A carbon fibre reinforced plastic (CFRP) adhesively bonded single lap joint sample is used for comparing the detection of different defect types using pulsed phase thermography (PPT). Firstly, a polytetrafluoroethene (PTFE) insert, of the type widely used to simulate defects in composite materials, was added to the bond line of the joint.Liquid layer kissing defects were simulated using silicon grease. PPT clearly identified the PTFE but not the silicon grease contamination. The PPT identified the silicon grease defect when the joint was loaded. It is postulated that kissing defects can be detected using thermography if a small load is applied to the joint, as loading opens the defect and produces a gap that provides sufficient thermal contrast for detection. Thermoelastic stress analysis (TSA) is used to validate the approach. On-site application is addressed both in terms of the load application and the use of low cost infrared (IR) detectors.

Vibration based loading has been successfully used to facilitate out of laboratory inspections using thermoelastic stress analysis enabling stress based non-destructive assessment of structures. An initial plate study verified the technique. A laboratory demonstrator of the onsite implementation was created to facilitate the development and assessment of a suitable loading device. The developed system was then taken on-site at a coal fired power station during a scheduled outage period. Vibration loaded thermoelastic stress analysis was successfully applied to welds in high pressure steam drain lines in-situ.

The maximum depth for which a defect of a given size can be detected in thermographic inspections is known as the 'probing depth'. For materials such as polymer composites, with low through-thickness thermal diffusivity, inspections are limited to thin materials or near surface inspection. With the aim of improving probing depth, the paper describes how established signal processing techniques are adapted for thermographic inspections. The procedures are implemented for both Pulse Thermography (PT) and Pulse Phase Thermography (PPT) inspections of laminated composite materials and sandwich structures. It is demonstrated that the adaptation significantly improves the probing depth, identifying defects that could not be identified using existing procedures. The applicability of the new approaches is discussed, paying particular attention to systematic and random errors resulting from equipment setup. Simple and efficient compensation methods to reduce the effect these errors are presented.

Quality assurance of adhesively bonded joints is of vital importance if their benefits are to be exploited across a wide range of industrial applications. A novel lightweight, low-cost, non-invasive embedded sacrificial sensor is proposed, capable of detecting damage within an adhesively bonded joint, which could also be used in a laminated composite structure. The sensor operation uses changes in electrical resistance, increasing as the sensing material area diminishes with damage progression. Initial tests prove the sensor concept by showing that the electrical resistance of the sensor increases proportionally with material removal, mimicking the sensor operation. Thermography is used to verify the current flow through the sensor and that any localised heating caused by the sensor is minimal. Short beam interlaminar shear strength (ILSS) tests show that embedding sensors in a composite laminates did not cause a reduction in material interfacial structural performance. Finally, the in situ performance of the sensor is demonstrated in quasi-static tensile tests to failure of adhesively bonded single lap joints (SLJs) with sensors embedded in the bond line. Prior to crack initiation, an electrical response occurs that correlates with increasing applied load, suggesting scope for secondary uses of the sensor for load monitoring and cycle counting. Crack initiation is accompanied by a rapid increase in electrical resistance, providing an indication of failure ahead of crack propagation and an opportunity for timely repair. As the crack damage propagated, the electrical response of the sensor increased proportionally. The effect of the sensor on the overall structural performance was assessed by comparing the failure load of joints with and without the embedded sensor with no measurable difference in ultimate strength. The research presented in the article serves as an important first step in developing a simple yet promising new technology for structural health monitoring with numerous potential applications.

No abstract