namely, that a column ascending to the summit of our atmosphere corresponds in weight with a column of water of thir ty-three feet in height, allowing the bas es, i. e. of the air and of the water, to be equal. Thus we find that where the air is withdrawn, by means of suckers, pis tons, valves, &c. from within a pipe, of which the lowest part is immersed in the water Contained in a well, &c. the fluid will rise to the height of thirty-three feet within the pipe, supplying the place of the air thus withdrawn, This is effected by the pressure of the atmosphere on the surface of the water ; whereby it is forc ed into the space formerly occupied by the air. Generally speaking, it is not a sudden operation ; for unless the well be very shallow, it will require many strokes of a pump to withdraw so much air as may so far rarify the residue within the pipe, as to allow the water to rise thirty three feet above its level. This is the greatest height to which water can be in duced by a sucking pump. In this con trivance the piston, A, see fig. 9, has a valve, which, as the rod draws up, is closed by the pressure of the air above it ; but in descending it opens, and allows the water, which had flowed into the lower part, whence the air was withdrawn, to rush through ; as the piston is raised again, the weight of the water forcibly oppresses the valve, until it finds a lateral passage at B, whence it issues, and in this manner any quantity may be raised. If the water has a direct issue, as in the common spouts of pumps, no further ap paratus is wanted ; but if it is to be re tained, or pass through any other pipes more elevated than the debouchure, B, there must be a small angular projec tion, as shewn by the dotted lines, to admit the valve C, also pointing up wards. In dry weather, or when the pump is not much used, the leather bind ing of the piston, as also the valvesp will become dry ; therefore it is necessary, on such occasions, to throw in a pail-full or two of water, to moisten them ; else the air will pass downwards as the piston ris es, and prevent that exhaustion on which the ascent of the water depends. It is ge nerally necessary to have a valve at the bottom of the pipe, to keep in the water drawn into it, in order that the labour may be decreased ; and that, if the pump ing be intermitted, there may be less trouble in bringing up the water within reach of the piston.
Where the water lays near the surface, a lifting-pump may be used. This is nearly the same with the former ; but requires the piston should be forced down beneath the level of the water in the well. In this it is not so indespensably necessary that the leather on the piston should fit so close ; though it is the ,better for so do ing. In the lifting-pump, the whole de pends on actually raising the water from the well as though it were done by means of a bucket ; this occasions many to apply that designation to the piston. The same precautions are necessary, if the water is to be passed into any pipe, as has been stated regarding the debouchure of the sucking-pump.
The forcing pump has a solid piston, as seen at Ain fig. 10, which, after the water has passed the valve at B, is pressed down, and causes the fluid to pass into the conducting pipe C, where there is also a valve d, to prevent its return. The
valve at B closes as the piston descends, while that at d rises, to allow its escape from the main pipe. When the piston rises, the water follows, as in the two for mer instances, through the lower valve B, while the smaller valve at d is also clos ed by the super-incumbent water in the conduit, and by the attraction of the piston the water rushing after it to pre vent a vacuum. In this kind of pump, the piston must fit extremely close; both on account of the intended attraction of the fluid from below, and to prevent its escape upwards, when the piston is press ed downwards.
The whole of those inventions, which raise water by alternate risings and fall ings of only one piston, are subject to the inconvenience of having the water issue in jerks, which, hr some instances, would prove highly inconvenient. To remedy this, a cistern should be placed near the debouchure, or spout, whence a small stream would flow with much less varia tion than from the spout itself. But the best mode of regulating the issue of wa ter is by aid of an air-vessel, as in a fire engine See PNEUMATICS.
To detail all the varieties of pumps that are in use would be both beyond the li mits of this work, and of no real utility to the reader : we shall, therefore, enter upon the description of the valves in ge neral estimation, and then proceed to give a brief account of hydraulic machinery.
The most common kind of 'valve con sists of h piece of stiff leather, such as is applied for soles in shoes, and is generally known by the name of pump-leather. On its upper side a piece df milled lead is rivetted firmly, and the part where it is to be fixed on the frame, or shell, of the piston is grooved, for the purpose of giv ing it pliancy, that it may work up and down, as if on a hinge. Fig. 11, shows the plan ; and fig. 12, the profile of this valve, which is cheap, simple, and easily repaired, though it has the defect of be ing liable to choke, and of not rising high enough to allow a sufficient passage for the water.
Fig. 13, shows a button-valve, which is merely a piece of turned metal A, having a shank, B, of about eight inches or a foot in length according to the depth of the block, x z. The shank passes through the bar C, at the bottom of the block, and is prevented from coming up too high by the stud or nut o, at its bottom. When the water rises, it forces up the button A, and passes through the hollow in the block, of which the superior part is expanded so as to fit the bottom, which, being the frustrum of a cone, necessarily fits close into the expanded part as the water presses it, after having passed up wards in consequence of the descent of the piston, which may either be solid, as in a forcing-pump, or valved, as in a lift ing or a sucking-pump. This valve may be applied to a piston, as well as to that part of the pipe which retains the water, that may be within reach of the piston's action. An improvement has been made to this valve, by adding a ball of some weight to the bottom of the shank B, and excavating the button, in order to reduce its weight in proportion : this insures the regular descent of the button to its seat.