The system of handling raw materials in barrows was almost universal until the middle of the decade beginning in 1891. Many attempts had been made to eliminate at least a portion of the labor required for handling the great quantity of material needed as furnace outputs increased. But these were either mechanically unsatisfactory, or else their effect upon the work of the furnace was detrimental, in many cases disastrous. The reasons for this, in so far as they are known, will be given at some length in describing operation; suffice it for our present purpose to say that probably ninety-five per cent. of all the iron made for thirty or forty years previous to the year mentioned was made from raw materials handled in filling barrows. These are so widely known that it seems almost unnecessary to describe them, but for logical completeness a brief description may not be amiss.

These barrows are universally two-wheeled, with at least one of the wheels loose on its axle to enable them to turn corners easily, and provided with stout legs on the rear end, which hold them level when standing. They are dumped by tipping forward on the axle, the wheels remaining stationary when they come against the dumping ring on the top of the furnace. On account of the vast difference in the specific gravity of coke, and ore, or limestone, a different type of barrow is generally used for handling these two kinds of materials. The ore and limestone barrows are as shown in Fig. 4, and the coke barrow as shown in Fig. 5. The latter are much larger and provided with a door in front, which is held shut by a latch, this latch being released from the rear a moment before the barrow is dumped. This construction is used in order to obtain a reasonable capacity for so light a material as coke. These barrows are built of steel plate, very substantially riveted together throughout, as they are subjected to continuous service and particularly to very heavy shocks when they are dumped, the nose of the barrow dropping forward on the iron dumping ring with a hard blow.

Fig. 4. Ore Barrow.

In ordinary practice the barrow weighs about eight hundred pounds and can carry a load of about fifteen hundred pounds of ore, but they are commonly not loaded quite as heavily as this, as they then become too difficult for one man to handle, even on good floors.

The capacity of the coke barrow is roughly one-half as much by weight, so that the same number of barrows of coke as of ore are required per charge, the ore charge weighing normally about twice as much as the coke charge.

The physical labor involved in wheeling the barrows from the stock piles to the foot of the hoist is in ordinary practice roughly the same as filling them by hand. Therefore to dispense with hand shoveling, and continue the use of filling-barrows, is only a half solution of the problem of eliminating manual labor in filling. Moreover, when the barrow system of filling is used it is necessary to have always one, and frequently two or even more men on top of the furnace to take the barrows from the cages to the furnace top, dump them and return them.

Fig. 5. Coke Barrow.

This labor is more expensive, man for man, than the labor at the bottom of the furnace, because it does not pay to have a foreman supervise the relatively small number of men required, and yet the operation of dumping, with which they are entrusted, is of the utmost importance, and with hand-filling, slight variations in the method of dumping, more frequently than any other cause, derange the work of the furnace.

For this reason it is necessary to secure a higher grade of labor for at least one of these top fillers by paying higher wages. Another reason for paying higher wages is that the men are under a certain liability to be overcome by the furnace gas when remaining for a long time on top. It is impossible to make the bell absolutely tight and keep it so all the while. Small particles of stock lodge on its seat and hold it open by an almost microscopic amount until the next dumping, when they are swept off, but very likely replaced by similar particles in some other location, so that the bell seldom shuts absolutely tight.

There are other points at which gas can escape, such as the test rod hole and various other places through which the gas will find its way in quantities frequently too small to be visible, but nevertheless tainting the atmosphere sufficiently to overcome men exposed for a long time to this poison, even in this dilute state. This possibility of accident, though real, is small, fatalities in the ordinary course of filling being almost unknown, nevertheless it is sufficient to necessitate a higher scale of wages.

For all these reasons the elimination of manual handling of the stock from the bins to the furnace top was the goal of furnace builders for many years. Blast-furnace engineers were often sure that they could build a top that would give results as good as those of hand-filling, but managers, knowing that they would be held responsible for the operation and that bad construction would not be an acceptable excuse for bad furnace work, were for a long time very chary of using any method of filling of whose results they could not be absolutely sure in advance, and this for a long time meant hand-filling exclusively.

The Duquesne Revolution

In the middle of the decade beginning 1891, furnace outputs had become as high as 400 tons per day in regular operation. The crews necessary to fill them by hand labor with filling-barrows were tremendous. Then, too, the capacity of the platform hoists had about reached their limit, and it became almost a matter of physical impossibility to fill furnaces for any greater output by this means. Accordingly when about 1895 the Carnegie Steel Company decided to build at Duquesne a plant of four furnaces with a capacity of six hundred tons per day each, these furnaces being about twenty-one feet in diameter by one hundred feet high, they determined to strike out on new lines and eliminate altogether hand labor for filling.

M. A. Neeland was then chief engineer of the Duquesne works, and he designed what has since become one of the standard methods of mechanical filling, commonly known as the bucket system. This was a success mechanically and metallurgically and was the beginning of a revolution in the construction and operation of blast-furnaces. It having been demonstrated that mechanical methods could be made to operate successfully on so large a scale, and with an enormous saving of labor, other methods to achieve these ends were soon devised, and as a result it is probably safe to say that three-fourths of all the iron now produced is made at furnaces mechanically charged.

The plan of eliminating labor wherever possible was carried out in all directions at the Duquesne plant, whose mechanical equipment throughout was the best obtainable at that time. Thus the construction of this plant marks a distinct epoch in the history of the blastfurnace in the United States.