Extreme cooling

[Citrus]

hi guys ,
have any 1 tried before nitrogen cooling ???
in singapore i cant get my hands on liquid nitrogen wat other extreme cooling can u guys reckon ???

[Phil]

There have been a few people that have used it, and also dry ice to get very high one off overclocks, mostly mad head japanese overclockers though. It's a one off thing and the hardware is usually destroyed in the process, they overclock real high until it evaporates, I've never heard of someone making a usable permanent liquid nitrogen cooling system.

[Citrus]

u saw www.octools.com ??? tha was way cool !!!
my problem is to get someone to get a maze2 and sell to me !

[Random Nonsense]

LN2 is dangerous, it literally pulls the CPU apart. the coldest u can go and remain "safe" is -40 celsius or thereabouts

[panhead49]

think of the o/c though............would be fun while it lasted.......i would look into water cooling i think.....

[Random Nonsense]

hehe i not a fan of 20 mins then fzzzt dead CPU, i prefer a rock steady moderate overclock

[cloasters]

There was a wonderful site in Japan I came across mebbe two and a half years ago. Folks in Japan can take their LN0 cooling very seriously. This man got C300A's up to 800 Mhz, using an aluminum block that was custom-cast for him.

His site also featured traditional Japanese flower arrangements, and was written in excellent English as well as Japanese.

Like an idiot, I didn't save the URL.

I'd never mess with liquid nitrogen, personally. AFAIK, no-one has created a system that's "perma-cooled" by the stuff.

[panhead49]

i agree, with todays cpu's there is no need to go to such an extreme.........but as said.....they go about there o/cing as an art form...to the extreme.......it would be fun though......dont you think.....

[Phil]

Originally posted by panhead49
i agree, with todays cpu's there is no need to go to such an extreme.........but as said.....they go about there o/cing as an art form...to the extreme.......it would be fun though......dont you think.....

I can think of better ways of having fun with the money wasted on killing the equipment and the ln2.
Like inflatable furniture for instance.

[panhead49]

i think it would be fun running LN2 for an extreme o/c "while" playin on the furniture......lol........

come on ....you can allways get another setup ......just dont melt the blow up "coffee table"....

[Phil]

Originally posted by panhead49
i think it would be fun running LN2 for an extreme o/c "while" playin on the furniture......lol........

come on ....you can allways get another setup ......just dont melt the blow up "coffee table"....

Some of us can't afford both, and if it was a choice between inflatable furniture or another setup, I think it's safe to say what most people would go for, am I right folks?

[panhead49]

uhhhhhh......let me think.......where are we,..a o/cers web site.....what would most people go for here .......furniture....or computer setup for o/cing.......i wonder.
lol

[Phil]

Originally posted by panhead49
uhhhhhh......let me think.......where are we,..a o/cers web site.....what would most people go for here .......furniture....or computer setup for o/cing.......i wonder.
lol

Remember it's inflatable furniture, not bog standard stuff!
You gots to be comfy don't you

[panhead49]

well ok.....you have a point.........if the remote to the 65" hdtv goes with it.....lol

[Citrus]

hmmm. i was thinking of changing a heatsink into a waterblock the fins act as the heat dispensers

[Aedan]

Yeah, you could do that. A standard airbased heatsink has a lot more surface area than most waterblocks.

I think a combination of all these ideas wouldn't be bad. Phil gets his inflatable furnature, Citrus gets his liquid cooled system. I figure that inflatable furnature is a great reservoir. You'd need to make sure that the piping and joints could take the pressure when you sit on your furnature though.

Brings a new meaning to warm and wet huh?

AidanII

[Kaitain]

A standard airbased heatsink has a lot more surface area than most waterblocks.

On finned surfaces it's not just about surface area. Air-cooled fins are designed with the cooling properties of air in mind - they typically need much higher surface areas because solid to air heat transfer is so slow. Solid to water on the other hand is much faster, so the surface area requirements are much less - this is how waterblocks manage to have so little.

Having too much surface area won't help, but it shouldn't hinder the effectiveness of the block much either.

[Citrus]

Originally posted by Kaitain


On finned surfaces it's not just about surface area. Air-cooled fins are designed with the cooling properties of air in mind - they typically need much higher surface areas because solid to air heat transfer is so slow. Solid to water on the other hand is much faster, so the surface area requirements are much less - this is how waterblocks manage to have so little.

Having too much surface area won't help, but it shouldn't hinder the effectiveness of the block much either.

meaning the surface area of the heatsink shouldn't be too big ???
i was thinking of using my coolerMaster heatsink

[Aedan]

Originally posted by Citrus

meaning the surface area of the heatsink shouldn't be too big ???
i was thinking of using my coolerMaster heatsink

No.. Meaning that the surface area can be too big, it's just that the extra surface area won't be doing anything. It won't hurt, but it won't do anything either. I believe that you can actually reduce the height of the fins and get the same level of cooling, but Kaitain can confirm that, being the closest we have to an expert on all things fluid flowing!


AidanII

[Kaitain]

I believe that you can actually reduce the height of the fins and get the same level of cooling,

Which CoolerMaster? When I get a few minutes, I'll put its details into my little model here and see what theoretical performance you should get. If you know what pumping rate you can manage (litre/min or gal/hr, I'm not choosy), that'd be helpful too (otherwise I'll guess one).

I have simulations for waterblocks that have been calculated to have tiny fin dimensions - the limiting factors are actually pressure drop and flow rate.

You need a certain water speed (m/s) to get effective cooling. If your waterblock is smaller, then a lower capacity (m^3/s) will provide that, BUT the amount of effort needed to force it through the block will be greater. You'll notice this as a pressure drop across the waterblock. Now, if you assume that you'll get a pressure of 0 bar.g at the outlet, then you can reasonably expect a pressure of >0 bar.g at the inlet. Make the block too small and you get a huge pressure at the inlet - enough to pop seals maybe. Make the block too big and you need to pump water at stupid rates to get the speed up.

It becomes a trade-off, and the deciding factor is your budget: how much can you afford to spend on metal? That determines the surface area you can afford. How much can you afford to spend on sealing your system against pressure? That's your upper pressure limit. How big a pump can you afford? That's your upper capacity limit. The best fit between them is the design you get.

The surface area curve flattens out - beyond a point, the benefit you get per cm^2 of fin area are pretty negligible. The point to aim for is the break-in-curve, where it just starts to flatten out. Then you're getting the best value for money out of the metal you've got.