not to be taken into consideration too. Of course either of those, I presume, would be sufficient. But it seems to me that either the clearance or the terminal pressure ought to be one of the elements; probably the terminal pressure would be better, because the varying amounts of compressions would affect the terminal pressure, while they would not have any particular bearing on the clearance. So that if the theoretical point of cut-off was taken in connection with the terminal pressure, there might be some satisfactory standard arrived at. My idea is that what we want to get at is the number of expansions, and that could be best got, probably, by taking in the terminal pressure as one of the elements.

Professor Robinson.-In answer to the question, if you will allow me a second word, the amount of steam admitted should not be taken as HE nor HC, but the compression line should be extended according to some steam lines from I to D to this top line. Now, the amount of steam admitted, I think, is DC, and that should be compared with the length DJ to obtain the ratio of expansion. As to what steam line ID should be, it is quite a question whether it should be the adiabatic steam line; but the curve to make on the board or on the diagram to establish D would be quite a question. There would be many points involved, such as the heat capacity of the cylinder walls, re-evaporation of steam, etc., just as in this curve FC.

Mr. A. Wells Robinson.-If this Society is to undertake the consideration of this subject, I should like to see it considered from two points of view. The first is the scientific point of view. I have no doubt that they will give that point of view every attention. The other point of view is that of the owner of the engine; and it seems to me that what we need to supply him with is some approximate system of obtaining just the information that he wants. It does not seem to me to be of much use to undertake a very elaborate system, to carry it out to, say, eight or ten places of decimals in an engine that is subject to a variation of 100 per cent. every five minutes; and what the owner of the engine wants to know in the majority of cases is the cost of his coal-pile, and how it can be reduced, if possible.

Mr. George H. Barrus.*—I heartily concur in the suggestion of Professor Magruder to include gas and oil engines in the scheme

*Author's closure, under the Rules.

point of cut-off, he will know how it will be understood by the purchaser and the expert who conducts the test.

Prof. S. W. Robinson.-What should be considered the point of cut-off? I think there should be some such term as effective cut-off or theoretical cut-off laid down and established by this committee when its work is done. Undoubtedly that question has been up in the minds of many gentlemen here. Taking this diagram Fig. 172 as drawn, this curve HFG, for instance, is the

actual curve given by the indicator. Now, at some point near F the actual cut-off or actual closure of the valve occurred, so that the actual cut-off or the actual closure of the valve would be found by projecting a vertical line from F up to E. That point would be not far from where the counter-curvature at Foccurs, and that is the point in the stroke of actual closure of the valve-but not the point of effective cut-off that I wish to call attention to. I think that some point C should be obtained by continuing the line GF back on some steam curve line FC to C to find the effective cut-off. That point should always be brought out in connection with tests. Attention is called to this important matter, and it is believed that the instructions to this committee that shall be appointed should be such as to require them to establish some point Cas the true theoretical point of cut-off and defining the line FC, even in the case where there is considerable curvature DF of the actual indicator line.

Mr. Barnaby.-I would like to ask Mr. Robinson whether he would measure that point from the end of the card or from the clearance line; also as to whether the terminal pressure ought

not to be taken into consideration too. Of course either of those, I presume, would be sufficient. But it seems to me that either the clearance or the terminal pressure ought to be one of the elements; probably the terminal pressure would be better, because the varying amounts of compressions would affect the terminal pressure, while they would not have any particular bearing on the clearance. So that if the theoretical point of cut-off was taken in connection with the terminal pressure, there might be some satisfactory standard arrived at. My idea is that what we want to get at is the number of expansions, and that could be best got, probably, by taking in the terminal pressure as one of the elements.

Professor Robinson.-In answer to the question, if you will allow me a second word, the amount of steam admitted should not be taken as HE nor HC, but the compression line should be extended according to some steam lines from I to D to this top line. Now, the amount of steam admitted, I think, is DC, and that should be compared with the length DJ to obtain the ratio of expansion. As to what steam line ID should be, it is quite a question whether it should be the adiabatic steam line; but the curve to make on the board or on the diagram to establish D would be quite a question. There would be many points involved, such as the heat capacity of the cylinder walls, re-evaporation of steam, etc., just as in this curve FC.

Mr. A. Wells Robinson.-If this Society is to undertake the consideration of this subject, I should like to see it considered from two points of view. The first is the scientific point of view. I have no doubt that they will give that point of view every attention. The other point of view is that of the owner of the engine; and it seems to me that what we need to supply him with is some approximate system of obtaining just the information that he wants. It does not seem to me to be of much use to undertake a very elaborate system, to carry it out to, say, eight or ten places of decimals in an engine that is subject to a variation of 100 per cent. every five minutes; and what the owner of the engine wants to know in the majority of cases is the cost of his coal-pile, and how it can be reduced, if possible.

Mr. George H. Barrus.*—I heartily concur in the suggestion of Professor Magruder to include gas and oil engines in the scheme.

* Author's closure, under the Rules.

for the proposed standard system of testing, and I have no doubt that the committee will adopt his suggestion. The printed forms used at Sibley College, which Professor Carpenter submits, will be of value to the committee when they come to consider in what manner the tables of results had best be presented. The question as to the proper location of the point of cut-off on the diagram, brought up by Mr. Barnaby, is important, and this is one which should be settled by the proposed standards. I agree with Mr. Robinson that the matter under consideration should be viewed not only from the engineering standpoint, but also from the commercial standpoint. The interests of the owner and user of an engine should certainly be remembered, whatever action is taken.

As to Mr. Stanwood's suggestion, that it would be well to investigate the relative economy of engines at different loads, it seems to me that, although very desirable, this is work for an independent committee rather than for the one proposed.

NOTE BY THE SECRETARY.-Mr. Barrus's paper was presented and discussed in connection with the paper of similar tenor by Mr. Bryan Donkin, of London, England, entitled, Extension of the Standard Uniform Methods of Conducting and Reporting Steam-Engine Tests." Readers and students are referred also to the discussion appended to that paper, which will be found as No. 786 of the current volume.

As the result of the discussion the Council was authorized and directed to appoint a committee to consider the question treated by Messrs. Barrus and Donkin. This committee, subsequently appointed, consisted of Messrs. Boyer, Barrus, Donkin, Jacobus, and Richmond.

DCCLXXXII.*

THE PROTECTION OF STEAM HEATED SURFACES.

BY C. L. NORTON, BOSTON, MASS.

THE investigation, of which this is a partial report, has been undertaken at the request of Mr. Edward Atkinson, and has been pursued during a large part of the years 1896 and 1897, and is yet uncompleted. The first object sought for was the relative efficiency of several kinds of steam pipe covering now upon the market. The second object was to ascertain the fire risk attendant upon the use of certain methods and materials used for insulation of steam pipes. Third, an attempt was made to show the gain in economy attendant upon the increase of thickness of coverings, and to show also the exact financial return which may be expected from a given outlay for covering steam pipes. Further information is given on many minor matters and conditions effecting the transfer of heat from a steam pipe to the surrounding air.

Method.

The method adopted is one which, so far as I know, is original. A piece of steam pipe is heated from the inside electrically. The amount of electrical energy supplied is measured, and hence the amount of heat furnished is known. If the steam pipe is kept at a constant temperature by a given amount of heat it is because that amount is just equal to the heat it is losing, for if the supply were not equal to the loss, the temperature would rise or fall. In other words, the heat put into the pipe is just equal to the heat lost from it by radiation, convection, and conduction. By measuring the electrical energy supplied I can determine the heat put in, and hence the heat given out or lost. It must be borne in mind that a given amount of electrical energy always produces the same definite amount of

* Presented at the Niagara Falls meeting (June, 1898) of the American Society of Mechanical Engineers, and forming part of Volume XIX. of the Transactions.

TB