ultrasonic testing of welds - any here do it?

Hi all

Training in UT of welds.

Anyone here do this - ultrasonic'ing welds?

Finding it a bit challenging and absorbing. On a course now.

So far, with test plates, broadly seem to be able to identify there is a defect there - locate it lengthways from beginning to end - and state where it is in the height of the weld with some idea of how tall it is.

However on finer points of drawing it together so can classify

- what type of defect it is

- where it is in the cross-section

- how wide it is in cross-section

- etc as you'd need to say "it's a lack of root fusion full root-gap about

3mm high" or "it's a 4mm high lack of sidewall fusion on the LHS starting at about 10mm depth" or "it's a slag inclusion about 3mm wide and high at (x, y)" - not got anything like there yet

Any got hints for someone trying to find their way in?

Richard S

Reply to
Richard Smith
Loading thread data ...

I did my classroom work 6 years ago at Hellier, and only now am I getting field hours as an inspector.

A lot of the work of defining the flaws is based on context. With experience you learn what kind of flaws occur in what kind of welds and in what locations. The UT machine allows you to set location, but experience will help a lot in interpreting the scan screen. Also the newer machines make it much easier to disallow non-relevant indications.

Make sure to read the D1.1 section on UT. It is much more useful than the UT manuals for actual UT scans. Remember that in shear wave scanning with an angle transducer the direction of the Transducer determines how certain flaws will show up. Try not to be fooled by weld geometry.

The ends of a flaw can be determined by when the indication drops to half of it's peak height.

Reply to
Ernie Leimkuhler

Thanks Ernie

Out of TWI and their UT-welds training course

formatting link

3 crazy weeks of taking it in turns to be on suicide watch for whoever was feeling lowest at any particular time. Brilliant time.

Got to look at AWS D1.1 for first time as stopped for rest while driving home last night. It's an amazing Standard - stands out.

Definitely all the real learning ahead of me. Already thinking of what I need to do. One thing would be to make defect-free welds (internally - no cracks or volumetric defects) but with external shape defects like over-penetrated root, excessive cap, excessive cap with narrow-angled cap-toe say approaching 90deg. See what sort of echoes you get. Particularly root shape. That caused me a lot of problems - "noisy roots" which were not actually defects.

Then make samples with l.o.r.p.'s (lack of root penetration) of different heights which you damn' well know are there and play with the echo dynamics and location measurements. And what probes give what echoes - 60deg and 70deg.

See if can achieve some single-sided l.o.r.f.'s (lack of root fusion)

- locate them with a magnet and iron-filings so you know where they are - and again, play with echo-dynamics with probes.

Can see myself spending a fair bit of time at my table in my apartment with the UT set I have on loan if I can get to make such weld samples.

Thanks for the encouragement and I hope you are getting on well.

Rich Smith

Reply to
Richard Smith

If you need to make weld cracks, use TIG and 309SS filler metal on steel. Crank your amps up and really cook it hot and fast. Almost guaranteed to make lovely deep cracks.

I did some work for a flex hose welding shop to figure out why their welds were cracking when they switched to 309SS fill. The problem was that they were welding too hot and too fast. When you get 309SS too hot it melts into the base metal too much which causes massive shrinkage cracks.

With practice I could make a weld crack every time. It was kind of fun for making test welds.

Reply to
Ernie Leimkuhler

Ernie

All my book learning leads me to say "hey yes - you would expect that!"

long comet-tail weldpool (planar solidification fronts meet in middle) + austenitic stainless => solidification cracking

What I don't understand is how you can then use it as a UT test sample. You'd have austenitic weld metal which I am told ultrasonic waves struggle to penetrate ??? Yet to try it and see it for myself. Keen to! Told that when grain size larger than wavelength, as in austenitic weld metal, scatter is chronic and UT unusable. ???

Rather like when driving putting on headlights to main-beam in fog - all you get is more white-out reflection (all "noise" no "signal")

Obviously it works for you - just trying to puzzle that one.

Anyway, Friday hopefully going to place where they've been teaching me UT - where they also do welder Codings (Welder Performance Qualification in ASME IX speak) - so they'll have some qualification-pieces I can UT.

Any suggestions on what to try that's worked for you and your students in getting an understanding of UT? Likely there will be some austenitic stainless welds around.

Rich Smith

Reply to
Richard Smith

The grain size is rarely so large that it blocks your signal.

Most UT in welding is on steel, and flat steel at that. You have to figure out how to scan things like Wide Flange Beam butt splices. The transducer is like a flashlight being shown into a dark room. Think about what direction the flashlight needs to be pointed to see the area being scanned.

Also keep in mind that standard shear wave transducers won't work on pipe under 24 inches diameter. For smaller pipe you need custom shoes to match the curvature.

Reply to
Ernie Leimkuhler

Ouch! We've just been UT'ing pipe down to 5inch (125mm) dia :-(

While respecting your point that a probe for a plate is not really suitable... As I found it - you have yet another probe motion in maximising your signal - rolling side-to-side (like one of the motions of a ship), as well as the forwards-and-backwards and swivel (beam-path & direction).

Thanks for all the message. Noted those good points. Must look for an exposed structural steel with a welded joint within reach of a electric supply socket :-)

Rich S

Reply to
Richard Smith

Some decades ago, when the extent of the intergranular stress corrosion cracking problem in large austentic weldments was being discovered, it was noticed that some very large flaws were completely missed or underestimated in size with the inspections being done at that time using UT equipment designed for steel. The problem was solved with the design of specialized UT equipment designed specifically for inspection of welds and HAZ in austentic reactor safe-ends and pipe, and a requirement that the new equipment be used incorporated in the ASME pressure vessel code sometime in the 1980's.

first reference I found:

formatting link

Development of Ultrasonic Transducer Testing Austenitic Stainless Steel Weld

cut and paste of summary:

  1. Requirement on transducer testing coarse grain material The coarse-grained and inhomogeneous structure in austenitic stainless steel weld results in severe scatterance and attenuation of ultrasonic energy. In order to increase SNR (Signal to Noise Ratio) for farther enhancing the testing sensitivity and detectability of defects, the incident acoustic wave should be low-frequency, long-wavelength, which behaves lower attenuation, more penetration and higher SNR. In addition, pulse duration cycles also
1 evidently affect the SNR of defect-echoes. Commonly, the more short duration, the better SNR is[2]. In a word, to successfully inspect austenitic stainless steel, the transducers with low-frequency, short-duration pulses, high-sensitivity and high- resolution should be used.
Reply to
Glen Walpert

Thanks Glen - brilliant - Rich Smith

Reply to
Richard Smith

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.