Phased Array Ultrasonic Examination

[MikeMet]

Need to settle an argument. Is phased array ultrasonic examination of welds considered a sophisticated version of shear wave examination? Or is it classified as something other than a shear wave examination?

 

[redpicker]

It can be, and in its usual application, yes, it is.

Phased array uses an array of smaller transducers that can pe "phased" by electronic control to produce anything from a full-compression beam to a angled "shear" beam to a true surface wave (what is it calles, a lambda wave?). I believe the method holds promise as being the "new best thing" in NDT, but the usual applications I have seen it used for are just building a angle beam to mimic what you would get with a wedge.

To me, the potential is to be able to scan a material using multiple angles to optimize inspection at a variety of orientations. This should also enable the inspector to examine an indication using a variety of angles and effective transducer sizes to provide a better evaluation of the indication. I have not seen this potential used, however.

rp

 

[dbooker630]

The way it was first explained to me is that PA is like using a 45, 60, and 70 degree wedge all at once.

 

[Duwe6]

"45, 60, and 70 degree wedge all at once." And every angle between them, in 1° increments. The power of PA comes from the computerizes displays possible. The simple, cheap(er) units just show you a fan-shaped scan. ". . used for are just building a angle beam to mimic what you would get with a wedge." is all that unit is good for.

A sophisticated unit like OmniScan can show the fan. and also can a store strip recording of a scan along a weld. That stored scan can be viewed as "looking down" through the weld, or as "looking sideways" through the edge of the plate and into the weld. When properly calibrated, all defects are color-coded, such that Red = rejectable, Yellow = borderline acceptable, and the cooler colors are acceptable reflectors.

Then the raw scan can be transmitted via the internet to any location on the planet, to allow someone else to 'read' the scan and decide if your characterizations of the defects found was accurate.

PA with recorded, indexed scanning will eventually kill radiography, and the need to tote around highly radioactive sources to make those RT's.

 

[metengr]

Yes. However, to settle an argument requires reputable references. The best two sources of information are ASME, Section V, Article 23, Ultrasonic Standards, under which ASTM E2491-06 is contained titled, Standard Guide for Evaluating Performance Characteristics of Phased Array Ultrasonic Examination Instruments and Systems.

Excerpt from E2491-06
4. Summary of Guide
4.1 Phased-array instruments and systems have similar
individual components as are found in traditional ultrasonic
systems that are based on single channel or multiplexed
pulse-echo units. These include pulsers, receivers, probes
and interconnecting cables. The most significant difference
is that phased-array systems form the transmitted ultrasonic
pulse by constructive phase interference from the wavelets
formed off the individually pulsed elements of the phased array
probes.
4.2 Each phased-array probe consists of a series of
individually wired elements that are activated separately
using a programmable time delay pattern. Varying the number
of elements used and the delay time between the pulses
to each element allows control of the beam. Depending on
the probe design, it is possible to electronically vary the
angle (incident or skew), or the focal distance, or the beam
dimensions, or a combination of the three. In the receiving
mode, acoustic energy is received by the elements and the
signals undergo a summation process utilizing the same
type of time delay process as was used during transmission.

 

[Duwe6]

Actually, ASME is a *very* poor reference for Phased Array. It has been avoiding the inherent problems of quanitfying and developing calibration/testing/certification requirements. ASME may be waiting to see what the NDT industry develops. Best reference code for phased array PAUT at this time is API.

There is quite a bit of info available at NDT.net, most of it excellent. Anything posted by Ed Ginzel or Mark Davis may be taken as fully accurate, with testing, peer review, and documentation behind it.