New Idea. Not Mercury.

[CPFitz14]

New idea.

Gallium. It melts at like 85F. Apparently in this alloy it melts at 51F/10C. Thats perfect for a water cooling system. You'd have a hard time getting your hands on enough of it, but here it is on ebay:

And its also apparently non-toxic. However it does say to wash your hands after touching it, and not to eat it. Duh...

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=413&item=3285058342&rd=1&ssPageName=WDVW

This LIQUID METAL alloy melts at the lowest eutectic point of the Gallium-Indium-Tin Eutectic, which is 51°F/10°C and a full 9°F lower temperature than the freezing point of Gallium-Indium alone. Unlike elemental Mercury, which has a rather high surface tension and beads up on a hard surface - THIS IMPRESSIVE ALLOY exhibits almost ZERO surface tension and wets almost everything including metals, glass, plastic, wood and ceramics just like a coating of paint. This alloy contains NON-TOXIC metals that are not regulated by the EPA for their toxicity

-CPFitz-

[CandymanCan]

and what would this be good for ?

[ares350]

I thin in general you want your liquid way below 51degrees farenheit if your gonna go through all the trouble of this.

though most everything but a simple radiator wouldnt be effective on it.

[CPFitz14]

Thats 10C though, I'd take temps that were in 10-20C anyday. I'm just trying to figure out how it would work right now...

You could put it in a heatpipe.

[blackjackel]

melting point is not the only thing that you need to look at when looking for something to watercool, there are a variety of things such as how much heat it has to hold....etc.

There is nothing on earth (i think) that is better than pure pure water for watercooling situations, and even if there were it would cost like 500 dollars to get enough of it to fill a loop.

besides, through that ebay acution you would need to buy like 50 of those itty bitty bottles to fill a loop, thats gona run ya 300something dollars.

but hey, what can i say, i love out of the box thinking, new ideas, & stuff like that, so good job on that <thumbsup>

[fletch]

Gallium is very expensive (more than gold I belive). Depending on other minor factors, the heat would have to transfer through the entire system before it would flow (in order to melt it first). The melting point is 30C, and since it is not liquid at room temperature, it would cause problems...

I don't know what to think of that alloy though. It might be heavy, and that could pose a problem. Other than that, a big pump might work wonders :D



If you are a Materials Science Engineer this just may prove very useful to that specialized application you had in mind

Hey... I'm a Materials Science Engineer... :D

[Korndog]

Originally posted by CPFitz14
New idea.

Gallium. It melts at like 85F. Apparently in this alloy it melts at 51F/10C. Thats perfect for a water cooling system. You'd have a hard time getting your hands on enough of it, but here it is on ebay:

And its also apparently non-toxic. However it does say to wash your hands after touching it, and not to eat it. Duh...

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=413&item=3285058342&rd=1&ssPageName=WDVW



-CPFitz-

Yea, i've heard about this stuff for telescopes because it can make a good parabolic mirror just by spinning the liquid. But from what i've heard, it can go down to -4F without freezing, or atleast a certain kind of it. They are even using it in thermometers now as well because its just as accurate as mercury.
Now the really question is how dense is this stuff? If its too dense, its somewhat pointless to try unless you can use it w/o pump, like a heat pipe, which would be very effective IMO.

[Sentential]

Imagine what that stuff would do to a pump:eek:!

[Korndog]

Originally posted by Sentential
Imagine what that stuff would do to a pump:eek:!

You'll probably need a pressure pump or a screw pump like the ones cathar was suggesting for watercooling.

[fletch]

Can't find a Ga-In-Sn phase diagram to see the density :(

[Korndog]

oh boy.. its density is .2348 lb/in^3
i think water is .0037 lb/in^3 :-/
someone correct me if i'm wrong, i hate the british/american measurment system :(

Edit:
http://www.aimsolder.com/troubleshoot/AlloysSpecialty.pdf
the link.. bottom of the page

[greenman100]

Problem #2:

Gallium
Specific Heat Capacity (c) at 300K
370 J/(kg.K)

Water
Specific Heat Capacity (c) at 300K
2100 J/(kg.K)

300K=27C

What's the point, with stats like that, unless that alloy's thermal properties are way better?

[fletch]

^ Good point - it would have to have a high flowrate.

[Blind Tree Frog]

There is nothing on earth (i think) that is better than pure pure water for watercooling situations, and even if there were it would cost like 500 dollars to get enough of it to fill a loop.
Mercury is. that's why people keep trying to figure out how to use it.

[fletch]

The thermal conductivity for mercury is a little over 10 times that of water, but the heat capacity for water is more than 30 times that of mercury. Which would matter the most?

[Korndog]

what about thermal conductivity of this stuff?

here, check out the last page under 4.1
http://accelconf.web.cern.ch/AccelConf/p01/PAPERS/TPAH012.PDF
they used this stuff for cooling. wish they had more details though.

[fletch]

Interesting... Wonder how they physically used it for cooling (where it ran, dissipated, etc.)

[Korndog]

Originally posted by fletch
Interesting... Wonder how they physically used it for cooling (where it ran, dissipated, etc.)

probably like how we do it.
Similarly, my physic teacher used huge heat exhangers that look like a transmission fluid cooler to cool the massive laser equiptment that he used when he worked for livermore labs and they generated lots of heat. I don't know why, but i want to try this stuff pumpless.. just depend on natural circulation, it works pretty nicly with water (like a 10c increase from normal), something with higher thermal conductivity should yield better core temps right?

[greenman100]

quick calculations:

capacity of a watercooling system with ~3ft of 1/2"ID tubing:

1.5L

density of callium: 5.9g/ml

so, that's 9400grams.

27g here (http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=413&item=3285982706&rd=1&ssPageName=WDVW) for $37.

that's $13,000 to fill 'er up on some Gallium.


I'll pass.

[greenman100]

Originally posted by Korndog


I don't know why, but i want to try this stuff


I'll tell you what. Send me the $13,000, and I'll send you an AMAZING system that gets SUB-ZERO temps!

[fletch]

^ LOL - once again, good point. Even if this alloy was excellent for cooling, it's not cost efficient at all to try this stuff right now.

[Korndog]

now wait a sec there
this isn't pure gallium for starters, it doesn't cost that much, and what i wanted to try is a natural circulation setup. Something with very thin tubing copper tubing and with less then 30 cm length of tubing.
I just like the benifits of get low temps, without having to worry about a pump failing and without a compressor or peltier.

[greenman100]

Originally posted by Korndog
now wait a sec there
this isn't pure gallium for starters, it doesn't cost that much, and what i wanted to try is a natural circulation setup. Something with very thin tubing copper tubing and with less then 30 cm length of tubing.
I just like the benifits of get low temps, without having to worry about a pump failing and without a compressor or peltier.


doesn't cost that much? for it to make sense, it'd have to be less that $600.

Find me a source of gallium anything for less than $.063 a gram.

[fletch]

I wouldn't rely on convection alone to cool my processor. When something has as low a specific heat as gallium (compared to water), it's going to get hot in a hurry. I would think temps would get very high with this unless some sort of assisted flow was there. Perhaps with the much higher thermal conductivity it will pull heat away faster, but I'll just say I wouldn't want to be the guinea pig :D

[Korndog]

sorry, i was refering to the alloy.
the alloy is only part gallium, if it has lower melting points, couldn't the specific heat be different as well?
All i'm thinking is that this could be a good replacement for the stuff they use in heatpipes, gallium at even high temps has low vapor pressure.

[CPFitz14]

it would be interesting to drill a small hole, remove the orginal heatpipe liquid, and fill it with this.

Probably wouldn't cost too much either. And i'd imagine if you contacted this guy on ebay, you might be able to find a lead as to where to buy it in greater quanities for cheaper.

[fletch]

^^ The specific heat will more than likely be around the same. Gallium I belive had higher specific heat than both of the metals it was alloyed with. I don't think it could bump it up to close that of water, though.

Using it in heatpipes sounds like it might be worth a shot. Probably not too expensive, as fitz pointed out. Go for it :D