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

Vacuum chamber cooling?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.


New Member
Sep 7, 2020
I thought of this idea awhile ago after I saw this video:
The liquid used here is called 3M Novec. The boiling point of this liquid I'm not sure of because the Gigabyte rep didn't specify which Novec liquid they were using. They are using their 2 phase cooling method which adds a condenser to the set up.

According to their spec sheet on their product page for liquid immersion found here 50c or 56c is the boiling point of the Novec liquid depending which Novec they used for their 2 phase immersion cooling method.

I understand the condenser is cooling the vapor being boiled at those lower temperatures, but I always was thinking of using a vacuum chamber in a reservoir instead since Novec isn't the cheapest liquid or readily available to use for the average Joe.
Has anyone ever attempted doing something like this? My thought was if you could pressurize a reservoir of water low enough to boil the water at near room temperature, you would then have an efficient way to produce water vapor or "steam" as I'll continue to call it. This steam could then be funneled to power PC fans, in much the same way a steam turbine works, but the fans would be cooling the water flowing over the radiator that would cool the water before hitting the reservoir. Since it's steam powered there's several ways to produce an X amount of steam cools with higher fan RPM. So in theory, the hotter the water gets, the faster the fans blow, either by some steam to electric generator conversion or by brute force steam power. The faster the fans blow the more cooling capability. The more cooling capability the less steam produced. Since the water atmosphere is a negative pressure you could set the boiling point a little above room temperature so that when the PC is off, the fans are off. When you turn on PC with a light work load, the fans would be casually cooling the radiator. When gaming, the fans would spin at their highest capable RPM or theoretically the amount of steam produced by brute force steam power.

At some point, the water can only boil so much, to produce so much steam. It might take some basic electric theory and circuitry, but you could make a circuit to trip X voltage to power X additional fans. Let's say we have a 280mm radiator with 4x140mm fans in a PP configuration. X amount of voltage produced by steam could trip 1 fan with higher voltages tripping more fans. So basically it becomes, the hotter the liquid, the more fans are activated or even less RPM of each if you prefer all fans going. It gets really interesting if you use the 3M method of using cooling coils as a heat exchange in a custom pressurized reservoir. I think at that point it would start to get pretty expensive, but would be a great step up to try after the steam powered fans method.

Let me know what you guys think. Anyone ever heard of something tried like this or can think of any catches to consider if I were to try it? (I'm seriously considering attempting this)
I need to think about this some more, but to check some numbers first ...

To get water to boil at:
20 degrees C, you'd need a pressure of 0.06 mbar.
30 degrees C, 0.09 mbar
40 degrees C, 0.14 mbar

These are achievable pressures with a high-quality rotary vane pump, but you can't allow water into these, so you'd need a [very expensive] dry scroll pump.

What you essentially want to do is use heat output to drive the fans, which is an interesting idea.

Next step would be to work out heat capacities, so that a chip producing e.g. 100 W doesn't overheat.

If I find time later I'll do some maths ...