The Lawrence Filter Bed.
THE Lawrence filter bed, which is now in use at Lawrence, Mass., and of which we herewith present a series of views, has proved remarkably successful in operation. The construction of the bed was commenced in 1892, and it was finished in the fall of the following year. It was designed by Hiram F. Mills, C.E., chairman of the State Board of Health, and was constructed under his direction.
The area of the filter bed is 2J3 acres, and its proposed daily capacity is 5,000,000 gallons, or 2,000,000 gallons per acre. It is separated from the Merrimack Kiver by an embankment made up of the material excavated from the site of the bed. This excavation was carried to a depth of seven feet below low water in the river, and the embankment is designer! to exclude the highest spring freshets. The general level of the surface of the filtering sand filled‘into the excavation is two and one-half feet below low water in the river, and is flooded to a depth of one and one-half feet by water, which enters through a two-foot gate and flows through an open conduit at the foot of the enbankment. This channel has a paving laid in Portland cement and lined with Portland cement mortar. The water from the conduit overflows into very shallow distributing conduits of the same material which cross the bed nearly its full width once in thirty feet, all as shown by the plan and sections, Figures 1 and 2. The bottom of the excavation has cross ridges directly below the distributing conduits and these thirty feet, c. to c., which are two feet higher than the depressions between.
In the depressions fora port of the way across the bed are small drain pijres, laid with joints so open that the spigot does not enter the bell, and these are surrounded by a fourinch layer of stones, two inches in diameter or larger, outside of and upon this layer of stones aic four other layers of selected stones each wider than the one beneath it, as shown in the sections, Figure 2, having diameters cf about 1⅛, •’.» :⅛. and 3*tfi inches. I he upper layer of stones is covered with about one inch of coarse sand which spreads somewhat beyond it. The full depth of the stones is about one foot, while the layer of stones, or the upper layer, spreads to a width of seventeen feet. ‘1’he surface of the bed of filtering sand has depression ranging in depth from one foot to nothing, extending nearly across the bed, once in thirty feet, and corresponding with the distributing conduits. When the lower part of the filtering material becomes full of water, the water rises up the slopes and finally covers the upper level of the sand to a depth of one foot, at which height it remains during the greater part of the day. The filtered water passes through the under drains to a brick conduit, and through the old filter gallery to the pump well.
The pumps run about nineteen hours a day. About five hours before they stop the inlet gates supplying the filter bed from the river, arc to be closed, and the pumps continuing will uncover the surface of the sand, and drain the sand nearly to the bottom and allow it to fill with air. The inlet gate is to be opened again about an hour before the pumps start, when the sand is filled and gradually overflowed. This intermission once in twenty-four hours is found by the experiments of the Massachusetts State Board of Health to be sufficient to give the filter the advantages of intermittent filtration.
The filtering sand is of two kinds, as shown bv the secretions. No. 70 sand covers the ridge of the bottom, with a depth of only three feet, and extends for the width of twenty feet. It has grains, the finer ten per cent, of which have diameters equal to and less than o. 3. M. M. or o. 118 inches, and when clean will allow 70,000,000 gallons of water to pass through any depth when the acting heads is equal to that depth.
Directly over the underdraining stones, for a width of ten feet, is somewhat finer sand, whose finer ten per cent, have diameters equal to and less than 0.25. M. M., or 0.098 inches, and which when clean will allow 50,000,000 gallons of water to pass through any depth, when the acting head is equal to that depth, or the distance through which the water moves. It is desirable that the head used shall not be more than one foot in either case, and the distance through the sand to the underdraining stones is in the former case snen or eight feet, in the latter hve or six feet, reducing the quantity of water that can pass either sand to 8,000,oooor 10,000,000 gallons per acre per day, which quantity will in time be reduced to perhaps 5,000.000 gallons per acre by the slow infiltration of fine river silt into the laxly of the sand. This silt will, however, be nearly all retained upon, or very near the surface, there ob. structing the entrance of the water; and to maintain an even passage of 2,000,000 gallans per acre per day, it will be necessary to remove from the whole surface of the bed from oneeighth to one-fourth inches of sediment and sand each month, and a larger amount during the spring freshet, when much sediment is brought down the river, Thesand removed is to be replaced by clean sand at the time. By the present arrangement of a single bed, it is necessary to remove the sediment daily or nightly, when the water is drawn below the surface.