Test Pressure for Old Line

OK Start from the beginning. What is the true purpose of carrying out a test pressurisation of anything with a placebo fluid?

If you are worried that the pipe is going to rupture by running the standard test, wouldn't that be an indication that the pipe should be replaced?

At the end of the day, you're going to have far more down time and money involved by patching/replacing a piece at a time. Not to mention the explosion hazard that undetected leaks could pose.

How about carrying out a cost-benefit analysis on the two proposals, the one of doing it and the other of not doing it?

If it was me I would fill it full of water and pump it up to its designed 425 PSIG rating and hold it there for a hour or two. If it leaks or bursts it junk and needs replacing if it holds I would take it to be good but still do regular retesting anyway.

ASME B31.4 Liquid Transmission Piping Systems, section 451.7 is Derating a Pipeline to a Lower Operating Pressure.

The resulting situation from lack of or improper testing. This doesn't include the pending lawsuits.

"on January 13 2012, an independent audit from the State of California issued a report stating that PG&E had illegally diverted over 100 million dollars from a fund used for safety operations, and instead used it for executive compensation and bonuses."

-A-

I assume that you have been doing ultrasonic thickness testing on a regular basis and know the minimum thickness and location of the thinnest point. With that information you can calculate the theoretical rupture pressure. As others have said if you are afraid of the pipe rupturing with an inert fluid you have a real problem.

Thanks very much. This helps me.

Interesting situation!!

Allways engineers have the insufficient information problem....

It would be appropriate to test it with the working pressure+the pressure variatons existing in your system at least !! (if you know them)

For future perspective - any monitoring device for leakages-will be appreciated

....and ofcourse - luck...

Times have changed. Due to the various certifications nowadays (Quality, Environment, Health & safety etc. we have to cover ourselves from all sides. Otherwise we can think of many ways of doing the job safely, but have to have documented criteria and evidence to satisfy the auditors.

I stand to be corrected, but 1.5 x max working pressure is the appropriate hydraulic test-pressure for your repaired line, not the original design pressure.

Can you quote the code please.

Mmmm... I did state under correction ...

I may have confused myself momentarily, between MPOP and MAWP and Service Testing, providing a snap response from memory only.

My Apologies.

You can refer to ASME B31.3, B31.1 & B16.3 for further info, or believe Abdel Halim Galala instead.

That's a good thing, isn't it?

The design pressure of LPG piping systems shall not be less than P = 285 PSIG, where the hydrostatic test pressure must be not less than P_h = 1.5 P = 1.5 x 285 = 427.5 PSIG, and you can use P_h = 425 PSIG.

Thanks for the info. This is what we have specified, but Operations find it difficult to get the pressure up to this using the available N2 and I thought of helping them out without violating any code. They have raised upto 350psig and it is quite sufficient from all angles except the reference to a code..

Hold it...this is the first mention of a gas pressure test...we were referring to hydro testing as far as I could ascertain...gas has it's own hazards and parameters, and to my knowledge, not 1.5 x pressure.

Someone help out here please.

Stop this immediately, before someone gets injured/killed.

Recommence the test using an incompressible fluid, like water.

30 year old 16" header at 400+psig with N2... What could possibly go wrong?

I kid. This guy's a nut. Do you suppose these types of questions come from engineers who have no idea about potential energy stored in a compressible fluid (gas) or, are they non-engineer types who come here looking to get advice from actual engineers?

I ask because we get lots of "I'm about to blow myself (or someone else) to kingdom come, what's the best way?" type of questions and, it would hurt my feelings if these questions were coming from people who were allowed to call themselves "Engineer."

-A-

It does make one wonder, indeed.

Sounds like this guy is on his way to be nominated for a Darwin Award. Lets just hope he doesn't take out any innocent by-standers.

Bleed the N2 pressure off immediately ! (Stay calm but urgent, as I am)...

Ensure there is more-than-adequate ventilation at the bleed points (N2 is an asphyxiant gas).

If /when you approach this forum again, please state the whole !@*%$ scenario.

Example:

1. You have a repaired gas line (you did not say how it was repaired). The repairs (please God) were not patches, but new spools flanged into the line.

2. Each spool should have been hydro-tested independently prior to installation.

3. The spool (s) should then have been mounted in the line with all due care (proper alignment, proper spec gaskets, torque settings/ flogging procedures et al)

4. Now all you require is a Service Pressure / Leak test. This is normally hydraulic, but by exception eg when the line is fixed to a pressure vessel and cannot be isolated, you may apply for permission to perform a gas test.

5. Generally, the maximum gas pressure allowed here would be 1.1 x normal max operating pressure, and it is done with great circumspection, in adherence to an approved procedure, with the N2 residence time as low as reasonably possible in the line.

I'm not going to research any further ASME for you, as I now believe it is essential that you do this for yourself.

Remember this, that under failure conditions, the line will fracture most violently under gas pressure, whereas under hydraulic pressure, it will merely rupture at a weak point and de-pressurise.

As per ASME Code B31.3, a pneumatic leak test (using gas like Nitrogen or air) can be carried out in lieu of hydrostatic test. In this case the pneumatic test pressure P_h shall be 1.1 P = 1.1 x 285 ≈ 315 PSIG.

Notes.

1. As per ASME B31.3, Para. 345.5.2, during pneumatic test, a pressure relief device shall be provided, having a set pressure not higher than the test pressure plus the lesser of 345 kPa (50 psi) or 10% of the test pressure.

2. For LPG piping systems, if the air is used for pneumatic test, and before putting the pipeline into operation, a full purging of whole pipeline with nitrogen is must to discharge all oxygen inside the pipeline.

Thanks for all input. We have successfully tested to 1.1 times the design pressure using nitrogen as this falls into the category of B31.4. This is what we have done through the past as well. I was only trying to find if there are any other alternatives covered by any international code since Operations wanted the line back in service ASAP, which is understandable due to the current LPG prices. For your infomation this is a line supplying Liquid LPG to the shipping terminal for export and is in Middle East.

It doesn't matter what the line does or the price of the stuff inside it. If the line fails a pressure test, it really ought to do so with a placebo fluid inside it under a safe test regime and not the pricey stuff. Environmental disasters can be pricey too.

IMHO, the preferred test is 1.5X design pressure, hydro. the compromise test is 1.1X design pressure, N2. The higher pressure test looks further into the future and is MUCH safer to preform.

Compromising on test procedures should NEVER NEVER NEVER! be driven by "production wanting it back in service ASAP." I don't care if it gets me FIRED!! I (all of us?) have seen too many examples of corners cut to get running. Have a meeting with your managers. Do a cost-benefit analysis of unplanned downtime due to catastrophic failure.

What you don't pay for up front costs a hundred (or several thousand) times more down the line.

-A-

The following is based on hydrostatic testing, a material Specified Minimum Yield Strength of 60,000psi, an OD of 16in., & a wall thickness of 0.312in.

The Design Pressure produces a stress of 10,897psi, or 18% of SMYS.

If the pipe is tested at 1.5 x Design Pressure, the test pressure would be 638psig, and the stress level would be 27%.

If the pipe were tested at 2 x Design Pressure, the test pressure would be 850psig, and the stress would be 36% of SMYS.

We once bot & installed pipe only to find out the longitudinal weld was suspect, so we tested it at 100% of SMYS. In your case, this test pressure would be 2340psig, or 5.5 times Design Pressure.