Tiny bubbles are key for future

[FallOutBoyTonto]

News link 1 (http://news.uns.purdue.edu/html4ever/030411.Mudawar.cooling.html)
News link 2 (http://www.scienceblog.com/community/article1470.html&mode=thread&order=0&thold=0)
News link 3 (http://www.xbitlabs.com/news/other/display/20030417174705.html)

They are all talking about the same thing, pump-less liquid cooling. Interesting read. But since I'm just a noob in the cooling world, I came to hear what the pros thoughts are about this new technology.

[jaydee]

Cathar brought something up similar before:
http://forums.procooling.com/vbb/showthread.php?s=&threadid=6416

I have no real comment on it at this point. Non of it makes any logical sence to me.

[Cathar]

When considering these "future cooling technologies" it must be remembered that these guys aren't gunning for the coolest possible CPU's, just keeping a manageable lid on CPU temperatures, being anywhere from 80-100C

[Phant0m51]

Umm...Those temps are quite a bit too hot for my tastes.

[Dunno]

This idea may be worth trying!

Lets say you make a water cooling system, similar to todays setups, that can handle a vacume; with solid pipes and a submersble pump.

Now if you pull a vacume in it so that the water boils just above your ambiant air temp at say 25 deg C.

(Boiling point is proportional to surrounding pressure;See heatpipes)

When the water is heated above that in the block on the CPU it will boil. I am not sure of the physics here but the phase change takes a lot of thermal energy and the bottem line is that water just cannot get hotter than its boiling point! Also its the insulating boundry layer that boils away and is replaced by nice cool water.

Your circuit would have to go CPU - rad - res - pump - CPU so that the bubbles had a chance to condense in the rad rather than causing the pump to cavatate. Also one might have to use a thicker pipe from the WB to the Rad to allow for the extra volume of steam bubbles. The exit from the water block would have to be at the highest point in the block to stop the steam from collecting in it. From there the floe rate should be high enough to transport the bubbles to the rad. It would also prbably be better to have the rad inlet at the bottem, contrary to Convection laws, so that the bubbles can rise up the channels as they condense.

I think in the case mentioned in the links they have it set up so that the rising bubbles are in the direction of the flow so that they help flow and also disrupt the boundry layer.

Also a Rant:
People seem to be ignoring my posts. I am not sure if this is because I am a Noob to these forums or wether its because I'm from South Africa and considered better qualified for hearding goats. I would get a nasty sunburn from hearding goats, just like any other European. (how's that for politically correct?) I have been water cooling since 286's (Direct die) and making water blocks on milling machines and cooling systems ever since! Also my meaningless ramblings are based on what I learned studying Engineering at univarsity. So There!

(Puts soap box away and goes in search of spell checker) :)

[Blackeagle]

Dunno,

Just how would you suggest we would get this vacum in the system? I don't see how this can be done with DIY methods or items purchased at any site I know of.

It would work if this can be done, but not other wise, and that is the catch.

As to why no one has responded to your posts, how many have you put up? This is the first one I"ve seen, although I havn't been posting as much due to work demands of late. And I'm sure your being from S. Africa has had nothing to do with it.

All the same, welcome to Pro Cooling!:)

[Phant0m51]

Ok, one problem with the giant heatpipe scenerio:

Sure, the water will only boil at a certain temp and not go above, and sure, it will take alot of that heat away from the processor. The problem is that, what if the processor DOES get hotter than the water will boil? where is that extra heat going to go?

[Dunno]

Blackeagle

The vacume could be pulled at a refridgeration service the same as they do for fridges.

THX for the welcome:)

Phant0m51

As I understand it when water boils it takes a large amount of energy for the Phase change reaction. More even than the specific heat of water. This means that water boiling of any given surface area removes much more heat than can be done by cooling it with water normally:

The bubble that forms is not an air bubble but a steam bubble that is at low temp due to low pressure.

Also the hot piece of copper is not staying dry; as soon as the steam bubble is big enough it breaks away taking the,now boiled, insulating boundry layer with it. The copper surface is now wetted with nice cool, turbulant water and the whole process starts again.

I'll do more research on this and post.

[trit187]

to get a lower boiling point, ie less pressure would be required, could a different liquid be used? this could reduce the extra design modifications needed.
but just getting water to boil is failrly simple but I don't know if you started introducing vacum conditions to a pump what the effects would do.. I guess it would be dependent on how the vacume was created...
would any liquids be able to fit the criteria that would be viable solutions?
I think it's be cool to at least try and see if it'd be viable

[Dunno]

Originally posted by trit187
to get a lower boiling point, ie less pressure would be required, could a different liquid be used? this could reduce the extra design modifications needed.
but just getting water to boil is failrly simple but I don't know if you started introducing vacum conditions to a pump what the effects would do.. I guess it would be dependent on how the vacume was created...
would any liquids be able to fit the criteria that would be viable solutions?
I think it's be cool to at least try and see if it'd be viable

Dont know about different liquids. Someone memtioned alchol; 80C, possibly refridgerants.

You could use a submersable pump in a container that could take the vacume.

[Phant0m51]

You can also heat up the object you want to vacuum, then seal it and let it cool off. Water expands as it gets warm, taking up more space, then as it cools down there will be lower pressure. Sealing it while it is hot would cause the loop to have a positive pressure, relative to the outside world. But technically it will have the same pressure inside as on the outside, so when it cools it'll have a lower pressure.

I don't know how well that would work, it's just a suggestion.