Ultrasonic inspection of pipelines for detecting cracks and corrosion using intelligent inline inspection (ILI) tools is a well-established method. In order to keep the resulting costs low, the target is to reach a first run success rate of 100 percent which imposes high demands on the ultrasonic transducers.

Flaw detection and sizing rely on complex algorithms that, for a distinct position, take into account not only the reflected signal of one but also the adjacent probes. These algorithms are based on simulations. Usually the characteristics of an ideal probe is the base of any computation. Therefore, any deviation from this theoretical behavior will influence the quality of an inspection run.

While acoustic parameters such as sensitivity and center frequency are important, geometric factors like beam offset (typically <0.3 mm) and squint angle (typically <0.5°) are critical as well. In many other applications, these variations can be compensated by adjusting probe position and orientation, but this is not practical in ILI systems due to the large number of probes. When extending the simulation and including these deviations the robustness of algorithms can be investigated. In parallel to that, measurements should validate the quality of the probes used.

Since ISO 22232-2, the standard for characterizing and verifying ultrasonic probes, does not provide specific guidance for this process, a new procedure has been developed. This presentation will demonstrate the results of characterizing a set of probes using this approach, highlighting the practical implications of these variations.