Dual-Fidelity Numerical Modeling Approach for Screen-Wick Heat Pipe Operability Assessment and Design Integration

Aiming at future design integration of heat pipes, this publication presents a dual modeling approach able to provide rather accurate HP temperature predictions at a low computational cost and in parallel to assess the operability of the device in terms of required wick capillary pressures. In essence, two models were developed. One stationary heat conduction model based on the imposition of a uniform vapor temperature and one transient conjugate-heat-transfer (CHT) model which provides detailed solutions for velocities, pressures, as well as temperatures. Both models share one empirical coefficient, which was calibrated against experimental data. After calibration, the solid and CHT models were compared. The CHT model showed independency of the chosen starting point, which allows using converged solid-conduction results as initial solutions for faster transient convergence. It also provided reasonable values for pressure losses on the wick, in comparison to previous experimental results. While both predicted wall temperatures and thermal resistances agreed well with the experiments, the heat conduction model underestimated wick pressure-drops, which made the necessity of combing both models evident. Calculations considering the effects of gravity were then performed. The CHT model predicted modified wick pressure-drops in the presence of a gravitational field, however with essentially no changes in temperatures and fluid velocities. Based on the results, key dimensionless parameters were calculated. They indicate the dominance of forced convection on the vapor core, which partially explains the thermal behavior in face of gravity, as well as corroborate the assumptions of wick local thermal equilibrium and laminar, incompressible vapor flow. Verifying such assumptions is a critical requirement for the model’s applicability. Under this condition, the combined application of both models proved to be cost-effective, able to provide reasonably accurate predictions of key parameters at moderate modeling and computational costs.