Slashdot Mirror


Liquid Metal CPU Cooling

IceFoot writes "Bored with water cooling? Try a liquid metal cooler. It's a proven technology, used in nuclear reactors for decades because it carries heat away much better than a heat sink, heat pipe, or water cooling."

12 of 494 comments (clear)

  1. Um, details? by Alcimedes · · Score: 5, Interesting

    I actually went ahead and read TFA, at least the linked one.

    Although there was talk of "high tech this" and "required for the future" type crap, there really wasn't a whole lot of meat to the story.

    Maybe I missed the link that actually describes what kind of metals they're using, how the pump actually works (it's magnets! doesn't count) and what types of temperatures they're able to achieve relative to other liquid cooling methods.

    Could someone please fill in the other half of this submission, this time with less Sales pitch and more Info?

  2. Re:That's a little... extreme by BrokenHalo · · Score: 2, Interesting

    What about common or garden mercury? Liquid at room temperature. Though you really don't want it to leak...

  3. Re:That's a little... extreme by thsths · · Score: 5, Interesting

    > it is probably a gallium alloy

    Or some other weird alloy. You can buy "liquid metal" for fun at http://www.scitoys.com/scitoys/scitoys/thermo/ther mo4.html . They even claim it is nontoxic (no cadmium or mercury).

    But I still wonder what this has that water does not? :-) Cooling wise, I mean.

  4. WRONG!!! by Svartalf · · Score: 2, Interesting

    Nobody ever gives consideration to Eutetic alloys of Gallium...

    Do a google search on Galinstan or Geratherm.

    You'll find you were wrong about the mercury.

    --
    I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
  5. Re:Chernobyl at home? by LWATCDR · · Score: 2, Interesting

    Yes you are wrong.
    1 Graphite is not a liquid or a metal. It is a form of carbon. Typically uses for pencil lead.

    2 Graphite is not used for cooling it is used as a neutron moderator. It slows neutrons so that they are more likely to be captured and cause an atom to fission.

    3 Chernobyl was water cooled.

    4. The Chernobyl did not show anything about liquid metal cooling. It did show that containment buildings should not be considered optional. Graphite moderated reactors have more than a few safety issues.

    Now Russia has had some fun with liquid metal cooled reactors. The Alpha class subs used them. If you ever shut them down the froze solid and would never work again. I hear that they have solved this issue.
    The US used a liquid metal cooled reactor on the second SSN made. The USS Seawolf. It had to many problems and was replaced with a water cooled reactor. BTW this USS Seawolf was retired many years ago. There is a new USS Seawolf and it has always used a water cooled reactor.

    --
    See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
  6. Concepts of Cooling by Dixie+Flatliner · · Score: 1, Interesting

    Cooling with liquid, in flux or stasis, would be inefficient within a transistor based environment, since it is more important to quickly dissipate small amounts of heat, then slowly dissipate immense amounts. The simplest way to upgrade a water/ethynol based cooler is to replace the coolent with two parts high performance synthetic engine oil (high lubricity dyno nascar oil is best) and 1 part synthetic stasis thinning agent. While it cannot absorb as much energy as metal (1/1000th that of copper) it dissipates it many hundred times faster, and will obviously not conduct a charge. The downside is it requires a 5 psi tested sealed system, since it can leak through even airtight connections.

    Also magnetic flow operated sodium sinks tend to polarize contact points over time (the reason that dusted and secured used plutonium rods form a distinctive magnetic nickel finish) which unless I'm mistaken would invalidate gates on a microprocessor.

  7. specific heat by ehudokai · · Score: 4, Interesting
    IANAC , and it's been a while since I took chemistry, although I did ACE it. Isn't the specific heat of water 4x greater than that of metal?, or does the specific heat of metal increase in a liquid state.

    IIRC, the specific heat of water is around 4 while most metals are around 1. This means it takes 4x the amount of heat energy to raise water by 1 degree than to raise a similar metal by 1 degree.

    therefore, what is the benefit of metal cooling?

    Just because they used it to cool nuclear reactors, doesn't mean its good for computer cooling.

    --
    This is just sig!
  8. Water vs. Metal -- Try Mineral Oil by N30F3AR · · Score: 5, Interesting

    For the record... Water is chemically the best cooling agent available. The reason that liquid metal is used in nuclear reactors is that it's much easier to contain in a closed environment than water because it has a MUCH higher boiling point. For those of you that are concerned about the electroconductivity of your coolant, I suggest mineral oil. Mineral oil is cheap, cools well and above all not electroconductive. Unless, of course, your CPU is generating heats at nuclear levels, then i guess liquid sodium might be the only route.

  9. Re:That's a little... extreme by bugnuts · · Score: 2, Interesting

    Very unlikely. Besides the poisonous aspects of mercury, mercury tends to dissolve all metals in contact with it.

    If you force mercury over a copper block, that block will be dissolved in a few months.

    Got some mercury? Drop a dime into it and watch what happens after a week or so.

  10. Re:That's a little... extreme by fshalor · · Score: 2, Interesting

    At the very least we get to show them another example of why such cooling is necessary. 28 comments, and the site's php rendering is already taxed.

    It's been 3 minutes, and I still don't have the pictures...

    "The ever increasing demands put on cooling solutions for semiconductor devices have never been greater than today" ... now that we're being slashdotted!.

    Hehe...

    Since there's almost no actual substance in the ad, I'll hold off commenting on its feasability. Unless the've matched some melting point to the cpu (and that would give them a very small window of max effectiveness), the'd be better off using water or ethylene glycol.

    Phase changes are evil and tougher to deal with.

    --
    -=fshalor ::this post not spellchecked. move along::
  11. Re:That's a little... extreme by Sique · · Score: 2, Interesting

    Another little clarification: German: Wismut. Discovered about 1540 in the Schneeberg mining region. Translated into latin by Georgius Agricola to Bismutum (there is no W in the latin alphabet). Therefore the chemical symbol Bi.

    Newest english-asskissing craze: Writing "Bismut" in german, because it looks so english and thus so scientific.

    --
    .sig: Sique *sigh*
  12. Re:Not very toxic. by joto · · Score: 2, Interesting
    I don't know why you absolutely want your kids to take apart your computer and eat the contents. I'm sure you have other toxic things in your house too...

    (Ever wondered what you use in your dishwasher? Or what some of the stuff in your garage does?)

    "Not very toxic at all" is quite comfortable for me. Of course, I also want to know how it reacts in case of fire, if it destroys my floor if it ever leaks out, if it's environmentally friendly, etc.. But according to the MSDS I shouldn't need to worry about any of those...