Cooling breakthrough for computers, car electronics

[Airbozo]

Thoughts?

Opinions?

Forgot to add link:
http://news.cnet.com/8301-17912_3-10359738-72.html?tag=mncol

[Kayin]

I have copyleft on a few breakthroughs. What are we discussing, again? Is there a pic that isn't loading for me here at work?

[slaveofconvention]

Errrr - I think I ate too much at dinner but I don't think that's pertinent - what are we talking about?

[Airbozo]

Woops, forgot to add the link....

http://news.cnet.com/8301-17912_3-10359738-72.html?tag=mncol

[Zephik]

This thread makes SO much more sense now. lol

Pretty interesting read, although I don't really have anything to add about it.

[slaveofconvention]

Well it's going to have to be a liquid with is 1) Non conductive and 2) boils well below 100c - I personally wouldnt be too keen on hardware which needs to run that hot before it'll cool effectively, or which will fry the rest of the components if it get a crack

Anything to improve cooling has to be a good thing though - there's only so far they can go with bigger and bigger heatsinks...

[TheGreatSatan]

So they discovered water cooling? No, wait that already exists

[slaveofconvention]

More like steam cooling :p And I think the big difference is they're running liquid THROUGH the chips, altho I may well be wrong on that - it's late and I only speed-read the article

[crenn]

Did anyone actually read the article?


Garimella and Harirchia have now determined that "allowing a liquid to boil in cooling systems dramatically increases how much heat can be removed, compared to simply heating a liquid to below its boiling point," according to their report.

"Boiling occurs differently in tiny channels than it does in ordinary size tubing used in conventional cooling systems," Garimella said in a statement.

[d_stilgar]

I would feel more comfortable letting something boil on a peltier vs right on my chip or even on some sort of waterblock or sink.

[blueonblack]

It makes a lot of sense. When a liquid changes to a gas it absorbs a HUGE amount of heat just in changing state. That's how refrigeration works. I forget the numbers exactly, but it takes a LOT more heat to change water that's 100 degrees Celsius to steam that's 100 degrees Celsius than it does to raise the same amount of water from 99 degrees to 100. Inducing the phase transition takes a lot more heat, so if you can force the phase change (by using a compound that boils easily like R-12 or any other refrigerant) you take the heat with it.

The basic science is sound and used all over the world every day. It sounds like they may have found a way to cause the phase change more easily with these microchannels. They could also be putting the channels in a partial vacuum, that will cause any liquid to evaporate more easily. The key to the whole thing (to me at least) sounds like the miniaturization of it. It's interesting and with today's nanotech and manufacturing processes, I see no reason why it won't work.

(Long and pedantic, I know, but you asked. :D)

[Drum Thumper]

BoB hit the proverbial nail on the head me thinks. However, I am a bit skeptical until I see more than just a blurb on cnet. Remember last year when the 'next big thing' was perforated chips? When was the last time any of us heard any more on that?

[Kayin]

It has to do with the reduction of surface tension, as I remember. I could be off, but I'm looking into microchannels at the moment myself...

[mDust]

They could also be putting the channels in a partial vacuum, that will cause any liquid to evaporate more easily.Liquid can't exist in a vacuum; it would just expand to fill the volume that was removed to create the vacuum.
There would need to be a ton of 'slack space' (some light, compressible gas) and very little actual liquid to allow for extreme expansion. If not, then they are just building a bomb that will kill everything in the computer when the user fires up their favorite game or hits the render button.:whistler:

This is really just half of an on-chip phase-change cooling system and isn't going to make current cooling systems obsolete...even the steam can only absorb so much heat, it still needs to be moved somewhere else. The expansion of the liquid flowing over the IC will absorb heat from the die and move it to another part of the chip, until that part of the chip cannot absorb anymore: <enter stage right> heat sink to the rescue!

Another thought: (on the positive side) the on-chip liquid might increase thermal conductivity for your crazy phase change system resulting in ultra-mega-uber low temps...at least until the coolant freezes.

[blueonblack]

Liquid can't exist in a vacuum;

That would be why I used the term partial. Meaning less than 1 atmosphere of pressure. The closer to vacuum, the easier it boils (as evidenced by the fact that liquids expand to fill the space vacated).

[mDust]

Sorry for restating the obvious there, but I went on to say there needs to be room for gas expansion. That was my point.