Cracks in dents or linear anomalies interacting with dents are a major pipeline threat. These combined anomalies represent challenges to the Mechanical Engineers that design ILI tools as they need to keep the sensor in an optimal position towards the inner pipe wall. Ultrasonic Crack (UC) tools consist in a sensor plate with a fixed incidence angle that depends on the coupling medium. This plate is then attached to the skids; these are in constant contact with the internal pipe wall. When the tool interacts with a dent, the incidence angle is not optimal; therefore, detection of any interacting feature is compromised.
By not having the optimum angles in the pipe wall, the amplitudes from the reflections caused by cracks will be attenuated. Depending on the magnitude of the attenuation, these might be below analysis thresholds meaning that an algorithm and/or analyst will not consider them as relevant signals. Up to this point, detection of interacting features sounds like a “guess “ or “luck”. So, how can we use UC inspection to detect the interacting features? How can operators manage their assets knowing that they have dents but there is an uncertainty if there are interacting features?
To answer these questions, a systematic approach had to be used. It consisted of multiple phases where 1.- The mechanical design of the tool was understood, 2.- Simulation campaigns to understand the ultrasonic pulse while interacting with the dent, 3.- Pump tests with artificial features, and 4.- Pump test with real features.
All of the data gathered through the different phases allowed the authors to understand the attributes from the features and conditions that influence detection and identification of cracks in dents. This derived in a performance specification stating the truth capabilities to detect interacting features in a dent. These learnings were applied to commercial inspections where the feedback loop is closed with the field verifications.