DRAFT Proceedings of IMECE: 2004 ASME INTERNATIONAL MECHANICAL ENGINEERING CONGRESS AND RD&D EXPO November 14-19, 2004. Anaheim, California. IMECE2004-60715 UNSTEADY FORCED CONVECTION IN PACKED BEDS (COMPUTATIONAL AND EXPERIMENTAL ANALYSIS) Ricardo Mejia/Universidad de Antioquia John R. Agudelo/Universidad de Antioquia Cesar Nieto/Universidad Nacional de Colombia Laura C. Villa/ Universidad Nacional de Colombia ABSTRACT In this work, a process of unsteady forced convection in a packed bed of spheres was experimentally and computationally analyzed. A device was designed and constructed in order to run the experiments in packed beds. It was used to carry out an experimental run in a packing of ten aluminum spheres, which tube-to-particle diameter ratio was 2,4. Methane-air combustion products were kept flowing into the packed bed at a constant inlet rate of 2.8 m/s and temperature of 369ºC. Packed spheres were heated from 25ºC to gases temperature. While heating, temperature of spheres, wall and gases at different positions were measure to follow unsteady process. On the other hand, computational simulation was carried out by modeling the ten-spheres packing under the same flow conditions of the experimental run. Physical properties of gases were kept constant and fluid flow profile was solved before heating process. Results of unsteady temperature variation in different positions showed good agreement with the experimental measures. This result allowed inferring that flow field calculations were a satisfactory representation of the actual flow field, since temperature field variation depends strongly upon flow field. In conclusion, it was shown that the Computational Fluid Dynamics (CFD) simulation is an accurate tool to analyze unsteady forced convection in packed beds. The device designed is a flexible and powerful tool to measure unsteady forced convection in packed beds. Features such as behavior of gas-to- solid heat transfer coefficient are fundamental questions to solve. CFD supported on experimental measures is the way to solve this and other several questions. INTRODUCTION Computational Fluid Dynamics (CFD) is a tool that allows studying gases flow inside thermal regenerators by numerical solution of continuity, momentum and energy equations. There is still a slight lack of comprehension of heat transfer and fluid flow features in previous models of packed beds, specifically those related to unsteady convection. CFD is an alternative method to determine thermodynamic variables behavior in packed bed thermal regenerators. For instance, it allows determining the effect of packed geometry and gases inflow conditions upon packed bed regenerators operation. A number of previous works such as [1]-[6] have shown CFD potential on studying heat transfer and fluid flow in packed beds. However, those works are limited to steady proceses. Besides, there are works as [7]-[16], that deal with non-thermal equilibrium models of heat transfer in packed beds. Nevertheless, those models do not deal with actual geometry but with a pseudohomogeneous geometry. In this work, an advance in the field of unsteady forced convection in packed beds is presented. This study applies experimental and computational models to assess thermal behavior of a regenerator under realistic operational conditions. NOMENCLATURE a,b,c Fitting constants p atm Atmospheric pressure [Pa] q Heat flux [W/m 2 ] T Temperature [K] t Time [s] T ∞ Reference temperature [K] Experimental Setup. Figure 1 shows a picture of the setup built by our group to measure unsteady forced convection in packed beds [17]. Its main duty is to control inlet flow conditions and measure temperature and pressure inside a packed bed thermal regenerator. This characteristic makes possible to know unsteady thermal behavior of the packed bed. Besides, the setup is able to work in a reasonably broad range of flow and temperature conditions (30–700 o C and 0–3m 3 /min) to explore unsteady forced and even natural convection in packed beds. Additionally, computational and theoretical results would be verified and supported by experimental results. Technical specifications of the setup are listed in Table 1. 1 Copyright © 2004 by ASME