ASME B31.3-2014 Depth of Incomplete Penetration

[Nathanael Vleeskens]

This has been bothering me for some time, I have noticed there is no longer a maximum depth of incomplete penetration on girth welds per Table 341.3.2 Acceptance Criteria for Welds. I remember in the past it was 1mm maximum depth, now it only refers to the cumulative length.

Hypothetically, what if you have a 6mm (1/4") thick weld and have 3mm deep incomplete penetration? Apart from having a bad welder, would this be acceptable to the code assuming it's cumulative length is acceptable... or am I missing something?

 

[SectionIX]

Since the Code lists a length criteria only, the Service Condition would dictate whether any or that shown in the Table would be acceptable.

While theoretically, IP having a depth of 3mm in a 6mm could be acceptable (based upon Service and length), it is unlikely that an owner/operator would accept this.

ASME B31.3 Clause 341.3.1 gives you the right of rejection for this type of circumstance: The type and extent of any additional examination required by the engineering design, and the acceptance criteria to be applied, shall be specified".

Although API 1104 does not apply, it does provide clarification for this type of imperfection:

9.2 RIGHTS OF REJECTION
All nondestructive test methods are limited in the information that can be derived from the indications they produce.
The company may therefore reject any weld that appears to meet these acceptance standards if, in its opinion, the depth of
an imperfection may be detrimental to the weld.

 

[XL83NL]

As I interpret it: assuming the 3mm depth is over a length not more than 25% of total weld length, yes. Not sure why this section was ruled out.
Ive tried finding info on it by reviewing posts like these, but with no success. Maybe some Code committee member/guru knows?

 

[DekDee]

XL83NL,
The proposal to revise this clause was accepted in 2013.
The rationale was even when you have access to visually inspect the internal weld it is very rare that it is possible to actually measure the depth of this discontinuity.
It is not possible to determine depth by MT, PT or RT and depth is not considered in 344.6.2 for UT.

If the conditions (root gap) were similar to the drawing in Fig 341.3.2 (c)then manual measuring would be possible but it was felt that the most common form of IP is tightly butted unfused root faces and that is why Note 7 was deleted.
Note 7 stated that that condition (tightly butted unfused root faces) was "not acceptable" but that actually contradicted the allowance given for acceptable length of IP.

Hope that helps,
Regards,
DD

 

[XL83NL]

At first glance, sounds like a strange argument to me. So because you cant visually measure it (at least, not in small diameter piping), nor determine it using any other standard type of NDE, youre allowed to have any depth of IP.

 

[DekDee]

If you can't measure it how can you regulate it ?

 

[XL83NL]

When looking at fig 341.3.2(d) for a possible occurence of IP, I can easily measure this, when the pipe ID is large enough. Sure, when the pipe is large enough (say 8"), and the weld is midway a lengthy and finished spool, you wont reach it; but that also counts for other means of visual examination of welds that are difficult to reach. For such cases there's in-process examination.
However, even if you cant measure it, doesnt mean (imo) it should be allowed. The trick then is to find some sort of criteria that allows one to determine if the defect is acceptable when you cant give it a hard number in terms of deviation.

 

[DekDee]

XL83NL,
My mistake - I meant (d) and not (c).
Yes, you are correct and that is what I was trying to say.
It was the committees opinion that the majority of IP cases were found to be tightly butted unfused root faces and they could not be measured.

For your last point, root concavity acceptance can (and regularly is) calculated by density on the radiograph.
Pretty sure a 3 mm deep IP on a 6 mm WT weld would be that dark there would be no option but to raise it with the client / owner for acceptance/rejection,
Cheers,
DD

 

[XL83NL]

Pretty sure a 3 mm deep IP on a 6 mm WT weld would be that dark there would be no option but to raise it with the client / owner for acceptance/rejection

Agreed. Any NDE firm, provided they understand the acceptance criteria for piping (which in my experience theyre often unaware of), should not accept such 'defects'even if the code allows it. However, this now puts the responsibility of interpretation and quality more towards 3rd parties and gives more room for discussion. I find decision to be a strange one by the Code committee.

 

[SectionIX]

You could use an undercut comparator shim to estimate the depth of IP on a radiograph...

 

[DekDee]

XL83NL,
I could probably write a small book on the things I find strange in B31.3.
That is why I decided to bite the bullet and volunteer.
What is the old saying ? - if you can't beat 'em, join 'em !

In your opinion it may be strange but a group of professionals who are far more experienced and knowledgeable than me have decided to delete it.
Regards,
DD

 

[XL83NL]

I have considered becoming a volunteer, but when you live outside the US, so it is hard (perhaps impossible?) to attend the meetings in your own spare time and at own expense. If these meetings where to be held in my country (Netherlands), I would consider them

 

[DekDee]

You do not have to attend any meetings.
There is an International Review Group that is part of the B31.3 Committee (I live in SE Asia).
You just need to have a sponsor / mentor who looks after you whilst you go through the Prospective Member period.
If I can assist further drop me an e-mail - sbrwelding at hotmail.com

I find it quite funny now I know you are based in Europe - you think B31.3 is strange, I think the majority of the EN/ISO codes are a nightmare and if someone mentions PED I run as fast as I can in the opposite direction.
Regards,
DD

 

[XL83NL]

Ill drop you an email, thanks for the idea.

I personally prefer the 'US approach' on pressure equipment design. You have a code for (almost) everything and it's nested in the law. The AI decides. No discussions. Period. What I also dont like is how some of the young EU standards, like EN13480, are vague on certain aspects, or incorrect, or inconsistent. Materials are a pain in the *** to purchase, even for a plain 316 welding tee Ive had a hard time getting it. On the other hand, the codes do provide a more prescriptive way on some aspects ASME lacks rules for, e.g. fatigue and creep design.

With the PED for EU, what I often find to be the bigger issue is that a lot of pressure vessels shops cherry pick in codes or code sections. Satisfying PED has become prio 1, satisfying a code prio 2. What it ends up with, in my experience, it that neither is satisfied. Imo the shops (especially the smaller ones) dont have a fricking single clue about what's actually in a code book, or the PED. The Nobo's sometimes dont know either, or just dont care. What you want is a shop with ASME stamp, but that makes the vessel 50% more expensive. The people who do the budgetting decide, and you know what it ends up with then. Theyre just interested in scoringthe next big job with a max on their 'fair' share.