Practice in regard to these matters in the past has been based on general experience and the belief of the furnace man or engineer that stoves of about such a size would be sufficient to supply the hot blast for a given furnace. It is not a rule but it seems to be according to general practice that the diameter of the stove shall be about the same as the bosh diameter of the furnace, and that its height shall be as a minimum that of the furnace and from that up to twenty or twenty-five per cent. more.

Until recently but little attention was paid to the quantity of heating surface furnished per foot of stove height, but there is evidence that this condition is passing and that within a few years stoves will be bought on their heating surface, practically as boilers are to-day. This has been impossible except in the most general way in American practice until now, because the hard driving of American furnaces, combined with the enormous percentage of extremely fine ore used at seventy-five per cent. of them, has made an exceedingly dirty gas, and this dirt depositing on the checkers in the way already described has reduced the efficiency of the stove in proportion to the extent in which it has been allowed to accumulate, so that stoves kept clean and under good management at one plant are capable of heating a given quantity of blast to a much higher temperature than similar stoves at another plant where they are not properly cleaned.

To a large extent these conditions have been past control and absolutely past all calculation. General experience has, therefore, been practically the only guide. With the introduction of primary cleaning for all the gas, that which goes to the stoves will be in vastly better condition than it is now, the stoves will stay clean for far longer periods and will have not only a higher but a more definite efficiency.

The advantages of smaller checkers already pointed out, which become available under those conditions, will cause competition among stove builders to give the maximum amount of heating surface per stove, and we shall gradually reach the place where the capacity and economy of stoves are as well understood as that of any other apparatus around the plant, a condition which certainly does not exist at the present time.

There has been in recent English and American literature little or nothing on the hot-blast stove from the operating point of view, but in December, 1912, Mr. A. N. Diehl published before the Engineers' Society of Western Pennsylvania a very valuable paper giving the results of long and careful investigation of the subject, and in October, 1913, in a discussion of the subject of gas cleaning before the American Institute of Mining Engineers he contributed additional data on the subject.

His tables I and II (pages 230 and 231) from the first-mentioned paper are reproduced here and will repay careful study.

With clean gas the efficiency of the stove is raised considerably, according to Mr. Diehl's later figures, from 56 per cent. to 64 per cent. This is not a decrease of 8 per cent., but of 12 1/2 per cent. (8/ 64) in the gas consumption of the stove and, what is much more important, it brings with it the possibility of obtaining higher blast temperatures with a given equipment of stoves and by this means reducing the coke consumption of the furnace.

Table I shows the advantage of having a sufficient amount of heating surface for the given conditions, and as the use of clean gas makes permissible the use of smaller checkers and thereby increases the heating surface, it may be said that the effect of clean gas is cumulative in its influence on stove design and operation. The best practice as brought out by Mr. Diehl's paper and its discussion favors the use of not less than four and preferably five square feet of heating surface per cubic foot of blast blown per minute measured by piston displacement. When four stoves are used this would mean from one to one and a quarter square feet of heating surface per stove for each cubic foot of blast blown per minute.

This will obviously vary to some extent with the conditions. When high heats are regularly required, as with the use of lean and refractory ores, more heating surface should be provided than with soft and reducible ores which ordinarily do not need such high heats. On the other hand, when there is no use for the surplus gas it may be better to save in the size of the stoves and burn the gas less economically in them to get the desired heat.

Table No. 1 Excerpt From Stove Tests

Note: Side Combustion Stove as above, volume of brickwork is for checkers only. Center Combustion Stoves, Duquesne. Side Combustion Stoves, Foreign.

Table No. 2. Showing Heat Losses In Stove Practice Due To Inefficient Combustion Of Fuel Gas

* Col. No. 11 - Over and above that required to oxidise the CO present in the flue gas.

+ Col. No. 13 - Per stove per hour and based on coke having a calorific value of 12,200 B. T. U. per lb. as received. + Col. No. 14 - Based on tour stoves per furnace, or 72 stove hours per day on gas and 24 stove hours per day n blast Note: All gas volumes and calorific values are at Standard Conditions, 62 F. and 30 in. Hg. Above test Numbers correspond to those given in Table No. 1.

Since this section was originally written there have appeared accounts of recent experiments in Germany in which the heat imparted to the stove was greatly increased by increasing the velocity of the gas through it, first by an air jet in the combustion chamber and in the later experiment by use of a fan to increase the draft. It is claimed that by the use of this system so much heat can be stored in each stove that only two are needed per furnace.

The accounts are too incomplete, and it is too early in the development to say what the final result will be, but with the vast improvement made in the operating conditions of stoves by clean gas there is no doubt that many improvements will follow and that increasing the velocity of the gas and blast through the checkers is one of the best ways to increase the rate of heat transmission.

As the efficiency of stoves increases it is very likely that fans will be used to provide draft for them for the reason that high efficiency can only come with low stack temperatures. Chimneys only provide good draft with fairly high temperatures and with a consequent waste of much heat. Therefore, it seems natural to expect that this conflict of requirements will be solved by the use of a fan whose performance becomes increasingly better as the temperature falls and which gives absolute control of conditions of draft, rate of combustion, etc.

The limit of possibility in reducing stove capacity is set by the amount of heat it is possible to store in a given weight of brickwork with a reasonable increase in its temperature, for if too great a drop in the temperature of the brickwork be necessary more gas is required and this limitation must be carefully considered in any plan for reducing the stove capacity.

Within the last year or two where clean gas is available stove heating surface is increased by making the side of the checker bricks wavy instead of straight. No detailed results of this practice are available.

Three stoves with clean gas and high gas velocity may easily be better than four without those advantages, but whether it will ever pay to reduce the number of stoves below three may well be doubted until overwhelming proof is offered.