Duotight gas leak indicator?

[cactusgarrett]

I just converted my gas side over to 4mm EVAbarrier and DuoTight fittings, and I'm hoping someone can lend some insight as to whether or not I have a leak.

I hit the system with pressure, and with quick disconnects off the kegs, shut off valves (before the in-line regulators) and let it sit overnight. When I come back to it the next morning, there's a noticeable pressure drop on the gauges, and when I charge the lines again (turn on valves before the regulators) the gauges jump back up to the set pressure. Disconnected from kegs and isolated from the gas tank, should these lines hold/maintain pressure? Is this indicative of a leak? I tested the QDs by submerging in a glass of water (no leaks there), but obviously I can't do that to the in-line check valve or in-line regulator junctions.

Overall, the setup is similar to this (including the blue clips). After the gas tank and regulator, there are 5 independent lines, so I'll illustrate just one for simplicity:
Gas tank/regulator -> DuoTight in-line ball valve -> DuoTight in-line regulator -> DuoTight check valve -> DuoTight low profile ball lock gas QD

 

[Bobby_M]

Back all the way up. If your primary regulator has a shutoff valve, shut that. Set the pressure to 20psi and then shut off the tank. Does the 20psi still hold for an hour? If not, you have a leak in the regulator or between the tank and reg.

 

[cactusgarrett]

Back all the way up. If your primary regulator has a shutoff valve, shut that. Set the pressure to 20psi and then shut off the tank. Does the 20psi still hold for an hour? If not, you have a leak in the regulator or between the tank and reg.

It does, and I have been shutting it off after charging the lines. So basically the closed line that I'm monitoring is ball valve + Duotight regulator + check valve + QD. THAT closed system isn't maintaining the initial pressure over a handful of hours.

 

[Bobby_M]

Take the tubing out of each duotight connection and reset it in the fitting. Many leaks on this system is based on not pushing the tubing past the second oring. A light smear of keg lube on the end of the tubing and a slight twist as you insert will help a lot.

 

[cactusgarrett]

Yeah, I tried resetting all the fittings, but no dice. I'll try again with some lube and a twist. Is this tubing good for only so many inserts before needing to be snipped for a fresh sidewall?

 

[BongoYodeler]

Also make sure you have good, 90° cuts on all your tubing. I had one that I could not stop from leaking, even though it looked fine. I changed Duotight fittings and it still leaked. I ended up cutting an inch or so off and re-seating it into the fitting and that fixed it. Yours is not exactly the same situation as mine, but it wouldn't hurt to check all your cuts.

 

[DuncB]

Duotights and EVA barrier do not like tight bends. Make sure that keezer lids etc when shut aren't bending something.
If you get the correct sized allan key you can judge the depth of penetration for the tube. Measure the distance and then mark the tubes with pen before ramming a good straight cut end in. I've never had to twist or lube my connectors to fit them together. Have about 35 or more push fit joins on my system but I did have a leak when I first set up and did empty a tank so it's worth looking and finding the problem.

 

[cactusgarrett]

Yeah, no tight bends or anything - this is a standup fridge through-wall run, so it's all sweeping turns. I've been using a duotight snipping tool, so i'm confident in the cuts. I've also marked every insertion with a sharpie to confirm I'm pushing each one in fully, but I'll re-cut each connection as a last resort, as well as lube and turn the tubing.

 

[TenaCJed]

Check for micro cracks on the duotight fittings themselves. I had nothing but issues with the fittings on the gas side, to the point where I went back to standard swivel fittings on gas. I have no issues on liquid side, but I found several fittings had micro cracks, even the elbow on the flow stopper keg filler cracked on me. Now that is not something I use much and definitely put no strain on it.

It is easy to overtighten the fittings that thread onto something else, those cracks tend to be easier to see. The micro cracks I am talking about is just below where the tubing goes in where it gets thinner.

Maybe I just got unlucky, but I gave it a try and replaced some pieces just to see those replacements also get small cracks in them. I have seen plenty of others that have not had issues, no idea why I had issues.

 

[bartichex]

Also, keep in mind that after you insert the tubing into the duotight fitting, you must firmly pull the tubing outward in order to seat the compression collet and engage the metal teeth.

 

[DuncB]

@bartichex
Good reminder but he's using the blue clips under that collet so must have it pulled out for the clip to fit on.
I don't use those clips just push pull and hope!
Which has been fine so far.

 

[day_trippr]

fwiw, I have almost 50 ptc connectors in service, mostly JG by far, then some DMfit (beer shank connectors), and just five Duotights (regulator outputs), and never pulled back on the EVABarrier tubing with any of them. I don't think it's necessary: similar to Sharkbite connectors the "tooth" angle on these fittings pretty much guarantees they positively engage just by stuffing the tubing to the stops...

Cheers!

 

[cactusgarrett]

I removed the check valves from the circuit, so it's basically:
( Ball valve / Duotight regulator / ball lock QD )

I'm still seeing a pressure drop (from 15 psi) overnight in all five lines after re-cutting, lubing, reconnecting (w/ twists). Just to confirm - when charged and disconnected, this closed circuit should maintain pressure, right? I'm running out of options on what to test. Next will be replacing the line between the shutoff ball valves and the Duotight regulators, then I don't know what to do.

 

[Yesfan]

Can you swap out the ball lock QD? Do you have a spare regulator you can swap out to confirm the Duotight one is not the culprit? If you have some of the Kegland carb caps, you could put one on an empty 16oz Coke bottle, hit it with 15psi to make sure it isn't the line to the keg that's leaking too.

 

[cactusgarrett]

All five lines (including QDs and regulators) are behaving the same, which is suggesting either these closed circuit lines don't hold pressure when disconnected from a keg overnight, or I have leaks in all 5 lines at the ball valves or regulators. I'm more apt to believe the former.

I have carb caps, so I suppose I could hook one up and see if it behaves when attached. I've dunked the QDs in a glass of water with the line charged and I don't see any bubbles, so IF it's leaking there, it's pretty slow. No one has explicitly stated that these QDs are supposed to hold pressure when disconnected from a keg, though, so that is still a wildcard to me.

 

[Deadalus]

That the five branches all leak is indicative of one of the pieces in the branch not working for all five branches, which you have noted. It could be the tubing, I was reading the reviews for the ball valves at Morebeer and one reviewer noted the following when asked if the ball valve could be used as a CO2 shutoff.

BEST ANSWER: Yes it can, as long as you are connecting it to 8mm tubing. The tubing needs to be very round and free of defects. Tubing that is slightly oval in the cross section may leak and you will lose all of your CO2. This happens a lot when the tubing is coiled too tightly and stored or shipped in that state.

Do you have rigid pieces installed between some of the pieces? If you do try making a branch using just those between fittings and not the flexible tubing. And for fun, make a branch using just flexible tubing.

 

[DuncB]

How about just building it up one part at a time ie reg to ball lock. THen reg to elbow and inline reg to the ball lock already tested. Then add in a T etc.

But as it's all assembled can't you just dunk that whole board in water? Put the pressure up to 35 psi and open up each component one by one?

 

[cactusgarrett]

So from submersion in water at 40 psi, I found that one of the ball valves has a slight leak at the knob when turned to one position. However, testing the other four valves (and check valves) separately shows those are all okay. Why would the other four lines be dropping pressure if they're isolated from the one bad ball valve by four good ball valves?

 

[jdauria]

I wound up with a leak with my DuoTight connections on gas side myself yesterday. Have DuoTight fittings on my manifold with one of the gas lines a longer line for when I am filling kegs via closed transfer. After filling the keg, I used that line to attach the keg since my other two lines are in use and gas was coming out on manifold side. Somehow the tubing had slid out and was no longer connected to the 2nd inner o-ring. Not sure how that happened, but once I pushed it back in, all was good.

 

[sibelman]

The ease of use for these fittings is somewhat compelling, at least on the beer side where one might occasionally change out tubing (to adjust length, or due to age). And beer side leaks are easily seen.

But on the gas side, leak chasing threads like this make me wonder whether PTC is worth it, given that EVAbarrier tubing can work with barbs without huge difficulty. What am I missing?

(yes, I know I'm being something of a curmudgeon)

 

[cactusgarrett]

Barbs don't avoid the issue when I'm wanting to use EVAbarrier check valves and individual in-line regulators. The crux is that where i would use a barb, I'm pretty confident I'm not leaking from (tank regulator, ball lock disconnect).

 

[TenaCJed]

Yes, the draw is the individual in-line regulators. The funny part is the gauges are so small you cannot read them, so you have to use blowtie or something else with a larger gauge to reliably set your pressure.

I do still use the regulator and check valves, but everything else is barb to include 4 port gas distribution. I did not use the EVA barrier ball valves.

 

[day_trippr]

To me the primary draw of PTCs is enabling a simplified tubing architecture compared to the old school barbs, swivel nuts and band clamps, plus tail pieces, etc.
Secondary consideration is the ease of reconfiguration, eg the fermentation gas keg purging system I use that is shared between two chamber fridges, either or both of which may have a batch bubbling away...

Cheers!

 

[cactusgarrett]

Update:

I took the regulator manifold board down, isolated each line separately and recut, refit, etc. until each line held pressure through the whole thing overnight. Reassembled the board, put all circuits in place, held fine. Once each line was confirmed to be holding, I put the blue collar clip in place to indicate each joint was kosher. Last night I re-installed the board onto the side of the upright fridge and all five lines dropped 15 psi overnight after the shutoff valve. CO2 tank to shutoff valve held fast, though.

Re-twisted the lines at the regulators to ensure hanging the board isn't torquing the lines. Gonna re-submerge the check valves and QDs in water, again, if it loses pressure over today.

 

[McMullan]

I'm really leaning back towards old-school swivel nuts and barbs, tbh. Much better and more reliable, especially on the gas side. I'm not a fan of too many push fittings on the liquid side either, when you notice how much gunk builds up in them. Tubes need to be recut to tidy them up before reconnecting after cleaning and push-fit systems just aren't designed to be dis-/connected indefinitely. Eventually it's going to fail. All over the floor I quite like the EVA barrier tubing, though. The Duotight compatible secondary regulators need to be supplied with regular 1/4" female threads. In the meantime, a compatible stem to AN4 male (to receive a barbed swivel nut) is the way to go.

 

[cactusgarrett]

Update:

I took the regulator manifold board down, isolated each line separately and recut, refit, etc. until each line held pressure through the whole thing overnight. Reassembled the board, put all circuits in place, held fine. Once each line was confirmed to be holding, I put the blue collar clip in place to indicate each joint was kosher. Last night I re-installed the board onto the side of the upright fridge and all five lines dropped 15 psi overnight after the shutoff valve. CO2 tank to shutoff valve held fast, though.

Re-twisted the lines at the regulators to ensure hanging the board isn't torquing the lines. Gonna re-submerge the check valves and QDs in water, again, if it loses pressure over today.

Currently looking like this. I just heat-gunned the lines out after the regulators for a 90 bend downward to try and minimize torque at the collar, but even with the blue clips, I'm not sure if that's kosher. If it loses pressure again I suppose i can pull down the entire assembly and see if it leaks without the weight of the lines pulling on itself.

 

[Deadalus]

Currently looking like this. I just heat-gunned the lines out after the regulators for a 90 bend downward to try and minimize torque at the collar, but even with the blue clips, I'm not sure if that's kosher. If it loses pressure again I suppose i can pull down the entire assembly and see if it leaks without the weight of the lines pulling on itself.

View attachment 766559

This isn't what it looked like after you lost the 15 psi on each line correct? When you say " Reassembled the board, put all circuits in place, held fine. " what did it look like? Originally you mentioned ball valves and check valves.

One common part you did mention is that you installed the blue clips. Are you getting them installed all the way in? Maybe they aren't manufactured correctly (too thin/too thick)? You could try without them or even split them and use them on 2-3 circuits.

I suspect you could find small clamps to support the connection after the fitting and before the bend if you wanted to.

 

[cactusgarrett]

This isn't what it looked like after you lost the 15 psi on each line correct? When you say " Reassembled the board, put all circuits in place, held fine. " what did it look like? Originally you mentioned ball valves and check valves.

Right. This is an earlier pic (before installing the ball valves), but the orientation is the same. Ball valves are installed between the Ts and the regulators, and check valves are inside the fridge. I'm questioning if the weight of the lines themselves is pulling down too much on the ball valve/regulator connection or the regulator-tube-out interface. It's the only thing I can think of, considering it all worked when it wasn't put up onto the fridge.

 

[Deadalus]

Were there lines inside the fridge while testing with the fridge on? Maybe the air in the lines is being cooled causing a pressure drop.

 

[cactusgarrett]

On day one they were, but I thought that the cooling might be an issue so all subsequent testing was done without them in it. The current status (even though my drawing has them in the fridge) has them outside the fridge.