Ultrasonic weld inspection using phased array ultrasonic testing (PAUT) is used to detect and size weld defects across many different industries and is incorporated in American Welding Society (AWS) structural and bridge welding codes and American Society for Mechanical Engineers (ASME) Boiler and Pressure Vessel Code (BPVC). In one or two encoded scans, PAUT can generate 2-D and 3-D weld defect profiles. In this article, the basic principles of phased array weld defect sizing are presented on the Olympus MX-2 platform. The basic phased array A-scan, S-scan, and C-scan data are discussed along with how to use phased array measurement cursors to size weld flaws.
Comparing Conventional Ultrasonic and Phased Array Sizing Techniques
Weld defect sizing is performed differently between conventional ultrasonic testing (UT) and phased array ultrasonic testing (PAUT). In conventional ultrasonic testing, the -6 db sizing technique is used generally used. This is a manual process that requires the inspector to find the maximum reflection amplitude from the weld defect and then moving the transducer in the scan axis and index axis directions until the maximum reflection amplitude is decreased by 6 dB or 50%. The scan axis length provides the length of the weld defect. While moving the transducer in the index axis direction, the depth of the weld defect is recorded at the upper and lower 6 dB points to provide weld defect height or dimension in the thickness direction. These are the two most important dimensions for fracture critical analyses. Weld defect sizing using phased array technology is always performed digitally in the height direction using the S-scan. Along the weld axis, it may be performed manually or digitally depending on if the scan was encoded.
Figure 1: Conventional ultrasonic testing A-scan used for weld defect length and height sizing.
Phased Array Ultrasonic Testing Data
A PAUT sectoral scan, or S-scan, uses a fixed number of piezoelectric elements and time delays, or focal laws, to electronically steer through a range of angles. Each focal law generates an A-scan that is color coded across its range. The A-scans generated at each angle are used to construct the S-scan. Similarly, a C-scan is generated from the S-scans acquired as the ultrasonic transducer and wedge are scanned across the weld axis. Encoded phased array data is generally presented in A-scan, S-Scan, and C-scan displays as shown below in Figure 2. The A-scan is shown on the top left with the yellow vertical and purple horizontal axes measuring amplitude (%FSH) and depth (inches or millimeters), respectively. This particular PAUT A-scan additionally has time compensated gain (TCG) and three amplitude references superimposed onto it; Standard Sensitivity Level (SSL) at 50% FSH, Automatic Reject Level (ARL) at 89% FSH, and Disregard Level (DRL). This are specific to American Welding Society bridge welding and structural code and will be discussed in a separate article.
Figure 2: Phased array ultrasonic testing (PAUT) A-scan, S-scan, and C-scan displays.
The PAUT sectoral scan (S-scan) is shown on the top right. In this case the S-scan was configured to generate ultrasonic shear waves from 45 to 70 degrees in 1 degree increments. As a result, the phased array S-scan consists of 26 total A-scans. The yellow amplitude and purple depth axes are also present in the PAUT S-scan. However, a colored pixel is now assigned to the individual A-scan amplitudes based on the color map shown in the right of the S-scan. The phased array S-scan consists of an additional index axis which tracks the position of the weld defects relative to the weld center line or some other location reference.
Lastly, the PAUT C-scan is shown across the bottom of the display. The C-scan consists of many S-scans, or slices, that are acquired as the PAUT encoder is moved along the weld scan axis. While the phased array C-scan is populated pixel by pixel with color coded amplitude it does not provide a 2-D top view that is generated with traditional straight beam C-scans. The teal colored axis is the scan axis which is common between PAUT and conventional ultrasonic C-scans. The gray axis on the left, however, is the phased array focal law or angle. In conventional ultrasonic testing, the gray axis would track the index axis position. The phased array C-scan must be used in conjunction with the S-scan to establish index axis position. Assuming a PAUT encoder resolution of 0.040”, over a 15” distance, results in 375 total S-scans. Since each PAUT S-scan consists of 26 A-scans, a total of 9,750 A-scans were used to generate the phased array C-scan shown.
Figure 3: Phased array ultrasonic testing (PAUT) S-scan weld defect sizing.
Sizing the height of a weld defect is accomplished in the phased array S-scan. In this case, the blue cursor is the active focal law and is displayed in the PAUT A-scan. The maximum amplitude is roughly 30 %FSH. The focal law cursor is shifted downwards in the S-scan until the measured A-scan amplitude is approximately 15 %FSH. The green ultrasonic axis measurement cursor is placed at that depth. The same procedure is repeated on the PAUT S-scan in the upwards direction and the red ultrasonic measurement cursor is assigned to that depth. The weld defect height is the difference between the two cursor positions, roughly 0.50” in this case.
Figure 4: Phased array ultrasonic testing (PAUT) C-scan weld defect sizing.
Sizing the length of the of a weld defect is easily accomplished in the phased array C-scan. In this case, the blue cursor is the active PAUT S-scan from which the weld defect depth, height, and index axis position are measured. Using the S-scan blue data cursors, the maximum amplitude of the weld defect indication is determined. The 6 dB technique is applied to the left and the red C-scan cursor is established. The same process is repeated to the right and the green C-scan cursor is assigned to that position on the PAUT scan axis. The total length of the defect in approximately 1.75”. The two indications are considered a single defect because they are separated by less than twice the length.
Comments