I know i will get flamed alot for even posting this, but i have so far acquired some older computers that are disposable, so me and some of the other people i know that have experience working with liquid nitrogen, are going to be running test experiments with many different set-ups, and trying to find a way to make it useful, but not a gigantic blow to the pocket, much like phase change. We are also working with a chemist to supply this, as well as for advice upon differenet types of metals to use for the cooling block, and that different stuff, then it will be vacumized to keep the temperature from changing much, keeping the nitrogen in the pipes, which i will not disclose our pipes yet, but either way this is for some fun to see if we can get anything done at all, so you guys post ideas if you have any about something that would be useful for this.
Working with some others on Liquid Nitrogen
XgamerGt04
1 - 20 of 27 Posts
Joined
·
42 Posts
I know i will get flamed alot for even posting this, but i have so far acquired some older computers that are disposable, so me and some of the other people i know that have experience working with liquid nitrogen, are going to be running test experiments with many different set-ups, and trying to find a way to make it useful, but not a gigantic blow to the pocket, much like phase change. We are also working with a chemist to supply this, as well as for advice upon differenet types of metals to use for the cooling block, and that different stuff, then it will be vacumized to keep the temperature from changing much, keeping the nitrogen in the pipes, which i will not disclose our pipes yet, but either way this is for some fun to see if we can get anything done at all, so you guys post ideas if you have any about something that would be useful for this.
Joined
·
22,879 Posts
I just think as long as you keep the nitrogen in its liquid state, it will be kickass. Let us know how it goes!
Joined
·
42 Posts
The liquid nitrogen staying nitrogen is currently the least of my problems, i believe we have a simple solution for that, use a dewer with a pump from it (
don't know where we found it) and connect that to a part of the cooling system, with a three way head(cept the one for the dewer can be closed, and the one for the vacum sealer can be closed) so we create a vacum in it then seal its valve off, and then put the liquid nitrogen in, and then seal that one off, then let it run, and it goes into a vacummade resivior (everything is a vacum basically) then from their to our specially made heat sink (not gonna tell what its made of) then we may or may not have a radiator (think we need one?) and then back to the resivior.
don't know where we found it) and connect that to a part of the cooling system, with a three way head(cept the one for the dewer can be closed, and the one for the vacum sealer can be closed) so we create a vacum in it then seal its valve off, and then put the liquid nitrogen in, and then seal that one off, then let it run, and it goes into a vacummade resivior (everything is a vacum basically) then from their to our specially made heat sink (not gonna tell what its made of) then we may or may not have a radiator (think we need one?) and then back to the resivior.
Joined
·
190 Posts
How can you have a heatsink with liquid nitrogen? It will absorb heat from the air rather than release it...
Correct. Liquid nitrogen is not inherantly cold - its only when its allowed to evaporate that it cools down to its famous temperature. Liquid-nitrogen cooling is good for achieving impressive overclocks (I believe someone managed to make a 500MHz system run at nearly 2GHz once), but only for as long as the nitrogen lasts. You cant sustain it, at least not unless you want to build a very large phase-change compressor system.
Also, careful about oxygen dilution. If you start to feel at all lightheaded during the experiment, get out immediately.
And your pump might be a problem too - if its just pumping compressed liquid nitrogen, fine. But if its cold, it can be sold enough to make plastics fail and shrink metal to the point where seals leak.
Also, careful about oxygen dilution. If you start to feel at all lightheaded during the experiment, get out immediately.
And your pump might be a problem too - if its just pumping compressed liquid nitrogen, fine. But if its cold, it can be sold enough to make plastics fail and shrink metal to the point where seals leak.
Joined
·
42 Posts
The heatsink i was talking about is made like a water block, and it looks like the one where they did the 5ghz overclock, just smaller, we've been talking with some other people (chemiestry teachers, ac specialist, everything that we can get to) so the heatblock will have small holes, that will allow some of the LN to evaporate, but then the rest will be sent back and recompressed, then sent through the pump back to the heatblock, and that's what we have so far.
Joined
·
1,687 Posts
You do know that Nitrogen compression systems are huge, heavy (you'll need a forklift to move it) and probably draw more power than your home right.
Next if you only plan to allow a "bit" to evaporate, you don't need a nitrogen compressor since in your plan most of the nitrogen will stay in a liquid state. Unfortunately you fail to mention how you'll remove heat from the loop as you don't mention a radiator anywhere.
What tyep of pump do you plan to use? Something industrial no doubt. Also how will you deal with the added pressure? The brittleness of the metals? Any ideas on hwo to deal with condensation?
If you have the skills and ability to build yourself a liquid nitrogen loop, why not just build yourself a cascade, it's definitely easier.
Next if you only plan to allow a "bit" to evaporate, you don't need a nitrogen compressor since in your plan most of the nitrogen will stay in a liquid state. Unfortunately you fail to mention how you'll remove heat from the loop as you don't mention a radiator anywhere.
What tyep of pump do you plan to use? Something industrial no doubt. Also how will you deal with the added pressure? The brittleness of the metals? Any ideas on hwo to deal with condensation?
If you have the skills and ability to build yourself a liquid nitrogen loop, why not just build yourself a cascade, it's definitely easier.
Joined
·
42 Posts
I believe that this experiment will be trashed soon, our first attempts failed, the heatblock works for a small ammount of time, but then the holes in the metal started to form cracks, the pipes cracked after a few hours, and we eliminated the loop, just a path to the heatblock and then it evaporates, because the cracks grew to large, more ln that was supposed to evaporate did, and a few of us got a bit light headed during the experiment, so we either have to trash it, or redesign the heatblock, and find a way to keep the holes from cracking.
Joined
·
1,687 Posts
Quote:
Definitely stop playing if you get light headed, if you want to experiment with LN2, do it outdoors.
Did you machine your own block? I'm rather surprised that the block and pipes cracked, what type of metal are you using?
Originally Posted by XgamerGt04 I believe that this experiment will be trashed soon, our first attempts failed, the heatblock works for a small ammount of time, but then the holes in the metal started to form cracks, the pipes cracked after a few hours, and we eliminated the loop, just a path to the heatblock and then it evaporates, because the cracks grew to large, more ln that was supposed to evaporate did, and a few of us got a bit light headed during the experiment, so we either have to trash it, or redesign the heatblock, and find a way to keep the holes from cracking. |
Did you machine your own block? I'm rather surprised that the block and pipes cracked, what type of metal are you using?
Quote:
Cracks... really... *applauds*
Originally Posted by XgamerGt04 I believe that this experiment will be trashed soon, our first attempts failed, the heatblock works for a small ammount of time, but then the holes in the metal started to form cracks, the pipes cracked after a few hours, and we eliminated the loop, just a path to the heatblock and then it evaporates, because the cracks grew to large, more ln that was supposed to evaporate did, and a few of us got a bit light headed during the experiment, so we either have to trash it, or redesign the heatblock, and find a way to keep the holes from cracking. |
Joined
·
4,702 Posts
Send Postaldudeleo an E-mail on how to do it, He's got a pretty insane looking auto-recharging LN² system.
I'm still waiting for some result Leo... *Stomps foot impatietly*, lol..
http://www.overclock.net/showthread.php?t=13663
I'm still waiting for some result Leo... *Stomps foot impatietly*, lol..
http://www.overclock.net/showthread.php?t=13663
Joined
·
71,082 Posts
KLKL this sounds good ..let us know how you get on
Joined
·
42 Posts
we we're using some of the older windows 98's with about 64 mbs of ram and such, so far we're back to designing the heat block to keep it from allowing to much LN from escaping, but we're working in a well ventilated area now, until we devise a way to get rid of the gas that is expelled, possibly make a template for holes to be drilled into the pc, and have some sort of way to catch the nitrogen gas in a container.
Update
Today we finished the redesign of out heatblock, and we kept it working for quite sometime, the LN comes from the dewer atm, because we havent designed a containment for the Ln, so it goes from there to the heatblock via some coppertubing that is used in air conditioning, then from the heat block it evaporates slowly from small holes in the block. Our test today showed that the processor was about -260 something fahrenhiet, but it only last for a short time atm, because the heat block still cannot last for more than a few hours, but we are getting closer to our goal, and at this rate we may find a good combination for cpu and possibly gfx before the end of this year.
Update
Today we finished the redesign of out heatblock, and we kept it working for quite sometime, the LN comes from the dewer atm, because we havent designed a containment for the Ln, so it goes from there to the heatblock via some coppertubing that is used in air conditioning, then from the heat block it evaporates slowly from small holes in the block. Our test today showed that the processor was about -260 something fahrenhiet, but it only last for a short time atm, because the heat block still cannot last for more than a few hours, but we are getting closer to our goal, and at this rate we may find a good combination for cpu and possibly gfx before the end of this year.
Joined
·
648 Posts
Wow, just noticed this thread. Sorry if it took a while. You don't need pumps. The contents of the dewar are under pressure so just screw on a tube of copper and open the valve; it will almost instantly flow through the tube and if you put a hole in the tube then make a stopper, you can easily control the start and flow on the spot.
(This function because the dewars have a lttle valve that allows the gass to slowly exit when it evaporates as to keep pressure form compiling inside the dewar and ripping it. If you block the tube, even with your finger (don't recommend it), the gas will evaporate from the dewar valve which opens auto under a certain pressure.)
Anyway, you can't have a closed space around otherwise it will help cracking the enclosure or the evap itself when pressure builds. Aka just punch a hole in the evaporator.
Thats all i can think of at the moment. I can safely say that 20 liters of liquid nitrogen can last about a week or more if used non-stop. I used it for 2 days in a small evaporator. And only used 2-4 liters.
The best cooling is to use the liquid on the CPU(s) and gas on EVERY other component. To get a massive overclock, you can just keep the temperture at a steady 0. Liquid nitrogren will give you lots of headroom which is why it works well but a very well (literally nuclear plant quality) chilled water cooling can technically get the same result.
(This function because the dewars have a lttle valve that allows the gass to slowly exit when it evaporates as to keep pressure form compiling inside the dewar and ripping it. If you block the tube, even with your finger (don't recommend it), the gas will evaporate from the dewar valve which opens auto under a certain pressure.)
Anyway, you can't have a closed space around otherwise it will help cracking the enclosure or the evap itself when pressure builds. Aka just punch a hole in the evaporator.
Thats all i can think of at the moment. I can safely say that 20 liters of liquid nitrogen can last about a week or more if used non-stop. I used it for 2 days in a small evaporator. And only used 2-4 liters.
The best cooling is to use the liquid on the CPU(s) and gas on EVERY other component. To get a massive overclock, you can just keep the temperture at a steady 0. Liquid nitrogren will give you lots of headroom which is why it works well but a very well (literally nuclear plant quality) chilled water cooling can technically get the same result.
Joined
·
15,303 Posts
Quote:
Keep in mind that you need to run antifreeze in your setup.
Originally Posted by postaldudeleo Wow, just noticed this thread. Sorry if it took a while. You don't need pumps. The contents of the dewar are under pressure so just screw on a tube of copper and open the valve; it will almost instantly flow through the tube and if you put a hole in the tube then make a stopper, you can easily control the start and flow on the spot. (This function because the dewars have a lttle valve that allows the gass to slowly exit when it evaporates as to keep pressure form compiling inside the dewar and ripping it. If you block the tube, even with your finger (don't recommend it), the gas will evaporate from the dewar valve which opens auto under a certain pressure.) Anyway, you can't have a closed space around otherwise it will help cracking the enclosure or the evap itself when pressure builds. Aka just punch a hole in the evaporator. Thats all i can think of at the moment. I can safely say that 20 liters of liquid nitrogen can last about a week or more if used non-stop. I used it for 2 days in a small evaporator. And only used 2-4 liters. The best cooling is to use the liquid on the CPU(s) and gas on EVERY other component. To get a massive overclock, you can just keep the temperture at a steady 0. Liquid nitrogren will give you lots of headroom which is why it works well but a very well (literally nuclear plant quality) chilled water cooling can technically get the same result. |
1 - 20 of 27 Posts
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
