IMPROVEMENTS IN STEAM ENGINES. 01 



two hot-wells. The head of hot-water suffered to remain on 

 these valves must be moderate, or they will refuse to open; 

 for it must be remarked, that great part of the contents of the 

 air-pump is an elastic gaz r which suffers compression and is 

 not expelled, if the weight on the valve be too great. The 

 action of this air-pump is evident from the drawing. The 

 expulsion of the contents of each chamber creates a vacuum 

 in the other, which draws in the contents of the condenser; 

 and thus they act equally and alternately, agreeing in their 

 operation with the alternate condensation of the steam in the 

 opposite chambers of the cylinder. Experience proves this to 

 be a real improvement. 



The principle which has been applied to the construction 

 of the air-pump, is that upon which the main pump of our 

 water-works is constructed. A section of this pump is an- 

 nexed which perfectly explains it. 



This pump has so many advantages that, had the corpora- 

 tion of 1SOO permitted its disadvantages, (of which I shall 

 presently speak,) to be remedied by the means then proposed, 

 I have no doubt, but that I might now recommend its general 

 adoption, wherever a double steam-engine is used for pumping. 

 The drawing in Plate II, Fig. 9, will explain its construction; 

 A the working barrel. B the piston. C the feed-pipe. D the 

 rising mam p pe. F the valves which supply the working barrel. 

 G discharging valves in the ascending pipe. H the air-vessel — 

 The valve E, in the rising pipe, and the air-vessel H, are not ad- 

 ded to our pumps. The want of one or other of them, has these 

 disadvantages: as long as the engine makes only 1 1 or 12 strokes 

 per minute, no inconvenience whatever is perceived in the work- 

 ing of the pump-. But in the engine in the center-square, which 

 raises the water in an 18 inch pipe 5 1 feet, and which has less re- 

 dundant power than that on Schuylkill, the attempt to work 

 faster than 12 strokes per minute is vain; and, as it appears 

 tome, from two causes: 1st, whenever the piston is at its ut- 

 most ascent or descent, and makes a momentary stop, the 

 whole column of water follows the shutting valve, acquiring 

 momentum as it falls. The range of our valves is 16 inches, 

 the column therefore descends at an average 8 inches. It 



p 



