Professor

Correspondence.

Professor V. DWELSHAUVERS DERY observed that in his view the

Dwelshauvers. Author's experiments were unexceptionable, and made under conditions which gave them considerable importance. It was so much the more to be regretted that they were insufficiently detailed. In fact, to study them, and to investigate the bearing of the walls of the cylinders after the method of Hirn, it would be necessary to have for each cylinder separately the volume of the dead space, and the volume occupied by the steam at the commencement of each of the phases of admission, expansion, exhaust, and compression; the pressure of the steam at each of these periods; the work of this pressure during each of these phases; the mean diagram for each side of the piston; the external radiation of each cylinder and of each receiver; the quantity of heat transferred by the steam in each jacket. In fact, the data ought to be sufficiently complete to allow of the application of the six equations of the practical theory. With the information supplied in the Tables, only the first equation could be established, the equation of control, the balance-sheet of the trial. Without doubt, some conclusions relative to the facts might be deduced from it; but it was not clearly elucidated either in respect of the reason of, or of the method of the economies realized. He had applied so far as possible his method of investigation to these experiments. It consisted in bringing into account all the quantities of total heat expended, and in expressing them as fractions of this total heat. This was composed of two parts; first, the heat brought to the cylinder by the steam, Q, reckoning from 32° Fahrenheit; secondly, the heat Q' of the jacket steam, total Q+Q'. This heat ought to be found again (1) in the external work accomplished, represented by T thermal units; (2) in the external radiation E; (3) in the heat given up to the condenser composed of two parts, namely, that which had been furnished to the cold water, C, and that which the condensed steam possessed above 32° Fahrenheit, c; and for the sum (C+c).

By this process, it would be seen clearly that No. 56 gave the highest efficiency; that for some unexplained reason the external radiation E was in that case very feeble; that the heat Q' E supplied to the steam in the first three trials was very nearly the same. The reason then of a greater efficiency in trial 56 was not apparent; but it was, however, a question which deserved to be cleared up. The Paper contained sufficient data to determine the thermal efficiency, f, that of the fluid, f2, that of the mechanism, f, and consequently the ultimate efficiency f1 f2 f3 = f. By thermal т-т -71 efficiency he meant the fraction

T

in which was the absolute temperature of the source of heat (of the steam in the boiler), and 1 was the absolute temperature of the cold condensation water, of the cold body. The temperature of the cold water varied greatly from one trial to another, insomuch that the possible fall and the thermal efficiency varied for the highest and the lowest as much as 5.5 per cent. from the mean. He called f, fluid efficiency, the relation between the work indicated and that which should be obtained from a Carnot ideal motor with the same expenditure of heat and the same fall of temperature. Lastly, f3 was the relation between the brake HP. and the indicated HP. From the highest to the lowest the difference varied about 5 per cent. from the mean of this efficiency, but the reason for this variation was not very apparent. In each of the two series, with and without steam in the jackets, the efficiency of the mechanism was greatest at mean speed. With steam in the jacket it was least for the highest speed, and without steam for the lowest speed. If, therefore, it

Professor

Dwelshauvers.

Professor

Dwelshauvers.

were desired to determine either the benefit due to the jacket, or that due to the influence of the speed, it would be necessary to disregard f, and f, and consider only f2. At the outset he would consider the efficiencies deduced from the data in the Paper :

The last line, the product of fif2 represented the relation between the heat equivalent to the indicated work and the total heat expended. The economical effect of the jacket was shown by a comparison of the trials, made at approximately the same speed, 1 and 4, 2 and 5, 3 and 6, with respect to the value of ƒ1⁄2 only :

The mean advantage was 0.194, or 19.4 per cent., the highest being in respect of the highest speed. To find the advantage resulting from high speeds, he would compare in the same way trials 1 and 2, 2 and 3, 1 and 3 with the jacket, and trials 4 and 5, 5 and 6, 4 and 6 without the jacket. This would give:

The maximum efficiency of the fluid was therefore obtained at mean speed; but the highest speed gave greater efficiency than the lowest. He hoped that the data in the Paper would be ex

tended, so as to allow of a more thorough analysis of the thermal Professor phenomena observable in the experiments.

Dwelshauvers.

Mr. BRYAN DONKIN, Jun., said he would like to know how the Mr. Donkin. radiation experiments had been made. It would much increase the interest of the Paper if the Author could add any experiments giving the result of each set of steam-jackets separately, so as to determine their relative effect; also if he could divide in the same way the returns of the total weights of jacket-water. If an experiment could be made with the three engines coupled in the usual way, as in mill and marine engines, it would be of great value. His new hydraulic brake dynamometer was especially interesting and complete, and was likely to be much used in similar experiments on account of its small size and its easy application. The great difference between the feed-water into the boiler and out of the air-pump was startling; but no doubt the Author would be able to reduce this in future experiments, or determine the reason for the differences. As the balance-sheet system of heat in and heat out had been used in this Paper, as in many other similar trials, and had been so largely adopted of late years by engineers, it might be of interest to state that, after some trouble, Mr. Donkin had succeeded in tracing when and by whom it originated. Professor Dwelshauvers Dery, of Liége, informed him that it was first employed by Mr. Hirn in the following Paper :-"Mémoire sur l'utilité des Enveloppes à Vapeur."1

Mr. A. C. KIRK could not refrain from a note of warning, that Mr. Kirk. results got from a model, however carefully and correctly the trials might be carried out, should not by any means be implicitly accepted as applicable to a large engine. As an illustration, the late Mr. Froude's observations on the behaviour of ship models towed in a suitable tank, while they might in any case have been of some interest, would have been of little value had not that illustrious experimenter found a law of comparison between the results from the models and those of the full-sized ship, and been able to prove by experiment that the law held true. It might, perhaps, be said that so far as the action of the walls of the cylinder was concerned, the law of comparison was simply that (under the same conditions of temperature, of expansion, &c.), while the weight of steam used in the cylinder would be as its cubic capacity, the action of the sides of the cylinder would be in proportion to the surface exposed to the working steam. But this should by no means be hastily assumed, and it might possibly be

1 Bulletin de la Société Industrielle de Mulhouse, vol. xxvii. (1855), p. 105. [THE INST. C.E. VOL. XCIX.]

S