TESTS OF DRAINAGE PUMPING PLANTS. 21 



TEST OF PUMPING PLANT IN JEFFERSON DRAINAGE DISTRICT NO. 3, LAFITTE, LA. 



DESCRIPTION OF PLANT. 



This pumping plant is used to drain a tract of about 5,000 acres. 

 The average lift of the pumps is about 4 feet, and at times the 

 maximum lift is 10 feet, 



It consists of two 48-inch double-suction centrifugal drainage 

 pumps direct-connected to two 16 by 36 inch simple noncondensing 

 Corliss engines of the girder-frame type. The engines are fitted with 

 gravity release gear and a special governor for emergency use in case 

 the pumps lose their priming. The cut-off is adjustable by hand while 

 the engine is running. Steam is furnished to the engines by two 

 Brownwell horizontal return tubular boilers of 15*0 horsepower each. 

 The fuel used is Mexican crude oil. For starting the boilers with oil, 

 steam is supplied by a 10-horsepower boiler fired with wood or coal. 



The suction and discharge pipes are respectively 25 feet and 20 feet 

 long and are enlarged so that the areas of the intake and discharge are 

 four times the area of the discharge nozzle of the pump. The 

 suction pipes are tapered uniformly from one end to the other, while 

 the discharge pipes are tapered and flattened so that the outer ends 

 have a rectangualr cross section with the sides rounded to a radius 

 of 24 inches. A steam ejector is installed on each pump for use in 

 expelling the air in priming. 



While this plant was guaranteed to deliver at normal load 55,000 

 gallons per minute against a total head of 6 feet at a speed of 100 

 revolutions per minute, it was decided to test, it at a much lower 

 total head and somewhat greater capacity and speed. 



FIRST TEST. 



The first test of this plant determined the capacity and efficiency 

 of the pumps for a rather wide range of heads and speeds. During 

 the test the head increased rapidly, making it necessary to take 

 observations on the pumps as frequently as possible. It was decided 

 that a boiler-test run under such varying conditions would be of 

 little value. 



Tests were run oil each unit separately. In making the tests indi- 

 cator cards were taken, the speeds of the engines and pumps were 

 recorded, a Pitot tube traverse was made in each suctiort pipe, the 

 -ii'tion olid discharge canal gages were rend, and the dynamic head 

 was obtained by means of mercury manometers connected to each 

 suction pipe Dear tlie pump flange and to the discharge pipe near the 

 discharge flange. 



At tli<- beg inning of the test of unit No. 2 the difference in water 

 levels was 3.38 feet. This difference rapidly increased t<» 6 feet, at 

 whirl) point the attempt was made t<> keep the speed of the pumj 



