A "cool" idea

[ThatOnePerson]

I got to thinking today about ways I could cool down the water in my liquid cooling system, effectively, something that isn't the size of a refrigerator and most of all, cheap.  So I started looking around and stumbled across aquarium chillers, namely the ones made for say a 10 gallon tank. Now the water in the reserator is only about 1.25 liters so I'm worried about possible freezing since it says the chiller can get things to 20 degrees. But, the anti-corrosion stuff mixed with the water says it's freezing point is -9C. Will this lower the freezing point of the water as well? I could really see this working seeing as it requires a 1.25 inch wide hole. And it just so happens the opening on the reserator is that size..exactly. Any thoughts/ flaws in the way I'm looking at this?

http://www.coolworksinc.com/iceprobe_aq ... hiller.htm

Edit: The physics side of me activated and remembered heat always transfers to cold, so I dont think it would freeze regardless. Assuming it fits any other complications you guys see?

[Wii]

Looks like it will work. I don't think it would freeze at all either.

[NickN]

The only issue with this is condensation. it does not need to freeze in order to create condensation so I would be very careful in monitoring a project like this. You need the temp of the water to only drop a small amount and the heat of the CPU to compensate enough to maintain the least condensation buildup possible. Anything that invokes temperature inversion may invoke condensation if the ambient air and humidity meet defined criteria.

[ThatOnePerson]

Is there any sort of equation or a "baseline" at which the temperature would create condensation? I mean, obviously a cold drink that isn't frozen will get condensation on it but I think it should have something to do with the amount of temp difference rather than the actual temps themselves correct? Like the dew point I suppose.

[NickN]

Condensation occurs where moist air comes into contact with air, or a surface, which is at a lower temperature. Air contains water vapor in varying quantities, its capacity to do so is related to its temperature - warm air holds more moisture than cold air. When moist air comes into contact with either colder air or a colder surface, the air is unable to retain the same amount of moisture and the water is released to form condensation in the air or on the surface.

You would need to know the relative humidity and the dew point of the room on a daily basis, and, the temp of the line with the water moving through it to calculate the amount of condensation

You can easily insulate the lines however what is difficult to insulate without covering it in silicone and letting it kick solid, is the CPU block.

The amount of condensation you may be dealing with will vary from day to day based on the ambient conditions of the room or tower. that is why I said a certain amount can be controlled by not allowing the temp to drop too far.. the heat from the CPU alone will maintain a balance and not allow condensation to form. Also, the flow rate will also play into that. The faster the flow, the higher possibility of a good amount of condensation forming.

It would be an experiment to find out how it works and even if you find it works great one day, three weeks later the ambient room environment may make for a major shift, allowing a large amount of condensation to form.