Abstract

Large Eddy Simulation (LES) is a step beyond widely used Reynolds Averaged Navier Stokes formulation (RANS) with turbulence model. LES is unsteady, therefore, predicts evolution of turbulent eddies in time. The only physic that is not resolved is the evolution of so-called sub-grid eddies, these are turbulent eddies smaller then local grid resolution. These eddies are modeled mainly by increasing local dissipation via so-called sub-grid model. LES has widely used as research tool rather then design tool due to its computational intensity. There are two factors that make LES appealing as design tool in air-cooling thermal management. First, those flow problems that occur in this industry are considered to be low Reynolds number because of board size and air speed when compared to problems in aerospace or power-generation industries. This is a solid advantage since required mesh resolution, cost of LES, is directly proportional to Reynolds number. Second, with new high-speed (2-3 GHz) work stations in the market, the phrase, intensive computation, has been redefined. Other factors include employing non-conformal grid that allows local mesh refinement inside regions of interest and existence of faster solvers. This work is an attempt to introduce LES to electronics cooling industry. First portion of this paper is devoted to explore/invent building blocks for reasonable simulation. They include mesh requirement study, implementing periodic boundary conditions, finding ways to filter pre-heated flow entering periodic condition, and examine an effective way to induce flow instability to create turbulence. Second portion of this paper is devoted to employ LES for better understanding of flow/thermal physics that occur in board level analysis and heat sink modeling.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call