Computational Framework for Computational Fluid Dynamics Analysis of Loop Thermosiphon Based Embedded Cooling Systems

Abstract

Loop thermosiphons (LTSs) are highly effective two-phase heat spreaders, enabling significant heat transport through passive liquid–vapor phase-change, particularly beneficial in electronics cooling. However, prior studies on LTS simulations often lack sufficient clarity regarding critical modeling assumptions—particularly the selection of mass relaxation coefficients in the Lee phase-change model—and omit detailed analysis of the interplay between key numerical and physical parameters. These gaps present challenges for thermal engineers and researchers in developing stable, reliable volume of fluid (VOF) based computational fluid dynamics (CFD) simulations. In this study, we address these issues by proposing a computational framework that systematically examines the impact of parameters such as numerical time-step, flow regime, and mass relaxation coefficient ratios on stability and convergence. By monitoring and controlling mass variation, we demonstrate stable simulations with time-steps ranging from 1 × 10−4 to 5 × 10−4 s, using both turbulent and laminar flow assumptions and density-ratio-based mass relaxation coefficients. Our findings also highlight that an explicit discretization scheme combined with Geo-Reconstruct for volume fraction calculations significantly enhances stability. This framework thus provides a clear, systematic approach to LTS VOF modeling, offering a practical “recipe” for ensuring numerical robustness and guiding thermal engineers in navigating complex simulation settings.