Has anybody thought about liquid Helium cooling

[Gerti]

I just learned in Physics thst liquid helium reaches a temperature of 3K (kelvin) that means -270C.It's just above absolute 0 where the silicon works best because it's atoms can't vibrate.
I wondered if somebody has used it in cooling a CPU instead of LN2.
A gig proc could reach 3 GIGs at 3V and it probably wouldn't last more than a minute.
LOL

 

[William]

Gerti (May 17, 2001 09:14 a.m.):
I just learned in Physics thst liquid helium reaches a temperature of 3K (kelvin) that means -270C.It's just above absolute 0 where the silicon works best because it's atoms can't vibrate.
I wondered if somebody has used it in cooling a CPU instead of LN2.
A gig proc could reach 3 GIGs at 3V and it probably wouldn't last more than a minute.
LOL

Well there is one problem, Liquid Helium is extremely difficult to work with, and probably kinda expensive. Also, getting things that cold can cause stuff to crack. Liquid Nitrogen is cold enough, i don't really thing Helium is needed. And FYI Liquid Hydrogen is not as cold, I am sure you knew that, but some people seem to miss it. H2 is larger than Helium.

 

[Gerti]

I know that liquid He is very expensive and very cold but I can tell you that it's mainly used to cool the magnets on an MRI scanning machine.There are large currents induced to create a strong magnetic field so they use liquid He to cool it down.
It doesn't the magnets at that temp so i don't think that it would the CPU unles you apply a force on it.
Dunno what u mean that H2 is heavier than He.
If u mean it is less dense then it is but He has a larger relative molecular mass so it's atom is bigger then H

 

[William]

woops, i meant it is a bigger molecule than helium That was a typo. I had thought about liquid He, and I can understand why you would use it for magnets, but liquid Nitrogen is more than cold enough. I think it is just easierand cheaper just to add a much larger peltier, say a 300W or something like that.

 

[Clemson Physics Geek]

We happen to use both liquid helium and liquid nitrogen in the materials lab I work in. I assure you, you don't want to mess w/ liquid helium b/c it costs more than milk. Our magnet actually has a liquid nitrogen jacket around the liquid helium b/c liquid nitrogen is much cheaper. The big problem would be insulation. Liquid nitrogen is stored in double-walled vacuum dewars. In THEORY you could use one of those as your resevoir and set up some sort of pumping scheme, but you couldn't recool the nitrogen like you can water, you're have to vent it and just fill the thing up occassionally. If there's anyone out there w/ WAY more money than you know what to do with, then please try this, I'd love to see the results. The other consideration is that the properties of materials drastically change when you start getting that cold, so the chip might not even work.

 

[William]

the easiest ultra cool setup is to setup an acetone dry ice slurry and use a slurry pump. The slurry pump would be exepsnive, but it will be around -78C.

 

[Shadow ÒÓ]

are you nuts? I once had an ice cube stick to my tongue and burned the heck out of me.......no way I'm playing with anything colder than tap water anymore!

Seriously........there are drawbacks to it....and anyone who had access to it wouldn't even consider it because of them.

The MRI machines need coolant added occasionally.....the shortest way to explain it, and the way it was explained to me, the teacher had to add the coolant to the machine by removing a type of glass rod. The rod slipped and he grabbed it to keep it from falling out of instinct. needless to say he was hospitalized for quiet some time....he no longer has fingerprints.....and has to wear special gloves (this is from many years ago) to help keep what skin is left on his hands from cracking and bleeding. N2 can do the same damage.

I'd say there are less than 1% of the members here who are even qualified to talk about extreme cooling.....and even less who are capable of handling the material. Stick to peltiers.......that's plenty cool. =)

 

[Fink]

We use NMR machines in my line of work - essentialy a fancy MRI (same principal, different application). The superconducting magnet in the heart of such systems is cooled by He, and that stuff is dangerous and expensive. One of the techs on staff took a chunk of his pants and skin off his lower leg from exposure - not fun.

Want to see something fun? Let all the He boil out of the NMR and wait until the magnetic field collapses. The EMP does some scary stuff to computers...

I work with liquid N2 often, not to difficult, but it can give you some nasty frostbite if you are careless.

 

[Gerti]

Ok guys I see it's quite expensive and dangerous but i just thought what kind of results if any at all do you get by cooling the cpu with liquid He.

 

[Slain]

(l)He is a non starter. Silicon transitors actually cease to funtion an extremely low temps, something quantum no doubt. I can't remember the exact figure off the top of my head but if I find out I'll post it.

 

[Clemson Physics Geek]

Anyone ever pour liquid N2 over their hand? You can actually do it b/c the nitrogen forms a vapor barrier than insulates your hand. Just don't let it soak into your clothes! I've seen people put it in their mouth and then immediately spit it out but I'm a little leery of that.

 

[Slain]

I have dipped my whole hand into a dewar of (l)N2 just make sure you have *dry* hands. Any moisture encourages it to stick, I know this from personal experience :'( . Never tried it in my mouth, guess you need a dry mouth for this. Then again if you were going to do it, you probably would be a bit dry mouthed anyway!!!

 

[Clemson Physics Geek]

I've dipped too...dipped too far once and got some caught underneath my watch...stung but didn't do any permanent damage