[enso]

I continue to be driven crazy trying to find the CO2 leak in my system.  I have replaced hoses, redid connections and gaskets, everything I can think of.  Still leaking.  I believe I know the general source of the leak from shutting off different parts of the system.  It seems to be somewhere between the manifolds in the fridge and a T of shutoffs after the primary regulator.  I am hoping I am wrong but starting to wonder if it is the bank of 3 secondary regulators.  I have checked the connections and they appear to be sound.  So I am wondering if there is a leak in the regulators themselves.  Man I hope I am wrong.

How can I rule this out?  If there is a leak can it be fixed?  They are perlick regulators.  Used I think.  I bought them from American Science and Surplus.  They were working well for several years.

[tschmidlin]

Have you tried spraying with soapy water to find the source of the leak?

[saintpierre]

Assuming a pressurized keg and your keg/connections don't have a leak wouldn't a pressure drop in the regulator (when you turn the gas on) indicate a leak?

[gordonstrong]

Yup, spray with soapy water and watch for bubbles. Keep some handy in a spray bottle for troubleshooting leaks.

Are you able to swap regulators and see if you still have the problem?

[bluesman]

A spray bottle of Starsan will also do the trick,  plus it will help keep your beer fridge sanitized. 

[enso]

Yes, I have been spraying all connections insescently with star san, even dunking ones that I can into a container of star san.  I just can't find any leaks.

Here is what I have done briefly.    

I first checked the primary regulator.  Sprayed the connection between the tank and regulator.  No bubbles.  Charged up the regulator noted where the needles where on the gauges and shut off the CO2 bottle.  Left it for about a day.  No drop in pressure.

The line going out of the primary regulator is a short hose connected with a swivel hose barb and it travels to a T junction with 2 shutoff valves.  I sprayed the connection coming out of the primary with star san with the CO2 open.  No bubbles.  I then dunked the T connection with the shut-off valves turned off.  No bubbles.

One shut-off goes directly to a barbed swivel attached to a flare fitting QD.  I opened that valve and dunked the QD in star san.  No bubbles.  Shut the valve.

The other side of the T goes to a bank of 3 secondary regulators.  I turned the pressure control screws all the way open so no pressure was going out of the regulators and opened the valve at the T.  I sprayed the barb fitting going in to the regulators, the fittings between the regs, and the end plug.  No bubbles.

The first 2 regulators in the series are each connected to a two way manifold inside the kegerator.  The third reg. goes directly to a barb swivel and flared QD.  I shut all the manifold valves off and had the QD not connected to a keg.  I then pressurized one reg. at a time, noted the pressure, and turned off the gas back at the T.  I come back in about 8 hours and...  The pressure has dropped.  I repeat the process with each regulator individually.  Same deal.

I replaced all the hoses going in to the regulator bank and traveling to the manifolds and QD with new 3/16" beverage tubing and checked them again.  Still had a pressure drop.  If left long enough it would drop all the way to 0 by the way.

So?  Is it the regulators?  Can anything be done?  Am I missing something to check?

 

[gordonstrong]

Could you post a picture? I'm having trouble visualizing your setup.

Are the bank of three regulators one unit, or three individual regulators hooked together somehow? Can you isolate/replace the first regulator in the series? That seems to be a common part of your problem.

Did you test the manifolds inside the kegerator? Are you able to cap off the out side of the regulators?

I tend to think you're right that it's a problem with those regulators but I would try to eliminate all down-stream components to verify that theory. I don't know how much you can attack pieces of the regulator, though. If it's one unit, what model?

[bluesman]

A picture would certainly help. 

I think you need to isolate (leak test) each component independently through process of elimination which is done one step at a time.  If the keg holds pressure then check the regulator by swapping it out with a known good regulator,  etc...  If you don't have a second regulator you have to rely on soapy water to test each component from the cylinder to the keg.

[enso]

Okay, here we go.

Starting at the CO2 bottle the line out goes to a T with two shut-offs.  One goes to a QD as shown, the other travels up to the bank of secondary regs.  

http://www.flickr.com/photos/68291744@N00/4926411257/in/photostream/]http://www.flickr.com/photos/68291744@N00/4926411257/in/photostream/

Next we have the three secondary regulators.  They are attached via pipe nipples and the end of the line is plugged.

http://www.flickr.com/photos/68291744@N00/4927004748/in/photostream/
the front

http://www.flickr.com/photos/68291744@N00/4926408647/in/photostream
the back

http://www.flickr.com/photos/68291744@N00/4926412545/in/photostream
the end.

Next, the first 2 travel on to manifolds inside the kegerator.  Normally these are inside.  I pulled them out and reattached to test them easier.

http://www.flickr.com/photos/68291744@N00/4926406739/in/photostream/

The third regulator just goes directly to a QD.  I did not take a picture of that.

That give a clearer picture or perhaps suggest a problem area?

The regulators have a sticker with the following information
"Perlick corp. Model 40115, Secondary CO2 compressed gas regulator 366M, 1800 Maximum Inlet Pressure."

[enso]

Please stand by, techincal difficulties...  Where are my PICS!?

Sorry folks.  Best I can do is post the links to the flickr page.  I cannot seem to embed the pics properly.

[bluesman]

I think you need to isolate each component one at a time and through process of elimination you'll detect the source of the leak.  Start with the CO2 cylinder and work through it one component at a time using one regulator in-line to the keg.  Attach a pressure gage to the keg and pressurize the keg to 30psi.  Use soapy water and start the process of elimination by introducing one component at a time.

[tschmidlin]

You can take the regulator bank apart and isolate them individually for testing.  It'll be a total pain, but it's the best way to do it I think.  I'm going to be doing the same thing pretty soon, I've got a leak somewhere in the new system I just put together.  I'm hoping it's just one of the kegs.

[enso]

I think you need to isolate each component one at a time and through process of elimination you'll detect the source of the leak.  Start with the CO2 cylinder and work through it one component at a time using one regulator in-line to the keg.  Attach a pressure gage to the keg and pressurize the keg to 30psi.  Use soapy water and start the process of elimination by introducing one component at a time.

Pressure gauge?  Something else I need to buy?  ...and then still have a leak!

[enso]

So, still no luck.  I made up a different soap solution for leak checking rather than star san.  I used dish soap and some glycerin.  I sprayed everything!  I mean everything, connections and all sides of the regulators.  Not a sign of bubbles, other than from the spraying action itself.  I have built a keg pressure tester/bleed valve.  Not sure if that will tell me anything as I have not even gotten to the stage of connecting the kegs...

This sucks.

[jalynn2]

Watch the gas connectors on the kegs -- I have had leaks there myself even though the o-rings look good. I solved my problem with a light coating of keg lube on the o-rings on the valve.