Weld Repair ASME B31.3 1

[LB_FLO]

Hello experts,

I am looking for your thoughts on the amount of times a weld can be repaired. From reading through ASME B31.3 I cannot see any guidance on a specific number. From searching online and through forums the usual answer seems to be initial weld plus 2 repairs is good practice.

We have a job that has had multiple repairs including some that have been repaired twice. Our customer has questioned this but to me we have followed the code and can refer to ASME B31.3 para 341.3.4 (f)
If any of the defective items are repaired or replaced, re-examined, and a defect is again detected in the repaired or replaced item, continued progressive sampling in accordance with (a), (c), and (e) is not required based on the defects found in the repair. The defective item(s) shall be repaired or replaced and re-examined until acceptance as specified in para. 341.3.3. Spot or random examination (whichever is applicable) is then performed on the remaining unexamined joints.

We have carried out 100% x-ray on all welds which was agreed with the customer at the beggining of the job.
The pipe is 1.5"NB SCH40 304/L.

Thanks for your thoughts.

 

[DekDee]

Progressive sampling has nothing to do with amount of times a repair may be made.
It is usually noted in project specifications and if it is not (or not noted in any other contract documents) then the customer gets what they receive.

 

[weldstan]

On such small diameter pipe, why did you not fully remove the weld and replace? I certainly would have required such.

 

[DVWE]

Did your customer require 100% RT in their specifications, or was it something mutually agreed upon by both parties and not in any contract of any type?

If your customer specified 100% RT, then they should have also specified the amount of repairs allowed. If they didn’t, then you have nothing to worry about, and as per DekDee above, progressive sampling has nothing to do with this scenario.

Were these welds butt welds, or socket welds? I find it odd to require 100% RT on small bore stainless steel……

The devil is in the details; she also wears prada.

 

[r6155]

The welder is disqualified and requires training, a new welder is required.
The company may be disqualified.

Regards

 

[LB_FLO]

@DekDee
Thanks for your reply.
Could you point me in the right direction if the progressive sampling section is not relevant?

@weldstaan
Each nozzle (18) had up to 12 joints so to start again would of been very time consuming.

@DVWE
The 100% RT was requested by the customer. It was accepted as it was not thought there would be such problems which this small pipe and material.
Nothing in their specification to indicate amount of repairs allowed.
It was butt welds, staight pipe and elbows.

 

[david339933]

The only place you may find the number of repairs allowed is in the customer specs. You can't progressive sample...you are already shooting 100%. What additional welds would you shoot?

 

[DekDee]

First thing you should do is find alternative employment for your welder/s - or serious re-training / re-qualification at least.
Second thing you should do is explain politely to your customer that there is no code mandate for maximum number of repairs - it is something that should be stipulated by the customer via project specifications or contract documents.
Do an internet seach on multiple repairs of s/s welds - I think you will find you may be OK.
It is usually specific exotic materials that have a maximum of 0, 1, or 2 repairs allowed.
Hope that helps,
Cheers,
Shane

 

[r6155]

Once on the first day of inspection I found serious defects in the welds that show a wrong company. It was for stainless steel pipes.
I stopped work. My company looked for another manufacturer and the work was much better.
There are many things that do not allow going around so many times and delaying the solution.

Regards

 

[Dhurjati Sen]

Replies by weldstan, r6155 and Dekdee are very apt.
If a joint is repaired quite a few times, the metal degeneration that takes place is enormous (that too by below average welders and maybe poor fabricators).
You may carry out an in situ metallography of the original base metal adjacent to the HAZ. This could give you a basis for complete rejection and probable blacklisting of the contractor.

DHURJATI SEN
Kolkata, India

 

[DekDee]

LB_FLO
Looking again at your original post it raises more questions in my mind.
You have no project specifications that relate to NDT so I can presume it is not critical piping.
Therefore, one would presume B31.3 NFS which is 5% NDT - you have been performing 100% NDT.
The more you inspect the more chances you have of finding a defect.
You did not mention how many welders got repairs - that is the most important question.
Progressive sampling as per B31.3 is exactly that - progressive.
Testing each welder and monitoring the results - you should never arrive in the situation you are now. (especially if you are testing 100% - pretty easy to see repair status of all welders)
Welder gets close to a company or project maximum repair rate they are closely monitored and if necessary removed for re-training or re-qualification.
One important thing to remember too - with root defects (LOP or LOF), if the welder was not competent enough to get a compliant root run on first attempt what makes you think they are competent enough to perform the repair ?
Welds do not have to be repaired by the original welder - put your best welder on root repairs.
Some food for thought,
Cheers,
Shane