My Liquid Metal Experience

A couple fo weeks ago while rendering a video 100c on my cpu caught my eye, I knew that it was time to do something about it and after a nice research on google I decided to try them all. Undervolting, fan profile and replacing thermal paste, but I still wasn't happy with the results. Then I heard about this magical compound that would drop 10c even on a cleaned system! So, I was like, let's go for it! Today I got my tube and was really happy about it and after applying kapton tape and the lm the temps were even worse than before I knew something wasn't right therfore opened notebook again and while I was changing the tip 3 tear sized drops landed in my opened notebook, FUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUU, one splashed in the gpu, another one in the battery bay another one on the heatsink. Ok, I can fix this, tried to suck the lm back in the tube but popped the plunger out and more lm spilled all over the place So I unscrewed the tip and slowly pushed the punlger back in the barrel then put the tip back on and sucked in more lm and did this for 30 mins. Surprisingly I didn't loose any lm, anyway after applying more the second time temps were better but not what I expected. Too be sure I didn't put too much I opened it up again one last time and it was all fine.

RESULTS HERE:

The laptop is an msi gt70 barebone from medion.



-0.090v repasted with prolimatech pk-2 twice and fan set to 100% when gpu/cpu hits 80c Also, there is 1c difference when the gpu is not overclocked.



Thermal Grizzly Conductonaut with cpu undervolt and fan set to 100% when gpu/cpu hits 80c



Thermal Grizzly Conductonaut with cpu undervolt and new fan curve!

If I'm not mistaken it looks like that the gpu heatsink is not making proper contact.






Well, thanks for reading I hope you had fun and if you have any suggestions let me know!

 

scary read....glad you cleaned it up and it is fine now.

 

When using conductonaut, make an initial small drop appear out of the needle nose over something over than the laptop, and use a slow twisting motion, not a push. The plunger can actually twist quite a bit. Then drop that somewhere safe, then making sure none will spill out (small backwards motion), now you can apply it over your chip, and then twist the plunger slowly until another drop starts to appear. Then keep twisting (do not push it, twist) slowly and then another drop will fall right on your chip. Repeat if you need a second drop. Being in a hurry causes accidents with this stuff.

If you are pasting the GPU as well, then pull the plunger back so the drop is gone, move it away, make a drop appear over some other object, twist a small drop out, then with no more drops showing, go back to the GPU and twist another drop there.

Getting the very first drop out is what often causes accidents or too much to squirt out.

 

To solve the contact issue on the gpu heatsink I probably have to replace the thermal pads with thinner ones right?

 

Liquid metal doesn't build much on top of die. Check the pads imprint from the components. This can be done without LM. But this doesn't look bad. And Liquid metal don't make the big temp differences on GPU vs. what you will get on Cpu.

 

So I don't have to worry about the die burning out even if the max temp is 66c

 

66C ain't a problem. This is... http://forum.notebookreview.com/thr...15-owners-lounge.825553/page-45#post-10825773

 

That wasn't what I meant , the die not making proper contact and perhaps having higher temps than what the core shows or is that not possible?

 

The temp from die should be quite accurate. With bad contact you would see awful temp results. Be sure pads have proper contact. Because you don't know the temp from rest of the components on GPU graphics card.

Bad contact from Heatsink will will degrade LM quite quickly. And You will fast find out if this is a problem..

 

Ah okay then, thanks for the clarification!

 

I'm glad I don't play around with things I cant control and will break my notebook.