Steady-state and transient microscale temperature measurements by multispectral method and photons counting. (25th April 2016)

• Record Type: Journal Article
• Authors: Pierre, Thomas; Rodiet, Christophe; Rémy, Benjamin; Degiovanni, Alain
• Is Part Of: Applied thermal engineering. Volume 99(2016:Apr.)
• Page Start: 343
• Page End: 351
• Publication Date: 2016-04-25
• Subjects: Steady-state and transient microscale temperature measurement -- Multispectral -- UV-visible wavelength -- Photon counting -- Least-squares; Heat engineering -- Periodicals; Heating -- Equipment and supplies -- Periodicals; Periodicals; 621.40205
• DOI: 10.1016/j.applthermaleng.2015.12.134
• Languages: English
• ISSNs: 1359-4311

Abstract

Highlights: A three wavelengths pyrometer able to focus on a micrometric surface collects the photons emitted by this surface. The wavelengths are selected according to several criteria and a global function takes into account the spectral parameters. The photonic emission is a random phenomenon and the Gaussian law estimates an average photonic flux for each wavelength. The temperature is estimated by a least-squares method from a multispectral pyrometer. Abstract: This work presents steady-state and transient microscale temperature measurements by optical and multispectral method in the ultraviolet-visible wavelengths. Regarding the classic laws of radiative heat transfers, the photon counting is preferred to the energy measurement. The photonic emission is a random phenomenon and is here measured with statistical laws such as the Gaussian law. The spectral dependence of the optical apparatus transfer function is taken into account by a linear function and the temperature is estimated by inversion with a Levenberg–Marquardt algorithm. Measurements are performed with a small-dimension blackbody developed in the laboratory and with a Chromel wire. Both of them are capable of heating up above 1000 °C. The diameters of the observed surfaces are 10 µm and 5 µm, respectively. The theoretical criterion of wavelength selection and least-squares parameter estimation allow for a difference between the estimated and controlled temperatures lower than 4%.