• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

New Member- designing a immersion, subzero system

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
Don't get us wrong CraigRochester, we love to see crazy over the top stuff. We are just trying to keep you from some expensive mistakes.

If the fluid is strictly to eliminate frost build-up, you could omit it and concentrate on a Zero humidity cabinet for the system.
 
Hi Soulcatcher,

I hope y'all don't mind my joking around, just trying to have fun. I do really appreciate everyone's feedback and see from the content of the responses that you guys know what you're talking about.

Re zero humidity cabinet, I do that stuff for work. For fun, I like to do stuff I'm clueless about. ;-)
 
Hope the container is hermetically sealed though, when the fluid boils you don't want to breathe it in.
 
ATMINSIDE said:
Hope the container is hermetically sealed though, when the fluid boils you don't want to breathe it in.
Click to expand...

Yes- that's my plan. I'm looking at Silicone Oil and other dielectric fluids as I'm barfing over the cost of the 3M stuff. I'll probably go with a welded aluminum box, bolt on gasketed lid and 2" Rubbatex insulation. If I can work it, I'll have tripple pane window. Window is so I can collect my $20. ;-)
 
You realize the bet was for Novec 7000, right?
If you change the fluid it changes system properties.
 
Wouldn't the Novec 7000 ionize/de-ionize/change properties with a metalic or (worse) ferromagnetic materials as the container? You would certianly have to go with an inert material that does not bind to the 7000 fluid. But I'm also not sure what the fluid chemical compound is.

Also, the CPUs demonstrated in the videos are Intel i3s... They have a TDP of less than 100watts. Your probably going to try out a 140watt CPU in this bathtub experiment. I'm pretty sure your making a very fancy explosion.
 
Unless you extremely pressurize this system, it WILL boil. Why? Because you're 100% going to get nucleate boiling on the IHS, from there it will create a vapor barrier, causing transitional boiling until it collapses in on itself.

Also, vapor expansion takes up thousands of percent more space by volume than its liquid constituents. I can't speak for novec, but water takes up ~1600 times volume for the same mass of water. I.e. "sealing with a gasket" might not be enough.

This stuff appears incredibly dangerous, and theres more variables than you've considered at work. Do your homework, or in my not so humble opinion, leave it as an idea to investigate and discuss and nothing more.

edit; Also, any leak in the system will cause a pressure drop, potentially causing the system to lower its latent heat for vaporization -> more boiling, potential to a catastrophic failure.
 
ATMINSIDE said:
Only going to pay me $1??

Power draw depends on what CPU you're using. Take the size of the IHS and the TDP of the chip, there's what you need.

And, thinking about it, I'll change to 99.5%.
The thermal conductivity of Novex 7000, until phase change occurs, is abysmal at best :)
Click to expand...

Hi ATMINSIDE,

I found a calculator on line for a flat plate calc. If I got all the units right, looks like I owe you $1. :-( How'd you like your payment?

Even if I could crank up the liquid flow to 100 ft/s (30 m/s), which isn't very feasible, I'd still only get about 60W heat dissipation.

Forced Convection on Isothermal Flat Plate in Free Stream

PlateCalc.jpg
 
CraigRochester said:
Thanks! I'll need every cent to make this tank rig.

Has anyone worked with "conformal coating"? Sealing the MB components from liquid and using something like a Glycol solution for the bath?

techni-tool
Click to expand...

conformal coatings are prolific in industry based electronics. LN2 and phase change guys use conformal coatings as well.

And the "dangers" of the novec come from inhalation. Thats the concern. If you're boiling it off in a closed room, it might be an issue. Food for thought. Heres the msds http://www.acota.co.uk/assets/data-centre/msds/3m/3mnovec7000msds.pdf
 
I've had some time this weekend to do some reading on this subject and found this very good web article: Advances In High-Performance Cooling For Electronics

I've chosen a name for this project; rather than bathtub cooling: "Jacuzzi cooling". I've changed my "user requirements" a bit:

  • Liquid Immersion of MB to achieve a total subzero environment and address condensation (thinking low viscosity Silicone oil)
  • "Direct Expansion (DX)" cooling of immersion liquid using standard refrigeration (window air conditioner guts) NOTE: work to be performed by a licenced refrig tech.
  • "Heat Spreader" blocks on CPU & GPU w/ spot forced convention (water jets maybe)
  • DX coils throughout the water bath for forced convection cooling of bath liquid @ evap coils.
  • Hermetically sealed Jacuzzi box
 
:muahaha:

Heat pipe technology uses boiling and condensing fluids to transfer heat from one end to the other. They are sealed and work quite well.
 
I put together a table of properties of bath fluids. I'm thinking the silicone oil is what I'll try.

Material Properties.jpg
 
Still a low thermal conductivity, you're going to boil it unless you use heatsinks on the chips.
 
ATMINSIDE said:
Still a low thermal conductivity, you're going to boil it unless you use heatsinks on the chips.
Click to expand...

roger that... I believe you this time. ;-)

I'm still mulling over using R-410A, which would require a full vacuum / 200 psi pressure vessel with crazy seals for the wires. :bang head
 
You must log in or register to reply here.
Back