the stoke-hold. This was done in order to facilitate the keeping of the water in the boiler at the right level, which would otherwise have been somewhat troublesome under the trial conditions. The levels of the water in the boiler itself, as well as in this auxiliary tank, were always made the same at the beginning and the end of each trial. The temperature of the feed, the height of the water in the glass, the pressure in the boiler, and the air-pressure and temperature in the stoke-hold, were noted every five minutes continuously throughout most of the trials, and in all cases every ten minutes. Arrangements were made for collecting samples of chimney gases, and for taking the temperature of those gases. This latter was noted every quarter of an hour, the thermometer used being a mercury thermometer containing compressed nitrogen over the mercury. The samples of chimney gases were collected about every hour by Mr. C. J. Wilson, F.C.S., and were afterwards analyzed by him at University College, London. Samples of the coal, which was "Nixon's Navigation," were collected on every trial. All the samples were subsequently mixed together, and resampled, and duplicate analyses made of the coal; this work was also done by Mr. Wilson, who also determined the calorimetric value of the coal by direct experiment. Samples of the ash were also collected and analyzed. The measurements thus made have sufficed to enable me to give you a detailed account of the expenditure of heat in and about the boiler in each of the trials. The torpedo-boat tested (No. 258) had two boilers; one of them was entirely disconnected and not used at all; the tests were made entirely on the forward boiler, and in all the following particulars about the behaviour of the engines and boat, it must be remembered that only steam from one boiler was used. In every trial a running start was made, as the boiler was so sensitive to changes in the fire or feed that greater error would have been introduced by the irregularities consequent on drawing the fire than were at all likely to follow the method adopted. The boiler and engines were allowed to work in the normal conditions of the trial for some little time before the start. The fire was then allowed to run itself as low as was consistent with the maintenance of the steam-pressure. The ashpit was cleared, and there was a weighed quantity of coal placed upon the stoke-hold floor. At a given signal the height of water in the gauge-glass was marked, as well as the level of water in the auxiliary tank, and the first portion of the weighed coal was put upon the fire. The time was noted at which each weighed quantity of coal was entirely put upon the fire, and until this happened no more coal was weighed out on the floor. The signal for ending the trial was given at a time, as near as possible to the intended duration of the trial, when all the conditions as to water-level, &c., were the same as those at the start. The diagrams of the trials show to what extent the coal consumption was constant throughout the whole of the trial, and afford, perhaps, the best test of the extent to which any difference in the state of the fire at start and finish could affect the result. It was intended always to begin and end with a falling pressure, and this was done as far as possible, but it was not found possible in every case. OBSERVATIONS IN ENGINE TEST. As it was not possible to test the boilers without running the engines, it was thought advisable and interesting to obtain all the particulars about the working of the engines that could be measured during each trial. The method of measuring the feed-water has already been described. In addition to this indicator diagrams were taken at or about the middle of each twenty minutes (in some cases fifteen minutes) during each trial, the number of revolutions were found from the counter, and the I.H.P. computed from each set of diagrams by taking into account the mean revolutions per minute during the period in which the diagrams were taken. As it was desired to keep the boiler-pressure the same in all cases, it was necessary in the low-power trials to throttle the steam very much on its way to the engines; the actual pressure available was, therefore, that given by the gauge attached to the high-pressure valve-chest. The reading of this gauge was noted every time diagrams were taken, as were also the gauges on the two receivers and the vacuum-gauge, as well as the engine-room gauge in connection with the boiler. BOILER. The boiler itself was one of Mr. Thornycroft's patent tubular boilers, having a heating-surface of 1,837 square feet and a grate-surface of 30 square feet. In trials "D" and "E," a portion of the grate was bricked up, so that it was reduced to 26.2 square feet. ENGINES. The engines were triple expansion, of your usual pattern, with cylinders 14, 20 and 31 inches in diameter by 16 inches stroke; the piston-rods were 21 inches in diameter. All three cylinders were jacketed. The feed-pumps were driven by the main engine. Separate engines were employed to drive the circulating pump and for the fan. The boiler supplied steam for these engines as well as for the main engine, and also for the donkey-engine and the steering engine. All the separate engines, except the steering engine, drained back into the condenser, so that the water-supply was not lost. Taking the whole experiment as a boiler trial this is, of course, of no importance, but taking it as an engine trial it must be remembered that the total feed-water includes that required for all the auxiliary engines, while, of course, the I.H.P. belongs to the main engines alone. The donkey-engine piston leaked considerably, especially upon trial "D," when its work was lightest, and the steering engine also lost some steam by leakage, so that the figures as given below, representing feedwater used per I.H.P. per hour, are very much larger than correspond to the real consumption of the main engines. COAL. The following is the analysis by Mr. Wilson of the coal used in the trials: I find by calculation that this fuel has a calorific value of 14,900 thermal units per lb., which is equal to that of 1.025 of a lb. of carbon. Each lb. of coal is therefore capable, if completely burnt, of evaporating 15 41 lbs. of water from and at 212° Fahrenheit. Mr. Wilson has made for me a direct determination of the calorimetric value of the fuel in a carefully tested calorimeter; this experiment shows the value of the fuel to be 15,450 thermal units per lb., which corresponds to 15,020 thermal units per lb. if the steam formed passes away in a gaseous condition. This value is only about 1 per cent. in excess of that due to calculation, a most satisfactory agreement. Each lb. of fuel requires 11.5 lbs. of air, theoretically, for its perfect combustion. The analysis of the ash shows it to consist of 86.5 per cent. of carbon, and 13.5 per cent. of inorganic and incombustible matter. (By ash, I here mean the material which fell through the grate bars into the ashpit during the trial.) The leading particulars of all five trials are given in Tables I and II. TRIAL "A," 21ST NOVEMBER, 1888. NATURAL DRAUGHT. This trial was the first in order, and may be taken as being rather in the nature of a rehearsal than a complete trial, but so many observations were made that I have thought it worth while to work out the results as completely as was possible. The trial lasted about five hours. The stoke-hold was open. The mean boilerpressure was 186 lbs. per square inch above the atmosphere, and the coal burnt was 334 lbs. per hour, or 11.1 lbs. per square foot of grate-surface per hour. The feed-water measurement on this trial was unfortunately rendered untrustworthy by an accident. The engine ran at 192·8 revolutions per minute, the pressure in the HP. valve-chest being only 50 6 lbs. per square inch. The I.H.P. was 150 3. The total coal put upon the fire amounted to 2.22 lbs. per I.H.P. per hour, but in this case none of the ash (which it will be seen contained over 85 per cent. of carbon) was re-used. The ash amounted to just 16 per cent. of the whole weight of the fuel. The samples of furnace gas collected during this trial were analyzed with the following results: Calculation from this analysis shows that about 24 lbs. of air per lb. of coal were used; that is about 2.1 times the quantity theoretically required. TRIAL "D," 26th NOVEMBER, 1888 (PLATE 3, FIG. 3). NATURAL Draught. This trial lasted 4 hours 57 minutes. The stoke-hold was open, the mean boiler-pressure during the trial 181·8 lbs., and the coal burnt per hour 203·3 lbs., or 7.74 lbs. per square foot of grate per hour. The feed used per hour was 2,281 lbs., which corresponds to an actual evaporation of 11 22 lbs. of water per lb. of coal. The temperature of the feed was, however, 76° 3 Fahrenheit, while the temperature of the steam was 380° 2, so that reduced to standard the evaporation was 13 40 lbs. of water per lb. of coal. The ash which fell through the grate amounted to about 47 lbs. per hour; this very large percentage being, of course, due to the very thin fire that had to be kept up on the grate. The whole of this ash was put back again on the grate, about a quarter of it in the fourth hour, and the remainder in the last hour. The coal line in the diagram of this trial shows very clearly its use as fuel. The detailed analyses of the chimney gases are given in the appendix to this report; the following is their mean :— The analyses given in this and all other similar tables are those of the gas in a dry condition, that is free from water vapour. I did not attempt to find out the weight of moisture in the gases by direct experiment. This omission does not sensibly alter the calculation by which the weight of air per lb. of carbon or coal is obtained; its greatest effect is to make a small difference in the specific heat of the gas. For example, the specific heat of the dry gas in trial "B" is 0.238, whereas the presence of gaseous steam, due to the hydrogen in the fuel, would increase this to 0.242. The mean specific heat of the gases has been taken as 0.24 in all cases. I find that the above analysis corresponds to only 18 14 lbs. of air per lb. of coal, which is 1·62 time the quantity theoretically required, but it will be noted that although the quantity of air is so small the combustion was very nearly perfect, only one of the samples collected showing any carbonic oxide. The heat of the fuel goes in greatest part to the heating and evaporation of water. The remainder partly heats the furnace gases, partly is lost by imperfect combustion and partly by radiation and other causes. I attempted to determine the loss by radiation, as I have been able to do approximately in some cases, by finding the amount of coal which it was necessary to burn in order to keep up the steam-pressure in the boiler, no steam being allowed to leave the boiler, and no water being put into it. The quantity, however, was so small that the errors in its measurement due to burning it in such a large grate prevented my obtaining any satisfactory results. The following short Table shows the way in which the heat of combustion was utilized by the boiler. It is perhaps unnecessary that I should give the calculations in detail: Heat expended in heating and evaporating feed-water. by radiation and otherwise unaccounted for. Per cent. 86.8 10.8 0.5 1.9 100.00 The small loss in furnace gases corresponds to their very low exit temperature, namely, 421° Fahrenheit. The very high percentage of evaporation, 86 8, represents the efficiency of the boiler, and is of course simply equal to the ratio between the actual evaporation and that theoretically due to the perfect combustion of the fuel, or 13.4 It is only right that I should say that this is the highest boiler efficiency I have ever found upon any trial with which I have had to do, if indeed it be not, as I almost think it is, the highest on record in any trustworthy manner. As regards the working of the engines on this trial, I can give you the following particulars: The engines ran at 165-2 revolutions per minute, but in order to keep the power down sufficiently it was necessary so to throttle the steam that the pressure in the high-pressure valve-chest was only 22.7 lbs. per square inch above the atmosphere. As the engines are intended to work also with a pressure of 200 lbs. per square inch in the valve-chest, it is a matter of some curiosity to examine their behaviour under these very special conditions. The mean pressure in the HP. cylinder was 13.0 lbs. per square inch, in the MP. cylinder 9.5 lbs. per square inch, while in the LP. cylinder it was only 2.2 lbs. per square inch. The vacuum was 27.97 inches of mercury, which on the day of trial corresponded to an absolute pressure of 0.79 lb. per square inch. The total feed-water per I.H.P. per hour (that is, as already explained, the total feed-water per hour used for all purposes divided by the total I.H.P.) was 25 6 lbs., the I.H.P. being only 89 1. The coal per I.H.P. per hour was 2.28 lbs. The water coming from the jackets was collected and measured separately before being returned to the feed measuring tanks, it amounted to 5.6 per cent. of the whole feed-water. The leakage through glands, &c., together with the non-return to the condenser of the steam which passed to the steering-engine, made it necessary to add a little water from time to time in order to keep the water at its proper height in the boiler. This added feed was separately weighed before being poured into the hot-well (through which it passed to the measuring tanks), and amounted to 0.57 per cent. of the whole feed. TRIAL "C," 24th NOVEMBER, 1888 (PLATE 3, FIG. 2). AIR-PRESSURE IN STOKEHOLD, 0.27 INCH. In this trial it was intended to keep the air-pressure in the stoke-hold as nearly inch of water as possible. The actual average pressure during the trial, which lasted five hours and nine minutes, was 0.27 inch. The average boilerpressure was 171 2 lbs. per square inch above the atmosphere, and the atmospheric pressure for the day was 14.8 lbs. per square inch. The coal burnt per hour was 559 lbs.; the rate of combustion being therefore 18.6 lbs. of coal per square foot of grate-surface per hour. The ash which fell through the grate during the trial amounted to 6.8 per cent. of the weight of coal, but was nearly all put back on the fire before the end of the trial. The feed-water evaporated amounted to 5,852 lbs. per hour, the feed temperature being 78° Fahrenheit, and the steam temperature 376° Fahrenheit. The actual evaporation per lb. of fuel was therefore 10.48 lbs. of water, which reduced to standard amounts to 12.48 lbs. of water, per lb. of coal. The airpressure in the chimney, measured close to the place at which the furnace gases were collected, was 0.03 inch of water. The mean temperature of the chimney gases was 540° Fahrenheit. By a most unfortunate accident to the case containing the furnace gas sampling bottles, these were all broken, so that I am not able to give the analysis of the furnace gases for this trial. The results, however, from trials "D" and "B," between which this trial lies, allow a fairly accurate estimate to be made of the quantity of the furnace gases. In trial "D," with natural draught, 18.1 lbs. of air were used per lb. of coal; in trial "B," with inch of pressure in the stoke-hold, 17.4 lbs. of air were used per lb. of coal. It cannot be far wrong therefore to take the consumption of air per lb. of coal in trial "C" as about 17.8 lbs. The heat balance for this trial will therefore stand thus: Heat expended in heating and evaporating feed-water. Per cent. 81.4 15.0 |