Kegging Setup CO2 leak - driving me insane

[Grott]

Tape = no no, you'll get bits of the crap down the inside of the regulator, great way to turn it in to a paper weight.

Sorry but there is no way it can get into the reg, it's on the outside thread of the gas bottle and the nut of the reg screws on/over this.

 

[DJ_L3ThAL]

It won't be the thread on the type 30 connection unless you are deaf, as 50bar leaking out of a thread will always be an audible leak. As mentioned above the nylon washer gives the high pressure seal to the cylinder, thread tape is a HUGE no no and will cause leaks as it doesn't let you compress the nylon washer effectively. You should be able to do the regulator up hand tight and then a 1/8 to 1/4 turn with a spanner to tighten, if it leaks then the washer is damaged and needs replacing. Again this will be audible and not a minor leak due to the full high pressure.

One full turn is NOT fully open, try spraying the top of the cylinder handle with soapy water. You maybe surprised what you find. Also the Inlet stem thread I mentioned...

 

[DJ_L3ThAL]

This shows what I mean for tightening the regulator on and also regarding opening of the cylinder valve.

https://youtu.be/Qa7P5qKf1e4

 

[Grott]

As mentioned above the nylon washer gives the high pressure seal to the cylinder, thread tape is a HUGE no no and will cause leaks as it doesn't let you compress the nylon washer effectively.

To understand is to learn. The thread on the gas bottle that secures the reg using the type 30 nut is no where near the nylon washer and has no effect in letting "you compress the nylon washer effectively". Could you explain how. Only interested.

 

[DJ_L3ThAL]

To understand is to learn. The thread on the gas bottle that secures the reg using the type 30 nut is no where near the nylon washer and has no effect in letting "you compress the nylon washer effectively". Could you explain how. Only interested.

Sure mate happy to, Gascon have good info on the cylinder thread types here http://www.gascon.com.au/content/products/pdf/Valve_Summary_AS2473.3.pdf

You can see the flat face of the type 30 male. This is where the nylon washer sits and is squeezed up to make the seal as you tighten the nut over the thread. Hope that makes it clear. Cheers.

 

[Grott]

Thanks for that, appreciated. No tape!

 

[Lionman]

My 6KG keg king bottle is pretty much empty after carbonating and dispensing about 15 kegs over 10 or so months. It runs through a 3 way manifold that is connected to two kegs.

I'm fairly sure it should be able to do about 10 times this amount so I'm guessing I have a leak.

When I set up the system I did the spray down check with starsan and found one leak where the output on the reg wasn't tight so I fixed that.

I couldn't find any other leaks after that. I'm a bit lost as to where to look now. Do I have to remove the entire system from the fridge and test all the components in the bath?

What is the most likely culprit? Where the reg attaches to the bottle?

 

[Grott]

A 6kg bottle should enable you to carb and dispense about 45 x 19l kegs, so you have a slow leak over 10 months. You need to go through a step by step process.
1) turn manifold off, put 40psi out of bottle and turn bottle off. After a couple of hours check the gauge and if it hasn't dropped the bottle/reg/hose connections and at the manifold are not leaking.
2) if not leaking then remove gas disconnects from kegs, open manifold turn bottle on so 40psi is up to the disconnects. Turn bottle off and check pressure after a couple of hours.

If there is a drop in pressure the leak is from manifold or disconnect connections. If no leak the the kegs need to be looked at.

If you follow this let us know the result so we can work from there.

 

[Lionman]

A 6kg bottle should enable you to carb and dispense about 45 x 19l kegs, so you have a slow leak over 10 months. You need to go through a step by step process.
1) turn manifold off, put 40psi out of bottle and turn bottle off. After a couple of hours check the gauge and if it hasn't dropped the bottle/reg/hose connections and at the manifold are not leaking.
2) if not leaking then remove gas disconnects from kegs, open manifold turn bottle on so 40psi is up to the disconnects. Turn bottle off and check pressure after a couple of hours.

If there is a drop in pressure the leak is from manifold or disconnect connections. If no leak the the kegs need to be looked at.

If you follow this let us know the result so we can work from there.

Roger that, thanks for the guidance.

I'll follow these steps and see what I can find.

I guess every time the kegs are touched a test should be done similar to what you have mentioned confirming no leaks? I guess you could get a small leak at a post or lid seal that you would need to check for.

I think I have been a bit complacent, need to improve my kegging procedure.

 

[DJ_L3ThAL]

Good quality fittings, setup correctly and maintained (ie. orings on keg posts lubed and/or replaced as needed) won't need rechecking. My current 2.6kg bottle has lasted almost a year (15/7/2015 was swap date) and has pushed out MANY kegs and fluid when cleaning/flushing. Haven't weighed it in a while to see how much is left but it proves you can quite easily have a totally leak free system. I don't even turn the cylinder valve off, ever. It's located in my lounge room also.

If you systematically leak check everything bit by bit you'll find one or more leaks and can fix them as you go. Don't rule out your kegs leaking either.

 

[Jack of all biers]

If you systematically leak check everything bit by bit you'll find one or more leaks and can fix them as you go. Don't rule out your kegs leaking either.

Yep, about week two into my systematic leak check of a new system (KK kegerator S4). Major leaks fixed and now only a minor leak(s) that is/are SLOW, but still to go. Soapy foam trick has not really found it/them, but maybe I'm not persistent enough. I filled two kegs up with water and carbed to test the whole system. I tested the regulator alone and initially found a leak, but it holds pressure fine now over a couple of days. I checked the line, stage by stage and thought it was holding (over a day or two), but when I connected the kegs, I found the pressure going down. I went away for a week, so disconnected the kegs to double check the keg. The line was pressurized and the bottle turned off. High Pressure gauge 5600kpa, low pressure gauge 10 psi. After 7 days the High pressure gauge was 4000kpa and low pressure just under 10psi (my initial reading may have been by eye so by angle slightly off). Checked again today (another 24 hours after the 7 day check) and HP 3400kpa and LP just under 10psi. Now I really have tried to find that F*&king leak, but it must be minuscule or pressure changes due to temperature have resulted in a change of pressure in the line (a cooling occurred here in Adelaide this last week). Is this even theoretically possible?

Am I frustratingly chasing a gas leak that is not there or should I just give up and turn the bottle off every time it's not in use or should I keep persisting with finding that F*&king leak :unsure: I hear what DJ is saying and want that to be my system, but really am struggling to find the VERY SLOW leak.

EDIT - The major leaks were found by the whole system being removed from fridge unit and manifold and disconnects dunked under water, which found one disconnect leaking from top (tightened to fix, which resulted in one major leak being fixed). Leaks from reg to bottle fixed by tightening nut and reg to line by tightening nut. Now soapy bubbles aren't finding anything. This was all done prior to this last week in which the change in HP has been 2000kpa over 8 days with the bottle off and kegs disconnected.

 

[Chap]

Don't rule out your kegs leaking either.

I lost a 2.6kg bottle in a matter of days. Turns out the out post internal o rings were no good. Replaced and lubed, no further issues.

Like DJ said, don't rule out the kegs, take a systematic and elimination approach and hopefully you find the culprit asap

 

[enoch]

I lost a 2.6kg bottle in a matter of days. Turns out the out post internal o rings were no good. Replaced and lubed, no further issues.
Like DJ said, don't rule out the kegs, take a systematic and elimination approach and hopefully you find the culprit asap

The gas post external oring is my first port of call for a leak. They are rubber and they perish and crackers eventually.
Run your finger round them before refilling. I just use water qd orings now as they are cheap and I change ever 12 months or so.

 

[Rocker1986]

My 6.8kg cylinders last about 11-12 months, I haven't kept track of the number of kegs that have gone through them though. I guess it's about 4-5 kegs every 2 months or so, so around 25-30 all up, this includes the soda water keg that takes up residence when there's a spare spot available. I also use the gas for pushing cleaner and rinse water out of each emptied keg. I suppose it also depends on how highly carbonated the beers are kept too. I like mine on the higher side of mid-range.

In any event, swapping a gas cylinder once a year isn't something that bothers me. If they were only lasting a month or two then there'd be a problem.

 

[FarsideOfCrazy]

. Now I really have tried to find that F*&king leak, but it must be minuscule or pressure changes due to temperature have resulted in a change of pressure in the line (a cooling occurred here in Adelaide this last week). Is this even theoretically possible?

Hi Jack,

I'm having a problem with a slow leak also. But this comment made me take notice. Yes temperature can effect the pressure in your c02 system.

I had/have a leak in my system, firstly it was the hose on the barb had been bent and had split the line, fixed that, still leaking. Going crazy couldn't find a leak anywhere. Thought about submerging my reg, don't do this as the gauges aren't water proof and can be damaged. But I found another leak by a fluke. While spraying soapy water around the gauges some ran down onto the front face of my micromatic reg. A lot of bubbles appeared from where the face screws in to the body. This part is sealed by the diaphragm behind it. Removed the front face to make sure the reg was clean wiped a small amount of keg lube around the contact edge of the diaphragm (not sure if this was wise) tightened it all up, and hey presto, no more leaks while it only had a disconnect connected to a short piece of line.

Left it for 3 weeks no pressure drop except when the temp dropped by about 15 degrees, Check on google you can find a image on gas c02 pressure in relation to outside temperature.

Then I put the one way valve back in, JG fitting version, and the gas disconnect (different one from the first test) and now the leak is back.

I've just purchased some new gas disconnects with barb fittings and I hope that when I get a chance to connect everything up and check it for leaks, it doesn't effing leak!

 

[DJ_L3ThAL]

Yep, about week two into my systematic leak check of a new system (KK kegerator S4). Major leaks fixed and now only a minor leak(s) that is/are SLOW, but still to go. Soapy foam trick has not really found it/them, but maybe I'm not persistent enough. I filled two kegs up with water and carbed to test the whole system. I tested the regulator alone and initially found a leak, but it holds pressure fine now over a couple of days. I checked the line, stage by stage and thought it was holding (over a day or two), but when I connected the kegs, I found the pressure going down. I went away for a week, so disconnected the kegs to double check the keg. The line was pressurized and the bottle turned off. High Pressure gauge 5600kpa, low pressure gauge 10 psi. After 7 days the High pressure gauge was 4000kpa and low pressure just under 10psi (my initial reading may have been by eye so by angle slightly off). Checked again today (another 24 hours after the 7 day check) and HP 3400kpa and LP just under 10psi. Now I really have tried to find that F*&king leak, but it must be minuscule or pressure changes due to temperature have resulted in a change of pressure in the line (a cooling occurred here in Adelaide this last week). Is this even theoretically possible?

Am I frustratingly chasing a gas leak that is not there or should I just give up and turn the bottle off every time it's not in use or should I keep persisting with finding that F*&king leak :unsure: I hear what DJ is saying and want that to be my system, but really am struggling to find the VERY SLOW leak.

EDIT - The major leaks were found by the whole system being removed from fridge unit and manifold and disconnects dunked under water, which found one disconnect leaking from top (tightened to fix, which resulted in one major leak being fixed). Leaks from reg to bottle fixed by tightening nut and reg to line by tightening nut. Now soapy bubbles aren't finding anything. This was all done prior to this last week in which the change in HP has been 2000kpa over 8 days with the bottle off and kegs disconnected.

Unfortunately it took me hours of pain. See this video of a tiny leak I found on a brand new higher quality (non Keg King) manifold which I am currently using (without leaks).
https://www.youtube.com/watch?v=yaUmyhP-c8Q&feature=em-upload_owner
I had to remove the valve, clean the thread and re-apply some Loctite thread sealant. Two wraps of high quality teflon tape should do the same but for long term use the thread sealant used as per manufacturer's instructions is the go as it forms a pressure bond when squeezed and does not need Oxygen/Air to set.

Again, I HAVE put my whole regulator under water, I just made sure I dried it straight away by blowing out with an air compressor etc. This was how I found the steam thread leak mentioned earlier in the thread. If you don't dry parts then yes you will have problems.

I'm having a problem with a slow leak also. But this comment made me take notice. Yes temperature can effect the pressure in your c02 system.

Left it for 3 weeks no pressure drop except when the temp dropped by about 15 degrees, Check on google you can find a image on gas c02 pressure in relation to outside temperature.

While you are correct that ambient temperature affects the pressure of any gas in any system. Gaseous CO2 pressure fluctuations in a beer system (even commercial size) due to ambient temperature is a negligible effect. You would not see it or measure it really on the gauges we have. More likely you are seeing fluctuations in the gauge bourdon tube system due to the metals expansion/contraction with ambient temperature changes.

The pressure versus ambient temperature charts you are likely referring to are relevant for the CO2 cylinder only, as the Liquid-Gas equilibirum (balance) will change with ambient temperature. So I'll quote a rough pressure of 5000kPa in your CO2 bottle, but this in summer could be much higher as the liquid CO2 boils off more CO2 into gas to stay in equilibirum, the reverse happens when it is cold. The same effect is seen when using a BBQ LPG gas bottle for a while and you see condensation or ice (at the case swaps when we smash them) on the bottle, because the cylinder is using heat from the air to boil off the liquid LPG to replace it as gas because you are withdrawing gas from the bottle. It wants to stay in liquid-gas equilibirum.

The other thing you are seeing where you pressurise the system, then turn the cylinder off and the high pressure gauge drops over time and the low stays relatively the same means you most likely have a leak downstream of the regulator, or on the regulator outlet or main body face. Because the low pressure side is supplying gas to the leak from the high pressure side. So this hopefully pin points it.

You can of course simply connect one short run of nylon to a ball valve or quick disconnect off your regulator and use that to pressure/leak test and see. This would rule out the regulator if you don't want to put it all underwater. But underwater is the only way to see TINY leaks (refer my video above), even the commercial grade leak detecting VRV fluid I use at work wouldn't have picked that up.

Hope that makes sense.


==== Also.... just realised I'm the OP haha, so will provide some more info on the path I followed, helpful or unhelpful ====

For others reading this thread, persistence pays. But I think sadly I learned this but starting with high qualitry fittings greatly reduces the amount of small leaks you need to chase. Some of the Keg King fittings I had originally wouldn't leak one minute but when disturbed or moved leaks to buggery. Not only can the gas post on a keg leak, but the gas disconnect can leak, take it apart with a large flathead screwdriver from the top and luber up the o-ring seals (inspect it first as any rough bits or cracks means it WILL be leaking). It is easy to dunk these under water.

As for orings, someone mentioned plumbers grade, I'm paranoid about food grade stuff so always use silicone where I can. The red silicone orings are awesome and I bought all types I needed from eBay and have enough for myself and my future children. Here are the order numbers for the types available.

Keg post 111
Keg dip tube 109
Keg lid 417? (Check, I have not purchased this one)
Keg PRV 104
Keg poppet (universal) 105

Quick Disconnect small for the pin 103 (2.38x7.14mm, measured 2.6x8.0mm)
Quick Disconnect large for the top cap 111 (11.11x15.86, measured 10.8x15.0)

Vessel wall (bulkhead) 1/2" BSP 20ID x 26OD mm

Can be ordered here: http://stores.ebay.com/Oringsandmore

 

[TheWiggman]

Well here's another lesson that you should never doubt me
I dunked the reg underwater today (off work, crook as a dog) and found a small leak under the adjustment handle. I sealed that area up, so I'm confident the diaphragm has a leak - likely in the middle where it is fused to brass. Short of getting replacement diaphragms, a new reg is in order.

 

[Lionman]

I shut the input on my manifold off and shut the tank off and the line has maintained 10psi for 24 hours. Doesn't appear to be a leak at the reg then, I think.

Edit: actually, the tank pressure gauge may have dropped. It was hard to tell because there is almost no pressure in the tank but when I turned the gas back on the needle jumped a little.

I took the reg off the tank to go and exchange it and noticed a tiny black oring in the outlet port and of the tank. It sure where that was meant to go.

 

[Jack of all biers]

Thanks DJ and Farside for the feed back. :beer:

Unfortunately it took me hours of pain. See this video of a tiny leak I found on a brand new higher quality (non Keg King) manifold which I am currently using (without leaks).

I had to remove the valve, clean the thread and re-apply some Loctite thread sealant. Two wraps of high quality teflon tape should do the same but for long term use the thread sealant used as per manufacturer's instructions is the go as it forms a pressure bond when squeezed and does not need Oxygen/Air to set.

Cheers for the video DJ, I did dunk my manifold under water, as I was sure (given others experience) that my KK manifold would leak. I watched it like a hawk for 10-15 mins and not one single bubble. One of the top covers of one disconnect did bubble, but that was easily fixed with a screw driver.

The other thing you are seeing where you pressurise the system, then turn the cylinder off and the high pressure gauge drops over time and the low stays relatively the same means you most likely have a leak downstream of the regulator, or on the regulator outlet or main body face. Because the low pressure side is supplying gas to the leak from the high pressure side. So this hopefully pin points it.

You can of course simply connect one short run of nylon to a ball valve or quick disconnect off your regulator and use that to pressure/leak test and see. This would rule out the regulator if you don't want to put it all underwater. But underwater is the only way to see TINY leaks (refer my video above), even the commercial grade leak detecting VRV fluid I use at work wouldn't have picked that up.

Hope that makes sense.

Makes sense and I guess I haven't been as methodical as I thought. I will have to give it a crack I guess as I don't really want to have to live with a leak as it will bug me everytime I look at the thing. Thanks for the tips. The reg LP is 10 psi (as set) and HP is down to 2500 kPa today (day 9 of test) so definitely a leak somewhere in the system. Sigh, once I eventually get this sorted I will move on to the kegs. Hopefully no leaks there. FIRST WORLD problems I guess. Will have a :chug: and get over it eventually.

EDIT - Oh and thanks for the link to the O'ring seller. Typical as I just paid almost 5 times the price for a few from KK. Doh. (smacks forehead)

 

[abyss]

Great info DJ.