View Full Version : open cycle R718 A/C that also provides high purity water?

23-07-2006, 07:22 PM
I got the idea while thinking about how a "direct compression" distiller works.
Basically, tap water is let into an evaporator chamber through an expansion valve. There is a vacuum in a chamber so the water "cold boils". An independent heat exchanger cycle cools water that then flows through a radiator that is used to cool the room. (This way, it would be easy to clean the evaporator.) Some water is pumped out to clean out any impurities left behind, using a counter-flow heat exchanger to precool the incoming water. The waste water then subcools the product water and is then either used or disposed of. A centrifugal compressor (either oil-less or using vegetable oil) takes the vapor from the evaporator chamber and pumps it into the outdoor condenser, where the vapor recondenses into a liquid. The (very pure) liquid water is then pumped into a storage tank for use. Since tap water contains some dissolved gases, a purge system after the condenser pumps out non-condensables.

I can just imagine its applications. What about a water vending machine that uses the cooling function to cool the product water? Or a bottled water factory that is cooled down by the distiller?

Erik Detroit
23-07-2006, 11:27 PM
Back of the envelope calculation gives me a vacuum pump of 3600 liters/min displacement for every kW useful cooling. I have a 5 cfm (cubic foot per minute) vacuum pump (about 2 liters/min) that could easily fit into a vending machine. I don't think a 3600 l/min would fit? Did I miss something?

24-07-2006, 01:11 AM
The vacuum pump only takes out the non-condensables. The water is condensed and pumped out of the condenser as a liquid.
The centrifugal compressor develops the pressure difference needed to make water boil on the low side and condense on the high side. A solid state inverter enables the compressor to spin very fast and therefore be small. (It might be necessary to use multiple stages, but I'm not sure.)

Erik Detroit
24-07-2006, 02:01 AM
I called it a vacuum pump instead of a compressor since you need to keep the boiling water at 0.008 bar. I get how it works, and it will work just fine, but my point was that you have to move an enormous amount of water vapor, and thus the machinery is going to be physically large. 3600 liters/min needs a large pipe!

26-07-2006, 04:54 AM
The compressor only has to work with a small pressure difference (as opposed to almost 1atm for a regular vacuum pump) and it would be more like a turbocharger than a normal compressor.
Google "R718" and you'll find designs that use a centrifugal compressor in a closed cycle.

26-07-2006, 11:26 AM
I think Eric is right. It's not such a small pressure difference. For water to boil at 7.2*C (a normal temperature for an AC heat exchanger) the pressure must be 1kPa (the atmospherical pressure is 100 times higher, 100kPa, so you can understand the pressure difference!) As water boils, the pressure increases, so you need to exhaust all the vapours. So, anywhay, you need a pump which can sustain a large flow and a pressure difference of almost 1bar.

Johnny Rod
26-07-2006, 11:56 AM
Don't forget that where there is water there are bacteria.

27-07-2006, 04:18 AM
Don't forget that where there is water there are bacteria.
Two words: UV lamps.

At 15C evaporator and 60C condenser temperatures, the pressure difference is about 0.18 atm. In the real world, the condensing temperature shouldn't be that high, so the pressure difference should be even lower.

I think my idea is certainly an interesting one, although it will very likely need lots of modifications to be practical.