Tracking down a very small CO2 leak ...

Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

2puttbird

Well-Known Member
Joined
Mar 6, 2011
Messages
111
Reaction score
1
Location
Jefferson City
Hi All,

I'm having trouble locating tips on finding an ellusive CO2 leak, so I thought I would start a thread.

I built my keezer in the summer of 2011, and my first 5 lb bottle of CO2 lasted over a year. My second lasted about 6 weeks, which was quite a surprise. I immediately suspected the bottle (I don't really have anywhere to get bottles refilled, so I have to swap bottles when I need a refill) and/or the plastic washer. I talked to the bottled gas guy, and said that it's pretty rare for a bottle to leak, and he also said that the washers usually last longer than one bottle, but they're cheap so I replaced it when I traded.

After installing the third bottle, I ran a few simple troubleshooting tests ... each time, I pressurized the system and then shut the main valve on the bottle and waited to see if it would hold pressure ... it didn't.

Since then, I've just been closing the bottle valve in between uses until I had a chance to take the keezer offline for closer analysis and repairs ... and that time came this week, when both kegs finally became empty at the same time. :(

I unhooked and removed the kegs, closed the valves on the distribution header, unplugged the controller, and let everything come to room temperature. I pressurized it and then shut the valve on the bottle. I just checked it, now 5.5 hrs later, and the secondary pressure has dropped from 11psi to 9psi ... not as bad as I had expected, I guess, but still a leak.

I bought the system pre-assembled. The CO2 hose is clamped onto the regulator with a crimp clamp. The other end of that 4' hose is secured with a worm gear clamp to the fitting that is threaded into the header. Unless the valves are not completely sealing (possible, but probably not the problem, since the leak existed when the kegs were connected also), my leak must be in those few joints, right?

What I'm having trouble imagining is how a change in bottles would have affected any of those joints. I'm not tough on the hardware. And I replaced the plastic washer between the bottle and the regulator, which was the most obvious (or so I thought) culprit.

Thanks for any tips you might have.

Cheers!
 
Thanks for replying ... I am familiar with that process, but unfortunately it requires enough of a leak to actually blow a bubble ... my leak requires 5.5 hrs to relieve 2 psi of pressure in a 4' hose, and it took 6 weeks to empty a 5 lb CO2 bottle, so it's a very small leak!
 
I have no idea ... my understanding is that it's standard hose used for this purpose. My next step is to get a new clamp, cut new ends on the main line, and reinstall them to see if that helps, even though I can't imagine how changing the bottle would have affected those connections.

Thanks for the reply.
 
Sounds like you only changed the bottle and the washer...if you still have the old washer, put it back in and check...could have a hairline leak on the new washer.
 
Is it possible to get your suspected leaks in a bucket of water? Even with a small leak if you can, you should be able to see air bubbles forming
 
Duh! Increase the pressure! Sometimes I'm too close to the problem to see other potential solutions.

NivekD ... the second bottle leaked out with the original washer, and then I changed it when I got the third bottle ... sorry that I didn't explain that very well ... but that's definitely something I need to verify anyway, thanks.

Nunner ... it would be a pain, although with all the rain we're getting I may get a chance yet! ;)

Thanks for the replies, guys ... they're really a big help!
 
Def put it up to 40 psi and submerge everything.

I have had 2 bad check valves, so make sure you are putting those underwater too!

And don't rule out the kegs, if you don't get a good seal a keg will bleed a tank dry pretty quickly.
 
look in the plumbing supplies section of your favorite hardware store for leak detection soap. It has more than just simple soap and water. Even a very slow leak will eventually bubble up. Combine this with increased pressure, and you should find it soon enough.

There are a bunch of things that can be used to improve the bubble performance of soap and water. For a food safe DIY formula try equal parts Dawn dishwashing liquid, water, glycerine, and an HEC based personal lubricant like K-Y or the Walmart brand Equate. It will be thick and gooey and slow to evaporate, better for keeping it in place until a bubble forms. These bubbles will be long lasting too.

Rinse well when you are done, you don't want everything to be slippery the next time it gets wet.
 
Thanks for all of the helpful replies, guys. Bought some high-end bubble stuff (sorry if I'm getting too technical) and cranked it up to 40psi, but couldn't find anything. Even used it on the regulator, including the relief valve.

Detached my CO2 header and dunked it in a bucket of water ... nothing. Also dunked the keg connectors, just in case my valves weren't working properly. Don't think it would be wise to turn my CO2 bottle (to which my regulator is directly attached) upside down in the bucket, so I'm coming up empty.
 
At this point you may want to start to valve-off the system, then later go back and see where pressure was lost. Turn everything off with pressure on and up to 40psi and let it sit, then starting from the tank (tank off), open each vlave looking to see if pressure has dropped. That includes disconnecting the kegs.
 
Thanks for the reply. It's currently sitting pressurized with all of the valves off, so the secondary gage is reading the pressure of gas in the regulator, the main hose, and the distribution header.

The pressure is dropping, albeit very slowly. It's down to about 35psi from the original setting of 40psi a few hours ago. I haven't done the calculations, but given the relatively small volume of gas that is being measured, I think that 5psi represents a very small amount of lost gas, which is probably why my second 5lb bottle lasted 6 weeks.

So the only thing left is the regulator ... I did put bubble soap on the regulator joints a couple of times (and there are a lot of joints, as you know), but didn't see anything.

It looks like I can get a good quality dual-gage regulator for $50-$60, and at this point that isn't looking like too much money to spend. Any recommendations?
 
Hi all.

Bought the 742-2, which allowed me to rip out the header and primary gas line. Ran an overnight pressure test and it was solid as a rock. I'm back up and running!

Thanks, everyone, for the tips and suggestions!
 
Back
Top