Optimizing rectangular fins for natural convection cooling using CFD

• Published in: Thermal science and engineering progress 2020-06, Vol.17, p.100484, Article 100484
• Main Authors: Adhikari, R.C., Wood, D.H., Pahlevani, M.
• Format: Article
• Language: English
• Subjects: Dynamic-Q algorithm; Heat transfer and flow patterns; Natural convection cooling; Optimum fins; Rectangular fins
• ISSN: 2451-9049; 2451-9049
• DOI: 10.1016/j.tsep.2020.100484

•The heat transfer and flow features of optimum rectangular fins are characterized.•The flow from the channel ends strongly affects the heat transfer rate from the fins.•Multi-parametric CFD optimization using dynamic-Q algorithm has been implemented .•The fin optimization results are validated against the experimental data . Efficient heat transfer from finned-surfaces under natural convection is a critical consideration in the design of heat sinks for microelectronics and thermal devices. Previous experimental studies on natural convection from rectangular fins on horizontal surfaces have shown that maximum heat transfer per unit base area occurs within a narrow range of fin spacing in tall fins, and over a wider range of fin spacing in shallow fins. However, the heat transfer and flow characteristics of optimum fin configurations and the combined effects of fin geometrical parameters have not been sufficiently scrutinized in the literature. This work is primarily focused on characterizing the optimum heat transfer and flow patterns using three-dimensional, steady-state, laminar, conjugate heat transfer simulations, which serve as the fundamental basis for determining optimum fins. The main fin design parameters are the fin spacing, height, and length. Multi-parametric computational fluid dynamics (CFD) optimization was conducted to determine the combined effects of fin spacing, height, and length on heat transfer. To find a global optimum fin design, we performed a multi-parametric optimization using the dynamic-Q algorithm, rather than through a single parameter variation as is commonly done in previous studies. It was found that the fin spacing, height, and length strongly affect the flow through the fin channel ends, and hence the temperature gradient and heat transfer rate from the fin surfaces. The results demonstrate that the dynamic-Q optimization could efficiently find the global optimum CFD solutions for the design objectives: the maximum heat transfer per unit base area and the maximum heat transfer per unit base area and fin weight. The results are validated through the experimental results available for typical fin designs, as well as an experimental measurement of an example fin design in this work.