Is water-cooled THAT good?

[cwiz]

Reynolds number=(density*velocity^2*hydrolic-diameter)/viscosity.

Prandlt number=viscosity*Cp/Thermal conductivity

Prandlt, Nusselt, reynolds, schmidt, as well as many other numbers interact algebraicly

Heat transfer increases proportionally with Reynolds number, but also with viscosity.

its a tricky topic as to which to use, distilled water or antifreeze. the distilled water flows faster because it has a lower viscosity. the lower viscosity raises the reynolds number of the fluid across the waterblock(hydrolic diameter is associated with the geometry of the waterblock). reynolds number increases inversly with viscosity, but heat transfer increases directly with viscosity and reynolds number. Antifreeze, although handicaped by its lower reynolds number, it's higher viscosity causes more friction with the radiator and tubing walls causing greater heat transfer out of the system.

it is quite complicated... to learn more, google fluid mechanics, convection heat transfer, reynolds number, nusselt number, prandtl number... it goes on.


the Cp, or specific heat capacity, of water is higher than that of antifreeze (ethylene glycol/water mixture) and so the water is capable of carrying more heat in it with less of a temperature change of the fluid. and as we all know, heat transfer is driven by temperature differences! this means that you may need a larger reservior if you are going to use antifreeze.

as you can see, the same things that advantage one in a certain respect will also detract from it in another. it is quite a design problem... i gotta stop typing and get back to work... designing heat exchangers, curiously enough

 

[dwarf]

you definetely do NOT want to run regular anti-freeze/coolant made with ethylene glycol, as said, the viscosity is significantly higher than what this equipment was made to run, and can really cause problems with buildup, especially in the more popular/higher performing impingement water blocks.

There simply is not a better solution than distilled/deionized water at the moment, there have been numerous other things tested, and they all had significant drawbacks.

I would also recommend against MCT-5, besides it being a lower performing liquid to use in the loop, the main point of MCT-5 is that "it won't fry your equipment" which just isn't true.. MCT-5 is just as likely to fry your hardware as Distilled/DI, there have been casualties and survivor stories on both sides, the difference is one of these doesn't cost 100+ dollars to fill most loops.

 

[RockyZ]

http://svc.com/fxphprf32.html

Finally decided on this coolants. I don't feel like going around looking for dyes and the perfect coolant etc. This seems like a good choice.

Going to wash out everything in my loop with distilled water and vinegar. Run the system with distilled water. Drain it, and pour in the coolant to make sure my loop is clean enough.

I was thinking of getting a temperature indicator for my loop. Would the water be as hot as the CPU? (Only thing I am cooling in my loop) I know how inaccurate the mobo indicators are.