So I’m a refrigeration tech with some electronics manufacturing experience. But I’ve never combined the 2 skillsets so I’ve been toying with the idea of building a large vapor chamber to cool a computer via direct immersion in a refrigerant. I know its about as far from practical as you can get but it sounds like fun.
Ignoring all of the many many other problems with doing this for now the one thing I’m not sure about is how well the electrolytic caps on the various components would survive. I would need to pull a fairly hard (500 micron) vacuum on everything before I charge it with refrigerant. I know most electolytic caps aren’t vacuum rated but I’m not sure if that just means you can’t have them operating in a vacuum or if they will immediately pop if you just subject them to hard vaccum period. Additionally while I am planning on using a low pressure refrigerant (probably some R-123 substitute but I’m definitely still working on that part) the components would all still be subject to pressures of up to about 20 PSIG at the high end. Beyond that point I would probably have an active cooling system kick in just for safety sake. I’m not sure how well the caps in particular would survive being immersed in a liquid under 20 PSIG pressure.
Does anyone here have any experience subjecting electrolytic capacitors to hard vacuum or elevated pressure? At what point do they just pop?
Not too sure about the vacuum effects, I suspect the electrolytics wouldn’t last long as they are built to handle a certain pressure then pop to vent in a controlled manner in the event of failure. The positive pressure under operation is also likely to inject liquid refrigerant into the components and into layers of the PCB and such, that cant be good for any of it, that would definitely kill capacitors by displacing and or dissolving the electrolyte fluid.
As for the longer term, I know that pretty much all of the phase change fluids you would likely use act as pretty strong solvents in their liquid states, so I doubt the hardware would survive terribly long.
There are immersion cooled computer systems using an inert liquid like Perfluoro(2-methyl-3-pentanone) but that is a different process to phase change refrigeration.