In the third decade of the nineteenth century James Beaumont Neilson, who from plain beginnings had made for himself a distinguished position as a gas engineer and authority on matters of combustion, was consulted by a furnaceman as to the reason for the very unsatisfactory work of his furnace, particularly in summer.

This started Neilson on a long investigation. Among other things he experimented with the use of preheated air on gas burners and on blacksmiths' fires and found that the effect was a marked increase in the brilliancy and intensity of the combustion. In 1828 he took out a patent on the use of preheated air in the blast-furnace, but even after surrendering two-thirds of his patent to furnace owners he had the greatest difficulty in obtaining a chance to make a trial of the process.

Finally after several years of effort and after overcoming the most factitious and unreasonable objections to changes in insignificant details such as the shape of the blast pipe, he made a demonstration and, while the increase in blast temperature was only 80° F., the results were so obvious that the system was at once adopted and higher temperatures applied.

Then he was at once thrown into another difficulty - that of preventing the overthrow of his patent by a combination of Scotch furnace owners who took a pledge to break his patent virtually "by legal means or otherwise" in order to avoid the payment of his moderate royalty of 1s. per ton, though the saving effected amounted to 32s. per ton.

It is rather a relief to realize that such piracy, which we have been taught to regard as the product of our own times, flourished even in those "good old days".

It is still more pleasant to know that after innumerable trials which went thrice to the House of Lords, Neilson and his associates won, and that on the royalties he finally received he was enabled to pass the closing years of his life in comfort and leisure devoted very largely to what would now be called welfare work.

The results of his invention upon the operation of the furnace were probably very much beyond Neilson's most sanguine expectations, and within a few years the warm blast was used in producing by far the largest proportion of all the iron made. The reason for this saving so disproportionate to the amount of heat introduced by the warm blast will be explained in a chapter on the thermal action of the furnace. We have here only to deal with the apparatus used for heating the air.

The first apparatus used was in all cases a nest of pipes of some sort in a masonry chamber, first fired independently with coal, later with waste gas from the furnace itself. The gas burned around the pipes while the blast on its way to the furnace passed through them. In early days these stoves were set on the top of the furnace, or if not immediately on top of it at the level of the top and just to one side, as has briefly been described in connection with boilers, so that the burning gas from the open furnace top could be drawn into them. This had the disadvantage that the cold blast had to be taken to the top of the furnace and the hot blast brought back to the bottom with the loss of heat consequent upon that journey, and as soon as the closed top was introduced the stoves were set at the ground level, with great advantages in first cost and in convenience.

The temperature to which the blast could be heated in these stoves was strictly limited by the fact that the pipes constituting the heating surface were rapidly destroyed if a certain temperature, found by experience to be about 900° F., was exceeded by even a small margin, and if the stoves were fired while no blast was passing through them to cool them off, they might be completely destroyed in a period of an hour or two.

To meet this condition the late Sir William Whitwell conceived the idea of applying to this service the regenerator previously invented by Sir William Siemens, which consisted of a pair of chambers filled with brick work, through which ran a great number of passageways arranged to present as much surface as possible.

Through these passage ways of one chamber the hot waste gases of combustion were passed for an interval sufficient to heat the bricks to a high temperature; then the gas was cut off from this and thrown into the second chamber, and the air to support combustion was drawn through the first chamber, from which it absorbed heat by direct contact with the hot surfaces of the bricks. When a sufficient amount of air had passed to lower the temperature of the bricks below that which would impart the required temperature to the air, the gas currents were again reversed in relation to the chambers, the gas being put back through the first one and the air currents being drawn through the second.

For melting crucible steel, the first purpose to which Sir William Siemens applied his invention, the hot air for combustion was drawn through the checker work under the action of the stack draft, the pressure within the checkers, therefore, differed by only a few inches of water from that of the atmosphere, and masonry construction throughout was sufficient for tightness under this difference of pressure.

In order to apply this system to the blast-furnace, however, the chambers had to be so constructed as to withstand the pressure of the air after compression to the pressure required for the blast-furnace, because obviously the blowing engines could not handle blast heated to a temperature of 1000° F. or so. Therefore, the blast must be compressed first, and after heating must go directly to the furnace.

For this reason Sir William Whitwell built his stoves in the form of cylindrical chambers with a top either cylindrical or dome-shaped to prevent deformation under pressure. The bottom was flat because prevented from deforming by the weight of brickwork upon it. The shell of these chambers was of iron (now steel) plate, riveted together air-tight and strong enough to resist the maximum blast-pressure expected.

There was also another change in the conditions as compared with those of the steel-melting furnace. There the gas to do the heating had already been burned and was simply hot waste gas. The waste gases of the blast-furnace, on the other hand, are not particularly hot, but are highly combustible, and so must be burnt for use in the regenerator and in order to provide a gas burner and combustion chamber these firebrick stoves required, therefore, four connections. First an opening for the gas burner, second an opening to the draft stack, third a connection for the cold blast from the blowing engine, fourth a connection for the hot blast to pass to the tuyeres of the furnace.

All of these openings were required to be alternately closed air-tight and opened at frequent intervals, and must retain their tightness under conditions of temperature injurious to any metal available for construction. This meant that all metal parts around these openings had very largely to be water-cooled. The brick stove, therefore, brought many annoyances and difficulties of operation in its train, and, while it gives blast temperatures about 50 per cent. higher than those permitted with the iron-pipe stove, its first cost is greater, and this with the inconveniences and difficulties of operation and maintenance resulting from the conditions described above, retarded its development somewhat, so that a few plants have continued to use the iron-pipe stove, but except in charcoal practice they are rapidly disappearing.

With this general view of the situation we may proceed to describe the two classes of stoves.