Help me find my gas leak! -- NEVER MIND!

Homebrew Talk

Help Support Homebrew Talk:

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

tomakana

Supporting Member
HBT Supporter
Joined
Jun 20, 2017
Messages
87
Reaction score
30
Location
Southeastern PA
This has been driving me crazy for a couple of weeks now. I had a couple of gas cylinders go empty quicker than I expected recently in my keezer - I knew the first one was partially filled to begin with so it didn't strike me as odd, but when the second went dead, I realized I had a leak somewhere. Spraying everything down with sanitizer identified a leaky PRV in one of my kegs, so that's been pulled and replaced - however, I think I still have a very small leak somewhere, small enough that I can't track it down after coating everything in soap water or sanitizer. So I would love for some guidance, even if it's to tell me my logic is all wrong and I need to relax. So, I've attached a picture of my current setup (it's been pulled from the keezer to facilitate replacing some of the parts) with some labels.

I've got a PRIMARY regulator attached to the CO2 tanks, and a SECONDARY regulator that normally sits in the keezer. There are 4 valves in play, #1 on the PRIMARY and #2-#4 on the SECONDARY as cutoffs to three lines going out to the kegs (labeled A, B, C). In the picture, the lines are not attached to kegs, they just terminate in ball lock disconnects.

As I've always understood, a properly setup system should hold pressure until it's released, ideally when you're pouring a cold beer from your tap. So, if I open the CO2 tank and pressurize this system without it being attached to anything, it should register pressure on the various gauges (as it does in the picture), and then once I turn off the flow from the tank, the pressure should stay at that level until it's released. If I pressurize the system and the pressure on the gauges drops, then it's a sign that there's a leak.

Question 1 - is everything I just wrote about pressure staying steady in a system with no leaks accurate?

If that's correct, here's what's currently happening. In the picture, the system is pressurized and all of the valves are closed, which should isolate the PRIMARY and the SECONDARY from each other and the SECONDARY from the gas lines (valves 2-4 are closed). The pressure on the PRIMARY stays at the same level indefinitely (or for well more than 24 hours). The pressure in the secondary drops over a number of hours (looks fine after 90 minutes, then 6 hours later the pressure has dropped in half), which I'm interpreting as indicating that there is a leak - if there is, I can't find one. It seems to be slow enough that it's not creating bubbles in soap water anywhere on the regulator body.

So - two more questions for the collective:

Question 2 - My plan at the moment is to remove and retighten valves 2-4. They are new MFL valves I added to begin using the duotights - otherwise the rest of the SECONDARY regulator is stock as I got it 5 years ago. If that eliminates the slow leak, then I'm good for this step - if it doesn't, any other suggestions for next steps?

Question 3 - Once I eventually eliminate any issues with the SECONDARY regulator, am I correct that following the same process while opening a single additional valve would identify any problems with the gas lines? For example, pressurizing the system with valve #2 open, then closing all valves other than #2 should then tell me if I've got a leak in gas line A if I see the pressures in SECONDARY dropping again, correct? However, if the pressures hold, then the connection at the other end of line A is fine, and I can perform the same check with valve #3/line B, etc.

Thanks in advance for any suggestions/guidance. This may be a roundabout way of testing all of this, but after spending a lot of time trying to track down the leaks with the whole system pressurized and not having any success, I'm trying to see if I can at least narrow my search a little. On the bright side, the kegs are camping out in the keezer, which means I'm actually doing a real lagering on two of them that I usually don't have the patience to do!

Regards,
Tom
gas-system-upload.jpg
 
A couple of days ago I realized I had a CO2 leak in my keezer. I have a dual primary regulator providing 11 and 15 PSI feeds to it, the former to an 8-way manifold, the latter to a Y feeding a pair of wheat beer kegs. I narrowed it down to the 11 PSI side yesterday, so I closed all of the manifold valves at that time, and am about to go open them one by one until the leaking branch is found, at which point there's a near zero chance it's not the keg because the gas infrastructure dates to 2019 and has never been modified.

I'm pretty sure it'll be my Ballantine IPA keg on tap as it's the most recent keg swapped in....

Cheers!

[edit] It was the most recent keg swapped in, but that actually was my HBT Julius clone. When I went to check out the gas side it jogged my memory that I thought the QD snapped on in a rather hinky fashion but I didn't give second thought to that at the time.

I removed it and snapped it back on and it did the same hinky thing, kind of a momentary hitch before fully engaging, so I swapped that QD for a spare, opened all the manifold valves and regulator valve long enough to charge everything back up, then close the regulator valve and the manifold valve for the Julius keg. Tomorrow I'll open the regulator valve and expect no sound of gas moving, then I'll open the manifold valve and expect the same.

Checked out the gas QD, took it apart and it looked like there might have been some dried beer inside, so I cleaned that all up and am letting it dry before reassembling it and testing it...
 
Last edited:
Glad you found the culprit!
Please keep us updated to where your leak problem lied. Someone else may benefit from it someday.

BTW, with a thread title that defies correct spelling ( ...me-craxy.733915/) I'm surprised you were able to trace or find it at all. ;)
 
BTW, with a thread title that defies correct spelling ( ...me-craxy.733915/) I'm surprised you were able to trace or find it at all. ;)
Hah - it was actually the first thread in the "Similar Threads" suggestions when my message posted, so the gnomes in the machine managed to find it for me.


At the risk of jinxing it, I think the 'dunk the whole manifold in a bucket' technique has born fruit, although it was still nearly impossible to find. It looks like there was a VERY tiny leak at the connection point for the last gauge in the string (the gauge above valve #2 in the picture). But even in the water, I was getting a bubble about every 15-20 seconds, so it took a while to figure out where they were coming from. I pulled the gauge and retaped/retightened it. No more bubbles, and it's under pressure now to see if it holds. I'll check in the morning to see what's up...fingers crossed!
 
Back
Top