Slashdot Mirror
New 3D CPU Water Cooling Method
captain igor writes "According to this story on Wired News, a new company launched by researchers from Stanford has come up with a way to layer a silicon network of tiny tubes on top of a microprocessor. The system then uses a solid-state motor (no moving parts!) to pipe cold water through the silicon network. According to the article, this system can handle 1000 watts (yes, a kilowatt) per square centimeter."
Now my PDA can wee-wee in my pocket.
"A microprocessor... is a terrible thing to waste." --
GeneralEmergency
I'm still wary about pumping water into my computer. What if one of those microthin pipes were to burst? Then you'd get a microsized stream of water shorting out your not-so-micro-priced processor. I'll stick to windtunnels and heatsinks... maybe a heatpipe or two.
Forget about the cooling, tell me more about that pump! /me googles electrokinesis ..
apparantly it uses osmotic pressure to drive it, how cool is that?
A computer without Microsoft is like ice cream without ketchup.
...before someone at IBM notices their use of 'MicroChannel'?
This sig left unintentionally blank.
It's not a solid state motor. I dare say, there's no such thing. By definition, a motor turns, therefore it has moving parts. In fact, the word "motor" appears nowhere in the article, so I'm not sure where the submitter dreamt that up.
It's a solid state pump that moves an electrolyte through it using osmotic pressure.
-Todd
"The details of my life are quite inconsequential..."
Maybe they were Slashdotted?
Wait a minuite....
This is pretty cool, and I thank the poster and slashdot editors for putting up the story. I just thought that this was funny:
:)
Apple, Intel, DARPA and Cooligy did not respond to requests for comment.
Well that includes just about everyone mentioned in the article, so exactly where did the information come from? I see, I'm reading a posting about an article about another article about information gleaned from a website. Oh, well at least they told me
If we put in a miniature steam turbine we can generate power to charge laptop batteries and perhaps add a steam whistle to the sound system. Actually, since I can't get any sound out of my laptop a steam whistle would be a nice addition!
From excellent karma to terible karma with a single +5 funny post...
/. is acting weird, so someone will probably have posted a similar idea, but ...
If you could figure out a way to sew this into material, then you could have some really "cool" (literally) clothing. I'm sure people like the Army would be very interested in a suit or body armour that offered effective cooling, esp in the desert where a system with a motor could be undesireable. I know it would be sweet to get a set of motorcycle leathers with something like this built in (those Texas summers get a bit toasty).
Cooling with water reminds me of primitive automobiles and their liquid cooling . Most of the energy of an auto is wasted HEAT.
,just google(optielectronics) but most are used for high speed networking. Optical buses are achievable in my opinion right now. Logic gates are another story.
Something new needed in chip technology. Moores law is about to END.
How about opical-electronic computer chips. Lets reduce heat ! These chips already exist
A typical stove top burner is order of magnitude 1000 watts spread out over around 500 sq cm: so were talking order of magnitude less than 10 watts per sq cm.
if I take a teaspoon of water an put it on a sq cm of stove top and it boils in far less than a second. really almost instantly so its probably like less than a tenth of a second.
Thus if this thing is going to not explode, the flow rate required to avoid boiling at 1000 watts /sq cm is going to be on the order of hundreds to thousands of teaspoons per second.
If I take a tiny swizzel straw and try to suck through it I cannot suck 1000 teaspoos per second. Since my ability to suck is probably within an order of magnitude of the cavitation pressure for atmospheric pressure water a pump trying to flow this stuff through an equally small crossecttion may not be able to sustain such a flow rate. And any on-chip pump is probably going to have a simmilar crossection for its fluid intake port. (off -chip is another matter)
unless this thing is actually flowing the water based on the steam pressure itself, I'm skeptical that this can meet the claimed specs.
but I assume these people aren't fools. Perhaps the science reporter slipped a few digits.
Some drink at the fountain of knowledge. Others just gargle.
I wonder how long until we have nanotubes running all through the processor. There's a professor at my school doing research on 3-D photolithography, which would allow much more complex structures to be built out of crystalline silicon. This sounds like a good application.
This side up.
I thought the network guys had finally blocked Slashdot. It's a good thing it came back before I finished loading my gun.
Considering the consequences of a little algea or whatever in tubes so small, I'm sure they'll provide the coolant(likely non-water) and perhaps even an on board Closed coolant system.
Considering the size of 3rd party coolants shown on site's like Tweak3d.net I wouldn't be suprised at all if the setups didn't look like some of ThermalTakes larger models.
If most of the tubing is kept in the in-die, and the motor is solid state (not sure what size we're talking about) then I'd envision something that would leak about as mutch as an air cooled system. hehe.
"Oh... There it goes... my brain stopped" - Ed from Ed, Edd, and Eddy.
1000 watts (yes, 0.9765625 kibiwatts) ?
Actually, I wonder what the theoretical limit is on converting waste heat back into electricity in a laptop... would it be worth the extra weight? Even if it's NOT worth the extra weight, it might be fun to do it just because it can be done.
Off the top of my head, though, I'm not aware of any laptop-scale device for generating power from a heat source.
When the first supercomputer was built on Seymour Cray's farm by 34 guys with 1 PhD among them (a junior programmer) the key technology turned out to have been refrigeration devised by a kid from the Amana Colonies. Seymour spent his career fighting heat as he strove to get path-lengths between components smaller thereby driving up power density as the cube of his system's scale and speed. He most certainly would have approved of the focus if not the approach taken by the Stanford team.
Seastead this.
How much force would it take to burst a pipe? I would think that would be instant death for your cpu... imagine THAT for a blue screen of death -- "Sorry, your CPU has drowned. Go buy another one!"
stuff |
Put it in the main CPU, not just the GPU. That way we can get rid of the screaming banshees/cooling fans in our towers. (So you can leave your favorite p2p running overnight without the whirrrrrrrring)
Wow! Not only can I run Java on the
processor - it can brew it too!
I can't wait to get my NeverEmpty
coffee cup on ThinkGeek!
When you hear hoofbeats, think horses, not zebras
No, actually, they're not because the G5 is excessively hot, nor are they for show. They are for maximizing the efficiency of the 9 (VERY low speed) fans in moving heat out of the system with minimal airflow
People assume that because the G5s have a extremely well-engineered cooling solution that the G5 is also extremely hot. It's simply not true, it's all about noise reduction.
Random and weird software I've written.
I agree, and it looks like someone (or a couple someones) used up all their mod points nailing this thread offtopic.
Now I consider it ok to mod the first post or maybe a couple subsequent posts offtopic, but threads like this also serve to let people know that they aren't the only ones experiencing problems. This can prevent a lot of head-banging or wondering whether their ISP is choking. So please don't blanket mod everything in a thread like this offtopic. It's rude and counterproductive.
Fer chrissakes, mods, use some sense.
SB
It's old. The more humans I meet, the more I like my cats. At least they are honest.
my post erred because the reason the water boils is not the heat flux but the stored heat in the stove top coil. The transient delivery of this stored heat vastly exceeds the rate of power delivered to the stove and thus the water boils fast. but this would not be sustained.
I withdraw my original answer.
Some drink at the fountain of knowledge. Others just gargle.
They claim the potential to move 1kW through this surface, but they don't mention the conditions.
If you make it really cold on one side, and really hot on the other this could happen by itself.
Think of your cooler, it doens't leak heat much on a cold day, but on a hot day it will warm up much quicker.
Change your temperature difference, the heat flow rate will change.
On your boiling water, take steady state water evaporation vs energy input. Your 1kW Burner isn't going to be boiling thousands of teaspoons per second. You have to heat it up to the boiling point, then apply the energy to vapourize it.
Energy=Power*time = Mass * Heat of vapourization +Mass * Temperature Change * Heat Capacity
More people should take physics.
The engineering problem is getting Thot to be the microprocessor temperature, not the exiting cooling water temperature - this would give you much better efficiency, but at the possible cost of cooling power.
It's not wasting time, I'm educating myself.
Just reverse the flow of coffee in -> urine out.
And that would be....urine in -> coffee out? No thanks!
-g.
Not so, you could link the nanotubes to larger "arterial" and "venal" tubes to move the heat off-chip totally, then have another heat exchanger where the primary cooling circuit is cooled by a second water circuit, which because there's more room off-chip could be a flow of tap water in and water passed to a drain on the way out. This should prove pretty effective.
"'I pass the test,' she said. 'I will diminish, and go into the West, and remain Galadriel.'"
- JRR Tolkien.
Someone will have to doublecheck this for me, but I seem to recall distilled water as being very likely to damage any system due to simple chemistry. Last I recall from high school chem was that impurity concentrations will travel from highest to lowest, and that if the metal involved has enough of a charge to it, it'll just leach out into the water resulting in local pitting of the metal (and eventual failure of the surrounding structure).
Other then that, isn't it more of a matter of finding the right liquid vis-a-vis thermal density vs. size of liquid molecule? I bring up the latter as I recall reading that the size of the molecule vs. the diameter of the tube (hydrostatical effect?)
Pity about the fishtank cooled processor. It'd keep the tropical fish nice and toasty. In a pinch you could always short out the board and have dinner available pretty quickly I'd figure (anyone have any ideas on how fast it'd fry fish?)