Good Morning Everybody,

We have following situation during fabrication of a pressure vessel :

There are 3 Long seams in a fabricated Shell section.
Welding of all 3 seams are completed and all NDT are balance for these seams.
There are some stringent dimensional requirements as per client's specification. Due to high thickness, significant distortion (peak-in/out) had occurred. Now production people is planning to correct it. For that, they have removed some 50% of one side of weld in one particular joint by grinding. And planned to re-weld it to negate the distortion effect.

Shell material is SA387 Gr 11 Cl 2 with SA 240 Type 410S cladding having thickness of 44.5+4.5 mm.
Pre-heating, Interpass Temp., DHT & ISR/PWHT are required as per WPS.

Now my question is
(1) Is it allowed as per general engineering practice of pressure vessel fabrication ?
(2) Can I call it a repair and can use WPS of weld-repair ?
(3) Is it a case of NCR ?
(4) Is it acceptable as per Code ?

Please suggest your views on this.

Regards,

NN

Now my question is
(1) Is it allowed as per general engineering practice of pressure vessel fabrication ?
(2) Can I call it a repair and can use WPS of weld-repair ?
(3) Is it a case of NCR ?
(4) Is it acceptable as per Code ?

nikesh_in

First, I will document the problem by NCR. Then disposition the repair by welding using the same WPS used during the fabrication of the vessel. This is an inprocess repair and acceptable per Code, besides you have not Code stamped the vessel.

leem

Thank you Leem for reply.

Impact Testing is required in this job.

Is it not required to use WPS for weld-repair in this case, as it includes more inputs through re-welding ?


Regards,

NN

NN

This is inprocess repair and you can use the same WPS during the vessel fabrication.

leem

Any other comments / views from other members or specially, AI on the same matter ?

Thanking you all.

Regards,

NN

I basically agree to Leem,

1) it is a repair
2) according to QCM, an NCR is required, incl. AI acceptance to the scope and method.
3) I suggest the same WPS
4) NDE shall be repeated
5) Watch the aggregate time at temperature when re-PWHT, the tensile and impacts may be voided (UCS-85) and the PQR might not cover the PWHT.
6) In my opinion re-welding does not make the vessel better. re-rolling, or controlled heating may be an alternative.

I hope it works.
DK

Thank you Dr. Koelbl for your reply.

Please see my comments as below in blue font:

1) it is a repair
- Manufacturer is not agreeing to this. For him, it is in-process correction and it is not repair. I believe it must be called repair.

2) according to QCM, an NCR is required, incl. AI acceptance to the scope and method.
- Job is still with production people and not yet offered to QA/QC. Can it be still case of NCR ? Vessel is non-coded vessel.

3) I suggest the same WPS
- As per client specification, separate WPS is required for repair. PQR for this repair-WPS not yet qualified. My opinion is that this WPS for repair should be used in this type of re-welding after qualification.

4) NDE shall be repeated
- No NDE has been done yet. 100% TOFD UT and MT to be done before PWHT & after PWHT. NDE & PWHT will be done after re-welding only.

5) Watch the aggregate time at temperature when re-PWHT, the tensile and impacts may be voided (UCS-85) and the PQR might not cover the PWHT.
- Main welding WPS has been qualified with PQR having Minimum SPWHT (2 Hrs.) & Maximum SPWHT (10 Hrs.). Same will be done with WPS & PQR for repair too.

6) In my opinion re-welding does not make the vessel better. re-rolling, or controlled heating may be an alternative.
- I agree with you. But manufacturer is not agreeing on this.

As I have given some more details to your views, what would be your suggestions on them.

What is your opinion on this ?

Regards,
NN

NN,

As there is no definition of REPAIR in section VIII, maybe NQA-1 definition helps :


But in my experience the difference is:
When the weld was made according to a WPS, and the welder or welding supervisor decides to grind and re-weld part of the joint, it is an in-process repair, which is not considered a REPAIR in the sense of the definition, so let us call it CORRECTION.

When the weld is declared finished by fabrication, and later during dimensional inspection, or NDE it is found to be "nonconforming", it is clearly a REPAIR. That is a NONCONFORMITY, consequently the restoring action is a REPAIR.
It could be called "rework" if not special grinding activities were performed before weldingand if not the welding variables are different (at least weld prep). AND IT IS NOT IN PROCESS, becuase the process welding was finished.

You say it is NON CODE, so who is the Inspector, and what does he/she say?
And what says the Manufacturers QC manual, or is it non-code and non-QC System?
The problem point seems to be the client spec. A REPAIR WPS is required, but why not support the WPS with the PQR from the production WPS? So the customer gets his repair WPS and the extra work is filling one form and sign and date it.

DK

Thank you Dr. Koelbl for your reply.

You are right.
Procedure demonstration (PQR) is best suited for this repair procedure qualification.

Thank you for all your comments.

Regards,

NN