i think this a total stupid question and wont work but...Can liquid nitrogen be put into a water cooled set up with out mods or slight mods?

wont the nitrogen crack the tubes and wont there be mad condensation?

LN evaporates when it hits the air. It wouldn't last long. Your stuff would need to be heavily insulated because there would be condensation. And yes, plastic parts would become brittle when exposed to sub-zero temperatures.

When they do record-breaking overclocks they insulate a metal tube that sits on top of the CPU block and fill it with the nitrogen.

I've never heard of anyone using N20 in an H20 cooling setup, probably because like you said it would crack the tubes. Most N20 coolers I know of involved directly putting the Nitrogen on the PC...as in dumping it on the CPU.

Condensation can be fixed, if you look at the way phase change coolers work you'll see that there are ways around it.

If you want a super-chilled liquid cooling system that could get the CPU below ambient, one idea would be to drill an entry and exit hole large enough for the tubing into a minifridge, then put your resevoir in there. Slide the tubing into the fridge, hook up the resevoir inside the fridge, then seal the holes with a caulk gun or something.

Then you'd have to worry about condensation in the socket, and you might still run the risk of blowing out the fridge's compressor, since there's an endless supply of heat inside.

LN2 is something like - 320 degrees Fahrenheit, tends to make metals extremely brittle and subject to shattering, would probably freeze your pump solid without special lubricants, would likely cause such a temperature differential on your mobo as to crack it, the CPU socket, or both.

Besides, where are you going to get a supply of LN2? Dunno what a large cryo dewar container would cost, but you would have to resupply it on a regular basis.

When I belonged to an astronomy club we checked into the cost of a LN2 generator unit - the smallest available - and it was about $27,000 discounted if we submitted the purchase order on the club letterhead. Not including the cost of compressed N2 gas bottles either. You have to use pure N2 as a source because LN2 has a lower boiling point than LOX (liquid oxygen); therefore if you try to liquify air you are going to get LOX first which can be highly explosive. As I recall, the output was something like 3 liters per day. Oh, and the electricity costs were rather substantial to boot.

So, other than risking explosion, cracking and ruining your mobo and CPU, and shelling out huge amounts of money, it's a really good idea

A dewar large enough to supply the nessary amount LN2 for any reasonable amount of time would cost hundreds, if not thousands depending on your location. It would cost about the same to fill the thing and transport it.

So yeah, phase or a chilled liquid system are your only options if you want LN2-like temps.

ya the minifridge wouldnt work..you would have to get a large cooler and fill it with ice and run tubes threw it and have the radiator sit atop of it or someone else....

i was going to do that...but then wooooooo nitrogen sounded sweet...i knew it wasnt going to work..


iam stickin to the cooler idea

Have you heard of bong coolers? They're pretty awesome. You can get sub-ambient temps for about $50. All you need is a showerhead, a couple lengths of PVC pipe, a bucket, and a pump powerful enough to move the water.

http://en.wikipedia.org/wiki/Bong_cooler
http://www.overclockers.com.au/techstuff/a_bong/
http://www.wc101.com/guides/bongs/
http://www.ocforums.com/showthread.php?t=321055
http://www.overclockers.com/articles389/index.asp
http://www.xtremesystems.org/forums/showthread.php?t=78543&highlight=bong



Oh, and about your cooler full of ice idea. If you do go through with it, along with the regular ice, throw 5 or 6 pounds of dry ice in too. It'll make the regular ice last alot longer, and net you better temps.

The cooler idea won't work so well, regardless of the cooler. They are not built to handle constant heat sources.

The best methods for cooling below ambient temperatures are peltiers for a cheaper alternative, and phase-change if you can afford it and want to go extreme.

Everyone worries about the tubes, very understandable. What about the pump. Will it even work at those temps?

Everyone worries about the tubes, very understandable. What about the pump. Will it even work at those temps? Can you say antifreeze?

Use a heat exchanger instead of a radiator on the water side of the loop, and connect the heat exchanger to a phase-change system. LN2 isn't going to work here.

It is quite easy to overclock with liquid nitrogen, that is of course if you are Tom's Hardware. If you really want to go for the gusto here is a how to, but be prepared to get your wallet out. Note the video download on page three it may be old but it's still fun to watch.

5 GHz Project: CPU Cooling With Liquid Nitrogen
http://www.tomshardware.com/2003/12/30/5_ghz_project/

what about liquid co2? Not as cold as liquid nitrogen. I think it still would break the tubes....

I WILL GIVE SOMEONE 5 BUCKS THROUGH PAYPAL IF THEY GIVE ME ANOTHER LIQUID TO USE IN A WATERCOOLING LOOP THAT IS BETTER HEAT CONDUCTOR AND COLDER THAN WATER(BUT NOT TO COLD) AND IT HAS TO BE EASILY AVAILIBLE( :x :x :x :x )

First if all colder then water? Water isn't naturally cool, it's temperature is determined by the ambient temp of the environment the water is in and the technique used to cool it (if applicable.).



Anyway, here's my second shot at reorganizing this post.

From your posts it sounds like you're trying to mix two, completely different forms of cooling.

Any liquified gas such as LN2 or liquid CO2 wouldn't work at all in a standard water cooling loop. The first reason is that almost, if not all, liquid gases are WAAAAAAAAY to cold for standard parts to handle. The second is that, well, it's impossible. For you to understand why it's impossible you have to know how water cooling and LN2/dice cooling work.


Water cooling uses conduction cooling to cool a pc. LN2, dice, ect., use evaporative cooling.


The principles of evaporative cooling are a little hard to understand, but I'm pretty sure I can explain it. For a liquid to turn into a gas it requires energy, but it can't just create energy, since energy can be neither created nor destoryed, just transfered. So it has to get it from somewhere else, in this case, from heat energy. When using evaporative cooling to cool a pc the heat source is whatever component you're trying to cool, the cpu, gpu, whatever. Energy is transfered from the heat source to the liquid. Less energy = less heat. When the liquid has enough energy it turns into a gas. Then both the gas and the transfered energy become a part of the atmosphere, removing them from the pc completely. The colder a liquid is, the more energy it requires to turn into a gas.

Another example of evaporative cooling us the creation of dry ice. Liquid CO2 cant exist at regular atmospheric pressure. That's why it turns from a solid to a gas. All you need to make dry ice is a tank of liquid CO2. When the tank is depressurized the liquid immediately evaporates, this lowers the temp of the liquid CO2 that hasnt vaporized, since the evaporating liquid needs energy, creating ice. The ice is then packed into pellets, blocks, ect., and shipped out.

Conduction cooling does just that, conducts heat. In a water cooling loop cold liquid starts by running across the hot components, the water absorbs and thus removes the heat from the components, cooling them. The now hot water moves to the cool radiator, where the heat energy is transfered yet again, then removed from the pc and into the atmosphere through the use of fans. So the water cools the components, the radiator cools the water, and the fans remove the heat. Same end for heat energy, just a different way of getting there.

The best fluid to use in water cooling would be an anti-freeze/water mix, possibly with some water wetter thrown in. Water has one of the highest specific heat capacity of all liquids, the only thing I can find that's better is liquid ammonia, and that would be insanly hard to use safely and effectively in a wc loop. Specific heat capacity is the amount of energy it take to raise the temp of an element by 1ºc. The higher the specific heat capacity, the more energy an element can absorb before it's temperature is raised. The water wetter and anti-freeze pretty much do one thing, heighten the boiling point and lower the freezing point, essentially raising specific heat capacity of the liquid.


So now that the basics are covered, I can explain WHY it's impossible.

A water cooling loop is closed, meaning that the only way to expel heat is from conduction cooling, since there are no exits for gases and the energy they contain to escape, so energy can't be removed from the system. To turn LN2 (or a liquid gas like it) from a gas into a liquid you need extreme cold. Regular conduction cooling can't get below the normally hot (when compared to LN2) ambient temp of the room, because it expels energy into the atmosphere, and there's a certain amount of energy already in the atmosphere. As I said before, energy can be neither created nor destroyed, only transfered. If the gas cant get to the necessary temp for it to turn into a liquid, the evaporative cooling process cant start over again, and you end up with a WC loop filled with gaseous nitrogen.

Why not deuterium? Or heavy water? It has superior heat displacement propertied than regular water and you can make it yourself, it will take a long long long long time, but it can be done. Wiki explains how you can do this on your own, but the easiest method really is the peltier, you can also make this yourself, peltier are the simplest methods for cooling out there.

Okay, I want the five bucks. Add some CFC or commonly known as freon, to your water, it boils at room temp and will act as a cooling agent, you will have to seal your setup from air so it doesnt evaporate out. Al Gore will cry if you do this, but his tears cure cancer, so it all balances out in the end.

I've already given you my email via PM, I'm so buying slurpee's with that finsky!

Unfortunately, there is no such thing as liquid CO2 at atmospheric pressures. Solid CO2, aka dry ice, sublimates or evaporates directly to gaseous form at 1 bar (sea level) atmospheric pressure. If you want a liquid (recommended for pumps ) you'll have to raise the pressure to 5.1 atmospheres and cool it to -65 degrees Fahrenheit. If you raise the pressure to about 100 atmospheres it'll liquify at +88 degrees Fahrenheit.

Really, your practical choices are water, peltier or phase change unless you are Bill Gates in disguise ...

You forgot one other reason liquid nitrogen or liquid CO2 won't work in a water cooler... They Are Under Pressure. Nitrogen or CO2 would simply cause the water cooling system to Explode before it even got cool enough to crack the tubing.

Water is probably the Best liquid heat conducter there is.

Other liquds or adatives are often used in place of water due to there anti-corrosive or anti-freeze properties, but they don't conduct heat any better than water. You'd need to find a liquid that is denser than water in order to conduct heat better.

Now that I think about it, Mercury might be a better conductor than water but since it won't wet with most materials, it'd probably act as an insulator unless you made your radiator tubing and "water block" out of gold.