24 
HII.LKTIX 121G, U. S. DEPARTMENT OF AGRICULTURE. 
Molds. 
Sampling. 
Preparation of 
specimen. 
Capping. 
Caring and 
storing. 
be used. (The sizes most commonly used are 6 by 12 inches and 
8 by Id inches.) 
."'.. ("i The molds shall he cylindrical in form, made of non- 
absorbenl material, and shall be substantial enough to hold their 
form during the molding of the test specimens. They shall not 
vary in diameter more than one-sixteenth inch in any direction, 
nor shall they vary in height more than one-sixteenth inch from 
the height required. They shall be substantially water-tight so that 
there will be no Leakage of water from the test specimen during 
mold;' 
(b) Each mold shall he provided with a base plate having a 
plane surface and made of nonabsorbent material. This plate 
shall be large enough in diameter t<> support the form properly 
withoul leakage. Plate glass or planed metal are satisfactory for 
this purpose. A similar plate should be provided for covering the 
top surface of the test specimen after it is molded. 
4. (a ) Concrete for the test specimens shall he taken immedi- 
ately after it has been placed in the work. All the concrete for 
each sample shall he taken from one place, A sufficient numher 
of samples — each large enough to make one test specimen — shall 
be taken at different points so that the test specimens made from 
them will give a fair average of the concrete placed in that portion 
of the structure selected for tests. The location from which 
each sample is taken shall he noted clearly for future reference. 
(b) In securing samples, the concrete shall he taken from the 
mass by a shovel or similar implement and placed in a large 
pail or other receptacle for transporting to the point of molding. 
Care shall he taken to see that each test specimen represents 
the total mixture of the concrete at that place. Different samples 
shall not be mixed together but each sample shall make one speci- 
men. 
5. (a) The pails or other receptacles containing the samples 
of concrete shall he taken as quickly as possible to the place 
selected for molding test specimens. To offset segregation of the 
concrete during transportation, each sample shall he dumped into 
a nonabsorbent, water-tight receptacle and, after slight stirring, 
immediately placed in the mold. 
i /; i The test specimens shall he molded by placing the concrete 
in the form in layers approximately 4 inches in thickness. Each 
layer shall he puddled with from 25 to 30 strokes of a five-eighths 
to three-quarter-inch bar about - feet long, tapered slightly at 
the lower end. After puddling the top layer, the surface con- 
crete shall he Struck off with a trowel and covered with the top 
cover plate which will later he used in capping the test speci- 
mens. 
6. Two to four hours after molding, the test specimens shall he 
capped with a thin layer of stiff, neat, cement paste in order that 
the cylinder may present a smooth end for testing. The cap can 
best he formed by means of a piece of plate ;rlass one-quarter 
inch thick and of a diameter ii or :-; inches Larger than that of 
the mold. This plate is worked on the fresh cement paste until 
it rests on top of the form. The cement for capping should he 
mixed to a stiff paste some time hefore it is to he used, in order 
i (1 avoid the tendency of the cap to shrink. Adhesion of tin- con- 
crete to the top ami bottom plates can he avoided by oiling the 
plates or by Inserting a sheet of paraffined tissue paper. 
7. At the end of -is hours the test specimens shall he removed 
from the molds and buried in damp sand, unless cardboard molds 
are used, in which case test specimens may he buried in damp 
sand without removal of the mold, thus permitting shipping of 
the test specimens in the molds. 
8. (") The lest specimens shall remain buried in damp sand 
until Id days prior to the date of test. They shall then he well 
packed in damp sand or wet shavings and shipped to the testing 
Laboratory, where they shall he stored either in a moist room or 
in damp sand until the date of test. 
