Conductonaut and RockitCool Copper IHS --> anyone regularly redo theirs?

[aa777888]

i7-7700k 5.0GHz 1.35V Vdroop AUTO
Conductonaut
RockitCool Copper IHS
MX-4
Corsair H115i
EVGA 270 FTW K

I knew that when I put it all together it might be a crapshoot, vis a vis liquid metal vs. copper IHS, but I finally had to redo the re-lid because of rapidly escalating temperatures.

After temp's climbed 20C over the last three months, I knew I had to get off my butt. Things were getting unstable. I blew the dust out and redid the MX-4 with no changes, so that confirmed the problem had to be under the copper IHS. I pulled the IHS off and, sure enough, the Conductonaut was not only rock hard, but it seemed that gravity had also migrated it down to one end of the die (CPU/motherboard stands vertical in my case). The copper was discolored but I really don't care about, that, of course we expect gallium migration into the copper. But this was ridiculous. I had no choice to resort to Scotch Brite to polish the IHS back down to flat because it was very bumpy!

All back together now and almost working fine. Temp's are good, back to where they were, but I've got some instability to iron out, don't know why. It was rock solid before this.

It is pretty clear this might become a once a year preventive maintenance requirement. Any thoughts on this?

BTW, getting 55C average at 50% CPU.

 

[DaveB]

i7-7700k 5.0GHz 1.35V Vdroop AUTO
Conductonaut
RockitCool Copper IHS
MX-4
Corsair H115i
EVGA 270 FTW K

I knew that when I put it all together it might be a crapshoot, vis a vis liquid metal vs. copper IHS, but I finally had to redo the re-lid because of rapidly escalating temperatures.

After temp's climbed 20C over the last three months, I knew I had to get off my butt. Things were getting unstable. I blew the dust out and redid the MX-4 with no changes, so that confirmed the problem had to be under the copper IHS. I pulled the IHS off and, sure enough, the Conductonaut was not only rock hard, but it seemed that gravity had also migrated it down to one end of the die (CPU/motherboard stands vertical in my case). The copper was discolored but I really don't care about, that, of course we expect gallium migration into the copper. But this was ridiculous. I had no choice to resort to Scotch Brite to polish the IHS back down to flat because it was very bumpy!

All back together now and almost working fine. Temp's are good, back to where they were, but I've got some instability to iron out, don't know why. It was rock solid before this.

It is pretty clear this might become a once a year preventive maintenance requirement. Any thoughts on this?

BTW, getting 55C average at 50% CPU.

I'd blame the RockitCool Copper IHS. I delidded my i7-8700K about a year and a half ago, reused the original IHS and never had any issue.

 

[trents]

Don't the instructions for LM TIM products point out that it should not be used on exposed copper surfaces? I think that is pretty common knowledge. Or is it aluminum?

 

[MaddMutt]

Don't the instructions for LM TIM products point out that it should not be used on exposed copper surfaces? I think that is pretty common knowledge. Or is it aluminum?

^ It's not to be used on aluminum Period... BUT it will discolor copper.
All I have read about is : The use of regular TIM under the IHS - CAN and MAY - dry out over a 6~8 months. From what I have read - You are suppose to use LM under the IHS (Re-application is not required) except if you are going to be benching Sub-Zero.
Metal shrinks when cold, thus it may or may not properly contact the IHS. In this usage, it is Recommended to use regular TIM (KPx, Thermal Grizzly, Gelid, Ect, Ect) as it will not shrink.

I would investigate why you had to re-apply the LM. Was it because you did not do it right or something else??
I use the Coollaboratory Liquid Ultra LM on my i7-8700k but I will change that to Thermal Grizzly Conductonaut after reading this review
https://www.hwcooling.net/en/coollaboratory-liquid-ultra-vs-thermal-grizzly-conductonaut/

 

[aa777888]

Don't the instructions for LM TIM products point out that it should not be used on exposed copper surfaces? I think that is pretty common knowledge. Or is it aluminum?

It's aluminum that's the big no-no. The gallium will immediately destroy it. Check out youtube, it's pretty crazy! However, where copper is concerned you are only supposed to get some migration into the copper, which is non-destructive but does change the color of the surface. I don't really give two hoots about what it looks like, only that it works really well.

 

[trents]

Did you spread the LM on both the underside of the heat spreader and the die?

 

[aa777888]

Sure did!

 

[Rev_Night]

I'm bumping this thread to see if anyone else who delided their cpu has had to reapply the liquid metal (under the IHS) after ~6 months. This shouldn't be happening. As I am about to delid my 6700K soon, I want to know what I can do to avoid this.

 

[aa777888]

@Rev_Night -- In response to your PM, I have no idea why what happened happened.

Since this thread is bumped I'll add that the first go around on this lasted from June 2018 to Oct 2019, i.e. about 17 months, give or take. Hard to say when things went over a cliff because the change was gradual, but it didn't bad enough to motivate me until about Oct 2019.

The re-re-lid has been doing fine since Nov 2019, so about 10 months now. We'll all have to wait until maybe Jan 2021 to see if it does it again. I'm keeping a sharper eye on it now.

 

[Rev_Night]

So 17 months mostly good, fix, 10 months good.

Strange