BigAkita has a new review up on an interesting type of thermal transfer compound called Liquid Metal. The name isn't fluff, either... this stuff is liquid.


Check it out here.

BA...
Good Read....Brown or grey squirrel?

I am curious about this "melts at -20" thought...Does this mean if I put it on my chip under a phase head at -60F that it would liquify and run everywhere on my motherboard? I have to admit I did not know there was a metal that liquified under extreme cold.....
Just crazy......
Digi

Well since its already in a liquid form digi im sure you wouldnt have to worry about that, but does that mean that once the chip gets even colder....lets say -50 will the compound start to stabilize and actually harden?

I'm sure that if you used a cooler that got below the -20C limit it'd solidify.

it reminds me of the liquid metal from this website http://sci-toys.com/scitoys/scitoys/...uid_metal.html but their stuff eats aluminum. or its the gallium stuff that eats it. the other liquid seems similar with a -20 degree melting point. they talk about super cooling where the alloy stays liquid under its melting point

Quote: Originally Posted by carbongraphite but their stuff eats aluminum. or its the gallium stuff that eats it.

FYI = That is the same stuff. . . (I think)

BigAkita references the Alumnium eating characteristics in his "Final Thoughts" section

Quote: Originally Posted by BigAkita The hint that this stuff is a gallium mixture is because of the big warnings at the Coollaboratory website to not use the substance with an aluminum heatsink. The gallium will alloy with the aluminum and actually turn some of the aluminum into a liquid and you end up with a big hole in the center of your heatsink which probably will not assist you with your CPU cooling.

here is dex's attempt at it
looks like dex got it spred a bit better

nice review BA
http://forums.pcapex.com/showthread....t=liquid+metal

Quote: Originally Posted by Digital-World BA... Good Read....Brown or grey squirrel? I am curious about this "melts at -20" thought...Does this mean if I put it on my chip under a phase head at -60F that it would liquify and run everywhere on my motherboard? I have to admit I did not know there was a metal that liquified under extreme cold..... Just crazy...... Digi

I would try It but my container is made of solid aluminum. Maybe when I build my solid copper container.

Hehe....was there a Pimprig rating on this product? no pimp-hat found

no because...

"I have to assume it is a gallium compound. And since I've made that assumption, I cannot recommend this product. Yes, it cools, but the potential for damage to the motherboard or other components is too great if someone were to spill it or if some leaked out between the CPU and heatsink and it went unnoticed."

I'm a scientist so I like experiments... BA, I'm sure you have to have some scrap aluminum lying around. Why not dab a bit of this on there and see what happens?

Good review overall. Sort of a weird product. Ships in a needle-bearing syringe from Germany, acts like mercury, doesn't spread to well, but cools pretty well.

Oh, my other question was if you think the cooler you use is so beefy that the actual thermal compound doesn't make much difference? What if you just used some generic compound? I am racking my brain trying to come up with a test that would be more "thermal compound"-limited, but not getting anywhere

Hmm. I'll try that compound with my Vapochill Micro.
And I'll be applying it with a latex glove because using a textile ( cotton, rayon, etc.)
such as a paint brush or a q-tip sounds inane. (Coollabs should be laughed at for the very idea)
Not only would it be an uneven layer because of all the swirls and what not--
you'd actually be leaving fibres of the swab or brush behind.

As far as the composition of metals in that alloy-- Gallium
is definitely there but it could also include Indium, and Silver. Tin and Bismuth could also be used.

It is likely a composite of Gallium and Indium
doped with trace amounts of silver for Thermal conductivity.

If you want a better way to test TIMs (thermal interface material) would be to create an adjustable CPU simulator and stress it under various loads (up to 150w) a lot of heatsinks today state their solution is capable of dissipating in excess of 120 watts (some at 150w) if you stress the heatsink like that-- only the variables of airflow and thermal interface will be left.

that was very interesting. Nice Review

I found that the best way to get it to spread properly is to use your finger. Rubbing the stuff for about 15-20 secs seemed to allow it to spread just as easily as other TIMs.

When its in a spreadable state, the old credit card trick worked a charm .

Im still using it on my overclocked Opteron now (plus w/c), and getting very good temps, usually not going over 34c load.

Give it another shot BA using your finger, you'll probably get better results