Reservoir for waterchiller?

[aasmaukr]

Hiya, I built a chiller for a 8800GTX and a E6600.




It's insulated and ready for mounting now, but my customer will need a reservoir. What would you recommend? I'd prefer one that's got room for a Eheim 1250 so we won't have to insulate the pump.

It's not going to be very low temp, maybe about 0*C max. load.

Thanks

[Ramaistro]

What about making a reservoir out of pvc tubing in the middle of the coil ?
It is a nice place to place because it will safe you room and with insulation you can insulate the coil and reservoir at the same time.

[epion2985]

Don't use a rez, don't need one, the heat exchanger is rez enough. If you add a rez you add more coolant to the loop, which in turn will take longer to pull down and will have more surface area to loose cold to. In terms or performance I like to use no rez, as there is really no point.

[aasmaukr]

Okay, isn't it going to be kinda difficult to get all the air out of the system then? How would you add coolant?

Also, I doubt the coolant will be below -25*C. What coolant would you have used? Denaturized alcohol (red ethanol) and water?

[aasmaukr]

We put a T-connection right behind the pump and added a 50/50 mix of distilled water and red ethanol.

The chiller pulled down the temp very fast and we let it get down to -27.5*C before we started dual prime on the E6600. Surprisingly the temp didn't go any higher than -26.5*C.

I'll post some pics later

[n00b 0f l337]

I would have put a resevoir in the middle of your coax...
T-lines good too though.

[The Water Dog]

i am extremely pleased with my new chiller, pictures comes also when all is finished. Have to make a nice house to the chiller
thanks to aasmaukr for all help

[epion2985]

Sounds good. As for coolant methanol/water is a little better then ethanol/water, if you are looking to tweak.

[serialk11r]

Sounds good. As for coolant methanol/water is a little better then ethanol/water, if you are looking to tweak.

Thermal conductivity is 0.1 W/mK more well that's 4%...

[epion2985]

Thermal conductivity is 0.1 W/mK more well that's 4%...

spare us your analysis...

The conductivity is better, so is the capacity. Also it has a lower freezing point, meaning you need less of it, meaning you can have more water in the mix, meaning more conductivity and capacity across the coolant.

[serialk11r]

spare us your analysis...

The conductivity is better, so is the capacity. Also it has a lower freezing point, meaning you need less of it, meaning you can have more water in the mix, meaning more conductivity and capacity across the coolant.

Well lets see:
Ethanol
Thermal Conductivity (at room temp) 0.17W/mK
Specific heat capacity 2.44 J/gK
Melting point -114.3C

Methanol
Thermal Conductivity (at room temp) 0.2W/mK
Specific heat capacity 2.55 J/gK
Melting point -97C
(values are rounded)

Don't believe me, do a search and you'll get something like that. Higher specific heat capacity, higher thermal conductivity, higher melting point. For dry ice cooling methanol should be better but the fumes are more toxic, which sucks...

[epion2985]

Those are correct values, however I am still missing your point. There is benefit, it it huge? no, but then no one said it was so its a moot point, thats why I used the words "tweak", ie getting the most out of as given system.

While the numbers may not be very different they are different enough to be useful, if you don't like it don't use it, methanol is no more harmful then ethanol in the given scenario as there are no fumes.

In a way ethanol may be worse as with methanol you use less and more water and the resulting mix has a lower flash point, making it safer form a fire stand point.