Pin Hole leak in HFW pipe during field test

[nickypaliwal]

There is a 18" 10.3mm thick API 5L X65 HFW pipeline which passed all the tests including hydro test in mill however there is a pin hole leak observed near the seam weld which was observed in field hydrotest.

It was difficult to find the location of the leak as the leak is small and was not even detected during full UT.

Many other areas are showing pressure drop however the location is still untraced.

As I read on internet, it looks probably cold weld/penetrator defect.

Is there a way out to confirm if the failure is penetrator defect and what should be the corrective action?

 

[SJones]

J H McHaney et al, Advance Planning Advised To Manage Failure Investigations, Pipe Line & Gas Industry, January 1999, pp 83 - 90

Steve Jones
Corrosion Management Consultant

http://www.linkedin.com/pub/8/83b/b04

All answers are personal opinions only and are in no way connected with any employer.

 

[brimmer]

Cut out the section, send it to a metallurgical lab... they should be able to tell you the defect / cause of failure.

Corrective action is a little tricky, as this you would expect to be a one off defect, or one off severe defect enough to fail. There is no way to reliably detect this type of defect in the line by inspection methods (ILI, etc.). You could send extra joints to the lab when you do failure analysis, i.e. cut the entire joint out at least, and see if there are more defects. As you mention, it is a very small leak (Defect has no length), so additional hydrotesting may not burst any of these features. Hopefully your regulator is not PHMSA, and you can get a reasonable plan, or the regulator has enough technical knowledge that he doesn't make you spend a bunch of money for nothing (I have yet to encounter a technically competent regulator in the US). I would go about trying to prove this was a one off defect, hopefully you don't find anymore of these. I would also challenge the pipe manufacturer, in terms of weld quality, QA/QC and why this was missed. I have seen this before, and their 2 second hydrotest at the mill simple misses these, and because the defect has no length, UT misses these as well.

 

[SJones]

Besides cold welds and penetrators there are a number of other failure initiators:

https://primis.phmsa.dot.gov/matrix/FilGet.rdm?fil=7684

This compendium might be a useful resource to bookmark:

https://primis.phmsa.dot.gov/matrix/PrjHome.rdm?prj=390

There's no regulator to worry about, it's in the Wild Wild Middle East, only an irate end user who is now staring down the barrel of a full pipeline replacement if they can't get a handle on pipeline integrity.

Steve Jones
Corrosion Management Consultant

http://www.linkedin.com/pub/8/83b/b04

All answers are personal opinions only and are in no way connected with any employer.

 

[nickypaliwal]

@ Steve Thanks for the links. First link one of the same link that I reviewed. Based on the available information, I feel the penetrator defect may be the most probable reason. The second link above is new and I will try to investigate more for the relevant information.

@brimmer: We plan to recommend to send the samples to lab however I was looking for any specific test that can lead us to conclusion. The section of pipe was sent for lab analysis but the report looked fine in all aspects. Only thing which was little suspicious was the impact test values which varied a lot in one of the test. However we ignored the same since the pipe was not commissioned and ailed only during field hydrotest.

Regarding defect identification during manufacturing, The only method that I found in 2-3 papers to detect such defects is high frequency photography which probably would not be a full-proof method.

Although I do not have welding background but the weld parameters used by vendor looks to be in acceptable range.

Please add anything which may help us to proceed in right direction.

 

[LittleInch]

nicky,

I can't offer too much technical knowledge, but I can advise that before you go any further and send your pipe off, you search around and find a world expert in seam welding as this could get very nasty very quickly. You don't want to go and do some tests which then destroy the evidence.

Worst case is that the entire pipe batch is essentially "contaminated" and needs to be dug up and replaced. Your comment "Many other areas are showing pressure drop however the location is still untraced." doesn't sound good - any update?

It might just be "one of those things", but that won't stop the sh1t flying around when the various costs and delays are added up.

Record everything in writing, keep your own copies of e-mails etc and get prepared for battle...


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

 

[nickypaliwal]

There is not much to add up till now but there are 4-5 locations/leaks expected in a pipe more than 250KM long. At present, it is not suspected to be spread all over however more details are yet to come. I will keep updating with the information.

 

[brimmer]

I might suggest you find a lab that does proper failure analysis? The way to see what the defect is would be to freeze it, and break it open at the defect site, you should get a nice cross-section and determine what it is. This was not done???

 

[SJones]

Another problem that might be faced as a result of all the chopping, rewelding and more hydrotesting is that of "pressure reversal." Something to think about. Thermatool used to have a nice set of documents about HF welding quality and defects, but probably can only be found via Google these days:

https://www.google.co.uk/search?q=T...X&ved=0ahUKEwjhq9Xl3dfSAhUFcBoKHcA2A8wQsAQINg

The principal prevention methodology is: selection of coil supplier, and correct establishment of welding and heat treatment parameters with reliable monitoring and alarming of the same. What parameter monitoring data can the manufacturer produce?



Steve Jones
Corrosion Management Consultant

http://www.linkedin.com/pub/8/83b/b04

All answers are personal opinions only and are in no way connected with any employer.

 

[SJones]

I missed out edge preparation as a prevention mthodology. Apologies for that.

Steve Jones
Corrosion Management Consultant

http://www.linkedin.com/pub/8/83b/b04

All answers are personal opinions only and are in no way connected with any employer.

 

[nickypaliwal]

@brimmer We recommended the same and still it needs to be done.

@Steve We also considered edge defect as a probable cause however the defect is at a limited locations and we can not do anything to confirm it now. Now we are just looking forward to track the defect locations and to decide whether entire pipeline can be used with little repair on defected areas.

Also like to add that the UT result (U3-N10) was used by the manufacturer which did not trigger the alarm however shows a little indication (small peak) in the leak location. However by looking at UT report, it seems likely possible to catch it by more stringent acceptance criteria(U2-N5). This may be considered a learning.

Failure analysis is still required to be carried out.

 

[EdStainless]

If the root cause is edge prep or alignment then I would worry more.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[weldstan]

s has been stated the root cause can be many things, including minor electrical interruptions/variances during welding. In at least one instance, flattening tests at the seam led to determination that edge prep was the root cause. Small penetrators may be more discernable using magnetic flux leakage equipment rather than UT.

 

[nickypaliwal]

Flattening test was done and it passed. Failure Analysis is still under progress.

Nickypaliwal

Materials & Corrosion Engineer