Liquid Metal Thermal Paste

[max_Jaker]

the card above the other card will run hottest,....

[Ray_GTI-R]

I still don't know why your Metal didn't solidify in the freezer, how long was it inside? -24 C is way below room temperature

Sorry for the delay .... I tried to reply ages ago - foiled, dagnabbit! A repair-install of XP that was o/s since March 2005 did the trick.

The LM was in the freezer for long enough to drop below -24C ... obvious, I know! But it was at least 30 mins, probably longer but not more than 4hrs.

I've done some more physical tinkering with this and it's soooo weird. I've had a blob on a cotton bud for two weeks with an ambient of +/-20C ... no evaporation. LM simply spread out no problem. This time I tried it on stainless steel. Still very liquid!

Then I dry-wiped it and then Brasso'd the remaining - um - smear (!). Polished up a treat!!???!!!

What's all this worrying about, then? Why bother?

Well, what I want to do is get an optimum application method (for myself, coz I'm a thumb-fingered so-and-so) that will ensure that I get best thermal contact with minimum potential damage, given that this stuff is truly very, very conductive. And very, very liquid after a reasonably long time.

Ponder ... As most other thermal "paste" is non-conductive - if I get that application wrong, the dangers are small. No other damage. With this stuff, if I get any drips anywhere conductive& we're talking "big cost", replacement of major components etc etc.

Shoot me down, I don't care ... but make it factual

[IFMU]

http://www.frostytech.com/permalink.cfm?NewsID=46586

[freecableguy]

^^ what a dip, the directions were clear...

[Fairydust]

At least he didn't kill his Pc with it. RTFM if you don't know what you are doing.

[Susquehannock]

Many thanX for performing the test & sharing your results Fairydust.

BTW ....... the reason Brasso works to remove traces of this product is because of the
abrasive properties. Any mild abrasive will likely work in a similar fashion.

Products like Brasso used for polishing usually contain protective additives (wax, oil, etc ...)
which must then be removed before reapplying a TIM. Personally I'd just lapp the surface with
wet sandpaper & clean water instead.

[aMp]

Hey FCG, any results?

[J-Mag]

the card above the other card will run hottest,....

Not true, my GPU1 is below GPU2 and it runs hotter (lian li inverted V series, case it is in)

[HARDCORECLOCKER]

Any news about ???

Just ordered some of this stuff - some ppl with more experience here ?

[Entsafter]

I tested it with watercooling and lapped cpu's! Very impressive results -> 8-10°C drop under load compared to arctic ceramique

[HARDCORECLOCKER]

I tested it with watercooling and lapped cpu's! Very impressive results -> 8-10°C drop under load compared to arctic ceramique

Holy $hit........ - 8-10°C !!! So maybe 4-6°C to AS5, really impressive!!

Did U have some probs to clean up the surface of CPU & cooler ?

Regards to Switzerland - Bern is a beautiful town, but beware of the Aare.............

[MaxxxRacer]

8-10 isnt possible from merely different thermal paste unless you were in Extreme temperatures. IE 100C+ or -100C or less.

[HARDCORECLOCKER]

Hm - so anyone else used that liquide stuff ? Plz post here - experiences welcome.............

[Thrilla]

LoL
AS5 is bad too, it fried my video card.

Anyways, I've developed a liquid metal that does not eat aluminium
http://www.theubercomputers.com/revi...tim/paste.html

I've been using my own paste for a long time, pwns everything ATM, cuz it's really thin, works as an interface material rather than a layer of glue.

[MaxxxRacer]

Thrilla, all liquid evaporate.. that is a principle of physics. Just pointing that out as you said your liquid thermal paste doesnt evaporate.

Even mecury will evaporate, despite its very high atomic mass.

BTW, what material is your paste. You never mentioned.

[Thrilla]

Oh I never new that, but compared to the rate, it should be almost "non evaporating"
I do not want to mention the material yet, all I can say it's non toxic, and the MSDS (material safety data sheet) is pending, so far so good.

Btw thanks for the head up

[MaxxxRacer]

Ahh ic. Going to sell it or something? If not, why the hesitation?

[Trice]

Because someone will steal it!

[HARDCORECLOCKER]

Any new experiences of cleaning? Don't wanna get a damaged surface on my copper cooler........

[Fairydust]

After having used it on a few components, I found some which could be cleaned from the traces and some which required a bit of abrasive polishing, all depending on the durface quality. If you have a near perfect surface it comes off rather easily.

For what it's worth, I doubt the traces have any negative impact on performance as they mostly fill in the gaps of the surface.

[HARDCORECLOCKER]

O.K. - here's some important news I figured out. Don't need to tell this stuff is best and all the blabla U already know but I found out a way for better cleaning the DIE.

This stuff is agressive and even if U use a copper cooler U will see traces of it after removing the paste........... - but no problem at all, tried 3 times a new charge of this stuff and had always the same good temps even the copper looks ugly now.....

But it's possible to clean the DIE. The first chip I used (Opteron IHS off) with the liqiud metal I tried to clean with a dry paper and after that with some Acetone. But I was not able to get the DIE clean - still some shadows on it.

On my second CPU (also Opteron IHS off) I tried a different procedure :

I didn't touch the DIE with the paste and I gave some drops of Acetone onto the DIE.

Waited 1 minute and then started cleaning it with a dry paper. Worked perfect for me, the DIE is looking like a virgin's !!!

So maybe this agressive stuff first needs some time to react with the Acetone before it can be removed without leaving some traces......

[Budwise]

what would you guys recommend on a bare core? Im using AS5 for now, but if there is some new stuff that is better im willing to give it a try...

[Thrilla]

Coolermaster NanoFusion is better, but not as sticky, takes some effort to apply it. But the good thing is it won't short close traces on a motherboard like AS5 would. I don't think it's out yet... or is it?

[MaxxxRacer]

AS5 isnt conductive, only slightly capacitive.

[largon]

This stuff is very close to galinstan (alloy of gallium, indium and tin) used in liquid metal thermometers...

I think I'll grab one of those thermometers and "lure" the goo out...