Ivy Bridge Temperatures Could Be Linked To TIM Inside Integrated Heatspreader: Report

Smoogels · 20 May 2012 at 21:56

Ollie1132 said:
IB are actually considered cooler chips in the sense they generate less heat than SB but since they can't dissipate it as well as SB does when they are overclocked they get hotter. Even that shouldn't concern anyone

Thanks for the speedy response. Made me settle for a ivy bridge. Thanks again

Ollie1132 · 20 May 2012 at 21:59

Aha np

Freddie1980 · 20 May 2012 at 22:09

hanluc said:
My results are with the IHS back on the CPU, I tried to try it without but the die wouldn't make contact with my CPU block unless I had the CPU retention holder up on the MB, but then I couldn't get enough downward pressure to the contact pins because the pillars were too high, the screws were bottoming out, as to why liquid pro was so good my guess is what I said in post 274, but it's just a guess.

So once you have taken off the IHS how do you reattached it? Can you get away with just applying MX4 around the edge to get it to hold?

Jokester · 20 May 2012 at 22:17

You just hold it in place whilst you clamp it with the socket retention bracket.

-Mike- · 20 May 2012 at 22:18

MX4 isn't an adhesive. I'm guessing the retention plate on the motherboard clamps it down pretty hard, given this plate has marked every cpu I've used.

moogle · 20 May 2012 at 22:21

-Mike- said:
MX4 isn't an adhesive. I'm guessing the retention plate on the motherboard clamps it down pretty hard, given this plate has marked every cpu I've used.

It's because it's moving across the surface under pressure. Even 775 clamps it down, but only vertically.

Cotti · 20 May 2012 at 22:37

Just started clocking my 3570k today got a max temp of 74c at 4.7ghz 1.280v, I'm reasonably sure it will do 4.8 before hitting 90 but Ill probably hang around at 4.7ghz for now.

Sure is a hell of a lot quicker than my first revision q6600!

Cotti.

Krono5 · 20 May 2012 at 22:47

hanluc said:
I don't think it's a case of "cheap" TIM, because I tried mx-4 in between the IHS and die and the temps actually went up, it took me a few goes to get it back to what it was before. I rate mx-4 quite highly, I've used it for years, I just think think the combination of paste>metal>paste>metal is acting as an insulator. Coollaboratory liquid pro, is liquid metal and doesn't seem to be affected by this.

Yeah, that was what I was getting at with it being "normal tim" instead of solder (not specifically the type of tim but that it wasn't solder). As solder kind of bonds the two surfaces together so there's not really any heat insulation at all, as if the IHS and die were effectively the same item. So extra good TIM may act in a way that the heat transfer/it's heat conductivity is enough to reach a critical point where heat can escape as fast as it is being produced.

The Solder on Sandybridge chips has a thermal conductivity of around 80W/mk. After doing some research MX-4 has a thermal conductivity of only 8.5W/mk which is probably enough for use between the IHS and CPU cooler Heatsink, but Liquid pro has a thermal conductivity of 82 W/m·K which is pretty much the same as solder and this may be required to dissepate the heat into the IHS fast enough.

Krono5 · 20 May 2012 at 22:51

There's also the possibility of it being viscosity related, or particles being able to get closer to the Die.

-Mike- · 20 May 2012 at 23:08

Krono5 said:
The Solder on Sandybridge chips has a thermal conductivity of around 80W/mk. After doing some research MX-4 has a thermal conductivity of only 8.5W/mk which is probably enough for use between the IHS and CPU cooler Heatsink, but Liquid pro has a thermal conductivity of 82 W/m·K which is pretty much the same as solder and this may be required to dissepate the heat into the IHS fast enough.

/Goes off to find the W/m-k of IC diamond

http://innovationcooling.com/ICDDatasheet.htm

-Mike- · 20 May 2012 at 23:16

Do ocuk sell this liquid stuff?

moogle · 20 May 2012 at 23:16

I'm glad I kept my syringes of Liquid Pro

Sasso · 20 May 2012 at 23:19

-Mike- said:
/Goes off to find the W/m-k of IC diamond

http://innovationcooling.com/ICDDatasheet.htm

hummm liquid pro you say..... surely though.... it's just going to run out onto everything xD

moogle · 20 May 2012 at 23:20

-Mike- said:
Do ocuk sell this liquid stuff?

Yes they do

http://www.overclockers.co.uk/showproduct.php?prodid=TH-005-CL

I first used mine on an Athlon X2 CPU and it my heatsink survived it but it did eat into the CPU IHS a bit.

hanluc · 20 May 2012 at 23:34

moogle said:
Yes they do

http://www.overclockers.co.uk/showproduct.php?prodid=TH-005-CL

I first used mine on an Athlon X2 CPU and it my heatsink survived it but it did eat into the CPU IHS a bit.

It must have been aluminium then, liquid pro is said to be unsuitable for aluminium IHS for that reason.

hanluc · 20 May 2012 at 23:38

Krono5 said:
Yeah, that was what I was getting at with it being "normal tim" instead of solder (not specifically the type of tim but that it wasn't solder). As solder kind of bonds the two surfaces together so there's not really any heat insulation at all, as if the IHS and die were effectively the same item. So extra good TIM may act in a way that the heat transfer/it's heat conductivity is enough to reach a critical point where heat can escape as fast as it is being produced.

The Solder on Sandybridge chips has a thermal conductivity of around 80W/mk. After doing some research MX-4 has a thermal conductivity of only 8.5W/mk which is probably enough for use between the IHS and CPU cooler Heatsink, but Liquid pro has a thermal conductivity of 82 W/m·K which is pretty much the same as solder and this may be required to dissepate the heat into the IHS fast enough.

That seems like a good explanation, it certainly tallies up with what I found.

moogle · 20 May 2012 at 23:38

hanluc said:
It must have been aluminium then, liquid pro is said to be unsuitable for aluminium IHS for that reason.

Maybe, I didn't know how to tell.

I had to use some steel wool to scrub it off and the markings were still there and readable

Did have pretty low temps for a 90nm CPU using the stock heatsink though

Krono5 · 20 May 2012 at 23:48

hanluc said:
That seems like a good explanation, it certainly tallies up with what I found.

If someone replicated the results it would be pretty conclusive as Jokester only really noticed a difference in temps when completely removing the IHS, so if he for example were to switch out the MX-4 he was using (I think it was MX-4?) for Liquid pro and noticed a significant drop in temps we'd have a much better idea of the exact reasons.

I would do this myself when I get my 3570k, except I'm too poor to risk doing it wrong, so will prob leave it unmodded for a few years until I'm almost ready to upgrade again

-Mike- · 20 May 2012 at 23:50

Could someone brigher than myself please explain the following

A/S 5

Thermal Conductance:
>350,000W/m2 °C (0.001 inch layer)

IC Diamond

Thermal Conductance: 4.5 W/m-K

How do you convert these figures into comparible ones?

Krono5 · 21 May 2012 at 00:15

W/m-k = watts per meter kelvin = the SI unit.

1 W/mK = 1 W/m.C

only difference is that is converted into 0.0001 of an inch from a meter.

To convert that to the same units, simply mutiply/divide by 0.0000254 (which is 0.001 inches in m or vice versa)
= 8.89 W/m-k.