The material commonly employed for the surfaces of asphalt pavements consists of a mixture of asphalt cement, powdered limestone, and sand. The mixtures used in different places have varied considerably in character, according to the nature of the materials available and the amount and consistency of the bitumen employed. Experience has gradually devel oped the practice in different places, the work at first being largely experimental, defects in the early work being corrected by modifications in later mixtures. Much of this work has been done by very haphazard methods and without any careful analysis of the causes of defects and failures, or of the differences in materials used in different places.
Sand. It is common to grade sand in size by sifting through sieves of 10, 20, 30, 40, 5o, 8o, too and 200 meshes to the linear inch, finding the percentage which passes each sieve and is caught by the next finer one. The portion which passes the zoo-mesh sieve is too fine to be considered as sand, and is classed with the stone powder which is added as filler. The sands used for asphalt surface mixtures are much finer than those employed in cement mortars. In sand for this purpose most of the sand is usually fine enough to pass the 40-mesh sieve. In some instances, the bulk of the sand will pass through the 8o and ioo mesh sieves; in others, the larger portion will only pass the 4o and 50-mesh sieves. A grading of sizes in the sand is desirable on account of reducing the amount of voids to be filled by the asphalt, and frequently when a natural sand of correct sizes is not available, it is possible to secure a proper relation of sizes of grain by mixing two or more sands of differing sizes. In general, sand so graduated as to leave a small percentage of voids is desirable in order that the interstices may be fully filled with bitumen, but the sand of greatest density is not necessarily the best for this purpose, as there may be instances where the percentage of voids would not admit of a sufficient amount of bitumen.
The voids in the sand are commonly tested by filling a measure with packed sand and then deter mining the quantity of water that can be added to it. A better method with fine material is to determine the specific gravity of the sand and also the weight of a given volume of it, the volume of voids being the difference between the measured volume and the volume of solid matter represented by its weight.
The sand used should be hard and tough in order to resist wear well, but its chemical character is not of special importance, although there seems to be a difference in the adherence of bitumen to the surfaces of different sands. The reason for this difference is not apparent and cannot be judged in advance of actual trial. The shape of grain does not usually appear important. Rounded grains often pack more easily and form a more dense mass, but it has been sometimes thought that they move more readily upon each other and form a less firm surface.
Filler. The filler used in asphalt paving mixtures consists of very .finely ground mineral matter mixed with the bitumen for the purpose of rendering the surface more dense, and giving stiffness to it. The material commonly used for this purpose is ground limestone, although a number of other materials have been employed. Ground clay may make a good filler; slaked lime, and Portland or natural cement are also used, especially good results being obtained with Portland cement.
The filler should be finely ground; nearly all of it is usually required to pass a 200-mesh sieve, but the finer portions are too fine to be graded by the use of sieves. For this purpose the method of elutriation may be employed. By this method the powders of different degrees of pulverization are separated by observing the times required to settle after being shaken in a vessel of distilled water, the portions which settle in 15 seconds, one minute, and 30 minutes being deter mined, and the relative fineness of different samples thus compared.