From the time of the closed top or soon thereafter, some furnacemen claimed that the gas should be taken off either from the center of the furnace or at several points symmetrically arranged around the furnace top. Others claimed that the gas, having gone through the stock column, had done its work and that the location of the outlet was a matter of no importance, and that one large outlet located on one side was cheaper and simpler than any other arrangement and just as good.

In connection with furnace tops, I have shown a design of top arranged to take the gas out at the center, which worked successfully for many years and gave but little trouble. It was, however, impossible to make the top quite so simple in this way as with the outlet on one side, and so the great majority of furnaces built in the United States up to twenty years ago were constructed on the latter plan, although in discussing operation I shall show what is to my mind an indisputable proof that a single side outlet produces an unsymmetrical passage of the gas through the furnace and cannot lead to the best work.

As furnaces and their outputs grew they produced a vastly larger quantity of gas, so their designers provided more gas outlets, two and even up to five or six. Good practice now very generally uses four, though in my judgment three properly located are sufficient.

Downcomers

This introduction of multiple outlets introduced other difficulties into the design of the plant.

With a single outlet the downcomer could drop almost vertically to the ground close to the base of the furnace, but with this growth of output, necessitating multiple outlets, came the vast increase of dust which I have mentioned and the necessity for providing some means for its removal. As long as the gas traveled on a steep downward slope the dust would accompany it, but when the dust reached the horizontal passages leading to the stoves and boilers, and from these to the stack, it settled in them and obstructed them so rapidly as to constitute a serious obstacle to continuous and satisfactory operation.

Means therefore had to be provided for the removal of the dust at the foot of the downcomer and for this purpose chambers, at first only a little larger than the gas mains were introduced, but as conditions grew worse and worse these were increased in size and required therefore to be placed further from the furnace in order to give the necessary room for operation.

This threw the foot of the downcomer away from the furnace at its base, while the necessity of reaching more than one outlet threw its top back from the closest point on the top of the furnace shell, to the center line of the shell when two outlets were used, and finally to the back or opposite side when four or more were used. These two factors flattened the slope of the downcomer from almost vertical to a slope well below 45°.

In some of the earlier installations these conditions were made still worse because it was considered necessary to have the downcomer enter the top of the dust catcher in the center of the latter and this reduced the vertical height at the same time all the other factors above mentioned increased its horizontal reach. I think that the slope of the dust catcher on some of the plants built about the time of the Duquesne revolution was not more than 30°. These were extremely unsatisfactory in operation because not only would dust lodge in them but also coke and other materials from the charge; they did not free themselves under any circumstances and no matter how large they might be built, they filled up until the velocity of the gas through them was great enough to scour a passageway through the accumulation on the bottom slope, this of course involving high top pressure and consequent leakage on every joint not tightly sealed.