It has been recognized for many years that firebrick exposed to the conditions in the blast furnace would wear back until the amount of heat carried off by their bases about equaled the amount brought in by their inside surface, after which wear took place much more slowly. It has also been recognized that in the standard construction satisfactory life of lining can only be secured by a corresponding thickness, and that as this thickness wore away the departure of the furnace from its original and presumably correct lines became greater and greater, generally with more and more unsatisfactory working.

About fifteen years ago with the idea of overcoming these difficulties the experiment was tried in Germany of making the lining of the fur-nace exceedingly thin, only a few inches in fact, and cooling its exterior surface so as to maintain its original shape indefinitely. The idea has superficially much to recommend it, and after several years' trial in Germany it was introduced into this country, first at the Lucy Furnaces in Pittsburg, under the management of Mr. James Scott.

The German construction consisted of cast-iron staves with flanges on their exterior surface and jaws like flanges on the inside, into which the bricks were securely fastened. These were built up into a furnace shaft very much in the same way that the bosh -jacket shown in Fig. 166 is built up, but the exterior was cooled by a water film or spray instead of the pipes cast in shown in Fig. 166.

The construction adopted at the Lucy Furnaces was different. There a steel shell was built much smaller in size in proportion to the lines of the furnace than the ordinary shell, since the thickness of the brick with which it was lined was only about one foot, from top to bottom. On the outside of this shell were riveted annular troughs, each about two feet deep by six or eight inches radial width, and arranged so that water pumped into the top one flowed down through all the others in sequence to the bottom. In spite of great pains to secure absolutely air-tight construction, some water leaked into the furnace through the rivet holes and the results wore unsatisfactory, though subsequently these difficulties were overcome.

Soon after this two thin-lined furnaces of this type were built at the Illinois Steel Works, and subsequently one of these was rebuilt with cast-steel plates instead of the boiler-plate shell and riveted buckets. This, however, did not prove satisfactory and the latter construction was restored. A bucket type of thin-lined furnace was also constructed at the plant of the American Steel & Wire Company in Cleveland.

A thin-lined furnace was subsequently built by the Detroit Iron & Steel Company according to the plans of their general manager, Mr. Max McMurray. This solved the problem of saving the original shell for supporting the top works of the furnace, which had not been done in any of the other constructions, columns having been run all the way up from the mantle to the top for this purpose.

In the case of the Detroit furnace the expedient was adopted of building a smaller shell with a thin lining inside the original shell and cutting a number of doors through to give access to the inner shell for examination, this shell is cooled directly by water sprays at the top, these form a film of water which runs all the way to the bottom. The outside shell also presents the advantage of preventing the wind from blowing the water off the stack, thus leaving it dry on one side, and making a disagreeable spatter below, both very important points. This construction is shown by Fig. 171 (page 264).

The Tennessee Coal & Iron Company has also had a thin-lined furnace in operation for a number of years. This has a very heavy boiler-plate shell with spray pipes arranged around the top. A series of galleries are arranged at different levels around this shell and outside these galleries is a sort of housing of steel slats to prevent spatter and to prevent the wind from blowing the water off the shell.

A different construction still has been used by Mr. Frank C. Roberts in a furnace built by him for the Warwick Iron Company, shown in vertical section in Fig. 172. In this the top of the furnace is supported by separate columns running up from the mantle, as previously described, and the shell is a boilerplate shell, but instead of trusting somewhat to chance in the matter of the distribution of the cooling water around it, this is accomplished by a helical trough shown in larger section in Fig. 173 (page 266).

The trough is supported by tap bolts which do not go all the way through the shell and therefore make leakage through it impossible. The bottom edge of the trough is spaced away from the shell by distance blocks as shown. Part of the water runs out the space at the bottom of the trough and that which cannot get out this way runs around the helical trough. In this way a perfect distribution of water is secured around the circumference of the shell and the danger of wind blowing the shell dry on one side is eliminated.

A compromise, one-half thin-lined and one-half standard construction, has been in use for several years at the Port Henry Furnace of the Northern Iron Company. This construction shown in Fig. 174 (page 267) is that of the former manager of the company, Mr. F. E. Backman, the design being carried out by Frank C. Roberts & Co. This furnace has a circular mantle shown in section, resting on the vertical portion of the columns at the level of the top of the bustle pipe. From this other columns run up on a slight inward batter to a second mantle some twenty-five feet higher. On this second mantle rests the upper portion of the inwall which is of standard construction. From this point down to the bosh there is a boiler-plate shell with rings of angles for supporting the brickwork, riveted to its inside surface at intervals of five or six feet. This shell makes a joint on top of the cast-iron water-cooled segmental bosh jacket. The bosh jacket is lined with nine inches of brick and the nearly vertical portion of the shaft above it with thirteen inches. This construction is intended to prevent the gouging out of the lining above the top of the bosh which has been described above. This furnace has now been in operation for several years and is said to have given good satisfaction.