Speaker
Description
Thermochromic liquid crystals (TLC) provide high spatial resolution surface thermography by mapping temperature to color. Quantitative use, however, requires robust calibration with the support of IR thermography, and careful control of optical artifacts. However, direct IR-referenced calibration in the final configuration is often impracticable in water-flow experiments where the surface is viewed through transmissive layers (e.g., water columns and polymer windows). Such layers can significantly attenuate thermal radiation in typical MWIR/LWIR bands, preventing reliable measurement of the true surface temperature behind the interface. In this work, we present a calibration-transfer procedure that enables re-use of an air-based TLC calibration in experimental configurations where imaging occurs through transmissive layers (water and plexiglass), which alter the recorded RGB intensities and therefore the computed hue.
A TLC coating (Hallcrest SPN300R27C18W, nominal red start 27 °C, bandwidth 18 °C) was applied to a curved rod bundle surface. Calibration was performed in air by simultaneously recording RGB images using a DSLR camera under white LED illumination and reference temperature maps using an infrared camera (FLIR A655sc) while the surface temperature was swept by controlled electrical heating. Hue values were computed from the RGB intensities of images, and mapped to temperature via pixel-wise calibration curves to account for spatial non-uniformity induced by surface curvature, viewing angle and geometry.
In our facility, the experiments have been performed in the Aupinel closed-loop water facility, featuring a vertical aluminum test section with plexiglass (PMMA) optical access and a hexagonal channel containing electrically heated rods.
The surface is observed through approximately 2 mm of water and a 1 cm plexiglass window, motivating a calibration-transfer approach from air to the water–plexiglass optical path. We propose a practical correction strategy based on per-channel multiplicative factors (kR, kG, kB) that compensate the recorded RGB values to an “equivalent air” response prior to hue extraction and temperature conversion. The correction factors are derived from a spectral model of the optical chain, by integrating the product of the TLC spectral respons, camera spectral sensitivities, and the wavelength-dependent transmissions of water and PMMA, following the Beer–Lambert law. In addition, specular reflections from the curved rods are mitigated using a two-polarizer arrangement to suppress polarized reflected components, improving hue stability in regions prone to glare.
A preliminary verification was performed to isolate plexiglass influence: identical scenes were recorded with and without the 1 cm plexiglass window, with exposure and white balance locked to the white LED source. After hue-to-temperature conversion, the resulting temperature difference map showed small discrepancies, with a maximum deviation of approximately 0.2 °C over the analyzed region. The validation is being extended to combine water+plexiglass optical paths, and to operating conditions representative of water-flow experiments.
This methodology enables quantitative TLC thermography in water-flow facilities where direct IR-referenced calibration is not feasible through water and polymer windows, by compensating spectral transmission effects and suppressing specular reflections.
