10 replies to this topic

[GaryNConcord]

Disclaimer on the front end.  I've just started kegging a few months back so still learning and mostly ignorant.  

 

Hooked up a 5 lb cylinder this past week.  Next day pressure had dropped from ~1000 psi to 650.  It has held there there for 3 days.  This morning I took the two kegs and cylinder out of the kegerator and drenched every connector/fitting/relief valve/gage with soap solution.  Nary a bubble anywhere.

 

I've come up with 3 possibilities:

1-I've got a very small leak that leaked at the higher pressure, but is holding now.

2-The pressure drop was due to cooling the cylinder from room temp to ~45 F.

3-Little magic gremlins are screwing with me...  

 

#2 doesn't seem likely to me-seems like a big pressure drop for the temp change, and 3 is unproveable.  

 

Anyone with any ideas or insight?

 

Kegs have relatively new o-rings, however I'll be replacing them next refill just to cover that base.  

 

Gary

[neddles]

I'm betting on #2.

[JMcG]

I usually see a drop in measured pressure as I keep my gas canister in the cooler, too.  If you are carbing that will also drop it further as the beer aborbs the CO2. You could also take it out and weigh the canister to see if it has lost significant amt of gas.

[Darterboy]

One way to be sure is to submerge your entire draft system, including suspect kegs, in the shallow end of a swimming pool while completely rigged for serving. A member of our club did this to finally discover where his leak was a couple of years ago. Good last resort method.

[GaryNConcord]

Darterboy, that may be the best excuse for putting in a swimming pool I've seen  .

Next time I'll weigh my cylinder before I hook it up, and then again if I see the same drop.

Thanks for the responses!

[ChefLamont]

Your CO2 cylinder will always read around 650 pounds until it is almost done.  I dont know how/why it read 1,000 pounds.

 

CO2 in the tank is in liquid form.  It sits in equilibrium with the gas.  The temperature of the tank determines the equilibrium point and therefore the temperature,  so it will read 650-700 the whole time until the liquid is gone and the pressure actually starts going down.  This is why I never bothered putting a high pressure gage on my systems.  They tell you very little. 

 

The better ways to test is one or more of these:

 

1. Use a starsan or soap spray to find leaks.  Spray everything down and look for bubbles.

2. Submerge things in water and look for bubbles.  I wouldnt submerge the regulator though.

3.  Pressurize the system then shut the gas off at the tank. The pressure one the low side should remain pretty consistent for at least a day or two, but really longer.  This also assumes, if attached to beer, that it is fully carbed and you are not pulling beers off the keg.

[JMcG]

The pressure reading is dependent on the tank temperature.  Lower temperatures mean less CO2 in gas form, thus lower pressure reading in the tank.  Agree with all said about starsan/soap spray and submerging to find leaks, but weighing the tank (without your regulator) and subtracting the tare weight (TW stamped on the tank) will tell you how many pounds of CO2 remain in the tank.

[Stout_fan]

Your CO2 cylinder will always read around 650 pounds until it is almost done.  I dont know how/why it read 1,000 pounds

Room temperature of 451°F

[Brauer]

When I used to maintain house CO2 systems, this is what I would do each time I changed a tank, to make sure there were no new leaks:

3.  Pressurize the system then shut the gas off at the tank. The pressure one the low side should remain pretty consistent for at least a day or two, but really longer.  This also assumes, if attached to beer, that it is fully carbed and you are not pulling beers off the keg.

Once you know there is a leak in the system, this is a good way to find where the leak is:

1. Use a starsan or soap spray to find leaks.  Spray everything down and look for bubbles.

Another option for finding leaks is to isolate sections of your system and look for pressure drops when you open up the next section. For example, if you have a valve after your regulator, close it with the system pressurized and the tank valve closed, and see if there is a loss in pressure that would indicate that the leak is at the regulator. Then, open that valve and remove the QD at the keg, to see if the leak is at the keg end.  If you have a manifold with valves, you can also use that to test each leg in the system in the same way.

[GaryNConcord]

[font="calibri;"]Mystery solved:  Got hold of a phase diagram for CO2.  I can’t cut and paste, but if you google -Praxair, Carbon Dioxide Safety Precautions- it’ll take you to the document I’m looking at (p.2) .  Any CO2 phase diagram should give you the same thing though.[/font]

[font="calibri;"]First off, CO2 condenses to a liquid with a gas phase on top in the cylinder at room temperature.  That kills using the ideal gas law.  The tank pressure will be the vapor pressure of the CO2, which is a function of temperature.  When I brought the cylinder home, it rode in my truck which doesn’t have AC, on a day in the low 80’sF.  That matches up with the diagram showing 1000 psi when the temp is between 80 and 90F.  (The graph stops there.  If I understand it correctly, once you exceed a max temp of somewhere around 85-90 for CO2, discrete phases of liquid and gas no longer exist and pressure will ramp up with temp.)  Any rate, the 1000 psi makes sense for the conditions the tank was exposed to.  [/font]

[font="calibri;"]My kegerator is set at about 45F.  Following the curve down, at 45, the vapor pressure is right at 600psi.  Throw in a little allowance for error on the part of the gage, and I’d say what I observed was pretty much dead on for what I should have seen  for the temperatures.  [/font]

[SchwanzBrewer]

(P1*V1)/T1 = (P2*V2)/T2

 

If volume is constant that variables V1 and V2 drop out. Make sure your units are right.

 

So...

 

P2 = P1*T2/T1

 

Which is basically saying that the pressure drops (P2) when you decrease T2.