Detection of dis-bond between honeycomb and composite facesheet of an Inner Fixed Structure bond panel of a jet engine nacelle using infrared thermographic techniques

Abstract: Carbon fiber reinforced polymer (CFRP) composites are preferred for their specific strength and toughness. As fibers are the main load-bearing constituent of composites, fiber breakage has a significant effect on their strength and stiffness. The complex nature of damage involving oriented fiber breakages across multiple layers has posed a challenge to manufacturers and end-users alike. While detailed investigations of the damage have been carried out using micro-CT scans, assessment of oriented fiber breakages with field-deployable non-destructive techniques would facilitate our understanding significantly. A new scheme deploying induction thermography to detect fiber breakage and identify its orientation is proposed. It is experimentally demonstrated on samples with realistic fiber breakage produced in a controlled manner. Further, a numerical model capturing the proposed inspection mechanism is described.

Abstract: Induction thermography is a non‐contacting, non‐destructive evaluation method with a wide range of applications. A deeper understanding of the detectability of cracks requires fundamental knowledge about the induced current density distribution in the component under test. A calculation of the current distribution provides information how much current is flowing at which location of the component, how a crack disturbs the current density, how much heat is produced at which location of the component, and how the heat diffuses to the surface. The heating process depends on the type of crack. On the one hand there are cracks which can be detected mainly by direct observation of the heating process due to an increased current density, and on the other hand there are cracks which can be detected mainly because of a modification of the heat diffusion. This paper presents an analytical model for the calculation of the current distribution, including the back‐flow current along with finite‐element calculations. Furthermore, two new crack models are presented for a better description of real cracks.

Abstract: A unified approach, considering three possible heating mechanisms: fiber (Joule losses) and fiber crossover junction (dielectric hysteresis and contact resistance), to identify dominant heating mechanisms during induction processing of conductive fiber reinforced composites is presented. Non-dimensional parameters were proposed to identify the relationships between heating mechanisms and process and material parameters. Parametric studies showed that junction heating mechanisms dominate fiber heating for the material systems considered, with the exception of relatively low contact resistance (< 10 3). Results for dielectric hysteresis and low contact resistance were consistent with individual models in the literature. A design map relating the three mechanisms is presented that can help identify the dominant heating mechanism, given the properties of the composite.

Abstract: For joining and repair of continuous fiber thermoplastic composites, induction heating has been viewed a strong candidate. Induction heating employs an applied alternating magnetic field, which induces a rotational emf in a grid of conductive carbon fibers, which are then used to carry resulting currents. In continuous carbon fiber crossply composites the available paths for “eddy current” loops are along the network of conductive carbon fibers. For this to occur, an electrical transfer must take place between crossing fibers in adjacent plies. Tests involving variable thicknesses of interply neat film layers have been performed to provide insight into the mechanisms taking place. These tests indicate that the primary mechanism for heating in such laminates is dielectric losses in the polymeric region between fibers in adjacent planes that form the conductive loop. Therefore, heating is not uniform in such composites despite a uniform magnetic flux. Heating patterns were viewed using liquid crystal materials and E-type thermocouples. Several factors leading to nonhomogeneous thermal distributions have been considered, including current density effects, internal emf cancellation, and rotational field effects. Global and local considerations are addressed, a localized model is proposed, and the corresponding theory is developed qualifying the early results. Additional testing has supported the theory.

Abstract: Dynamic thermography with inductive excitation is analysed as an alternative to magnetic particle inspection or to eddy current testing. Given by the relation of the electromagnetic skin depth, the thermal penetration depth and the crack dimensions to be detected, different regimes for defect detection are identified. The effect of the crack parameters length, depth and inclination angle are discussed. In ferritic steel, at induction frequencies of 100-200 kHz, perpendicular open cracks with a length of 7.5 mm were detectable when their depth was minimum 0.15 mm. For inclined cracks, the sensitivity is even higher. Experiments were performed on cold and warm forged steel components. The signal-to noise ratio obtained from defects was usually high, the critical limitation on technical surfaces is the background due to surface roughness and due to surface contamination. A series investigation on forged components showed a good probability of detection and a low false alarm rate compared to magnetic particle testing. The short testing times of a few 100 ms per object view will allow short cycle times for mass products.