Odd D1.2 question

Date 02-02-2016 19:17 Edited 02-02-2016 19:22

Working for a customer, we shot some digital RT on a part. Part is 6061-T6 AL, it is a CJP weld with a backing. Now when we did RT we got some of the typical lines you would see on a CJP weld with a backing. Not so so much of "wagon tracks" per say but straight lines with intermittent spots of what appear to be possible slight suck back or maybe even undercut at the root upwards. We cant remove the backing strip, all we have is the RT to look at. Now myself a lowly level 1 RT, I dont have to make the call. But asking as a visual inspector, Had they removed the backing and it is VT'd, would any type of below surface on the root be classified in undercut? or profile? maybe even underfill in a way since the total filling of the groove is not full? Acceptance on this weld per D1.2 2014 per table 5.3 there are a few different ways to approach it. Curious any input.

Now this one one frame of 24 there are much worse areas but I just wanted to show the idea. The part failed in RT in many places.
I wont get to cut the backing, just asking hypothetically if we did how you might approach it.

here is an image -------> http://imgur.com/a/BqKmr

Ive circled other problem areas

J

By 803056

Date 02-03-2016 22:45

How about providing a cross section through the joint so we have a better idea of what we are looking at? Knowing the configuration of the joint can provide some clues as to what is going on. Some if the indications I saw may have nothing to do with the weld.

Nice radiograph. Love that digital stuff, but I would have to be working with Space-X to afford the equipment.

Al

By MRWeldSoCal (***)

Date 02-03-2016 23:57

Al

This is all I really have as far as images go. I could load up the other 24 but this is the idea. It is a long box made from plate, with V-groove CJP welds running long way with a .188" backing strip. The box is approx 14' long with 4 CJP welds holding it together. The backing strip is stitched or tacked throughout the backing which you can see a few in the image. The images are shot downward to the plate so the welds and defects in the images are as if you are inside looking out. There are some spots that have great fusion, you cant even tell there is a weld there, then tracks begin to appear.

The digital stuff is pretty awesome I agree.

Jordan