Turning my co2 extractor into a cryochiller?! Gas expert needed! :)

Hey friends, I had built a couple smaller scale co2 extraction units (think 1/2lb every 8 hours) using an air drive pump (Products - Maximator GmbH)

So I’m able to ramp up to about 1600psi on the one that I have setup right now.

The extraction time just sucks nuts compared to doing large batch ethanol, especially with the efficiency losses in the pneumatic drive system.

I know the sublimation of the co2 can produce some real deep cooling. I’ve made dry ice blocks on the bottom before when I was just starting out on the builds.

I was thinking using the co2 to chill the ethanol down to -40 from lets say -18 or something. Does anyone have any recommendation for parameters to maximize flow/chill in terms of the psi? Does it make sense to only bring it up to subcritical pressures (800-1000), it has a faster flow rate at subcritical. I can make sure it hits supercritical pressures beforehand, it just reduces the flow rate/pump speed.

I’m wondering how fast/slow this would be, or if its even worthwhile. Or if anyone has any experience doing something like this with a different gas? Is this a good choice? If its just bubbling through alcohol I could switch it to nitrogen or something?

Also, is this a super bad idea for any reason? Lol.

You want to run at 900psi and place a brazed plate heat exchanger behind your needle valve and before your separators. You are creating a r744 refrigeration cycle where your waste heat will be transferred to what ever is cooling your gas boosters.

Your brazed plate should be sized for the flow and heat load that you have on your ethanol side. You can install a TXV to control the flow to the heat exchanger. Look for R744 expansion valves and TXV’s.


Kaori Brazed Plate Heat Exchangers are a good choice as are Alfa Laval. A simple needle valve can work as a expansion valve. You could use the waste heat from your gas boosters to boil your ethanol, you will make alot of hot water with those boosters.


I’m using an equilibar valve (https://www.equilibar.com/back-pressure-regulators/supercritical-back-pressure-regulator/) so I’m able to set the PSI pretty easily, I’ll rock out 900 and see what happens.
I have a heat exchange coming off the pump, its a bent stainless tube style, if I put a fan on it I can get it down below 30c by the time it hits the first chamber (currently product tube). Seems ilke I might be ready to rock :stuck_out_tongue:

Is this an efficient cooling method vs a standard chiller?

Good call on using the waste heat, I actually have an evaporator system running for the ethanol that I was going to run on a heater so this might be a viable option. I’ve got a few other uses for warming as well. Thx!

What are some standard temperatures this can hit? I guess at least -60 I’m guessing?

1 Like

With your separators (suction pressure) running at 120 psi you will hit -60f. Your cooling power will be dictated by your booster power. I find that cooling power is usually a multiple of 2x to your air power to your booster. (IE 10 hp air compressor running a booster at near full load will produce 20 hp of cooling power.)

1 Like

Thanks! This is super useful.

I have a 15hp compressor so it should get some good power behind it.

Is there a calculator out there for chiller equivalence in hp? I found an air conditioning calculator, is it a similar conversion? Or a way I could get some ideas as to how long it would take for a litre to go from -18 to -40c.

I guess it would be about 22kw equivalent? I guess i could try and compare some of the numbers from the immersion threads. I’ll poke around a bit :slight_smile:


Jesus after sitting here going through this feed I can’t help but he soooo curious as to see what all this looks like and be there to see it go down. This is why I love coming to the site!!! So many people with knowledge that intrigues me! After all the chiller issues I’ve dealt with over the last two days I wish I had half the know how these two have! My brand new out of the box 48L chiller wouldn’t cool below 17 degrees. Got it sorted out now but damn I’m still up at 4am stressing about work. Goodnight. :zzz:


If you have a chiller that won’t reach temperatures low enough for you application you can add a r744 cascade chiller to your existing chiller to expand its power and low temperature chilling capacity.

A cascade system functions with two refrigerants. Your standard chiller will condense the r744 at around 60f well into the most efficient range of regular refrigeration. The condensed high pressure r744 will then be expanded through a needle valve or TXV into a heat exchanger. The liquid r744 will expand and evaporate in the heat exchanger and create temperatures down to -60f depending on the pressure on the suction side of your pump. You need a gas compressor that can handle pressures up to 1000psi, a good option is a Copeland scroll compressor for R744 or a blackmer oil-free reciprocating pump.

Alternatively you can just use direct expansion and vent your suction side instead of compressing it for reuse.


omg! you make this sound almost trivial…

makes me want to go get my refrigeration tech license.

thank you @JonaaronbrayAzoth


So i’ll whip up a diagram. But I essentially have a downtube running into my collection chamber. Then the gas goes from the chamber to the pump from a port in the top.

I’m curious if I can just fill the collection chamber with ethanol and essentially “bubble” the expanding gas through the ethanol?

Also, thx so much for the help so far!

You need the energy from the expanding Co2 to be stored by a medium like the brazed plate heat exchanger. If your separators are jacketed then you could run your solvent through them essentially turning them into heat exchangers.

Yeah the separators are jacketed. I was thinking of just pulling vacuum on the jacket to ensure the least amount of cold loss, though its just solid stainless throughout so maybe even taking the time to insulate these things might be key. I’ll snap some photos and a diagram to give a better idea of what I’m talking about.


I think the threads got confused

"I would suggest something to function as a restriction point between the high pressure extraction vessel and the lower pressure separators. The expansion point can be a needle valve or a small nozzle to restrict the flow and allow for lower pressures to be maintained in the separators. The lower the pressure you run on the separators the lower the cooling temperature will be. You will still need a water chiller to run a closed loop cooling cycle in your system. The heat from compressing the separators to higher pressures has to be displaced somewhere for the refrigeration cycle to continue. Your chillers need to be able to hold a steady 5c under load for you to get the best performance from the co2 cooling.

Your separator’s jackets will function as your heat exchangers. Pump the extraction solvent through the separator jackets flowing top to bottom for best exchange. Running your separators at around 250psi should cool your solvent down to -15f, 140psi should cool down to -30f"

Just gonna put this in there to continue convo

Here’s a link to the image of my setup. So basically I can get rid of the product tube entirely.
The eqiulibar valve acts as a pinched tube/regulator valve. It runs with a nitrogen backpressure and only allows throughput at the desired psi (I’d set to 900)

My question is, on my collection chamber, Can I just fill the 8L cylinder with alcohol? (lets say 6L full), then as the gas expands into it, the alcohol would be acting directly as the heat exchange? There is a downtube towards the bottom, the gas would expand and come out the top of the chamber at 120psi back into the pump (it isn’t clear on the diagram, but my gas flow back into the pump actually comes from the top of the expansion chamber into the pump)

The other option, is fill the jacket of the collection chamber with the etoh, and allow the gas to expand in the full chamber and chill the fluid inside the jacket instead.

I’d imagine the etoh direct expansion would be the most efficient?