66 BULLETIN" $49, U. S. DEPARTMENT OF AGRICULTURE. 
Compression tests. — The specimen to be used in compression tests shall be a cylinder 
not less than 6 inches in diameter and 12 inches high. When the diameter of the 
largest particle of aggregate runs over 2 inches an 8 by 16 inch cylinder is recommended. 
The mold should be of metal. A suitable type of mold consists of a 12-inch length 
of cold-drawn steel tubing 6 inches inside diameter, split along one element, and closed 
by means of a circumferential band and bolt. Suitable forms can also be made from 
galvanized iron. Forms should be tight and should rest on level nonabsorbent bases. 
At least three specimens should be made to cover any single point in a series of tests. 
Only one cylinder of a kind should be fabricated at one time. 
In testing compression specimens the speed of the moving head of the machine 
shall travel approximately 0.05 inch per minute when the machine is running idle. 
The bearing plates of the testing machine shall be brought into direct contact with 
the end of the specimen, and a spherical bearing block shall be used on top of the test 
piece. The diameter of the bearing block shall be approximately the same as that of 
the specimen ; the radius of the ball in the block should be not over one-half the radius 
of the test piece. As the testing head of the machine is brought down upon the top 
of the cylinder, the lower section of the adjustable block should be oscillated to and 
fro to insure a central bearing and to avoid pulling the cylinder to one side. 
The results of the tests of individual specimens should be reported. 
48. PROPOSED TRANSVERSE TESTS OF CONCRETE. 
Field specimens. — A slab 30 inches long, 8 or 12 inches wide, and of a depth equal 
to the depth of pavement should be employed. This specimen should be molded 
at the edge of the pavement with its long dimension parallel to the length of the 
road. The forms for separating the test piece from the remainder of the road should 
be made of sheet metal and should be submerged about three-fourths of an inch 
below the finished surface of the pa\-ement. In order to provide bearing surfaces 
and a uniform thickness at the center, three strips of wood 3 inches wide should be 
placed on the subgrade with their axes rimning transversely with respect to the axis 
of the test specimen and wide faces parallel to and equidistant from the top of the 
pavement. The boards should be placed near each end and at the center of the 
length of the test piece. Specimens should be tested over a 24-inch span and loaded 
at the center. 
Laboratory specimens. — A specimen 12 inches wide, 8 inches deep, and 30 inches 
long, to be tested by center load over a 24-inch span, is suggested for laboratory 
fabrication. The methods of proportioning, mixing, molding, and curing of the 
transverse test piece should be similar to the method previously outlined. 
The modulus of rupture S T may be found from the following expression: Thus 
oc p 
S T = T-p- in which P is the center load in pounds and b and d the breadth and depth 
in inches of the slab, respectively. 
49. PROPOSED TEST FOR CONSISTENCY OF CONCRETE. 
For the determination of the consistency either in the field or in the laboratory 
the committee proposes the use of the 4 by 8 by 12 inch conical frustum, as shown in 
figure 35. 
In making the test, the thoroughly cleaned frustum should be placed on a level 
nonabsorbent surface and filled with 3-inch layers of concrete. During filling, the 
mold should be held down by the operator placing his toes on the lip at the bottom 
of the mo d. As each layer of material is introduced, it should be puddled by a 
stirring motion with a one-half-inch rod to uniformly distribute the material. 
After the upper layer has been placed the top shall be struck off and the mold 
removed by slowly pulling it vertically tipward. The height of the frustum shall 
be measured and the slump calculated from the difference of the height of the mold 
and the frustum. 
