In some ways, you can compare this to other NDT methods. For instance, when you do MT and you get a linear indication running longitudinally down the center of a weld, and it "looks" like a crack (meaning that the indication has a crack-like shape to it), you can call it a crack with a fairly high degree of confidence, even though you cannot actually see the discontinuity - all you can see is the indication.

With manual shearwave UT of welds (this is what you referred to) it is not quite so straightforward. An experienced (I would underline that word if I could) UT tech can often determine what type of discontinuity is causing a certain indication - but not always. In order to be able to do this, the UT tech must be armed with all of the relevant information, such as: material type, welding process, joint configuration and dimensions, welding position, etc, etc. With this information, together with the basic determinations that you can make about the shape of the discontinuity from probe selection and manipulation, and from the other characteristics that can sometimes be determined by how the indication appears on the screen (e.g. does it walk?, does it show facets?, etc, etc), a good experienced UT tech can sometimes be able to determine with a fairly high degree of accuracy the discontinuity type. It is not easy to do, and most of the UT people I've worked with in my career cannot do it well.

Most if not all codes recognize this inherent difficulty in determining discontinuity type. They then use other characteristics (such as amplitude and length) for determination of acceptance.

All that said, I would agree with your interpretation of AWS D1.1 Section 6 UT. It does not permit the UT operator to arbitrarily accept a discontinuity simply because the operator believes it to be porosity.
Mankenberg

kipmank summed it up pretty well with his post. I haven't been "bouncing sound" as long as some of these folks on here so, the only thing I thought of to add was that when you pay attention to the shape of the echos that are returning on the screen, and thier orientation you should notice a pattern after a while. These patterns are what kipmank is referring to. Along with the orientation of the indication within the joint, a sharp edge(ie. from a crack) will have a very sharp, distinct, narrow, thin signature where as a pore of porosity will reflect a signal that might show up as a blunt, sometimes wider, signature. I feel that the lack of fusion is the hardest for me to figure out by simply looking at it's signature on the screen and most of the time the welder won't even see it when they gouge it out, yet when you recheck it after the repair, the signal is gone, no more reflectors.
John Wright

I too would agree with Makenberg & John. The UT industry has recognised the limitation(s) of exact identification and many technicians use terms like "planar", "spherical", "irregular" or "cracklike". In addition to assessments from probe signals and scanning techniques mentioned above, I also rely on accurate mapping using beam path overlays.

Here is a sample. The joint details are drawn on the graph paper For the overlay I use the transparent sheets sold for use with overhead projectors. They are quickly and easily made and I file the sketch and overlay in the job file along with the report and technique sheet.

http://img.photobucket.com/albums/v377/thirdeye2/welding01/2513e75c.jpg

~thirdeye~

Yes, an experience UT tech may be able to tell the nature of the discontinuity. In my experience, most UT IIs (including myself) are reliable with location and not necessarily capable of discerning the particular discontinuity (nor under many codes are they required to). I am planning on attending a class before the end of the year to improve this aspect of my ability to perform UT competently. Also, I agree with your interpretation of D1.1.

Without trying to sound abrupt, but if you failed (or would have failed) the weld, why was it not repaired and accepted before the UT II came to your shop to do his inspection? I do not present any welds for outside inspectors to inspect that have not already been accepted with an in-house inspection.

QCCWI: I agree!! If it is rejectable based on the rating then reject it and repair it. Who exactly is this guy doing favors for by attempting to accept/reject based on his characterization of the indication. There is always the chance this kind of thing can bite you in the butt later down the road because of the liability involved. There are a lot of UT guys and CWIs who can characterize an indication with pretty good accuracy based on its location in the weld when plotted out. Even then the rating of the indication takes precedence.
Regards, Donnie Mann