I'm force carbonating my first batch of homebrew. I bought a 5-gallon Cornelius keg and switched out an empty CO2 tank with a fresh one. The tank pressure started out at 500 psi, but when I turned up the pressure on the keg (filled most of the way with beer) to 20 psi the tank pressure started dropping rapidly. It settled out at 400 psi. The tank pressure did stop dropping, so I don't think I have a leak. Does this sound like a normal amount of CO2 to use up when force carbonating?

3 Answers 3


The tank-side pressure is mostly measuring the temperature of the CO₂ tank, not the actual amount of CO₂ remaining. Above the triple-point of CO₂ at about 517 kPa ~ 75 psi, the pressure reading is dominated not by the amount of CO₂, but the temperature, which relates to the fraction of gas vs. liquid in that equilibrium. You'll notice your CO₂ tank will basically stay around 500-600 psi until it quickly degrades down to nothing over the course of the last few days. Like propane, measuring the weight of the tank (minus the tare weight of the tank itself) is the best way to understand how much CO₂ is actually used.

If you're force carbonating to approximately 2 volumes of CO₂ for a 5 gallon batch, that's … I think around 75 grams CO₂/batch. A 5lb CO₂ tank is about 2300 grams, or about 30 batches. You'll probably lose a lot more to leaks, headspace and over-pressure venting than actual forced carbonation.

  • +1 I see different tank-side pressure readings depending upon where the tank is stored.
    – mdma
    Commented Mar 4, 2014 at 7:18
  • You also have to account for CO2 used for serving and keeping it carbed while serving since more CO2 goes in when beer comes out. IME this uses up a noticeable amount of CO2 also.
    – paul
    Commented Mar 4, 2014 at 18:42
  • Thank you. The tank/keg are holding a constant pressure, so I don't think I have leaks. (But I'll definitely be water testing the system before I do this again). I guess I just lost a lot more CO2 to headspace and over-pressure venting than I was expecting. Commented Mar 5, 2014 at 14:18
  • Use soap (not just water) for your leak testing. That initial dump of CO₂ can cause the tank itself to cool a bit, which might explain the initial pressure drop.
    – jsled
    Commented Mar 5, 2014 at 14:47

I've seen a couple of pages now that list numbers on the milage of a 5 lb tank. From what I gather you can carbonate 5-7 5 gallon corny kegs with a 5 lb tank. You can serve 15-22 5 gallon kegs with a 5 lb tank.

Links on how many kegs you can force carbonate with a 5 lb tank: http://www.homebrewtalk.com/f35/how-much-co2-force-carbonate-333455/ http://www.kegoutlet.com/how-many-liters-does-a-5lb-co2-tank-carbonate.html

Links on how many kegs you can force carbonate with a 5 lb tank: http://www.micromatic.com/beer-questions/how-many-beer-kegs-dispensed-out-co2-tank-aid-89.html


My 5 pound CO2 tank has always sat at about just under 1000 PSI for the reading on the tank pressure until it gets to almost empty. That's at a room temp in the 70's.

When you had that 500 to 400 PSI drop, over how much time was that and also, did you do the shaking the keg method to quickly force carbonate during that time or did you just turn on gas at 20psi on new full (uncarbonated) keg and leave it sit undisturbed?

Shaking the keg method will move enough CO2 to change the temp of the tank and that could definitely create a quick drop in pressure reading of the tank, but then once you stop shaking and leave sit long enough for the tank to return to room temp, that pressure reading drop from the tank changing temp should go back to what it was at room temp. If you just hooked it up and did no shaking and are letting it absorb the co2 just sitting, then I don't think gas will flow fast enough to change the tank temp.

Soapy water worked up to a foam is a great way to check for leaks. But what I usually do first, is pressurize the system - i.e. have everything hooked up to keg and valves open through the lines to the keg - then once it looks to be holding a constant pressure, I completely close the rotating valve that is directly on the CO2 tank (i.e. valve PRIOR to gas hitting the tank regulator) and don't close any other valves. Then you can quickly see if your system is leaking pressure out anywhere because all your pressure gauges will drop to 0 over a short time if it is (5 minutes if bad leak - few hours or half day max if slow leak). If you don't lose any pressure on the gauges for a day with that main tank valve closed (and of course no serving of beer), then you can be pretty sure you're completely leak free. Of course if the beer is still carbonating and hasn't reach carbonation equilibrium, those pressure gauges will drop down maybe even to 0 as the beer in the keg is continuing to absorb co2 with that main tank valve closed. But once you get some practice with it, you can tell the difference in time between an actual leak and the beer just continuing to absorb co2.

I would say 5-7 five gallon kegs on one full 5 pound CO2 tank is maybe even on the low side of what I get. Definitely can force carbonate AND server that many if no leaks and not blowing gas off for anything else. (I tend to use CO2 a lot -- I prefill my empty kegs with CO2 before transferring the beer to the keg to reduce oxidation and remove any outside air - CO2 is heavier than plain air so it will form a nice blanket of co2 in the bottom of the keg. I also fill any other vessel with co2 prior to transferring beer to it if I want o2 free transfer).

  • The pressure drop was pretty quick. I watched the needle drop over maybe about 10-15 seconds. I didn't shake the keg at all. Commented Jun 12, 2014 at 16:45
  • Thanks for your answer. I'll be back to read this again next time I keg a batch. Commented Jun 12, 2014 at 17:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.