Non-Code Vessel Repair

[UW1981]

We have a situation where it seems to make sense to repair non-code vessels (like cracked compressor bottles). In some cases, it may be a repair we make in order to keep our processes in-service until we can get a code vessel manufactured. In other cases, the repair may be permanent....although we are still work on the criteria, the decision would be based on service, pressure, and location (ie, the higher risk, the more likely we would end up replacing the vessel). The question is - what vessel information do we absolutely have to have in order to make the repair in accordance to API 510 - NBIC-NB23 - and ultimately ASME Sec VIII (which is where these other two documents tend to send you). We can perform PMI and hardness testing, but do we have to go to this level or can we simply assume the worst metallugy and go from there. If we conduct PMI and hardness testing, is this enough information to meet the requirement of API and NBIC?

 

[unclesyd]

Comeback with a few more details on the type of repair that you are contemplating along with the ratings and environment that the equipment sees.

 

[UW1981]

Well....it can all vary.

Typically, we are repairing cracks on the nozzle-to-shell weld...though it could be at the nozzle-to-flange weld, or a shell weld. But pin-holes in these same areas are not unheard of.

It is most often compression service (so we may have some vibration), but some of these repairs may be in large inlet receivers (ie, built many moons ago and not code stamped).

The rating could range from 25 psig to 1,000 psig. Sweet or NACE service (up to 5% H2S), and could have as much as 2% CO2. Figure all of this is water saturated.

Temperatures could be as high as 300 F (compressor discharge bottles), but nothing usually below -20 F (cold winter day).

The size of these vessels could range from 2'X4' to 12'X40'.

I can try to be more vague....

 

[metengr]

The question is - what vessel information do we absolutely have to have in order to make the repair in accordance to API 510 - NBIC-NB23 - and ultimately ASME Sec VIII (which is where these other two documents tend to send you).

Reply;
The in-service inspection and repair codes API and the NBIC (NB 23) were developed to assure that repairs/alterations to pressure retaining items are performed in a manner that follows the code of construction (COC) requirements to assure continued safe operation. This means that materials of construction (MOC)need to be known or identified for pressure retaining items to assure proper weld repairs are performed using traceable (COC) materials and approved repair methods in accordance with a documented Quality program.

If you intend to use the NBIC or API for inspection and weld repairs to non-code pressure retaining items, you need several basic elements to your program;


-Identify all materials of construction, original dimensions and locate all shell girth, seam and appurtance welds for the non-code vessels.
- Perform periodic inspections of these non-code vessels to evaluate changes in wall thickness over time and to inspect shell seam girth and appurtance welds using nondestructive testing.

If during the course of inspection you find some type of in-service damage that requires weld repair, the weld repairs can be performed using the NBIC or API for guidance.

 

[UW1981]

MetEngr - thanks for the reply...

Here's the thing...we can perform PMI to get a pretty good idea of the material we are repairing (head, shell, nozzle, etc). And we might even be able to use PMI to get an idea of the original weld material...but that, I believe, is a little dicy. The specific original weld procedure is lost since this is a non-code vessel.

So....is PMI adequate (without knowing the original weld metal/procedures) to make a 510 repair.

Also, is it acceptable - in lieu of PMI - to make a conservative assumption as to the material in question and based the repair procedure on that assumption?

We can gather all the dimensional data and weld locations, and these vessels are inlcuded in our Integrity Management plan.

 

[metengr]

UW1981;
Since these are non-code vessels with no reference information, we need to be assured that the original vessel material is of a chemical composition and heat treatment that it can be satisfactorily welded using welding procedures qualified to ASME Section IX. So, performing PMI and obtaining surface hardness of the vessel steel(s) in question would be beneficial in my mind. This information would also allow you to select the proper material needed for flush patch plate weld repairs or whole-sale replacement of shell course sections. I would not assume worst case for any repairs related to these vessels.