• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Liquid Nitrogen Container and experiment

G

gclg2000

Gawd
Joined
Jun 27, 2004
Messages
634
I've said before that im' a student at UWF. I'm a medical/chemistry student there and we have access to alot of cool machine, substances and labs. But the cool thing is i talked to the lab worker today about possibly getting some Liquid nitrogen to do a experiment on CPU cooling with. I should be able to conduct a "experiment" on campus in a lab with a tube that i will make from copper obviously. Hopefully within the next month or two, i'll get around to actually do the overclock on whatever chip i have at that time.

Below is your classic LN2 containter design that i will be using. I'm going to take a ton of picutres and video when i do it. Hopefully it should be pretty cool run. I'm currently building a small single stage in addtion to my cascade cooler, so i plan on using the single stage on the North Bridge whil LN2'ing the CPU. SHould be a fun run.

Please post any experience or comments you've had with LN2.

LN2%20Container2.JPG
 
The tube should be made of an insulating material, otherwise it's going to be pulling alot of heat from the ambient air into your nitrogen. I also don't know that the base design would be most efficient. What I can see happening is when the nitrogen boils, it's going to form a bubble in the "cup" that will fill it up, momentarily disallowing liquid from replacing it until it breaks loose and rises. You may even end up with a static layer of gaseous nitrogen in each of the cups, basically an insulating blanket where you want the most heat transfer. A sandblasted surface on a flat base might be a good alternative, providing many nucleation points for gas bubbles but without having means to trap them.
 
oh ok cool zero. How did you do your LN2 experiment? Can i see the design you used so i can have a better run at it??? And yes those holes are just what i envision to be shot (not sand) blasted.
 
Good plan for the tube..his plan will be fine as long as the pits arent machined to deep and he shot blast (not bead blast) the suraface,then puts closed cell foam over the whole tube..besides...Liqued Nitrogen is -196*c..who is gona give a damn if it loses 10*c????


EDIT:: Guess i have to clean out my shot blasting box and get it ready to blast some copper..i got to blast my evap also...
 
gclg2000 said:
oh ok cool zero. How did you do your LN2 experiment? Can i see the design you used so i can have a better run at it??? And yes those holes are just what i envision to be shot (not sand) blasted.
Click to expand...


I've never done it myself. I've seen others do it though, most often an insulating container is used (could be something as simple as a Styrofoam cup). My conjectures regarding the container material and the "cups" stem from my studies as a materials engineer and my interest and tinkering in cooling devices and thermal properties of materials. So, don't praise my word as the absolute truth. I would be willing to defend them, however I wouldn't be astonished if I was wrong. You're building a system that almost approaches the scientific definition of chaos :D
 
BitchBreaker said:
Liqued Nitrogen is -196*c..who is gona give a damn if it loses 10*c????
Click to expand...


I guess there's no sense in being conservative in a practice of overkill , eh? :p
He won't lose any cooling power, the nitrogen will still boil at the same temperature. The stuff isn't cheap or everlasting though, so the slower it's consumed, the more time you have to deep-freeze your CPU
 
Tom's Hardware already tried this. They successfully managed to set a new world record CPU speed if I remember right. Look up "The 5ghz project" Not that it wouldn't be frickin' awesome if you did this on your own, but hey, if you're gonna do it, you might as well try to improve their design a bit :) See what you can come up with.

zer0signal667 said:
The tube should be made of an insulating material, otherwise it's going to be pulling alot of heat from the ambient air into your nitrogen. I also don't know that the base design would be most efficient. What I can see happening is when the nitrogen boils, it's going to form a bubble in the "cup" that will fill it up, momentarily disallowing liquid from replacing it until it breaks loose and rises. You may even end up with a static layer of gaseous nitrogen in each of the cups, basically an insulating blanket where you want the most heat transfer. A sandblasted surface on a flat base might be a good alternative, providing many nucleation points for gas bubbles but without having means to trap them.
Click to expand...

Perhaps the design that the koolance blocks I've seen use would work. The one where the whole surface is a bunch of little pyramids
 
DarkMonkey said:
Tom's Hardware already tried this. They successfully managed to set a new world record CPU speed if I remember right. Look up "The 5ghz project" Not that it wouldn't be frickin' awesome if you did this on your own, but hey, if you're gonna do it, you might as well try to improve their design a bit :) See what you can come up with.
Click to expand...


Tomshardware did that along time ago on a very older chip. I have a 560J chip which is much newer and if done right could hit 6ghz (fairly easy if mobo is modded properly).
 
DarkMonkey said:
Tom's Hardware already tried this. They successfully managed to set a new world record CPU speed if I remember right. Look up "The 5ghz project" Not that it wouldn't be frickin' awesome if you did this on your own, but hey, if you're gonna do it, you might as well try to improve their design a bit :) See what you can come up with.



Perhaps the design that the koolance blocks I've seen use would work. The one where the whole surface is a bunch of little pyramids
Click to expand...


There has been a handfull of 6+ ghz runs now.
 
I am a Ph.D. student in a chemistry lab. I use LN2 pretty regularly to flash freeze biochemicals and proteins and so forth.

Not having put a ~150W heatsource at the bottom of a LN2 dewar, I can't tell you exactly what is going to happen with your design, but my guess is that you're going to boil it off so fast that the escaping gas is going to blow a lot of the LN2 out of your tube (gun-barrel more like) geyser-style. I'd guess you won't make it through POST before it's mostly or all gone, and meanwhile you'll probably be splashing LN2 on your MoBo, which could be bad in terms of thermal shock cracking.

Better to have a much wider reservoir. Maybe you could have a narrow CPU contact and then a brief rise to a much wider radius barrel. That would provide extra reserve and be much less likely to be elevated en mass out of the tube. If I were in your place I'd want something that could hold upwards of 3-4L of LN2.
 
Not being a student, I tend to prefer calculations and experiments to guesses.

LN2 has a latent heat of vaporization of 199.1 J/g. With leakage heat added to that from the CPU, figure you'll boil away about a gram of nitrogen per second of operation. A liter of LN2 will last 12 to 14 minutes.

Geyser? Well, a gram of nitrogen gas at 77 K will have a volume of ~0.28 liters. To see how your dewar might react to 0.3 liter per second of gas production, fill it with water, drop a length of tubing into it, and exhale through the tubing. Normal adult lung capacity is 3-5 liters with a deep breath; did you make a geyser?
 
yeah, it won't boil off that fast. I'm a scinece student too and have used the stuff before and it doesn't go away that fast. I've frozen samples too...its not that fast. Smashed golf balls, flowers and all that fun stuff......all the video i've seen it lasts much longer than that.
 
zer0signal667 said:
I've never done it myself. I've seen others do it though, most often an insulating container is used (could be something as simple as a Styrofoam cup). My conjectures regarding the container material and the "cups" stem from my studies as a materials engineer and my interest and tinkering in cooling devices and thermal properties of materials. So, don't praise my word as the absolute truth. I would be willing to defend them, however I wouldn't be astonished if I was wrong. You're building a system that almost approaches the scientific definition of chaos :D
Click to expand...

Oh i thought you had done it before. Since you talked so much about it.

My bad.

BUt yeah, it should work out fine. The picture really does no justice to what a shot blasted surface would look like. I feel when that is done it will be great. Besides all the tubes i've are essentially just like this. I was just posting this up for the board on how i'm gonna conduct it.
 
Whats the point? Seems like the setup to something I would see on the 5 o'clock news after you splashed some on yourself. I would be more impressed if you designed a storage system that let you use this for more than 15 minutes.

Just being critical cause this post brought forth an image of joe schmoe's trying this in thier garage.
 
gclg2000 said:
I've said before that im' a student at UWF. I'm a medical/chemistry student there and we have access to alot of cool machine, substances and labs. But the cool thing is i talked to the lab worker today about possibly getting some Liquid nitrogen to do a experiment on CPU cooling with. I should be able to conduct a "experiment" on campus in a lab with a tube that i will make from copper obviously.
Click to expand...

Hi, I'm Joe.

I love [H]ardforum so much.
 
How much heat do you think was in that golfball? Now how about the heat coming from a running 150W floodlamp? Not nearly the same thing.



gclg2000 said:
yeah, it won't boil off that fast. I'm a scinece student too and have used the stuff before and it doesn't go away that fast. I've frozen samples too...its not that fast. Smashed golf balls, flowers and all that fun stuff......all the video i've seen it lasts much longer than that.
Click to expand...
 
Congrats! The first time is always memorable. :p

BitchBreaker said:
Yea...well...i had sex last night!....but dident get to break out the Liqued Nitrogen :(
Click to expand...
 
But he's not going to have a liter of LN2 in there is he? More like 300 ml, and by your math, 300 ml of LN2 will boil off in about 3.5 minutes. Just about enough time to boot into BIOS, change his settings to what he wants, and reboot.

If you don't think that producing 300 ml of gas a second in a narrow container that can barely hold about 300 ml of LN2 isn't going to produce considerable splashing, you're certainly welcome to do the experiment yourself. Make sure you watch closely from the top. :rolleyes:

HeThatKnows said:
Not being a student, I tend to prefer calculations and experiments to guesses.

LN2 has a latent heat of vaporization of 199.1 J/g. With leakage heat added to that from the CPU, figure you'll boil away about a gram of nitrogen per second of operation. A liter of LN2 will last 12 to 14 minutes.

Geyser? Well, a gram of nitrogen gas at 77 K will have a volume of ~0.28 liters. To see how your dewar might react to 0.3 liter per second of gas production, fill it with water, drop a length of tubing into it, and exhale through the tubing. Normal adult lung capacity is 3-5 liters with a deep breath; did you make a geyser?
Click to expand...
 
I can't measure the volume of the proposed container simply by looking at, and I don't see and dimensions mentioned. Where'd you get 300ml? Anyways, the cool thing about containers with open tops is that you can pour more into them!

If you don't think that producing 300 ml of gas a second in a narrow container that can barely hold about 300 ml of LN2 isn't going to produce considerable splashing, you're certainly welcome to do the experiment yourself.
Click to expand...
Didn't I just suggest testing the splashing/geysering potential before use? I even came up with a cheap simple method. I could've sworn you even quoted it. :)
 
For the continuing arguments from all these experts that have done this around here.....


The tube will be either 2 or 3" wide with at least 18" in height or so. I guess i could make the CAD drawing the EXACT height. I just figured the "general" idea would be what everyone would understand
 
gclg2000 said:
Oh i thought you had done it before. Since you talked so much about it.

My bad.
Click to expand...

Like I said, I was only speaking out of conjecture with knowledge on what I consider to be relevant topics. Even personal experience won't tell you how surface finish on a micro-scale changes surface energies and fluid flow. Then again, it might not even matter in your case... who knows. You might try being receptive to peoples' advice too, before you dismiss them as being "wannabe-experts". Either way, I'll keep my conjectures to myself. But good luck, and have fun.
 
Don't ask for opinions if you don't want them. Now play nice or I set the forum servers on fire.
 
Okay, back ON topic: I've seen the tube trick done many times before. Be it either friends at a college, or on the Screen Savers (before it was raped by G4). You put the nitrogen in the tube, start the computer up, and keep pouring more in as it evaporates. It IS possible people, just be sure to have a foam insulator on that tube.

Now, if someone could figure out Direct Die cooling with a tube setup.
 
Could you JB weld a copper tube to your cpu? Leave the core untouched and just pour ln2 right on it :D

Or would the JB eat the silicon?
 
i have a feeling you would not wanna try that....besides you live in alaska...why would you wanna do that when you can stick it down in the snow 20 feet :p
 
Unless u plan on making a exhaust valve into this experiment ur gonna have one large faliure in my opinion. Ever notice how a dermatologists freeze guns all have exhaust valves. The pressure builds up as the gas boils over. Well now that i think about it the actual Liquid nitrogen Tanks dont have exhaust valves. But they are also really really sturdy tanks. To do this without the exhaust valve would seem like overlly dangerous.

Exhaust valve=more wasted Liquid Nitrogen
Exhaust Valve= no Explosion

I used to work for a company that made those guns and i did quite a bit of RD "playing with" Liquid Nitrogen. Funny how it wont actually burn u unless u stop it from moving. Can dump a vat over ur head and feel nothing. (i think) never tried that one. Though its definetly fun to crank the tank on full blast as someone walks up to them. Will scare the person silly as Steam and ice cold liquid shoots across the room only to vanish harmlessly.
 
solo16 said:
Unless u plan on making a exhaust valve into this experiment ur gonna have one large faliure in my opinion. Ever notice how a dermatologists freeze guns all have exhaust valves. The pressure builds up as the gas boils over. Well now that i think about it the actual Liquid nitrogen Tanks dont have exhaust valves. But they are also really really sturdy tanks. To do this without the exhaust valve would seem like overlly dangerous.

Exhaust valve=more wasted Liquid Nitrogen
Exhaust Valve= no Explosion

I used to work for a company that made those guns and i did quite a bit of RD "playing with" Liquid Nitrogen. Funny how it wont actually burn u unless u stop it from moving. Can dump a vat over ur head and feel nothing. (i think) never tried that one. Though its definetly fun to crank the tank on full blast as someone walks up to them. Will scare the person silly as Steam and ice cold liquid shoots across the room only to vanish harmlessly.
Click to expand...


What's going to explode? There's nowhere for pressure to build up, it's an open container.
 
crazyman_130 said:
....besides you live in alaska...why would you wanna do that when you can stick it down in the snow 20 feet :p
Click to expand...


Duh, its only -20 F here... :D
 
nvm then i thought u were gonna try and put it in a closed container/piping/radiator setup.....I should read more closely next time. Sad that im taking your word for it that it was mentioned before ;) but too busy to inspect the thread for flame bait. So either way no big deal.
 
Yeah we gotcha. lol all ya had to do man is read the #1 post and look at the pic.
 
lol...i wouldnt wanna pour the stuff over my head...i might try and pour it over my hand after reading more in depth about it...oh btw when i say hand i mean my finger first and if it survives then two and so on.
 
You should go take a look at http://www.tomshardware.com and look at their video section...they have already built a cooler similar to what you are talking about and you might find something useful.

...Dan
 
crazyman_130 said:
lol...i wouldnt wanna pour the stuff over my head...i might try and pour it over my hand after reading more in depth about it...oh btw when i say hand i mean my finger first and if it survives then two and so on.
Click to expand...

I stick my hands in LN2 pretty often. In the lab, we use it to flash freeze little sample tubes. They usually float, so the easiest way to get them back out is just to reach in and grab them. It won't hurt you if you're pretty quick about it because the flash boiling around your hand forms a layer of bubbles.
 
DarkMonkey said:
http://www.tomshardware.com/cpu/20031230/5ghz-01.html Look at the arcticle date, December 30, 2003. I'm quite sure back then, the 5.25 or whatever ghz they reached was a new record... anyways, they've got a video and go into great detail about how they made the cooler and how they had everything set up to deal with all the problems.
Click to expand...


That is my favorite READ ever, I saw that years ago when it first came out and read ever word and was in aww for days, think it is the coolest thing ever :D :D :D
 
R!P13y said:
That is my favorite READ ever, I saw that years ago when it first came out and read ever word and was in aww for days, think it is the coolest thing ever :D :D :D
Click to expand...
Same, even did a powerpoint on it for a school project =] was simply amazing
 
You must log in or register to reply here.
Back
Top