Heat-pipe waterblock

Cova · 08-20-2002, 11:43 AM

Surfing around today I came across a few interesting looking heatsinks, and I can't help but ponder whether we could use this technology somehow with water-cooling. If we had a similar base as used in the heatsinks shown in the linked page, but a piece of tubing over the heat-pipes (leave about 1/4" gap around the heatpipe) and flowed water down over the heat-pipes, collected it around the base of the block (have all the outter-tubes end a bit short connected to a collector base around the block itself) and then send that water back to the rad/pump.

Maybe this isn't the best way to water-cool the heatpipes. But it seems as though most of the discussion around here currently centers on the most effective way to get heat from core to water as fast as possible. If this moves heat away from the core to a large heat exchanger faster for air-cooling, could we not use the technology somehow but use water to keep the hot side of the heat-pipes even cooler. It wouldn't be much below, but if the hot end of the heat-pipe was cooled to very near ambient, could the block on the core (doubtful the core itself) go sub-ambient? I'm thinking/hoping that with a well designed water-cooled heat-pipe system that we may start getting core temps within 5C of ambient under load.

And..., before I forget..., the link.
http://www.frostytech.com/articlevie...articleID=1193

bigben2k · 08-20-2002, 12:04 PM

Nice!!! I'm going to file THIS LINK for posterity. It's got the most amount of info (although still not complete) about making your own phase change unit.

jtroutma · 08-20-2002, 04:30 PM

Actually I have been thinking of how to encoorperate a heatpipe/watercooled block for some time now. The main problem is that heatpipes work best (again) when the difference between the hot side and the cold side is greater.

Heat pipes are simply sealed tubes with some sort of liquid sealed inside in a vacuum. The tubes are not filled completely on purpose as well. When one side of the heat pipe gets hot, it evaporates the liquid and moves the heat to the top or other side of the heatpipe (which is naturally cooler) where it condenses back into a liquid again.

What we would have to do is locate a liquid that boils at a lower temperature than what most heatpipes are made to boil at and then keep the temp lower with our water setups.

For example, I believe that the new Shuttle Mini cases use H20 heatpipes to keep the P4s cool. Well H20 in a vacuum will probably begin to boil off at about 47C. What we would need is a liqid that would boil off at say 20-25C and then our watercooling setups would need to keep the hot side at about 30-32C.

Correct my numbers or thoughts if necessary but I believe that is the way to go in order to be able to use a good heatpipe setup with watercooling.

#Rotor · 08-20-2002, 04:38 PM

so what is the two water-carrying pipes that goes into a waterblock, non other than two "heat-pipes", really....

granted they are not self propelling, but they do have the ability to carry a significantly larger amount of heat, than what the closed loop heatpipe will be able to carry, simply because they are woking with water as the transfer medium.

Roscal · 08-20-2002, 05:46 PM

A waterbloc (Aquacomputer) using heat pipes is already in project

Barbs are hidden but there are the holes to plug them into the bloc and the 2 grey tubes are heat pipes.

Quickmcj · 08-21-2002, 03:48 AM

Is that the EVO ???

It looks like besides, the two grey tubes....

Havnt. seen that pic at aqua-computer forum......

Where do you have it from ?

Roscal · 08-21-2002, 04:34 AM

Yes it's an EVO with heat pipes (project). It was a news on a french web site

I don't know if this bloc is already realized at this time..

Brians256 · 08-21-2002, 05:43 PM

OK, I have to swing into this because it is so interesting. BigBen2K, that has to be one of the most interesting sites I've read in the last 6 months. Thanks!

As for heat pipes, I think you are barking up the wrong tree. Yes, it would be good to move heat from a small area to a larger area so that the cooling method requires less vigor. However, I don't believe that the cost effectiveness of a heat pipe can compare to the simple changes needed to increase water velocity and decrease the skin effect. Note: please prove me wrong, as I'm just spouting my theories instead of experimentally proven fact.

A heatpipe should have a hard time competing against a moving body of water which is close to the surface of the CPU. The higher the water velocity (and the thinner the skin effect), the thinner you can make the bottom wall of the waterblock.

I just think that it's easier to fiddle with getting a better pump, larger diameter hose, lower restriction radiator and so on. I read a web page saying that heat pipes were really finicky and not all that good at moving heat compared to even simple water cooling close to the die. Remember, every extra layer we add to the coolant system means an extra layer of resistance to the heat flow (unless you add peltiers or other active heat movement technologies).

Brians256 · 08-21-2002, 05:57 PM

Dang. Sorry for not noticing that your original link, Cova! I didn't notice that BigBen2k's link was from the 3rd page of the article you referenced. Doh!

#Rotor · 08-21-2002, 08:11 PM

a heat-pipe, less than 2 inches long, is that not kinda like missing the point all together..... I sugest a chat with TN from OCN... he is the residual heatpipe expert of the world.. if not, then well... he should be...

he's been heatpiping his xeons since long ago...

link1

link2

08-20-2002, 11:43 AM	#1
Cova Cooling Savant Join Date: May 2002 Location: Canada Posts: 247	Heat-pipe waterblock Surfing around today I came across a few interesting looking heatsinks, and I can't help but ponder whether we could use this technology somehow with water-cooling. If we had a similar base as used in the heatsinks shown in the linked page, but a piece of tubing over the heat-pipes (leave about 1/4" gap around the heatpipe) and flowed water down over the heat-pipes, collected it around the base of the block (have all the outter-tubes end a bit short connected to a collector base around the block itself) and then send that water back to the rad/pump. Maybe this isn't the best way to water-cool the heatpipes. But it seems as though most of the discussion around here currently centers on the most effective way to get heat from core to water as fast as possible. If this moves heat away from the core to a large heat exchanger faster for air-cooling, could we not use the technology somehow but use water to keep the hot side of the heat-pipes even cooler. It wouldn't be much below, but if the hot end of the heat-pipe was cooled to very near ambient, could the block on the core (doubtful the core itself) go sub-ambient? I'm thinking/hoping that with a well designed water-cooled heat-pipe system that we may start getting core temps within 5C of ambient under load. And..., before I forget..., the link. http://www.frostytech.com/articlevie...articleID=1193

08-20-2002, 12:04 PM	#2
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Nice!!! I'm going to file THIS LINK for posterity. It's got the most amount of info (although still not complete) about making your own phase change unit. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

08-20-2002, 04:30 PM	#3
jtroutma Cooling Savant Join Date: Aug 2001 Location: SLO, CA Posts: 837	Actually I have been thinking of how to encoorperate a heatpipe/watercooled block for some time now. The main problem is that heatpipes work best (again) when the difference between the hot side and the cold side is greater. Heat pipes are simply sealed tubes with some sort of liquid sealed inside in a vacuum. The tubes are not filled completely on purpose as well. When one side of the heat pipe gets hot, it evaporates the liquid and moves the heat to the top or other side of the heatpipe (which is naturally cooler) where it condenses back into a liquid again. What we would have to do is locate a liquid that boils at a lower temperature than what most heatpipes are made to boil at and then keep the temp lower with our water setups. For example, I believe that the new Shuttle Mini cases use H20 heatpipes to keep the P4s cool. Well H20 in a vacuum will probably begin to boil off at about 47C. What we would need is a liqid that would boil off at say 20-25C and then our watercooling setups would need to keep the hot side at about 30-32C. Correct my numbers or thoughts if necessary but I believe that is the way to go in order to be able to use a good heatpipe setup with watercooling. __________________ Athlon64 X2 4200+ @ 2.5Ghz (250FSB x 10) OCZ VX 1GB 4000 @ 250FSB (6-2-2-2 timmings) DFI LANParty nForce4 Ultra-D SCSI Raid 5 x (3) Cheetah 15K HDDs LSI Express 500 (128MB cache) OCZ PowerStream 520W PSU ATI X850XT PE (Stock) DTEK WhiteWater + DTEK Custom Radiator Eheim 1250

08-20-2002, 04:38 PM	#4
#Rotor Cooling Savant Join Date: Feb 2002 Location: Dione, sector 4s1256 Posts: 852	so what is the two water-carrying pipes that goes into a waterblock, non other than two "heat-pipes", really.... granted they are not self propelling, but they do have the ability to carry a significantly larger amount of heat, than what the closed loop heatpipe will be able to carry, simply because they are woking with water as the transfer medium. __________________ There is no Spoon....

08-20-2002, 05:46 PM	#5
Roscal Cooling Savant Join Date: Apr 2002 Location: North of France Posts: 198	A waterbloc (Aquacomputer) using heat pipes is already in project Barbs are hidden but there are the holes to plug them into the bloc and the 2 grey tubes are heat pipes. Last edited by Roscal; 08-20-2002 at 05:49 PM.

08-21-2002, 03:48 AM	#6
Quickmcj Cooling Savant Join Date: Jun 2002 Location: Denmark Posts: 109	Is that the EVO ??? It looks like besides, the two grey tubes.... Havnt. seen that pic at aqua-computer forum...... Where do you have it from ?

08-21-2002, 04:34 AM	#7
Roscal Cooling Savant Join Date: Apr 2002 Location: North of France Posts: 198	Yes it's an EVO with heat pipes (project). It was a news on a french web site I don't know if this bloc is already realized at this time..

08-21-2002, 05:43 PM	#8
Brians256 Pro/Staff Join Date: Oct 2001 Location: Klamath Falls, OR Posts: 1,439	OK, I have to swing into this because it is so interesting. BigBen2K, that has to be one of the most interesting sites I've read in the last 6 months. Thanks! As for heat pipes, I think you are barking up the wrong tree. Yes, it would be good to move heat from a small area to a larger area so that the cooling method requires less vigor. However, I don't believe that the cost effectiveness of a heat pipe can compare to the simple changes needed to increase water velocity and decrease the skin effect. Note: please prove me wrong, as I'm just spouting my theories instead of experimentally proven fact. A heatpipe should have a hard time competing against a moving body of water which is close to the surface of the CPU. The higher the water velocity (and the thinner the skin effect), the thinner you can make the bottom wall of the waterblock. I just think that it's easier to fiddle with getting a better pump, larger diameter hose, lower restriction radiator and so on. I read a web page saying that heat pipes were really finicky and not all that good at moving heat compared to even simple water cooling close to the die. Remember, every extra layer we add to the coolant system means an extra layer of resistance to the heat flow (unless you add peltiers or other active heat movement technologies).

08-21-2002, 05:57 PM	#9
Brians256 Pro/Staff Join Date: Oct 2001 Location: Klamath Falls, OR Posts: 1,439	Dang. Sorry for not noticing that your original link, Cova! I didn't notice that BigBen2k's link was from the 3rd page of the article you referenced. Doh!

08-21-2002, 08:11 PM	#10
#Rotor Cooling Savant Join Date: Feb 2002 Location: Dione, sector 4s1256 Posts: 852	a heat-pipe, less than 2 inches long, is that not kinda like missing the point all together..... I sugest a chat with TN from OCN... he is the residual heatpipe expert of the world.. if not, then well... he should be... he's been heatpiping his xeons since long ago... link1 link2 __________________ There is no Spoon....

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)