The balls i i are supported by the bent levers h f, and as they are made to revolve with the fly wheel axis, by means of a band passing round the pulley c, any increase in the speed of the engine will cause the balls to diverge. The moment this takes place, the shorter arm of the lever n is depressed, and as the extremity l is connected with the steam-pipe by the throttle valve, the supply of steam must of necessity be diminished, and the speed of the engine reduced.
As the working power of the engine depends very materially on the accurate fitting of the piston, it may be adviseable to examine some of the modes of effecting this important object.
Mr. Smeaton, who greatly improved the atmospheric engine, coated the under side of the piston with elm or beech planks about two inches thick; the wooden bottom being screwed to the iron with a double thickness of flannel and tar, to exclude the air between the iron and the wood. By the adoption of this improvement, its property of conducting heat was reduced, and the wood having been previously jointed, with the grain radiating in all directions from the centre, was not liable to expand by the heated steam. This piston was kept air-tight by a small stream of water continually falling on its upper surface; but in Mr. Watt's engine he was compelled to make the piston fit tight without any other media than the oil that was employed to lubricate it.
The piston is now cast with a projecting rim at bottom, which is fitted as accurately as possible; the part above the rim being about four inches less than the cylinder, thus leaving a circular groove for the hemp which forms the packing. To keep this in its place, a lid or cover is put over the top of the piston, with a projection which enters into the circular groove for the packing, and pressing upon it, the plate is forced down by screws, which work into the body of the piston. By this means the packing is made to fill the internal part of the cylinder with tolerable accuracy, and thus prevents for a time any steam passing between the piston and the cylinder. When, however, by continued working, the packing ceases to fit, it occasions a waste of steam, to remedy which, the cylinder cap must be removed, and as this is attended with a considerable degree of trouble to the engine-man, it is seldom attended to till a considerable loss of power has arisen. There are two improvements on the piston, by which this inconvenience is to a certain extent obviated.
In the first, by Mr. Woolfs, each of the screws is furnished with a wheel or nut, and these are all connected together by means of a central wheel, working loose upon the piston-rod in such a manner, that if any one of the screws be turned, a similar motion is given to the remainder.
In a piston thus constructed, there is little difficulty in drawing down the packing, by applying a key to the square head of the projecting screw, employed to communicate with the rest: the key-hole being afterwards closed by a cap.
The second contrivance is by Mr. Barton, a diagram of which, accompanied by a piston as it is usually constructed, is shewn beneath.
In the first piston, the screws i i are made to compress the packing h h, by acting upon the plate n n, the piston-rod r being firmly attached by the nut c.