Abstract
Theoretical and experimental investigations of a novel gas-dynamics device having no moving parts yet performing the function of a compressor, are described. This device, called the “Aerothermopressor,” exploits the possibility of raising the total pressure of a high-speed gas stream through cooling of the gas. When placed at the exhaust of a gas turbine, the Aerothermopressor will reduce the exhaust pressure, thereby improving both fuel economy and power capacity per unit of air flow. Basic elements of the apparatus comprise a nozzle which accelerates hot gas into an evaporation section; a water-injection system which delivers finely atomized water into the high-speed stream; an evaporation section in which the gas is cooled and most of the water evaporated; and, finally, a diffuser in which the gas stream is decelerated and the static pressure increased.
Although the Aerothermopressor is simple in structural arrangement, the physical processes occurring within it are exceedingly complex in their details. The simultaneous effects on the gas stream of droplet drag, evaporative cooling, area variation, and wall friction lead to many regimes of operation, including the hitherto unknown passage from subsonic to supersonic speeds in a constant-area duct. Theoretical calculations of a one-dimensional nature, involving for the gas stream the equations of continuity, momentum, and energy, and for the liquid-droplet cloud the equations of motion, heat transfer, and mass transfer, have been carried out on a high-speed, electronic digital computer. The theory reproduces all the behavior patterns of experimental units and is in generally good quantitative agreement with the experimental data.
The results of experiments on a small-scale, constant-area unit of 2.13 in. diam are presented and compared with theoretical calculations. The experiments and theory both show that a net stagnation rise is possible only with gas flows greater than about 2 lb/sec; below this value the detrimental effects of wall friction completely absorb the gains due to cooling. In the range of 25 lb/sec a net stagnation pressure rise of about 10 per cent seems assured, while 20 per cent seems possible.
Early tests on a recently completed medium-scale unit of 25 lb/sec capacity have already demonstrated a net overall rise in stagnation pressure.