The natural life of machinery, it must be remembered, is not wholly a question of the resistance to wear and tear, but it is subject at times to the greater hazard of depreciation from subsequent inventions. After the Border City mill fire the treasurer wired George H. Corliss that the underwriters wanted to discount ten per cent. from the cost of an engine which Mr. Corliss had built only the previous year, and wanted him to come on and defend the price of his engine which had only been used for a year, before the commission that was trying to adjust the loss. Mr. Corliss wired back congratulating him on the small amount of the discount. After the whole adjustment was effected and the matter was all over Mr. Corliss said to me: "In that year I had made other inventions which had depreciated the value of my past work, and therefore if that man was to expend the same amount of money on an engine of this kind he could have got a great deal better engine." Those of you who knew Mr. Corliss and his characteristics can well appreciate the unction which he would give to such a statement. A good many years ago, when George Draper was engaged on the development of the spinning frame for spinning filling yarn, I was in a small private shop where it was working very well. Turning to me he said: "When this machine is done the mules in the cotton mills will be thrown out of the window down a chute because they are not worth the labor to take them down stairs." While as a matter of fact Mr. Draper's anticipations were not so fully realized, yet on certain classes of cotton yarn it did depreciate the value of the mule. It stopped the production of any more and it did cause a depreciation of a great many cotton mills. There is a book on the subject of the depreciation of machinery by Ewing Matheson, an Englishman. The questions involved in the valuations of water powers seized by a city, was gone over very thoroughly by General Butler, as counsel for the city of Boston, in what are known as the Sudbury River cases, in which he introduced, I believe, to American practice for the first time, a great many English judicial rulings on the subject, which were accepted by the commission and have passed into American practice.

Mr. Main.-What Mr. Rockwood says is about right, that you can estimate the value of the machinery by considering the depreciation in a general way about as nearly as in any other way. I think, however, that an examination of the machinery should be made in order to be sure that the depreciation has not been more

rapid than would ordinarily be expected, through excessive speeds or severe work, or longer hours run than usual, or by negligence. The inability to separate the depreciation from wear and tear and from that due to improvements as mentioned by Mr. Woodbury is spoken of in the paper.

MULTIPLE-CYLINDER

DCCLX.*

STEAM ENGINES.-EFFECTS

OF VARIATION OF PROPORTIONS AND VARIA-
BLE LOADS.

BY ROBERT H. THURSTON

AND

LOUIS L. BRINSMADE.

(Member and Past President.)

(Junior Member of the Society.)

THE following paper is intended to exhibit the results of an experimental investigation of the relative efficiencies, with varying loads, of the ordinary high-pressure triple-expansion engine, of the compound of usual proportions at similar pressures, and of an intermediate type, already somewhat familiar to the engineering world, through the enterprise of a member of this Society mainly, in which the high-pressure element of the compound engine is made exceptionally small; the effect being, practically, that which would be produced by the suppression of the intermediate cylinder of the usual construction of the tripleexpansion machine. The machines employed in this research were, in fact, the available combinations of the largest of the triple-expansion "experimental engines" of Sibley College, and the combinations adopted were:

1. The triple-expansion engine in its usual condition.

2. The intermediate and the high pressure elements combined to make a compound engine of usual proportions-three to one. 3. The low and the high pressure elements combined to produce a compound of the peculiar sort above mentioned-seven to one.

The subject here taken up for study and experimental investigation was first brought into view by the remarkable results reported by Mr. Rockwood to the American Society of Mechanical Engineers, as given by test of singularly proportioned engines, in which the total expansion was that made appropriate to pressures for which triple-expansion engines were customarily used, where great economy was sought, and yet in which he

* Presented at the New York meeting (December, 1897) of the American Society of Mechanical Engineers, and forming part of Volume XIX. of the Transactions.

adopted a compound system having an abnormally small highpressure cylinder. The engine was thus very similar in its general arrangement and proportions to a triple-expansion engine with the intermediate cylinder omitted; the exhaust from the high-pressure cylinder passing directly into the lowpressure element, an intermediate receiver being employed of suitable dimensions to give a clean drop and unobjectionable fluctuation of pressures. The economy reported for this case was both unexpected and unprecedented. The following study of the general case will show, at least in part, how this unexpected and singular result came about.

The characteristic feature of this new idea in steam-distribution was a large "drop" between cylinders; but "drop" has never been regarded as a desirable feature in itself, either practically or thermodynamically, and it is interesting to trace out those phenomena and conditions which have here made such a feature a source of advantage. It will further be interesting to see that the thermodynamic case, practically and theoretically, aside from the finance of the matter, conforms to our earlier ideas of the essentials of maximum efficiency.

The Transactions of the American Society of Mechanical Engineers for the year 1892, vol. xiii., contain the first account of the results of experimental investigation of the relative value of the peculiar form of compound engine which it is the purpose of this paper to discuss.*

The engine there described was a triple-expansion engine, built from the designs of Mr. Rockwood for the Merrick Thread Co., of Holyoke, Mass. The dimensions were 12, 16, and 243 inches diameters of cylinder, 36 inches stroke of piston. for the high and intermediate cylinders, and 48 inches for the low-pressure element. Receivers were placed between the cylinders, and the cylinders and receivers were jacketed. The machine was rated at 175 horse-power. A separator was used at the engine and returned the condensed and entrapped water to the boiler through a "steam-loop."

The results of test were reported as giving from 12.67 to 13.06 pounds of feed-water per indicated horse-power per hour, with steam at 142 pounds by gauge. The best figure was given when developing 199 horse-power. When the intermedi

Two-cylinder vs. Multi-cylinder Engines," by Messrs. Green and Rockwood, presented at the San Francisco meeting, May, 1892, p. 647.

ate cylinder was thrown out of action, the best figure was 12.76 and when developing 180 horse-power. The conclusion was apparently justified that the two forms of engine were of prac tically equal efficiency. The conclusion might further have been reached that, when costs are considered, the compound, as here proportioned, was the better of the two styles of engine. Still another conclusion would seem to be by this experiment fully justified that the new system of proportioning the compound engine is decidedly the better where loads are in any considerable degree variable, as giving better opportunity to meet the demand for variable power with least sacrifice of thermodynamic and engine efficiency.

In the discussion of this paper it was remarked by Mr. Cooper that" if two cylinders will secure an economy commercially equal to that obtained by the use of a greater number, then two cylinders are enough"-a conclusion which is axiomatic but none the less important. The italicized word-the italics are introduced by the writer-is, however, of special importance; since the maximum thermodynamic efficiency and a minimum consumption of heat, steam, and fuel is not, by any means, necessarily coincident with, or even approximate to, the condition of maximum commercial economy. The true statement of the engineer's problem is always: What construction and what steam-distribution will give the largest return on the investment in building an engine to supply a specified amount of power?

It is to be further particularly noted that the question which here arises is not at all whether two or three cylinders should be adopted in a stated case, but, and especially, whether, it being decided that, for business reasons, the compound is better than the triple-expansion machine, the two-cylinder compound should have the usual or an exaggerated cylinder ratio. The question is not whether a two-cylinder is better than a three-cylinder series engine, but whether novel proportions are to be adopted with the older type of engine.

A comparison made by Mr. F. W. Dean, of the performance of an engine designed by Leavitt with one designed by Rockwood, the one having a cylinder ratio of four to one, the other of seven to one, later gave rise to further and interesting discussion.*

* Transactions A. S. M. E., vol. xvi., No. 619, p. 169, 1895.