MrJoeyJiffy - 22-7-2011 at 18:15
Hello all! Long time lurker, first time poster!
I searched and didn't see anything involving this so I posted, if this is a repeat please forgive me. The title basically explains it, has anyone ever
used a computer water cooling kit( ex. http://www.xoxide.com/swiftech-apex-ultima-kit.html) in place of the pump/bucket/ice water technique commonly used for simple distillations?
If you're not familiar with PC water cooling its basically just a radiator, pump, some coolant, tubing, and some water blocks that take the place of a
traditional heatsink/fan combo usually used for cooling chips.
I'd like to try to use one of these setups(minus the water blocks of course) for my condensing. Has anyone ever done this before? In theory it seems
like it would work fine the only thing I could think that may be bad is if the liquid is too cool could that cause the condenser tube to become overly
brittle? I wouldn't think that would be an issue but I was hoping someone with some more experience might be able to give me some insight before i
drop a few bills on a setup.
Paddywhacker - 24-7-2011 at 00:07
Yes, that was my first setup when I was distilling my own spirits, but the radiator, even with fan, was not up to keeping the liquid cool. Maybe it
would work for something with a much higher boiling point. Including the CPU block and clamping a peltier device on to it did not make it much
better. I ended up using the pump and tubing with a 4 gallon water reservoir. The large reservoir took all day to warm up, and then it was left to
cool down.
The ideal would be to put the reservoir in a freezer, but I haven't organized that yet.
The pump broke down eventually and was replaced by one from an aquarium shop.
hkparker - 24-7-2011 at 00:58
Yea, distillation generally needs to be kept way cooler then a water cooling kit can provide.
MrJoeyJiffy - 25-7-2011 at 20:34
After finding out finally that the liquid in a water cooling system is usually only around 30C I agree it definitely wouldn't work, are there any
other alternatives to the ice bucket/pump to your knowledge?
smuv - 25-7-2011 at 21:19
A computer processor puts out what 75w max. A hot plate is 1kw, even at 10% duty cycle, the energy transferred from the hot plate --> liquid
--> gas --> cooling system will probably overload it.
hkparker - 25-7-2011 at 22:33
mines rated at 95W 
I don't see why an ice bucket pump isn't acceptable. I guess its not the most convenient thing every but it gets the job done well.
not_important - 26-7-2011 at 08:32
smuv points out the important factor, the input heat flux vs the capacity of the cooler/radiator. The second important factor is that with only
passive heat transfer you can't get any cooler than the surrounding air. Evaporative cooling of the radiator can do better, with increased
complexity. TEC units can do even better, but their relatively low capacity and low efficiency make them impractical for any size of setup. Ice is
simple and cheap most places.
smuv - 26-7-2011 at 20:35
Use a garden hose and reducers to get it to tubing. OR...buy a slip-n-slide and use the cheap plastic fitting it comes with (and have a party... 
).
watson.fawkes - 27-7-2011 at 03:38
The heat exchanger in a typical CPU cooler is designed for using air as the fluid in the cold sink, but there's no particular reason that you couldn't
use ice water. What you'd be doing is converting a liquid-gas heat exchanger to a liquid-liquid heat exchanger. Admittedly you can't use the fan that
comes with the kit to move ice water. You'd need to acquire a pump to move the fluid in the cold sink. You'd also need a manifold to hook up the pump
to the heat exchanger.
For the highly motivated, this could be used to get sub-0 °C working fluids (e.g. glycols) into a distillation path.
peach - 27-7-2011 at 09:06
30C is not really good enough for some solvents even at atmospheric pressure. DCM, for instance, will try getting round a condenser at that
temperature.
I'd be tempted to just remove the pump and use it with the cool box method. I use that all the time. It's usually below 20C in the UK outside and is
only 10 or so for long parts of the year. The cooler fits tens of litres of water in it which means that, even without a state change, it can absorb a
lot of heat without changing it's temperature much. I rarely ever need to bother with ice in there. If you live somewhere hot, it's ice and a lid for
you.
Here's an option I've wanted to try for a while, but I think the guys at the local tip might be getting curious as to why I keep messing around down
there.
One chest freezer, drop your pump in, drill two holes in the lid, one for the cold out and the other for the warm back. Seal the inside up with
sealant and fill it with anti-freeze, or use a bin full of anti-freeze. Engage the super freeze. A day or two later, you'll have a huge thermal mass
at -30C, and a pump to help pump heat out as it comes in. Such a cooler from a laboratory supplier, whilst looking much more high tech and performing
in a different manner, would set me back a thousand or three.
If you were getting fancy, you could also drop plastic bottles in there loaded with something to freeze solid at specific temperatures. There's
basically no distillation you'll be doing at home that will unfreeze and heat that amount of mass up.
NHZ - 9-8-2011 at 19:54
I have been reading this forum and must say, very informative 
Although I cannot
contribute
much by way of chemistry, I still feel compelled to contribute something.
After viewing this thread, the idea of thermal electric cooling came to mind. A copper rod sealed
inside a piece of glassware like a test tube would mimic what is being discussed here with the ability
to provide sub zero temps if so desired.
If a 100W TEC would not be enough, they can be 'stacked' with the cold side of the plate connected
to the copper..
Just an idea, and perhaps there is something im overlooking as im new to anything related to
chemistry... conflicts etc. Just thought id throw in my 2 pennies 