The following text may have been generated by Optical Character Recognition, with varying degrees of accuracy. Reader beware!

'TAE Marine REVIEW 31 ficed and two screws used, only one propeller per shaft "should be adopted. Before leaving this subject, it may be as well to impress on those undertaking such designs, and while excellent results can be and' have been obtained with this system of propul- sion, it is only by very careful attention to the design and arrangement of propellers for efficient working it is im- possible to assume revolutions and later on to design a pro- peller to suit. The determination of the details of eating arrangements to secure the best results also' necessitates considerable practice; there are few formulae for guidance, such 'as- are found in reciprocating engine work. The general features of the design and the action of the | steam in the blades of a Parsons turbine is,. probably, well known to most of the members here. The total expan- sion of the steam is approximately adiabatic, and is sub- divided into a number of steps, in each of which is main- tained a certain dynamic relationship between the steam speed and that of the vanes. Each stage consists of a ring of stationary blades, which give direction and velocity to the steam, and a ring of moving blades that immediately convert the energy of velocity into useful torque. The total torque on the shaft is due to the impulse of steam enter- ing the moving blades, and to reaction as it leaves them, this process being repeated throughout the turbine. The best relationship of steam speed and blade speed is largely a matter of opinion, and equally good results have been obtained with wide differences. Probably 'little is gained by departing in any way from the best theoretical value. - The vane speed V7 is, like many other things, provided that the number of rows of blades is correct, a matter of con- ' venience and is modified by experience; wide differences in this also can be made without tangibly affecting the efficiency. Assuming a speed such as is given in table 1, the diameter of the turbine is found from the formula: _. -Blade velocity.in ft p. sec. X 228 Re Pie Diameter '" The velocity acquired by the expansion of steam (=VS) may be found by calculating the B. Th. U. drop between the two pressures, whence, OP Sem aaas/ se hi Us: The ratio of blade speed to steam speed being settled, the drop at each row may then be determined, which will give the number -of rows of blades required, since the initial and ter- minal working pressures are known. No formula such as pt ALN 33,000 but experience has shown that there are various formulae which can be arrived at from past practice, which will give a very suitable combination of dimensions. Such methods are exists for determining the main dimensions, 'the result of long and costly experiments, and it is scarcely surprising that builders should forbear to publish them. The blade velocity adopted in practice varies largely ac- cording to the class ,of work; generally it is much lower in marine work than with the high revolutions that are adopted for generators in electrical work. While that for . marine turbines can barely exceed 200 feet per second as a maximum, without great sacrifice of propeller efficiency, it is More usual to adopt a far lower speed, say 100 to 140 feet per second. The Rateau and Curtis types of turbine require far higher velocities, the former about one and one-half times and the latter at least twice these speeds, and in marine work it is almost impossible to obtain them. ~ The blade heights are made variable to allow for the ex- pansion of the steam and the desired adjustment of veloci- ties throughout the turbine necessitates intricate calcula- 4 2 tion, involving more space than is at the author's disposal. The width and pitch of the blades are partially arbitrary and, to some extent, a matter of strength and convenience. Ample longitudinal clearance can be allowed, varying in practice up to 3% in. with large blades, but the tip clearance -- is more important, and must be-made as fine as possible. *The blade tips are usually ground up and their diameters made accurate to thousandths of an inch: clearance depends largély on temperature and diameter, and varies in amount from thickness of a dime to that of a dollar. The blade heights should vary from about 4 percent to 14 percent of the mean diameter, in which case this clear- ance only represents a very small percentage of loss. Vari- ous metals have been used for the blading; generally a form of cheap brass has been employed for both blade and packing piece. Steel blades have been used in this country, while the best metal yet found for work with high tem- perature steam is an alloy containing 80 percent of copper and 20 percent of nickel; this being especially strong and durable, though almost pure copper gives excellent results. The usual form of caulking piece adopted is shown by the shaded portion in Fig. XXII, which outlines the action of steam in the blades. The distorting effect of caulking on large cylinders is distinct unless they are well designed, and the assembling process may appear slower than it really is. The author is convinced that this form of caulking-piece must largely disappear in favor of a machine-divided strip, such as some firms are now using. This form of blading not only ensures perfect accuracy but is a remarkably cheap one, saving several processes over the old method. The slotted strip (see Fig. XXV) is associated with a shallow shroud ring into which the blades are fitted, being thus held at the tips as well, the whole arrangement forming a very strong and durable method of attachment. As to blade- stripping, nearly every case may be _attri- buted to carelessness in design, manufacture and operation. The unequal distortion of cylinders by heat--a matter of great importance in marine turbine work where flexible shaft couplings are impossible--accounts for most stripping; the passage of grit or water is not so formidable as might be imagined. Any whipping of the turbine spindle will destroy the tip clearances (and the blades too), but careful design should never permit this; defective or unsuit- able blade material is another preventable cause, and there is also the possiblity of bad workmanship. In a properly designed and well-made turbine the chance of blade- -strip- ping is very minute. The bearing pressure permissible in marine turbines is about 80 to 90 pounds per square inch; the circumfer- ential speed of the shaft, however, must be kept down to about 25 to 30 feet per second, if such a pressure is used; in electrical work 50 pounds and 50 feet per second is mote usual. Under these conditions and with ample oil supply the necessity for adjustment will be very rare. Rigid bearings, lined with white metal, are used. The centrifugal stresses are very low in marine * achias spindles; in fact about 200 feet per second is practically the highest vane speed obtainable, in which case the stress will not exceed 4,000 pounds per sauate inch, but usually it is very much lower. The propeller thrust is balanced by the action of the steam in a very satisfactory manner, but this calculation involves knowledge of effective thrust along the shaft in order to accur+ ately proportion the annular area on which the steam pressure is to act. The result is that the spindle is in compression and the cylinder in tension when the turbine is working; a mar- gin is always allowed so that the propeller thrust exceeds the steam thrust.