|
|
Water Block Design / Construction Building your own block? Need info on designing one? Heres where to do it |
Thread Tools |
07-03-2005, 06:59 PM | #1 |
Cooling Savant
Join Date: Feb 2004
Location: W. Sussex, UK
Posts: 329
|
Combining liquid metal with water/liquid cooling?
I was just pm'ed a link to saphires liquid metal cooling thingy and got an idea.
Could liquid metal be combined with water/other liquid? The reson it would be good: Water has high specific heat, low conductivity Metal has low specific heat, high conductivity. It would be possible to get avery high amount of surface area. In theroy this would possibly be the best block ever.. The problem would be containing the liquid metal (which would probably be mercury). for the liquid metal systems, the pump is just electromagnets round the tube, to push the coolant round the system. Electromagnets could be used to contain the mercury, and keep it in contact with the base, preventing most if not all of it getting washed round the system. Magnets around the system would filter/reclaim any mercury that was moved, and return it to the blocks. If this would work, then its probably a LONG time before any company invests in it for reseach and it couldnt be done without the proper equipement. Acctually I think this could be done now, with current tech, the only limitation being it would have to use copper pipe all the way round, braized if possible. No flexible parts and a specially designed pump. so couldnt be used in any PC, because no flexible tubing.. but a prototype couldbe made to test the theory.
__________________
Waterblocks |
07-06-2005, 09:54 PM | #2 |
Cooling Neophyte
Join Date: Apr 2005
Location: San Ho!!
Posts: 1
|
I am totally new here, but I have to say this place is very informative with many skilled and knowledgable members.
The system you are talking about would probably perform very well indeed, but I myself probably would not venture in that direction unless someone guranteed me it would never leak. Messing with water on electrical compenents is one thing, but mercury and your health is another. All I am saying is I would not risk my health over a few MHz! But anyway, I too have been very interested in the proposal made by Sapphire Tech. Right away I thought about the implementation of the cooling device on a CPU. The use of Jets and whatnot probably would not longer be needed and low resistance like the Swiftech blocks would be the target because of moving a liquid with a, I am assuming, much higher viscousity than water. I also imagine that one single 120mm radiator would be more than engough to cool the hottest CPU with that cooling. Can't wait to see what the future holds, Bryan D. |
07-06-2005, 11:24 PM | #3 |
Cooling Savant
Join Date: Dec 2004
Location: Orlando, Florida
Posts: 383
|
Thermal barriers is a problem here, just like cooling a heatpipe with water. Heat slows down when it is transferred from one material to another.
If you are going to cool liquid metal with water... what's the point of containing the liquid metal in the block? It just doesn't seem efficient to me. For a system of this complexity and cost, jet impingement would probably offer better performance. Why not just have a liquid metal coolant loop with a larger radiator? Personally I'd like to experiment with liquid metal. It could potentially eliminate a very expensive part of watercooling: the pump. |
07-13-2005, 03:46 PM | #4 |
Cooling Savant
Join Date: Feb 2004
Location: W. Sussex, UK
Posts: 329
|
I dont think that will be the problem here, as the mercury is liquid, so will be very different to water hitting a solid metal.
Your probably right about complexity.. but then sapphire found a way round it. Whats to stop this idea being used with jet impingement? I think mercury would be much more expensive than a pump though (O/T) Also I dont quite understand why heatpipe coolers have many sealed tubes, and never have a waterblock like thing, filled with coolant, with the pipes connected to it... wouldnt it be work a lot better?
__________________
Waterblocks |
07-27-2005, 01:23 PM | #5 |
Cooling Savant
Join Date: Jul 2005
Location: california
Posts: 429
|
The metal used as the coolant is gallium and not mercury as most people think. It is non toxic and easy to obtain but it can't be that easy to setup otherwise sapphire jump on it and build several new cooling setups and dominant the cooling market.
|
07-27-2005, 03:44 PM | #6 |
Cooling Neophyte
Join Date: Jul 2005
Location: CALIFORNIA
Posts: 50
|
I still think jet impingement blocks would perform better than liquid metal, but i'm still interested in liquid metal due to its silence. Yes, we are all silence enthusiasts invading this forum from SPCR!
|
07-27-2005, 09:49 PM | #7 | |
Cooling Savant
Join Date: Dec 2004
Location: Orlando, Florida
Posts: 383
|
Quote:
Frosty tech recently reviewed it here: http://www.frostytech.com/articlevie...articleID=1645 |
|
07-28-2005, 12:41 AM | #8 |
Cooling Savant
Join Date: Jul 2005
Location: california
Posts: 429
|
http://www.xbitlabs.com/images/news/...up_diagram.gif
http://www.nanocoolers.com/ is creator of the sapphire cooling Heres a pic of the setup and the nanocoolers provides some more info on what they do. Looks like it is using a tec and using the liquid metal as a coolant would be a good idea. Technology is still young so has a lot of maturing to do. |
Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
Thread Tools | |
|
|