APPARATUS FOR RAISING WATER. Q*J 



The float in the veflel B, is a fmall cylindrical copper Description of 



veflel, one inch in height and two inches in diameter. It conftruftedVor 



was filled with water and clofed before it was fixed in its railing water, 



place. Though equal in bulk only to one ounce and three ^ a ' ,r . conden fr 



. . " . r i. . in an inverted 



quarters of water, yet it is quite Uifncient to move the fyphon fyphon. 



which would work a larger machine. 



The pipe C is half an inch in diameter : E, K, G, L, are 

 fmaller. The valves are leather. 



This machine, when in good order, raifes water nearly 12 

 feet above the ciftern, at the rate of 20§ pints per hour, and 

 performs all its operations as well as can be defired. When. 

 firft fet to work, the cock is fo placed as to clofe the top of 

 the pipe C, until the condenfed air begins to collect in the 

 veflel B, and then the communication is opened to the cham- 

 ber F. If the communication was open at firfl, the water 

 would be expelled from F into the ciftern, while the preflure 

 was inefficient to clofe the valve in the pipe I. 



The chambers F H being fmall, the fyphon moves fre- 

 quently ; but in a working machine thefe veflels fhould not 

 only be broad and fliallow, but capacious, that the wear of 

 machinery may be reduced to its ulmoft extent. 



To determine what quantity of water flows through the 

 apparatus, I fixed a fpout upon the top of the afcending water 

 pipe ; but in doing this I entangled the float in the veflel B, 

 that it could not be made to work the fyphon. The effluent 

 water, in this unemployed ftate of the machine, including 

 half a pint difplaced from the veflel B, amounted to 8| pints. 

 The fuperflucus water from the ciftern R meafured 8| pints 

 alfo. If the machine had been working the wade water would 

 have been Iefs^ as part would have been drawn off by the 

 fyphon. 



From feveral trials, this model appears to raife water above 

 the rate that might be eftimated by the experiment of March 6, 

 previoufly related. 



In actual practice, I think the allowance for wafle and 

 working the buckets, of one third or perhaps only of one 

 fourth of the fupply, will be fufficient ; then fuppofing the 

 apparatus fo adapted to the fupply, as 28 or 30 gallons from 

 the cittern will raife one gallon 18 feet, fo 84- or 90 gallons 

 will raife one gallon 44 feet, by three afcending columns. 



Th» 



