steam-engines are to be worked by the steam generated by the release of pressure from this water, and the valves are to be arranged in such a way that the steam shall work at 130 lbs. pressure. A reservoir 8 ft. in diameter and 30 ft. long, containing 84,000 lbs. of heated water at 250 lbs. pressure, would supply 5250 lbs. of steam at 130 lbs. pressure. As the steam consump tion of a condensing electric-light engine is about 18 lbs. per horse-power hour, such a reservoir would supply 286 effective horse-power hours. In 1878, in France, this method of storing steam was used on a tramway. M. Francq, the engineer, designed a smokeless locomotive to work by steampower supplied by a reservoir containing 400 gallons of water at 220 lbs. pressure. The reservoir was charged with steam from a stationary boiler at one end of the tramway. Cost of Steam-power. (Chas. T. Main, A. S. M. E., x. 48.)-Estimated costs in New England in 1888, per horse-power, based on engines of 1000 H.P. 28. Total yearly expense, lines 10, 20, and 27.. 24.087 ... 80. Total if all ex.-steam is used for heating... When exhaust-steam or a part of the receiver-steam is used for heating, or if part of the steam in a condensing engine is diverted from the condenser, and used for other purposes than power, the value of such steam should be deducted from the cost of the total amount of steam generated in order o arrive at the cost properly chargeable to power. The figures in lines 29 29.855 83.248 12.597 and 30 are based on an assumption made by Mr. Main of losses of heat amounting to 25% between the boiler and the exhaust-pipe, an allowance which is probably too large. See also two papers by Chas. E. Emery on "Cost of Steam Power," Trans. A. S. C. E., vol. xii, Nov, 1883, and Trans. A. I. E. E., vol. x, Mar. 1893. ROTARY STEAM-ENGINES. Steam Turbines.-The steam turbine is a small turbine wheel which runs with steam as the ordinary turbine does with water. (For description of the Parsons and the Dow steam turbines see Modern Mechanism, p. 298, etc.) The Parsons turbine is a series of parallel-flow turbines mounted side by side on a shaft; the Dow turbine is a series of radial outward-flow tur. bines, placed like a series of concentric rings in a single plane, a stationary guide-ring being between each pair of movable rings. The speeds of the steam turbines enormously exceed those of any form of engine with recip. rocating piston, or even of the so-called rotary engines. The three- and fourcylinder engines of the Brotherhood type, in which the several cylinders are usually grouped radially about a common crank and shaft, often exceed 1000 revolutions per minute, and have been driven, experimentally, above 2000; but the steam turbine of Parsons makes 10,000 and even 20,000 revolutions, and the Dow turbine is reputed to have attained 25,000. (See Trans. A. S. M. E., vol. x. p. 680, and xii. p. 888; Trans. Assoc. of Eng'g Societies, vol. viii. p. 583; Eng'g, Jan. 13, 1888, and Jan. 8, 1892; Eng'g News, Feb. 27, 1892.) A Dow turbine, exhibited in 1889, weighed 68 lbs., and developed 10 H.P., with a consumption of 47 lbs. of steam per H.P. per hour, the steam pressure being 70 lbs. The Dow turbine is used to spin the fly-wheel of the Howell torpedo. The dimensions of the wheel are 13.8 in. diam., 6,5 in. width, radius of gyration 5.57 in. The energy stored in it at 10,000 revs. per min. is 500,000 ft.-lbs. The De Laval Steam Turbine, shown at the Chicago exhibition, 1893, is a reaction wheel somewhat similar to the Pelton water-wheel. The steam jet is directed by a nozzle against the plane of the turbine at quite a small angle and tangentially against the circumference of the medium periphery of the blades. The angle of the blades is the same at the side of admission and discharge. The width of the blade is constant along the entire thickness of the turbine. The steam is expanded to the pressure of the surroundings before arriving at the blades. This expansion takes place in the nozzle, and is caused simply by making its sides diverging. As the steam passes through this channel its specific volume is increased in a greater proportion than the cross section of the channel, and for this reason its velocity is increased, and also its momentum, till the end of the expansion at the last sectional area of the nozzle. The greater the expansion in the nozzle the greater its velocity at this point. A pressure of 75 lbs. and expansion to an absolute pressure of one atmosphere give a final velocity of about 2625 ft. per second. Expansion is carried further in this steam turbine than in ordinary steamengines. This is on account of the steam expanding completely during its work to the pressure of the surroundings. For obtaining the greatest possible effect the admission to the blades must be free from blows and the velocity of discharge as low as possible. These conditions would require in the steam turbine an enormous velocity' of periphery-as high as 1300 to 1650 ft. per second. The centrifugal force, nevertheless, puts a limit to the use of very high velocities. In the 5 horsepower turbine the velocity of periphery is 574 ft. per second, and the num ber of revolutions 30,000 per minute. However carefully the turbine may be manufactured it is impossible, on account of unevenness of the material, to get its centre of gravity to correspond exactly to its geometrical axle of revolution; and however small this difference may be, it becomes very noticeable at such high velocities. De Laval has succeeded in solving the problem by providing the turbine with a flexible shaft. This yielding shaft allows the turbine at the high rate of speed to adjust itself and revolve around its true centre of gravity, the centre line of the shaft meanwhile describing a surface of revolution. In the gearing-box the speed is reduced from 30,000 revolutions to 3000 by means of a driver on the turbine shafts, which sets in motion a cogwheel of 10 times its own diameter. These gearings are provided with spiral cogs placed at an angle of about 45°. For descriptions of the most recent forms of steam turbines, see circulars of the Westinghouse Machine Co., Pittsburg, Pa., and the De Laval Steam Turbine Co., Trenton, N. J.; also paper by Dr. R. H. Thurston in Trans. A. S. M. E., vol. xxii., p. 170. Rotary Steam-engines, other than steam turbines, have been invented by the thousands, but not one has attained a commercial success, as regards economy of steam. The possible advantages, such as saving of space, to be gained by a rotary engine are overbalanced by its waste of steam. Rotary engines are in use, however, for special purposes, such as steam fire-engines and steam feeds for sawmills, in which steam economy is not a matter of importance. DIMENSIONS OF PARTS OF ENGINES. The treatment of this subject by the leading authorities on the steam-engine is very unsatisfactory, being a confused mass of rules and formulæ based partly upon theory and partly upon practice. The practice of builders shows an exceeding diversity of opinion as to correct dimensions. The treatment given below is chiefly the result of a study of the works of Rankine, Seaton, Unwin, Thurston, Marks, and Whitham, and is largely a condensa tion of a series of articles by the author published in the American Ma chinist, in 1894, with many alterations and much additional matter. In or der to make a comparison of many of the formulæ they have been applied to the assumed cases of six engines of different sizes, and in some cases this comparison has led to the construction of new formulæ. Cylinder. (Whitham.)-Length of bore stroke + breadth of pistonring to in; length between heads = stroke + thickness of piston sum of clearances at both ends; thickness of piston = breadth of ring thickness of flange on one side to carry the ring+thickness of followerplate. Thickness of flange or follower.... % to 1⁄2 in. 34 in. 1 in. 60 to 100 in. Clearance of Piston. (Seaton.)-The clearance allowed varies with the size of the engine from 1% to 3% in. for roughness of castings and 1/16 to in. for each working joint. Naval and other very fast-running engines have a larger allowance. In a vertical direct-acting engine the parts which wear so as to bring the piston nearer the bottom are three, viz., the shaft journals, the crank-pin brasses, and piston-rod gudgeon-brasses. Thickness of Cylinder. (Thurston.)-For engines of the older types and under moderate steam-pressures, some builders have for many years restricted the stress to about 2550 lbs. per sq. in. is a common proportion; t, D, and b being thickness, diam., and a constant added quantity varying from 0 to 1 in., all in inches; p, is the initial unbalanced steam-pressure per sq. in. In this expression b is made larger for horizontal than for vertical cylinders, as, for example, in large engines 0.5 in the one case and 0.2 in the other, the one requiring re-boring more than the other. The constant a is from 0.0004 to 0.0005; the first value for vertical cylinders, or short strokes; the second for horizontal engines, or for long strokes. Thickness of Cylinder and its Connections for Marine Engines. (Seaton).-D = the diam, of the cylinder in inches; p= load on the safety-valves in lbs. per sq. in.; f, a constant multiplier thickness of barrel.25 in. Thickness of metal of cylinder barrel or liner, not to be less than p × D + 8000 when of cast iron.* (2) Thickness of liner when of steel p X D + 6000+ 0.5 66 metal of steam-ports 66 valve-box sides = 0.6 × f. (4) When made of exceedingly good material, at least twice melted, the thickness may be 0.8 of that given by the above rules. Thickness of metal of valve-box covers = 0.7 × f. 66 cylinder bottom = 1.1 .6 cylinder flange 66 covers 66 X f, if single thickness. 66 = 0.6 Xf, if double 66 cover-flange valve-box" = 0.9 × f. = 1.2 Xf. 64 66 66 66 = 1.0 Xf, when there is a false-face. = 0.8 xf, when cast iron. = 0.6 Xf, when steel or bronze. Whitham recommends (6) where provision is made for the reboring, and where ample strength and rigidity are secured, for horizontal or vertical cylinders of large or small diameter; (9) for large cylinders using steam under 100 lbs. gauge-pressure, and This is a smaller value than is given by the other formulæ quoted; but Marks says that it is not advisable to make a steam-cylinder less than 0.75 in. thick under any circumstances. The following table gives the calculated thickness of cylinders of engines of 10, 30, and 50 in. diam., assuming p the maximum unbalanced pressure on the piston = 100 lbs. per sq. in. As the same engines will be used for calculation of other dimensions, other particulars concerning them are here given for reference. The average corresponds nearly to the formula t = .00037 Dp +0.4 in. convenient approximation is t = .0004Dp+0.3 in., which gives for Diameters. Thicknesses.. ....... 20 30 A 10 40 50 60 in. .70 1.10 1.50 1.90 2.30 2.70 in. The last formula corresponds to a tensile strength of cast iron of 12,500 lbs., with a factor of safety of 10 and an allowance of 0.3 in. for reboring. Cylinder-heads.-Thurston says: Cylinder-heads may be given a thickness, at the edges and in the flanges, exceeding somewhat that of the cylinder. An excess of not less than 25% is usual. It may be thinner in the middle. Where made, as is usual in large engines, of two disks with intermediate radiating, connecting ribs or webs, that section which is safe against shearing is probably ample. An examination of the designs of experienced builders, by Professor Thurston, gave t = .005D Vp +0.25. D being the diameter of that circle in which the thickness is taken. Thurston also gives Marks gives t = 0.003D Vp. He also says a good practical rule for pressures under 100 lbs. per sq. in, is to make the thickness of the cylinder-heads 114 times that of the walls; and applying this factor to his formula for thickness of walls, or .00028pD, we have t = .00035p D. which is equal to .0005pD+.25 inch. Whitham quotes from Seaton, t = 2 2000 Seaton's formula for cylinder bottoms, quoted above, is (4) (5) (6) t = 1.1f, in which f= .0002pD+.85 inch, or t = .00022pD + .93. Applying the above formula to the engines of 10, 30, and 50 inches diameter, with maximum unbalanced steam-pressure of 100 lbs. per sq. in., we have |