ON FURNA CES. 397 



up the entering gas and air, from the temperature of the atmosphere, 

 to a temperature nearly equal to that at which the spent flame leaves 

 the working chamber ; and the effect of sending in hot gas instead of 

 cold gas, is not that the gas is much hotter when it reaches the top of 

 the regenerator, but simply that the bottom of the regenerator, where 

 the gas enters, is less cooled down ; and, on reversing, the burned 

 gases, after traversing it, escape at a higher temperature to the chim- 

 ney. In practice, the heat of the gas, as it leaves the producer, is in 

 most cases purposely thrown away, by leading it through overhead 

 sheet iron " cooling tubes," in order to obtain a better pressure, at the 

 furnace, from the syphon action between the ascending hot column 

 and the descending heavier cool column, and to remove the greater 

 part of the vapour of water that it generally contains. 



Again, in the Ponsard furnace, as has been pointed out, when only 

 the air, to burn the gas in the working chamber, is heated by the 

 waste flame, the quantity of heat carried into the regenerator is about 

 twice as great as is required to heat this amount of air ; and however 

 perfect the action of the regenerator may be, however great may be 

 its extent of surface, the burned gases necessarily escape to the 

 chimney at a high temperature. 



If, on the contrary, the air to supply the gas producer, as well as 

 that to burn the gas, were heated by the waste heat ; and the hot gas 

 from the producer were led direct, without passing through a regene- 

 rator, to the working chamber, and burned there ; the volume of air to 

 be heated would be sufficient to take up all the available heat of the 

 waste flame, and the heat that is otherwise inevitably lost, either by 

 the chimney in the one case, or from the cooling tube in the other, 

 would be rendered available. 



The system of burning powdered fuel, that has been worked out by 

 Mr. Crampton, is another remarkable deviation from the ordinary form 

 of flame furnace. 



A model of such a furnace, arranged for mechanical puddling, is on the 

 table. The coal, burned, is first ground between ordinary millstones to 

 such fineness that it will pass through a sieve with thirty holes to the 

 linear inch (which Mr. Crampton estimates may be done, on the large 

 scale, at a total cost of less than a shilling a ton) ; and the powdered coal 

 is supplied, at any required rate, by a mechanical feeding arrangement 



