WATER. 



like the fprings of a travelling carriage, whofe jolts are in- 

 comparably lefs than thofe of a cart, and by this means 

 really enables a given force to propel a greater quantity of 

 water in the fame time. 



The ftrcam produced by the aflillance of an air-veffel is 

 almofl perfeftly equable, and as much water runs out dur- 

 ing the returning of the pifton as during its aftive llroke ; 

 but it muil not be imagined that it therefore doubles the 

 quantity of water. No more water can run out than what 

 is fent forwards by the pifton during its effeftire ftroke. 

 The continued ftream is produced only by retaining part of 

 this water in the air-veffel during t!ie llroke of the pifton, 

 and by providing a propelling force to aft during the pif- 

 ton's return ; but it cannot enable the moving force of the 

 pifton to produce an increafedcfFcft : for the compreflion which 

 is produced in the air-veflel, more than what is necelFary for 

 merely balancing the quiefcent column of water, reafts on the 

 pifton to refift its compreflion juft as much as the addition of 

 a column of water would do, the height of fuch column being 

 fufficient to produce the required velocity of the efflux. 



Machines for working Pumps. — The beft method of work- 

 ing pumps from a firft mover which afts with a rotative mo- 

 tion, is by means of cranks ; and if two or more pumps are 

 to be aftuated by the fame machine, the cranks for them 

 fiiould be placed at regular intervals round the centre, fo as 

 to produce a continual aSion. 



It has been obferved, in our article SrEAM-Enginc, that 

 the reciprocating motion obtained by a crank is very un- 

 equal, even when the rotative motion of the crank is quite 

 uniform. This renders the motion of the pifton in the bar- 

 rel of the pump irregular, for at the top and bottom of the 

 barrel the motion of the pifton is very flow, but when the 

 pifton is at the middle of the barrel the pifton moves 

 quickly. This property is a great advantage in working 

 pumps, becaufe it puts the column of water in motion with 

 a lefs fudden fhock ; but it has been very generally miftaken 

 and confidered as a dcfeft, and many ingenious contrivances 

 have been propofed, by means of racks and pinions, to give 

 an uniform motion to the pifton-rods of pumps. Thcfe have 

 never fuccecded in praftice, and have always been laid aiide. 

 Tlie attempts of mechanicians to correft this unequal 

 motion of the pifton-rod are mifplaccd ; for if it could be 

 done it would greatly injure the performance of the pump. 

 As this is a favourite fpeculation, and new attempts to 

 perfeft it are conftanily making, we think it right to fliew 

 the reafon of their failure. 



Suppofe the firft mover to move uniformly with a rotatory 

 motion, and that the machinery is fo conttrufted, that the 

 pifton-rod will be moved up and down with a regular mo- 

 tion, or that the velocity of the pifton fliall be at all times 

 the fame, whether it is at the top or bottom, or in the 

 middle of its courfe. In this cafe, at every reciprocation, 

 the column of water in the main pipe muft be fuddenly 

 Brged into motion from a ftate of reft, and the machine 

 could not perform one ftroke, if the velocity of the firft 

 mover did not ilacken a little, or if the different parts of the 

 machine did not yield by bending or compreflion. Thefe 

 Itrains would be fo fudden and violent, that no flrength of 

 materials could withftand the violence of the fhocks at every 

 reciprocation of the motion. This would be chiefly ex- 

 perienced in great works which are put in motion by a 

 water-wheel, or fome other equal power, exerted on a large 

 mafs of matter, of which the machine confifts. The water- 

 wheel, being of great weight, moves with fteadincfs or uni- 

 formity ; and when an additional refiftance is oppofed to it 

 by the beginning of a new ftroke of tfie pifton, its quantity 

 of motion is but little afleftcd by this addition, and it pro- 



ceeds with very little lofs of motion. The machine muft 

 therefore yield a little by bending and compreflion, or it 

 muft break to pieces, which is the common event. 



A crank is free from this inconvenience, becaufe it ac- 

 celerates the pifton gradually, and brings it gradually to reft, 

 while the water-wheel moves round with almoft perfeft uni- 

 forrnity. It has been ftated as an inconvenience of this flow 

 motion of the pifton at the beginning of its ftroke, that the 

 valves do not (hut with rapidity, fo that fome water gets 

 back through them ; but this is a miftake, becaufe the valves 

 always fall by their own weight as foon as the water ceafes 

 to flow upwards through them. Now when the pifton be- 

 gins to move with its flow motion towards the end of the 

 ftroke, lefs water is caufed to flow through the valves, and 

 in confequence they clofe gradually, and will be fully fhut 

 by the time that the pifton becomes motionlefs, and before it 

 begins to return. This is ftiewn in the large machines, fuch 

 as that of London-bridge, where the pumps are worked by 

 cranks, and the valves clofe imperceptibly ; but in a fteam- 

 engine of the fame power, the fhock occafioned by the fliut- 

 ting of the valves is exceedingly violent. In fhort, by a 

 judicious application of the crank and a fly-wheel, or an air- 

 veflTcl, and by employing two or three barrels to the pump, 

 the evils of the reciprocating motion of pumps may be com- 

 pletely remedied, and on this account we confider, that if a 

 rotatory pump could be brought to perfeftion, it would have 

 no fuperiority over an accurate pump with a ftraight barrel. 

 Jlfr. Smeaton's proportions for a tiuo-horfe pump machine for 

 raifing water are as follow : horfc-track thirty feet diame- 

 ter ; great cog-wheel nineteen feet diameter, with 144 cogs ; 

 this gave motion to a trundle of feventeen ftaves, fixed upon 

 an horizontal axis, which carried a caft-iron fly-wheel of ten 

 feet diameter, and the rim three inches by three inches. On 

 the extremity of the horizontal axis was- a crank of a foot 

 and a half in length, which, by means of a connefting-rod, 

 gave motion to one end of a working beam or lever of feven- 

 teen feet long, which was poifed on a centre in the middle 

 of its length, and at the oppofite end was an arched feftor 

 for the chain, by which the pump-rod was fufpended. The 

 pump was a fucking-pump, fix inches diameter in the barrel, 

 and the length of the ftroke was three feet. A weight was 

 applied to the end of the beam over the crank, which was 

 fufiicient to balance one-half of the column of water in the 

 pump. In this machine, when the horfes walked two miles 

 and a half per hour, they made two turns and one-third per 

 minute. The trundle and fly-wheel made twenty turns per 

 minute; the pump made the fame number of effeftive ftrokes, 

 and raifed upwards of a hundred barrels ale meafnrc^irr hour. 

 By the counter balance and the fly-wheel, the refiftance to 

 the horfes was rendered perfeftly uniform. 



The Pump Machine at Blenheim, which was erefted by 

 Mr. Alderfea for the duke of Marlborough, is thus de- 

 fcribed by Mr. Fergufon in his leftures. The water-wheel 

 is underfliot, and is turned by the fall of the water running 

 down an inclined plane, and ftriking the floats of the wheel. 

 The extremity of the pivot or gudgeon is formed into any 

 number of cranks ; for inftance fix, that is, three at each end 

 of the axis, more or lefs, according to the force of the fall 

 of water, and the height to which the water is intended to 

 be raifed by the engine. As the water-wheel turns round, 

 thefe cranks move as many levers up and down, by the iron 

 connefting-rods. Thefe levers alternately raife and deprefs 

 the piftons of the forcing-pumps by other iron rods, which 

 are attached to the oppofite ends of the levers, and as one 

 is raifed the oppofite pifton is dcpreffed. Pipes go from 

 all thefe pumps, to convey the water which they draw up 

 (to a fmall height) into a clofe ciilcrn or box, from wliich 

 G 2 the 



