430 445 
340 330 
425 430 §725 l650 
320 34() 
400 400 =I =1505
=I = 
=I ~ ~765 
----··· 
* " Wet" mixtnre. 
trrwo specimens. 
tElastic limit of steel passed in case of one specimen. 
§E lastic Jimit of steel passed in case of two specimens. 
ifElastic limit of steel passed in case of three specimens. 

TABJ, E 27 
DO:\D S'l'REl\G'l'H OF GRAVEL 0 0:\0RE'l'E, SEl~IES II 
(Each Hesult Determined from the 'l'est of One Specimen. 'l'he Diameter of tbe Bar 
Was %" with Deptb of Embectmcnt of 7" . Plain Round Rods Were Used) 
Sacks of 
cement per cu. yd. of concrete 
l.66 
3.80 t5.54 *5.86 (5 .82) 
5. fl3 (5 . 96) 
8 .45 
9.67 Jl. 55(1J.50) 
*rl'hc "wet" 
Unit bond stress in 
Unit bond stress in 
pounds per square 
pounds per square 
Ultimate unit bond 
inch at sli p of 
inch at s1ip of 
stress in pounds
.0002 of an inch 
.001 of ao inch 
per square inch 

45 340 400 
355 
370 
560 380 
mixture. 
trl'hc "very wet" mixture. !Elastic limit of steel vass 1ci 
60 
145 
70 
150 
145 
75 
485 
455 
345 
490 
415 
520 
655 
535 
435 
660 
l680 
460 
680 
410 
5:30 
510 
535 
615 
510 
650 
580 
!735 
365 
510 
[)65
400 
400 
1690 
170 
170 500 
490 
555 
!730 !765 
!&'30 
~71 5 -~ 71 5 
!715 !754 
1690 
t690 
!600 
J>hysical Properties of Dense Concrete 
rrABJ....E 28 
BOND S'l'RENG'l'H OF GRAVEL CO="CRE'l'E, SERIES III 
(Each Result the Avelage of 'l'ests of Two Specimens Except as Noted. The Diameter 
of the J3ar Was %" with Depth of Embcdment of 7" . Plain Round Rods Wer~ Used) 
Sacks of cement per 
cucoi~·~ et~f 
l .OJ 
sro 
;) 88 
8 .20 
9.18 
10.86 
~mt bo~cl strnss 1n Uo1t bond stress in 
I 
pounds l)Ct square l)Ounds per square Ultimate unit bond 
mch at slip of 111ch at slip of stress in pounds

!Ag:002 ~g:o m~:ge .. A~~lf o:;: _me~ _A;:er sq:::e 1ne~ge 1 ~~~28 cl_:~ 3 mo._ 12 days12s da~2-.'._1:_':.:._ 12 days,28 days~ 
70 120 235 80 1 130 2•15 120 165 285 
a ~ ~ ~ m ~ ~~ ~~5 1 *660 440 5:J5 !740 I t750 425 1 520 530 450 560 620 1704 !720 H 5 580 *565 450 I 640 *620 580 !710 t680 
1 
•125 *475 460 *535 615 !730 t680 
--·-----· -··---------------------------------···­
*Result of test of one specimen. 
tElastic limit of steel passed in case of one specimen. 
!Elastic limit of steel passed in case of two specimens. 
TABLE 29 
BOND S'l'RE:s'G'l'H OF BROKE=" S'l'Ol\E OONOHE'l'E, SERIES I 
(Each Result the Average of Tests of Three Specimens. Diameter of Bar was %" with Depth of Embcdment of 7". Plain Round Rods Used) 
Sacks of 
i 
Unit bond stress in J Unit bond stress in Ultimate unit bond 
pounds per square inch pounds per squ are inch stress in pounds
cement per 
cu. yd. of 
Ht slip of .0002 of an inch at s lip of .001 of an inch per square inch conc rete 
___________;__________,_
_________ 
Age i:_daysr ge_:s days Age ~2 clays Age zs cl ays Age 12 days 
*50 *75 *110
1.63 




275 *265 280 'J(295 3103.75 
3~5 4•JO 445 1400 570
5.90 
325 315 850 335 5157.65 
2:30 280 245 320 470
9.45 
205 235 245 260 555 
*rrwo specimens . 
t1.. 1~1:;Lic li rnit of steel pa ssed in case of one soccimen . 

1·1.ao 
'!'ABLE 30 
Age 23 clays 
*345 
!595 565 
t&5 495 
BO:\D S'l'RC\GTH OF BROKE=" S'l'O:s'E COl\ORWl'E, SERIES III 
(Each Resu lt the Average of r 
ests of rl'hree Specimens, except as Notecl. rrhe Diameter 
of llar Was 
Sacks of 
cement per 
cu. yd. of 
concrete 
J .61 
3.63 
:1.( .J 
7.38 9.0-2 
10.53 
*'l'wo specimens . 
%" with, Depth of Embcclment of 7" except for Last Half, Where Det)th Was 5". Plain Hound Rods were Usccl) 
Unit boncl stress in 
pounds per squa re inch at slio of .0002 of an inch 
Age IAge IAge
~-.'.:!_:_~ 28 cl.:_:= 3 mo._ 
55 105 00 
*7j0 *820 *765
~ l~ I ~ 

*560 715 *575 
~ ~ ~ 
U ni t bond stress in 
Ultimate unit bondpounds per square 
stress in pounds .001 of an inch 
inch at slip of per square inch 
Age j Age IAge Age I A~e I Age
12 c~_: ~ cla~ ~~
12 daysl~ clays~ 
65  ll5  130  100  135  180  
~*760 *640  1  ~ 775  1  = *730  ~*860 *800  l  ~!1015 !885  I  ~tll45 !1055  
~  - •  - m  ~  

tJ~ lastic limit of str.el passed in case of one specimen . !Elastic lim it of steel passed in case of two ~J)ecimens. 
University of Texas Biilletin 
PERMEABILITY 'fES'fS 
Scope of the investigation. For these tests, made only in Se­ries I, twelve specimens of the standard: consistency .yere made from which thirty six small specimens were prepared. One half of these were of broken stone concrete and the other half of gravel. For each percentage of cement, there was one large specimen. To find the permeability, these specimens were sub­jected to water pressures of ten, thirty, and fifty pounds per square inch. In addition, a specimen of the ''six sack'' gravel concrete, the ''wet'' mixture, was prepand and tested. 
111oiilds. The moulds used in making the specimens for the permeability tests consisted of pieces of No. 24 galvanized iron pipe four inclw:; in inside diameter and fourteen inches long. 
Mixing, placing, etc. The moulds with the inside walls free from grease, clean and bright, were placed on a greased galvan­ized iron plate. The concrete for each percentage of cement was mixed in one batch as already described and put in place in layers of about three inches, being tamped with an iron rod one inch in diameter and weighing two and one half pounds. The specimens were smoothed off on top and after one day were stored under water without removing the moulds. After twelve days from the time of mixing, they were removed from the water and left in dry air until tested. 
Testing. The speicimen was prepared for testing by sawing 
three equal lengths out of the middle without removing the cov­
e,ring, making three pieces four .inches in diameter and four . 
inches long. On the top and bottom of each of these but separ­
ated from the concrete by rubber packing, two and one half inch 
pipe flanges were placed. Connecting these flanges were five 
bolts which were tightly drawn up so that no water could leak 
around the packing. One opening was bushed down for 0onnect­
ing a pressure pipe, while the other opening was left free for 
water leaking through to pass out. This arrangement reduced 
to a minimum the effect of any leakage between the concrete 
and the wall of the enclosing pipe mould. 'l'he area exposed to 
the water pressure was about 514 square inches. 
These small specimens, six at a time, were connected with the 
city water mains by means of three-quarter inch pipes. To the 
Physical Properties of Dense Concrete 
main inlet pipe, a pressure gage was attached together with a 
valve and waste pipe for regulating the. pressure. When a test 
was to be run, the pressure was set and the amount of leakage 
was obtainE·d by4l catching and taking the time required. This 
~as repeated at intervals after the specimens 'had been subjected 
to pressure for different periods. 'l'ests like the above were 
made for unit pressure of 10 pounds per square inch at the a?,e 
of one month, 30 pounds per square inch at two months, and 50 
pounds per square inch at three months. The leakage was caught 
and accurately weighed whenever of sufficient amount, while for 
the more impervious specimens, the appear.ance of the free s1n·­
face was noted as to whether it was wet, moist, or dry. 
After the completion of these tests, the iron cover was re­moved and the condition of the sides carefully examined. The specimen in each set showing up worst in respect to the ''pock­ets" in the concrete: or apparent leakage along the walls (if this specimen also appeared to show the greatest leakage) was then excluded in arriving at conclusions in regard to the permea­, bility of the concrete. Otherwise the results from all these spe­
cimens were recorded. 
After this inspection, the spe:cimens were dressed down on 
the sides with an emery wheel to take off the thin skin of neat 
cement which had flushed to the surface due to the tamping of 
the concrete. They were then thoroughly dried out until 
they ceased to lose weight. On removal from the drying oven, 
they were imm~rsed in water and kept there until they ceased 
to gain in weight. The total increase in weight was then com­
pared ·with the initial weight when dry. In this way a measure 
of their porosity was obtained. 
Resitlts. The results of the permeability tests are shown in 
tables 31 to 36 and those of the porosity tests are given in tables 
37 and 38. 
University of Texas Bulletin 
TABLE 31 
P£R.)1£ABILI'l'l: 'l'ES'l' Of<' GRAVfl, CO:\'CRE'l'E UNDER PRESSURE OF TEN 
POUKDS P ER SQUARE INCH, SERIES I 

(Age at 'l'imc of 'l'est About One Month) 
passing tbrough' in 5 minutes and con­
Sacks of cement per cu. ycl. of concrete  No . of spec. Number of grams of wadi ti on 1 br . 6 hrs . 1 day --­---­-­-moist 3.0 'U ) --­-­--­-moist m oist mois t 2 dry 2 dry 1 moist 1 moist dry on edge on edge clry dry clry dry clry dry ter  
l.IJ6 3.80 *5.82 5.96  
8.45  
9.67  Iclry dry clry _ _ _ ._____ _  
11.50  1 S 11 clr > dry d1 y  

of surface of specimen 
2 days 
1. 3 
0.01 
dry 
dry dry 
2 dry 
1 moist 

after being under pressure 
3 clays 4 days 5 clays 6 days 
o.75 1.C8 1.0 
- ---moist  1 ---­O. OG4  mo ist  clry  
dry  dry  dry  
clrl"  
dry  clry  dry  
2 dry  2 clry  J. dry  
1 moist  l moist  Z moist  

on eclge ~n edge ~n edge on e~ 
2 dry I 2 dry11 moist 
l mOist 
on edge Ion edge 
*The "wet" mixture. 
'l'ABLE 32 
PER~IEABILITY TES'l' OF GRAVEL CONCRE'l'E U:'<DEli PRESSURE OF THIRTY 
POUNDS PER SQUARE INCH, SERIES I 
(Age at Time of Test About 'l'wo Months) 

Sacks of cement per cu. ycl. of concr13 te 1.66 3.SO  :Ko . Of spec. Kumber of grams of water passing· through in 5 minutes and con­dition of surface of specimen after being under pressure _2_~1 ~­~-~~~.~~ays_~~~~lay:__ _~~~ 15.7 i 15.5 11.0 8 .0 6.J 6.3 6.3 d ry dry dry clry 0.01. 0.01 wet 1 dry 1 dry  

il5 .8'2 5.96  3 2 moist 2 moist on edge on edge clry clry clry dry dry dry dry dry clry dry dry l dry 1 moist dry dry dry dry dry dry on edge 2 dry dry dry dry l moist dry dry on edge ,--J--­_ 2_d_r_y_ __2_d_r_y___2_d_r:;­-2-._d_r_y_ -2-d-ry­--­2 dry ­3 / mo:st 1 moist 1 moist l moist 1moist 11 moist 11 moist on edge on edge I  
9.67 _____ 11. 50  

*The "wet" mixture. 
Physioal Propel'ties of Dense Concrete 
'l'ABLE 33 
PERMEABILl'l'Y 'l'EST OF GRAVEL CONORE'l'E UNDER PRESSURE OF FIF'l'Y 
POUNDS PER SQUAllE INCH, SEHJES I 
(Age at 'l'ime of 'l'est About Th ree Months) 

Sacks of 
cement No. Nurnber of grnrns of water 1rn ssing through in 5' minutes and cou­
per cu. of <lition of surface of specirncn after being under pressure 
yd. of spcc.11-------­_'.'~ncrctc __ ~~1~.'.::.~J~I-= days ~ays_I .J days ~~~=---~~ays J..(6 3 28.l 31.() 25.2 23.8 21.3 li.3 Jb.6 14.2 
--~~I 2 dry I clry moist moist =moist =moist =~o·~t moist"~ 
·AJ .82 dry clry dry clry dry dr)' dry 
--5~96 J2 ~l~----;;---,~-~;------dry-d~­
---S.~ ~~d~1-di;,------d~-d~,-~.-­
---~Y-l dr;--1 ;;;:;-!____ ----------­
0.67 moist 2 mo1i,t z n1u ~t I~ mo1 ..:t moist moist moist 
1 dry 
)] .3~ rnoist moist moist rnoi st 2 moist moist moist 
1 
*The ''wet" mixture. 
TABLE 34 PllH)JF.AB!Ll'l'Y 'l'F.S'l' OF BROKEC\' S'l'ONJ<~ CONCRE'l'E UNDER PRESSURE OF T EN POUKDS P ER SQUARE !:\'CH, SERIES (Age at 'l'imc of '!'est About One Month) 

No. 
of spec. 
_~ 

I 3 
Number of grams of \n1ter passing through in 5 rninutrs and con ­dition of surface of specimen after being under pressure 
~­
~~~­
wet 3.75 
d ry 
moist 
1 dry dry 
2 moist on cclgc 
drydry drydry 
d1"y dry 
_:__~~­
J .92 
0.03 
l dry 
2 moist on edge 
clry 
2 dry 
l moist 

2 dry 
1 moist OD edge 2 clays 1~~~ 
1.92 
0.92 
0.17 
O.OJ. 
1 dry 
1 dry 
2 moi~t 
Z moist 
on edge 
on edge 
dry 
dry 
2 dry 12a;.;­
1 moist 
1 moist 
on edge 
on edge 
~~~-I~~~~~~YS 
0.69  J .00  
0.00·1  0.01  
1 dry 2 mo ist on edge  
dry  dry  2 dry 1 moist  

dry dry 
cJry 
University of Texas Bulletin 
TABLE 35 
.l:'ER\!EABILII.Y 'l'ES'J' OF BROKE:\ S'l'O:\'E 00:\'CRE'l'E UNDEii PRESSURE OF 
'l'liIR'l'Y POUXDS PER SQUARE INCH, SERIES I 
(Age at 'l'ime of 'l'est About 'l'wo Months) 

Sacks of  
cement  No.  
per cu.  of  
yd. of  spec.  
concrete  
I.63 3.75  
5.9:>  3  
1 .63  3 3 
9.45  
-­ 
11.30  3  

r-..·umbcr of grams of water passing through in 5 minutes ancl con· 
dition of surface of specimen after being under pressure 
1---------·~--·~-----------------­
l !Jr. 
~
moist 
cl ry
I-a;:;­
I dry 
moiEt
Ion edge 
Sacks of cement per cu . yd. of concrete  
1.63  
3.7:) 5.90 7.63 9 .43 11 .3()  

1 l1r. 5().0 
moist on cclge 
dry 
3 dry
I 

~r-~ 
II 
moist 
on eclge 
ABSOHP'J.'TO~ 
1 clay I 2 days
6 ilrs . 
__:~~I 
0.05 ---:--­0.04 
moist clry on edge 
-a;;--a;:;­
2 dry dry l moist 
moist moi~t 
on cclgc on cclge 
6 hrs .  1 day  
CS .6  48.2  
moist  moist  
clry  2 dry J moist  
2 dry 1 mois t  2 dry 1 moist  

2 dry 
~~ 
0.02 
clry 
-a;:;­
2 dry 
1 moist 

1 dry 
2 moist 
1on cage 
2 days 
•15 .0 
m?t 
2 dry ·1 moist 
2 clry 
1 rnoist 

1 
moist Imoist I moi:;:t jslightJy 
on edge on edge n eclgr~ moi~t 'l'ABLE 37 
3 clays 
4 clays 
5 clays 
6 days 
-~-~
__6_._1_ 
__6!._ 
l dry 
().02 
dry 
1 wet 
----.------­
dry 
dry clry 
c11 y 
-a;:;--a;:;­
-ai;­
1 dr.v 
1 dry 1 dry 
1 dry 
2 moist 
2 moist 2. moist 
2 moist 
2 dn1
2 dry 2 dry 
1 moist 1 moist 
1 moist 
/ 
TABLE 36 
n:HMEAnILI'l'Y 'l'ES'l' OF BROKE:\ STO:\'E 00:\'0RE'l'E UKDER PRESSURE OF 
FIF'.l.'Y POUNDS PEI~ SQUARE I:'.\OH, SERIES I 
(Age at 'l'ime of Test About 'l'hree Montbs) 

Xo. 
/ Sumber of grams of "·3.tcr passlngi tl1rougb in 5 mfnutes and cou­of 
dition of surface of specimen after being under pressure spec.Ii'-------·-------------------· 
'!'EST OF GHAVEL OONORE'l'E, SERIES I 3 days 


32.6 wet 
2 dry 
1 moist 

4 days 
29 .6 
wet 
1 dry 
2 moist 

dry 
1 dry 

_2 mo~s_: 
sligl1tly
moist 
5 clays 
35 .8 
l dry 
2 moist 

1 rlry 
2 moist 

6 days 
3:2 .i 
wet 
2 (1ry 
l moist 
:l dry 

1 m oist 
1 clry 
_..:. mo:st 
slightly

moist 
(Each Resul t the Avcruge of 'l'ests of Three Specimens) 
~---------c;-----·-----------------------­
Sacks of ce-Percentage of increase in weight after -hours immersion 
ment J)Cl' cu. ycl. of concrete 24 hrs . 
1.ll6 5.75 3.8() 4.6• 
*5.82 5 .21 !'i.96· 3.42 
8.45 
9.67 
11.50 3.46 
*rihe "wot" mixture. 
45 hrs. 
4.'/() 48 hrs. 72 hrs. 76 hrs. 96 hrs. 
5.87  
5.25  
3.·16  
3.77  3.79  
3.96  4.00  
3.46  

6.00 
4.78 
3.47 
PhysioaZ Properties of Dense Concrete 81 
'l'ABLE 33 ABSOR.PTIOS 'l'ES'l' OF BROKEN S'l'O:\!E OONORE'l'E , SERIES I 
(Each Result the Average of 'rests of 'l \hree Specimens) 
Sacks of c;,-Percentage of increase in weight after -hours immersion 
mcnt per cu . 
yd. of concrete 2J hrs. 15 hrs. 

1. 63  6.15  
3. 75  5.40  5.47  
5.9J  4 .. J6  4.50  
7 .6:J  
9.n  3.70  
11.30  3.88  

48 hrs. 72 hrs. 
6 .25 
4.62 4.64 
3.76 
76 hrs.  96 hrs .  
6.35 5.52 4.57  
3.91  3.77  

A study of thESe tables will show that under a pressure of 10# per sq. inch and at an age of one month, the "six and eight sack" gravel concretes are impermeable, the "ten and twelve sack" are practically so, while the "four sack" leaks slightly and the• ''two sack'' freely. The broken stone concrete at the same age and under the same pressure shows very much the same results. The ''eight sack'' is absolutely impermeable, the "six, ten and twelve sack" are almost so, while the "two and four sack'' leak rather freely. 
For a pressure of 30# per sq. inch and an age of two months, the condition is practically the same as at the earlier age except that the lrnkage is greater for the "two and four sack" mixes, while the ''ten and twelve sack'' show a slight amount of mois­ture passing through. The ''six sack'' broken stone concrete SE.'ems to show up better than at the age of one month. 
The three months test at a pressure of 50# per sq. inch shows the same relative impermeability for the different mixes with however a somewhat larger amount of leakage. In the gravel concrete· the ''six sack'' and ''eight sack'' specimens still show no leakage, while in the broken stone concrete at this pres­sure some, if not all of the specimens of each mix, show some moisture pm1sing, but this is least with the "six and eight sack" mixes. 
Contrasting the impermeability of the brok<'n stone and gravd conerete as a whole, it may be said that the gravel concrete appears more impermeable than the broken stone concrete of similar proportions but ·that there is very little difference be­tween the two. 
Th0 sp<'eim<'ns of the "six sack" "wet" mix are seen to be 
University of Texas Biillet-in 
practically impermeable at all three prESsures and compare very favorably with the ''six sack'' standard consistency mix. 
ABRASION TESTS 
Scope of the investigation. For these tests, made only in Se­ries I, ten specimens of standard consistency we·re prepared. Half of these were of broken stone conc,rete and the other half were of gravel. To find the resistance• to \vear, each specimen at the age of 28 .days was placed in a brick rattler and sub­jected to the action of a charge of steel cubes for 2,000 revolu­tions and the percentage of loss in weight · obtained for 500, 1,000 and 2,000 revolutions respectively. In addition, one spe-. cimen of ''six sack'' gravel concrete, the ''wet'' mixture, was made and tested. 
Moulds. The moulds for the specimens to be used in the abra­sion tests were made of No. 16 galvanized iron bent in to cir­cular shape forming a cylindrical shell twenty eight inches in inside diameter and eight inches high. On the inside of this was placEd a second hollow galvanized iron cylinder twenty inches in outside diameter and of the same h2ight as the larger cylinder. This was braced on the inside to keep it from chang­ing its shape and was separated from the walls of the larger piece by spacers which served to accurately center it. These two parts were set on a galvanized iron sheet and constituted the mould with which could be made a concrete ring twenty eight inches in outside diameter, four inches thick, and eight inches high. 
Mixing, placing, etc. The concrete was made as already de­scribed and placed in the moulds, being tamped with a tamper weighing about 13 pounds. The rings were reinforced with two ~ inch round steel rods extending completely around the cir­cumference. Ea.ch piece was placed about equal distances from the top and bottom, and the inside and outside walls. This reinforcement was necessary to· ensure the specimen's holc1ing together when being handled and tested. One specimen of each kind except the "two sack" concrete was made. The day after making, the mould was removl:.'d and the specimen st0red until tested. 
Physical Properties of Dense Concrete 
Testing. A brick rattler Vl'as used for making the test. After 28 days seasoning, the ring was carefully weighed and placed in the machine. A charge of 155# of cast steel pieces 411z"x 2%" and of 114# of 11h" cubes 1was placed on the inside and the open end closed with the movable plate on the machine. The rattler was then operated at a speed of 30 r. p. m. After 500 revolutions, the specimen was removed, weighed, and replaced. This was repeated for another 500 revolutions, after which, the machine was run 1000 revolutions more, making a total of 2000, when the specimen was finally removed and weighed. The con­dition of the surface was carefully noted and a record madn showing the uniformity of wear. 
Resiilts. The results of this test are shown in Table 39 for the gravel concrete and in Table 40> for the broken stone. The percentages of loss are obtained by taking the total loss in weight for each weighing and expressing this as . a percent of the orig­inal weight. Particular attention is called to the column headed ''Remarks;'' which indicates the condition of the wearing sur­face. 
The curves in Fig. 41 show clearly the results of the tests so far as loss in weig·ht is concerned. From these it is seen that the gravel concrete for most of the; differmt mixes loses less than the broken stone concr~te, but by referring to the table the latter is seen to wear down le·aving a smoother surface than that of the former for the leaner mixes, while both give good results for the richer mixes. 
'l"ABLE 39 
ABRASION 'l'ES'l' OF GRAVEL OOKORE'l'E, SERIES I 
(Eacb value determined from test of one specimen at age of 28 days.) 
Sacks of cc­
PcrC'cntage of loss after being revolved in rattler. 
ment bcr cu. 11---­
yd. of concrete 
500 
revolutions 
1.66 
3.32 *5.82 
3 .80 

5 .96 


1.0-2 
1.25
0 .75 

8.45 
0.47
9.67 
0.58
11.50 
1000 
revolutions 
5.87 
2.28 
2.49 
1.79 
l.34 
1.46 2000 
revolutions 
10.62 
5 .28 
5.1 7 3.3'l 
~ .09 
~.21 
Remarks 
Fair \Yem·. Few projecting gravel. Good wear. No projecting gravel. F ine wear. Very even. 
*'l1he "wet17 mixture. 
University of Texas Bulletin 
12  
II  
Q  
I  
10  
-­GRAVLL  CONCR ETE  
~  - - STONE:  CONCRETE:  
\ ~ \  •  WET MIX GRAVEL A9e cB do'is  CONCRETE:  
8  \ \\  
\\  
\  
if) if) 6 0 _J  \  \ \\  
I­r.  ~:'\  
LL] 5 u  "\ \\  \"'I"'-. \ u..... <oo I'\. ....z~~  
~  \ \"  ""  ""-... <'oo o R., ~  ...........  u 0  
l:!. \ \~  ~~o/a~•~~.  
'\ ~  ~ ~t--.u  
"'1( ~­ 0  ---­0 ~  
~.-~Re...  
- 0  
0  
0  2  3  5  6  7  0  9  10  II  IZ  
SACK~ Or  C[M[NT  Pm  CUBIC  YARD  Or  CONCR[T[  
Fig.  41.  Results of Abrasion Test on Hollow Cylinders of Gravel  
Concrete and Broken  Stone  Concrete,  Series  I.  

Physical Properties of Dense Concrete 8~ 
'l'ABLE 40 
ABHASIOX 'l'ES'l' OF BHOKDi STOXE OONOHETE, SElUES I 

tEarh value cleterminccl from test of one specimen at age of 28 days.) 
Sacks of ce­Percentage of loss after being revolved in rattler. 
ment per cu . 
yd. of concrete 500 1000 2000 
revolutions l'C\'Olutions revolutions 
------·11---------­
1.6~  
3.75  2.67  4 .97  8.74  
5.90  1.29  2.81  5.56  
7 .63  0.00  1.98  4.43  
9. 45  0.65  1.65  3.96  
11.30  o.n  l.12  3.41  

Remarks 
Not very goocl wear. No projectinll' stone 
Even good wear. No ]lrojecting stone. Very good wear . No J)rojecting s tone. 
Very good wear. Very even.