WATER GAS
721
a rate of about 45 1b. per minute. It will be found that the “ down run ” supply can be adjusted to give a rather greater flow than. the up steam, rougbly about 2 1b. per minute more.
Steam meters, as used on water-gas plants, are calibrated to read pounds of steam passing per minute. There are many types, most of which are constructed to read direct on a dial, or to give an autographic record. Kent’s steam meter (Fig. 447) is based upon the principle that steam flowing through a constriction in a pipe falls in pressure, the pressure drop being approximately proportional to the square of the velocity of flow. The constriction consists of an orifi.ee plate which is bolted between two flanges, the size of the orifice being such that the desired pressure drop at the maximum discharge is obtained. The pressures from the inlet and outlet sides of the orifice are conveyed through copper pipes to the two portions 'of the diaphragm, the movement of which corresponds to the differential pressure created by the steam flow. The diaphragm movement is then transmitted through a gland to the indicating mechanism. The loss of pressure due to the insertion of the orifice amounts to about -j 1b. per square inch at maximum flow.
Levelling the Fuel-Bed
If the best results are to be obtained the surface of the fuel-bed should be levelied as far as possible. Owing to the faet that the charging door for coke is invariably situated above the centre of the generator the fuel tends to lieap up in a conical mound, the apex of which is just beneatli the point of charging. Con-sideration of the isothermals in the fuel-bed, however, will show that they are of a concave form with the lowest point of the curve in the centre of the bed, and the
higher extremities reaching up towards Fig. 448.—Fuel-bbd Spbeadbk. the periphery of the fuel. The surface
of the fuel should, therefore, conform to the line of isothermals. That is to say, the shape at the surface should be concave rather than convex. In order to overcome the drawbacks resulting from the heaping of the coke towards the centre, various forms of meclianical spreaders have from time to time been experimented. with. Thuman 1 proposes to preclude heaping of the fuel-bed by means of the apparatus shown in Fig. 448. It will be noted that distribution around the periphery of the
1 B.P. 7532/18, and 15537/17.
3 A