Define, Optimize and Validate Detection and Sizing Capabilities of Phased-Array Ultrasonics to Inspect Electrofusion Joints in Polyethylene Pipes

General project net improvements and benefits are an emerging ultrasonic testing (UT) phased-array (PA) techniques, procedures, reference libraries, and guidance for inspection of polyethylene (PE) joints were developed; an advanced detection, imaging, and sizing capabilities of the UT PA techniques were validated; and a good agreement between UT PA simulations predictions, UT PA experimental results, and destructive testing was achieved.

Net improvements specific to electrofusion (EF) joints:

• The PA UT procedures developed during this project were able to detect planar flaws in the fusion zone (represented by implanted aluminum disks) as small as 1 mm.
• Due to electrode connectors, identification tags, and general coupler geometry approximately 15 to 20% of the EF weld could not be adequately inspected. Actual inspection coverage will change depending on manufacturer. Inspection coverage can be improved by removing scanning obstructions such as electrode connectors.
• Probability of detection (POD) a90/95 = 6 mm was achieved by a PA UT operator having several years of PA experience and at least 3 months experience inspecting PE joints. Statistically, this means that 90% of flaws of this type and size (6-mm diameter), would be detected in 95% of the inspections. During this project each operator who participated in the blind trial study for EF joints was given 80 to 90 flaws in order to assure the detection and sizing data was statistically valid.
• For flaw sizing purposes, a 95% limit against under sizing (LUS) value was calculated from all the flaw sizing data obtained. A 95% LUS value is the value that needs to be added to flaw measurements for this type of inspection to ensure that at least 95% of the flaw measurements are conservatively oversized. For the planar fusion flaws used during this project, the 95% LUS for the best PA UT operator was equal to 8 mm.
• It may be possible to identify EF cold joints by measuring the heat-affected zone (HAZ) to determine if the heat input was sufficient. Results from this project showed very good agreement between the PA UT measured HAZ and the heat input. Further work needs to be done to determine the relationship between heat input and joint strength.
• Several PA instruments, probes, and scanners are available on the market that can be used by inspection companies with trained operators to inspect EF joints.

Net improvements specific to butt fusion (BF) joints:

• PA pulse-echo, PA tandem, PA pitch-catch, and PA time-of-flight diffraction (TOFD) techniques were evaluated for inspection of BF joints.
• Although the PA pitch-catch technique produced a clean, low-noise display, the detection capabilities were not as good as the other PA techniques.
• PA TOFD did a good job detecting planar flaws from approximately mid-wall thickness to the inside surface.
• PA pulse-echo and PA tandem produced similar detection results being able to detect flaws having a size 1 to 2 mm in through wall height.
• The BF ultrasonic PA procedure was unable to detect the presence of weak joints in samples that were welded using non-standard parameters. Only limited destructive verification was performed on these samples since this was outside the project scope.
• PA pulse-echo was the only technique that was able to provide some measurement of through wall flaw height.
• Using the BF inspection procedure developed during the project, flaws near the inner surface were detected better than those near the outer surface. Small flaws having a through wall dimension of 1 to 2 mm in height were difficult to detect if they were located in the outer 25% of the wall thickness. This was primarily caused by the difficulty in obtaining high beam steering angles because of the material velocity. Inspection coverage could be improved by designing ultrasonic angle beam wedges specifically for PE materials.
• During the field trials on BF joints, the PA techniques performed nearly identical to what was observed in the lab trials and no flaws were detected during the three field trials.
• The scanner unit developed during the lab trials performed sufficiently in the field trials; however, it was noted that modifications should be made to the scanner to reduce problems experienced with pipe curvature. Curvature in the pipe was the result of spooling of the pipe during manufacture. The primary modifications involve making the scanner shorter and centering the probe within the scanner frame.

Conclusions specific to saddle fusion (SF) joints:

• Because of the relatively complex shape of the saddle fitting and joint, the saddle joints required two ultrasonic techniques for the fitting used for this project. The first scan was a scan around the rim of the saddle, and the second technique was a scan on the sides of the saddle fitting. Using the two techniques, flaws having a through wall height of 1 to 2 mm were detectable.
• While 100% coverage of the fusion zone could not be obtained, these two techniques provided approximately 90% coverage and are capable of detecting flaws that would cause a leak path.
• Due to the saddle joint geometry, specialized scanner frames would need to be acquired or designed to perform encoded scans on SF joints.