A self-method for resolving the problem of apparent LWIR emissivity for quantitative thermography up to 130 °C

In a previous work, we succeeded in connecting normal LWIR apparent emissivity to the spectral one of an aluminum nitride ceramic plate. The key problem was the knowledge of the effective spectral bandwidth in use in the system. Hence we have developed an analyzer which permits to identify the spectral bandwidth of IR system using only its raw data. It proceeds by minimizing the dispersion from linearity of the characteristic thermosignals/integrated radiance over a temperature range of the IR system. The capacities of the analyzer are tested for five commercial cameras. Each of these systems exhibits a similar formatting process implemented during the thermogram recording. The effective spectral bandwidth shows plausible values. It varies significantly from one model to the other and the residual non-linearity is connected to the NETD of the IR system. The robustness of the apparent emissivity measurements is also tested with the aid of emissivity reference of 0.5. The overall accuracy of the method is less than 1%, depending on the specular or diffuse part of the reflected irradiation. Applied in field situation, the method is suitable to detect absolute variation of emissivity of less than 6 ⋅ 10−3. We use the analyzer to determine the spectral bandwidth of a commercial 320 × 240 microbolometer uncooled IRFPA camera which had already served to characterize the normal LWIR apparent emissivity of the aluminum nitride ceramic plate. By using the spectral response of the two major microbolometer sensor technologies, the general formulation of apparent emissivity matches our apparent emissivity measurements. An agreement better than 0.6% in absolute value and a less than 6 ⋅ 10−3%/°C dispersion are found over the entire temperature range [40–130 °C].