ASME B31.3 Acceptance criteria for welds 1

[engrom]

Peers,

As per ASME B31.3 Table 341.3.2 Acceptance criteria for welds, is it important to meet all the acceptable values mentioned in the criteria. For eg. lack of root penetration for Normal and Category M fluid - Girth, Miter groove and branch connection shows B as acceptance criteria. Notation B has two acceptable values in the table. One for depth i.e. <1mm and <0.2Tw (both are less than or equal to) and the other for cumulative length i.e. <38mm (also less than or equal to) in any 150mm weld length. Does this mean that both depth and length should be within the limits to accept the weld. RT cannot detect the depth as it is 2D image, so if in this case when the depth cannot be determined, is it possible to reject the weld? Is there an interpretation of this available from ASME?

Appreciate your help.

 

[metengr]

You do not need an interpretation, read NOTE (6) for acceptable value limits.

 

[engrom]

Metengr -Thanks. I've read it already, which does not answer my question. My query is whether both depth and length of IP shall be within the acceptable criteria to pass the weld or not. In the table for notation 'B' has two line items which are not separated by an 'and' or an 'or'. RT reports normally determines the length only, but depth is just an assumption based on the image density compared to parent material. If we're not able to determine the depth of IP for line with class 600# or above and cat M fluid, does the weld qualify based only on the length of IP?

 

[metengr]

engrom;
Thanks for re-stating your query. My reply based on review is yes, both depth and cumulative length criteria would apply if using this table because both are stated under symbol B. Similar to symbol C for lack of fusion or even slag inclusion for symbol G.

 

[engrom]

Thanks Metengr. Sorry for another question related to RT reporting. One of the testing contractors states that as per ASME V it is not required to mention the defect size in the report if the discontinuity complies to the code criteria. They reports defect size of non-compliant welds only. My point is that, even though the defect complies to the code, it is important for the client to know the defect size especially for LOP, UC etc. to monitor weld defects while in-service as to how internal erosion/corrosion affect the piping integrity, especially, LOP etc, when the size is closer to extreme limits (i.e. 38mm). Does the code state not to report size of compliant defects?

 

[Kiwi2671]

Engrom,
Just remember it cannot be a compliant defect.
If it is compliant it is a discontinuity, when something does not comply with a specific code it becomes a defect.
Regards,
Kiwi

 

[Kiwi2671]

engrom / metengr,
I have posted this question before on another welding forum and could not get a definitive answer (even from members of ASME code committees)- maybe metengr can help.
In an internet cafe in Vietnam so do not have access to my codes but my query is:
B31.3 allows up to 38 mm of IP (both edges of the root unfused) but allows no Incomplete Fusion.
The pictures detailing the joints show Incomplete Fusion of the root Due to Misalignment but what if you have Incomplete Fusion of the root without Misalignment ?
How can not fusing both edges of the root be acceptable up to 38 mm but not fusing one edge of the root is unacceptable ?
Your thoughts ?
Regards,
Kiwi

 

[metengr]

Kiwi2671;
My opinion based on code committee interpretation. We need to keep the two weld defect terms separate for better understanding - lack of fusion (LOF) and incomplete penetration (IP). LOF can be anywhere from the root to the weld cap, and isolated, thus it is controlled by planar depth limits in the weld as a function of weld thickness and less than or equal to 1mm.

IP simply means the weld joint was not completely filled with weld metal at the weld root and deeper into the weld region. IP is limited to length and depth concurrent (for symbol B) because only the root is affected either on one side or both sides, and in this case LOF at the root could be treated as IP. Also, type of weld and fluid category service establish if both IP and LOF are enforced concurrently based solely on weld thickness and length (symbol C).

Incomplete penetration is illustrated in two sketches in Figure 341.3.2 in B31.3, c and d. In c, incomplete penetration is caused by pipe end misalignment with incomplete filling of the root on one side of the weld joint. However, in d incomplete penetration is shown without misalignment and is due to lack of fill at the weld root affecting both sides of the weld joint.

 

[engrom]

Thanks 2671, yes you're right... Sorry for my wording compliant defect. I meant compliant discontinuity. is it required to report the size of compliant discontinuity? If not reported how do the client know about the size as in the case of IP the significance of 1mm and 38mm are quite important. Appreciate your thoughts as well metengr.

 

[stanweld]

B31.3 does not require the reporting of compliant flaws. As the purchaser of the NDE services, you may require such reporting.

 

[jharris3]

Yes Stanweld's last statement and metengr are both right.

The owner is required to establish engineering requirements for the system integrity for new service and also for future maintenance programs.

Something said above that I would like to touch on. There are other ways to obtain a permanent record of the examined weld such as visual inspection, use a camera and/or scope w/ camera, UT-phased array and UT shear wave. The last two are expensive at times because the industry won't let go of RT which leaves NDE companies without cash to purchase the necessary calibration blocks, however, if the disposition of the discontinuity is that important then they should be more seriously considered. These methods are at times more sensitive to the discontinuity type and size than RT and if the discontinuity becomes a problem in the future...that means the entire weld would be inspected again anyway.

LOF being so important...RT is not as sensitive to sidewall LOF as UT-PA...

 

[SJones]

Kiwi,

This is where some oil and gas operators make use of Note 1 to Table 341.3.2 and call up ISO 6520-1, ISO 5817 and the IIW Reference Radiographs. It saves a lot of tears in the darkroom arguing between IP and lack of root fusion. The major argument is getting the contractor to spend the three thousand US on the IIW graphs!!

Steve Jones
Materials & Corrosion Engineer

http://www.linkedin.com/pub/8/83b/b04

 

[Kiwi2671]

metengr,
Thank you for your response but I am still confused.
This interpretation makes me even more so.

Interpretation: 13·16
Subject: ASME B31.3-1993 Edition, Chapter VI Inspection, Examination, and Testing
Date Issued: November 21, 1994
File: B31-94-o59
Question (1): In accordance with ASME B31.3-1993 Edition, Addenda a, if a weld has not
penetrated to the inside comer of the root face on one side of the joint and the remainder of the
weld is sound, does ASME B31.3 classify this defect as lack of fusion?
Reply (1): No, refer to Fig. 341.3.2 sketch (c).Question (2): Can lack of fusion occur at the root of a groove weld joint?Reply (2): Yes.
Question (3): Is 1.5 in. of incomplete penetration in any 6 in. weld length for girth and miter
groove welds acceptable for normal fluid service?
Reply (3): Yes, per acceptance criteria listed in Table 341.3.2A.

The answers to Question 1 and 2 seem to be contradictory.
The description in Question 1 is what we class in the Southern Hemisphere as Lack of Root Fusion (one edge of the root unfused)
Question 2 replies yes to "Can lack of fusion occur at the root" - what is lack of fusion at the root if not the description in Question 1 ?

The main question I am trying to get answers for is how can one edge unfused be not acceptable (if it is classed as lack of or incomplete fusion at the root)and two edges unfused is apparently acceptable up to 38 mm in length

Regards,
Kiwi

 

[metengr]

Kiwi;
Here is my view of below the line, which most code committee members will know what this means

Question (1): In accordance with ASME B31.3-1993 Edition, Addenda a, if a weld has not
penetrated to the inside comer of the root face on one side of the joint and the remainder of the
weld is sound, does ASME B31.3 classify this defect as lack of fusion?
Reply (1): No, refer to Fig. 341.3.2 sketch (c).

Reply: The code committee reply is correct, this is classified as IP. The inquiry used lack of penetration, this implies IP.

Question (2): Can lack of fusion occur at the root of a groove weld joint?Reply (2): Yes.

Reply; This is what I mentioned to you in an earlier post, LOF and IP can be one in the same type of defect but only at the root location.

Question (3): Is 1.5 in. of incomplete penetration in any 6 in. weld length for girth and miter
groove welds acceptable for normal fluid service?

Reply (3): Yes, per acceptance criteria listed in Table 341.3.2A.

No issue based on the acceptance criteria in the table for that edition.

 

[SJones]

Try and picture how the two types of imperfection will look on a radiograph.

Steve Jones
Materials & Corrosion Engineer

http://www.linkedin.com/pub/8/83b/b04

 

[Kiwi2671]

Steve,
That is why I am so confused.
I spent 4 years as a Radiographer.
A single straight line on the graph in the root area was classed as Lack of Root Fusion.
Two parallel straight lines on the graph were classed as Lack of Penetration (Incomplete Penetration in US terminology).
One is classed as a discontinuity and is acceptable and one is classed as a defect and is not acceptable.

metengr,
I appreciate your responses but need more help.
If you have a graph with a 32 mm long jet black straight line in the root area and no way to visual or UT the weld (and there is no misalignment)how do you differentiate between Lack of Fusion at the Root (unacceptable) and Incomplete Penetration at the Root (acceptable) ?
Regards,
Kiwi

 

[SJones]

That's when the IIW Reference Radiographs come into their own.

Steve Jones
Materials & Corrosion Engineer

http://www.linkedin.com/pub/8/83b/b04