Browse/search for people

Publication - Professor Robert Smith

    3D ultrasound characterization of woven composites


    Tayong, RB, Mienczakowski, MJ & Smith, RA, 2018, ‘3D ultrasound characterization of woven composites’. in: 44th Annual Review of Progress in Quantitative Nondestructive Evaluation. American Institute of Physics (AIP)


    Recent studies on the Non-Destructive Testing (NDT) of composites for the aerospace industry have led to an understanding of ultrasonic propagation in these materials [1]. Techniques for enhanced ultrasonic imaging of the internal structure of composite laminates containing unidirectional fibers have been proposed and tested in a laboratory environment. For the automotive industry, textile composites are often preferred and widely used. The reason for this is that these types of composites offer good mechanical performance, with resistance to delamination and reduced manufacturing costs. In this study, two models are developed and shown to be suitable to characterize the woven specimen. The first model is a 1D analytical model that makes simplified assumptions and the second is a 3D time-domain Finite Element (FE) model developed [2] for advanced understanding of the woven composite response to an ultrasonic excitation. For each of the proposed models, three parameters are defined and used to analyze the structure behavior. They are the instantaneous amplitude, instantaneous phase and instantaneous frequency. These parameters are employed to track the in-plane fiber orientation and the ply-interface location and for the sentencing of features. Three different specimens with the following weave type: 3D orthogonal, 2D plain and Multilayer stitching were considered and scanned (using a focused ultrasonic transducer) to validate the proposed models. As a preliminary study, the work only focuses on the Orthogonal weave specimen. The results obtained from experimental, analytical and FE modeling, B-scan and C-scan are compared, discussed and presented in terms of the above defined parameters.

    Full details in the University publications repository