Slashdot Mirror
Wriggling Heat Sinks
YourHero writes "Purdue researchers have come up with a new way to cool chips, in about 2 years. Just build a bunch of little piezoelectric fans (the waving kind, not the spinning kind). Since they don't spin, no bearings, less self-generated heat. Since they don't have magnets, no electromagnetic noise problems. And, of course, super-efficient. A press release and abstract for your reading pleasure. Formal presentation at THERMES 2002 Jan 15th."
The same reason cars use gasoline instead of electric or battery power.
sounds like you'd just be waving the heat bye bye
"The concentrated circuits in a semiconductor computer chip can generate more heat per square centimeter of chip area than an area of equal size on the sun's surface."
Is this true? If so I have so much more respect for my heatsink....
Green-voting, republican-registered, socialist-libertarian.
You can get them now. ARM CPUs, for example. If you're willing to stick at a 233 MHz CPU.
The innovative fans will not replace conventional fans. Instead, they will be used to enhance the cooling now provided by conventional fans and passive design features, such as heat-dissipating fins.
Oops. Looks like the editor didn't read the article....
Does this surprise anyone?
Linux - Because Mommy taught me to Share.
Um... because gasolene has a higher energy to density ratio than steam or conventional batteries?
Or are you trying to say that making cooler chips is too difficult?
GPL Deconstructed
It would be nice if they made it self-cleaning too. Those things accumulate a lot of dust. I use my dad's 5.0HP shop vac to clean the ones I have. It makes a loud ZZZZZZZZ and sucks the dust right out of there! Sounds like that would break one of these. Neat idea though. :D
Doesn't anybody think it's cool to be noisy anymore? I mean, say what you will about being distracting and all that, but I'd love to impress my friends with a PC that sounds like a lawnmower. it's POWER! it's TOUGH! it's AMERICAN!
sometimes worrying about things like noise is too girly, even for me.
spacefem.com
However, they're not cheap. Pricing starts at $149. Additionally there is a Piezoelectric Resonant Blade Element. Interesting stuff. Hopefully mass production of piezoelectric fans will lower their price to the average customer range.
What do you think of MusicCity now?
Half seriously, though, you might think of superconducting chips to eliminate the heating due to the resistance in aluminium/copper wires. But AFAIK you can't build logic circuits entirely out of superconductors. The siliconductors (sic :-) we now use, require current to pass
through potential differences (energy gaps in the crystal structure). Power dissipated equals
current times potential difference, period. And there are lower limits for the voltage imposed
by the semiconductor used.
Until we get something entirely different, I'm quite happy to put my geekineering effort into the design of better cooling. I'm sure it can be almost as fun as inventing new kinds of logic chips.
Escher was the first MC and Giger invented the HR department.
This is just an excuse for designers to make CPU's less efficent and more power hungry.
ImagineWashington Post: Dec 13, 2018. Details are now emerging about the accident that irradiated much of Germany on Tuesday. Nothing is as yet confirmed, however, initial reports indicate that a heatsink was somehow removed from an AMD processor (PR rating 10,000,000). A bizzare terrorist group with the initials THG may have been involved. Containment was lost, and critical mass was reached almost immediately. AMD representatives have issued a statement in the wake of the carnage: "Obviously, they were using an improperly designed motherboard."
Here's a picture of an old-style piezo fan
You used to be able to buy piezo fans for the old Mac Classic (read the list near the bottom of the page).
IOW, piezo fans have been around since the mid-to-late 80's. Now, yes, I'll admit that they weren't very efficient (as in, they didn't move a lot of air)... but the concept has been there for eons.
--
Don't like it? Respond with words, not karma.
I don't mind the noise, but dissipating heat in general would be a good thing.
The thing they need to do is make chips that run cooler. And yeah, Crusoe's do run cooler but they don't perform optimally in a task-switching environment.
Cooling the CPU is fine, but the heat has to go somewhere and a better solution is to go back to the drawing board and figure out how to reduce the heat output in the first place. PLEASE.
----
I don't have a solution, but I certainly admire the problem.
What are the chances of the conventional ball bearing fans, in the very computers that are doing all the mathematical modeling, will go on strike??
Self-preservation is quite a motivator.
I have had one of these fans cooling my sterio for years. I got mine as a sample while working for a crystal manufacurer in 1984. It makes VERY little noise but does not even begin to move enough air to cool a modern CPU. These new ones would have to be 10 times more powerful.
I didn't see any specs as to the rate of air flow these things can produce. Assuming they are optimized given the 'provided' mathematical models. Any chance these models are as easy to understand as the instructions they put on coke cans?
The article stated that these fans could have blades up to an inch long, anybody have any opinions on whether this could replace the large fans in cars that are used for air flow over the engine and radiator? This would make working on your car while it is still running a little bit safer. But of course the saying "Make something idiot proof and somebody will make a better idiot."
And since the topic of energy consumption was brought up, how about using these instead of ceiling fans in our homes. Being that I have never seen one of these in action I bet you could make them look aestically attractive at least to us nerds. Sort of like having a huge rack of all black stereo equipment.
I don't know what kind of chips these researchers are using, but the kind I use build up heat a lot faster, and thus need to be cooled constantly, not just every two years.
lame jokes brought to you by:
"when life gets complicated, I like to take a nap in a tree and wait for dinner" - Hobbes.
They're used in inkjet printers - they're in ink some cartridge when an electric field is applied to them and they change shape, forcing the ink out of the I also hear the they used them in the ipod for some sort of playlist control mechanism.
I'd much rather trust my components to one large, well made fan with some intelligent ducting inside the case to deliver the air flow where its needed. I think this is one area where some of the big system manufacturers still have a big advantage over a typical 'roll your own' case. Small cooling devices are just too fragile and unreliable, and multiple points of failure are unacceptable, especially in server applications IMHO.
Endless arguments over trivial contradictions in books written by ignorant savages to explain thunder in the dark.
Ok, I just got this be-yoo-t-ful image in my mind:
Imagine the piezoelectric fan on a larger scale, not just waving a metal+ceramic blade (single flexible surface area), but creating an undulating sheet about the size of a letter/a4 size piece of paper using stripes of piezoelectric flexion areas that create a wave every 2-3cm. Now combine this with the latest in flexible printed circuitry top and bottom (or 2 layers top and bottom, for the really adventurous). I'd imagine you might also need periodic non-flexible stripes (ends?) for components and connects that can't be made flexible. Then add a lower-power processor and put it into an enclosure only slightly larger than the wave height, such as, say, a laptop computer housing. What do you have?
You'd get a motherboard that cools itself by cilia-like swimming/undulation movement that pushes air (against the enclosure) across its surface silently.
You'd get quieter rackmount systems, with 1U or "blade" servers that self-vent. ("Ah, yah need tah balance yer server there, buddy, the blades are outta sync.")
You get a laptop that you might enjoy putting in your lap. (On second thought, I'm not sure I want to sit next to someone on a plane with a two-stroke laptop...)
just my $0.02
-Jon Espenschied
I think not...(*poof*)
I see no reason why the same technology could not be applied to modern CPUs and computers. It would be energy efficient to say the least..
On a side note, if you want an interesting geometry problem, try to mathematically design a pyramid out of cardboard for a specific height and base.
I've been looking into this a lot recently, and there's some pretty (ahem) cool developments on the cpu front recently, with x86 architectures.
Some people point to the VIA C3-800, but if you have real computing needs, steer clear. It runs comparable to a Celeron 400, which is almost, but not quite adequate for general computing. Instead, check out the old reliable suppliers. The shift to .13u means a lot. Frequencies are so high and chips are so powerful that underclocking has become a real option. A good general target for fanless operation is about 12 watts. You can go higher with good case airflow, or lower if you're dealing with troublesome ambient temperatures.
Right now, you can take the Intel Tualatin pIII 1.13GHz (28W), cut the bus speed to around 100MHz, cut the voltage down to about 1.1v and be right in the target range. Of course you won't know exactly w/o experimentation on your cpu, but it *should* be doable. If you're worried about losing efficiency to bus speed, remember that you can compensate by running it on one of the PIII DDR chipsets that are now available (upping effective bus speeds to 200MHz) or waiting until February, when Intel says they'll release a similar part themselves. Additionally, the 512k (vs 256k) cache on the pIII-s will offset lower bus speeds. Just check out the specs of the PIII-M LV models at developer.intel.com and ask how they got to those low wattage numbers with the same core. Since the last fanless G4 was 400MHz and claimed (in its wildest fantasies) to be a supercomputer twice as fast as a pIII, a fanless 800MHz pIII is not insignificant.
Even better, surprise, is AMD. The current mobile palomino runs at 1.1GHz, 1.1v, 25w. This is clearly just an underclock of the current 1.75v desktop XPs. But what it tells you is that the AMD architecture is very open to undervoltage at lower clock speeds.
Now if you consider AMD's forthcoming die shrink, things really look good. Zdnet.de reported (unsourced) that the Athlon 1.73GHz processor would drop from about 75W to 45W after the changeover. Depending on how far you could drop the voltage, you could be looking at a 1-1.2GHz part running at about 10W! Fanless! Now imagine (a beo..no) 2 of these in a well ventilated case, with an MPX board -- 2GHz of dead silent AMD power! Wooo!
Alright, I'm calmed down. Back to your original point. It's really a shame about the alternative architectures. Every time I think of venturing into the embedded market, I get brushed off by the 2x price, 1/2x power rule. But since the ARM and PPC don't seem to be generating any economies of scale, at least mainstream processors are progressing fast enough to make cool, cheap and fast a real alternative.
Kill, Tux, kill!
Fans, fans fans. Might as well use a Tesla Turbine to move ungodly volumes of air with very little noise. No fan blades, no resonance with the heat sink blades to make loud whine or buzz. Just the hiss of moving air over the heat sink blades.
However, solid state heat transfer has been around for ages. I would love to find an advert for a 12-volt refridgerator for camping that I saw back in the 1970's. It used a pezo film between two heat sinks, one on the inside and one on the outside. Apply the voltage, and heat was moved actively into the outside heat sink, enough to chill your beer and keep the fish fresh on the trip home.
Put such a film between the chip and a heat sink. Gosh wow, a cool CPU.
Bob-
The Ludwig von Mises Institute. The reasoning individuals economics
PowerPCs are cooler running chips.
Or did you mean cooler running IA-32 chips that can run Microsoft Windows?
$x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
$x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
Multiple points of failure in a system without cycles, such as A->B->C->D->E are bad. If you're going from A->E, B, C, or D could fail and ruin the whole thing. However, if you had A->B & A->C & A->D, etc... then more points of failure are a Good Thing. Now look at another case. If you have a device A that does 100% of the work, and it fails, than you have 100% failure. If you have devices A, B, C, and D, each doing 25% of the work, and one fails, you only have 25% failure. More points of failure is good in this case. With a ton of these tiny fans, if some fail, the system continues to work without damage. Think, write, read, think, rewrite, think, preview, post. Try it.
Why bother.
They say you can use a surface of these fans each of which is only a hair's-width long to cool chips. Do they have any concept of the idea of dust?? Every six months or so I have to take one of those cans to my fans to remove the huge air blocking clumps that seem to clog up the entire fan. Are we gonna have to start purchasing a special cleaner that we have to dip these into every couple of weeks? My monitors have a pretty much permanent grey film that doesn't wash away on them from a year or two of the Los Angeles smog. I'd hate to see what particles that small will do to the effectiveness of these fans.
I'm out of my mind right now, but feel free to leave a message.....
The article refers to low *electromagnetic* noise. It also points out that these piezo fans will require conventional fans to dissipate heat further away. Their action will lead to increased heat transfer at the chip itself.
So, if I understand the article correctly, they won't in themselves lead to any reduction in audible noise.
Oh well!
A little planning goes a long way...
Why don't you just propose a design for such a better chip? I thought so.
Here's my design: A processor that isn't catering to the 2% of computer users that need all the power they can get, but is sold across the board to the other 98% as well.
aren't claiming its anything new.
When I did brain surgery on my ancient Mac to slap in an extra meg and a half and a SCSI interface I had to install the 'flapper' fan or whatch my case do a Dali "soft watch."
Gluing a piezo fan onto the chip is not very smart anyway. And it does generate some 'flexing' heat where there is the least air motion. And it makes noise like a butterfly on speed.
You don't get something for nothing. Moving air other than by convection causes turbulence which causes vibrations. Vibration IS noise. Which is more irritating, a flapping buzz or a whirling whoosh? Its a matter of taste.
MSBPodcast.com The opinions expressed here are my own. If you don't like 'em... Think up your own stuff.
Hey baby, why don't you follow me downstairs to the computer room. I've got a wriggling heat sink that you need to check out.
------
Let me give you the lowdown
the answer there for is case design.
look at Apple's case design. it is elegant, cables are out of the way, and the heat is disepated much more than in an ATX case.
I am the Alpha and the Omega-3
I really agree with your assessments.
Anyway, if the case is relatively open inside, all you need is a power supply with really decent venting (like the Enermax 300W unit with its double fans I'm using right now), a decent CPU cooler and a expansion slot fan to vent the hot air out of the lower portion of the system case. I've never had any heat-related failures.
It doesn't change the fact that cooler fast chips can be packed more tightly (fit in smaller spaces and achieve higher densities in servers, due to heat and power constraints)
It's almost always better to be more efficient; basic laws of physics and all, when you have constrained resources like we do... you get more done and more bang for the buck.
GPL Deconstructed
Yes, that's what i'm trying to say. If making cooler chips was the cost-effective way to make chips, that's what they would be doing.
The chip manufacturers are only trying to make the most profit possible, not to make cooler chips, unless the market demands it to the extent that it would be more profitable.