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

LN2 cooling: what do I need?

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

Careface

Member
Joined
Jul 4, 2004
Location
Auckland, New Zealand
So ive decided to take the plunge in the deep end and go to some LN2 cooling for some quick suicide runs/benches, but I had a few questions because..well, since im currently running on stock cooling, im not to sure what taking this big of a jump will be like.

Basically what I was thinking was that I could get a block of copper, lap it smooth; create a copper tube and get it welded to the block. That should be no problem to handle. Insulating with foam will be easy as well. As for mounting it, I'll have to figure it that out when i come to it. Getting hold of the LN2 should be no problem either, thats been sorted. 30L should do for a good days benching im thinking.

Now where I get hazy is what I should be doing about condensation.. Ive read there you need to put some dielectric grease on the CPU and socket and surrounding the socket.. but on the toms hardware 5ghz project video, I couldnt see any grease at all when he was using LN2, so whats the deal with this?

My next problem is that I hear that Venice's have the cold bug.. i cant remember if this is true or not, i get mixed responses about that so yeah, does anyone know if the cold bug is present? If it is, ill be usingthe LN2 on my fully unlocked AXP2000+, but id rather it be on the venice.

Anyway umm summary:
1)What precautions should I take for condensation other than foam/a heater running on the back of the mobo around the socket.

2)Do the venices have the cold bug?

3)Anything else im forgetting other than general safety precautions?

Thanks for any help, Id really love to do an LN2 bench, as im sick of stock cooling, and I wont be able to get the swifty APEX h2o kit for a while, so why not?

Careface*
 
Going from air cooling to LN2 in one jump is not realistic. It's not impossible but if I were you I would try DI first.

I have never seen a heater used under LN2. Most guys want every ounce of cooling they can get. Running LN2 your risking hardware to go as fast as possible, any extra heat is not wanted.

Even with the best insulation you will still get condensation. Grease the socket, nail polish or coat the board surface with something waterproof, use fans, plastic bags or saran wrap along with the container insulation itself.

Your Venice will not run under LN2 temps, sorry. DI maybe, but you won't know that until you actually try it. Most .130nm A64's will work under LN2 temps.

First build your container. If you get that far then you can worry about the rest ;
 
i say make your tube and dry out dry ice, just make sure you do a good job with isulation. most dry ice tubes have a ring at the bottom where the mounting plate can sit (slide it down the top. they're covered in fitted pipe insulation. you want a piece for the bottom that's a little too long so it gets compressed by the mounting and creates an airtight seal around the socket. doing that you don't need dielectric grease, although be forewarned that when you go from -70c in the socket to 20+c room temp you will get condensation formed all over the socket and will need to let it dry out.

please don't go right from air cooling to ln2, your chip wont work and you're almost guaranteed to kill something.
 
thanks for all the replies, I never actually thought of using dice.. im guessing that making a dice container (in acetone bath, right?) would be easier (and less dangerous).. and would probably not be cold enough for the venice cold bug to kick in? i guess ill read up a little more on dice cooling, as you guys are probably right: stock air -> LN2 is too much of a jump, but i think i could do a dice jump before my h2o arrives.

Careface*
 
You should also be carefull about not cooling down the chip to fast if you do go to L2N eventually. If your container is fairly thin (which it should be) and you just poor L2N in, your chip will go from ambient to minus a whole lot in virtually no time, and it may crack.
 
Back