Forgot your password?
typodupeerror

Micro-Pump is Cool Idea for Future Computer Chips 96

Posted by ScuttleMonkey
from the bad-pun-already-in-the-headline dept.
core plexus writes to tell us that Engineers at Purdue University have designed a tiny 'micro-pump' cooling device that can be used to circulate coolant through the channels etched on an individual chip. From the article: "The prototype chip contains numerous water-filled micro-channels, grooves about 100 microns wide, or about the width of a human hair. The channels are covered with a series of hundreds of electrodes, electronic devices that receive varying voltage pulses in such a way that a traveling electric field is created in each channel. The traveling field creates ions, or electrically charged atoms and molecules, which are dragged along by the moving field."
This discussion has been archived. No new comments can be posted.

Micro-Pump is Cool Idea for Future Computer Chips

Comments Filter:
  • by Anonymous Coward on Wednesday April 26, 2006 @07:33PM (#15208698)
    The smallest particle in the coolant would block it.
    • The coolant is supposed to be built in, doh.

      The way I see this working is having the micropump embedded INSIDE the chip so the surface can be attached to a heatsink and dissipate heat more efficiently.
    • Maybe it's just me, but 100 microns seems like a fairly large particle.
      • you still do get blockage with 100microns micro channel. You have to use fairly distilled water (unless they are using different coolant) so there are not much ions floating around, and not too distilled so the water is not corrosive which may eat away the channel.
        • Why do we have to pump water through the microchip ?

          Yeah sure, it gives us probably almost the most efficient way to cool things down since water has excellent figures on leading and consuming heat. But it's rather complicated aswell, one problem is stuff can get stuck in the microchannels, the other one is that it has to be really tight and it get's very fragile.

          Why won't amd and intel and all the other `hottie makers` just include heatpipes into the design of the cpu ? it would help a lot
          • yay, input boxes use fixed width font, slashdot doesn't ... anyway i' guess you'll understand what i meant (and maybe how wrong i could be on the idea...)

            smaller version here then
            H_H_H
            xxxxx
            xcxcx
            xcccx
            gotta learn to use preview ...
    • True, in fact it's a kind of catch-22. Dirty water is more conductive than clean water. Using an EHD drive to pump the water is ingenious, but if the water's too clean it won't efficiently pump through the channels, if it's even slightly dirty, there will be mineral build up which will block the channels and fry your chip. Perhaps some other electrolytic fluid will work better, but I can't think of any with better thermal conductivity that won't boil at room temperatures. Good idea though. Using a refrigera
    • Perhaps they could use ferrofluid [wikipedia.org] instead of water. The same electromagnetic field would cause far more flow and with a lot more force. Particulates in ferrofluid are on a nano scale and are kept from agglomerating by use of a surfactant, which should help prevent clogging in the channels. The additional force of the fluid would not only increase flow (and thus cooling) dramatically, it should also help keep the channels clear as the force pulling the fluid through the channels would probably be sufficient
  • by jlseagull (106472) on Wednesday April 26, 2006 @07:34PM (#15208705) Homepage
    Chips fabbed in 3D have numerous advantages - short trace lengths, higher density, etc. However, the problem with all of them is getting the heat out with today's convective cooling technologies. This technology will allow multiple cores in 3D to operate without overheating, and that's a good thing as the number of cores in personal computers and servers continues to increase.
    • Eh? But wouldn't it be more difficult to deal with the silicon in three dimensions? I'm not an engineer, but I imagine that the wafers are much easiers to use, (especially with my understanding of how they create transistors using semi-conductors).
      • Yes, it's more difficult, but you (potentially) get large gains in speed due to shorter wire lengths. I'm not up on all of the technology, but some of it centers around fabbing layers separately and then bonding them together (similar to PCBs), and others are all-at-once methods.

        This cooling method is also valuable for MEMS systems.
      • by ShakaUVM (157947) on Wednesday April 26, 2006 @08:18PM (#15208931) Homepage Journal
        The trouble is that routing on chips isn't done by hand anymore. An algorithm crunches away on a design and spits out what it found to be the most optimal layout for the given parameters. So if you have to start pushing things around by hand in order to make room for cooling channels, it could break your design.

        I'd say the solution to it would be to lay out the cooling channels just like other routes in the die, and set the parameters up somehow in the routes would be relatively well distributed for maximum heat absorption.
    • by thpr (786837) on Wednesday April 26, 2006 @08:18PM (#15208926)
      Yes, 3D is a neat application, but cooling is not the only challenge in 3D semiconductor electronics. Another perspective on 3D is available in Business Week's More life for Moore's Law [businessweek.com] article.

      For example, one of the assumptions that exists on a semiconductor wafer before it is printed is that it is effectively flat (a typical peak to valley range on a modern wafer within the expected field of a chip is on the order of 175 to 200 nm)

      Polishing to that accuracy once structures have been placed on a semiconductor wafer is difficult. Getting a consistent layer of material when you are polishing an uneven surface (uneven due to vias [connections] to the other layers of silicon present) is downright challenging. Another problem with printing transistors on anything but a pure wafer is the issue of reflection. Thin layers of materials on a semiconductor are semi-transparent and not perfectly vertical. Those angled and curved structures produce reflections. Those reflections can cause problems in printing later layers (because of constructive and destructive interference of the light used to expose the photoresist). Those reflections mean that modeling the exposore process of a 3D semiconductor is a VERY challenging task.

      Such items are not of concern today, because the later structures placed on the wafer are generally metal lines or capacitors for DRAMs or lenses for image sensors, etc. These are all large and some level of imprecision is acceptable. While variation can cause differnet RC characteristics in metal lines, the timing models in the library or other models can account for this variation. In fact, Matrix Semiconductor has been producing 3D DRAM since about 2004, which shows that heat isn't necessarily the problem, and DRAMs (and memory in general) are a reasonable application for 3D technologies (likely because the capacitors are generally large in relative terms).

      Transistors, however, are much more sensitive to variation, and the variation in later polishing used today is too rough for the effective printing of transistors. While I don't doubt that there are situations where the density will be valuable, I think 3D processors and custom chips (in consumer electronics, et al.) are as much an economic issue as a cooling/technical one. (in other words, with my understanding of current roadmaps, you will decrease semiconductor yield to such a degree that 3D may not be economically viable, even if the cooling problem is solved.)

    • This sounds like a good idea, but can someone with some knowledge in thermal dynamics shed some light into how the effectiveness of these micro pumps? I mean, come on, can those thin layers of liquid weighing less than a few grams actually cool a 20W device? If so, how fast must the pump be running?
  • It seems that using a field to push ions around would create "inertia" or damping on the signal that is pushing it.
     
    Wouldn't this impact performance or timing issues within the chip?
     
    • These ions don't have anything to do with the electrical signals for computation. The ions are the coolant, so you're just simply inducing an electric field to push them along.

      But you're right, it would cause issues if they were actually using them for signals, but it's simply so they can move the coolant around.
    • At first I thought they were using the computational signals to push, but after going through the article more carefully I realized that.
       
      Having said that however, it seems that doing this would seriously complicate chip design as most of the actual work in chip design is dealing with electromagnetic concerns.
       
      seems like this would be juju of the worst kind.
    • Are you referring to the electrical field generated by the electrodes in the channels?

      The ions themselves dont generate an E-field large enough to affect anything above the molecular scale. And I would assume that the engineers have designed the channel so that there is no E-field present outside of the channel, which is entirely possible.

      Essentially what they're doing on a large scale is moving electric dipoles using several differences in potential. The really cool thing about taking advantage of water's
  • Wouldn't it be easier to do the cooling on the chip and use something that conducts heat very good on the chip? I mean, I would rather have lanes of some conductive, non moving material instead of some liquid running through my CPU, if you don't mind.
    • Yup, like copper for example. I guess that would be too simple - you can't get venture capital for adding copper bars to a chip...
    • by Ungrounded Lightning (62228) on Wednesday April 26, 2006 @07:57PM (#15208828) Journal
      Wouldn't it be easier to do the cooling on the chip and use something that conducts heat very good on the chip?

      A conductor would have to be thick, which would take up a lot of space.

      Moving s liquid with high heat capacity (such as water, which has ENORMOUS heat capacity) means you can move the heat out by transporting the liquid, rather than by conducting the heat THROUGH it. The liquid can then drop off the heat at the heat sink in a leisurely fashion on its way through. Heat only has to move by conduction across distances measured in molecular diameters rather than inches.
      • I have always thought that mercury would be the ideal liquid for this sort of closed loop system - it has a thermal conductivity / capacity of somewhere around 14x that of water (and thus would be 14x more efficient.) Granted there is that whole 'highly conductive and highly corrosive' aspect of mercury, plus the nasty side effect of being toxic - but if the system was properly engineered to take those into account, why not go for broke?
        • Well here in the US the EPA would kick the manufactures a$$

          one thing most people don't realize is that on top of being toxic and crosive and jsut plain out bad for everything mercury is also extreamly hard to clean up.

          the only thing that can be used to correctly clean it is pure sulfer - everything else makes a partial bond and doesn't clean it up, and afterwards you end up with something that is just as bad if not worse than what you started with....

          while i agree it would be neat and more than likly would
          • There's still mercury in my Thermostat in my apartment. I'm sure there's mercury being used in lots of other places too. There's no reason why we couldn't put it in computer chips. There's already a lot of toxic materials in a computer. I don't think 1 mL of mercury is going to make that much of a difference.
            • Except that no company would do it because of the environmental problems. That thermometer you own that still uses mercury has obviously been in your possession for quite some time, I know where I live I haven't seen mercury thermometers availible for sale in probably twelve years.

              Case in point, I work for a comapny that, among other things, disposes of mercury. Right now we have probably just under a thousand cubic feet of gas/water meter componenets in our building. Each one has only the tiniest drop o
      • Moving s liquid with high heat capacity (such as water, which has ENORMOUS heat capacity) means you can move the heat out by transporting the liquid
        Unfortunately, it's not easy to move liquid through a pipe which is, if you'll excuse the technical jargon, pretty damn narrow [urbandictionary.com].
    • Actually, silicon is a very good thermal conductor, although not as good as copper (150 W/mk, vs. 400W/mk for copper.) However, as a heat transfer mechanism, both of these pale in comparison to moving water. A heat pipe (http://en.wikipedia.org/wiki/Heat_Pipe [wikipedia.org]), which transports heat by evaporating water at the hot end and condensing it at the cool end, can have an effective thermal conductivity many times better than copper. Moving vapor or liquid is also a lot more effective at moving heat long distance
  • by syousef (465911) on Wednesday April 26, 2006 @07:41PM (#15208736) Journal
    ...and what's the term for a blocked CPU? Constipated???
  • ok, bad pun is outta the way :)

    Having wet, electically charged canals in the middle of a CPU sounds weird upon reflection. And aren't they huge compared to the circuits, where's the room?

    hmm

    Article says challenges include sealing it to prevent leaks...DUH.

    Nor did they say the chip included OTHER circuits yet...THE WHOLE REASON for this cooling to exist.

    Interesting idea, nothing more and won't be for some time.

  • by Xiph (723935) on Wednesday April 26, 2006 @07:45PM (#15208756)
    So for those of you who did the same.
    This system works in multiple ways, it has an ionisation pulse that travels along the water lines
    The pulse ionizes the water the ionized water is dragged by the pulse
    the pulse alters the shape of a small membrane, boosting the pump.
    as for the efficiency
    We have shown that the power input required is in the microwatts, but you can get milliwatts of cooling
    that being said, it's still work in progress, and they (according to the article) haven't solved leakage problems yet.
    • We have shown that the power input required is in the microwatts, but you can get milliwatts of cooling

      I think it's important to point out that this is a meaningless statement. Cooling is not about creating heat flow, that happens naturally, cooling is about lowering resistance to heat transfer.

      Even in the case where you assume the milliwatts of cooling is an improvement over what load could normally be sufficiently cooled, it's not meaningful without knowing the original power dissipation. An impro
  • From plants (Score:4, Informative)

    by piotru (124109) on Wednesday April 26, 2006 @07:47PM (#15208769) Homepage Journal
    Simplifying, the plants are thought to use similar idea to transport viscous liquid within their vascular system - phloem. Beautiful!
    Link: http://www.cas.muohio.edu/~meicenrd/ANATOMY/Ch9_Tr ansport/phloem.html>
  • by kimvette (919543) on Wednesday April 26, 2006 @08:16PM (#15208918) Homepage Journal
    Why not implement oh, I don't know, say, a Peltier Junction directly into the heat spreader? Since you KNOW there is going to be a heat sink (no warranty if no heatsink is used) then any overheating concerns from running the junction without a heat sink are moot.

    KISS (Keep It Simple, Stupid) -- they're over-complicating the solution. Fluid directly in the chip might be a good idea, but let conduction and natural convection handle the heat transfer to the heat spreader. Don't over-complicate this thing with a pump that can break the second a nanometer particle gets into the system.
    • The problem with peltier coolers, as far as I know, is that any one with a decent wattage rating will require its own power supply because of the power that it alone draws; you can't just plug it into a spare molex. It consumes more power than it moves.

      http://www.heatsink-guide.com/peltier.htm [heatsink-guide.com] has more information.
    • Peltier coolers are so inefficient, they are terrible at cooling anything. You note no one uses one on any heatsinks right now, don't you?

      Peltier coolers are incredibly inefficient, and the power they waste turns directly into heat (of course), so you end up heating up the thing you wanted to cool down.
  • I read "tiny micro pump" and stopped, thinking someone has spammed Slashdot with penis-enlargement scams ...

    "I'm tellin' ya baby, it's not mine!"

  • by imgod2u (812837) on Wednesday April 26, 2006 @08:23PM (#15208953) Homepage
    I'm no expert in ASIC design but that doesn't sound like the best thing to have in your extremely sensitive high-speed signals. I assume this field will remain constant and won't provide noise for the chip (or at least I hope) but it will introduce an electrical bias that needs to be planned and compensated for during the chip's layout.
  • Cooligy System (Score:1, Interesting)

    by Anonymous Coward
    Here's a system that is doing something similiar, but on a larger scale. http://cooligy.com/micro_channel_cooling.html [cooligy.com]
  • Goodbye to Overheating problems, the next problems will be bleeding Chips!!.
  • by Anonymous Coward on Wednesday April 26, 2006 @09:05PM (#15209158)
    Not only is this literally "cool," but many geeks are used to operating a micro-pump..........

    [crickets...]
  • I expect and demand every cool discovery to be prefixed with "nano"
  • Why bother building the pump into the chip at all? What's wrong with a "mini" external pump, pushing the coolant through simple channels with the traditional pressure gradient? The MEMs on the chip take space that computing HW could occupy. Maybe some MEMs valves and sensors to optimize flows through varying areas as they heat differently during different processing tasks. But the pump can be outboard, where logic HW would be less useful because of signal latency.
  • ...could screw up your chip forever. Its not like you can roto-rooter the thing. The puritity and perfection of the fluid to flow through channels like this would itself be prohibitive for now.
  • Did any one else read the headline as "Micro-Pimp"? I was really curious to see what midget pimps were going to do for the furture of computer chips.
  • by Rob Simpson (533360) on Thursday April 27, 2006 @12:37AM (#15210050)
    "Innovative cooling systems will be needed for future computer chips that will generate more heat than current technology"

    Except for supercomputers, servers, and hard-core gamers with air conditioning, who is going to want chips that will generate substantially more heat than current chips? If CPUs alone start using hundreds of watts of power, people are going to take notice, and even the most naive shopper will start taking this into account. Already, Intel has realized [wikipedia.org] that their ridiculous space heaters are a dead end.

  • How do you put a sphy.... uhmmm sphygnomo.. mona.. uhmm (google, copy, paste) Sphygmomanometer on a CPU anyways?
  • by Aimak (652182)
    Although not many details are given, it seems to me this people just adapted an existing analytical technology called Capillary Electrophoresis [ceandcec.com]. The piezo pump is a clever addition to the system to improve the micro-liter per second flows typically obtained in CE technology.

    I wonder where and how they want to hang the liquid reservoir with the cooling solution. The processor may have to come then with an attached infusion bag like those you get at hospitals.
  • Now if our CPUs could drown, somewhere near the "printer on fire" error, there needs to be a "CPU is downing" error, right?
  • "The prototype chip contains numerous water-filled micro-channels, grooves about 100 microns wide, or about the width of a human hair."

    Being that I live in the north, I am a bit skeptical about water being inside of the chip. They didn't mention anything about how it can handle cold temperatures.

    If you were to transport an item with one of these cooling mechanisms in the winter time (perhaps to a repair location) is there the potential that the water in the channels could freeze? Would it be capable of wi
  • More to the point is this not like the catapiller drive used by the Red October in Hunt for Red October?
  • Very impressive - but pointless I suspect. Channels that small won't hold much fluid, which means very little cooling capacity.

It is not for me to attempt to fathom the inscrutable workings of Providence. -- The Earl of Birkenhead

Working...