Determination of the convective heat coefficients by the pulsed photothermal method

Abstract

We present some elements in order to improve the non-detructive procedure for the measurement of the local heat coefficients between a flat plate and an air flow. We use the pulsed photothermal radiometry method which consists of analysing the transient temperature on the front face of a wall after a sudden deposit of luminous energy by a lump. The infrared camera measures the temperature evolutions at different abscissa from the leading edge of the plate, in order to deduce the local heat coefficients for the same experiment. The results of three identifications, based on a one-dimensional model at any abscissa with constant heat transfer coefficients, are compared. 1. Introduction The evaluation of the heat transfer coefficients between a wall and a flow is necessary for the control and the dimensioning of the thermal systems used in the energy processes. The installation of the flux and temperature sensors on the surface of the wall generates, often, considerable disturbances of the heat transfer. This is why the use of a measurement technique without contact has undeniable advantages. For that, we use an pulsed photothermal method. Being able to be employed in-situ, this technique consists in depositing of energy on the front face of the wall, and recording the temporal evolution of the surface. This method was introduced [1 to 3] to determine the thermophysical characteristics of materials, in particular, the thermal diffusivity. It was then extended to evaluate the heat transfer coefficients on the wall subjected to a convective heat exchange [4 to 6] or in the ducts [7]. These authors used a one-dimensional conduction model in the plate and the partial temporal moments of the temperature to identify the heat transfer coefficient. For the measurement by a transient method the steady state exchange coefficient, it was supposed in above mentioned works that the heat transfer coefficient remains constant during the pulsed experiment. This assumption is debatable if the quantity of energy sent over the wall during the pulse affects considerably the boundary layer. In order to specify the limits of validity of this assumption and the effects of the pulse duration on the results of measurement, we present here a comparative study between the results of three identification methods. All these methods are based on the assumption that the heat transfer coefficient remains constant in time. Moreover, by the use of the infrared camera, we measure the spatial evolution of the convective heat transfer coefficient over the plate.