Pressure testing gas line. I lost 1psi overnight...

[ripsnortMN]

Hey guys, moved my black pipe propane line to a different wall in the garage. I pumped 70psi into it with a shutoff at each end "before the diaphragm" and this morning my gauge read 69psi.

I soaped all the joints and I can't find a leak. I don't want to disassemble all the pipe unless I know it's leaking. How can I tell if its sealed or not? It is five degrees colder in there than last night. Could that drop me 1psi from 5 degrees colder? Any help would be awesome guys.

 

[tcianci]

IMHO, the drop is probably from the temperature change.

 

[regguy1]

IMHO, the drop is probably from the temperature change.

I agree, a leak will have a much higher drop in my experience

 

[larry4406]

I would agree likely a temperature effect. Bubble test all of your fittings, not just the joints. I have had porous cast fittings that leaked thru the casting and not at a joint.

 

[norcalpablo]

I have no academic background in this area, so I’m sure some of this is totally wrong…but as I understand it (ideal gas law), if volume is fixed, pressure is directly related to absolute temperature. Since you’re in Minnesota, I assume you’re saying the temperature dropped five degrees Fahrenheit. Maybe you started at 65F (or 291.48 Kelvin) and ended at 60F (288.71K). The temp decreased 2.78 degrees Kelvin, or a .95% change. A .95% pressure drop from 70 psig is .665 psi, accounting for most of the observed drop.

Add any parallax error (assuming analog gauges) and the fact that your line isn’t filled with an ideal gas (moisture in compressed air might condense, etc) and I’ll bet you’re in good shape!

 

[jhelrey]

Gas companies will pressurize a system up to 15 PSI or there about and time it for 13 mins. If it does not drop in that time, they are allowed to leave. My times and pressures maybe wrong but you are good to go.

 

[mrobins297aaa]

your good, soap bubble all the joints and like someone said before all the fittings incase there is any pin holes in the castings.
sometimes all that testing just creates leaks since ng is around a 1/4 # of pressure and your pumping the system up to 70psi you make things that would never leak, leak.

around here the gas company doesn't pressure check and we never did either.
you just pencil mark the the face of the most sensitive dial on the meter and if you have any kind of leak the pointer will move, I think its the 1/2 cubic ft dial, they even call it the checking gauge.
when we did new housing we would pipe the whole house to there templet and call for the meter they would come install the meter, mark the gauge, watch the gauge for a little bit and if it didn't move that was it, we never check any of the joints.

oops just noticed he has propane, still most of the info is related

 

[ptschram]

I have no academic background in this area, so I’m sure some of this is totally wrong…but as I understand it (ideal gas law), if volume is fixed, pressure is directly related to absolute temperature. Since you’re in Minnesota, I assume you’re saying the temperature dropped five degrees Fahrenheit. Maybe you started at 65F (or 291.48 Kelvin) and ended at 60F (288.71K). The temp decreased 2.78 degrees Kelvin, or a .95% change. A .95% pressure drop from 70 psig is .665 psi, accounting for most of the observed drop.

Add any parallax error (assuming analog gauges) and the fact that your line isn’t filled with an ideal gas (moisture in compressed air might condense, etc) and I’ll bet you’re in good shape!

Good show! Very good show.

 

[ripsnortMN]

Well guys, its been 48 hours since i filled it with air and it looks to be about a 1.5 pound pressure loss. Two days ago I filled it to 70psi and now its a hair below 69psi. Can I call it good? Or should i give it another day?

 

[sberry]

It would be whizzing out at 70 if there was a leak, wouldn't take over night, as long as it is isolated you could pump it as hard as you wanted. If you soaped this deal and no bubbles you are good to go.

 

[Frank The Plumber]

It should be fine. You have to remember that you most likely filled the lines from an air compressor. The air coming out of an air compressor is elevated in temperature above ambient by a few but varying degrees. 70 to 69 is good to go over 48 hrs. Plus you lost your first pound initially and have not lost more. And you bubble tested. It should be golden.

 

[Steevo]

When I pressure tested my gas line into my shop, I lost a few pounds the first day, then gained a couple when it warmed up the next, then lost a few more, etc. After a week, I was down 10# from where I started, and no joints were leaking with soap bubble test. I finally determined that I had a leaky gauge (cheap china made ****), and bought a good quality one. It held stead except for temperature variation for six months, when I disconnected it to install the heater.

 

[trbomax]

The real deal here is that the propane supplier should be testing it. If a new appilance or new lines are installed they have to test it before supplying gas. This is a liability and insurance issue for them and for you. I'm not telling you to do it or not to do it,just how it is.

 

[rockchucker]

Boyle's, Charles and Gay-Lussac's Laws.

http://en.wikipedia.org/wiki/Gas_laws

The pressure exerted on a container's sides by an ideal gas is proportional to the absolute temperature of the gas.

So yes with a temperature decrease of 5° a 1 PSI drop is very feasible.

 

[ripsnortMN]

The real deal here is that the propane supplier should be testing it. If a new appilance or new lines are installed they have to test it before supplying gas. This is a liability and insurance issue for them and for you. I'm not telling you to do it or not to do it,just how it is.

Its supplying a modine hotdawg. That has been in service in the garage for two years. I am just rerouting my line. It runs off of a 100LB tank which is outside.

 

[Frank The Plumber]

Just don't put any pressure on the control valves, you'll mess them up. Or against the gas meters either.

 

[trbomax]

Its supplying a modine hotdawg. That has been in service in the garage for two years. I am just rerouting my line. It runs off of a 100LB tank which is outside.

Thats a different thing. I've never used 100 pounders! How long does one even last running the heat?

 

[ripsnortMN]

Just don't put any pressure on the control valves, you'll mess them up. Or against the gas meters either.

I have shutoffs at each end.

 

[ripsnortMN]

Thats a different thing. I've never used 100 pounders! How long does one even last running the heat?

Believe it or not it lasts most of the winter on one tank. I only heat it a few days a week. I will usually go through 1.5 tanks a year. So it only costs me about $100 a year.

 

[rlitman]

When I pressure tested my gas line into my shop, I lost a few pounds the first day, then gained a couple when it warmed up the next, then lost a few more, etc. After a week, I was down 10# from where I started, and no joints were leaking with soap bubble test. I finally determined that I had a leaky gauge (cheap china made ****), and bought a good quality one. It held stead except for temperature variation for six months, when I disconnected it to install the heater.

Considering that the gas pressure is well under 1PSI I would think that to be a little overkill.

On that note, consider that whatever temperature the air leaves your compressor, compressing the air in the pipes will heat it up (it is "adiabatic compression"). Even if it didn't get colder overnight, the air will cool as the heated air loses heat to its surroundings.

 

[ishiboo]

Considering that the gas pressure is well under 1PSI I would think that to be a little overkill.

On that note, consider that whatever temperature the air leaves your compressor, compressing the air in the pipes will heat it up (it is "adiabatic compression"). Even if it didn't get colder overnight, the air will cool as the heated air loses heat to its surroundings.

Yup, I agree. The LP company pressure tests with all of the final devices connected - 70psi may very well kill your gas valve, and testing without it connected will not ensure the device supply does not leak.

2 weeks ago I had my tanks combined into one 1000 gallon and relocated so its no longer visible from the house. I have the service slip next to me, they charged both lines to 10 PSI and let one sit for 9 minutes and another for 11 minutes. This is for the low-pressure in the house only, I am not sure if they tested the higher pressure line between the two single-stage regulators separately. Those are at 10 PSI operating. Any line inside the building can only legally be 11" WC or so, which is less than 1/2 PSI. Yup, less than .5 PSI.

 

[Frank The Plumber]

The reason that you would pressure test at a higher pressure say 70 psi is because the pipe sealant may tend to leak at those pressures. By charging the lines to 70 you are dislodging anything that could cause trouble later. Copper lines can hold together unsoldered at 50 psi so we run them up to 120 PSI. I test most of my fresh install gas lines at 120 also. At 120 you can hear it. I've pressure tested lines that these guys pass at 10 PSI and find a bunch of leaks. They just do that because they can use a hand pump.

 

[HoosierBuddy]

To the OP. There is no chance that you have a "hazardous" leak based on your reported testing. When gas companies (or propane companies) do testing...that's what they are trying to determine.

Based on extensive pressure testing I've done over the years I've come to the conclusion that all threaded joints leak to some microscopic degree. Molocules of gas will escape over time. Even certain types of pipe (all plastics for example) will bleed through molocules of gas over time, right through the pipe wall. These leaks are not even detectable by any means you are likely to have at your disposal. So...I'm going to say I'm quite confident your gas line IS leaking, but this will not create a hazard for you.

A little more information I want to give everyone on pressure testing:

Don't test against a valve. They will leak through eventually. Test against a plug, cap, or blind flange.

Pressure testing is a strength test as well. That's why it is important to test at a higher pressure than the system will see in use. However, 70 PSI seems excessive for a propane system that will likely run at less than 14 inches w/c.

When "soap" testing, don't use soapy water. Use a commercially available leak testing solution. They are many times more effective than using some of your wife's dishwashing liquid mixed with water.

And here's one I learned the hard way that almost no one in the world knows about (except me and now YOU)...when you pressure test polyethylene pipe to high pressure with air NEVER vent the air after the test by throwing open the valve and letting the air rush out at high speed. If you do, you can generate high voltage potentials due to static electricity buildup in the pipewall that can cause microscopic sparks that are capable of pinholing the PE pipe. So...your pressure test was good...but you just created several pinholes, each a few microns in diameter, in your pipe when you let the air out. Instead, bleed off your air slowly through a partially cracked valve (that you just removed the plug from...because you never test against just a valve.) This will provide some protection against static buildup along the inner wall of the plastic pipe.

That last one was an expensive lesson to learn.

Phil

 

[wolftrucking]

Thinking allong the lines of temp and psi think of why you check your tires cold. warm tires give a higher reading than cold tires. Keep in mind propane is heavier than air so it will not escape like N.G. if there is a leak. If you have any doubts or even a gut feeling cough up the coin after you hook up the furnace and have the fuel company test it. Even if it checks out its money well spent and you wont level the neighborhood. By The Way, get a report and store it with your homeowners insurance papers so if anything were to go wrong you have your a$% covered. Rember God gave you a gut feeling for a reason best to follow it.
good luck.