CCCXXVII. ON THE FRICTION OF PISTON PACKING RINGS IN STEAM CYLINDERS. BY JAMES E. DENTON, HOBOKEN, N. J. A MEASUREMENT of such friction has been made with a measuring device which consists essentially of the following arrangement: A cylinder M (Fig. 75), 6 inches bore by 9 inches stroke is fitted with a piston A long enough to permit a packing ring Cone inch wide to occupy the position shown. The ordinary packing rings B, B, preventing the access of steam into the space* immediately surrounding C. The latter is supported upon the outer ends of the levers, D, D, which are pivoted at 0, and have their inner ends coupled to the rod E. Motion is given to the piston A and its attachments through the piston rod P. Motion being in the direction of the arrow, the friction of the ring C tilts the levers D, thus compressing the spring F. The resulting movement of the rod E, relative to the incasing tube N gives motion to a pencil lever J through the pitman G and the crank H. Consequently the motion of the pencil Sperpendicular to the plane of the paper is proportional to the amount of friction of the ring C. The pencil makes a diagram resembling a rectangle upon paper fastened to a board K, K. The ring C is cut once and is provided with a device by the means of which its tension may be adjusted by a spring. Means are also provided whereby the ring may be drawn together so as not to touch the sides of the cylinder. When in the latter condition the spring Ft is calibrated by loading the rod E at Q with known weights and noting the resulting movement of the pencil S. Fac-similes of diagrams are given below. No. 1 (Fig. 76) is the * This space is kept drained of condensed steam. The spring actually resisting the motion of the levers D is the torsion of the pivots 0. The spring F is merely shown as an illustration of spring action. friction at 65 pounds tension per square inch upon the spring C (a boiler pressure of 67 lbs., revolutions from 36 to 115, and any cutoff from to ), when the parts of the piston and cylinder are thoroughly devoid of lubricant through having been soaked in naphtha. The scale of the diagrams is 250 lbs. per inch of width each side of the center line CC; that is, the distance d being about of an inch, the friction of the ring C is 250 about 80 pounds, of force for the down stroke. 5 Similarly the distance d, being about Diagram No. 2 (Fig. 77) shows the friction for a feed of cylinder oil of drop per minute * or one drop in two minutes. It gives a coefficient of friction of about 5%. B B Diagram No. 3 (Fig. 78) is for an oil feed of one drop per minute and shows an average coefficient of about 3%. Both diagrams, Nos. 2 and 3, afforded unsatisfactory lubrication, the piston groaning at the ends of the stroke when the engine was run slowly, and the film of oil found upon the interior surfaces was a sticky black paste showing by chemical analysis about 50 per cent. of iron. APPARATUS FOR MEASURING Diagram No. 4 (Fig. 79) is for a feed of two drops per minute, and shows a coefficient of about 1%. The oil upon the interior surfaces for this diagram indicated practically perfect lubrication, as it retained its natural color and was uncontaminated with iron. All of the diagrams were taken after the engine had been run at the respective feeds for about 8 hours. * The oil used was an average cylinder oil of good reputation, and 5,000 drops represented a pint. All effect of inertia is eliminated by making the pivot 0, of the levers D, coincident with the center of gravity of the ring, the levers D and the rod L with its attachments. DISCUSSION. Prof. R. H. Thurston.-I have been very much interested in the device which Prof. Denton has described. It strikes me as very ingenious and likely to be very useful. I should think the results would probably give us some facts and some figures that would be of great value. It is one of the prettiest things that I have seen in a long time, and I am glad to note the results reported. Mr. George Schuhmann.-The apparatus described by Prof. Denton is a very ingenious instrument, but the conditions under which the tests were made do not correspond with actual engine practice, because the rings B, B keep the steam away from the ring C, and so prevent certain particles of steam from forcing their way between the inner surface of the cylinder and the outer surface of the ring, which has some influence on the friction of the ring. This is very fully described in Prof. Robinson's paper on "A Rational System of Piston Packing," Vol. II., p. 27, and also in a paper by the same author, "Back Pressure on Valves," Vol. IV., p. 150. While I have not made any experiments on piston rings, I have made enough with valves to satisfy myself that his theory of what he calls "the pressure of the creeping fluid" is correct. For this same reason I cannot see how Prof. Thurston's experiment of measuring the friction of the piston can give exact results when there is no steam at all in the cylinder. I should think the best way to ascertain the friction of the piston rings would be to take an indicator card of the engine running light and with the rings in position, then remove the rings and run the engine without them, take another card, and the difference will be the friction of the rings. Of course, there would be a waste of steam owing to leakage around the piston, but I do not think this would prevent the indicator from giving an accurate account of the power developed. Mr. Wm. E. Crane.-Prof. Denton speaks of water in steam being a lubricant. Some years ago a 30 x 60 Corliss engine was erected, and supplied with steam through a 10" pipe over 200 feet long. The engine was started up and run for nearly a month with uncovered pipe. During this time it required a large amount of cylinder oil. The pipe was finally covered, and the amount of oil required was materially reduced. After some four years it became necessary to replace this pipe with a new one, and while the new pipe remained uncovered there was the same difficulty of properly lubricating the cylinder, which difficulty was removed by covering the pipe. Of course, the condensation was very great in the uncovered steam pipe, and showed that the presence of water in the steam required more oil for lubrication. Prof. Denton also mentions the fact that, where steam is cut off short, more oil is required than where the cut-off is late. If we could entirely prevent cylinder condensation, the problem of cylinder lubrication would be much simplified; and, if there were no re-evaporation, no more oil would be required with a short cutoff than with a long one. Some one has claimed that oil is not needed in a cylinder, and as proof of such statement mentions the fact that many engines have been run for years where it was not possible to get oil in the cylinder. It is true that engines were so run, but it is equally true that the friction in the same is very great. This might be illustrated by a homely incident. On the old style of locomotives they used a plain slide valve for the throttle valve, and, to make these throttle valves work more easily, an oil cup was placed in the dome just over the steam pipe, with a small pipe leading down to the valve. It was the duty of the fireman, before any pressure was raised, to pour some oil down this cup and pipe, which oiled the throttle valve. Should this small oil pipe become knocked one side, the oil would go into the boiler instead of into the steam pipe. A locomotive was sent to do duty on a branch road, and the engineer, fearing that this oil pipe might have become displaced, would not allow the fireman to put any oil in, for fear that it would go into the boiler and cause foaming. The fireman had all the switching to do, and was anxious that the throttle should work easier, and one morning he disobeyed orders and poured some oil into the cup. When he started the machine out of the house, he saw that the throttle was all right, but thought he would let the engineer find it out himself. The train was made up and the signal given. The engineer, according to his usual custom, braced his feet, took hold of the handle with |