The effect of variation in phased array element performance for Non-Destructive Evaluation (NDE).

This paper reports the results of an investigation into the effects of phased array element performance on ultrasonic beam integrity. This investigation has been performed using an array beam model based on Huygens' principle to independently investigate the effects of element sensitivity and phase, and non-functioning elements via Monte Carlo simulation. The purpose of this work is to allow a new method of array calibration for Non-Destructive Evaluation (NDE) to be adopted that focuses on probe integrity rather than beam integrity. This approach is better suited to component inspections that utilise Full Matrix Capture (FMC) to record data as the calibration routine is uncoupled from the beams that the array is required to produce. For this approach to be adopted specifications must be placed on element performance that guarantee beam quality without carrying out any beam forming. The principal result of this investigation is that the dominant outcome following variations in array element performance is the introduction of beam artefacts such as main beam broadening, raising of the noise floor of the ultrasonic field, and the enlargement or creation of side lobes. Specifications for practical allowable limits of element sensitivity, element phase, and the number of non-functioning elements have been suggested based on a minimum amplitude difference between beam artefacts and the main beam peak of 8 dB. Simulation at a number of centre frequencies has led to a recommendation that the product of transducer bandwidth and maximum phase error should be kept below 0.051 and 0.035 for focused and plane beams respectively. Element sensitivity should be within 50% of mean value of the aperture, and no more than 9% of the elements should be non-functioning.