Paper 10: The Influence of Moisture on the Efficiency of a One-Third Scale Model Low Pressure Steam Turbine

The rapid increase in blade-tip diameters and peripheral speeds of low pressure turbines in large 3000 rev/min turbo-generators has presented the designer with many difficult mechanical and aerodynamic problems.

To assist in the aerodynamic development of such blading, design studies on an experimental low pressure (l.p.) turbine were started early in 1959. Economic and technical considerations limited the choice to a one-third scale model steam turbine capable of running at three times the normal rotational speed to preserve full-scale working Mach numbers on the blading. Overall output and steam consumption were measured on a hydraulic dynamometer and by volumetric tanking respectively. The inlet steam temperature was controlled by a direct injection desuperheater so that the expansion could be kept dry for traversing or reduced to design inlet temperatures for normal wet running.

Three multi-stage sets with last row blade diameters corresponding to 90-in, 120-in, and 136-in full-scale turbines have now been tested in the experimental turbine and the wet performance of the largest forms the subject of this paper.

The overall wetness losses on the model 136-in diameter turbine have been assessed from a series of seven tests in which the inlet superheat and rotational speed were varied whilst maintaining fixed inlet and outlet pressure levels. To isolate the stage moisture correction factor (α), however, where

a stage-by-stage approach was adopted, in which the dry stage efficiencies were initially established from interstage traverses under dry steam conditions.

Two wet steam analyses were made, the first assuming equilibrium and the second supersaturated expansion, but the choice seemed immaterial since the moisture correction factor was almost the same for both.

In the case of the supersaturated expansion calculation, it was necessary to establish the point of reversion from supersaturated to near equilibrium expansion (the Wilson point) and supplementary water extraction results were used to establish the maximum supersaturation ratio. These suggest that the maximum level is nearer to 3 in the model turbine than to the value of 4–6 quoted for convergent-divergent nozzles.