EVALUATION OF PERFORMANCE OF INTEGRATING SPHERES FOR INDIRECT EMITTANCE MEASUREMENT

摘要

Integrating sphere reflectometers are used in the majority of instruments used for spectral reflectance measurements of non-specular (scattering) samples. Knowledge of directional-hemispherical spectral reflectance (and transmittance, in the case of semi-transparent samples) allows one to derive the spectral directional emittance, which is of considerable practical interest for numerous heat transfer and pyrometric applications. The indirect method of emittance measurement is commonly used in cases of low emitted flux such as for samples near or below room temperature. For direct emittance measurement facilities, the indirect method has also been used for accurate non-contact measurement of sample temperature. Along with a number of advantages for reflectance and emittance measurement, integrating spheres possess a considerable number of potential sources of error relating to the characteristics of the sphere and their deviation from ideal behavior. This paper describes an evaluation of the effects of sphere related error sources on a sphere reflectometer's performance by means of a computational model. This effort was undertaken in the framework of a current NIST project to build a facility employing indirect measurements of spectral directional emittance. A description of the modeling program and its assumptions are given. Examples of modeling results for several materials and sphere parameters (such as the wall reflectance and degree of wall coating specularity) and their effects on emittance values are presented and discussed. A discussion of the developed techniques and conclusions as well as the prospects of their practical application to design optimization and uncertainty evaluation of integrating sphere reflectometers and emissometers is presented.