• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Cooling a e6600 Conroe with a TEC

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

datura3

Member
Joined
Apr 7, 2005
Well I have my new system set up with an e6600, p5B-dlx, and 2gigs of mushkin 8500 ram. It's time for some peltier cooling! I plan on testing a couple of set ups. I have a 172w peltier that I will run on 12v to provide 86w of cooling power. I also will test a 226w peltier and show the difference between the two. I hooked up the 86w last night and should have some numbers soon. I also plan on testing the peltier with a standard psu and a Meanwell 320. I think the new PSUs with 620w should be able to handle a small peltier.
 
Last edited:
yeah lets see some load temps...and maybe a max OC vs air cooling too, if possible.
 
Gautam said:
Whoa, that's way better than what I would've expected. But how does it load? :p

Yeah, that's what I thought too :) 20C is awesome for such a small pelt. Especially since it only pulls about 4A and I would be fine running it on the PSU. But, unfortunately it loads like crap. 65C! A nice deltaT of 45C lol. I think it is the waterblock setup I'm using on it. It just can't pull the heat away. Right now I have a NexXxos XP on it which is a really bad block for a peltier. Plus, I really need about hundred psi clamping pressure to make it perform.

I have a mcw6000 that I'm going to build a peltier retention mechanism for. I think the 60xx swiftech blocks are probably some of the best for a TEC.... actually the longer fused pins may be better than swiftech's new block that uses a 226w TEC (MCW6500-T). What's great about the mcw6000, is it is large enough to fit a 437w on there.... so I plan on building something that I can test out a couple of different pelt setups. A 86w, 172w, 226w, 437w, and 226 with a 437w and compare them accurately. I probably won't run any of them 24/7. Ha. I just like playing around with these things. I'll head to the machine shop in the next few days and build something. Right now I'm pretty busy, so there may be a delay, especially with the 437w since I need to get the cash to buy another meanwell.
 
Last edited:
Ok, so I had a spare 1/2 hour today during work to machine a cold plate. I decided to make a single cold plate that will fit all three sizes. It needed to fit all of these sizes:

40x40mm (86w, 172w)
50x50mm (226w)
60x60mm (437w)

The cold plate I designed is 62x62mm and .375 inches thick. This was about optimal for the 50x50mm using the formula in the sticky here. The other two sizes won't be optimal, but I think they will be ok. I designed it kinda after the swiftech MCW6500-775T cold plate. Basically, I needed that bezel because the coldplate would hit the caps surrounding the socket. This has only been lapped down to 600 grit sand paper for now. I may need to make some more modifications so I'll hold off on the serious lapping.

Here is the MCW6500-775T. Notice the one I made below is similar :)
MCW6500-775T.jpg

Here is my old MCW6000 block with cold plate I just made.
blockandcoldplate.jpg
Here are a couple of pics with a 40x40mm peltier in the pic
blockandcoldplate3.jpg
blockandcoldplate2.jpg
 
Last edited:
Ok, I just machined the top plate out of aluminum and acrylic. Now I just need to build the bottom plate and screw it all together :)
retentionmechanism.jpg
 
Thats pretty nice although you should know that halving the voltage also reduces the amperage(i'll find the exact amount in the morning since its bed time but I think its about 42w based upon the formula V=IR)

looks like a pretty good project so far though.
 
looks schnazzy. must be nice having a mill at your disposal. can't wait to see that piece in action
 
speed bump said:
Thats pretty nice although you should know that halving the voltage also reduces the amperage(i'll find the exact amount in the morning since its bed time but I think its about 42w based upon the formula V=IR)

looks like a pretty good project so far though.

So halving the voltage doesn't halve the watts.... ha, it's been a long time since I took physics. Anyway, if you could give me a hand what would happen if I halved a 172w and the 437w that would be very helpful.
 
iirc, power = I^2 * R, I = V / R, therefore power = V ^ 2 / R, and R is constant. so, P1 / (V1 ^ 2) = P2 / (V2 ^ 2) or V2 = sqrt(P2 * (V1 ^ 2) / P1))

i haven't double checked with google but I think thats right

*edited to make my sentence closer to english.
 
Last edited:
splat said:
iirc, power = I^2 * R, I = V / R, therefore power = V ^ 2 / R, and R is constant. so, P1 / (V1 ^ 2) = P2 / (V2 ^ 2) or V2 = sqrt(P2 * (V1 ^ 2) / P1))

i haven't double checked google with but I think thats right

Thanks, that will be helpful. Now I just need to do the math :)
 
speed bump said:
I'm lazy here have a calulator that will do it for you.

http://www.sengpielaudio.com/calculator-ohm.htm

Awesome. That's pretty handy, thanks. One last question. If I use half of the voltage. i.e 12v for a 24v TEC, do I halve the Imax or Vmax? For example, would a 172w TEC (24v, 7.9 Imax) at 12v be:

41w TEC using half of the Amps

or

86w TEC, which is half the wattage, but still uses 7.1 amps.

Not sure which way is correct.
 
Gautam said:
Halve vmax. The current more or less remains constant.

Ok, that makes sense. The 172w peltier was running at 95w when using 12v. So when I say 86 above, it was actually running at 95w- oops :) That would explain how it was not overloaded and how it got the Dothan down to -5C.
 
Back