TRUE TEC Module (with pics)

[Sam__]

For heat conduction, aluminum has a k value (heat conduction) of 250 W/m*K whereas copper has a k value of 400W/m*K so copper is almost twice as good as aluminum.

Did you contact a local machine shop? Tell them your interested in copper 110. They might not have it in stock but I bet they can get it pretty easy. Most shops always have some aluminum 6061 in stock if you go in that direction.

thanks a lot for the advice .. ill be sure to ask around a bit more now.

[STEvil]

Yeah, however it is slowing down heat transfer as well. You really want to have as close to ambient hot side as you can, and I think removing that alu plate, or making it much much thinner might give better results.

Have you noticed what is holding the aluminum to the copper? Steel.

Only way to do this better is using very tough plastic really. 1 degree difference if that probably.

Also as has been known and shown time and time again thicker cold plates only help to stabilize temperatures when used in this type of application. Search around for examples back in the P3/P4/K7 days.. and earlier.

[Bloody_Sorcerer]

http://www.overclockers.com/articles305/
still totally legit for cold plates.

[ninjaskija]

Well some results could be nice..

[Jorlen]

Short answer: 2.

I am not an expert. But I do recognize two limiting factor of the heatpipes. Hotter pipes means increased pressure which means more of the substrate in the pipe is in gas form all the time. Which lowers the amount of working fluid. I dont know what temperature you will end up with, but its not likely this will have any impact.
The other limiting factor is the rate at which fluid returns to the evaporation end of the pipe. Think about it like this. The more energy that goes into the pipe the faster the evaporation occurs. Thats a linear relationship. Twice the energy, twice the flow of evaporated liquid (gas). The speed at which liquid returns to the hot end of the pipe does not increase in this manner at all. Gravity will help increase flow somewhat, providing the pipes are mounted vertically.
So at a certain point the capillary liquid transport in the wick-structure to the hot end will be less then the gas-transport to the cold end. At this point heat transfer via evaporation stagnates. At this point increasing the load further will raise pipe temperature and if run out of control the problem mentioned above might become noticable.



(edit spelling)

Jorlen, are you saying that if you heat a heat pipe above its maximum wattage it will

1. completely stop transfering heat except for it's outer shell

or

2. keeps transfering its maximum wattage and the excess heat must travel through the outer shell

[scifikg]

Gravity will help increase flow somewhat, providing the pipes are mounted vertically.

I've read that the affect of gravity is highly dependent on the type of heatpipe used. I'm not sure what type is in the TRUE, but it will definitely be mounted horizontally.

[ArtosDracon]

The TRUE uses chemical filled pure copper 8mm OD DOM tubing w/o internal channeling. The only better heatpipe is a powerdered metal tubing instead of DOM but, the powdered metal heatpipes can make a pretty notably difference, especially in longer heatpipes. I don't know what they use inside the pipes though. TRUE's heapipes are rated at 37W each at their production length, initial numbers had them rated at 35W but, they were shortened slightly for production, I happen to think thats why they're having isses with their bases not being flat, the tubing bender was designed around the pre-production pipes that were slightly longer, thats also why there is a gap at the edges of the base on the heatpipes with a tighter curve to them. It's unfortunate they didn't tool a mandrel bender instead. Also, the heatpipes will become less efficient when they're overloaded. Up to the 37W heatload they'll be fine but, when you cross the 37W barrier they slowly start dropping efficiency because the chemical doesn't re-condense as quickly or completely at the top of the pipe so when overloaded to say 40W they will drop back and only transfer about 34.5-35W instead of the rated 37W. Much beyond 40 and it's going to drop off more quickly. I would move your fans up about 1.5cm so you get some flow across the tips of the heatpipes, that'll be where they're going to do most of their condensing if you're overloading them.

The bolts holding the aluminum retention bracket are standard steel not stainless right?

[ArtosDracon]

I've read that the affect of gravity is highly dependent on the type of heatpipe used. I'm not sure what type is in the TRUE, but it will definitely be mounted horizontally.

you mean the motherboard will be horizontal and the heatpipes themselves vertical right?

[scifikg]

I would move your fans up about 1.5cm so you get some flow across the tips of the heatpipes, that'll be where they're going to do most of their condensing if you're overloading them.

The bolts holding the aluminum retention bracket are standard steel not stainless right?

I'll move the fans too see if that helps. The bolts are stainless steel.


The TRUEs are mounted like this with the motherboard perpendicular to the ground.

[THE JEW (RaVeN)]

http://www.ocforums.com/showthread.php?t=370139

Similar to the other OC link, but also a good link on plate thickness discussion.

[STEvil]

It would be interesting to see how your 2-True setup performs without the TEC's compared to a single TRUE

[Sam__]

It would be interesting to see how your 2-True setup performs without the TEC's compared to a single TRUE

that would be a very intresting test,

[scifikg]

It would be interesting to see how your 2-True setup performs without the TEC's compared to a single TRUE

Wow, that is a good idea. I wish that I had thought of that before installing all that insulation. It would have been a cool comparison. I'll put it on the list and give it shot later. This is about the time of year that I spend most of my free time outdoors. Plus, I'm anxious start using my gaming machine for games again. Probably won't get around to it till this summer.

[scifikg]

INITIAL TESTING OF TRUE TEC MODULE

Both probes read 23.5C before I turned the PC on.
Both TEC power supplies are set to 9.2V.


The core temperature in the bios is obviously wrong, but at least I know it is getting cold.
Temperature cold plate = -8.5C
Temperature hot plate = 42.0C



In vista, both core temperatures are stuck. This is the stable overclock I had with just air cooling.



Here is an idle shot. You can see where I put the 2 TEC power supplies.
Temperature cold plate = -11.8C
Temperature hot plate = 40.7C



Here is a full load shot. I ran superpi 32M. This is a 11.5 minute test. These are the highest temperature that I saw during the test. The temps rose to this level after a minutes or so into the test and stayed around there for the remainder of the test.
Temperature cold plate = -7.0C
Temperature hot plate = 42.6C



Front view of case with Mr. T.



PC power is on the left and TEC power is on the right. You can even see the blue relay that turns the TEC power supplies on with the PC.


CONCLUSIONS:
The cold plate temps are lower than I expected at -11.8C idle and -7C load. It's hard to say for sure what the core temperatures are but they must be pretty damn cold! Next step is to push the overclock higher, but I'm gonna take a little time to do some gaming before continuing.

[Sam__]

INITIAL TESTING OF TRUE TEC MODULE

Both probes read 23.5C before I turned the PC on.
Both TEC power supplies are set to 9.2V.

The core temperature in the bios is obviously wrong, but at least I know it is getting cold.
Temperature cold plate = -8.5C
Temperature hot plate = 42.0C

In vista, both core temperatures are stuck. This is the stable overclock I had with just air cooling.

Here is an idle shot. You can see where I put the 2 TEC power supplies.
Temperature cold plate = -11.8C
Temperature hot plate = 40.7C

Here is a full load shot. I ran superpi 32M. This is a 11.5 minute test. These are the highest temperature that I saw during the test. The temps rose to this level after a minutes or so into the test and stayed around there for the remainder of the test.
Temperature cold plate = -7.0C
Temperature hot plate = 42.6C

Front view of case with Mr. T.

PC power is on the left and TEC power is on the right. You can even see the blue relay that turns the TEC power supplies on with the PC.


CONCLUSIONS:
The cold plate temps are lower than I expected at -11.8C idle and -7C load. It's hard to say for sure what the core temperatures are but they must be pretty damn cold! Next step is to push the overclock higher, but I'm gonna take a little time to do some gaming before continuing.

Impressive! I was really quite doubtfull about how it would perform but you have proven me wrong. good on you.

[KTE]

You ran two 32M, one for each core, or you just did a one core test?

[evil-98]

wow thats amazing! very impressed with the results. looks like you might be getting some business

[ericab]

wow thats amazing! very impressed with the results. looks like you might be getting some business

agreed. if you sell the plates & TECs as a kit, put me down for one.


....sorry if this has been answered b4, but what results would you expect on a quad core q6600 ?

also, what powersupply is it your using to power the tecs? can i get a link ?

finally; how do you think a (1) Thermalright IFX-14 would fair in this test as opposed to (2) 120 ultras ???

[scifikg]

You ran two 32M, one for each core, or you just did a one core test?

I just started the program and ran a 32M test. I just figured it used both cores. Does it not? I'm certainly up for running other kinds of tests if you guys don't think 32M superpi is enough.

[Jon Boy Deluxe]

I just started the program and ran a 32M test. I just figured it used both cores. Does it not? I'm certainly up for running other kinds of tests if you guys don't think 32M superpi is enough.

Try a 30min run on OCCT, or even a 15min run

OR prime95

They will stress your cpu and the cooling ability of your TRUE tec. I really like what you have done here.

[scifikg]

agreed. if you sell the plates & TECs as a kit, put me down for one.


....sorry if this has been answered b4, but what results would you expect on a quad core q6600 ?

also, what powersupply is it your using to power the tecs? can i get a link ?

finally; how do you think a (1) Thermalright IFX-14 would fair in this test as opposed to (2) 120 ultras ???

I haven't really thought about selling it. I'll have to give it some thought and talk to my machinist.

It is hard to say how a quad will perform over a dual. I figure my cpu wattage output for that test was about 120W and from your sig I'd say your quad is outputing at least 190W. I'm using a cpu calculator. I plan to push my dual to about 200W so maybe that will give you some idea, but I've read that quads can output as much as 300W.

The power supply is a S-320. The 12V version.

One IFX-14 is only slightly better than one TRUE, so cooling these 2 TECs with an IFX-14 isn't enough cooling power.

[ericab]

ok thanks scifikg;

based on that calculator,
a q6600 g0 with an OC voltage of 1.38, and a speed of 3720Mhz outputs about 180 watts... probably wouldn't achieve sub temps at load, but it probably would be better then my idle temps of 27C...

[FrogBite]

Idle temps of 27C are awesome in my opinion.

[KTE]

scifikg, check post#70, that'll be a real cooling test for you and us
Yep, 32M only loads one core.

Is the Cu you used Alloy 11000?

ericab, 1.38v 3720MHz TDP.. what was your starting Vcore?
I don't think it'll be that much, maybe 160W but if you give me the correct starting voltage, I'll let you know.

[ZOMGVTEK]

Wow...

Thats actually what i initially thought you would get, but i started to get a little iffy after everyone was commenting. I wanted to do something like this for a while. But a quad is a whole lot of heat... Maybie...