In-line solution to Prevent CO2 bottle leak down ?

[beren]

Want to detect a leak? Use a co2 monitor. Won’t matter what the flow rate is if it’s in an enclosed space the numbers will build.

 

[whoaru99]

Want to detect a leak? Use a co2 monitor. Won’t matter what the flow rate is if it’s in an enclosed space the numbers will build.

Yeah, I get it.

My recent comment was essentially to that point re. flow-type measurements.

 

[WesBrew]

Since someone had to blow up my pic and point out the band clamps I hadn’t gotten to yet. Got off my azz and finished

 

[WesBrew]

A CO2 monitor in the kegerator is something I didn’t know about and likely wouldn’t have thought of. That is really the only place I’ve had a leak. Good ideas coming out of this thread

 

[day_trippr]

I saw this years ago and haven't thought about it until just now so I looked it up;
https://www.ontariobeerkegs.com/micro-dbl-inline-leak-detector.htmlI'm wondering if one of those kegland FOB detectors could be modded to do the same thing?

That meter needs roughly .1 lpm to show a flow. That would be a monstrous leak ime...

Cheers!

 

[IslandLizard]

Since someone had to blow up my pic and point out the band clamps I hadn’t gotten to yet. Got off my azz and finished

You, know you've got beer in your gas line, yes?

The one thing about using Oetiker clamps over barbs is that you can't simply take the lines off to clean and dry them.
That's one of the reasons to use MFL fittings, everywhere.

Or even better, EVA Barrier line with p-t-c fittings.

 

[WesBrew]

Didn’t even notice . Sht. looks like I’ll be doing that one over. Whatever can go wrong….

 

[SFC Rudy]

Good idea!
But for leak detection a much smaller flow range is needed, such as 0.01 l/min, or perhaps you'd get away with 0.1 l/min.

This gauge will measure 100-1000 mL/minute (.1 - 1.0 L/min). Most connect CO2 at night and find an empty bottle in the morning. Some math; a 5lb tank hold 300 liters of CO2. If it empties in 12 hours, 300 liters in 12 hours is 25 liters/hr which is .416 liters/min which this gauge would detect. If it took 24 hours to empty that would be .21 L/m which this gauge would also detect.

 

[IslandLizard]

This gauge will measure 100-1000 mL/minute (.1 - 1.0 L/min). Most connect CO2 at night and find an empty bottle in the morning. Some math; a 5lb tank hold 300 liters of CO2. If it empties in 12 hours, 300 liters in 12 hours is 25 liters/hr which is .416 liters/min which this gauge would detect. If it took 24 hours to empty that would be .21 L/m which this gauge would also detect.

Thanks for the calculations, a 0.1 - 1.0 l/min gauge will show those common, slow leaky flow rates. Yes!

The gauge should be placed right after the regulator's output, before any manifolds, splitters or Ts.

While on topic of "leaks," I had the PRV on one of my regulators starting to release around 35 psi, although the range of the regulator goes to 50-60 psi. I had a glass rinser attached to that channel, which really needs that higher pressure to do some serious spraying/rinsing. Oh yeah!

We heard it starting to blow off badly at 40-45 psi, so dialed it back, more and more , and ended up using it at 30 psi. Got to see if I can find a higher threshold PRV for that regulator...

 

[Broken Crow]

This is really bugging me... Since the brain injury, I've done my best to re-engage the neural pathways to my lost memory/knowledge/skills; Guided neuroplasticity..and it's worked to some extent. Between 2014 and 2020 I couldn't even comprhend how a screwdriver worked, I could still use one from memory but the intuition and comprhension of it's function eluded me until, in trying to relearn basic electrical systems and design my own BK controller (Thanks for ALL your marvellously detailed post and schematics @doug293cz !!!) a neuron finally reconnected and in an instant I regained a massive amount of basic problem solving skills and much of my basic electical knowledge..literally: POOF! There it was!
Anyway, This thread is exactly the kind of problem I would have siezed on prior to my injury and not rested until solved, so here's where my brain went:
Part of the IR spectrum is absorbed by CO2...One should be able to build a small inline unit, similar to a RIMS tube or inline oxygenation/carbonation unit, but at one end have an IR emitter, adjusted to the CO2 sensitive range, and at the other a simple diode...Due to Brownian Motion, there'll always be some variance which you use as your baseline measurement, when significant more photons are lost, it would indicate a greater mass (owing to the motion of a leak).
This is the sort of thing I would have tried to build before 2014, but I simply no longer hace the knowledge or skills...not to mention; my sense of proportion.
Maybe one of you sound scientific-minded could work with it, or at least tell me that I'm way off track.

 

[WesBrew]

piggybacking on this... My keg, half full, started blowing 95% foam, coming out too fast too. did all of the usual. checked post, dip tube and gasket. no issues. cleaned lines and faucets. No change. bled gas out of keg and filled, same fast n foamy pour. Blew gas through the liquid-out on the keg to rule out overcarbonation. no change. 5' vinyl lines at 10/11psi, 3/16id.
Decided to check the keg pressure with a gauge and it read 15psi ! I bled the keg down to zero and did it 2 more times. same 15psi reading at the keg through a spunding valve/gauge although the regulator was set to 10/11.
I disconnected the gas from the keg, turned the bottle off, turned the regulator to zero, bled out the pressure in the regulator with the relief valve, turned bottle on, slowly adjusted the reg up to 10psi. all ok. I then bled the pressure out of the reg with the relief valve and the regulator shot up to 15psi. I did this 5 more times and it would consistently rise 4-5psi over my setting. Thoughts? I'm tapped out, literally.

 

[sibelman]

@Broken Crow ,I'm as to hear you've been regaining mental function, and hope the improvement continues.

My intuition is that a CO2 detector would do the trick if the leak was in a small enclosed space (keezer/kegerator). Less likely to succeed in a kitchen or garage because of how quickly gases mix.

No DIY required, as there are fairly cheap detectors out there.

 

[whoaru99]

@WesBrew

How much CO2 is left in the cylinder?

I've read, but not confirmed myself, some regulators will exhibit output pressure rise as the regulator inlet pressure drops.

 

[WesBrew]

@WesBrew

How much CO2 is left in the cylinder?

I've read, but not confirmed myself, some regulators will exhibit output pressure rise as the cylinder/supply/inlet pressure drops.

It was just filled

 

[WesBrew]

don’t know if I found”it” but checked the last possible thing I could check. Crush washer. Dug it out , put a new one in. I cant repeat the regulator overpressure problem anymore. Reg set at 10psi goes back to 10 when the pressure is released. the gauge on the keg reads about 12.5. That’s closer but not quite. It doesn’t make sense that the washer change on the bottle would help anything but a leak at that spot.

 

[whoaru99]

It was just filled

Possibly the regulator is getting extra cold then acts a bit flakey if you are moving the gas/changing the pressure quickly?

 

[Broken Crow]

@Broken Crow ,I'm as to hear you've been regaining mental function, and hope the improvement continues.

My intuition is that a CO2 detector would do the trick if the leak was in a small enclosed space (keezer/kegerator). Less likely to succeed in a kitchen or garage because of how quickly gases mix.

No DIY required, as there are fairly cheap detectors out there.

Still would be nice to have an inline detector though... I was hoping some mad scientist on here might bite.

 

[day_trippr]

I think a solid gas leak detector would use a vertically-oriented in-line sight glass filled part way with water - like a bubble gauge, only larger. Connected to the output of a regulator, even the tiniest system leak would show up...

Cheers!

 

[sibelman]

This could be a genius innovative answer. Needs to hold pressure too though, yes?

 

[mashdar]

I think a solid gas leak detector would use a vertically-oriented in-line sight glass filled part way with water - like a bubble gauge, only larger. Connected to the output of a regulator, even the tiniest system leak would show up...

Cheers!

Are we talking in-line air lock?

Might also solve the shocking lack of keg air locks.

 

[day_trippr]

Not an air lock. Something like this, only bigger:

This could be a genius innovative answer. Needs to hold pressure too though, yes?

Absolutely. Creating a leak is not a solution to anything

My point about "bigger" is it has to handle expected flows that could be comparatively high and space and height would be needed to avoid having the water conveyed up the gas line to a keg...

Cheers!

 

[sibelman]

This is a cool and low tech alternative to Bracconiere's digital scale method. Not necessarily cheaper, but it's got beautiful bubbles instead of boring red numbers.

 

[day_trippr]

Bracc's scale would take days or even longer to pick up what I think one might see immediately if the bubble thing actually works

If anything the bubbler is going to be sensitive - it's going to pick up carbonation if a keg isn't sitting at equilibrium, for instance, which could mask an actual leak. Then again, that's a multi-keg problem, and if one has a manifold, isolation is a matter of flipping manifold valves. Shut off the gas to a keg that's still carbonating and see if there's still bubbling going on...

Cheers!

 

[mashdar]

Biggest issue may be blowing off all your liquid upon charging a fresh keg. Maybe needs a bypass? Inline ~1psi PRV might work if a 1psi with sufficient capacity exists?

 

[day_trippr]

Right, that's the kind of scenario one would need to avoid. But I think as long as the operator remembers when swapping in a fresh keg that the head space pressure might be below the regulator setting, and to thus use the shut-off valve in between them wisely (ie: start closed and s l o w l y open it), all should be well...

Cheers!