Custom water cooling using vapor chamber/heatpipes?

[_larry]

Hey guys,
Vapor chambers and heatpipes are usually hollow, correct? Would it be possible to drill out two ends of those and pump liquid through there? Obviously you would need some type of fluid that won't corrode the copper. The 7950 Windforce 3 that I have has a fairly nice heatsink that I could experiment on. I have been thinking about getting an Arctic cooler anyway if my theory doesn't hold any water. (pun not intended LOL)

 

[Bo$$]

Hey guys,
Vapor chambers and heatpipes are usually hollow, correct? Would it be possible to drill out two ends of those and pump liquid through there? Obviously you would need some type of fluid that won't corrode the copper. The 7950 Windforce 3 that I have has a fairly nice heatsink that I could experiment on. I have been thinking about getting an Arctic cooler anyway if my theory doesn't hold any water. (pun not intended LOL)

Sometimes they solder the ends into the heat sink base plate rather than using them as a continuous tube. you may face a trouble there, but again it could work fine..

I'm sure it would be fine but since i've never attempted this i can't help you!

Hopefully someone more experienced posts here

 

[buildzoid]

The insides of vapor chambers and heatpipes is usually coated in cintered metal so you will have major issues with that. you'd probably be better of taking a vapor chamber with fins soldered to it and then running water through the fins instead of the chamber. Basically your problem is that the inside of both vapor chambers and heat pipes has very little surface area.

 

[Sasqui]

From what I've read, heatpipes have a wick, so that they can be at any orientation without having to worry about the condensed heat exchange fluid dripping back to the heat source.

http://en.wikipedia.org/wiki/Heat_pipe

 

[TheBrainyOne]

The insides of vapor chambers and heatpipes is usually coated in cintered metal so you will have major issues with that. you'd probably be better of taking a vapor chamber with fins soldered to it and then running water through the fins instead of the chamber.

This.

Basically your problem is that the inside of both vapor chambers and heat pipes has very little surface area.

I think you are wrong. Heat pipes and VCs have lot more surface area than simple tubes. But this surface area comes at the cost of a lot of resistance, which would make the solution impractical.

 

[lilhasselhoffer]

Let's wait a moment, and go back to theory here.

A heat pipe is not a simple vacuum tube. A Vacuum tube would be an excellent insulator. Heat pipes have a low pressure environment because they are modifying the heat required to boil the fluid inside them. Basically, a heat pipe works by heating one end until the fluid inside vaporizes, pumping the vapor to a cooler area where it can condense, and then pumping the condensed fluid back into the heated area (this may be a misnomer, as capillary action generally brings the fluid back rather than an actual pump.

The reason that heat pipes are used, instead of pumped liquid, is because the energy required to phase shift a material is huge when compared to changing that material to another temperature within a phase. To the layman, bringing water from 50C to 51C requires less energy than going from 100C liquid to 100C steam.

So, I have to ask the question of what your end goal is. Assuming that you want to modify an air cooler to be a hybrid air and water cooler, you'll actually be spending more money to cool less efficiently. A radiator will cool much more efficiently than the fins of an air cooler.


You can technically slice open the end of a heat pipe, remove the contents, and pump water through it. The problem is simply that you aren't going to be as efficient as the heat pipe was, and it's less actual cooling than a proper water block. It would be worth it if you've got a free and damaged heat pipe setup, but if you're modifying something that already works I'd recommend you give this idea up immediately.

 

[Sasqui]

The reason that heat pipes are used, instead of pumped liquid, is because the energy required to phase shift a material is huge when compared to changing that material to another temperature within a phase. To the layman, bringing water from 50C to 51C requires less energy than going from 100C liquid to 100C steam.

AKA latent heat absorbed by coolant to vaporize it, and latent heat expelled by coolant to condense it.

Virtually all heat pipes are passive refrigererators. The heat source (CPU/GPU) provides the heat to vaporize the coolant, the air over the fins provide the cooling to re-condense.

If you don't have gravity to get the condensed coolant back to the heat source, it won't work. That's why they devised the "wick"

 

[d1nky]

Or copy Asus' new hybrid design (poseidon) a waterblock within an air (vapour) block?!

https://www.asus.com/Graphics_Cards/ROG_POSEIDONGTX780P3GD5/

 

[lilhasselhoffer]

Or copy Asus' new hybrid design (poseidon) a waterblock within an air (vapour) block?!

https://www.asus.com/Graphics_Cards/ROG_POSEIDONGTX780P3GD5/

Note that this is not a hybrid design. What you have cited is a cooling system designed to run on air, or on water. It isn't a modification that uses heat pipes as water cooling conduits.

 

[d1nky]

Note that this is not a hybrid design. What you have cited is a cooling system designed to run on air, or on water. It isn't a modification that uses heat pipes as water cooling conduits.

ive just read about 3 posts saying how inefficient 'using vapour tubes as water pipes' would be, so i merely offered an alternative modification.

all that is needed is a metal shim between air block and die, with some kind of water chamber.

 

[sneekypeet]

There was a kid here at one point that did exactly this. From what I recall, he cut open a Zalman CNPS9500, braised fittings on the end, and pumped water though the cooler since they used one continuous heatpipe that ran through the fins. It is very feasible to do this, you just need a cooler like the Zalman so you aren't fitting to 4/6/8 separated heatpipes.

 

[_larry]

From what I can tell, the heat pipes and vapor chamber stem directly off of the copper part that meets the GPU core. One of the three fans started making bad noises, so I ripped those off and have 2x92mm antec fans strapped to the heatsink. It never gets over 70 C even when overclocked to 1100/1425mhz. It also isn't the quietest solution.. Either way, I would be paying around $80 for another solution unless I got the Gelid cooler which is about $54.

 

[Vario]

Even if it was more efficient, you'd run the risk of a leak. Don't bother. Also not sure an Arctic cooler is any better than the Windforce, the Windforce coolers are as good as air cooling gets, probably just need to repaste it.

 

[_larry]

Even if it was more efficient, you'd run the risk of a leak. Don't bother. Also not sure an Arctic cooler is any better than the Windforce, the Windforce coolers are as good as air cooling gets, probably just need to repaste it.

The Arctic one seems to be a LOT better on paper. 92mm fans instead of the 80mm on the windforce, two more heatpipes, and separate little heatsinks for the memory and VRM.

 

[eidairaman1]

Even if it was more efficient, you'd run the risk of a leak. Don't bother. Also not sure an Arctic cooler is any better than the Windforce, the Windforce coolers are as good as air cooling gets, probably just need to repaste it.

meh, cant OC that high with em, plus they are the least compatible with several other motherboards around