264. The usual specification for sand is that it shall be " clean, sharp and siliceous." We have shown that it need not be siliceous, and we have also noted that one authority considers that it need not be sharp, though this latter does not appear to be proven; let us see what interpretation should be given to the word "clean" if it must be retained in all specifications for sand.

Mr. E. C. Clarke, in the tests for the Boston Main Drainage Works, showed that "clay in moderate amounts" (ten per cent, to thirty per cent, of the sand) "does not weaken cement mortars." Calcareous marl might be considered an impurity, but we have seen that M. Alexandre found that sands containing this material gave excellent results. On the other hand, there seems to be no doubt that loam, peaty matter or humus will very materially decrease the strength of mortars, or even destroy them entirely. Likewise, decayed particles of some kinds of stone, or grains which readily break up into thin scales, should be strenuously avoided.

265. Detection Of Impurities

Clean sand when rubbed in the hand will not leave fine particles adhering to it, but should the sand not prove to be clean, the character of the impurities should be investigated before finally rejecting it. When there is not time for making proper tests, it will, of course, be safest to use only such sand as has no foreign matter whatever; but when strictly pure sand can only be obtained at great Cost, tests may show that a small percentage of impurities may be tolerated.

Another simple test, beside the one of rubbing in the hand, is to place a little of the sand in a test tube filled with water; if any impurities are present, they may separate from the sand on account of their lighter weight, or if in a very fine state of division, the water may be rendered murky in appearance. This test is not absolute, however, especially for calcareous sand, as the fine particles of limestone will give the murky appearance to the water, although not objectionable except on account of their extreme fineness.

The use of poor sand will result in a larger proportionate decrease in strength for a mortar containing a large amount of sand than for one made with a small amount. The effect of incorporating various foreign substances in cement mortar is treated in Art. 49. As some of these substances may occur in sand, the article referred to should be read in connection with this subject.

266. Sand Washing

When impurities occur, they may sometimes be removed by washing, but such work must be carefully inspected if the foreign matter be of a really dangerous character.

In the construction of the Canal at the Cascades, Columbia River, Oregon, quite an elaborate concrete plant was established, which had in connection a sand and gravel washer and separator.1 This consisted of a tube about two and one-half feet in diameter and seventeen feet long, made of one-quarter-inch boiler iron and revolving about an axis slightly inclined to the horizontal. Angle irons were riveted on the inside of the tube to carry the material up on the side and drop it again, while a spray of water issued from a perforated pipe inside the tube. The materials were separated by screens near the lower end of the tube. The material contained considerable earthy matter and is said to have been fairly well washed by this process.

Another style of sand washer was designed by the contractors for the construction of Lock No. 3, improvement of Alleghany River.2 The sand contained earthy matter and some coal, the latter being hard to remove by ordinary processes. A large barrel or tank, nine feet in diameter and seven feet high, was provided with double floor, the upper one being pierced with one-inch holes. Paddles were attached to a vertical shaft in the axis of the tank and revolved by suitable gearing, while water was forced into the space between the two floors. The water finding its way through the holes in the upper floor, passed up through the sand and overflowed at the top, carrying with it the coal and sediment. The Cost of washing is said to have been about seven cents per cubic yard, but it is evident that methods of handling would have to be quite perfect to keep the Cost at so low a figure.

1 Report of Lt. Edw. Burr, Report Chief of Engineers, 1891, p. 3334.

2 W. H. Rober, Engineering News, Feb. 16, 1899.