Effect of Experimental Variables on Flash Thermal Diffusivity Data Analysis

Abstract

Effect of Experimental Variables on Flash Thermal Diffusivity Data AnalysisJames N. Sweet Sandia National Laboratories P. 0. Box 5800 Albuquerque, NM 87185 ABSTRACTFlash thermal diffusivity data is usually analyzed with a thermal model which assumes axial heat conduction and uniform illumination of the flashed surface. For high accuracy data reduction, it becomes important to bound the errors caused by radial heat flow and by nonuniform laser beam profiles. These effects are examined analytically for a case in which the incident laser beam is confined to a radius smaller than the sample radius. The dependence of the output of an averaging detector on the magnitude of the radial heat transfer coefficient is presented and the linear dependence of radial and axial loss sensitivity coefficients is discussed. From this discussion, we conclude that inclusion of radial loss effects in analysis of the thermal response of multilayer structures is not important unless the radial loss factor is very large. Analytical results are presented for the temperature vs. time response of a two layer composite sample with interfacial thermal resistance and high thermal losses at the sample faces. The use of these results to reduce data for two multilayer samples is presented to show the utility of new data reduction techniques.

The Apparatus for Thermal Diffusivity Measurement by the Laser Pulse Method

“…Finite-pulse time effects can usually be corrected by using the equation (16) For this to be valid, the energy pulse must be represented by a triangle of duration T and time to maximum intensity of {3T as shown in Fig. 10.…”

“…For this shape, numerical factors are p = 0.50, Kl = 0.27057, and K 2 = 1.9496; and the duration t p = 1 ms. Equation (16) is incorporated into the program DIFUS for the finite time effect correction. Procedures for correcting for finite pulse time effect are also elaborated in the literature.…”