monglobonglo

Honorable
May 15, 2012
7
0
10,510
I realize that one of the main purposes of a resevoir is to allow for expansion/contraction of the coolant, as well as a "trap" for air bubbles.


Still, most resevoirs are sealed. they have an inlet and an outlet, and a fill port that is sealed tight 99.9% of the time.


with the system sealed like that, any expansion or contraction of the fluid, or any air bubbles that evolve from the liquid (dissolved in the water ----> gas) will cause the pressure inside of the system to rise above ambient pressure.


that increased pressure will push against the tubing and may lead to leaks.


the solution is install some sort of valve in the system.


assuming a typical reservoir, with 3 ports (in, out, and fill), you would install this hypothetical valve in the "fill" hole.


then, any pressure would bleed out, and any negative pressure caused by a contraction of the fluid (possibly due to cold weather or prolonged periods of idle, etc.) would cause the valve to allow air back into the system.


I have searched on the typical websites that specialize in water cooling (dangerden, frozencpu, performancepcs, etc.) and I have not been able to locate anything like this.


there are plenty of "valves," but they are typically designed for control of liquid through the system.


if anyone has any suggestions, I would greatly appreciate it.
 

rubix_1011

Contributing Writer
Moderator
The way that air and water pressurizes is different and the heat being applied to the volume of water isn't going to increase this by very much- likely negligible to the human eye. Watercooling loops don't really operate under any real pressure and there isn't really a need for venting. If there is air in the loop, air will compress based on water expansion, and if there is no air in the loop (which is ideal, but not realistic that air doesn't reside somewhere) the water that expands so minimally that flex in the tubing would likely account for this expansion.

I get where you are going with this, but the temperatures and pressures you are theorizing don't exist in a normal watercooling loop. However, you could force them to happen by severely limiting radiator space and running a high heat load to the loop where the delta-T was incredibly high.