Effective thermal diffusivity measurement using through-transmission pulsed thermography: extending the current practice by incorporating multi-parameter optimisation

Abstract

Through-transmission pulsed thermography (PT) is an effective non-destructive testing (NDT) technique for assessing material thermal diffusivity. However, the current literature indicates that the technique has lagged behind the reflection mode in terms of technique development despite it offering better defect resolution and the detection of deeper subsurface defects. Existing thermal diffusivity measurement systems require costly setups, including temperature-controlled chambers, multiple calibrations, and strict sample size requirements. This study presents a simple and repeatable methodology for determining thermal diffusivity in a laboratory setting using the through-transmission approach by incorporating both finite element analysis (FEA) and laboratory experiments. A full-factorial design of experiments (DOE) was implemented to determine the optimum flash energy and sample thickness for a reliable estimation of thermal diffusivity. The thermal diffusivity is estimated using the already established Parker’s half-rise equation and the recently developed new least squares fitting (NLSF) algorithm. The latter not only estimates thermal diffusivity but also provides estimates for the input flash energy, reflection coefficient, and the time delay in data capture following the flash event. The results show that the NLSF is less susceptible to noise and offers more repeatable values for thermal diffusivity measurements compared to Parker, thereby establishing it as a more efficient and reliable technique.