brought out in the electrical field. It has been touched upon by various electrical societies, and has been discussed in the electrical journals, but it properly belongs to the domain of the mechanical engineer. The process was invented, and has been developed by one of the brightest electrical men in the field today, who I am sure would not have induced the investment of capital for the building of expensive machinery unless he was certain that there was something in it. In regard to the burning of the metal, if I may be permitted, I will read a letter from Professor Dolbear, which is quite interesting, and is supported by the tests which have been made later. This letter was in reply to an inquiry from the editor of the Engineering and Building Record, who had seen a statement that a certain European process of welding had burned the metal. He referred this paragraph to Professor Dolbear, who replied as follows:

COLLEGE HILL, MASS., Feb. 5, 1888.

SIR: I have made nearly a hundred tests of the tensile strength of electrically. welded bars of iron, steel, and other metals. The results were of such a character that I can state positively that with Thomson's welding process it is possible to weld both wrought iron and steel so that the weld is as strong as the same cross-section in another part of the bar; that the appearance of the fracture is fibrous for iron, and generally granular for steel, the strength of this granular steel being on some samples as high as 125,000 pounds per square inch; that the process is such that the welding is homogeneous from necessity. I had a number of bars welded by an expert blacksmith, and a number of similar ones by the electrical process, for comparison, with the result that the electricallywelded bars were much stronger than those welded by the ordinary process. The bars were of various sizes, up to an inch and a half for iron and three-fourths of an inch, octagon, steel.

A. M. DOLBEAR.

There is, perhaps, but one direct use to which this process is applicable for electrical work, that is for the making of joints in conductors. One of the weak points in electrical construction is the joint. Joints in the copper wire are twisted, and sometimes soldered, the work being very often poorly done, and, as Mr. Woodbury will tell you, connections of that character are liable to cause fire and give a bad reputation to the electrical business; and I might say right here, that although electricity is considered a mystery, we really understand its laws, and what it will do, and how it will do it, although we know not exactly what it is; consequently if everything is done well, if lines are properly constructed and the wires are properly insulated, there will be very

little of this trouble that you hear about, which comes from the cheap construction and the poor methods that are generally in use when an industry of the kind is first being developed.

Mr. Oberlin Smith.-I think that to any one who knew anything about ordinary blacksmithing (which, probably, that critic who spoke about burning up steel might not have done), there would have been no fear of steel being burned by this process any more than in a blacksmith's fire, and not as much. There is nothing mysterious about the action of the electricity on the metal. It simply heats it. That heat is under perfect control. There are no impurities coming up out of the fire-sulphur, smoke and other stuff-to damage the steel; no danger of the fire getting ahead of you; no danger of heating the bottom of the bar while the top remains cool. Of course, we have to heat steel up to a certain definite degree to weld it at all-that is, barely up to its melting point at the surfaces to be united. Now as this process heats it more uniformly, neatly, under much better control, and with a great deal more cleanliness than any possible blacksmith's fire can do it, there cannot possibly be as much danger of burning steel by it as there is in that case. It can be protected by flux, and in other ways, if necessary. All that there is to do to it is to bring it up barely to the melting point, and press it together. It is therefore impossible that the objection in question should have any force.

Mr. Woodbury.-My reply to the interrogatory relative to the electric welding of a ring was very naturally a general answer to a general question. The resistance of the joint in the first place varies with each stage of the process, according to the form and the temperature, and as the mechanical contact of the softened metal becomes more perfect, and as the temperature becomes greater, and finally, as the cross section becomes larger, we bring in conditions which modify the electrical conductivity between the ends of the articles being welded.

It is natural that such questions should be asked in regard to welding a ring, because the process uses electricity of greater current than ever before produced, and the results are without precedent.

One of the old stories among the members of the Boston Bar is that in a litigation over the construction of certain wheels, Daniel Webster was counsel upon one side and Rufus Choate upon the other. Mr. Choate wrought himself into a fury to show that those

wheels could not be made in that way-that it was impossible to make them, and that the whole thing was a fabrication-and after he had very brilliantly stated his points, Mr. Webster motioned to an assistant to lift a cloth which covered certain matter, and merely said for his argument, "Your Honor, and Gentlemen of the Jury, there are the wheels;" and so, in answer to some of the criticisms, I can say, "There are the rings."

The question was asked relative to the enlargement of the joints at the portion of the weld. One difficulty with ordinary hand butt-welding has been the reduction of the cross section by the blacksmith. Here the cross section is somewhat enlarged, and that has not been considered a defect in chains; on the contrary, it has a tendency to prevent a chain from kinking. In some of the special forms of this welding apparatus, there are a pair of swedges that strike a blow on the metal as soon as the weld is effected, for the purpose of reducing both surfaces to a uniform size, as, for example, in the joining of certain classes of work, the operator, by placing his foot upon a treadle, strikes a blow upon the weld and reduces it to a uniform section.

Pieces can be welded with a burr so small as to be unnoticeable for most work; and by shaping the ends to a proper convexity, the weld can be left of size uniform with the remainder of the bar. There is an addition to the process of chain manufacture to which I have not alluded, because it has not yet been developed to a commercial basis, and that is the method of making an electric welded chain by machinery passing the rods into the machine where they are cut, bent, and joined, then welding the chain in a thorough manner, and passing it out at the other end of the device. There have been numerous other processes of manufacture where electric welding is to be used in connection with special or automatic machinery in methods which would never have been thought of in the application of ordinary process of welding where any use of heat in a machine would be entirely unfeasible.

It has been my purpose to limit the paper entirely to a consideration of what has been done without any speculations into the future tense of the possibilities of this new art of electric welding invented by Prof. Elihu Thomson.

CCCXVI.

ON THE DISTRIBUTION OF INTERNAL FRICTION OF

*

ENGINES.

BY ROBERT H. THURSTON, ITHACA, NEW YORK,

(Member of the Society.)

INTRODUCTION.

In earlier papers, read at various times before the American Society of Mechanical Engineers, the writer has called attention to the fact that the variation of load in steam-engines is not productive either of the method or of the amount of engine-friction that has been commonly assumed by earlier authorities on that subject. It was shown that the formula of De Pambour, which makes the internal friction of the engine proportional to the load on its piston is not usually correct, and probably is never so, with any familiar form of engine, or under any conditions often met with in practice. It was further shown that, under the conditions of usual practice, and at all ordinary speeds and pressures of steam, the resistance of the engine itself, its internal friction, remains sensibly constant, and that the so-called friction card of the machine represents practically the friction of the engine when fully loaded, the indicated power without load being sensibly the measure of the wasted work of the engine when in operation under load of whatever amount.

The literature of this branch of the subject of steam engineering is very meager, and the results of experiment in this field, if any have yet been systematically made, are not recorded in any works as yet consulted by the writer. The very natural supposi tion that the friction of an engine is always composed of two parts, the one the friction of the engine unloaded, a constant, and the other a quantity measuring the added friction due to the imposition of the load, and variable directly with that added load, seems to have been accepted by all writers from De Pambour, the first

* Friction of Non-condensing Engines. Trans. Vol. VIII., No. CCXXVIII., and Vol. IX., No. CCLXV.

to attempt to consider the subject, to the period of investigation by the writer. On the other hand, however, engineers familiar with the operation of the engine have been accustomed to take a diagram with the steam-engine indicator, the engine being unloaded, as representative of the friction of the machine at all times. This was probably taken as so representative simply because it was usually impossible to obtain any measure of friction of loaded. engine, and the friction card was thus the best approximation that could be secured. Rankine would seem to have suspected that the assumed formulas of De Pambour might not be exact, as he remarks, "Our knowledge of the amount of energy so lost is still very vague and indefinite;" but he also states (Steam-Engine, art. 292) that "in most cases which occur in practice, a result nearly agreeing with that of the preceding formula, is obtained by supposing that the whole of the prejudicial resistance is proportional to the useful load." De Pambour gives the value of the coefficient by which the load is multiplied as about 0.14, and Rankine asserts this to agree with practice. Weisbach attempts to produce a formula for this waste, assuming Morin's values of coefficients of friction, but his results are very greatly in excess of those to be given as determined by investigations made to ascertain its amount by experiment; they also seem to be based upon entirely inaccurate assumptions, and are evidently quite as unreliable as are those of De Pambour and Rankine.

The first investigation undertaken systematically to determine the law and the methods of waste by internal friction in the steam-engine were, so far as yet known, those directed by the writer, the scheme being the securing of constant conditions, except in the one direction in which variation was to be produced, for the purpose of noting the extent and the law of variation of friction with variation of the one element studied. Thus: The engine was placed under the usual and standard working conditions, but without load, and a friction diagram was taken as a measure of the power wasted in friction of engine alone. The conditions being kept constant in all other respects the load on the engine was varied from this minimum up to and beyond the maximum rated power of the machine, and the indicated compared with the dynamometric power in every case, the difference measuring the engine friction for that power and load. In other cases, the speed of the engine varied, the power and all other conditions being kept constant; the same method applied when the power,