200 BUCKINGHAM: BALANCING OF TURBINES 



make a sketch design and determine approximately the form of 

 the curve of pressure distribution thru the turbine at the given 

 power. It is then easy, if the diameters of drum and shaft have 

 been fixed, to decide what pressure should act on the H.P. end 

 of the drum in order that steam and propeller thrusts shall be 

 balanced. The pressure drop to this point is to be taken upon 

 wheels, and the remainder on the drum. If the data are good 

 and the computations correctly made, the rotor and shaft will 

 float at some speed close to that assumed. 



In general, it will be found that the length of the drum computed 

 in this way is much less than would seem advisable if the question 

 of thrust-balancing were not considered — there will be too many 

 wheel stages and the efficiency will suffer because not enough 

 blade-rows can be got into the prescribed length. With a tur- 

 bine of large diameter, a satisfactory compromise is impossible. 

 The writer has seen one design of turbines for a battleship, in 

 which the drum was long enough that at full power the steam 

 thrust was about 60 per cent greater than the propeller thrust, 

 while the drum was still much shorter than was desirable from the 

 point of view of economy. 



The obvious escape from the conflict of requirements is the 

 adoption of some double-flow scheme. This complication may 

 not be at all worth while in small machines, but for large powers 

 when a division of the turbine into two lighter parts is advan- 

 tageous in itself, the double-flow arrangement offers a simple 

 solution of the difficulty. Let the H.P. rotor consist of the wheel 

 stages and the first drum while the L.P. rotor consists of a second 

 drum — combined of course with the backing turbine. The 

 steam flows aft thru the H.P. turbine, thence to the after 

 end of the L. P. drum, and exhausts at the forward end of 

 the L.P. casing. The two drums thus exert opposing thrusts 

 and no matter what the pressure at the beginning of the H.P. 

 drum, the intermediate pressure may be so chosen as to give the 

 resultant steam thrust due to both drums any value we please. 

 We are thus able to design the arrangement of stages as seems 

 most desirable, regardless of thrust, using a long drum and only 

 a few wheels. We then split the drum at the right point on the 



