water cooling v. fan\heatsink
On Aug 13, 10:36 am, Frank McCoy wrote:
In alt.comp.hardware.pc-homebuilt Mac Cool wrote:
rickman:
The large size of the radiator would provide lots of
cooling area eliminating the need for a fan. I don't know for sure
how well this would work, but even if it is not cooler than a fan, it
will be nearly silent.
A good quality water pump is nearly silent. If you want to eliminate the
fans, go with an evaporative system like I described to the OP. I don't
believe you will get enough water movement through hoses using convection
to prevent the CPU from overheating. You would need a cylinder of water
sitting atop the CPU and some method, probably a fan, of cooling the water
as it rises.
A. You don't need to restrain yourself to just water.
B. Use heat-pipe methods.
Water is a pretty good coolant with a high specific heat and ready
availability.
How would a hobbiest work with heat pipes other than using a stock
heat sink incorporation heat pipes? I was not aware that you could
construct your own heat pipes in any practical way.
Actually, wouldn't the design I described using convective movement of
water be pretty much like a heat pipe? I guess the difference is that
the water never goes to the liquid phase... at least I hope not!
I did a little reading, and it looks like you could use Butane to form
a closed loop, passive two phase cooling system at just a little over
2 bar. This would give a vapor temperature of about 25C. Of course
it might be necessary to have a slightly higher working temperature to
make the radiator work effectively. Pentane has a boiling point of 36
C which is higher than desired for cooling the CPU, but will make the
radiator more effective and not require a pressurized system. Of
course both of these are flammable, but with Butane a leak would
quickly depressurize the system and the computer would alarm from an
over heat condition. But clearly, this requires a lot more thought...
|