Isolating the Effect of Geometric Porosity on Supersonic Mars Parachute Drag

A Controlled CFD Investigation Using Historical Mission Porosities

Abstract

Introduction: Mars parachutes must operate at supersonic speeds, and past missions have used different canopy designs and porosities. Their aerodynamic results were obtained under different test conditions and numerical methods, so the effect of porosity alone cannot be determined from reported drag coefficients. This study aims to determine how geometric porosity affects the drag coefficient and flow field of a supersonic disk-gap-band (DGB) parachute, using computational fluid dynamics (CFD) with all other conditions held constant.

Methods: A three-dimensional, 48-rib DGB canopy based on the Viking 1 profile was built in Autodesk Fusion 360. Only the annular gap was changed, to give porosities of 9.3%, 10%, 12.5%, 15% and 22.4%, corresponding to previous Mars missions. Steady-state compressible Reynolds-averaged Navier–Stokes simulations with the k-ω SST turbulence model were run in Autodesk CFD at 450 m/s (approximately Mach 2) and 230 K, with a gas constant appropriate for a CO₂-dominated atmosphere. The mesh procedure, domain and number of iterations (200) were the same for every configuration. Drag coefficient was regressed against porosity (n = 5).

Results: The uncompensated drag coefficient decreased from 5.935 at 9.3% porosity to 5.518 at 22.4%, a reduction of about 7.03%. The largest drop occurred between 12.5% and 15%, and the change between 10% and 12.5% was comparatively small. The negative correlation was statistically significant (r = −0.890, p ≈ 0.043). Contours showed high pressure ahead of the canopy, and in all configurations a low-velocity separated wake with reverse flow and a total-pressure deficit. The extent and structure of these features varied with porosity.

Discussion: Porosity measurably influences drag and wake structure, and its effect appears to extend beyond the reduction in projected solid area. Comparing historical parachutes on porosity alone may therefore be misleading. The results differ from earlier NASA experiments, which found little porosity effect. This may be because those models were deployed behind a forebody and moved during testing, whereas the present canopy was isolated and fixed. The study has no mesh-independence analysis, uses steady RANS, and depends on the reference-area definition, so the results are not quantitatively validated. Future work should establish mesh independence and turbulence-model sensitivity, examine domain blockage, reconstruct Pathfinder, Viking and Huygens geometries, and compare against experiments on a matching isolated canopy.