Guidance on Piping Inspection after Water Hammer Damage

[ric_bm]

Introduction:
We recently built a compression plant. Each compressor has a suction scrubber. These scrubbers have process drain lines connected to a 6" SCH40 A-106 Gr-B header which runs 2 km. to a tie-in with existing plant.
Design Code: ASME B31.3

Problem:
Due to process issues, this header suffered "water hammer" damage in different sections two months after it started working. At some sections we have some "dents" due to impact with other pipes and, at some other sections, we have "gouges" due to impact with supports.
For the worst case scenario, see a photo and a sketch:

https://files.engineering.com/getfi...aca-e9ac8eff8079&file=Worst_Damage_Sketch.jpg

https://files.engineering.com/getfi...7-b778-89da31b568a2&file=Damage_1_Picture.JPG

Solution Approach:
We have checked each spot with: visual inspection, penetrant liquid testing, PA ultrasound testing and thickness measurement. However, neither our client nor us have defined which acceptance criteria needs to be applied.
At this point, we are trying to show that the remaining thickness in the worst case scenario (6,54 mm remaining thickness / 7, 11 mm nominal thickness) is acceptable per API RP 574 which states that the Barlow formula can be used.

Guidance Required:
I'm not looking for a definite answer. Just trying to understand:
a) what steps need to be taken in order to inspect the components, asses the damages and (hopefully not) repair the components.
b) which standards need to be followed for assesment? So far, I see two options:
1-API RP 574 for a simplified assesment.
2-API 570 with help from API 579 FFS - part 12 for a quantitative assesment.
c) which standars need to be followed for repair? So far:
API 570 section 8, which relies on ASME PCC-2 guidelines.

Thanks, Ricardo.

 

[gerhardl]

Such cases are always difficult to assess.

If this were about technical safety devices, all rules would have required all dented and damaged parts replaced with new, fully tested components. Reason: unknown risk and weaknesses not detected by tests and possibly shortening the components normal lifetime.

The question here should not be which tests and procedures to use, but who has the responsibility and cost for material and personell damage, should anything happen.

If this was an airplane you would probably be grounded until replaced.

On the other hand: I have seen factories allow repeated waterhammer and even repeated dust explosions in ventilation ducts because 'usually it is a limited damage'.

Your choice!

 

[MJCronin]

Every once in a while there is an opportunity for the bosses to actually earn their money .....

This is a situation where Management has to stand up, put on their big-boy pants and make a decision...

I recommend a search for an objective outside consultant who has experience with this problem....

This will be probably be expensive, but it is also probably fair to say that your existing staff has no experience in this area.

I also doubt that your Management will do this ...




MJCronin
Sr. Process Engineer

 

[LittleInch]

Why is this in "valve engineering forum"? Pipeline and piping would have been better.

You need also to look at hardness of the dents in particular.

Excessive hardness leads to fatigue cracks and failures.

The scrapes you can grind out and lose the stress concentration sharp edges, but the dents I would cut out or re-inforce with external sleeving (epoxy or steel)

B31.3 has some pretty healthy margins of safety for pipe wall thickness, but as ever the issue is risk acceptance and consequences of failure. These are not engineering decisions.

But whatever you do make sure you test it as a new piece of pipe with a full hydro.

If the damage is less than 10% of the line I would recommend replacing those parts, but if its spread out and would require complete replacement then I can see how the pressure builds for some sort of minimal repair scheme. The problem is the long term damage done to the pipe - maybe it will last decades, maybe it won't - especially if it under a fluctuating stress / pressure scenario.

Good luck and tell us what you decide to do!

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

 

[ric_bm]

Thank you guys for your feedback.