• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Continuous liquid nitrogen cooling?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

ln2dewar

New Member
Joined
Jul 12, 2014
Hey Guys,

I want to have my PC continuously cooled using liquid nitrogen and have continuous flow of liquid nitrogen to CPU. Only problem I see is that if I start cooling the air getting into the system, the entire thing would turn into an iceberg :-/ ( The snow and water would just short everything).
Since I work in the surface science laboratory and have used FTIR instrument that needed to be purged by nitrogen ( to remove air and water with it) I got an idea.

I could purge the entire PC with nitrogen. It would be cool to if you had only one input and one output for airflow on a PC assuming its completely sealed because that way one could purge it with liquid nitrogen boil-off ( completely dry nitrogen from the liquid nitrogen dewar). Once the gas inside PC is completely and exclusively nitrogen one could set up some kind of liquid nitrogen flow to CPU so no iceberg and :attn:

Let me know what you guys think?

Asphyxiation hazard maybe ?

Imagine if I also had a ln2 generating machine this could be a permanent fixture.

I'm telling you, just the insanity of the idea would make it worth the money.

Cheers
 
great idea but, things contract when they get cold, even worse when very cold.
I took one of my rigs to a giant freezer at -40f. the video card would not maintain contact, a couple of fans, well lets just say they did not like it at all, in the -60f room the thing just would not even try to boot
 
Well... I suppose anything can be done... There is a machine that makes ln2 as well.

That said.. how you going to insulAte? How you going to maintain a temp? Some chips, Intel, have cold bugs...

It's just not a realistic 24/7 cooling method. WHY 24/7 ln2 in the first place?
 
great idea but, things contract when they get cold, even worse when very cold.
I took one of my rigs to a giant freezer at -40f. the video card would not maintain contact, a couple of fans, well lets just say they did not like it at all, in the -60f room the thing just would not even try to boot

Well how about just the CPU for prolonged periods of time ? ( while having the insides of PC purged with nitrogen)

BTW where do I buy heat-sink for intel extreme qx9650 that would support liquid nitrogen flow ?

Maybe I can use some all copper solution ?

Well... I suppose anything can be done... There is a machine that makes ln2 as well.

That said.. how you going to insulAte? How you going to maintain a temp? Some chips, Intel, have cold bugs...

It's just not a realistic 24/7 cooling method. WHY 24/7 ln2 in the first place?

No necessarily 24/7 but just long time ( say 10 hours of so).

I had no idea about cold bugs. Which chips have cold bugs?
 
Last edited by a moderator:
sealing the case will be the *****.
I weld titainium, in a converted glass beading cabinet, I have to stay below %4 o2 or it's trash.
keeping that sealed is just the devil and you might need to keep it that low also.
 
I don't mean constantly 24/7... You seem to want to switch it off and on? That constant freezing and thawing with be a moisture night mare.
 
Last edited:
You might be better off modifying a scientific lab freezer to accept a MB, CPU, graphics card, and basic PSU. No fans, needed if you purge it, fill it with an inert gas such as nitrogen or helium, perhaps argon. Run cabling from the inside to the outside and seal the fitting/busing area completely and flip the switch. Might work that way.
 
Your CPU will stop functioning somewhere between -80°C and -140°C. LN2 is -196°C.

You'd be far better off buying a Cascade cooler, they typically run somewhere between -90 and -120, and can run indefinitely.

Or buy yourself a used 2600k and overclock it to an easy 4.2GHz. That will be faster than the C2Q on LN2.
 
Or buy yourself a used 2600k and overclock it to an easy 4.2GHz. That will be faster than the C2Q on LN2.

^^^ This ^^^

For this to be viable you would have to catch all the vapor and condense it back into a liquid. This is actually being worked on believe it or not.

Not as a 24/7 solution but as a way to reduce benching costs
 
it is impractical at best to run 24/7 ln2. Get a better setup, more cost effective, imo.
 
I think that you identified the real challenge here. In order to get below detection limit of FTIR for water vapors I had to purge the spectrometerer with insane flow rate of nitrogen. Sometimes it even get's the moving mirror frozen :)
I could buy a big plastic box, mount a couple of electrical feethroughts ( power and intenrnet) put the PC inside. Somehow put the intake and output valve on it.
 
Man - Bro -Guy .....

Try investing in a Cascade unit. That'll run a Cpu 24/7 probably around -90c or colder.

My 2 cents.
 
Back