The proposal has been made by several blast-furnace engineers ana operators to apply an equalizer to remove the objectionable peaks from the charts of blast temperature, and several of these apparatus have been built. One of the few in this country was that built by John S. Kennedy and the late John M. Hartmann at the Musconetcong furnace, Stanhope, N.J., which has been most successful in operation.

This apparatus is practically a small stove without combustion chamber or valves, filled with very small checker work through which the blast passes on its way from the stoves to the furnace. When the blast is very hot from putting on a fresh stove the excess heat above the average is given up to the checker work in which it remains stored until the temperature of the blast falls below the average, owing to the cooling of the stove, when this stored surplus flows out and makes up the deficiency.

Mr. Kennedy has told me that this apparatus smoothed down the "saw teeth" on the blast temperature charts so that the maximum temperature difference for long periods continuously was less than twenty-five degrees. Without such an apparatus the drop in temperature between putting on and taking off a stove will run from 50° to 150°, depending on the number of stoves and how they are handled.

Another method for obtaining a uniform blast temperature is by the use of a cold-wind mixing valve. This valve is located on the hot-blast main into which it opens and is connected with a branch from the cold-blast main. By opening this valve cold blast direct from the engine can be admitted into the hot-blast main, thus cutting down the temperature of the blast by admitting a considerable quantity of this cold air as a fresh stove is put on the furnace and gradually diminishing this quantity down to nothing, at or near the time when the stove comes off. By this means the blast temperature may be maintained uniform, but this method is open to two objections: First, that the uniform blast temperature is, in that case, the lowest temperature furnished by the stoves instead of the average, and, second, the equalization depends wholly on the skill and fidelity of the stove tender, and if done properly requires such frequent attention by him as to militate against his assisting at other work.

Attempts have been made to provide an automatic control of this valve, which would relieve the stove tender of the duty of regulating it and would give perfect equalization of the blast temperature, but so far as known to me none of the apparatus designed for this purpose is now in regular commercial use.