TEC chiller for learning/practice project.

[xoggyux]

Hi,
I want to make a TEC cooler for my Computer (eventually) but to get used to TECs I decided to do a "quick" project. A can chiller (actually I am hoping it will cool down from ambient to ~5C.)

My goal: I'd like to cool down a can ~400ml liquid, from ambient to ~3-5C in a reasonable amount of time (10m or lower would be ideal.)

I have some questions:
How much power do I need.
Is it better to get a single, high powered TEC or several "mid" power TEC(s) (I was actually thinking in the latter, to spread out the cooling better and have more area to install heatsinks.)
How much power (electricity) would I need for my goal.


I have these materials already available:
PVC tubings in which AA batteries fit nicely (I actually made super long batteries using this "technique" for a different project.) I have two long sticks of 5AA (~7.5V each stick, so about 15V total if connected in series.)
-This will be enough? (ignore the power needed to power the fans to dissipate heat, I will manage to squeeze another power source for them.)
I also have DC power adaptors (such as radios) and an old 300W PSU which I might use to make a wall adaptor to conserve batteries when needed.
I have a few stock heatsinks + lots of fans to keep the TEC's hot side cool (or close)
Thermometers.
A wider can (I got a "spray" can, which I cut in half and a soda/beer can fit perfectly inside, with just a tiny bit of air around.) Depending on how effective the device is, I could fill it up with water when being on use.


Thanks for input.

[cdolphin]

assuming water...instead of a different liquid:
1 j / c
= 400J per cc of energy that needs to be removed from the fuid...
Do you have a TEC on hand? I am not sure what you are asking exactly.
If you have a 50W 15V TEC that can transfer 35W of heat (i.e. it draws 50W of energy, produces 50W but transfers 35W) which is as I believe likely to find, it should theoreticaly take you about ten seconds per degree C (assuming the TEC is at its optimal temperature).
I am not experienced with TECS per se but if you have one, just try it out. The best way to find out how they work is to test them...You will want to pull a voltage across the tJunction and watch it whirl.

[untrueparado]

a can would be difficult to cool. i dont know how you would be able to attack the surface. the bottom is concave, and the sides are obviously round, and the top is for the mouth so gotta leave that clean. youll need some sort of special sink to attach to it and draw heat out of the can. most tec chillers are obviously flat for cpus so i dunno what youre gonna do. an easy way would be to buy a piece of copper and wrap it around the can and weld it to a flat copper base and then use as5 to attach to the tec.

i dunno what you mean by "fit perfectly" but if the sides arent touching, it does no good. and tin doesnt transfer heat nearly as good as aluminum or copper. you can always fill it up with water but thats even more stuff to cool.