Not logged inAmerican Welding Society Forum
Forum AWS Website Help Search Register Login
Up Topic American Welding Society Services / Certifications / Question about coupon discontinuities
- - By jsdwelder (***) Date 04-12-2016 20:33 Edited 04-12-2016 22:32
I have an example situation that maybe someone can give me some insight into. Suppose we have a test weld that is a 5 inch long plate and it exhibits undercut that is 1/32" deep the entire length of the plate, which obviously would result in that undercut running the entire width of the coupon (3/8" plate thickness in this example, so resulting in two 1 1/2" root/face bend coupons). Does this undercut count against the allowable discontinuities? AWS D1.1 standard in this example.
Parent - - By welderbrent (*****) Date 04-12-2016 23:32
Okay, first I hit the wrong button and lost my response and then my internet connection disappeared... it has been a long and bad couple of days already this week so I am not feeling real generous.  Here goes anyway:

First, you can't possibly use 5" plates for that test and be conforming to D1.1.

Second, what does the code say about undercut for coupons?

There, generous enough for now.

Parent - - By jsdwelder (***) Date 04-12-2016 23:54
5" was just an example of a plate length. Use any length you want. My question is more about if you had any length of visually acceptable undercut do you count that in total accumulated discontinuities? My thought is no but I was looking for a second opinion.
Parent - - By welderbrent (*****) Date 04-13-2016 00:36

First, I apologize for my first post.  My bad week does not need to be taken out on you and it was wrong.  Let's start over as far as your question goes.

For any others not understanding, D1.1 calls out 7" plates in Clause 4 when doing welder qualification on limited thickness regardless of side bends or face and root which is the preferred method of most of the CWI's I know and many here on this forum.  5" is the minimum length when doing an unlimited thickness, 1" plate with side bends.  Takes less material so you can get by using less material.  But I don't recall seeing anywhere that it says you can shorten the 3/8" plates just because you decide to go with the side bends instead of face and root.  And if you still aren't following: 1" plate or 3/8" with side bends takes 5/16" per slice for the bend which gives you a 5/16" X 3/8" part to bend for limited and 5/16 X 1" for unlimited whereas the face and root are cut at 1 1/2" so it takes more material. 

The area around the part JSD wants opinions on most: undercut.  Forget Table 6.1.  We are going to Clause 4 at Visual Inspection of Groove Welds which applies to both PQR/WPS and Welder Qual as you are referred back to there from 4.20.1 as there must be a successful visual examination prior to cutting and bending the coupons. 

So, in 4.9.1 and paragraph (4) we see no length prerequisite only that undercut shall not exceed 1/32".  So, if the theoretical undercut JSD is asking about is 1/32" or less then it is acceptable regardless of length. 

Now, at this point though I must ask a question: What is, and where do you find any reference to, "total accumulated discontinuities"?  I am totally unfamiliar with this concept.

Porosity of certain sizes is added together to make sure it does not exceed 3/8" within a certain linear length but I don't see that in  And, from Table 6.1 you still don't mix your porosity, undercut, overlap, and/or any other discontinuities.  Each stands alone with its own requirements. 

Undercut is undercut and in the question posed it appears to be acceptable regardless of length.  But, if I were the welder, I would not submit a coupon for bending that way.  It more than likely won't pass.  Fix it, run another plate, whatever it takes. 

He Is In Control (though I may not always act like it), Have a Great Day,  Brent
Parent - - By jsdwelder (***) Date 04-13-2016 01:25
I shouldn't have said total accumulated discontinuities. I should have worded it total accumulation(size not number of). So I guess I am right to think that if the undercut passed visual then it doesn't come into play when inspecting the bent sample.
Parent - By welderbrent (*****) Date 04-13-2016 02:01
"doesn't come into play", well, yes and no.  Just because it passed visual doesn't mean it won't be considered again upon examination of the bent coupon.  One must at that point go to and to make sure of the criteria of the bent coupon.  The undercut may induce a discontinuity which rejects it upon examination after bending. 

Parent - - By welderbrent (*****) Date 04-13-2016 02:02
Where are Al, John, and Lawrence when you need them :lol:
Parent - By 803056 (*****) Date 04-13-2016 02:15
I was waiting to see whether you would sink or swim Brent!

The undercut can run the entire length of the coupon, along both toes, and it is acceptable as long as it does not exceed 1/32 inch. As you pointed out Brent, the acceptance criteria is found in clause

I also agree that the welded joint in the 3/8 inch plate assembly is specified as 7 inches minimum, while the 1 inch assembly is 5 inch minimum.

I agree once more with your comment that the undercut has to be evaluated after the guided bend tests are performed to determine whether there are any cracks in the undercut regions.

Nice job Brent. You can swim!

Best regards - Al
Parent - By Lawrence (*****) Date 04-13-2016 11:22
You did a great job Brent... I have nothing to add at all  :)

Except for being sorry for not posting lately... Been in Florida visiting the Mother Ship !

I think I may have gotten myself on the Clause 5 task group :)

Beware water, oil, grease and *other* hydrogen based materials!

More about that off-line if you like Brent and Al
Parent - - By pipewelder_1999 (****) Date 04-13-2016 02:29
The question opens up this question... Is undercut a discontinuity? According to the definition " interuption in the typical structure....." then undercut on a bend specimen appears to be a discontinuity.

In reference to bends, D1.1 uses the term "discontinuity" and not a more specific term "open discontinuity" like used in some other codes. That could include those little wrinkles that occur if one was so inclined to follow the code verbatim.

If undercut is a discontinuity and you are doing face or root bends and the undercut doesn't miraculously dissappear (even if it was acceptable before bending) then after you get the bends done, you have a big decision to make.

Cannot tell you what to do but I understand the situation. Maybe a code committee person can make an official interpretation.

Hope this helps and have a great day.

Gerald Austin
Greeneville TN.
Parent - - By jsdwelder (***) Date 04-13-2016 18:08

That is exactly the situation. The problem I have is I do not want to make a big decision, whereas my opinion comes into play.  I want to do what the code directs me to do. Personally I would fail it if it were up to me, but AWS code trumps my opinion.
Parent - - By pipewelder_1999 (****) Date 04-13-2016 20:00
I think if you look at the code and it defines something as a "discontinuity" and that discontinuity appears on the convex face of the bend. The acceptance criteria is clear as to what it says.

I have seen some say that they do not blend in acceptable undercut on specimens before they bend them. That of course leaves a "discontinuity" based upon the definition. And they of course have a decision to make regarding verbatim interpretation of the code and deciding what the "intent" was. The latter of which is not addressed in the code.

Parent - - By Newt18 (*) Date 04-14-2016 03:28
Hi Jsdwelder. I would inform the welder of the high probability that the samples will fail when face and root bends are performed. I like to keep visual aids on hand for show and tell too. Almost always the welder will run another pass. If they elect not to, prep samples to code. I don't think I have seen a root or face bend not fracture and fail the 1/8" max length with the amount of undercut in your example.
Parent - - By jwright650 (*****) Date 04-14-2016 10:26
"I don't think I have seen a root or face bend not fracture and fail the 1/8" max length with the amount of undercut in your example." - Newt18

I agree, it will tear along that undercut when the face stretches as it bends.
Parent - - By jsdwelder (***) Date 04-14-2016 11:07
Not to discredit what you have seen, but I have witnessed a number of coupons that have not opened up any further. I appreciate everyone's input. Thanks guys.
Parent - By Lawrence (*****) Date 04-14-2016 11:19
I'm with you... I've done hundreds of root and face bends (limited thickness D1.1) that have had acceptable amounts of undercut as defined in this thread, that have survived guided bends... Typically as the coupon elongates the undercut stretches out a bit.

Now if there is a corner crack related to the undercut, I do NOT allow the 1/4" exception or an alternate sample.  So that's where I'm a hard A##
Parent - By pipewelder_1999 (****) Date 04-14-2016 11:34
I have never seen one with that length of undercut but I have seen a few with undercut that was visible after prep that was over 1/4" and never open up when bent.
I am sure the severity/angle of the notch at the root of the undercut would severly increase the likelihood of failure.
Of course what I have seen in the welding world is limited to just my experiences. Someone that maybe worked in a test lab or similar would have some representative experience.

If the bends cracked or otherwise open up I think the question may not have been put forth.
Parent - By pipewelder_1999 (****) Date 04-14-2016 11:05
How about some pics of the bends when they are done? Before and after would be great.
Up Topic American Welding Society Services / Certifications / Question about coupon discontinuities

Powered by mwForum 2.29.2 © 1999-2013 Markus Wichitill