Gas line pressure testing

[Oberlander]

The gas company recently ran natural gas line up my road and I opted to tie in. They ran the meter to the rear corner of my house and I installed a radiant heater in my pole barn as the first consumer to get the meter installed. Ran 1" PE with risers and Homeflex underground compression style fittings for the underground connections. Turns into black pipe inside building in 3/4" size. I pressure tested the system to 10psi for 72hrs before calling them out and had zero drop. They came and tested with manometer and indicated there was a leak. I have now tested it with my tester at 10psi for 13 days. It has dropped to 9.5 psi. Is it possible for them to detect a leak that I cannot? I have sprayed all joints and see no bubbles. I blame the .5psi drop on the recent temperature changes (20F to 70F to 30F) in my area and also the fact that I just fired up my compressor before filling so the air was hot when it went in. I have scheduled them to come back out to try again.

*Also note that the Teflon tape is only adapting the tester. All joints are sealed with Rectorseal 5.

Any help greatly appreciated.
Derek

 

[Movover]

Do you have the line capped off and not going into the heater? you should not be pressure testing the line with any appliances attached is what I was told especially at 10 PSI

 

[Oberlander]

It is capped at both ends. Not pressurizing against any appliances, valves, or meter.
Thanks
Derek

 

[Jlbc212]

Another way to do this is to "clock the meter." Turn off the shut-off valve at the heater. Make sure the gas is turned on at the meter. Take a picture of the reading at the meter. After a period of time (the longer you wait the more accurate the determination will be) recheck the reading at the meter. If the reading hasn't change there's no gas leak.

Where did the gas company connect their manometer? It could be the leak is at their connection.

You are correct about the pressure drop caused by a drop in temperature and the warm temperature of the air you filled the piping with. That is normal.

 

[Oberlander]

The gas company adapted their manometer and hand bulb pump to the meter connecting flange. They tested a couple times on a very cold day and stated it failed. I questioned their connections and the tech was bothered by that so I let it go and told him I would check into it further. They would not install a meter as I had already buried the underground connection points. He asked that I expose all underground connections for the return trip so I have dug them all back up and solution checked myself with no evident leaks.

Thanks for the help.
Derek

 

[gmcgeo]

The gas company adapted their manometer and hand bulb pump to the meter connecting flange. They tested a couple times on a very cold day and stated it failed. I questioned their connections and the tech was bothered by that so I let it go and told him I would check into it further. They would not install a meter as I had already buried the underground connection points. He asked that I expose all underground connections for the return trip so I have dug them all back up and solution checked myself with no evident leaks.

Thanks for the help.
Derek

on a ng system the only time i have used the manometer was to set the flow pressure of the meter on a new install. best way to test for a leak is to clock the meter that will show up a leak. the other thing if its say 70 deg out side when you pressure test it and its 10 psi and the next morning its 45 to 50 deg out i have noticed it will drop about 1/2 psi to 1 psi , now i came from Arizona and with the extreme heat this is where i have seen that do that. our inspectors out there will give a 1 psi drop a pass in a 24 hour time period

they can take a sniffer and test the lines on the outside and underground lines to narrow down the leak after the meter is there.

 

[Oberlander]

Thanks gmcgeo, I am hoping the return trip goes smoother and they install a meter. I just wanted the opinions of others as to whether my test results could possibly invalid.
Derek

 

[gmcgeo]

Thanks gmcgeo, I am hoping the return trip goes smoother and they install a meter. I just wanted the opinions of others as to whether my test results could possibly invalid.
Derek

hope it goes well for you!

 

[walta]

My guess is when they tested they pulled a vacuum. Do you have a vacuum pump and gage? It would be best to test the same way they will.
Walta

 

[mm08822]

0.5 psi drop in 13 days is nothing to be worried about. I agree temp differences is the source of the minor change. Try reading at same time of day assuming there arent big weather changes.

If you had a leak the pressure would continue to drop even if varying temps were creating noise.

Inspectors in my area want to see a 30 psi gauge reading 15 psi after 24 hours. 14 psi passes.

When the installer comes out next, be ready to soap up his connections.

13 psi is approx 50x their working pressure.

 

[HoosierBuddy]

Inspectors in my area want to see a 30 psi gauge reading 15 psi after 24 hours. 14 psi passes.

That's crazy.

In Indiana, state code requires customers lines be pressure tight to the appliance at the time of meter turn on.

Any unexplained pressure drop would fail the line.

To the OP....you indicate the line lost 1/2 PSI after several days. Is it still losing pressure?

If the issue is temperature fluctuation, it should stop losing pressure at some point. Heck, if it gets warmer it should gain pressure.

ALSO....I used to think dish soap was a reasonable substitute for leak detection solution and I found out I was wrong. I was unable to find leaks with dawn that were simple to find with commercial leak detection solution...so if you haven't gotten some....suggest you do so and check it all again.

Phil

 

[Oberlander]

Thank you all. In the first several days it dropped to 9psi and stabilized. Then with increase in temperature came up to 9.5 and is fluctuating between 9-9.5 with temp changes. Gas company returns tomorrow to check again.

 

[HoosierBuddy]

Thank you all. In the first several days it dropped to 9psi and stabilized. Then with increase in temperature came up to 9.5 and is fluctuating between 9-9.5 with temp changes. Gas company returns tomorrow to check again.

The pressure increase is a good sign.

Good luck with the gas company!

BTW 1 PSI drop would be explained by about a 55 degree F change in temperature in your test air. That's why the post above where they allow 1/2 to pass doesn't make sense.

Phil

 

[mm08822]

0.5 psi drop in 13 days is nothing to be worried about. I agree temp differences is the source of the minor change. Try reading at same time of day assuming there arent big weather changes.

If you had a leak the pressure would continue to drop even if varying temps were creating noise.

Inspectors in my area want to see a 30 psi gauge reading 15 psi after 24 hours. 14 psi passes.

When the installer comes out next, be ready to soap up his connections.

13 psi is approx 50x their working pressure.

That's crazy.

In Indiana, state code requires customers lines be pressure tight to the appliance at the time of meter turn on.

Any unexplained pressure drop would fail the line.

To the OP....you indicate the line lost 1/2 PSI after several days. Is it still losing pressure?

If the issue is temperature fluctuation, it should stop losing pressure at some point. Heck, if it gets warmer it should gain pressure.

ALSO....I used to think dish soap was a reasonable substitute for leak detection solution and I found out I was wrong. I was unable to find leaks with dawn that were simple to find with commercial leak detection solution...so if you haven't gotten some....suggest you do so and check it all again.

Phil

Let me rephrase that…..They require the use of a 30 psi gauge with a system charge at mid-range - 15psi. Fill the system to 15 PSI. Test duration is 24 hours.
This covers fluctuating temps and the vapor pressure component of wet air.

Indiana Fuel and Gas Code only requires a 10 minute test at 3psi - common to many locations. That's not a lot of time to see small changes. I'll take the higher pressure and longer duration.

10 - 15 PSI/24 Hours is done in many locations.

 

[metlmunchr]

If you charged the line at 70*F then a temperature drop of 22-23 degrees would drop your pressure by 1 psi if you do the calculation based on air as an ideal gas. To calculate based on psychrometric properties of the air, you'd have to know both dry and wet bulb temperature of the air, which wasn't given.

Pressure and temperature changes have to be calculated based on absolute temperature and pressure rather than gauge pressure and Farenheit temperatures.

 

[mm08822]

Everything can be converted and calculated but not necessary. You fill the system to a known psi. Expect a minor fluctuation based on time of day temps, humidity, where pipe is run. Everything comes to an equilibrium. Parts of system may also be exposed to sun, shade or entirely in basement during the test period.
As long as the system doesnt keep dropping pressure but rather holds it, all is good. Minor gauge fluctuations are not a concern.

 

[metlmunchr]

Everything can be converted and calculated but not necessary. You fill the system to a known psi. Expect a minor fluctuation based on time of day temps, humidity, where pipe is run. Everything comes to an equilibrium. Parts of system may also be exposed to sun, shade or entirely in basement during the test period.
As long as the system doesnt keep dropping pressure but rather holds it, all is good. Minor gauge fluctuations are not a concern.

Agreed. Just correcting the above statement saying it would take a 55* temp drop to drop the pressure by 1 psi.

 

[HoosierBuddy]

Agreed. Just correcting the above statement saying it would take a 55* temp drop to drop the pressure by 1 psi.

Metlmunchr is correct.

I pulled a rookie mistake on my calculations and used PSIg instead of PSIa.

It was a long day yesterday.

Phil

 

[mm08822]

If you charged the line at 70*F then a temperature drop of 22-23 degrees would drop your pressure by 1 psi if you do the calculation based on air as an ideal gas. To calculate based on psychrometric properties of the air, you'd have to know both dry and wet bulb temperature of the air, which wasn't given.

Pressure and temperature changes have to be calculated based on absolute temperature and pressure rather than gauge pressure and Farenheit temperatures.

Which directly supports what my inspectors require:
15 PSI setpoint
14-15 PSI actual after 24 hours.

20-30F delta is easy to find between day/night and day/day.

 

[TRWham]

Which directly supports what my inspectors require:
15 PSI setpoint
14-15 PSI actual after 24 hours.

20-30F delta is easy to find between day/night and day/day.

Our AHJ (for most projects anyway) wants 15 psig on a 30 psig gage so they can see the needle in the middle.

 

[Oberlander]

Well todays technician at least acknowledged my 17 days of 9psi and did his own 10 minute test, concluding it is holding. He installed the meter, set pressure, and even helped me purge the lines up to my new shop heater. I only wish the first visit went this smoothly. Thanks to all who shared their input, it is greatly appreciated.
Derek Oberlander

 

[HoosierBuddy]

Good deal. Just in time for 70 degree weather, right?

Phil

 

[Oberlander]

HoosierBuddy,
Yes my thoughts exactly! Cant believe they made me second guess myself into chasing a nonexistent leak. Month wasted waiting on ground to thaw so I could reexpose the connections.
Derek

 

[mm08822]

Last guy to replace my water meter was a bull in a china shop. Dinged 2 walls at separate times with his tool belt and bucket - just walking thru.
Also clueless - he mounts a new exterior meter read device next to the holes he left from the previous one. You couldn't cover the old ones with the new device?