DIY Cascade Refrigeration System (Part III)

To try everything Brilliant has to offer for free for a full 30 days, visit https://brilliant.org/HyperspacePirate/ . You'll also get 20% off an annual premium subscription.

This is part III of my video series on building a DIY vapor-compression casacade (i.e. multi-stage) refrigeration system. In the last two videos, i demonstrated the concepts needed to make a cascade system work and reached temperatures below -110C, but because i used low-powered reciprocating compressors meant for R600A, I wasn't able to build a system that ran continuously and only managed to evaporate low-temperature refrigerants in short bursts.

This time, I'm using a pair of rotary compressors extracted from 5,000 BTU window air conditioner units. The first stage will use propane (R290) as a refrigerant, and the second stage will use Ethylene (R1150). The compressors were made for R410A, so they're able to run as high as ~30 bar continuously without a problem. This will allow me to condense Ethylene at the first stage evaporator temperature of around -20C to -25C.

To extract more efficiency (and a lower temperature) out of the system, I use subcoolers on both stages; the return lines to both compressors are wrapped in capillary tubing coming from the condensers.

The lowest temperature I managed to achieve with this system was only -88C, but that's probably because I ran the second stage suction pressure relatively high (around 4-6 bar) in order to keep the mass flow rate high. Down to about -63C, the efficiency or "coefficient of performance" (COP) is about 10%, so there's definitely some room for improvement.

The best way to improve the ultimate temperature and/or thermal efficiency of this system would be to use a sub-cooler with a thermal gradient across it that looks basically identical to the regenerative heat exchanger in a joule-thomson system. When I experimented with this using my precooled joule-thomson cryocooler with 100% ethylene (effectively turning it into a cascade vapor compression system), I achieved nearly twice the thermal efficiency of the device I built for this video.

I originally planned to use this system to liquefy Methane. This is possible at -88C, but would require a very high pressure, and with the marginal cooling power at that temperature, it's probably not practical, so I'll be doing that in my next video using my joule-thomson cryocooler.

I'd also recommend using a higher-boiling refrigerant for the second stage - like R508 (if you can get it), Ethane (R170), Nitrous Oxide (R744A), or even Carbon Dioxide (R744) (although CO2 is only feasible if you keep the second stage evaporator ABOVE the -56C). The boiling points of Ethylene and Propane are far enough apart that the system is reaching pretty far.

Music Used:
Kevin MacLeod - George Street Shuffle
Kevin MacLeod - Bossa Antigua
Kevin MacLeod - Lobby Time
Kevin MacLeod - Elevator