Can a TEC freeze a 26" long piece of aluminum?

[typhoon43]

Strange question, but Google has left me still searching. I need to cool a 1/16" concave piece of aluminum that's 24" long by 1.5" across. I was wondering if I would be better off with multiple TEC's placed every 6 inches, or if one giant unit in the middlw would do it. There would be no heat load. I just need to cool the metal. I can't find any data on heat dissapation when the TEC doesn't have to fight a 100 watt CPU. I just need to be able to cool this metal to the point that ice forms. Would a 300W unit, or 4 75W units pull this off? Each TEC would have a 120mm fan/sink combo.

Thanks..

 

[Ice Czar]

http://www.pcper.com/article.php?aid=-105&type=expert&pid=3

The heat transfer through the wall follows a simple equation:

We can draw some interesting conclusions from this equation. First, heat transfer is proportional to the temperature difference on the object. If the temperature differential doubles, the heat transferred doubles. Second, the conduction coefficient "k" is proportional to heat transfer. If the conduction coefficient doubles, the heat transfer doubles. Alternatively, for the same differential temperature, twice as much heat may be transferred. The final observation is "L". As thickness increases, heat transfer decreases. Alternatively, to maintain the same heat transfer through a material twice as thick requires twice the temperature differential.

the last part being likely the pertinent part
its still dealing with a load, that of the ambient environment, the colder it gets the quicker it will draw heat into it from the environment

a single TEC has less chance of cooling the whole peice to the same temperature
whereas multiple tec would have a far better chance of a homogeneous temperature across its length

(think of why heatsinks look the way they do, the more metal the heat\energy must traverse the harder it is for conduction to make it worth while, which is why there is a practical limit as to how big a fin array can be without resorting to heatpipes to add to the transfer of a larger fin array, the very same reason watercooling is employed, to transfer from a point source to a far larger array by actually moving the matter that is heated)

so some of the questions youd need to answer are what is the ambient, and is the same temperature across the whole peice necessary. As to actually determining if a single TEC is up to cooling the total volume to 0C at a given ambient, yes at some power wattage it would be able to. But I doubt its the most effcient solution.

you do need to consider what its doing with the heat its removing from the peice\environment and the power its additionally converting into added heat to do that, if its dumping it back into the immediate environment its raising the ambient and will eventually fail dependent of that environments own fluid flow
(closed closet vs outside antarctica) regardless of how powerful it is (how great a temperature differential it can manage)

 

[typhoon43]

Thanks Czar. Great info as always. I'd be exhausting the heat outside the enclosure. I want to make a TEC powered ice-shooter slide

 

[DFI Daishi]

multiple small pelts is definately the way to go.

with an single 80 watt TEC, i can get frost to form within about 2 minutes. there's a good bit of heat do deal with, but it chills down a chunk of metal very quickly.

i think that you could get by with a bunch of undervolted 80 watt pelts, each with an a64 stock heatsink or equivalent.

why undervolted larger tecs? run them at full tilt for a while to quickly get a uniform layer of ice to bind on to the surface, then back them down and still get some ice forming, without having to deal with tonnes of heat.

things that i'm wondering about: are you planning on using rough surfaced metal to get a nice thick layer of ice so that you get a bit of dillution of the liquor as your pour, like with a normal ice block? since i'm guessing that you have access to machining equipment if you're thinking of making this, have you though about using a water cooled solution? how do you plan on powering this monster?

edit: on second thought, after a quick bit of rummaging and playing around, heatsinks suited for a 120mm fan would be a good idea. running this smaller 'sink for an extended period, the hot side id getting so hot that the cold side isn't getting down to sub zero.

 

[typhoon43]

Thanks Daishi,
I can get access to a machine shop if needed. Watercooling is out of the question for now, as I intend this to be portable, and run off 110V hoousehold. You think 4 x 80 watters with 4 120mm coolers could hold the heat off for an hour or two if ducted to the outside atmosphere?

 

[DFI Daishi]

ehm........i'm pretty sure pelts need DC.......i've use 12VDC pelts and i've seen 24VDC pelts. i've never seen a 110VAC pelt.

you need a switching PSU to run the pelts. maybe you already realize this, but the way you worded your comment.....i just want to make sure that you understand.

as for the 4x80watt pelts being enough.......i guess that depends on how much liquor you're serving up and how long you let the slide chill down before using it.