66 3 THE RAILROAD AND [ February, 1887. and, at the same instant, readings from the brake were ob- tained. A comparison of the power indicated by the diagrams and that shown by the brake gave a difference which measured the friction of the engine. During the trial, the engine, when working at its rated power, con- sumed. according to the indications of the diagrams, 28.2 pounds of steam per horse-power per hour, or, probably, between 35 and 38 pounds, allowing for the loss by cylin- der condensation, not accounted for on the indicator card, a very excellent performance for an engine of but 35 horse-power. The action of the governor was extraordi- narily perfect. The engine was adjusted to make 230 revolutions per minute under 90 pounds steam pressure. The observers reported that it made the same number of turns whether loaded or unloaded, an evident impossibility with a governor of this class, in which only approximate isochronism can be attained. The writer, to settle the question, counted the revolutions, minute by minute, with a hand-speed counter, and made it 230 revolutions with tween centers, a balanced valve with stroke of 2 to 4 inches, according to position of governor and eccentric, a fly-wheel 50 inches in diameter, weighing 2,300 pounds, the steam and exhaust-pipes having diameters of 2% and 4 inches, respectively, and the whole machine weighing 2% tons. The space occupied by the engine was 9 feet 4 inches in length, by 4 feet 8 inches in width, and 3 feet IO inches in height. Examining the above table of powers, it is seen that the difference between indicated and dynamometric power, the friction of the engine, varies somewhat, with varying steam pressures and varying total power; but in such manner as to indicate the controlling cause to be irregular in action, and possibly to some extent due to errors of observation and to accident. The maximum is four horse- power, the minimum about two horse-power. The usual difference is about three and the variations are irregularly distributed throughout the whole range of experiments. It is evident at a glance that the law of De Pambour does .»z/[£73900 OF/1 rm 0/1//va my //VD/0/1 T0125. 8 179/6 the whole rated load on the engine (35 to 40 horse-power), and 231 when entirely unloaded, the brake-strap being loosened until it could be shaken about on the pulley, by the hand, with perfect ease. This was repeated until no question could longer exist in regard to the matter. The variation with variable steam pressure was greater. The following are the data obtained from the brake and indicator readings : Number Steam Brake Indicator Friction of Card. Revolut’ns. Pressure. H. P. H. P. Diff, per cent. I 232 so 4-06 ML 335 45 2 229 65 $.98 7.58 2.60 1 34 3 230 63 .00 10.00 4.00 40 4. 230 69 7.00 10.27 3.20 32 5 230 73 8.10 11.75 3.65 32 6 230 77 9.00 12.70 3.70 29 7 230 _ 75 10.00 14.02 4.02 28 8 230 80 11.00 14.78 5.78 25.5 9 230 80 12.00 15.17 3.17 21 IO — 230 85 13.00 15.96 2.96 18,5 :1 230 75 14.00 16.86 2.86 17 12 230 70 15.00 17.80 2.80 15.75 13 231 72 20.1 22.07 2.06 14. 230 75 25.00 28.31 3.36 11.75 :5 229 60 29-55 33-04 3-15 9.5 16 229 58 34.86 37.20 2.34 6.3 :7 229 70 39-85 43-04 3-19 7-4 I8 230 85 45.00 47.79 2-75 5-8 19 230 90 50.00 52.60 2.60 4.0 20 230 85 55.00 57.54 2.54 4.4 This engine was 8 inches in diameter of cylinder, 14 inches stroke of piston, having a rod 44 inches long be- not hold, and» that it is as nearly correct to say that the friction of engine is constant as otherwise. The column of friction, as given in percentages of the total power, ex- hibits the same fact. There is continual, though some- what irregular, reduction of the ‘percentage of friction, throughout the range from the lowest to the highest power, and very nearly inversely as the power exerted. This is best shown by the curve given in the accompany- ing plate (Fig. 23), in which a smooth line has been drawn to represent as nearly as possible the mean of all observa- tions. It is evidently more nearly correct to assert that the friction of a non—condensing engine of this class is constant, and independent of the total power developed, than to accept the rule of De Pambour. The power for which the engine is proportioned is 35 to 40 horse-power. At this power, the friction of engine is but about 6 per cent. of the total, or less than one-half that assumed by De Pambour, and accepted as. correct by Rankine, for engines generally, and presumably for locomotives especi- ally. The result is exceedingly gratifying, and seems to the writer extraordinary for so small an engine. The repetition of the experiment upon an engine of an- other make, having a cylinder 9 inches in diameter and a stroke of piston of 12 inches, which would naturally give