Rapid ultrasonic array assessment framework method for evaluating ultrasonic array transducer performance

Abstract Ultrasonic array modelling is widely used to understand the performance of new transducer materials or geometries in a range of application scenarios. However, array modelling from first principles is both time-consuming and challenging due to the required knowledge of accurate acoustic and piezoelectric properties. This paper presents an alternative approach in which two elements are manufactured and characterised in the desired application scenario through a set of experimentally measured parameters. This experimental information is then used to construct a frequency domain model of an array of these elements. In this approach, the more challenging transducer modelling aspects are replaced by an experiment, and wave propagation modelling is used to simulate how those elements would perform in an array assembly. In order to characterise the elements and the application scenario the input pulse, attenuation coefficient, and both coherent and random noise amplitudes are measured. These experimental parameters are then used in a model to simulate the expected array image, from which the array performance, such as the signal-to-noise ratio, coherent noise scattering level, random noise level, and imaging artefacts, in the given application can be quantified. The approach is demonstrated with two commercial ultrasonic arrays and three test structure materials with differing levels of coherent grain scattered noise. Results across these various conditions show agreement within 4.2 dB between the simulations and their experimental counterparts in terms of image signal-to-noise ratio. Simulations with defects are also conducted and compared with experimental data for different array element geometries.