Design and optimization of a novel tesla valve air-cooled heat sink for improved thermal management

Abstract

Enhancing the thermal performance of air-cooled heat sinks while minimizing pressure drop remains a critical challenge in thermal management system design. This study proposes a novel air-cooled heat sink configuration based on the Tesla valve concept to address this issue. A three-dimensional conjugate heat transfer model was developed to investigate forced convection heat transfer across a Reynolds number range of 500 to 3000, employing a validated k-ω turbulence model to ensure accuracy near the solid–fluid interface. The thermal and hydraulic performance of the Tesla valve heat sink was compared with conventional configurations, including plate, pin, and plate-pin fin heat sinks. Results indicate that the Tesla valve design offers superior heat transfer characteristics, achieving the highest Nusselt numbers and the lowest mean airflow temperatures across the examined Reynolds number range. Additionally, its friction factor remained competitive relative to traditional designs. To further optimize performance, an integrated artificial neural network–genetic algorithm (ANN-GA) framework was developed, achieving a high predictive accuracy (R² = 0.98). The optimized configuration, validated by CFD simulations, demonstrated an error of less than 2.56 % and yielded thermal enhancement factors 10–17 % greater than those of the conventional plate fin heat sink for Reynolds numbers between 750 and 3000. These findings highlight the potential of Tesla valve-based geometries as high-performance solutions for next-generation air-cooled heat sinks.