A critical assessment of acousto-ultrasonics as a method of nondestructive examination for carbon-fibre-reinforced thermoplastic laminates

Abstract

The essential features of the acousto-ultrasonic (AU) technique are described, including the parameter known as the ‘stress-wave factor’ (SWF). AU measurements are comparative, so that a high degree of repeatability is required to provide adequate sensitivity for the detection of defects. Part I of the paper describes a systematic investigation into the effects of the principal experimental parameters upon the variability and sensitivity of SWF readings. Parameters studied are: transducer couplant and coupling pressure, inter-transducer spacing, injected pulse energy, amplifier gain, threshold and gate width. This work leads to the definition of optimum conditions for discriminating between the fibre and transverse directions of a unidirectional carbon-fibre/PEEK laminate. In Part II of the paper, these optimized instrument settings, and other, are applied to the detection of deliberately introduced defects in laminates of the same material. Defects had been introduced in moulding, also by impact, by crushing and by drilling through-thickness holes. Because of the scatter of replicate SWF readings, statistical methods are applied to establish the level of confidence with which various defects may be detected in comparisons with sound laminates. At the 0.1% significance level, the detection limit for holes was 4 mm in a 20-ply laminate approximately 300 mm square. Moulded-in defects 30 mm square, including missing areas of reinforcement and aluminium foil inclusions, could not be distinguished above the scatter. Impact damage, which has been shown in other work to be of the order of 15 mm diameter, was similarly undetectable.