Pressure dials when cylinder turned off- do I have a leak?

[DocAnna]

A bit of a question for those with cylinders of CO2: if you turn off your gas at the cylinder, do your primary and secondary pressure gauges drop at all over time?

So after clearing up from the mess of the roof (long story) I've finished plumbing in the brew space for CO2 supplies in several locations, each of the three outputs from the combined primary and secondary regulator go into manifolds, with each of their outputs on ball valves. If I turn off the gas at the cylinder when all the manifold outputs are closed the system should then be entirely closed, but I've found that over the course of about 2 days the pressure drops to zero on all the dials and primary. This suggests that somewhere there's a leak. I cannot though for the life of me find it. I've used leak detect spray all over the pipes, cylinder and everything in-between. I've now taken to trying to isolate the outputs one by one using a JG blanking plug, and still can't find the leak.

It's clearly a very slow leak - is there any other clever approach or can anyone suggest a different leak detect spray - I'm using this one at the moment Ambersil Leak & Flaw Detector Spray, Detector

 

[The Baron]

I am no expert Anna but the secondary will let the gas still go into the beer so I would expect it to go down as it is absorbed by the beer that is assuming you do not turn them off and only the main gas at the cylinder.
This will also eventually go down on the primary too I would think as that is connected to the secondary's.
Remember I am no expert on this.
Ps you are getting to the end of your woes

 

[gyurmaember]

You're ok, even with 1 day. Probably the pressure goes away through the wall of your pipes, it's perfectly normal.

The bottle I'm using for transfers purging etc. Has a whole roll of line connected it goes down to zero in half day after I cliose the bottle.

 

[MashBag]

Yes indeed. Difficult to say if it's a leak or adsorption depending on piping and nonreturn valves.

A light solution of fairy liquid in a wash bottle will help (gas on)

 

[Brewnaldo]

Probably just in the realms of normal leakage tbh. These systems are never 100% leak tight. The volume of your closed in system is absolutely tiny. A miniscule leakage will cause the pressure to drop. If its doing so over 2 days that's a negligible leak IMO.

To give you a real world example of what I mean.... When you put on Breathing Apparatus, this sort of check is part of the pre use procedure. You open the bottle then close it in to check the integrity of the hoses and connections up to the regulator. A Leak rate of 10 bar per minute or less is deemed acceptable for use.

 

[MashBag]

I do hope that's not a fire service rule? You only have 207 bar in the cylinder and should be heading out at 50??

 

[DocAnna]

I am no expert Anna but the secondary will let the gas still go into the beer so I would expect it to go down as it is absorbed by the beer that is assuming you do not turn them off and only the main gas at the cylinder.
This will also eventually go down on the primary too I would think as that is connected to the secondary's.
Remember I am no expert on this.
Ps you are getting to the end of your woes

The pressure drop occurs even when I isolate all the kegs - I'll post some photos later. Yes, getting there on the brew space recovery.

You're ok, even with 1 day. Probably the pressure goes away through the wall of your pipes, it's perfectly normal.

The bottle I'm using for transfers purging etc. Has a whole roll of line connected it goes down to zero in half day after I cliose the bottle.

I hadn't thought about pressure loss through the walls of the pipe - it does seem really quite a slow leak.

 

[Brewnaldo]

I do hope that's not a fire service rule? You only have 207 bar in the cylinder and should be heading out at 50??

10 bar an minute from the tiny volume of the lines equates to basically hee haw from the comparatively massive volume of the bottle. That's the point I was making. It's about volume. You will visibly see a decay in almost any high pressure, low volume system, especially one made up with push fittings etc.

There's an acceptable leak rate for all "leak tight" systems. Even in the likes of oil/gas/petrochemical plant. Taking 2 days to lose the pressure in some closed in tubing strikes me as a pretty good result tbh

 

[DocAnna]

There's an acceptable leak rate for all "leak tight" systems. Even in the likes of oil/gas/petrochemical plant. Taking 2 days to lose the pressure in some closed in tubing strikes me as a pretty good result tbh

Thank you - coming from someone with your expertise in such things is very reassuring. I'll stop prowling and spraying the leak detector everywhere looking for bubbles

 

[Brewnaldo]

Thank you - coming from someone with your expertise in such things is very reassuring. I'll stop prowling and spraying the leak detector everywhere looking for bubbles

Never hurts to stay vigilant on these things of course.... I have experience of replacing a bottle that had carbed a grand total of about 4 beers due to a shortfall of said vigilance

 

[dtashmore]

if your setup will fit in the bathtub that would be the final check.

 

[Jiffy176]

You are not the only one Anna, switched my 3/8 hose for 3/16 for better flexibility to prevent such a leak and then spent days looking for it. Gave up in the end and have left gas on. But this post reassures I am not going mad. Over this anyway.

 

[MashBag]

10 bar an minute from the tiny volume of the lines

Sorry, I can't get my head around this.
Whatever the volume, 10bar per minute did strike me as unacceptable. That is 10bar per min plus with the gas on.

If push fit is that bad, surely it is not fit for purpose.

Push fit works for water, which I know is a different molecular weight, but really.

As for it leaking through the pipe. Yes indeed HDPE is gas permeable but not like a sieve.

If this is indeed an acceptable failure/leak rate I am stunned that all of the installations all over the world leak co2 like this.

Just can't get me head around it....

 

[MashBag]

On a separate note. Well done Anna, I can't imagine the amount of work you are having to put in. All for the love of beer. Well done. Well done indeed.

 

[Brewnaldo]

Sorry, I can't get my head around this.
Whatever the volume, 10bar per minute did strike me as unacceptable. That is 10bar per min plus with the gas on.

If push fit is that bad, surely it is not fit for purpose.

Push fit works for water, which I know is a different molecular weight, but really.

As for it leaking through the pipe. Yes indeed HDPE is gas permeable but not like a sieve.

If this is indeed an acceptable failure/leak rate I am stunned that all of the installations all over the world leak co2 like this.

Just can't get me head around it....

The breathing apparatus thing is/and has been for years, taught to me as normal donning procedure by trained firefighters at me work. Whether fire and rescue at large use it I couldn't say. I could find out though.

As for processing plant, leak tests are generally done using nitrogen with trace helium because helium both leaks easier and is easy to detect in small concentrations. When you leak test a plant, the acceptable helium detection from any given flange or joint is not zero. Again, I can't put a number on it atm but it's not zero.

Likewise with routine LDAR (leak detection and repair) audits, there is a leak rate threshold below which, even though hydrocarbon is fetched by the LDAR meter, it is not considered a leak.

 

[Graz]

@DocAnna

Mine do this as well, primary and 3 secondaries. All secondaries isolated from the kegs with their built in shut off valves when I turn off the gas.

Quite confident I don't have a leak as when I did have one in the past it was quite evident from the rate I was getting through CO2. Traced to a rubbish push fit shutoff valve that producing a slow, steady, stream of bubbles when put in water. Now put in the bin

When I don't have any such leaks a cylinder lasts me ages so even though my regulators drop to 0 over a few days so pretty confident that even if it is leaking out somewhere the loss is insignificant.

 

[hoppyscotty]

I have a cylinder attached to my keggorator and it holds pressure so I see no reason for pressure to drop if you don't have a leak. My second CO2 cylinder I use in my brewery I will close up and release pressure in the regulator when not using for no particular reason just a habit I got into. But if I did leave it on I wouldn't expect it to lose pressure.

 

[DocAnna]

I've posted my CO2 plumbing diagram before on my own thread and photos of the setup here as well. I've had another go today at finding the leak, sprayed everything with all manifolds closed and open, still can't find anything. I'm leaving the cylinder closed unless I need to use the CO2 I think from now one.

https://www.thehomebrewforum.co.uk/threads/annas-brewdays.89471/post-1220066

 

[James]

I don’t believe you should be concerned by that leakage rate, as others said - calculate the pipe volume and do the maths. You’ll be putting more out yourself.
However, it’s still a potential hazard if something goes wrong. Weakest link and all. Gas cylinder valves aren’t really designed to be turned on and off frequently.
You might consider a CO2 alarm.

 

[MashBag]

I have a cylinder attached to my keggorator and it holds pressure so I see no reason for pressure to drop if you don't have a leak. My second CO2 cylinder I use in my brewery I will close up and release pressure in the regulator when not using for no particular reason just a habit I got into. But if I did leave it on I wouldn't expect it to lose pressure.

Thanks for the that. Mine never did. But I began to think it was just me expecting to much of gas tight fittings.