Problems with using fluorinert.

[c.ruel]

What would be the problems if one used fluorinert as the coolant for their liquid cooled system? For example:

What kind of tubing would you have to use? Would norprene be adequate?

What would be the issues with evaporation? How quickly would you have to get it into the tubing before it evaporated?

What kind of protective gear would you have to wear? What would happen if you did inhale it or get it on yourself?

Thanks in advance!

[Waterlogged]

Wiki might be of some use to answer some of your health related Q's. http://en.wikipedia.org/wiki/Fluorinert

As for using it in a loop, I've only ever heard of it being considered for direct HS contact cooling in a closed tank/case, not sure how well it will work in a loop.

[Wolf132]

Paying 100$ + per half gallon for something that's probably not going to outperform 1$ per gallon distilled water ?

[Buckeye]

It can't possibly cool any better then the Rad/Fan combo can remove heat & what your ambient temps are correct ?
So unless you had a better then normal water cooling setup to run this with how can it perform better ?

Am I mistaking how all this works ?

[Petra]

OP, Fluorinert is typically employed in situations where submersion cooling is necessary and/or desired... In a closed loop (non-submerged) cooling system, you'd be better off with distilled water. As for use in a closed loop goes, you may run into problems with Fluorinert's rather low surface tension (read: your setup may leak like a sieve)... rapid evaporation shouldn't be too much of a worry depending on the formulation that you choose, though (FC-72 boils at 133*F, FC-75 boils at 216*F, etc.).

If you plan on using the Fluorinert in a low temperature cooling environment (like chilling the Fluorinert by running it through a copper coil in a dry ice and alcohol bath), then be aware that it has a tendency to gel.

[c.ruel]

I want to create a pretty much maintainence free water cooling setup. Fluorinert is totally inert so no corrosion problems and won't need to be periodically changed like water.

I am also considering mineral oil but I think the performance of that might be dreadful. I am also not certain as to whether it will be a 100% maintainence free either.

[Conumdrum]

Well, there is no 'maintenace free PC watercooling setups unless you buy an H100i etc for your CPU.

It's a hobby and needs care and maintenance. Maybe WC isn't for you then.

Maybe you should head over to http://www.overclockers.com to the WC forum and read the stickies and tons of other helpful articles before you even consider such odd not recommendedliquids.

Mineral oil is not good in a loop and has very poor thermal properties next to distilled water and a biocide.

[mlwood37]

All so it expands and contracts and you need a breather in you system to compensate for this action. All so 1 Kilo is approx 1/2 a Lter and a typial system needs 1 Ltr and i can get it for ?25.00 to ?30.00 per kilo. The better option is Novac as its half as dense and less dangerous to the environment how ever the price increases 2 fold.

[Petra]

All so it expands and contracts and you need a breather in you system to compensate for this action.

I guess I forgot about that property... could probably use a volume compensating diaphragm to take care of that without opening the system to atmosphere.

[rge]

The main problem you will have with fluorinert is worse cpu temps than high end air cooling, and more expense and hassle. And high end air is relatively maintenance free along with better cooling.

High end air cooling, although only 120 mm fan size in surface area, has efficient heat pipes to transfer the heat to fins, before being cooled with air which has poor thermal conductance of .026 w/mK and sp heat 1 kj/kgK.

A 360 rad, uses water with thermal conductance of 0.6 w/mK and spec heat of 4.186 kj/kgK as the transfer medium, to increase surface area to 3 x 120 fan size, before having to then cool water with crappy air. Ie, both air cooling and water cooling use air to cool, but water allows efficient transfer of heat to a much larger surface area, ie like 3x 120 fan size rad before using air to cool, vs 120 fan size of air cooling.

Flourinert in server farms, typically use the fluorinert (non-conductive) to flow through the heat producing computer components, even though flourinert has poor cooling properties relative to water. Flourinert thermal conductance is .06w/mk, so heat will exchange from cpu to fluorinert ~10X more slowly than cpu to water would. Also the specific heat of fluorinert is 1.1 kj/kg/k, so fluorinert temps absorb ~ 1/4 heat of water while temps rising similar amount, and when the circulating fluid is hotter the cpu will be hotter. But server farms and supercomputers make it work by using phase change/chillers for heat exchange with fluorinert, they are not piping it to 360 rads.

But using fluorinert, with its relatively poor spec heat and very poor thermal conductance in place of water as the transfer medium to bring it to the rad to then cool with air, is not going to be any more effective than high end air cooler using heat pipes to transfer heat.