Crack of reo pad fillet weld
Crack of reo pad fillet weld
(OP)
Hi all,
The Duplex pressure vessel on this FPSO has developed a small crack (3/4" long) on the reo pad's fillet weld (12" NPS inlet nozzle), since last inspected 12 months ago. Salt ladden environment and warm H/C condensate + salty water inside, 400 psig pressure.
Is there any maximum size for this crack before you shut down the plant or the plant should be shut down immediately, the crack investigated (API 579 Level 3)and repaired.
Please advise.
gr2vessels
The Duplex pressure vessel on this FPSO has developed a small crack (3/4" long) on the reo pad's fillet weld (12" NPS inlet nozzle), since last inspected 12 months ago. Salt ladden environment and warm H/C condensate + salty water inside, 400 psig pressure.
Is there any maximum size for this crack before you shut down the plant or the plant should be shut down immediately, the crack investigated (API 579 Level 3)and repaired.
Please advise.
gr2vessels





RE: Crack of reo pad fillet weld
Could you state the precise location of this crack, e.g. middle or toe of fillet weld, vessel or pad side? Its a very different matter between a crack going into the shell of the vessel or the reinforcing pad.
If it looks as if the crack is going straight down into the shell from the toe of the fillet weld, then I suggest you obtain the services of a skilled manual UT guy to measure crack length and estimate crack depth because this is the one location where UT could be helpful. What is its operating temperature?
If its anywhere else then you might have a chance with eddy current of measuring its depth. I am no eddy current expert but my feeling is that Duplex parent material and filler will give problems to the E.T. technician.
Also you'll need to dig out the fillet weld procedure, check the vessel operating parameters and make a vibration analysis. On-pass all information to the operating company's stress engineer.
Good luck
Nigel Armstrong
Karachaganak Petroleum
Kazakhstan
RE: Crack of reo pad fillet weld
Was there an actual leak at this location? If not, it might have been a root crack that eventually propagated to the outer surface of the fillet weld on the repad. Hard to say at this point w/o more information. Have you checked other nozzle attachment welds?
RE: Crack of reo pad fillet weld
RE: Crack of reo pad fillet weld
I'm back from the site. The crack is on the fillet weld of the pad outer edge and the shell. It is contained in the plan at 45 deg, that is the middle of the fillet weld and seems now fully penetrated to the root of the weld. We still don't know if is propagating in width or not.
The plant was shut-down, just in case, but with some resistance from operations engineers. The crack is on the path of possible nozzle failure and the reinforcement was required because of high external loads. Could the nozzle fail with a bang?
Cheers,
gr2vessels
RE: Crack of reo pad fillet weld
No doubt there would have been resistance to shutting down, especially if no recovery plan was established. Is there an AI or inspection engineer on-board? Dont you have to present your plan to them? Repair will be carried out under which Code?
Without taking all that into account and the specific Code requirements, I would think your steps are to progressively grind the fillet weld restricting weld metal removal to about 2 - 3 mm at a time, MPI test after each skim to establish the continued presence and length of crack. This is an attempt to have the crack totally removed hopefully before it reaches the root and the adjacent shell. Surely repair methodology for defects that do not penetrate the pressure-retaining envelope, such as a crack confined to the weld metal of a reinforcing pad-to-shell fillet weld, will be significantly different from that for cracks which do penetrate into the shell.
Of course the other aspect is - why did the crack initiate and grow>
Nigel Armstrong
Karachaganak Petroleum
Kazakhstan
RE: Crack of reo pad fillet weld
You ask if the nozzle could fail "with a bang". A brittle fracture is possible in the duplex grades, but the materials are known to have fairly high Charpy impact toughness at low temperatures. That said, the conditions for a brittle fracture involve toughness, stress level and crack length. It's not a quickie answer. As previously mentioned, API-579 has procedures for such analysis.
ndeguy,
I'm not an NDE guy, but I was wondering how effective the magnetic particle technique would be in the case of duplex stainless steel. Most duplex that I've seen for vessels and tanks has been austenitic-ferritic. Does MT work for those grades? My guess was no, but I'm not sure.
Cheers,
Joe
Joe Tank
RE: Crack of reo pad fillet weld
I have seen things like this start to run as they are excavated.
Suggest LP, drill the ends, explore, maybe use some heat to get things to open up. May want to cut a window from the pad to check the vessel wall.
Don't know what you are dealing with as far as material. Could just be a bad weld.
RE: Crack of reo pad fillet weld
You are correct, some duplex grades are magnetisable whereas others are not. As it is a welded construction it may be best to replace the MT with PT even if the plate material is magnetisable, unless it can be demonstrated that the weld metal is also magnetisable by a Hall Effect meter - and not many FPSO's have those to hand! In the case of a crack the PT should show it up equally as well as the MT.
Thanks for the correction of my elementary error Joe - you have the makings of a damn fine NDT hand! My only excuse is that my single-tasking brain was concentrating on "How would I get it done" rather than remembering the material.
Nigel Armstrong
Karachaganak Petroleum
Kazakhstan
RE: Crack of reo pad fillet weld
rhg
RE: Crack of reo pad fillet weld
Very instructive replies;-hence conclusion for this event
- The nozzle is unlikely to enter a catastrofic, sudden failure.
- The nozzle pad fillet weld appears to have developed the crack due to a welding defect, accelerated by the stresses generated by external nozzle loads (perhaps not the best pipe stress analysis at the design stage)
- The repair has now been completed per code, in the presence of AI, lots of NDE specialists, fabricator representative, also the material saved from the crack excavation has been sent for analysis, particularly to detect the ferrite level.
- All been done by specialists, but perhaps a bit overconservative, having the whole plant shut-down for the investigation and repair.
- An extensive plan is being developed for the entire vessels lot to be closely scrutinised and the rutine inspections be intensified. The lesson learned from the forum is not to panic, but investigate and repair this defect, at the earliest convenient time to isolate the vessel, rather than shut-down the plant and causing obvious distress.
Thanks again,
gr2vessels