Steam:
Its Generation and Use

to demonstrate their unreliability in this particular. No. i2.—An attempt to avoid this diffi- culty and increase the heating surface in a given space. The tubes were expanded into both sides of wrought-iron boxes, openings being made in them for the admission of water and the exit of steam. Fire-tubes were placed inside these tubes No. 12. to increase the surface. These were aban- doned because they quickly stopped up with scale, and could not be cleaned. No. 13.—Water boxes formed of cast-iron of the full width and height of the bank of tubes were made of a single casting, which were bolted to the steam water-drum above. No. 14.—A wrought-iron box was substituted for the cast-iron. In this, stays were necessary car wheel metal; the headers^ having a sinuous form so that and were found, as is always the case, to be an element to be avoided wherever possible. It was, how- ever, an improvement on No. 6. A slanting bridge wall underneath the drum was introduced to throw a larger portion of its surface into the first combustion chamber above the bank of tubes. This was found to be of no special benefit, and difficult to keep in good order. No. 15.—Each vertical row of tubes was expanded at each encl — into a continuous header, cast of they would lie close together and admit of a staggered position of the tubes in the furnace. This form of header has been found to be the best for all purposes, and has not since been materially changed. The drum was supported by girders resting on the brick-work. Bolted joints were discarded, with the exception of those connecting the headers to the front and rear end of the drum and the bottom of the rear header to the mud-drum. But even these bolted joints were found objectionable and were superseded in subsequent constructions No. 13. by pieces of tube expanded into bored holes. 31