Detection and characterization of defects in composite materials using thermography

TL;DR: In this paper, a numeric model was developed to predict the result of an infrared thermal test in carbon fiber reinforced polymers with structural defects, and the model was validated with experimental tests, consisting of samples made of poly-methyl-meth-acrylate.

Abstract: Composite materials are widely used in the transport industry, particularly aeronautics, due to their light weight and enhanced mechanical properties. On the downside, these new materials are less known and their behaviour are unpredictable greatly because of their anisotropic properties, being difficult to simulate and understand their behaviour. To overcome these aspects regular preventive maintenance operations are necessary. In order to avoid major problems, Non-Destructive Testing (NDT) are a solution to assure the safety without destroying the components. From the vast possibilities available, field image NDT have proved to be fast, precise and easy to interpret. Image technologies usually capture and measure some type of electromagnetic radiation to identify a discontinuity in the radiation pattern that changes due to a component abnormality. Since the natural radiation emitted by objects at the ambient temperature is in the infrared waveband, this is used to create a temperature image, particularly in the middle and far-infrared wavebands. Better results can be obtained when is used a stimulation source to induce a temperature variation, highlighting any defect that a component may have. Two different types of tests can be done with: single or periodic stimulation. When using single stimulation (transient tests), a pulse of energy is applied to the object and observed its recovery to the previous equilibrium state. On the other hand, when performing periodic tests (lock-in), the stimulation is modulated in several cycles, with the amplitude response and phase delay being analysed. This work aimed to develop a numeric model able to predict the result of an infrared thermal test in carbon fiber reinforced polymers with structural defects. This work is divided in three main parts: field tests, numeric simulation and model validation. Several different field tests were performed with various test samples to identify the settings that produced the best results and give the highest sensitivity to defects detection. These tests were performed using the above refereed techniques, transient and lock-in. During the lock-in tests, it was identified an imperfection in the modulated sinusoidal stimulation, which was quantified and corrected improving the overall results. The samples used in the laboratory testes where simulated using the finite element technique in Matlab R ©. This manner it was obtained an internal view of the heat flow inside the component and provided the temperature evolution at the surface of the specimen during an entire test. With the results obtained from the simulations, it is possible to estimate the thermal response obtained with a certain type of test and waveform. Finally, the model was validated with experimental tests, consisting of samples made of carbon fiber reinforced polymers poly-methyl-meth-acrylate. At the end of this document it is presented a short comparison between the two thermal techniques and shearography.