Edge effects on the pressure drop through triply periodic minimal surface heat exchangers

Abstract

Triply periodic minimal surface (TPMS) geometries offer significant potential for compact, high-performance heat exchangers, but confinement-induced edge effects in small-scale test articles can distort the measured pressure drop and its extrapolation to full-scale performance. To the authors' knowledge, this is the first combined experimental and CFD study to vary the number of complete transverse unit cells, from 1 × 1 to 6 × 6, inside a square enclosure while the Diamond TPMS unit cell topology, its orientation and phase, the nominal wall thickness, and the streamwise length are held fixed. Pressure drop was measured with air ( Re = 220–7290) and simulated with CFD ( Re up to 8000); mean absolute differences are 1.4 to 5.6% for the 1 × 1 through 4 × 4 and 8.8% for the 6 × 6 and 13.7% for the 5 × 5. For Re = 3000–8000, the ratio of each pressure gradient to an extrapolated bulk limit collapses onto one exponential decay with cell count, with a mean absolute error of 0.040. The fitted confinement penalty relative to the extrapolated bulk limit is 150% at 1 × 1 and 12% at 6 × 6. CFD-based flow-field visualization illustrates the mechanism: corner recirculation behind rearward-facing lattice peaks and high-velocity channeling through diagonally opposing forward-facing corners, whose share of the cross-section shrinks as the cell count grows. The fitted relation predicts a confinement penalty below 5% at eight or more cells across. This threshold lies beyond the tested range. The relation is specific to the studied Diamond geometry, but the mechanism is a transferable framework for interpreting sub-scale TPMS test data.