Water cooling with no radiators

[batatas]

I would like to know if you know someone that has tried this before my ideia is just to test if it would work and maybe do something completely crazy if it works, but what i have in mind is having the processor and the gpu be cooled just by water no radiator or fans i have one 1000L deposit that i will fill with water and see how long the pc can be stress tested until the water reaches like 40 ºC and of course i need to monitor the temperatures of the cpu and gpu too, but i want to know what you think of this idea and if someone have done this before because that may help with some stuff but anyways i want to test the two 10 000 L deposits that i have i want to see how that will work.

[Bananasplit_00]

all you need for you to not need radiators is enough surface area that the heat can dissepate away, adding more coolant will just add more themal mass and slow down the heating. its an interesting project and id love to see someone pull it off with all the calculations and stuff but that system would never be even remotely mobile lol

[cluelessgenius]

one word:

 

filtration

[TidaLWaveZ]

Hope you've got miles of tubing.

[Ruffleduffle]

While @Bananasplit_00 is right thatwithout sufficient surface area for radiating of the heat, a thousand liters of water will take a very longtime to heat up. and the area of such a large reservoir will likely be sufficient anyway

[intender]

Bury them outside underground and pipe them back inside.  Instant sub ambient cooling.  Geothermal for pc!

[batatas]

I will probably try it just need to buy some stuff and put it all together... hey Linus may like to do a crazy thing like this too that would be dope but anyways i will try it and then maybe i will try on the 2 10 000 L ones too who knows right .

 

[batatas]

2 minutes ago, intender said:

Bury them outside underground and pipe them back inside.  Instant sub ambient cooling.  Geothermal for pc!

Hum that may be cool i can probably try that too.

[TVwazhere]

This would have been a perfect experiment back when whole room water cooling was a thing

[intender]

Just now, batatas said:

Hum that may be cool i can probably try that too.

I remember a guy on another forum probably about 12 years ago that cooled his pic with a coil out of about 30 feet of copper pipe and buried down about  6 or 8 feet in the ground.  at that depth it is near constant temperature year round, and he said it worked great.  The exposed lines might freeze in a cold environment though. 

[Wrb]

I think I saw someone do this while I was on pcpartpicker or something. The guy had (if I'm remembering correctly) 10l of fluid in one massive tank, and it worked fairly well because the fluid sat in the tank so long it managed to cool just enough so after one full loop the fluid was as I started.

[Ruffleduffle]

54 minutes ago, intender said:

I remember a guy on another forum probably about 12 years ago that cooled his pic with a coil out of about 30 feet of copper pipe and buried down about  6 or 8 feet in the ground.  at that depth it is near constant temperature year round, and he said it worked great.  The exposed lines might freeze in a cold environment though. 

also such a loop would be pain in the behind to drain and keep clean

[W-L]

4 hours ago, batatas said:

I would like to know if you know someone that has tried this before my ideia is just to test if it would work and maybe do something completely crazy if it works, but what i have in mind is having the processor and the gpu be cooled just by water no radiator or fans i have one 1000L deposit that i will fill with water and see how long the pc can be stress tested until the water reaches like 40 ºC and of course i need to monitor the temperatures of the cpu and gpu too, but i want to know what you think of this idea and if someone have done this before because that may help with some stuff but anyways i want to test the two 10 000 L deposits that i have i want to see how that will work.

You wouldn’t need that much fluid and it can be done but more effectively you would just have the tank itself passive radiate the heat out to the surrounding air.

 

There was a post a while back where this was tested, the temps would very slowly climb until it reach equilibrium and all the fluid was at running temp.