RELINING A BRICK-LINED RESERVOIR
It was customary ten or fifteen years ago for waterworks engineers and superintendents to construct large reservoirs with earth embankment and line them with brick. This method of construction has even been pursued since the advent of the concrete age. The usual method was to build the embankment on the interior on a one and one-half to one slope or a two to one slope,to cover this with a layer of clay puddle and then place on this a layer of brick laid on end with mortar joints or grouted joints. In some instances, a layer of asphalt was used either applied directly to the top of the brick lining or placed between the brick lining and the clay puddle. These linings in many instances answered the purpose for which they were intended in many instances answered the purpose for which they were intended and where an absolutely water tight structure was not essential, they proved very satisfactory. Of recent years, the necessity of purifying water for domestic use and in many cases of pumping it to the distributing reservoir has made a question of leakage in reservoir construction an important one in the economical operation of waterworks plants. Fortunately the advent of the concrete age and the cheapness with which reinforced concrete work can be constructed have furnished a ready solution of the problem. New water tight reservoirs of earth embankment design are now invariably lined with concrete. Many of the leading engineering firms use reinforced concrete structures almost exclusively as if properly designed, these structures as a rule can be constructed as cheaply as the earth embankment type and can be made absolutely water-tight. The writer in his specifications calls for a minimum loss of one vertical inch per 24 hours in reservoirs IS to 20 feet deep. He has constructed several reinforced concrete reservoirs recently and in not a single instance have they failed to meet the contract requirement. The problem of repairing the old reservoirs, built by the former method and now leaky, is not so simple. It always requires greater ingenuity to make good an old job than to construct an entirely new piece of work. Then too, the client does not feel like spending very much on repairs to an old structure that has already proven faulty. The result is that the engineer who undertakes such work is very’ liable to fall short of the mark and incur the everlasting displeasure of his clients. The reason for the failure of the old type of reservoir to be water tight can be ascribed as a rule to one of the following causes : Absence of expansion joints, lack of homeogenity in the lining. porosity of the lining, unequal settlement of earth embankment. Expansion joints were not used as the theory was that the joints between the bricks would open an infinitesimal amount thereby taking care of contraction but not to such an extent as to permit leakage. This result was not obtained in practice. In the Bellevue reservoir, to be hereafter described, there was very little leakage in the summer months but several hundred thousand gallons per day in the winter and this in spite of the fact that the brick work consisted of a two foot vertical wall circular in plan and backed with an earth embankment so that there was outside pressure against a masonry arch. Probably the most serious item is lack of homogenity of material. The slightest settlement i nan earth slope embankment at any point is very apt to cause the brick work to settle at the same point and to open the joints between bricks. Where concrete slabs are used, the slab will arch over any small settlement and it will take considerable settlement to break the lining. As concrete is the material that has proven so successful in new construction, it is naturally looked to for remedying the faulty construction in old work. It is especially adapted to lining work as it is comparatively cheap, can be made impervious when properly placed, can be placed in large units not subject to small settlements in foundation, and can be provided with expansion joints. Where the old lining has a good foundation and has not settled too unequally it will furnish an excellent backing for a concrete lining. This should be placed in blocks eight feet square and six inches thick. The blocks should be laid in checker board style with alternate blocks first and with balance of blocks after the first set have completely set up. There should be a three inch lap joint left on the bottom of the first set of blocks and 3-8 inch opening above this along between the adjacent blocks. After the second set of blocks have been poured, the 3-8 inch groove should be filled with a suitable asphaltum compound. The tops of all blocks should be finished with cement mortar before they have set up and this really forms the impervious coating to the concrete. However, all brick-lincd reservoirs do not have flat paved slopes. The Bellevue reservoirp of the Ohio Valley Water Company which supplies a large suburban territory west of Pittsburgh, consists of an earth embankment with a vertical lining of brick masonry 2 feet thick. The reservoir is 130 feet in diameter and 20 feet deep. It rests on solid rock foundation which several years ago was covered with a concrete floor. It lias a capacity of 2.000,000 gallons. It is supplied from a pump station on Neville Island where the water is obtained from a system of drilled wells anr pumped tothe reservoir under a head of 480 feet. The brick lining consists of brick masonry laid up with mortar joints and is two feet thick. It is backed as previously stated by an earth embankment. During the winter season the joints betwen the bricks opened up enough to allow as much as 400,000 gallons per day to leak out. An examination showed that this leakage must be through many infinitesimal cracks in the joints between bricks as there were no well defined points of leakage. The exterior embankment appeared to be in good shape and the leakage appeared to work down next to the wall in the shale formation below and thence through scams in the shale in a nearby ravine. While the leakage was not seriously threatening the stability of the reservoir, yet it was a serious proposition financially. The cost of pumping water at this plant against the high head prevailing, was at that time 3 cents per thousand gallons. This cost includes only fuel and labor. It was estimated that the leakage averaged 100,000 gallons per day per annum which amounted to a financial loss of $3.00 per day, or $1,095.00 per year. This capitized at 8 per cent would represent an investment of $13,688.00. Several schemes were proposed to stop this leakage, some of them involving patented water proofing methods, hut it was finally decided to reline the reservoir with an 8 inch concrete lining. This lining was designed to he constructed in sections twenty-nine feet long horizontally and extending the full vertical height of the reservoir. The sections were connected by a metal expansion joint. This was made of thin sheets of copper of No, 28 (iauge 12”x20’. The sheet of copper was corrugated and then folded at the center for a width of 4″. One edge was inserted in the section to be built first and when the adjacent section was constructed the other edge was inserted in it, leaving the fold between the faces of the adjacent sections. The sections were built alternately. The accompanying illustrationss show the method used by the contractor in handling form work. There were forms for four sections so that the work could progress continuously. It took a day to pour one section and while this section was being poured, the carpenters were bracing the section for the next day’s work and laborers were removing forms from sections that had been poured two day’s before and were setting this form for work to follow two days later. The form for each section was constructed in one piece with vertical ‘-mils and horizontal nailing pieces. These nailing pieces were 2×12 cut to the arc of the circle and spaced 2’-o” center to center. I” sheeting was nailed to these vertically and 2×10 struts also vertical were spaced on the hack of the form 5 feet center to center. The braces were nailed to these struts. The specifications stated that the contractor must not cut any holes in the concrete floor for supporting or bracing form work. Heavy pieces of timber were therefore laid on the floor of the reservoir entirely across it and braced against the wall on the opposite side. The form braces were then nailed to these timbers and were arranged as shown in the photograph. Kach section was poured in one day. J lie concrete was a 1-2-4 mixture and was placed wet and thoroughly spaded. The filling was made slowly so that the concrete in the lower part of the section would attain its initial set before the pressure from above could cause a deformation of forms. Upon the completion of a section it was allowed to stand two days before removing the form. The surface was then gone over with tools, imperfections removed, and a thin coat of liquid cement grout was applied. Upon the completion of the work, the test showed that the reservoir was absolutely water tight. The work had to he completed in 24 days according to the terms of the contract as while the reservoir was out of commission it was necessary to pump direct into the mains, which is a difficult feat under conditions existing at this station. The work was promptly executed bv the contractor W’thin the contract period. The total cost of the work was $2,(MM).00 so that the saving effected will pay for it in full in two years. The contractors were Pihl & Miller of Pittsburgh. The engineering including design and supervision of construction was handled by the firm of Chester & Fleming, of which the writer is a member.