Not logged inAmerican Welding Society Forum
Forum AWS Website Help Search Login
Up Topic Welding Industry / Inspection & Qualification / Backing bar indication help
- - By supermoto (***) Date 01-28-2009 14:49
I have been trying to get my OJT for UT.  I have gone through two community college UT classes, so I am very confident in what to look for except when it comes to the backing bar indications.  Whenever I see something and I see at the root the level II guy goes behind me and says it is a backing bar.  This just doensn't make sense be cause I figure my skip and surface distance for L1 and L2 per whatever material thickness, set the machine to that thickness, and I see these indications at the root of the weld with a depth within 1/8-1/4.  I have figured out a bunch of the indications being b-bars due to I can't be back far enough for it to skip into L2 even though it says it is in L2.  They appear to look like lack of fusion, a very crisp straight large indication. 

I am just not as confident to say if it is a backing bar indication because we have had a lot of indications that end up being lack of fusion at the root.  It makes me think that is a true indication. 
Parent - - By hogan (****) Date 01-28-2009 15:26
supermoto,
Yes b-up bar UT can be difficult. It was for me at first. But with the proper experience it can much easier. I have seen too many L-II's that call indication at the root B-up bars. There was an interesting situation that I happened a few years back. Our local municipality wanted to get an idea of the proficiency of L-II's ability regarding B-up bars (FEMA). They had a couple techs from all approved testing labs evaluate mock-up, all with b-up bars. Out of about 15 techs only 2 had the ability to make calls distinguishing the IF from B-up bars. From my experience, there are only two definite techniques for distinguish the two. First, Phased array. But currently this is not particle. Secondly, it has to be in the first leg. If you need to have the weld ground flush do it. At least on the first one to get an idea of how the signals evolve. Another not so technical, and not so accurate, is to look for dramatic changes in root signals. If you can scan parallel to the weld and you go from a +18 B-up bar signal to a +8 then I would want to spend some time drawing things out and maybe grind a section of the weld flush if not in my first leg. Experience is great. But unless the experience is doing things correctly, your wasting your time.
Parent - - By supermoto (***) Date 01-28-2009 16:47 Edited 01-28-2009 16:52
Well I work in a production shop and they want it out the door "yesterday" so we don't have time to do any grinding unless the code requires it to be done for that joint.

I was talking to our Level II and he said if he sees anything that is 1/32 before the end of L-1 sound path than he says it is the b-bar.  If it is 1/16 or more before the end of L-1 sound path than it is most likely lack of fusion or some other type of indication.

Hope this makes sense.
Parent - By dmilesdot (**) Date 01-28-2009 17:25
Supermoto, a couple of things come into play, first verify your angle so that your surface distance calculations are dead on, second verify the thickness of the plate so that again your distance calculations are accurate and your sound path is right on to determine the end of leg one.  D1.1 6.2has a good diagram of ways to examine a backing bar weld. They also show ways to de-mystify the signal. A lot of it comes down to experience.  I used to take welder quals that were one inch thick with a backin bar, ut them, cut them apart and prove to myself what I was looking at. 

Dave
Parent - - By hogan (****) Date 01-28-2009 19:45
QC, in my opinion has nothing to do with production. It has been that way ever where I have worked. Your job is not to get it out the door. What if you make a bad call and the part fails and people die. At least you made the production schedule? The code has requirements for grinding whenever you need to for UT inspection. I have rejected many, many indications that were in the last 1/32" of of leg 1. Then cut and performed WFMT, and there was IF at the root. So I call BS on your level II's statement. Why don't you ask him for a code reference on that one. Just remember your choices today can effect the rest of your career.
Parent - - By supermoto (***) Date 01-28-2009 20:01
Hogan,

I could be quoting him wrong or maybe you don't understand what I am stating about the sound path distance to help indicate a b-bar indication.

I am sure most of it comes down to experience, so that is what I will be working on.
Parent - - By hogan (****) Date 01-28-2009 20:18
supermoto,
I think that I understood what you were trying to communicate. It is just that I see his opinion as flawed. I would not be surprised to see someone who made a comment like that to be seen performing UT thru paint and claiming it was OK, it's a thin layer and I can see the B-up bar so I know I'm getting good sound.
Parent - - By Maxpayne575 (*) Date 01-28-2009 21:07
+1, most of the time a weld with a backing bar will have a characteristic reflector from the root zone. As stated earlier, slight changes in that indication could indicate a problem. Backing up to the third leg, is only an option if the exact angle and exact thickness are known, so you have to calibrate carefully, and record the thickness during your lam-scan. Grinding flush in the questionable area and scanning directly over it is my prefered approach, we use the D1.1, which alows us this option. In fact I believe it mandates it in some circumstances.

As far as your level II goes, I hope he's not contributing every root reflection to B-bar, if so he's cheating himself, and you I'm afraid. See if you can talk him into digging one of those out and "showing you" his theory.
Parent - - By trapdoor (**) Date 01-29-2009 18:37
Sometimes you have to make the call and have them gouge down to the root. If it is gone the next time you scan then you know it was a relavent indication. If not then it could be the back up bar or a welder who can't get things right in two tries. Also if you know the root opening yo can plot the weld out and see if a back up bar indication is even possible.
Parent - By Maxpayne575 (*) Date 02-02-2009 20:07
+1 careful plotting is a good thing to practice.
Parent - - By CWI555 (*****) Date 02-03-2009 01:43 Edited 02-03-2009 16:54
You are only as good as the knowledge you acquire through the experience of others. The caliber of those doing this training will be directly reflected in your abilities and a direct reflection of the teaching abilities of those who trained you. There were three key persons in my career that turned the lights on for me. George Lawrence, Emory Roberts, and Hector Diaz. I consider Emory to be the best I've ever seen for manual UT. George was a walking encyclopedia of UT theory, and Hector was the best AUT/TOFD hand I've ever seen. I was very very fortunate to have had the opportunity to pick their brains for their collective experience and knowledge and they were professional and conscientious enough to freely give of that knowledge to people who wanted to learn. They themselves were a collection of knowledge of those who mentored them and were not afraid to state so. There were others who filled in one hole or another in my knowledge base for UT, but those three are the cornerstones of my abilities.

Alternatively, if you are unfortunately saddled with being trained by people who are legends in their own minds, you will end up with the antithesis of the personal experience I describe.

I've seen a lot of techs(?) over the years who can only turn a machine on and go through the motions without any real understanding of what it is they are doing. As long as the machine tells them it's so, well that's the way it is. They either never took the initiative to learn the physics and mechanics behind UT or they were unfortunately trained by a con artist that only muddied their understanding of the method. 

I would be very very careful if I were being trained for any method. Look up the books, find anything you can on the method. Independently verify what it is you are being told by that trainer. In short, take the initiative and maintain a questioning attitude. If the person training you can't back up in writing what they are trying to teach you with authoritative text, they are probably just a legend in their own mind.

With that said, I'll try to address the technical aspects of your post.

Maxpayne stated it this way "backing bar will have a characteristic reflector from the root zone"
Which is correct in so far as it goes. Long experience will eventually reveal the "characteristic reflector".
However, what is it and why is it.

Looking at it from the back diffraction standpoint (similar to TOFD or aka tip echo method) an understanding of the echodynamics of the particle motion will help your ability to discern the difference between a benign backing bar signal and a flaw signal.
ASME Section V 2007 edition has attempted to alleviate the general misunderstanding of what that sound is doing, and the wave mechanics behind it. Appendix N of article 4 has given a fairly good basic version. NDT.net also has a lot of good information in this regards from the Euro perspective.

If you dig out this particular appendix note that for TOFD the positive and negative phase of the waveform is presented as a white and black line. Look at the images and you will note that the entrant surface and the back surface are 180 out of phase with each other.
http://www.kettering.edu/~drussell/Demos/reflect/reflect.html
A look at these waveforms will be enlightening. The hard backsurface of the steel creates a 180 phase change, passage through the opening of a fully penetrated backing bar will create the centerline 180 reflection, and from either edge of the opening a tip diffraction from either edge of the tie in (one on the front side of the main signal and one on the back depending on the distances involved they may both be on one side or the other depending on the time) the upper tip will be negative, the bottom of it will be positive, however; all of them will be surrounded by the 180 out signal as the base metal will give you a hard edge reflector as well. There are many other factors involved, but if you've read this, you now have a starting point to pick up the rest for yourself. (this post is getting long enough as it is)

Expanding your timeline with the RF screen for verification purposes, wheither or not your looking at a lack of pen, or a root area lack of fusion will be as clear as a neon sign once you've trained yourself to recognize the "characteristic reflector" as maxpayne put it.
For those who have long experience in it, they can typically recognize it with a fully rectified screen.

While that may sound like hocus pocus to the prospective UT technician, the physics of it don't change for any of us.

this is all information that machine will not give you. The machine assumes a straight line, and does not take into account beam spread, echodynamics/wavemechanics, nor does it take into account mode conversion (another tell tale sign of a backing bar when your scanning at 60 or 70 typically) and a host of other factors.

Learn the method,  not the machine is the best advice I can give you. Doing anything else just makes you an operator, and not a technician.

Regards,
Gerald
Parent - By hogan (****) Date 02-03-2009 15:22
Very well put.
Up Topic Welding Industry / Inspection & Qualification / Backing bar indication help

Powered by mwForum 2.29.2 © 1999-2013 Markus Wichitill