[SOLVED] Delidded 9700k now runs with high temps

[JJoyce]

Hi all,

As per title i've delidded my 9700k yesterday + copper ihs. (from rokit cool)
I'm going to upgrade to thre12700k soon anyway so why not?

I saw something like 10-15c drop, but today temps are similar to what i had pre-delid.

Any idea on what is going on?

Thermal grizzly - Conductonaut used

Waiting for the weekend to open it up and have a look.
Thanks all

 

[Phaaze88]

What did you use between the die and IHS?

 

[JJoyce]

What did you use between the die and IHS?

Thermal grizzly - Conductonaut

 

[Phaaze88]

And you 'painted' both the die and underneath the IHS(not the entire area, just the general area the die will make contact with)?
Hmm, not familiar with this scenario. I've used too little Conductonaut before, but bad thermals showed up right away.


It just sounds like the cooler came loose.

 

[JJoyce]

Yes i painted both sides, only thing is that i did not re-sealed.

Maybe after the installation the ihs somehow?

Well i've order some silicon, just gonna use a couple of drops to keep it in place

 

[Phaaze88]

I have read that resealing the IHS isn't necessary - that the mounting pressure from the cooler and its backplate can fill that role, but haven't personally done that.

So yeah, if you had used too little LM, it would've showed up right away... so mounting pressure must've loosened from hot+cold cycles in that short window of time.

 

[JJoyce]

Yeah that could be the explanation then.

Many thanks for your time

 

[helper800]

I am just curious as I have not used LM before. When you applied LM you do it in a thin layer correct? If so at what temperature does the LM have the viscosity to drip down in the bottom IHS (Motherboard is mounted vertically) if at all? It is my understanding that that LM can become more alike a liquid at higher temps. What prevents the undoing of the application of LM on the surface of the CPU die between the IHS and the chip? Thanks in advanced!

 

[Phaaze88]

@helper800
It needs to be balanced - not too little and not too much. It needs to be a little 'wet'. I guess it's like painting.
Apply too little, and expect to POST with 80C-ish temperatures. Apply too much, and expect to see some LM has rolled down the sides of the die after taking the IHS off.

I've got a layer of clear nail polish top coat over the little SMDs around the die to help protect them from LM contact.

What prevents the undoing of the application? Nothing really, as the LM slowly alloys with the surfaces over time.

 

[helper800]

@helper800
It needs to be balanced - not too little and not too much. It needs to be a little 'wet'. I guess it's like painting.
Apply too little, and expect to POST with 80C-ish temperatures. Apply too much, and expect to see some LM has rolled down the sides of the die after taking the IHS off.

I've got a layer of clear nail polish top coat over the little SMDs around the die to help protect them from LM contact.

What prevents the undoing of the application? Nothing really, as the LM slowly alloys with the surfaces over time.

Interesting. I never got into delidding or LM use myself but I have lapped an IHS and coolers before. I always feared destroying the CPU with delidding. I guess its an extinct craft for new CPUs anyways as they solder the IHS now which is good.

 

[Phaaze88]

Interesting. I never got into delidding or LM use myself but I have lapped an IHS and coolers before. I always feared destroying the CPU with delidding. I guess its an extinct craft for new CPUs anyways as they solder the IHS now which is good.

Yeah. I'm neutral towards it. Direct die mounted a 7820X with LM because my curiosity got the better of me; I was definitely scared of wrecking my cpu and I didn't have a Plan B.

Now that they're soldering cpus, just stick with good ol' paste on the IHS.

 

[JJoyce]

Yeah. I'm neutral towards it. Direct die mounted a 7820X with LM because my curiosity got the better of me; I was definitely scared of wrecking my cpu and I didn't have a Plan B.

Now that they're soldering cpus, just stick with good ol' paste on the IHS.

My 9700k was soldered, still don't know if it was worth it

 

[Phaaze88]

My 9700k was soldered, still don't know if it was worth it

In your scenario, it wasn't as beneficial, compared to older chips that were using the Intel TIM. I can't say whether it was worth it to you or not.

It greatly benefited 3rd - 8th gen as it reduces the gap and improves contact between the die and IHS.
The solder already does this pretty well.

 

[JJoyce]

In your scenario, it wasn't as beneficial, compared to older chips that were using the Intel TIM. I can't say whether it was worth it to you or not.

It greatly benefited 3rd - 8th gen as it reduces the gap and improves contact between the die and IHS.
The solder already does this pretty well.

Yes i was aware of that, but i was going for Lm + copper ihs, and i've orderd some grit papaer as well so i will lap it.

Hopefully i will see at least a 10c permanent drop

 

[Phaaze88]

You will lap what, and why?

 

[JJoyce]

You will lap what, and why?

I might have used the wrong term, not lapping but sanding.


Just checked my old ihs and it was not straight, i wonder if the copper ihs could be the same ...
So yes i will send (if necessary)the copper ihs

 

[Phaaze88]

https://www.gamersnexus.net/guides/3238-custom-copper-ihs-tested-on-intel-i7-8700k-cpu-rockit-cool

"The Intel IHS is also a non-flat surface – some coldplates are made concave to match the convex curvature of the Intel IHS (depending on your perspective of the heat spreader, granted), whereas the Rockit Cool solution is nearly perfectly flat. Most coolers have some slight conformity to mounting tension, flattening out coldplates atop a non-flat CPU IHS."
The die is monolithic and located in the center, so the slightly convex curve of the Intel IHS works for it; cooler mounting pressure is then focused at the center.

 

[JJoyce]

https://www.gamersnexus.net/guides/3238-custom-copper-ihs-tested-on-intel-i7-8700k-cpu-rockit-cool

"The Intel IHS is also a non-flat surface – some coldplates are made concave to match the convex curvature of the Intel IHS (depending on your perspective of the heat spreader, granted), whereas the Rockit Cool solution is nearly perfectly flat. Most coolers have some slight conformity to mounting tension, flattening out coldplates atop a non-flat CPU IHS."
The die is monolithic and located in the center, so the slightly convex curve of the Intel IHS works for it; cooler mounting pressure is then focused at the center.

Not only this was really intresting but it also changed my mind about re-sealing the ihs.

Again, many thanks !