✕

Advanced ultrasonic testing techniques such as Phased Array (PAUT) have been employed for years for weld-quality examinations. Major construction codes including ASME and AWS have implemented allowances for PAUT and other UT methods into their bodies of work over the past two decades. More recently, a newer UT technique, Full Matrix Capture/ Total Focusing Method (FMC/ TFM), has been gaining traction in the industry, and ASME started addressing it in the 2019 revision of the BPVC, in Section V, Article IV, Mandatory Appendix XI and Non-Mandatory Appendix F.

A commonality found throughout various codes for UT weld- quality exams is the requirement for a Scan Plan. This is true in most every code for PAUT, and it is also becoming the norm for FMC/ TFM. For example, in the ASME code, Scan Plans are required (SHALL!!!) for all PAUT and FMC/ TFM weld examinations. A Scan Plan is a documented inspection strategy that captures important information relative to the examination—typically including material and weld joint information, probe and wedge parameters, focal law/ beam-setups, a depiction of the extent of volumetric weld coverage, scanning patterns and probe positions, etc. Much progress has been made in the industry regarding PAUT scan plans over the years. Most PAUT instruments have very robust tools built into the on-board software these days, and there are even dedicated PC- based software programs that have been created to help in this regard as well.

FMC/ TFM shares many common attributes with PAUT. The same weld configurations can be scanned with both with FMC/ TFM and PAUT, so the part information and coverage required is basically the same—more on that later in this discussion. FMC/ TFM and PAUT essentially utilize the same types of probes and wedges. There are probe and wedge designs that may be optimized more for one technique versus another, but that topic is beyond the scope of this discussion. Things start to vary rapidly between the two techniques when we contemplate the focal law/ beam setup information and the depictions of weld coverage that are required by the codes. For conventional UT and PAUT, the creation and plotting of refracted and/ or beam-steered angles into an area of interest is valid, easy to understand and display. Not so much for FMC/ TFM, and that makes scan planning more challenging, and not as straightforward. There are no beams to plot with FMC/ TFM. So how do we show coverage? Many of the current FMC/ TFM software suites have a robust acoustic field simulator for TFM setups. (We will be showing the Evident OmniScan version referred to as A.I.M. in this paper.)

FMC/ TFM examination is often seen as a two-step process: data collection followed by data processing. The data collection process is carried out using various approaches: Full Matrix Capture (FMC) is considered the classic collection approach, but there are others. After the data is collected, processing algorithms are employed to help image and make sense of the FMC data. TFM is the currently considered to be the gold standard, in both medical and industrial UT, so we will limit our current discussion to that. There is an abundance of information on the basic fundamentals of FMC/ TFM and related- technologies in our industry, including online and also in formal training courses. Due to the limited breadth of this article, it is assumed that the reader already possesses the basic background knowledge of PAUT and FMC/ TFM, and we are simply going to focus on scan planning.

When we start the scan planning process, we have to first identify the properties of the part that will be examined. Part thickness, material type/ velocity, weld geometry—all of these play a vital role in our selection of the overall technique we will use, as well as the probes and wedges we decide to use. For the sake of this article, we will consider two different geometries- 1) Single- V Weld on a 0.500” (12.7mm) thick component, and 2) Double- V weld on a 1.00” (25.4mm) thick component, both carbon steel. The probe that will be used for TFM in this demo for both parts will be a common 5MHz, Linear Array, with 64 elements at a 0.024” (0.6mm) pitch, mounted on a 55° shear wedge.

When creating a scan plan for a weld, it’s good practice to break the weld down into simpler zones that can be individually assessed for their specific coverage needs. For most common welds, these can be grouped into four basic zones: 1) Root, 2) Fusion Zone, 3) Heat-Affected Zone (HAZ) and 4) Volume. Those are numbered in the above images with red letters. As the weld geometry gets more complex, more areas may have to be considered, but the basic four zones still apply in most situations.

The weld root is typically the first zone that is assessed for coverage during the scan planning process, as it is often seen as the most critical part of the weld. Typically with PAUT, this is accomplished by indexing the search unit as close to the weld toe as possible, to maximize 1^st leg coverage with as many angles as possible. Shown below are a common PAUT approach for accomplishing this for both weld geometries, followed by their FMC/TFM counterparts. Note: for each example in this demo, it will be assumed that access to both sides of the weld are possible. The index offset configured and described on the 90° skew will be mirrored on the 270° skew. In the case of single-sided access, a more complicated scenario would have to be considered.

After configuring root coverage, the next area to consider would be the fusion zone. This is the region where the weld metal should fuse with the base metal at the bevel. Common discontinuities here would be Lack of Fusion and cracking along the sidewall. With conventional UT and PAUT, studies have shown over the years that the angles closest to perpendicularity with the bevel should be used for fusion zone interrogation, with an optimum range of about +/- 5° from perpendicular seen as ideal. In some cases, you may be able to stretch that to +/- 10° with a PAUT sectorial scan, but the angles closest to perpendicular will always be best. For the welds in these examples, the designed bevel angle is 37.5°, making 52.5° perpendicular.

When configuring HAZ coverage, the code should give you measurement criteria for the extents of the HAZ. Always refer to your procedures for guidance on this and other parameters. For this discussion, we will use the ASME worst-case for pressure vessels- HAZ= 1”, or “T”, whichever is less, for materials up to 8” thick. For the 1” Double V, we will use 1” as our HAZ, and for the 0.5” Single V, we will use 0.5”. In the HAZ, we are primarily concerned with OD and ID cracks. Studies have shown over the years that shearwave angles between about 35°- 58° are considered optimum for yielding corner trap echoes on notch-like reflectors, so those angles are often chosen for HAZ crack detection with PAUT. With TFM, there are no angles to choose, but using a wedge with a nominal refracted angle in that range is a good place to start.

Volumetric coverage is typically the last weld zone that is considered during the scan planning process. This is primarily where your rounded indications such as slag and porosity are found. Those discontinuities reflect sound omnidirectionally, similar to a side-drilled hole, so the exact angles of coverage aren’t quite as important as they are for the planar defects located in other weld zones. Obviously, you could also have cracks and other planar defects in the weld volume, but if you have coverage, detection should possible. If the other weld zones are also configured with proven coverage, the volume should be covered without any additional measures, just be sure to verify.

In the paragraphs below, we will first build a scan plan for the 0.500” Single - V weld, followed by the 1.00” Double- V weld.

Single- V Weld- PAUT

For PAUT, the root, fusion zone, ID Haz, and volume were all covered for PAUT with an index offset of 0.500”. For the root, this was accomplished using angles 64- 70°. Fusion zone coverage was accomplished using angles from about 43-62°. Volumetric coverage was accomplished using the entire 2^nd leg. The only zone not covered at that index offset was the OD Haz. In order to get that, a second index position of 1.00” was required. This coverage is depicted in the images below.

Single- V Weld- TFM

For the FMC/ TFM approach, The root and ID Haz coverage were obtained with the T-T Wave Set, which simulates a 1^st leg shear wave. After the skip on the ID surface, T-T becomes TT-TT in the second leg, and that wave set provided coverage for the fusion zone, OD Haz, and weld volume. Note that only one index Offset Position was required- 0.500”, as the large aperture of the 64- element probe provided a larger footprint of coverage, compared to the PAUT sectorial scan. That means we were able to achieve full coverage with only one scan per side of the weld, as opposed to two per side with PAUT! Shown in the image below, the yellow zone box outlines the region of interest (ROI).

Double-V Weld- PAUT

The Double- V weld is essentially two single-V welds, so you basically have twice the geometry to obtain coverage for. Also, due to the large 1.00” thickness, multiple indexes were needed. A 40- 70° compound S- Scan was used for PAUT coverage here. Starting with an index offset of 0.500”, the root was covered in the first leg with angles approximately 60- 70°. The lower fusion zone was covered in the first leg with angles 52- 63°. The upper fusion zone was covered with angles 42-50° in the second leg. The ID Haz was covered in the first leg with angles from 40- 52°. The lower volume was covered in the first leg, and the upper volume was covered in the second leg. The only area missing was the OD Haz. A second index offset of 1.400” was used to provide OD Haz, at the end of the second leg, with angles from 40- 49°. The second index offset position also optimized the upper fusion zone coverage, using angles 49- 57°. The images below show those two scan positions and their respective coverage.

Double- V Weld- TFM

 TFM coverage for the Double-V weld is similar to that of PAUT. Two index offsets were required to obtain full coverage- 0.500” and 1.700”. The root, lower fusion zone, ID HAZ, and lower volume were all covered on the first index with the T-T Wave Set, as seen in the first image below. As seen in the second image, the near side coverage with TT-TT (second leg) was not optimum, it was mostly limited to the far side of the weld, which can be a very unpredictable region, especially after reflecting from the bottom cap. And just because the yellow zone box covers the desired area of interest, that does not mean that it was covered well. A second index position was needed to optimize TT-TT or second leg coverage. This is shown in the third image. The yellow zone box outlines the region of interest (ROI) in each example.