Question Liquid metal worse off??

Apr 13, 2019
11
0
10
I was having i5 4670K running at 4.2GHz at 1.24V . Wanting to have better OC performance, I replace the the thermal paste with Luquid metal. And the temperature just get worse. Before that , a 3 mins Aida 64 fpu test reach 93C max. And now it shoots up less than a sec. Does it mean I should return to normal thermal paste or did I do sth wrong??
 

izoli

Distinguished
Apr 29, 2011
597
0
19,210
When you say you replaced thermal paste, do you mean the paste applied between IHS and heatsink, or the TIM paste between die and IHS?

Edit: Assuming you mean between the IHS and CPU Heatsink, that will not give you very much of a difference in thermals. Usually no difference at all. If you want to do it anyways, I would work the gallium into an extremely thin layer that sticks evenly to the surface of both the CPU heatsink plate as well as the IHS, then carefully apply the CPU heatsink. This way both surfaces have a very thin layer of gallium already sticking to it and then the gallium coming together should easily form a bond.
 
Last edited:

USAFRet

Titan
Moderator
Before liquid metal, it gradually rise to 93. After that, it shoots up to the sky immediately......So should I go back to ordinary thermal paste?
THe problem here is:

You're going to take it apart
Remove the existing paste
Put on a new layer of the other stuff
Reassemble everything
And it will actually work, not instantly jumping up to 93C.

The problem is that you will attribute the better temp to the change of paste, not the proper assembly you did the second time.

The actual current problem is the assembly, not what paste is applied.
 
Apr 13, 2019
11
0
10
I place the liquid metal INSIDE, between IHS an TIM. And Im sure it form a bond for I place it upside down and the IHS stick to it. And that is what puzzle me...... I can think of nothing I did wrong......
 
Apr 13, 2019
11
0
10
Correction: before liquid metal ( ie Intel thermal paste) , temperature went up steadily and reach 93C in 3 mins time. Max power drawn 94.33. Now ( ie liquid metal) temperature surge to 100C+ in less than a second.
 

izoli

Distinguished
Apr 29, 2011
597
0
19,210
Between the IHS and TIM or the IHS and die? The process should have involved cleaning all the TIM off of the die and IHS with some alcohol pads then cover the cpu die and the underside of IHS that makes contact with die in a very thin layer of liquid metal then sealed it back shut with a gasket maker, after cleaning the previous gasket remains.

What did you use for polishing?

What delid tool did you use?

Did you clean the old gasket before making the new gasket? could be the gasket is too thick and the IHS isnt making contact with die?
 
Apr 13, 2019
11
0
10
Between the IHS and TIM or the IHS and die? The process should have involved cleaning all the TIM off of the die and IHS with some alcohol pads then cover the cpu die and the underside of IHS that makes contact with die in a very thin layer of liquid metal then sealed it back shut with a gasket maker, after cleaning the previous gasket remains.

What did you use for polishing?

What delid tool did you use?

Did you clean the old gasket before making the new gasket? could be the gasket is too thick and the IHS isnt making contact with die?
Yeah I cleaned with alcohol to remove all the thermal paste and applied liquid metal between IHS an die.

Before that, I used some metal polish gel (it writes)to polish thin and out of IHS

Aida 64 fpu