TRUE TEC Module (with pics)

[scifikg]

This air cooled TEC module is designed to directly cool a wolfdale E8400 on a DFI UT P35 T2R board.



The parts. The TECs are of the 62mm variety available on ebay



assembled profile view



Power Supply Voltage= 13.73V
Temperature hot plate = 49.4C
Temperature cold plate = -15.7C


Power Supply Voltage= 9.2V
Temperature hot plate = 33.8C
Temperature cold plate = -18.0C


more pics to come....

[twin_savage]

Nice work!
did you fabricate that yourself?

[scifikg]

Nice work!
did you fabricate that yourself?

nope, local machinist made it for me.

[STEvil]

What wattage are the TECs.. 320's?

[CL3P20]

What sort of delta do you expect to have using the dual-core? That is definitely a cool idea. Can't wait to see some results.

[ZOMGVTEK]

Awesome... I was thinking of doing the same. Im not a big fan of needing to liquid cool the hot side of tecs, regardless of its advantages.

[scifikg]

What wattage are the TECs.. 320's?

I think the TECs are TEC1-12730

What sort of delta do you expect to have using the dual-core? That is definitely a cool idea. Can't wait to see some results.

@9V and a 150W load from the CPU, I'd like to see -10C idle and 10C load on the cold plate.

[Duh]

anxious to see results with cpu

looks very very nice ..

[CL3P20]

Are you able to get enough clamp pressure on those TEC's with the hardware you have? I do not know the torque specs..but have heard that crushing the TEC is very important for developing proper heat transfer.

*10c will be a very nice load temp indeed for any aircooled TEC setup. Is this going to be for benching use or for 24/7?

[scifikg]

Are you able to get enough clamp pressure on those TEC's with the hardware you have? I do not know the torque specs..but have heard that crushing the TEC is very important for developing proper heat transfer.

*10c will be a very nice load temp indeed for any aircooled TEC setup. Is this going to be for benching use or for 24/7?

I don't have a torque wrench for this. I just cranked as well as I could with a regular old wrench.

This comp is for gaming. I figure that I 'll be using it 20 or so hours a week.

[Sam__]

wow... that looks like a beast. cant wait to see some results.

[hollo]

beautiful

but if you don't mind a little nit-picking the cold plate is too thin :/
(well, the parts of it around the edge mostly)
figure 3 on this page http://electronics-cooling.com/artic...v_article2.php shows the result of some experiments done with heat-plates of varying materials and thicknesses
you can see that with an 8x8cm square heat plate (the bottom two lines) the temperature continues to decrease even as the plate thickness increases over 16mm with a 1x2cm heatsource (because the extra thickness is required to allow the heat to transfer 'horizontally' through the plate), and your 62x62mm x2 cold-plate has 20% more surface area and about a 50% greater distance between the centre and the furthest edge from the centre than a 8x8cm plate
(copper has a "k" (thermal conductivity) of 400 W/mK, so the 2nd to bottom line would be a good comparison for the copper cold-plate)

[scifikg]

beautiful

but if you don't mind a little nit-picking the cold plate is too thin :/
(well, the parts of it around the edge mostly)
figure 3 on this page http://electronics-cooling.com/artic...v_article2.php shows the result of some experiments done with heat-plates of varying materials and thicknesses
you can see that with an 8x8cm square heat plate (the bottom two lines) the temperature continues to decrease even as the plate thickness increases over 16mm with a 1x2cm heatsource (because the extra thickness is required to allow the heat to transfer 'horizontally' through the plate), and your 62x62mm x2 cold-plate has 20% more surface area and about a 50% greater distance between the centre and the furthest edge from the centre than a 8x8cm plate
(copper has a "k" (thermal conductivity) of 400 W/mK, so the 2nd to bottom line would be a good comparison for the copper cold-plate)

Thanks, but lets be reasonable. Is it really fair to compare my cold plate to data from that review article? I can see some really important differences between them

1. That data was made with a single heat source but my cold plate has two.
2. The heat transfer coefficients are at least an order of magnitude to high.
3. 300W is too much for an e8400
4. The 8x8cm shape and the cold plate shape are very different

I think heat spreading is just way too complex to to declare that my cold plate is too thin based on that data.

[n00b 0f l337]

So you saw lower temps with less voltage? Basically showing you can't handle the heat of the TEC's at the higher voltage, your going to end up dropping the voltage even more when a heat load is applied.

[CL3P20]

So you saw lower temps with less voltage? Basically showing you can't handle the heat of the TEC's at the higher voltage, your going to end up dropping the voltage even more when a heat load is applied.

Not really. Lower temps at a lower input voltage, are related to the TEC efficiency. TEC's by nature are very un-efficient and when run at nominal voltage require massive amounts of cooling to run, even at spec.

Running below the TEC's nominal voltage, allows you a greater efficiency ratio..thus improving temps, both on the cold and hot side of the TEC, as well as using less current to run.

[n00b 0f l337]

Yes you just said a basic truth of TEC's, however that does not apply in this situation as your not moving any heat load Just the heat from the TEC's themselves. Lower voltage with no load giving you better temps means your hot sides too hot. And 50C is too hot.

[CL3P20]

Lower voltage with no load giving you better temps means your hot sides too hot. And 50C is too hot.

Mmm. I see your point...does make sense. Sorry for the misunderstanding.

*As for improving temps with that setup.. How about enclosing the sides with something to channel air through the coolers? I will ask around to see what sort of crush needs to be applied to those TEC's too. If my memory serves me correctly its in the neighborhood of ~60lbs..not sure though.

[scifikg]

So you saw lower temps with less voltage? Basically showing you can't handle the heat of the TEC's at the higher voltage, your going to end up dropping the voltage even more when a heat load is applied.

I don't see any problem running the TECs at 9V. I obviously don't expect to hit -18C on the cold plate but consider this. At 13V, I figure the module is outputing ~150W more heat than at 9V. That is about what the CPU will be putting out so I think the hot plate temperature will increase similarly under a heat load. I figure the cold and hot plates will increase 20C at worst. That corresponds to ~2C on the cold plate and ~54C on the hot plate with the TECs at 9V.

NOL, Why is 50C too hot? That is well within the operating temperature of the TEC and the specs are even givin with the hot side at 50C.

[Sam__]

Im intrested to see the clamp device... u just gunna use some nuts n bolts n do em really tight?

[n00b 0f l337]

It is within the operating perameters but its just a relatively hot temperature, ultimately that heat is passing back through. It means mainly that you are not dissipating all the heat you could be. Id honestly just suggest heatsinks WITHOUT heatpipes.

[scifikg]

Im intrested to see the clamp device... u just gunna use some nuts n bolts n do em really tight?

Thats right. I'm just gonna crank on them till it stops bleeding TIM

It is within the operating perameters but its just a relatively hot temperature, ultimately that heat is passing back through. It means mainly that you are not dissipating all the heat you could be. Id honestly just suggest heatsinks WITHOUT heatpipes.

N0L, Are you saying a heatsink WITHOUT heatpipes can move more heat than the TRUE? It is hard for me to believe this, but if you know a company that makes a heatsink like that I'd like to check them out.

[Vapor]

Heatsinks with heatpipes have a ceiling....'old fashioned' heatsinks don't.

[[XC] NetburstXE]

This looks really cool. What kind of temps do you think would be possible with a quad? I can't wait to see how this is going to turn out!

[Sam__]

yeah im intrested as to how it handles high heat loads. from what iv heard they can get quite bad and in some cases be worse that not having a TEC in the first place... hope this doesnt happen to you. gl

[n00b 0f l337]

As Vapor said, each heatpipe has a load for it, most around here are 35W or 50W. Past that they become quite useless.