Help Me Root a CO2 Leak

Brewing Processes
1

Need some help rooting the source of a CO2 leak.

Taprite T742 dual gauge regulator > connected to three way manifold (with check valves).

• With the gas line check valve closed, line charged, and the tank off – the beverage side reads 10 PSI for approximately 3 days

• With the keg gas lines disconnected, line charged, check valves open, and the tank off – the beverage side reads 10 PSI for approximately 3 days
• With the keg gas lines disconnected, line charged, check valves closed, and the tank off – the beverage side reads 10 PSI for approximately 3 days

• With the keg gas lines connected, line charged, check valves open, and the tank off – the beverage side reads 10 PSI for < 10 hours
• With the keg gas lines connected, line charged, check valves closed, and the tank off – the beverage side reads 10 PSI for < 10 hours

I have rotated through the three kegs, using a different gas line each time. With the line charged and tank off, the beverage side gauge reads 10 PSI for < 10 hours. Regardless if I use line 1, 2, or 3, attached to any of the kegs, the beverage side gauge will zero out after a few hours.

The loss of pressure appears to be keg independent. As I have rotated the lines to each keg and pressure still drops. This leads me to believe the equipment up to the gas-in post is sound and not leaking. Everything up to the keg has been water tested (not the tank) and no leaks found. I lube the gas posts and lid rings.

I have soap tested each keg numerous times, the lid ring, pressure relief, and gas post – no leaks. I have not tested the liquid side as the faucets aren’t leaking and there is no beer pooling at the bottom of the fridge nor is there any sign of beer running down the side of the keg.

I cannot root the source of the leak.

Any ideas?

2

Have you simply pressurized a keg, left both Bev and gas side disconnected and then checked the keg itself?

IME if there is a leak a tank will drain overnight, if not sooner, not 3 days later unless I’m not understanding your post.

It sounds like you tested everything gas side so I’d start looking at the Bev side even though you see no evidence of beer it doesn’t mean the co2 isn’t leaking out somewhere

3

You didn’t mention it, but I’m assuming you had a tank empty faster than usual?  How much faster?

Do you mean that after 3 days the low pressure gauge dropped to zero or that after 3 days it was still at 10 psi so you stopped that part of the experiment?

Are these kegs empty when you’re testing them or are they full of beer?

4

Have you simply pressurized a keg, left both Bev and gas side disconnected and then checked the keg itself?

  • yes, the kegs are/were pressurized and tested after a few days. All three kegs hold gas. Although I do not know how much PSI.

You didn’t mention it, but I’m assuming you had a tank empty faster than usual?  How much faster?

  • the 10 pound tank emptied sooner than it should but is a slow empty - months.

Do you mean that after 3 days the low pressure gauge dropped to zero or that after 3 days it was still at 10 psi so you stopped that part of the experiment?

  • PSI is set to 10. With the gas turned on, the beverage side reads 10 PSI. With the kegs disconnected from the gas circuit, the beverage gauge will remain at 10 PSI for at least three days. At some point after three days, it will zero out. I am not sure how long it should hold pressure - indefinitely?

Are these kegs empty when you’re testing them or are they full of beer?

  • They all have beer at varying levels.

When one keg is connected, pressurized to 10 PSI, and the gas bottle turned off, it will eventually drop the beverage size gauge down to zero. This occurs with all three kegs using different gas lines. For example, gas 1 to keg 2, gas 2 to keg 3; therefore it is happening to all kegs regardless of which gas line is connected.

Kegs are ball lock. Perhaps gas is escaping from the quick disconnect, as if the seals were bad. I just greased them up.

How can I test for a leak while the QD is attached to the keg post?

However, when disconnected, the gas stays put for a minimum of 3 days @ 10 PSI.

5

Are you using the fiber washer between the tank and regulator?  I ask because when I had a slow leak like this, I wasn’t aware that my tapright had a built-in seal.  Using the washer was actually causing the leak.

6

Is it possible that:

  1. There is no leak, and you just got a low fill on your last tank; or

  2. The fact that the low pressure gauge is dropping when you have a keg connected is because the beer in the keg is absorbing some of the CO2, and if there is a leak, it’s either between the regulator and the gas cylinder or somewhere in the regulator itself?

That said, the fact that the low pressure gauge stays at 10 for at least 3 days when the gas out line is closed, the regulator is pressurized, and the gas is turned off makes me think there is no leak between the reg and the cylinder or in the reg itself.

7

Every time I have had a Co2 leak it has been something simple. A white nylon washer has been missing. Thread tape needs replaced. Or a connection hasn’t been tightened. And don’t rely on hand tightening alone. Hand tighten then give a quarter turn or so with a wrench.

8

O-rings have been my main source of CO2 leaks. Once replaced, no more leaks.

9

Are you using the fiber washer between the tank and regulator?

  • no fiber washer, the Taprite has a o-ring built into the stem.

There is no leak, and you just got a low fill on your last tank;

  • it’s been happening tank after tank.

The fact that the low pressure gauge is dropping when you have a keg connected is because the beer in the keg is absorbing some of the CO2, and if there is a leak, it’s either between the regulator and the gas cylinder or somewhere in the regulator itself?

  • makes sense but these kegs have been sitting @ 40 F for many weeks now. I am uncertain that the beer will continue to absorb CO2 past it’s saturation/equilibrium point.

Every time I have had a Co2 leak it has been something simple.

  • it’s all taped-up and wrench tight. I even tightened the center stem of the QD gas connectors (little slot on top). As I mentioned earlier, when the gas is not connected to a keg, it maintains pressure for quite some time. Only when it is connected does the gauge “go flat” relatively quick.

Bear in mind that the kegs keep pressure, and expel gas when the pressure relieve valve is opened. I could understand if it didn’t blow off gas, that makes sense, but to have it hold pressure over time is odd.

O-rings have been my main source of CO2 leaks. Once replaced, no more leaks.

  • which ones? I have replaced the lid o-ring, gas post or-ring, and beverage O-rings. External and internal O-rings.

The beverage side gauge also drops from 10 PSI to 0 PSI, when pressurized and the tank off, whether or not the line check valve is opened or closed. The little valve under the regulator body, the feed to the manifold.

It would be helpful for someone to tell me how their gauge reacts and if it looses pressure over time with the bottle off and kegs connected and/or disconnected.

10

I replaced all of them: dip tube, gas post, and lid. I believe the dip tube o rings were my culprit.

My gage stays constant.  It is not a level indicator. It does not slowly drop as gas is used. It will simply be “good” then no pressure when empty.

I know this must be frustrating. When I install a tank I ensure I have a flat washer between the regulator and tank. I hook up my CO2 lines, attach them to the kegs with the tank valve off, pressure adjust knob all the way out, and levers off. I open the tank valve to get pressure to the regulator. I open the lever to the keg I am charging then I turn the pressure adjust knob to my desired PSI. I can hear CO2 rush into the keg. I burp the headspace three or four times to clear the air.

I don’t loose pressure until I burp the keg and it rebounds quickly. Otherwise it reminds constant.

I hope this helps.

11

Perhaps I misunderstood, but I thought you said in your initial post that the low pressure gauge stayed constant when the valve under the regulator body was closed and the tank was turned off?

If your gauges are going down when the valve under the regulator body is closed and the tank is turned off, the leak is either in the regulator or between the regulator and the tank.

I thought up a way to test for a leak between the gas disconnect and the keg post (if my logic is flawed, someone please let me know).  Hook a line up to a keg and turn everything on; close the check valve at the bottom of the regulator and turn the gas off; wait until the low pressure gauge drops to zero; quickly remove the gas disconnect from the keg; check for remaining pressure in the line by pressing the valve in the gas disconnect with your finger while holding it up to your ear to listen for gas escaping; if there’s gas still in the line there is no leak between the disconnect and the keg post; if there is no gas in the line that would indicate a leak (unless you weren’t able to remove the disconnect quickly enough and the gas escaped while you were doing so).

Here’s my thought process, with no new gas able to enter the line, it would all leak out if there was a leak between the disconnect and the gas post.  However, if the reason the gauge is dropping to zero is because the beer in the keg is absorbing some of the gas, there would still be equilibrium in pressure between the keg and the line, and there would still be some gas in the line.

I guess another way would to be hook everything up, turn everything on, and submerge the keg completely under water while watching for bubbles.

12

As I asked in my previous post, are you sure you have a leak and if so how? Does your tank still have co2 in it? If there is indeed a leak the tank will drain within 24 hours regardless of where the leak is. If you are simply looking at fluctuations in the pressure gauge and your tank hasn’t drained you don’t have a leak. You might simply have a bad pressure gauge:)

13

I’ve also found slow leaks when I don’t “seat” the keg lid properly at around 30 psi.  I believe I can thank Steve in TX for that one.

14

This is a slow leak, one that doesn’t dump a tank overnight, but one that will slowly kill it.

I believe I have narrowed it down to one keg. With the gas off for a few hours, it is always the one that takes gas. Although I cannot see any visible leaks with soapy water. The keg is 1/2 full.

Perhaps it is absorbing the gas - I don’t know …

15

Yes, kegs with liquid in them will absorb the gas.  That’s why your beer carbonates, it absorbs the gas.

16

Yeah, I get that … but only one keg seems to continually require gas after sitting w/the gas off for a few days. It has been under gas since mid-November, surely enough time to reach equilibrium. The other two kegs do not take on gas in the same way. The suspect keg, exhibits the same behavior with all three of the CO2 lines, taking on gas. There must be a leak somewhere but I’ll be damned if I can find it.

The posts are tight, greased, and not visibly leaking.

17

Found it.

This keg is my Pliny the Elder brew which had a hop bag for dry hopping. The bag was attached with dental floss and secured to the underside of the relief valve with a small stainless worm clamp.

Somehow, a small bit of floss found its way unto the relief valve opening and compromised the seal. One tiny little piece of floss had me chasing a leak for weeks now.

Only when I saturated the relieve valve with soapy water and let it sit did the leak become apparent.

18

Nice job!

19

Thank you JT.