Pressure test requirements for B31.8 in-service piping

[pjdpipingeng]

I have a question concerning the pressure test requirements, particularly with the required test pressure, for a 3" Sch 40 gas pipe. Some history: As already mentioned, the pipe is a 3" Sch 40 API 5L Gr B seamless pipe transporting methane gas. The line has been in service for about 3 years. Max. operating conditions are 39 bar/45C, design conditions are 48.6 bar/65C. The line was hydrostatically tested before commissioning at a pressure of 54.6 bar. Part of the line runs outside of industry boundary, which would put it in the ASME B31.8 scope, and the other part runs inside industry boundary, which would put it in the ASME B31.3 scope.

My questions are as follows: For determining structural integrity of the pipeline, it is being proposed to perform a hydrotest on the piping system. The company that will perform the test wants to test at the original hydrotest pressure of 54.6 bar. However, when I look in B31.8, it states that an in-service piping system must be tested at a pressure that will cause a hoop stress of at least 90% of specified minimum yield strength. When I did a calculation on what that pressure is, I get a pressure of about 268 bar. When I look in API 570, it states that hydrotest should be performed in accordance with ASME B31.3, which is 1.5 x design pressure, which turns out to be a lot less than 268 bar.

Is my interpretation correct? Should an in-service piping system per B31.8 be hydrotested at 268 bar? Wouldn't it be more reasonable to repeat the hydrotest at the original pressure of 54.6 bar? Its a little confusing to me.

It is my first time encountering this particular situation, so an expert's advice on this would be greatly appreciated.

Thanks!

 

[LittleInch]

I think your question fell into the weekend gap and might have had better traction in the pipelines forum.

however to try and address your points.

your figures don't add up / make sense.

It's not common to divide a pipeline up just by location in or outside a fence. B 31.3 is wholly unsuited for buried pipelines and unless you can somehow separate the two sections, just pick one design code ( 31.8) and use it from end to end.

B 31.8 varies the test pressure by location class and similarly sets the MAOP as a fraction of the test pressure.

The test pressure you quote (54.6) is exactly 1.4 x the MOP. This doesn't match any of the pressure test factors in table 841.3.2-1.

Equally the test pressure is too close to the design pressure to be of any use.

I don't understand why you're pressure testing a 3 year old pipeline??

However you are, so whilst your first point is correct - I guess you're looking at 851.12.1, I think item c applies in your case i.e. "..or those pipelines that operate at hoop stress
levels lower than maximum design pressure, the minimum strength test pressure shall be 1.10 times the MAOP.

If I was you I would go for a design pressure of 48.6 and test to at least 53.46 barg at the highest point. Then you don't limit the MOP in the future.



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[DekDee]

Just to add to LI's comments.
Many times I have seen an API 1104 pipeline come out of the ground and connect directly to a B31.3 piping system that then takes the product into the refinery.
An isolation valve separates the two lines - you would never consider hydrotesting the two together.
The same applies to B31.8 connecting to a B31.3 line - totally different design criteria so they should be tested separately,
Regards,
DD

 

[LittleInch]

Api 1104 is a welding code, used by b31.4 & 31.8. Splitting a pipeline into a 31.3 line before a pig trap is bad design, but otherwise yes, you often see a code break at some suitable point.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[pjdpipingeng]

Thank you LittleInch and DekDee for your responses.

LittleInch - While you are correct that you wouldn't normally divide a line up into 2 different codes, the way that the pipelines are divided up here in Saudi is basically what DekDee stated - All aboveground and underground pipelines outside of industry boundaries is covered by either B31.4 (liquids/slurries) or B31.8, and everything inside industry boundary is covered by B31.3. The pipeline that is going to/coming from industry always goes underground before entering/exiting industry boundary, and there is always an isolation flange kit at each end of the buried piping as it comes out of the ground, and usually an isolation valve on the industry side.

Having said all of that, I agree with LittleInch's question of why a 3-year old pipeline is getting hydrotested. Answer is that the owner of the pipeline is re-purposing the line for something else and wants to make sure that it is sound.

Concerning the pressure test at 54.6 bar is from an older version of B31.8 (2002, I think) which did stipulate a test pressure of 1.4 x MOP. I brought this up to the testing contractor, and mentioned that minimum version of B31.8 to use would be the 2013 edition, since that is the approximate year that the line was put into service.

Concerning 851.12.1c), I would tend to agree with LitteInch's comment that part c) might be appropriate to use in this situation, but the confusing part is do I compare the hoop stress at operating conditions to the maximum design pressure itself, or to the hoop stress at maximum design pressure? I'm thinking that it would be compared directly with the maximum design pressure itself, in which case the hoop stress at operating conditions (315 bar) is greater than the design pressure (54.6 bar), and thus part c) would not be applicable.

The default situation could be to use API 570, which states that the hydrotest pressure will be per B31.3, or 1.5 x Design pressure.

 

[LittleInch]

pjdpipiing
I was doing ok with your response right up until you got the point where you said the new Operating pressure was to be 315 bar (!!).

Pretty sure you're at or above yield at that point.

Care to add any more info?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[pjdpipingeng]

LittleInch,

Sorry for the confusion. The 315 bar (4569 psi) is the calculated hoop stress (Sh = PD / 2t) at an operating pressure of 39 bar (P = 566 psi). Maybe I'm not doing a correct comparison, which is where the confusion that I talked about in my last post is coming from. From what I understand paragraph 851.12.1 c) is saying, if the hoop stress (4569 psi) at operating conditions (566 psi) is less than design pressure (792 psi), then the test pressure would be 1.1 x MAOP. But if I am right, since 4569 psi is greater than 792 psi, then c) would not be applicable.

Like I said, maybe I'm not interpreting the code correctly and doing an incorrect comparison. That's where the confusion lies.

 

[DekDee]

LI,
Sorry for not responding sooner.
Just like to explain my reference to API 1104.
I am not an engineer and in 30 years of working with pipelines in the Southern Hemisphere I have never heard of B31.4 or B31.8.
From Welder to Radiographer to Welding Inspector to Quality Co-ordinator to Quality Manager - the only thing I ever heard was API 1104.
No idea what the design code was but presume it must have been one of the ASME codes you mentioned.
Thank you for the enlightenment,
Cheers,
DD

 

[LittleInch]

You may well be correct.

The wording is a little strange but it is clear from the entire clause that to establish the integrity by pressure testing they want to use the 90% SMYS rule unless, in section c, the stress in the pipe at operating pressures is so much less than "maximum design pressure".

Now "maximum design pressure" is a little odd and could mean the pressure at which the hoop stress is 72% of SMYS??

Also the main section 851.12 states that "Comparison
of new test pressures with previous test
pressures will demonstrate that the integrity of the pipeline
has not been reduced if new test pressures are equal
to or greater than previous test pressures."

Seems to imply that if you use the same test pressure it will all be OK.

A bit confusing.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.