ON THE STRENGTH OF STEAM BOILER MATERIALS. F was substituted, and, by means of notches 0, 0, at its upper rim, it was raised so high as completely to embrace the bath B as represented in plate IV. Standard of High Temperatures.-The standard of temperatures below the boiling point of mercury was the mercurical thermometer. Above that point the instrument adopted by the committee, was the steam pyrometer described in the American Journal of Science, vol. xxii. page 96, by Professor W. R. Johnson. At S, plates III. and IV., is the standard piece of wrought iron laid horizontally beneath the bar a, and kept in its place by the buoyancy of the mixture of tin and lead, the superior density of the latter metal serving to float the iron, and the higher specific heat of the former, 121 keeping the temperature of the bath more steady than it would have been if lead alone had been employed. As several improvements have been made in this pyrometer since the date of its first publication, which are conceived to be important in point of accuracy and despatch, it is proper that we should present a view of its structure as used in these experiments, together with a concise statement of the mode in which the latter were conducted. The instrument is founded on the supposition, that from a mass of water already in ebullition, a weight of vapour may be generated, by immersing a solid, of known weight and capacity for heat, which shall be proportioned to the temperature of such solid above the boiling point of water. In this Plate IV. ON THE STRENGTH OF STEAM BOILER MATERIALS. that when the equilibrium would be produced. To answer the conditions above iadicated, the steam pyrometer is constructed in the form represented in plate V. A, fig. 1, is a cylindrical boiler, 12 inches high, constructed of two concentric cylinders of tinned sheet iron, between which is a stratum n, n, half an inch thick, of dry charcoal-dust, (lamp black,) to serve as a non-conductor, and preserve the water within from loss of heat by radiation during the experiment. The bottom is formed of a single sheet of the same metal connected with the lower edge of the inner concentric cylinder, and rising in the form of a segment of a sphere to the height of 3ths of an inch, in order to present an enlarged surface to the action of the lamp L, which keeps the water in ebullition until the moment when the solid is immersed. t is a thermometer bent at right angles ths of an inch from the bulb, and passing along a tube opening into the boiler. A packing around this part of the stem prevents leakage, and the bulb being wholly immersed in the water, serves at all times to indicate its temperature, and particularly to mark the moment when that of the solid has descended so low as to cease generating steam of atmospheric tension. Mode af ascertaining the weight of vapour expended. The mode of suspending the boiler to the beam of the balance, is seen at m where the dotted prolongation of the beam B forms a forked curve rather greater than a semicircle, each extremity of the arc m, m, (fig. 2.) being turned inward so as to stand at right angles to the direction of the beam; this brings the two bearing edges which supports short hooks attached to loops on the opposite sides of the boiler, into the same line parallel to the main axis or knife-edge f, at the central part of the beam. These parallels are exactly 12 inches apart. The opposite arm of the balance beam is cylindrical, and cut into a fine threaded screw to within an inch and a half of the fulcrum f, where the beam is divided for a certain space into two portions ( x, fig. 2,) between which passes the upright rod of the supporting stand. The beam used during a considerable part of the experiments contained 150 threads in 1 foot of the length of the screw, and was an in inch in diameter. At the highest temperatures which the committee had occasion to measure, the number of threads passed over in one experiment, did not exceed 11, or so much as to measure from 1100 to 1200 degrees of temperature. By a careful measurement of different numbers of threads, selected at various parts of the screw-arm, (which was 16 inches long) it was ascertained that, though at the two 123 extremities the threads differed slightly from their mean length, yet at the middle portion, where the revolving counterpoise P is represented in the figure, no appreciable difference could be detected, and as this was the part always occupied by the weight, the instrumental error from this source may be considered altogether unimportant. In fact the extremes of the variation just referred to taken most unfavourably, could not have in the highest temperatures, amounted to more than 7 degrees Fah., which at points so elevated as 13 or 1400 degrees, would not be deemed a very important inaccuracy. But even this was avoided by occupying that part of the screw where the threads were of equal length; and by making an adjustment of the balance and weighing any given body placed on the boiler, with the counterpoise in different parts of the range selected for the experiments, it was easy to verify the accuracy of the measurements and determine precisely the error, if any had existed. But this method, when tried, only served to confirm the result of the other. The revolving counterpoise.—On the screw already described, was placed a revolving counterpoise P, which, together with its index I, placed on a neck projecting beyond the base of the cylinder, weighed 10517 grains; consequently, as each thread of the screw measured 0.07872 of an inch, the motion of the counterpoise over one thread was equivalent in effect to a weight of 100 grains applied at that end of the beam where the boiler is suspended, or 16 part of a revolution marked a difference of one grain in the weight of water in A.* The method used in determining, by calculation, the true adjustment of parts and the graduation of the scale of the steam pyrometer is equally appl cable, whatever may be the length of the weighing beam, or of the threads of the screw, and whatever the nature of the material employed for a standard piece, the latent heat of vapour, the kind of liquid from which it is produced, or the scale of thermometer to which we refer the indications, marked on the revolving counterpoise. Thus, if L be the length of the arm from f to m where the boiler is suspended, and the number of threads of screw on an equal length of the opL posite arm; then will --the length of a thread. Putting P-the weight of the counterpoise, and v-the weight of vapour which must escape in order that an equilibrium, destroyed by its loss, should be restored by making P move one revolution, that is one thread nearer to f, we shall have PL Р -L:v: P or Lv; whence v-. This may n n n be termed the mechanical relation of the instrument to the vapour produced. To determine the physical action of the standard piece, if be supposed = its specific heat; /= the latent heat of vapour from the liquid at its boiling point; the weight of the standard piece (ex The standard piece employed to produce vapour after having been heated in the bath of melted metal surrounding the bar under trial, was formed of wrought iron of the figure seen at S, its greatest length 23 inches, its diameter one and inch, and its weight 6336 grains Troy. This was sussuspended by an iron wire of an inch in diameter to the centre of a wire gauze cap w, the lower and smaller end of which entered the mouth of the boiler A, at the base of the funnel r. The upper diameter of this cap is 2 inches, and its height 2 inches, giving an area, including the top and sides, and of more than 134 square inches, or more than three times as much as the section of the boiler at its mouth. The object of employing this cap is to prevent the dashing out of water by ebullition-an effect which is, however, only liable to happen near the close of an experiment when the iron has descended to the temperature of maximum vaporization, and when the boiler contains too much water. The condenser.-In order to prevent all escape of vapour after ebullition has ceased, a cylindrical cap of tinned iron D, is placed over that of wire gauze, the instant that the pressed in the same denomination as that of P); the number of degrees to be placed on the circumference of the revolving weight; P=the degrees belonging to the same scale as those in which 7 is expressed; and (as before) the weight of vapour counterpoised by a single revolution of P; then the efficient cause of vaporization while the standard piece cools through degrees, will be represented by itz, and the effect produced must be expressed by vl; hence, is derived the equation lu itz, or V= Comparing this value Τ P itz of v with that obtained above, we get. n itz nitx P = In practice it was found most convenient to make t 100 degrees Fah. But by assuming t equal to the distance between the freezing and the boiling point of water under mean atmospheric pressure, a single standard piece would be sufficient to render the instrument universal in its indications. The curved surface of the revolving weight would only require to be divided into as many separate bands as there were different scales to be placed upon it, and graduating each band into as many equal parts as the particular scale comprehends degrees between the two points above mentioned. Thus we should have on the band marked Reaumer, 80 degrees; Centigrade, 100 degrees; Delisle, 150 degrees; and Fahrenheit 180 degrees. It may not be improper to remark that in applying the above formulas, the numerical value of also vary with the thermometrical scale. Thus, if for Fah. it be 1037 degrees, it will be for Centigrade, 576; for Delisle, 8634, and for Reaumer 4608. must See Amer. Journal of Science, vol. 21, p. 304. boiling point is attained. In general this cap is kept suspended at one side of the boiler A, as exhibited in the outline at D'. The lower rim of the condenser is furnished with an exterior welt or hem of silk, sewed to the tin by means of fine punctures near its edge. This serves effectually to prevent the escape of steam, and, besides allowing the operator to attend deliberately to the adjustment of the counterpoise, will admit, when necessary, the postponement of this process for a considerable length of time. The counterweight c, is to balance the standard piece S, with its suspending wire, and the wire-gauze cap W. As long as the water is kept boiling by the action of the lamp L; C is renoved from the beam and is replaced only after the condenser has been transferred from D' to D. The support E of the beam may be elevated or depressed on its sustaining rod by means of the tightening screw K. Immediately below the fulcrum f, is a small hole drilled horizontally into E, to receive a brass tap carrying a ball of an inch in diameter, through which passes the small index-wire i, so adjusted by means of the screwed counter-weight c, as to be accurately balanced on the tap g, as an axis on which it turns with no other resistance than what is due to the friction produced by its own weight. Near the extremity v of this wire, it is bent at right angles, and the pointed extremity directed to the side ef the beam where, at o, is a straight line of an inch long, serving to guide the eye in reproducing the level after an experiment. A little below this line, is a transverse hole through the beam in which slides the register s (figs. 1 and 2,) -a wire about of an inch in diameter, and 2 inches long. While the water in A is kept boiling by means of the lamp, the boiler continues to rise, and as the register now projects out beyond the index i, it lifts the latter, keeping the point v always directed a very little above the line o. But when W has been inserted in its place, with S suspended in the water, the additional weight, destroying the equilibrium, depresses the boiler, the base of which rests on the flat surface of the lamp L, the concavity in the bottom serving as an extinguisher to the flame. The index i, is, in the mean time, left at the level attained by the register at the moment before the immersion. While ebullition is proceeding, the operator pushes back the the register so as to project but little from the interior side of the beam; then observing the thermometer t, takes the condenser from the position D', and, at the instant, the ebullition ceases, covers the boiler with it, as at D, letting the standard piece remain in place; and having attached the AMERICAN FIRE-PROOF PAINT. counterpoise C, proceeds to bring down the boiler end of the beam, (which at first rises above the index i,) by causing P to revolve in the direction towards f. The number of revolutions being counted so many hundreds of degrees, he has only to add to their number 212°, in order to obtain at once the temperature by Fahrenheit's scale. As both the latent heat of vapour and the specific heat of iron enter into the calculation, in constructing the steam pyrometer, and as on both these points considerable discrepancy prevailed among writers who had treated of these subjects, it was thought important to attempt a direct solution of the question as presented in the particular case of this investigation. Two methods offered themselves, of verifying the calculations respecting the instrument. The first was, to heat the standard piece to any known temperature above 212°, and in that state plunging it into the boiling water to ascertain whether the amount of water vaporized weighed as many parts measured by hundredths of a revolution of C, as the standard piece had been heated in degrees above the boiling point. This method being the most direct, was first resorted to by the committee. As the standard piece S was at first made one or two hundred grains heavier than was supposed to be necessary, trial was made in the way just indicated, and as an excess of vapour was found to have been obtained, the standard piece was proportionally reduced in weight to that which has been already stated. The other method consisted in determining separately by direct experiment, both the latent heat of vapour, and the specific heat of the standard piece. The researches on these subjects were made in the manner and with apparatus described hereafter. (To be continued.) AMERICAN FIRE-PROOF PAINT. In November, 1837, a patent was granted to Mr. Louis Paimboeuf, of Washington, for a fire-proof paint, to prepare which he gives the following instructions in his specification, published in the Franklin Journal. Whether this American paint is in any way similar to the "composition for protecting wood from flame," lately patented by Mr. Davies, in England, which forms the basis of the Fire Preventive Company and with which various experiments have been made throughout this country, as it appears, has also been the case with Mr. 125 Paimboeuf's paint in Washington, we of course, are unable to say, as the specification of the English patent has not yet been enrolled : 66 My fire-proof paint," says Mr. Paimbœuf, may be prepared by grinding and incorporating the ingredients used, either with oil, or with water, as may be preferred. That prepared with water or other aqueous fluid, however, has one advantage over that prepared with oil; namely, it dries very rapidly, and affords the desired protection immediately; whilst that prepared with oil will not harden until the lapse of several weeks, depending upon the season of the year. "To prepare my paint, I take the best quick lime, such as when slacked forms an impalpable powder, and slack it by the addition of so much water, only, as is requisite to produce that effect, performing this operation in a trough or vessel, which may be covered over, so as to retain the vapour and heat as perfectly as possible, as upon this procedure I find that much of the effective strength of the composition is dependent. "When the slacking has become perfect, and the mass has cooled, I, in order to prepare my water paint, add either water or skimmed milk, or a mixture of the two, to the lime, in sufficient quantity to give to it the consistency of cream, or that of ordinary paint. When milk is not used, I add to the water a quantity of rice paste, made by boiling rice in water to a proper consistence, using about eight pounds of rice to every hundred gallons of the prepared paint. To every hundred gallons of this prepared lime mixture, I add twenty pounds of alum, fifteen pounds of potash, and one bushel of common salt. These are the essential ingredients, and the proportions such as I have found to answer well. If the paint is to be white, I find it advantageous to add to these ingredients about six pounds of prepared plaster of Paris, and the same quantity of fine white clay. When the paint is not required to be white, I substitute clean, well-sifted, hard-wood ashes for the potash; about two bushels being sufficient for the above quantity; in this case, also, I frequently add three or four gallons of molasses. "After mingling these ingredients, I first strain them through a fine strainer, and afterwards grind them together in a paintmill, when the paint is ready for use. When roofs are to be covered, or when crumbling brick walls are to be coated, I mix with my paint a quantity of fine white sand, in the proportion of about one pound to every ten gallons of the paint, as this addition will |