High frequency eddy current inspection of fibre placement in Carbon Fibre Reinforced Polymer Composites

This paper presents experimental and simulation studies of high frequency eddy current non-destructive testing (NDT) methods for inspection of carbon fibre reinforced polymer (CFRP) composites. It is well known that CFRP composite materials are widely used in industry as structural components, and products based on such materials are increasingly developed and introduced to market. In recent years, non-conventional eddy current method operating at high frequencies for NDT of composites begins to attract attention. It provides a non-contact, couplant-free option of inspection when compared with the other NDT methods currently used in industry. As eddy current relies on fibre fabrics in individual plies to form the current pathways, the eddy current paths in composite are unsurprisingly complex due to the structural features of composite laminates when compared with that in homogenous conducting metals for which conventional eddy current NDT is often used in industry. It will be noted that the knowledge is still limited so far about the eddy current distributions in composite laminates due to lack of effective theoretical and numerical models. This paper will focus on the development of a new numerical model that is suitable for use in simulations of high frequency eddy current inspection of fibre placements. An effective model is desirable for understanding the performance dependence of the eddy current testing method on the material properties, composite structural features, and other parameters involved. In this model, the fibre tow layout of individual plies is taken into consideration. The collection of the pickup signals by the sensor probe shows clear images of the fibre placement in different layers. The simulation results are compared with experimental results of fibre placement imaging. The model is useful for understanding the physical process for inspection of composites using high frequency eddy current methods, and for studying the influence of the controllable parameters involved.