92 Sir Charles A. Parsons [March 10, 



driven, and we shall allude to this gear later as likely to play a very 

 important part generally in future turl:)iue developments. 



In 18S4, or four years previously, I dealt with the turbine problem 

 in a different way. It seemed to me that moderate velocities were 

 essential if the turbine motor was to recei\-e general acceptance as a 

 prime mover. I therefore decided to split up the fall in pressure of the 

 steam into small fractional expansions over a large number of turbines 

 in series so that the velocity of the steam nowhere should be great, 

 and consequently, as we shall see later, a moderate speed of turbine 

 suffices for the highest economy. This principle is now^ universally 

 adopted in all except very small turbines, where economy is of 

 secondary importance. This arrangement of compounding turbines 

 also appeared to me to be surer to give a high efficiency, Ijecause the 

 steam was caused to flow in a non-expansive manner through each 

 individual turbine, and consequently in an analogons way to water in 

 water turbines, where high efficiency at that date had been proved. 

 I was also anxious to avoid the well-known cutting action of high- 

 velocity steam on metal. 



The close analogy between laws for the flow of steam and water 

 under small differences of pressure have been confirmed by experiment, 

 and the usual formula = ^ ^[/h, where // is the hydraulic head, gives 

 the velocity of issue from a jet for steam with small heads and also 

 for water, and we shall presently follow this part of the subject 

 further in dealing with the design of turbines. 



Having decided on the compound principle, it was necessary to 

 commence with small units at first, and in spite of compounding the 

 speed of revolutions was still high. 



Though, as we have said, the de Laval turbine appeared four years 

 later, the de Laval cream separators were in use prior to 1884, and I 

 had the advantage of seeing their beautiful means of balancing — the 

 supporting of the bearings in elastic rubber sleeves, which at 6000 

 revolutions absorbed vibration and allowed the bow 1 containing the 

 milk to rotate about its centre of gravity instead of its geometric 

 centre. The first compound steam turbine ran at 18,000 revolutions, 

 and had slightly elastic bearings. The turbine teeth or blades were 

 like cog-wheel teeth, set at an angle and sharpened at the front 

 edge, and the guide blades were similar. Gradually the form of the 

 blades was improved — curved blades with thickened backs were intro- 

 duced. The blades were cut off to length from brass material rolled 

 and drawn to the required section, and inserted into a groove with 

 soft l)rass packing distance pieces between and caulked up tightly, and 

 dummy labyrinth packings of various types were introduced. The 

 design was improved so as to reduce steam leakages and provide for 

 greater expansion ratios. 



The construction of a suitable dynamo to run with the turbine 

 involved nearly as much trouble as the turbine itself ; the chief 

 features were the adoption of very low magnetic densities in the 



