Slashdot Mirror
Liquid Metal Cooling in New ATI Video Card
MellowTigger writes "Water cooling is so passe, definitely 20th-century. What's the 21st-century geek to do to keep his gaming video card cool? Try the liquid metal technology that will be included in the ATI Radeon X850 XT video card using the cooling technology from Sapphire. This material is reported to be non-flammable, non-toxic, environmentally safe... and 65 times more thermally conductive than water."
It's Gallium. It's developed by NanoCoolers. I wonder if the name means it's a nano-technology. That would be exciting !
War doesn't prove who's right, just who's left.
A one slot and nearly-silent top end card.
Nice guess, it's indeed Gallium, as a google search will confirm you.
War doesn't prove who's right, just who's left.
61% Gallium
25% Indium
13% Tin
1% Zinc
Solid at 6.5C
Liquid at 7.6C
Correct me if I'm wrong, but isn't heat capacity more important than heat conductivity in this case?
The article mentions a special kind of pump with no moving parts, so I assume the liquid is moving around from some sort of radiator to the GPU and back.
Water has the highest value of heat capacity if I'm not mistaken (4186J/KgK), so in case of a moving cooling liquid, the higher the liquid's heat capacity, the less water will have to be moved to move a specific amount of heat.
What does heat conductivity matter, then?
Here's a little bit more info, no word on T1000's involvement.
However, I wouldn't be surprised to hear that owners of this new "metal liquid cooling" at a LAN party ending quite horribly.
I can see it start with a few leaks, the liquid metal all joins together... T1000 lives again...
I'm pretty sure they're using gallium. It melts at 85F, is nontoxic (unlike mercury), and is nonflammable (unlike rubidium, cesium, sodium, and potassium, the only other metals I know of that melt at reasonable temperatures for a graphics card). Gallium also has almost exactly 65 times the thermal conductivity of water.
Is this possible?
Last time I checked molten metal (burning death) and Mercury (deadly poison) was toxic.
AHHH, Nothing like a warm sip of heatsink juice to warm you up on a cold winter night.
--Greg
It's an alloy of gallium and indium
http://nwc.serverpipeline.com/news/54200844
Geeze, why does /. keep on linking to physorg, which has crappy articles and no links to real information about stuff.
Here's a more in depth article on X-bit. NanoCoolers has a pretty in depth description of the product. It's basically a watercooling loop but using a molten metal. The really cool part is that because the metal is obviously electrically conductive, they're using a DC current combined with some magnets to take advantage of Lorentz force to propel the fluid.
The liquid metal cooling topic was covered on /. before, eg:/ 03/1421243&tid=222/
http://hardware.slashdot.org/article.pl?sid=05/05
The technology probably derives from http://www.nanocoolers.com/products_cooling.php/
Sapphire is just the OEM manufacturer of ATi cards. For quite a while you could only get ATi branded cards but now you can get them from ATi or Sapphire. I doubt they have much to do with the technology besides licensing it.
Yes but probably not pure Ga but the GalInStan eutectic instead. Anyway though it's non toxic/non pyrophoric/non volatile is it still a METAL and a leak inside your computer would be instantly catastrophic for obvious reasons.
- "Hear that?! The percolations are imminent! Cease your ingress!"
It'll be between the liquidus and solidus lines, so you'll have a composition of liquid plus some small amount of alpha-phase.
Both liquid and solid: Recrystalizing state.
Oh well, what the hell...
Anyways, there are plenty of metal alloys that are liquid at or slightly above room temperature.
But even discounting alloys, there's are a few other elemental metals other than mercury that are liquid at room temp (assuming your room has a computer or two to keep it warm): Gallium melts at 29.76 degrees C and Cesium at 28.44 degrees (I'd keep the latter far from my computer though). Rubidium melts at 39.31 degrees, so it'd be liquid at the temperatures today's GPUs reach (but I'd keep that far from my computer too).
Did some research, found the following two patents:
T O2&Sect2=HITOFF&p=1&u=/netahtml/search-bool.html&r =1&f=G&l=50&co1=AND&d=ptxt&s1=nanocoolers&OS=nanoc oolers&RS=nanocoolers
T O2&Sect2=HITOFF&p=1&u=/netahtml/search-bool.html&r =2&f=G&l=50&co1=AND&d=ptxt&s1=nanocoolers&OS=nanoc oolers&RS=nanocoolers
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=P
And
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=P
It looks like they're using a Gallium/Indium (rare elements) alloy. This is certainly not environmentally safe from a chemical point of view as these are toxic heavy metals. I think by environmentally safe they are pointing to the "sealed" system that they are advertising. That is, they dont exepect the systems to leak, as they do not require any refilling.
Basically, their argument appears to be that it's safe because it cant get out, just like coolant in a nuclear reactor. This is actually a reasonable claim, however, we shouldnt take it to mean that the liquid metal coolant itself is evironmentally sound, just that the system, while in operation, is.
P.S. it appears they've also experimented w/ Lead/Bismuth, mercury, and Sodium/Potassium alloys.
Antec Phantom 350 PSU
Gigabyte 6800 fanless (only 12 pipes, but not a big sacrifice...though if this liquid metal stuff works it should make things easier the next time I upgrade)
Thermalright CPU heatsink with a 120mm fan on a Zalman fan bracket and set at minimum on a Zalman fanmate control.
All in an Antec 3700BQE case with quiet 120mm exhaust fan.
The annoying thing was that as soon as I got rid of one whining or droning noise I'd notice a slightly quieter one... Now, it's inaubible except in dead silence.
I would assume that they're using some kind of an alloy that has a lower freezing temperature. One possibility is an alloy of gallium, indium, and tin (also known as Galinstan) which has a freezing point of -20 degrees centigrade.
..wayne..
Please elaborate on MS DRM being cracked.
It was cracked in Japan and then quickly uncracked.
Rather, any of the phone-home restricted content forced an upgrade of the decrypter that was no longer crackable by the original method. T2 was liberated before the phone-home system had started to push out the new software.
The original method amounted to running the player under a debugger and looking for the decryption keys in a known location in memory, grabbing the keys and then using them to manually decrypt to a file. The new software checks for the presence of a debugger and refuses to run. I'm sure it is only a matter of time before that is also circumvented.
In theory, if you have not accepted an upgrade to the windows media system in the last month or so, all of the "on disc" restricted files could still be liberated. It is reportedly a fairly tedious manual process.
I think there is some discussion of the process on doom9.org if you want to dig deeper.
When information is power, privacy is freedom.
>>Liquid metal computer component cooling was discussed here not to long ago.
Indeed. link to Liquid Metal CPU Cooling 03May05
" Darn, the T2 joke has already been made *thinks of other joke to get karma..*"
Actually, Humor karma doesn't go on your karma record. Check the faq about karma.
Water takes one calorie of energy input per degree Centigrade raise in temperature.
Two words: relative density.
Specific heat is measured in J/KgK, you multiply by density and divide by a 1x10**6 to get J/cm**3K, which is what you're talking about.
The checks for a debugger are in the kernel. Good luck patching that out. :)
"But I'm still right here, giving blood and keeping faith. And I'm still right here."
Buy an Eden-based machine if that's what you like, but I don't see how it's "wrong" for a computer to have liquid cooling.
For most of the history of computing, liquid cooling has been associated with the ultra-high-end computing. I once saw a picture of the pipes that carry the liquid to cool one of the NSA's supercomputers, they were 8" or 10" pipes, if I recall.
Besides, cars get faster and more powerful, televisions get larger, stereos get more powerful, houses get bigger, and cometic surgeons use more and more implants. I don't see anything atypical about computers competing for insanity, either. =)
steve
Oh, you're not stuck, you're just unable to let go of the onion rings.
Right, and the specific heat of water is 1 calorie/(gram*degree C) which is approximately 1 calorie/(cc*degree C), depending on the density of your water. That's little-c calorie.
Therefore, the original poster is wrong. Water only absorbs one calorie for each degree, not a kilocalorie. However, he is correct that gallium has a poorer specific heat:
Gallium: 370 J/(kg*K)
Water: 4184 J/(kg*K)
And it's, apparently, only about 6 times as dense (if Google hasn't failed me). So the same volume of gallium would heat up about twice as fast as water (correct me if I'm wrong).
Then again, things like heat transfer are probably better with gallium, and it might be easier to cool than water, so who knows?
I've come for the woman, and your head.
Actually, the electronic damping on the C6 Corvette does not depend on liquid metal in the shocks as you assert. What the shocks actually contain is oil in which are suspended tiny ball bearings (for lack of a better word). These ball bearings are coated with a special substance that not only allows them to float in the oil, but also distributes them uniformly throughout the oil. These ball bearings are then acted on by electromagnets to vary the viscosity of the oil. To quote Car and Driver:
Well, it may be solid, but that doesn't mean it isn't absorbing heat. Melting typically takes significantly more energy than would be absorbed by a temperature increase, so the melting metal would probably actually cool your CPU better than the liquid result.
In addition, metals (by definition, I believe) have good heat conduction properties (that's why some heatsinks use metal heat pipes these days), so a large portion of the metal would probably end up melting and start flowing. You probably wouldn't just get a bubble of super-heated liquid gallium frying your CPU. You'd just start off with a normal-ish metal heatsink until the gallium melted.
Of course, there are other reasons not to use gallium, as other people have mentioned here.
I've come for the woman, and your head.
Well, sodium is non-toxic until it comes in contact with any living thing, where it will combine with water to form sodium hydroxide (aka Liquid Plummer). It's like the tree falling in the forest question, but in my book, sodium is toxic.