The interesting discovery of the want of proportion between the temperature and elasticity of steam is due to our countryman Perkins, whose experiments to prove it are given in the Franklin Journal, vol. iii. To this cause he ascribes the tremendous explosions that suddenly take place as well in low, as in high pressure boilers. These explosions having taken place when no apparent reason could be assigned for them, have been ascribed to the formation and inflammation of hydrogen gas in the boiler. But this opinion is contrary to true chemistry. On this subject professor Hare observes, that red hot iron may decompose the steam slowly, but it can do so only by absorbing oxygen, which must lessen the quantity of elastic matter in the boiler, and render the evolved hydrogen gas inert, as by itself it cannot explode.% Dr. Jones adds, that the hydrogen would be altogether incapable of producing an explo sion without admitting atmospheric air, which air must find its way into a vessel filled with vapour in a state of tension, and exciting a pressure outwards much greater than that of the external air.i• Arago admits the possible formation of hydrogen gas from the contact of steam with metal heated to redness, mixed with steam; this gas will pass into the cylinder of the engine, and not being capable of con densation, will be thence expelled at a great expense of power. He can allude here only to condensing engines. Upon the supposition that the oxygen, an admixture of which with the hydrogen is necessary to produce an explosion, would be furnished by the at mospheric air in the water of the boiler, he answers, that this water is warm, and therefore contains but a small quantity of air; and this air, as fast as it is dis engaged from the water, passes in a state of mixture with the steam into the engine. Farther, the oxygen of this air would combine with the heated metal more readily than that contained in the water would do; and thus, if hydrogen gas should be produced from the water, it would find nitrogen present with which to mix, but no oxygen.I In connexion with the subject of explosions refer red to the overheating of the boiler, through the defi ciency of water, we may mention a paper in a late number of the American Journal of Science (vol. XIX. p. 294), by Mr. Walter R. Johnson, in which he de tails several series of experiments undertaken ex pressly to ascertain what effect might be produced by a known quantity of metal heated to redness, and then plunged below the surface of water at boiling tempe rature. As the experiments were made with all clue precautions to avoid loss of temperature by the water during '..he process, to determine the weight of steam produced, and its relation to that of the metal em ployed, and to expend the whole of the surplus heat of the latter, above boiling temperature, solely in giv ing the elastic form, there cannot be a doubt that they furnish very accurate practical data by which to judge of the effect of an overheated boiler, when sud denly supplied with hot water.
These experiments demonstrate, that at a dull red heat in day light, each pound of wrought iron is ca pable of generating one ninth of a pound of steam; in other words, that nine pounds of metal in that state, are sufficient to give the elastic form to one pound of boiling water under atmospheric pressure. Of cast iron, it appears that pounds are adequate to the same effect.
The lower arch only of a cylindrical boiler 20 feet long, 30 inches in diameter, made of e inch rolled iron, would, together with a returning flue one foot in diameter, and of the same thickness, weigh not far from 1500 pounds, and would consequently, if made red hot, be capable of producing more than 160 pounds of steam, with a bulk of upwards of 4400 cu bic feet—more than fifty times the contents of such a boiler—and all this tremendous energy would be de veloped in a time not exceeding thirty seconds! Mr. Thomas Earl has also given a neat demonstra
tion on this subject, in the Franklin Journal, vol. vii. p. 156, and in accordance with the above theory.
2. Obstructions to the Escape of Steam.
The obstruction to the escape of the steam from the causes noted, are so intimately connected with the safety of the boiler, as scarcely to require any remarks to enforce the propriety of preventing it from taking place, by constant attention to the safety-valve. The only instance of the adhesion of the safety valve to its seat, that has been recorded in the United States, is one that is said to have taken place in the North River, New York, and of which the following account has been published by Mr. J. B. Calhoun.
" In the summer of 1829 I was engineer on board the steam-boat Legislator, of Hudson; and noticing that the engine was working faster than common, and not seeing any steam flow as usual from the safety valve, I went to the fire-room, and was told by the fire man that he had on twenty-one inches of steam, and that the was up against the As the safety-valve was loaded to carry only sixteen inches, I became alarmed, and taking hold of a cord that ran over a pulley, and was attached to the lever of the safety-valve, I attempted to raise the valve, but could not; on going to the top of the boiler where the safety-valve was, I found all right there; that is, there was no extra weight on the valve. I then slided the weight into the length of the lever, up to the fulcrum, where the weight was merely nominal, still the valve did not rise; I became confounded, and taking hold of the lever lifted on it pretty stoutly, and continued lifting for some seconds, when all of a sudden, with an explosion like that of a small field-piece, the valve opened, and the steam rushed out violently for some time before it got down to the usual pressure, the en gine being at work all the time. There was no water on the valve, nor any visible obstruction to its rising of its own accord after the steam got beyond the pres sure of sixteen inches, which it had invariably done before; the valve and the valve-seat were both of the same metal. For many reasons I had not placed full reliance in the mercurial steam-guagc, but had always entire confidence in the correctness and safety of the safety-valves; but in this case I was deceived, and per haps in a few moments more an explosion might have taken place, for I had no doubt, if a small rod in the steam-guage could have had a free passage through the boiler deck, it would have denoted thirty instead of sixteen inches.
" It is usual to have the steam-guage so graduated, as to show as many inches of steam as the engine will take, and to have the safety valve loaded so as to agree with the steam-guage, believing that when the steam guage indicated sixteen inches of steam, all the surplus steam would escape through the safety valve. Many engineers do not blow off any steam when the boat stops to make a landing, but depend wholly on the safety valve rising of itself, after the steam has risen a little above its required height. This has been con sidered a safe way of proceeding; but the case above stated, shows that it is wrong to depend too much on the safety valve." Mr. C. recommends that when a boat stops to land passengers, the safety-valve be raised, let the guage indicate what pressure it may. Before tne occurrence described, he believed the cause of boilers exploding was almost invariably the want of a sufficient quantity of water. He now thinks, that some explosions may be attributed to the being de ceived by the safety-valve not rising as was expected by the engineer.* The two following cases illustrate the necessity of attending to the proper form and construction of the safety-valve.