Originally Posted by sultanofswing
Figured I would put this here since it is in a water cooled system.
So on all my watercooled Builds I use Liquid Metal on the GPU as it usually gives me the best results. I ran a 1080ti on a EK Full cover block for 2 years without any single issues with the Liquid Metal itself(Conductonaut).
My setup now is a 2080ti on a Heatkiller Nickel plated block and I am seeing some weird things happening with the liquid metal.
I initially put the block on with liquid metal and temps were great I was happy. I wanted to flash some Bios's to the card but since my card is a 2019 version all the old Bios files don't work, so I pulled the card back out and used my CH341A Programmer to program the BIOS chip.
I cleaned the Liquid metal off, applied a new layer and off I went. Card ran 5-6c hotter than it did before even with the same settings with the new bios. Tore card back apart and the LM application was fine but I noticed something strange.
The top of the die had what appeared to be hard pieces of metal stuck to the silicon. I had to use a credit card to pick the pieces off they were stuck to the die pretty good.
Got all that cleaned off and felt like I didn't want to deal with the LM again and just used Kryonaut.
Put card back together and temps were fine but I couldn't keep it below my target of 40c like it was with LM.
Again card came back apart, Applied LM and temps were back to normal. About a week later I started noticing temps were slowly creeping up past when I first installed, Pulled the card back apart and found the same pieces of hard metal stuck to the die.
So here I sit now, back on Kryonaut as I am not sure what the issue is with the LM. Maybe a bad batch of it? At any rate sorry for the long read hopefully someone has possibly ran into this issue before?
If that is nickel plated copper (AND NOT NICKEL PLATED ALUMINUM), then grab some 1500 grit sandpaper and roughen up the surface with about 40 strokes of the 1500 grit so it isn't perfectly smooth.
After you're done, wipe the surface with 91% alcohol and a microfiber cloth, then apply 1/4th of a size of a grain of rice of liquid metal onto the heat block, then spend about 15 minutes wiping it around the area of the heat block that will contact the die. Keep spreading and wiping it around even if it's already fully spread--just keep wiping it. DO NOT apply downwards force. Just wipe it and spread it normally.
After about 10 minutes or so, you will start noticing that the liquid metal stops feeling so 'liquid' and starts feeling more paste-like (but a very thin paste like). This is your cue you're getting somewhere.
After about 15 minutes, apply a new same sized drop on top of the old layer and just wipe it around fast. This should be very fast and shouldn't take more than a minute.
Then apply a very small dab to the silicon and wipe it around there like normal.
Report the long term stability later.