Danamics - a revolutionary cooling system?!

[Boogerlad]

exactly. pressure buildup because it's closed loop

[bobbobson]

Blow off Valve! woot.

[krogen]

If the metal won't boil then the chance of it leaking under pressure shouldn't be significant...

So a closed loop is not a problem.

[Serra]

correct me if i am wrong here bt wouldn't there be some gas that is emitted from the mercury as its heated???

what are they doing with those emissions ?

and the pressure build up also,,

There is no chance this would be mercury. Other liquid metal solutions exist which are substantially less toxic (read some above posts, whatever this is is likely to involve gallium + tin alloys).

[twin_savage]

Does anyone here have experience with electromagnetic pumps?
what kind of head pressure can they accomplish.

[cegras]

It doesn't have to be completely metal to work though does it? Couldn't it be metal particles in a suspension. This would be relatively safe and cheap. Moving the metal particles with a magnetic field should move the liquid medium along with it. In fact, if they're going to use an electromagnetic pump, would the main metal have to be iron?

I just wonder how much of a pressure increase there is from room temperature to say 70C. Unless the liquid is completely non-volatile up to that temp, there's got to be some increase. If so and if it's not extremely well made, you'll eventually get some stress fractures and then you can kiss your mobo and cpu goodbye.

I think the melting point of a metal is still very far off from it's vaporizing point. Try mercury, it boils at ~300 C.

Or you can look at it intuitively - mercury thermometers are rated up to 100 C at the very least, so you're safe up until there.

[Kai Robinson]

The old Soviet Submarines used to use Lead Bismuth as coolant, not mercury

[ryboto]

I think the melting point of a metal is still very far off from it's vaporizing point. Try mercury, it boils at ~300 C.

Or you can look at it intuitively - mercury thermometers are rated up to 100 C at the very least, so you're safe up until there.

I have a 300C mercury thermometer in the lab...so, you're pretty safe from vapors at the temps inside a PC.

[oohms]

Water still has a much higher heat capacity

[lookmomnobrains]

In general when commercially speaking of liquid metal, it is zirconium-based alloy. Some example compositions are listed below, in atomic percent::
• An early alloy, Vitreloy 1:
Zr: 41.2 Be: 22.5 Ti: 13.8 Cu: 12.5 Ni: 10
• A variant, Vitreloy 4, or Vit4:
Zr: 46.75 Be: 27.5 Ti: 8.25 Cu: 7.5 Ni: 10
• Vitreloey 105, or Vit105:
Zr: 52.5 Ti: 5 Cu: 17.9 Ni: 14.6 Al:10
• A more recent development (Vitreloy 106a), which forms glass under less rapid cooling:
Zr: 58.5 Cu: 15.6 Ni: 12.8 Al: 10.3 Nb: 2.8


(source: Wiki lol)

[cegras]

Water still has a much higher heat capacity

Repetition of this is getting annoying.

We know that, but if the heat it conducted away from the hot spot and then dissipated with greater efficiency through a well designed radiator, then in this case liquid metal > water cooling.

I'm guessing with something like this that a high flow, with a large radiator, would work.

[hennyo]

I think the point of using liquid metal as a coolant, is that you can have a non mechanical pump that is both small and powerful enough that doesn't use much power that it could be reasonably used in a standard heat sink.

[Serra]

In general when commercially speaking of liquid metal, it is zirconium-based alloy. Some example compositions are listed below, in atomic percent::
• An early alloy, Vitreloy 1:
Zr: 41.2 Be: 22.5 Ti: 13.8 Cu: 12.5 Ni: 10
• A variant, Vitreloy 4, or Vit4:
Zr: 46.75 Be: 27.5 Ti: 8.25 Cu: 7.5 Ni: 10
• Vitreloey 105, or Vit105:
Zr: 52.5 Ti: 5 Cu: 17.9 Ni: 14.6 Al:10
• A more recent development (Vitreloy 106a), which forms glass under less rapid cooling:
Zr: 58.5 Cu: 15.6 Ni: 12.8 Al: 10.3 Nb: 2.8


(source: Wiki lol)

Actually, those are "solid" at room temperature. They're called liquidmetal because they were made by a research group called, ta-da, "liquidmetal". Think of it like glass - it's technically a liquid. Same applies here.

Source: The wiki you quoted lol

[saaya]

this has been tried several times already, id be surprised if they can get it to work but... good luck. iirc its a highly hazzardous metal mix and didnt really work well at all...

[hollo]

Water still has a much higher heat capacity

the same amount of heat energy released from a chip would cause a higher temperature rise in a liquid metal loop than a water loop, and that'd improve the cooling power of the radiator

[twin_savage]

i've been reading up on this specific type of pump, it runs a current through the "liquid metal". thats going to heat up the working fluid and kind of work against itself. i'm sure the resistivity of the "liquid metal" is much higher than that of copper

[Serra]

i've been reading up on this specific type of pump, it runs a current through the "liquid metal". thats going to heat up the working fluid and kind of work against itself. i'm sure the resistivity of the "liquid metal" is much higher than that of copper

"this specific type of pump"? The website for "Danamics" is maybe the most vague, vaporware-looking thing we've seen in a good while, not once mentioning anything about their "technology"... and you know what kind of pump they're using in a "revolutionary" system?

Not trying to harp on you or anything, I just want to remind people that what we know of this product is:
1. Nothing
2. It will have to be expensive
3. It likely won't work very well

Even NanoCoolers, the US company that designs stuff like this, publicly said awhile ago that liquid metal cooling is only about 5% better than water cooling and 30% better than air... while costing many, many times more. That's why they haven't brought anything to the consumer market.

Now, let's let this vaporware thread die.

[cegras]

i've been reading up on this specific type of pump, it runs a current through the "liquid metal". thats going to heat up the working fluid and kind of work against itself. i'm sure the resistivity of the "liquid metal" is much higher than that of copper

Resistances of most metals are pretty much nill given a large enough cross sectional area to send them through. Also, the spec's for the pump state that the pump operates on the order of mW.

Read moar.

[hollo]

"this specific type of pump"? The website for "Danamics" is maybe the most vague, vaporware-looking thing we've seen in a good while, not once mentioning anything about their "technology"... and you know what kind of pump they're using in a "revolutionary" system?

this http://www.rexresearch.com/emships/empship.htm

[MaxxxRacer]

heat capacity does not matter, what you want is heat conductivity

[twin_savage]

@1.5watts Lorentz-Force pumps can move about 20grams of fluid @ a 1psi pressure differential a second (this is from nanoclooers data), that translates to about 3 milliliters a second of flow rate. To use this kind of pump for anything xtreme, its going to have to be upscaled, thus there will be a need for a bigger EM pump which equals more current and more heat.

what i don't understand is why no one has tried using a fluid with greater heat conductivity in a normal water loop.

[cegras]

a) I highly doubt heat dump from a current running through the liquid metal will be significant.

b) A fluid with great heat conductivity is probably more volatile. Not so the case with liquid metals, which have a huge difference between melting and vaporizing points.

[oohms]

If the metal is liquid, it does not matter that much how well it conducts. Even if it were to conduct as well as copper, thats still pretty slow. (Have you ever held a copper bar on one end and heated the other?)

Having a higher heat capacity is better, because then it takes a lot more heat energy to heat up the water per degree, which is what you want in a waterblock.

The only reason i can think of for using liquid metal is to use a magnetic pump, so they don't have to rely on moving parts

[iddqd]

A gallium and indium alloy can be liquid at room temperature. Both are non-toxic in the sense that you could drink a few droplets of this alloy without a serious health risk. If you add tin to this you can make alloys that stay liquid all the way to -20C or even a little bit lower

You can drink a glasss of mercury without any adverse effects, too, as long as you don't do it standing upright (it might punch a hole through your digestive tract somewhere). You can't digest it anyway.

Mercury Oxide absorbed through the lungs is how Mercury can enter the blood stream

Edit: Also, I saw some articles on something similar about a year ago, except for video cards. Didn't Sapphire end up implementing it in some of their high end cards too?

[Loque]

*invests in Danamics*