# What temperatures can be reached in an air-to-air thermocompressor nozzle and why?

People are generally of the opinion that the boiler injector cannot be redesigned to run on air. In other words, an air-to-air thermocompressor that puts fresh air into a tank without a mechanical pump. Assuming there is a large quantity of high pressure air available to drive the injector, what is the procedure to determine the pressure, volume and temperature necessary to entrain large quantities of atmosphere or some other lower pressure air and put it into a 200 psi tank?

It has been suggested that the boiler injector is a bad analogy since the drive air cannot condense in the nozzle as it does in the boiler injector. If I knew how to compute obtainable temperatures, I would have a better idea whether or not air could condense for a split second so its latent heat could contribute to resisting the tank pressure.

As I understand it, the pressure will become velocity in the converging nozzle due to expansion and some of the pressure will come back in the diverging nozzle. According to PV/T = constant, there's no problem getting to -320 deg. F. with high pressure air, say 1000 psi or whatever, but this doesn't take flow or dynamics into account and I don't know where to start. So far this question is stumping people and search engines, so I am thinking the answer is unknown.

Even if there is no way for air to condense in a nozzle (which I doubt), I still need to know how to estimate the requirements of a conventional (non-condensing) air-to-air thermocompressor.

-
I think the pressure difference (high to 200 psi) on the nozzle has to create a supersonic wave to induce the flow. A pulse flow may work a check valve. An mechanical air motor driven compressor may be more efficient, possibly a free piston. – Optionparty Apr 13 '13 at 15:26
Like a reverse Glockemann diaphragm pump, but for air. – Optionparty Apr 13 '13 at 18:14