Neg. Pressure Watercooling

Seyeklopz · 10-20-2004, 02:45 PM

Has anyone setup a watercooling system with negative pressurized air in the reservoir? It would be a good idea since any leaks would involve air going in, not water coming out. It may have other advantages such as removing dissolved gases from the water. To prevent bubbles in the lines, the water could be exposed to vacuum before filling the system. Could also setup a kill switch that cuts power to the PC when the pressure gets too high( from a leak )

Only hardware difference would be hoses that won't collapse. You'd probably have to make sure the pump's motor housing is completely sealed. Depending on how low the pressure is, it might suck air through to rotor seal. You would need a fitting to suck the air out of.

bobkoure · 10-20-2004, 03:08 PM

Assuming a sealed system, pumps should continue to work, even though they're notably bad at suction.
However...
- you'd be reducing the boiling point of water and increasing the likelihood of cavitation in the pump
- having dealt with vacuum systems, it's actually a lot easier to keep water (especially the low pressure water we use) in than keeping air out
- beats me if this'd reduce waterblock heat transfer efficiency. there may be a relationship between the absolute force with which a water molecule is forced against a baseplate and the amount of heat transferred to it - but only some waterblocks seem to depend on this.
- I'm guessing there's a major effect that's right in front of me that I'm missing...

Why not set up a system and see - let us all know what you find. Might as well add some of those anti oxidants you were asking about earlier to see what they'd do, too

bigben2k · 10-20-2004, 03:08 PM

The impact won't be noticeable.

The effort(s) involved in maintaining a vacuum, far outweighs the benefit. The hoses would have to be solid (i.e. copper). The standard barbs won't hold a vacuum, so it'd be a custom setup, all the way.

It's best to just use a good airtrap.

Nice try though

Seyeklopz · 10-20-2004, 03:16 PM

OK, I'll give it a try... Just need some automotive vacuum hose, and different fittings. Copper pipes Ben?

bigben2k · 10-20-2004, 06:40 PM

Why not?

http://wc101.com/articles/coppertubing.php

ToasterIQ2000 · 10-21-2004, 10:58 PM

I run large passive (fanless & silent, low flow -- no OC or award winning temperatures achieved) radiators. There is enough coolant in the loop that the thermal expansion and contraction can swell silicon hose.
To compensate for this, I put a soft hose ( a bicycle tube ) up from the bottom into my reservoir, to let the coolant expand and contract to air pressure. One side effect of this was that the coolant was (excuse my colloqual visualization here ) always trying to leak out this air / liquid hole in the bottom. If such an expansion / contraction membrane is at the bottom, a pin hole anywhere in the coolant loop would allow air into that hole & cause the expansion / contraction baloon to shrink.
Now I'll admit that 30-40 cm of negative head is awfully little compared to even a moderate pump and can be completely overpowered by the pump. ( I run an Eheim 1046 ) Make sure the pump pushes into the rigid radiator and draws from the softer hosed half of the loop.
I have caught leaks as little streams of bubbles through clear tubing & air buildup at high points & natural bubble traps without wetting my pcb s.

10-20-2004, 02:45 PM	#1
Seyeklopz Cooling Neophyte Join Date: Jun 2003 Location: Saint John, New Brunswick, Canada Posts: 94	Neg. Pressure Watercooling Has anyone setup a watercooling system with negative pressurized air in the reservoir? It would be a good idea since any leaks would involve air going in, not water coming out. It may have other advantages such as removing dissolved gases from the water. To prevent bubbles in the lines, the water could be exposed to vacuum before filling the system. Could also setup a kill switch that cuts power to the PC when the pressure gets too high( from a leak ) Only hardware difference would be hoses that won't collapse. You'd probably have to make sure the pump's motor housing is completely sealed. Depending on how low the pressure is, it might suck air through to rotor seal. You would need a fitting to suck the air out of.

10-20-2004, 03:08 PM	#3
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	The impact won't be noticeable. The effort(s) involved in maintaining a vacuum, far outweighs the benefit. The hoses would have to be solid (i.e. copper). The standard barbs won't hold a vacuum, so it'd be a custom setup, all the way. It's best to just use a good airtrap. Nice try though __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

10-20-2004, 06:40 PM	#5
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Why not? http://wc101.com/articles/coppertubing.php __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

10-21-2004, 10:58 PM	#6
ToasterIQ2000 Cooling Neophyte Join Date: Oct 2004 Location: Gorky Island, WA USA Posts: 22	I run large passive (fanless & silent, low flow -- no OC or award winning temperatures achieved) radiators. There is enough coolant in the loop that the thermal expansion and contraction can swell silicon hose. To compensate for this, I put a soft hose ( a bicycle tube ) up from the bottom into my reservoir, to let the coolant expand and contract to air pressure. One side effect of this was that the coolant was (excuse my colloqual visualization here ) always trying to leak out this air / liquid hole in the bottom. If such an expansion / contraction membrane is at the bottom, a pin hole anywhere in the coolant loop would allow air into that hole & cause the expansion / contraction baloon to shrink. Now I'll admit that 30-40 cm of negative head is awfully little compared to even a moderate pump and can be completely overpowered by the pump. ( I run an Eheim 1046 ) Make sure the pump pushes into the rigid radiator and draws from the softer hosed half of the loop. I have caught leaks as little streams of bubbles through clear tubing & air buildup at high points & natural bubble traps without wetting my pcb s. Last edited by ToasterIQ2000; 10-21-2004 at 11:39 PM. Reason: oops wrong key

10-20-2004, 03:08 PM	#2
bobkoure Cooling Savant Join Date: Feb 2003 Location: USA - Boston area Posts: 798	Assuming a sealed system, pumps should continue to work, even though they're notably bad at suction. However... - you'd be reducing the boiling point of water and increasing the likelihood of cavitation in the pump - having dealt with vacuum systems, it's actually a lot easier to keep water (especially the low pressure water we use) in than keeping air out - beats me if this'd reduce waterblock heat transfer efficiency. there may be a relationship between the absolute force with which a water molecule is forced against a baseplate and the amount of heat transferred to it - but only some waterblocks seem to depend on this. - I'm guessing there's a major effect that's right in front of me that I'm missing... Why not set up a system and see - let us all know what you find. Might as well add some of those anti oxidants you were asking about earlier to see what they'd do, too

10-20-2004, 03:16 PM	#4
Seyeklopz Cooling Neophyte Join Date: Jun 2003 Location: Saint John, New Brunswick, Canada Posts: 94	OK, I'll give it a try... Just need some automotive vacuum hose, and different fittings. Copper pipes Ben?

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