4 ms·

I'm no refrigeration engineer, but I figure you might need some sort of pulsed cooling. Compress the solid so it gets hotter, run liquid coolant through it to e

I'm no refrigeration engineer, but I figure you might need some sort of pulsed cooling. Compress the solid so it gets hotter, run liquid coolant through it to extract the heat to a radiator. Then let the solid decompress, and switch to a coolant circuit that extracts heat from the interior of the fridge and dumps it in the solid. If you needed continuous operation of both coolant circuits, you could add a second block of solid material, and switch the cooling and heat-exhaust circuits between the two.

torpfactory 8y ago
Thanks! I think the idea you outlined would definitely work, but it is different than current refrigerants a few ways. For example: the secondary liquid loop you mention would probably vary in temperature with time. Right after the pulse starts the fluid would be very hot (or cold). As the solid material changes trends back toward the mean temperature, the secondary loop would follow. The secondary loop would need to then exchange heat with the system being heated or cooled. This would require the heat exchanger to work over a variety of temperatures for the fluid loop. This in itself is obviously not impossible but it does prohibit important optimizations to efficiency that you can get by assuming more steady operating conditions.
gruturo 8y ago
There is an interesting similar concept using magnetic fields instead of compression - some materials rise considerably in temperature when immersed in a strong magnetic field. Extract heat to a radiator, turn off the electromagnet, open the coolant circuit valve. It's called Magnetocaloric effect and they were considering it for in-car AC a few years ago, dunno what came of it. https://en.wikipedia.org/wiki/Magnetic_refrigeration https://en.wikipedia.org/wiki/Magnetic_refrigeration https://www.sciencedirect.com/topics/chemistry/magnetocaloric-effect https://www.sciencedirect.com/topics/chemistry/magnetocalori...

torpfactory 8y ago

Thanks! I think the idea you outlined would definitely work, but it is different than current refrigerants a few ways. For example: the secondary liquid loop you mention would probably vary in temperature with time. Right after the pulse starts the fluid would be very hot (or cold). As the solid material changes trends back toward the mean temperature, the secondary loop would follow. The secondary loop would need to then exchange heat with the system being heated or cooled. This would require the heat exchanger to work over a variety of temperatures for the fluid loop. This in itself is obviously not impossible but it does prohibit important optimizations to efficiency that you can get by assuming more steady operating conditions.

gruturo 8y ago

There is an interesting similar concept using magnetic fields instead of compression - some materials rise considerably in temperature when immersed in a strong magnetic field. Extract heat to a radiator, turn off the electromagnet, open the coolant circuit valve. It's called Magnetocaloric effect and they were considering it for in-car AC a few years ago, dunno what came of it. https://en.wikipedia.org/wiki/Magnetic_refrigeration https://en.wikipedia.org/wiki/Magnetic_refrigeration https://www.sciencedirect.com/topics/chemistry/magnetocaloric-effect https://www.sciencedirect.com/topics/chemistry/magnetocalori...