Concrete Aggregates

The importance of a careful grading of dif ferent-sized aggregates as affecting the strength of concrete, is graphically illustrated in the samples of natural bank sand and gravel and limestone screenings shown respectively in Plates 1 and 2. The samples are photographed and reproduced at actual size.

In Plate 1, both samples (A and B) were taken from the same natural bank of sand and gravel of glacial origin, at Attica, Indiana. They differ only in the size of the grains, due to screening; and give, in a general way, a fairly correct idea of the appearance of a well-graded sand. A, the finer sample, contained 34 per cent of voids; B, the coarser, 26.9 per cent. When made up into a 1 :3 mortar—that is, 1 part cement to 3 parts aggregate—the mortar from the coarser material, in one year, developed a compressive strength of 7,750 lbs. per square inch, whereas the finer material gave a mortar having a compressive strength of only 4,475 lbs. per square inch—a difference in strength of over 70 per cent in favor of the coarser material.

A similar relative superiority of coarse as compared with fine material, is shown by the samples of limestone screenings illustrated in Plate 2. Sample A was taken from a crushing plant near St. Louis, Mo., and shows a screening in which the grading is not at all uniform. Sam ple B came from a plant at Greenfield, Ohio, and represents the screenings separated from the crusher-run material by a screen; and it will be noted that, with the exception of the large proportion of fine material, the screenings are very well graded. Sample A contained 42.1 per cent of voids; B, 37.5 per cent. B gave a 1 : 3 mortar having a compressive strength, at one year, of 8,500 lbs. to the square inch, as com pared with only 4,908 lbs. per square inch for A. Here, again, the difference in strength was over 70 per cent in favor of the coarser screenings, showing that a large proportion of fine particles is detrimental to strength.

The relative percentage of voids in two mate rials, one of uniform size and shape, and the of irregular size and shape, is clearly shown by comparing the gravel of Sample B, Plate 1, with a mass of round shot of uniform size. In the irregular gravel, the voids equal 26.9 per cent, while in the shot they equal 47.6 per cent. That is to say, there will be nearly twice the space in gravel to fill with cement if the particles are round and all the same size.

Testing Sand.

Tests made by M. Rene Feret, a French engineer, with mortar made from dif ferent natural sands, showed great variation in strength, due merely to the different degrees of fineness in the various sands. He used a mix ture of 1 : by weight—that is, 1 part Port land cement to parts sand. At the age of five months, the mortars made from the vari ous sands showed varying degrees of tensile strength, in every case favoring the coarser material; and in compressive strength, the dif ferences were even more marked. The results are shown in Table III.

Similar tests were conducted by Sanford E. Thompson in 1905, on Portland cement concrete. The aggregates were all of the same kind of material, varying merely in the sizes of the particles; and the proportion of the cement to the total aggregate by weight remained the same. The results showed a strength of the con crete in some cases two and a-half times what it was in others.

Yield or Volumetric Test for Sand.

Refer ence has already been made in a general way to what is known as the "yield" or "volumetric" test, the purpose of which is to determine what one of two or more sands will produce the denser, and therefore the stronger, mortar in any given proportions. In a paper read before the National Association of Cement Users (June, 1906), Mr. Thompson described in detail the method of making this test, in substance as follows : If the proportions of the cement to sand are by vol ume, they must be reduced to weight proportions ; for example, if a sand weighs 83 pounds per cubic foot moist, and the moisture found by drying a small sample of it at 212 degrees F. (the temperature at which water boils and is driven off as steam) is 4 per cent, which corre sponds to about 3 pounds in the cubic foot, the weight of the dry sand in the cubic foot will be 83-3=80 pounds. If the proportions by volume are 1 :3—that is, one cubic foot of dry cement to three cubic feet of moist sand —and if we assume the weight of the cement as 100 pounds per cubic foot, the proportions by weight will be 100 pounds of cement to 3 times 80, which is 240 pounds of sand ; and this corresponds to 1: 2.4 by weight.