The value of d being calculated thus, we can deduce that of x, and we have for the weight of air 4·33x, and for that of the nitrogen 3.33x. But this supposes the air of the blast to be perfectly dry, whilst in reality it is always more or less charged with moisture, which adds, of course, the oxygen of the water to that of the dry air. 4 The equation y+ 8 11 myd+x, gives therefore for x 7 the oxygen of the air and its moisture united; but it is easy to calculate the values of dry air and nitrogen. It is well known that at the mean temperature of 12° to 13° C = (54° to 56° F), the humidity amounts to about 0-0062 of the weight of dry air, and the oxygen contained Thus if we put z to represent the oxygen derived from dry air and 4.33% that of dry air itself, we have— In reference to this indirect determination of the mass of air blown through the twyers, it is worth observing, that though it is not rigorously correct, as it depends on a sort of theoretical analysis of the pig-iron, it is nevertheless much more exact than any other mode. It may, of course, be rendered as exact as possible, by ascertaining by proper chemical analysis what is the true composition of the pigiron.* § 8. Agreement of the formulas with the complete analysis.— We may conclude from what precedes, that to get the composition of the escaping gases, it is sufficient to determine by experiment the ratio This allows of our calculat CO2 = m. ing exactly the quantities of CO2 and CO, and very approximately the proportion of nitrogen: and we deduce besides, with the same degree of precision, the weight of blast. All that is neglected is the hydrogen. But in blast-furnaces using coke, the influence of hydrogen is almost nothing. The hydrogen can only come from two sources, the moisture of the air, and the hydro-carburetted gas, which remains in * The other methods of determining the mass of air are the three following: The first consists in assuming the ideal working of the furnaces-that is to say, in supposing that all the carbon of the coke reaches the twyers intact, and then calculating the oxygen corresponding. But we have seen above (Sec. 5) that even in the case of good working the carbon reaching the twyers may be 16 per cent. less than the total carbon of the coke, which, of course, involves an erroneous estimate of the air. The second method, based on the formula for the outflow of air by a conical mouthpiece, gives a still higher figure. This formula, habitually used as given by d'Aubuisson, Karsten, and Weislarch, gives the volume of air which has no other resistance to overcome than the counter-pressure of the atmosphere; whilst, in fact, the blast meets in the furnace a resistance greatly above this, and which no experiments can determine. Besides this, all the air does not pass by the eye of the twyers; a quite notable fraction is thrown back outside. The third method, that based on the volume passed through by the piston of the blowing cylinder, is quite uncertain, because neither the losses by the quantity thrown back, above referred to, nor the losses by the ports and valves, and in the hot-blast stones, and from other imperfections, can be determined. even the best made coke—but this hydrogen is very little, and acts with the CO in reducing the oxide of iron. It also becomes transformed into steam in the upper part of the furnace. Only a small fraction escapes with the gases without being oxidized. Mr. Bell never found more than 0.001 of hydrogen in the gases of Cleveland and Ebelmen, at most, 0·0013 to 0-0018 in the three works of Vienne, PontEvêque, and Seraing. This small proportion has no influence on the working of the furnaces, and may certainly be neglected in the practical calculations with which we are now engaged. There is no question at present of blast-furnaces working with raw coal. Let us now show how well the formulas we have found correspond with the results found by Mr. Lowthian Bell by two examples: Let us take, first, the small furnace of Clarence Works of 1853 (see § 2), the height of which is 48 feet, and the internal capacity 6000 cubic feet. The consumption per ton of iron is 1.45 tons of coke, or 1.318 of pure carbon, and 6.83 of flux. We have in the formulas of § 17 On the other hand, the analysis of the gases gave z = 0.9767 × 1·531 = 1lb .495 and the nitrogen = 3.33 x 1.495 4.978, = which gives as the weight of gas per lb. of pig yielded Weight of dry gases 81b.571. and Mr. Bell's complete analysis gave 1lb .002 2lb -591 4b893, a difference of 0.085, or less than 2% 8lb .486. Let us now take, as a second example, the blast furnace of 1866 at Clarence Works, height 80 feet, capacity 11,500 cubic feet. The consumption per lb. of pig is 1.125 lb. of coke = 1.020 of pure carbon, and 0.683 of flux and limestone. α = 1lb .020 therefore b0lb 12 × 0.683 = 0.082 p = 11 ·020 + 0·082 — 0·03 — 1·072. 1.740 x= ·(44+56 × 0·6865)—0·642=1·863 — 0·642—11b.221 77 2= 0.9767 × 1.221 = 1lb.192 and N = 3.33 × 1·192 3lb .969. = which gives the following as the quantities of gas per lb. of we see thus that the agreement of the formulas is as exact as possible, and indeed, by pure chance, greater than we had |