148 



C. E. LINDSAY 



through initial design and construction 

 features, and by subsequent nioditications 

 to correct deficiencies and extend the sys- 

 tem for additional uses. The problems 

 and solutions we used are briefly described 

 in the remainder of this report. 



Toxicity of the system was reduced as 

 much as possible by the use of Johns- 

 Manville Transite pipe for those portions 

 of the lines which were outside the labora- 

 tory building, extending from the lower 

 end of the suction line into the building 

 and from the building to the sea-water 

 storage tank. Inside the building, main 

 lines were of 4-inch threaded hard rubber 

 with 2-inch hard-rubber distribution lines 

 carrying sea water to the aquarium area, 

 and half-inch hard-rubber lines to indi- 

 vidual laboratory sinks. Originally the 

 only metal in contact with sea water con- 

 sisted of a cast-iron foot valve, three cast- 

 iron elbows, a 60- by 32-inch steel plate 

 covering the top of the suction screen box, 

 six cast-iron flanges, and four threaded 

 steel pipe adapters used to connect the 

 pump to the suction box and main dis- 

 charge lines. Several types of protective 

 coatings have been used on metal parts 

 exposed to sea water. In order of use they 

 were: Nitrose, Atco carbon elastic black, 

 Gaco, epoxy resin, and powdered poly- 

 ethylene. None of these has provided com- 

 plete protection, but in general epoxy resin 

 and powdered polyethylene appear most 

 satisfactory. A major source of metal 

 contact was the steel suction-box cover, 

 which was finally protected with a i^- 

 inch-thick neoprene blanket used as a 

 gasket between cover and sea water in the 

 box. In time most of the cast-iron and 

 steel flanges and adapters were replaced 

 with threaded Uscolite pipe. 



The main pump from the beginning 

 was an Ace Hard Rubber Company 21/^- 

 inch WEFM pump driven by a 7i/^-horse- 

 power single-phase electric motor. The 

 standby pump was a 1-inch 1 -horsepower 



Gardner-Denver cast-iron centrifugal 

 pump. Because of cost as well as antici- 

 pated corrosion difficulties, valves were 

 kept to a mininnun. Originally only two 

 4-inch rubber-lined diaphragm valves 

 were incorporated in the discharge lines, 

 positioned so as to permit direct pumping 

 without use of the storage tank, or to per- 

 mit use of storage-tank water only and 

 allow^ dewatering of intake, suction screen 

 box, pump, and initial section of the main 

 discharge line. 



Protection of the pump from physical 

 damage or clogging is a vital considera- 

 tion. Damage from ice formation was 

 prevented by burying intake pipes under- 

 ground and incorporating the pump room 

 inside the main laboratory building. Ex- 

 clusion of bottom detritus or substrate ma- 

 terial such, as gravel has been achieved by 

 plugging the end of a 13-foot length of 

 Transite pipe and perforating this pipe 

 with 1/4 -inch holes spaced on %-inch cen- 

 ters over the full length of the pipe. Ex- 

 cept for occasional growth of seaweed or 

 accumulation of heavy barnacle or mussel 

 set on the interior of the pipe, this type of 

 screened intake has been entirely adequate. 



However, when using hard rubber as 

 a pump liner, it is also essential to elimi- 

 nate the abrasive damage caused by shells 

 of barnacles, mussels, or oysters which 

 have set on the interior of the pipe and 

 may later become detached and sucked 

 into the pump. To avoid this difficulty, 

 a suction screen box was cast as an integral 

 part of the building inside the pumproom, 

 and adjacent to the seaward wall of the 

 building. In plan view, the shape of the 

 box is a trapezoid whose bases are 54 

 inches and 28 inches ; altitude is 38 inches, 

 and depth 36 inches. It has a volume of 

 approximately 32 cubic feet. Two screen 

 slots were cast into the concrete. The 4- 

 inch suction line was cast into the box on 

 the short base of the trapezoid, and two 

 pump suction flanges provided on the long 



