This section is from the book "Cyclopedia Of Architecture, Carpentry, And Building", by James C. et al. Also available from Amazon: Cyclopedia Of Architecture, Carpentry And Building.
The best, hardest, and most durable broken stone comes from the trap rocks, which are dark, heavy, close-grained rocks of igneous origin. The term granite is usually made to include not only true granite, but also gneiss, mica schist, syenite, etc. These are just as good for concrete work, and are usually less expensive. Limestone is suitable for some kinds of concrete work; but its strength is not so great as that of granite or trap rock, and it is more affected by a conflagration. Conglomerate, often called pudding stone, makes a very good concrete stone. The value of sandstone for concrete is very variable according to its texture. Some grades are very compact, hard, and tough, and make a good concrete; other grades are friable, and, like shale and slate, are practically unfit for use. Gravel consists of pebbles of various sizes, produced from stones which have been broken up and then worn smooth with rounded corners. The very fact that they have been exposed for indefinite periods to atmospheric disintegration and mechanical wear, is a proof of the durabilty and mechanical strength of the stone.
There is hardly any limitation to the size of stone which may be used in large blocks of massive concrete, since it is now frequently the custom to insert these large blocks and fill the spaces between them with a concrete of smaller stone. But the term broken stone should be confined to those pieces of a size which may be readily mixed up in a mass, as is done when mixing concrete; and this virtually limits the size to stones which will pass through a 2 1/2-inch ring. The lower limit in size is very indefinite, since the product of a stone crusher includes all sizes down to stone dust screenings, such as are substituted partially 01' entirely for sand, as previously noted. Practically the only use of broken stone in masonry construction is in the making of concrete; and, since one of the most essential features of good concrete construction is that the concrete shall have the greatest possible density, it is important to reduce the percentage of voids in the stone as much as possible. This percentage can be determined with sufficient accuracy for ordinary unimportant work, by the very simple method previously described for obtaining that percentage with sand - namely, by measuring how much water will be required to fill up the cavities in a given volume of dry stone. As before, such a simple determination is somewhat inexact, owing to the probability that hubbies of air will be retained in the stone which will reduce the percentage somewhat, and also because of the uncertainty involved as to whether the stone is previously dry or is saturated with water. Some engineers drop the stone slowly into the vessel containing the water, rather than pour the water into the vessel containing the stone, with the idea that the error due to the formation of air bubbles will be decreased by this method. The percentage of error, however, due to such causes, is far less than it is in a similar test of sand, and the error for ordinary work is too small to have any practical effect on the result.
A pail having a mean inside diameter of 10 inches and a height of 14 inches is filled with broken stone well shaken down; a similar pail filled with water to a depth of 8 inches is poured into the pail of stone until the water fills up all the cavities and is level with the top of the stone; there is still 2 1/4 inches depth of water in the pail. This means that a depth of 5 3/4 inches has been used to fill up the voids. The area of a 10-inch circle is 78.54 square inches and therefore the volume of the broken stone was 78.54 X 14 = 1,099.56 cubic inches. The volume of the water used to fill the pail was 78.54 X 5.75, or 451.6 cubic inches. This is 41 per cent of the volume of the stone, and is in this case the percentage of voids. The accuracy of the above computation depends largely on the accuracy of the measurement of the mean inside diameter of the pail. If the pail were truly cylindrical, there would be no inaccuracy. If the pail is flaring, the inaccuracy might be considerable; and if a precise value is desired, more accurate methods should be chosen to measure the volume of the stone and of the water.
It is invariably found that unscreened stone or the run of the crusher has a far less percentage of voids than screened stone, and it is therefore not only an extra expense, but also an injury to the concrete, to specify that broken stone shall be screened before being used in concrete, unless, as described later, it is intended to mix definite proportions of several sizes of carefully screened broken stone. Since the proportion of large and small particles in the run of the crusher depends considerably upon the character of the stone which is being broken up, and perhaps to some extent on the crusher itself, these proportions should be tested at frequent intervals during the progress of the work; and the amount of sand to be added to make a good concrete should be determined by trial tests, so that the resulting percentage of voids shall be as small as it is practicable to make it. It is usually found that the percentage of voids in crusher-run granite is a little larger than in limestone or gravel. This gives a slight advantage to the limestone and gravel, which tends to compensate for the weakness of the limestone and the rounded corners of the gravel.
85. Broken stone is frequently sold by the ton, instead of by the cubic yard; but as its weight varies from 2,200 to 3,200 pounds per cubic yard, an engineer or contractor is uncertain as to how many cubic yards he is buying or how much it costs him per cubic yard, unless he is able to test the particular stone and obtain an average figure as to its weight per unit of volume.
Cinders for concrete should be free from coal or soot. Usually a better mixture can be obtained by screening the fine stuff from the cinders and then mixing in a larger proportion of sand, than by using unscreened material, although, if the fine stuff is uniformly distributed through the mass, it may be used without screening, and a less proportion of sand used.
As shown later, the strength of cinder concrete is far less than that of stone concrete; and on this account it cannot be used where high compressive values are necessary. But on account of its ,very low cost compared with broken stone, especially under some conditions, it is used quite commonly for roofs, etc., on which the loads are comparatively small.
One possible objection to the use of cinders lies in the fact that they frequently contain sulphur and other chemicals which may produce corrosion of the reinforcing steel. In any structure where the strength of the concrete is a matter of importance, cinders should not be used without a thorough inspection, and even then the unit-compressive values allowed should be at a very low figure.