Lysholm Machines as Two-Phase Expanders

Power recovery from the expansion of saturated liquids and wet vapours is discussed. Applications in large scale refrigeration systems, and in the recovery of power from low grade heat sources are reviewed. Twin screw machines are well suited for this function and a computer simulation of their operation in this mode is described. Performance estimates are compared with test results carried out by the authors using Rll3, us investigators using water and Japanese workers using R12 as the working fluid. Agreement at tip speeds of up to 30 m/s is good but mass flow rates are overestimated at higher speeds. Further experimental work is proceeding to resolve this. Adiabatic efficiencies of the order of 80% appear to be possible in large machines. INTRODUCTION The expansion of saturated liquids or wet vapours has been recognised as a potential means of generating power for more than seventy years [1] . Unfortunately, the practical problems of expanding a two-phase mixture efficiently proved an insuperable obstacle. Post world war two interest in geothermal power led to consideration of the use hot pressurised water or very wet steam, emerging directly from natural aquifers, as a power plant working fluid. The normal means of achieving this is to flash the fluid to some intermediate pressure, separate the steam from the vapour, expand the dry steam in a turbine and reinject the separated hot water. In the early nineteen seventies, Sprankle [2] proposed the use of the Lysholm twin screw machine to expand the "total flow" of steam and water to recover power without separation and thereby to increase the recoverable power by 50 100%. This stimulated a great deal of investigation of both twin-screw machines and turbines [3-7] as two-phase expanders. The results of these investigations were largely disappointing. Rotor sizes of up to 28 ft diameter were predicted for large scale geothermal power plant using screws while measured adiabatic efficiencies of such machines using water were only of the order of 50%. Performance analysis of these machines was limited largely to parametric studies [8] . Turbines, although smaller, were also found to be relatively inefficient, with water as the working fluid, and maximum predicted efficiencies hardly exceeded 65%. Using organic fluids in a closed cycle, Elliott [3] predicted maximum turbine efficiencies of up to 70%. Independently, one of the authors considered closed cycle systems with organic working fluids as a means of recovering power from hot water sources in a system similar to that proposed by Elliott but using a twin-screw expander [9] . The system, described as a Trilateral Flash Cycle (TFC) system is shown in Fig 1. The use of organic fluids reduces the size of the expander required by a factor of 10 and the volume ratio