Just thought I'd share this random project

Hey guys, haven't posted on OCN in forever, but this random refrigeration mock up made me think of putting it here just for shts and giggles.

So I scrapped this heatpipe cpu cooler out of an old Dell and didn't know what to do with it. I saw someone on the web had cut the pipe tips off and bridged them to make a small air to water heat exchanger for an induction heater project. So that got me thinking...maybe you can guess.



And, of course it just so happened that someone threw out yet another mini fridge.


I'm getting good at this.


The compressor is the smaller of the typical sizes found in these things. It's a 2.5cc displacement with 1.3 FLA. Not too impressive, but I've got a holiday to kill.

I formed the existing condensor tubing into a staggered array to work with a 120mm fan. I have a strong 25W delta made fan scrapped out of a Dell to cool it with.

A spark plug works perfectly as a former.


All assembled and ready to be vac'd down. I put it together with regular silver solder, partly because I want to see if it works for refrigeration and partly so I could build it on my desk with my butane micro torch. I have read some techs on the refrigeration forums have used it with success, so WHT. One plus is nitrogen is not as important, since the joints can be made at lower temperatures where the copper will not oxidize. The downside is risk of contamination with flux. I used flux very sparingly and quite frankly I don't care...it's just a quick mock up with a free no name taiwan compressor. No new liquid line drier, but I kept the lines pinched off, made the joints quickly, and vac'd down asap...again I don't care, as it's just a fun project. It did hold vacuum, so it's at least sufficient for testing.



So the final result after learning about just how to charge a cap tube system here (good read for beginners)
http://www.supco.com/images/pdfs/Manuals-Instructions/Cap%20Tubing%20Manual.pdf

The other part of the experiment is more testing with R-152a refrigerant which is almost identical to R-134a, but just a bit more efficient and at lower temps. The cap tube is the same as what came from the fridge minus about a foot or so. I have no idea what size it is, but I figured a foot less would raise the evap temp and capacity slightly higher. I used a 50W TEC module frozen to the bottom of the evaporator for a load. At 12V 4.1A, that's 50W right there, plus a 20 deg C temp difference across the module, indicating an additional ~30W being absorbed and transferred by the module into the evap according to the datasheet. Not considering the heat gained into the block of ice that is the rest of the evaporator, as well as not considering the thermal resistance between the module and the copper base that would mean a lower delta T across the TEC, it's probably right around 80W of cooling out of this wimpy little thing! The copper base was running at -9 deg C and the evaporating temp was -35 deg C at 10 psi. The cold side of the TEC was -30 deg C. Motor current held at 1.3A.



So efficiency increased by ~18% at the same time as a roughly 15 deg drop in evaporating temps (most fridge/freezer compressors are rated at -20 deg C evap). With a bit of cap tube experimenting and sacrificing a few degrees, it could probably get up to 90-100W (actually could probably measure at least that much at the base if I actually insulated the evap). Not bad for a piddly mini fridge compressor and a repurposed Dell heatpipe cooler.

The top bridging loops are that size due to my tube bending limitations.

Interesting you mentioned raising the temps. Most likely due to the quick and dirty build, there must have been either some small debris, or probably a bit of moisture that froze and partially blocked the cap tube during initial testing. The unit had not been shut off since the first start with just 20 psi of gas, so the initial low pressure must have frozen some moisture in the cap tube. I was quick in assembling it and vac'd it with my 2 stage robinair pump for an hour, but maybe it doesn't take much. Anyway, when I shut it off and let it defrost, it started up again a totally different animal with 32 psi low side and ~ 40 deg F on the coils. Now it's pretty much a kick butt micro air conditioner. That's kind of the direction I was expecting since the cap tube got about 30% shorter than the original fridge design. I had to dig out a 12V 0.85A fan for the evaporator to start getting some appreciable superheat, so given how easily the 12V 0.17A fan would max out the evaporator before the transformation, I would say it's cooling at least 200W or so. It would make a great internal computer air conditioner actually, but it would probably solidly outperform or at least equal a decent air cooled water loop as a CPU cooler with maybe a little extra for the ambient case space.. Too bad it's not a LGA775 mount or I would give it a try. I could probably cobble something together if I really wanted to. The only problem is condensation dripping. It sweats like crazy. Now it would really do better with a water cooled condensor, or at least an air cooled one with fins. Anything less than my annoyingly loud 25W 120mm fan hinders performance somewhat with the bare tubes due to the increased capacity.