REPORT OF EXPERIMENTS ON THE EXPLOSIONS OF STEAM-BOILERS. 119 A fact noticed during the experiments on fusible alloys was again verified in these experiments; namely, that the mixtures of metals require a considerable time to change their state of solidity or of fluidity; so that in the former case they may be heated above the true temperatures of fluidity, and in the latter case they may be cooled much below this temperature, without solidifying. The alloy used in these experiments appears to have put the apparatus very fully upon its trial in this respect, and the experiments were performed so rapidly as to give a further severe test. On the occasion devoted to the trials when the steam was not urged up with its greatest rapidity, the stem was drawn out at 268°, when more rapidly at 270°, and with the fire at its maximum intensity, when the water raised in temperature 24° in three minutes, the stem was drawn out at 274°. In other experiments it gave way at 256°. The range is 180 Fah., corresponding at ten spheres, to less than two atmospheres, under the test of very severe comparisons. There appears no reason to doubt, that when tested by no more rigid modes than practice would furnish, this apparatus would not only apply as an alarm to prevent undue heating of the parts of the boiler, but as a manageable and useful check, in ordinary cases, upon the safety-valve. Conclusions. was atmo The conclusions deduced from the foregoing experiments on metallic alloys may be thus stated. Ist. The impurities of common lead, tin, and bismuth, are usually not such as to affect materially the fusing points of their alloys. 2d. When mixed in equivalent proportions, tin and lead formed alloys, not presenting the characters of distinct chemical compounds, in definite proportions. The alloys between the range of one equivalent of tin to one of lead, and one equivalent of tin to six of lead, varied considerably in the interval between the temperature of commencing to lose fluidity, and that at which a thermometer, immersed in the solidifying metal, became stationary. These different alloys produced nearly the same stationary temperature in a thermometer plunged into the solidifying metal. 3d. Fusible metal plates covered by a perforated metallie disk, and placed upon a steam-boiler, show signs of Auidity at the disk before the steam has attained the temperature of fusion of the alloy of which the plate is composed. This fluid metal oozes through the perforations in the disk, and the plate thus loses much of its substance before finally giving vent to the steam. 4th. The under parts of the plate are not kept from fusion by a protecting film of oxide there formed. 5th. The thickness of the plate is not important, provided only that it is sufficiently strong to resist the pressure of the steam at temperatures below its point of fusion. 6th. The temperature at which the plates are cast, and the rate of cooling of the cast metal, do not affect the temperature at which the plates give vent to steam. 7th. The effect stated in conclusion 3d is explained by the nature of the alloys used, which are formed of portions of different fluidities; the more fluid parts are forced out by the pressure of the steam, leaving the less fusible. These latter in general are burst, not melted. 8th. By pressure in a receptable provided with small openings, this effect of separating the differently fluid portions of an alloy may be imitated. 9th. Fusible alloys, used to indicate the temperature of any part of a steam-boiler, should not be exposed to the pressure of the steam; at least, not in such a way that the separation of the differently fusible constituents of the alloys may be effected. Fusing Points of Alloys applicable to The Committee next proceed to give the results of their trials to determine metallic alloys proper to be applied to steam boilers. This problem admits, of course, of a great variety of solutions. The metals used were limited to tin, lead, and bismuth; but still different mixtures of these may be made which will give alloys of the same fusing point. The property which was most desirable in these alloys was a small range of temperature in changing from the liquid to the solid state. This property, it will be seen, is difficult to attain, and the less fusible alloys of the first table, as well as the more fusible ones of the third, do not possess it. For the higher temperatures, alloys of lead and tin are applicable, and the question is reduced to an examination of the fusing points of different mixtures. The greater proportions of lead might be inferred to give the higher fusing points, and the less proportions the lower ones. Beginning with the alloy of equal weights of tin and lead, the following table gives fusing points between 855° and 503° Fab. The stationary points were taken as already described; all the alloys in the table, except the first, were hard before the stationary point occurred, and therefore this point indicated, in these cases, some internal change in the solid, and did not correspond to the passage from the liquid to the solid state. This seems not to have occurred to Mr. Parke, who speaks of having taken this point as corresponding to that of congelation. It 120 REPORT OF EXPERIMENTS ON THE EXPLOSIONS OF STEAM-BOILERS. should be observed, however, that his table of alloys shows a variety in the fusing points, which is incompatible with the observations of the Committee, supposing the stationary point to have been taken in each case as the fusing point. The alloys passing gradually from the fluid to the solid state, an attempt was made to seize the more remarkable points, as referred to in the table; but these can only be considered as approximately determined. Direct experiments were, in most cases, made upon the temperature at which the metal refused to allow a metallic stem to be withdrawn. This was the case when the metal from the state of a soft solid, began to acquire hardness. REMARKS. 353+ * Began above this point. + In these three cases the alloy congeals in thin plates, at the surface, and is a sandy solid below, at the sides. A liquid, with solid portions, at stationary temperature. *Hardens in round masses, which, at the stationary point, are surrounded by a liquid. The table of alloys by Parke, before referred to, gives a considerable variation in the melting points of the alloys in the above table. He makes the stationary point of the alloy of eight to eight, 372° Fah.; that of eight lead and ten tin, 352°; that of eight lead to twelve tin, 336°; this latter being the most fusible of the alloys of lead and tin. That the alloy, in equal parts, has not a fusing point varying much from that just given, the Committee were able to determine from various specimens of metal. With pure lead and grain tin, they found, for eight lead and nine tin, nearly the same as the fore going, the stationary point to be, in different experiments, 3551°, 356°, and 3551°. With one specimen of common lead the stationary point of an alloy of equal parts of lead and tin was 356° Fah. This lead melted at 606°, and the tin at 4421°. The Committee have no greater reason to suspect the accuracy of their other results. In all these cases the stationary point occurs when the metal begins to solidify. It appears, then, by the foregoing table, that very little change is effected in the fusing point of the alloy of equal parts of tin and lead, by increasing the quantity of the more REPORT OF EXPERIMENTS ON THE EXPLOSIONS OF STEAM-BOILERS. 121 in quantity, the passage from the fluid to the solid state is by such minute mechanical changes, as to extend through a long series of temperatures. This is even more especially the case when bismuth also enters largely int the alloy; instructive examples of which occur in the following table : 352° Fused metal begins to lose fluidity. 307 Soft solid, penetrable. 279 Stationary point. 358 Fused metal begins to thicken. 466 Begins to thicken. These facts show, that in using fusible alloys, those should be preferred which contain the smallest quantities of lead: a similar reason would lead to the preference of those containing the smallest proportions of bismuth. Tin is nearly liquid at the stationary temperature; hardens by plates or small masses, and becomes entirely solid at this same temperature. Experiments were made to ascertain what quantity of bismuth could be added to tin without destroying the property just described. To one hundred parts by weight, of tin, one part, five parts and ten parts of bismuth, respectively, were added. The first alloy melted at 4394°, and had the general characters of tin in hardening; the second melted at 428°, and had these characters impaired; the third had no stationary temperature above 400°, and lost its fluidity by slow degrees. As it was thus shown that alloys of tin and bismuth presented no peculiar advantages, the alloys for temperatures below 355° Fah. were sought by combining the least quantity of bismuth which would give any requisite temperature with one of the alloys of the table on p. 104. For this purpose the alloy of equal parts of tin and lead was selected, as having appropriate characters in its solidification, and melting at nearly as low a temperature as any of the others in the table. It does not, of course, follow, that this allow when combined with a given quantity of bismuth, will produce as low a fusing point as some other would; a question which, if it were worth deciding, experiment would determine. A few trials on this head were made by the Committee. The following table gives the proportions of bismuth, which, added to an alloy of eight parts of tin and eight of lead, will give the temperature of the stationary points of an immersed thermometer between 355° and 326°. With the alloy which terminates this table the stationary temperature near the fusing point disappears, and another form of table is required for description. Bismuth. 122 REPORT OF EXPERIMENTS ON THE EXPLOSIONS OF STEAM-BOILERS. TABLE II.-Alloys of Tin, Lead, and Bismuth, melting between 355° and 326° Fah. Eight parts, by weight, of Tin, and eight of Lead. TABLE III.-Alloys losing fluidity between 313° and 246° Fah. The fusing points of the metals used in the foregoing alloys were, of the tin, 442° Fah., of the bismuth, 506, of the lead, 612°. VI. To repeat the experiments of Klaproth, relating to the conversion of water into steam, by highly heated metal. It being now well understood that an increase of temperature in a metallic surface may diminish the amount of vaporisation of a fluid placed upon it, the object of the following experiments was to study the phenomena attending the vaporisation of water by iron and copper, under different circum stances. * Stationary at 205o. i Stem drew out at 25°. Stem drew out at 235o. 232|| 226 1st. To ascertain the temperature at which a given small quantity of water will be vaporised in the least time, by copper, with different states of surface. 2d. To ascertain the same point for iron, in similar circumstances. 3d. To extend these deductions to the effect of introducing different quantities of water into copper or iron vessels, varying in thickness, in character of surface and heated by different sources, to various temperatures. A number of bowls, of these different metals, of as nearly the same figure as could be obtained, and of different thicknesses, were + Stem drew out at 264°. Stem drew out at 245o. REPORT OF EXPERIMENTS ON THE EXPLOSIONS OF STEAM-BOILERS. provided. The bowls were portions of spheres, of nearly three inches radius, and were eight in number, three being of copper and five of iron; four of these latter were of wrought, and one of cast-iron. For applying heat to the bowls. a cylindrical vessel containing oil, and another containing tin, were provided; the former was about nine inches in diameter and four high, and the latter six and a half inches in diameter and four high. These vessels were heated by Mitchell's* alcohol lamp, or in the very high temperatures, by a charcoal furnace. The bowls were furnished with handles, which projecting, overlapped the edges of the cylinders, serving as baths for the oil and tin, and were thus kept in place. The thermometers used in the experiments were carefully compared at the boiling point of water, and melting point of pure tin. The experiments first to be detailed refer to the vaporisation of drops of water in copper 123 bowls of different states of surface, from the Vaporisation of Drops of Water by 1. The bowl, No. VII., of copper, sevenhundredths of an inch thick, was polished, but not very highly, and then placed in the tin bath while fluid; the tin, on solidifying, kept the bowl in its place. The thermometer was placed in a small cylinder of thin sheetiron, containing mercury, the cylinder being as near the cup as possible. As the experiments progressed, the surface of the bowl became, of course, more and more tarnished; and after the two series of results recorded below were obtained, a third showed a marked effect from the oxidation, by the increased vaporisation. One hundred and twenty drops, nearly, from the tube used, made up oneeighth of a fluid ounce; the weight of one drop was, therefore, about 47 of a grain. A very convenient alcohol lamp, with a dranght through the wick, and a separation between the alco hol reservoir and the wick. The invention of Dr. J K. Mitchell. In this and other tables, the series marked descending, are those obtained when the temperature was alling; the ascending series were obtained while the temperature of the bath was rising. |