How much CO2 leak is OK?

[mistercameron]

I've been trying to diagnose this problem for a few weeks - I have a very slow leak in my system that I can't seem to get rid of. If I pressurize the line and close off the tank, within a day the pressure will have dropped by about 30%. Eventually it will hit zero. I've found no leaks during dunk tests, and I suspect the leak is too slow to notice. My setup is this: [reg]-[check valve]-[hose]-[manifold]-[check valves]...

I replace the check valve at the regulator, no luck. I then replaced the two check valves at the manifold - no luck. Everything has been reaped with about 4-5 wraps of the heavy yellow teflon tape.

Here's what I do know:
- Closing off the first check valve seems to hold pressure just fine - so it's not the regulator or the tank seal.
- Opening up the line to the manifold will leak somewhere between.
- Keeping the check valves open doesn't seem to change the rate of depressurization, so it's not the keg gas fittings.

This leads me to think it's the hose. The thing is - the hose is on there really right, but I do have worm clamps. I know those could be a problem.

Is there anything else I can try short of buying a set of Oetiker crimpers and clamps? Like - another way to seal the hose fittings? Or maybe a slow leak isn't a big deal. My reservation is that I don't brew often right now due to home and work schedules, so I could go months before having used up a tank on my own.

 

[Yooper]

How much is ok? 0.

But if you can live with it, and not have flat beer or a rapid loss of c02, then it's ok if you can stand it.

 

[j1n]

If you do a dunk test and there is a leak, you will see it. submerge the entire thing and sit and wait till you see the leak.

 

[mikescooling]

Nope fix your leak. It's in the hose or fittings replace one or the other and use worm drive thick wide clamps, so you can undo them if need be. All this testing is without a keg attached? If a conry is hooked on all bets are off, any of the 3 poplets, treads, or gaskegs can leak.

 

[COEBRA]

How much is ok? 0.

But if you can live with it, and not have flat beer or a rapid loss of c02, then it's ok if you can stand it.

Lol +1! 0 for sure. No leak is a good leak!

 

[brew_ny]

none

 

[grathan]

That's not a leak, mine depressurizes within 3 hours. I doubt the people saying 0 truly mean 0.

 

[mistercameron]

Nope fix your leak. It's in the hose or fittings replace one or the other and use worm drive thick wide clamps, so you can undo them if need be. All this testing is without a keg attached? If a conry is hooked on all bets are off, any of the 3 poplets, treads, or gaskegs can leak.

Yup - no kegs attached. Just the fittings and hose.

Dunk tests - no bubbles that I could see, except the fine bubbles that were on everything (I assume coming out of solution). The worm clamps I have are the standard width. Red CO2 hose.




Sent from my iPhone using Home Brew

 

[day_trippr]

That's not a leak, mine depressurizes within 3 hours. I doubt the people saying 0 truly mean 0.

Well, I hadn't said anything prior, but I'll say 0 leaks is the only way to remain sane over the long term. And given my 6 faucet keezer goes nearly a year with a puny five pound tank and my carbonation fridge gets every ounce out of a fill, I'd say I'm pretty confident in the whole "zero" thing...

Cheers!

 

[mistercameron]

Is it worth trying to seal the barbs better with Teflon tape, pipe dope, or other compound? Or possibly just go with two worm clamps on each end?


Sent from my iPhone using Home Brew

 

[mikescooling]

Have you tried soap bubbles or a lil water mixed with starsan. My old ball lock fitting used to leak when not connected to a keg. If you're looking for a leak connect to a keg (make sure it's an empty keg, so you don't have to factor the co2 going into the water/beer). So now hook up a empty keg (put the press up to 30pisg) and see if it leaks?

 

[b-boy]

Not all your connections need tape. The flared fittings shouldn't have tape, nor should your CO2 bottle to regulator connection. Sometimes the tape can cause leaking.

I feel for you. CO2 leaks are a pain, and they are expensive. A bottle of CO2 in my area is $25.

 

[mistercameron]

Have you tried soap bubbles or a lil water mixed with starsan. My old ball lock fitting used to leak when not connected to a keg.

The leak is between the first check valve at the the and the manifold. Even with the manifold valves closed off, the line will eventually lose pressure. See above re: dunking - no visible signs


Sent from my iPhone using Home Brew

 

[mistercameron]

Not all your connections need tape. The flared fittings shouldn't have tape, nor should your CO2 bottle to regulator connection. Sometimes the tape can cause leaking.



I feel for you. CO2 leaks are a pain, and they are expensive. A bottle of CO2 in my area is $25.

Pain is right. I dint have any flare fittings on this system. I was wondering about actually putting something in the barbs themselves if it's the tubing that is having trouble dealing over it.


Sent from my iPhone using Home Brew

 

[mikescooling]

The leak is between the first check valve at the the and the manifold. Even with the manifold valves closed off, the line will eventually lose pressure. See above re: dunking - no visible signs


Sent from my iPhone using Home Brew

It may take 5 minutes to blow a soap bubble, it's not the same as a dunk test. They make soap bubbles just for leak checking. It's a thick mix of soap and water.

 

[The_Bishop]

If I'm reading this right, you're saying that you pressurize the system then shut off the tank, right? Then your pressure slowly runs to zero?

Which gauge runs down to zero? The high pressure gauge? If that the case, your 'leak' is at the valve stem on the tank valve. Those valves are not designed to seal around the stem at a partially open position or closed, and they won't hold pressure post-valve (After tank) unless they're all the way open.

If you have a check valve after the regulator that's working right, even after the high side drops to 0 you should still have co2 pressure in your lines.

I went a little crazy with this when I built the gas system for my keezer. Found it out the hard way.

 

[mistercameron]

If I'm reading this right, you're saying that you pressurize the system then shut off the tank, right? Then your pressure slowly runs to zero?

Which gauge runs down to zero? The high pressure gauge? If that the case, your 'leak' is at the valve stem on the tank valve. Those valves are not designed to seal around the stem at a partially open position or closed, and they won't hold pressure post-valve (After tank) unless they're all the way open.

If you have a check valve after the regulator that's working right, even after the high side drops to 0 you should still have co2 pressure in your lines.

I went a little crazy with this when I built the gas system for my keezer. Found it out the hard way.

Yes - that's exactly the case. I watch the high pressure gauge slow drop to zero over a few days, at which point there's pretty much nothing left in the lines. My assumption has always been that closing the tank valve all the way down would leave the system in constant stasis as long as there were no leaks. The only confounding evidence was when I kept the first check valve off and everything seemed to stay pressurized.

This whole thing started when my tank fell off the chest freezer (don't ask), and not long after I drained a whole tank's worth in a few days. It appeared that nothing was damaged - no nicks or dings. I think that slow leak was solved by re-seating the main gas line, but it always seemed to have this slow leak afterward when I did the testing as I described above. I almost went as far as buying a new regulator, but wanted to be sure all other cheaper avenues had been exhausted before upgrading to the dual-pressure system I've secretly always wanted, but not yet *needed*.

I'll try pressurizing the line between the check valve and the manifold, whilst leaving the tank on. After a week I should still have pressure in the line if all is well.

Thanks so much - I never would have guessed that about the tank/reg design "feature."

 

[mistercameron]

For posterity sake, and in case anyone comes across this thread, the leak was fixed by really clamping down on the gas line. In a possibly related thread to this leak, I discovered one of my female brass nipples had a crack all the way through the fitting. I may not have noticed it earlier, but it could have been there and also be the source of the leak.