Simulation strategies for the Food and Drug Administration nozzle using Nek5000

Abstract

Computational fluid dynamics (CFD) is currently a versatile tool used for flow characterization in diverse areas of industry and research; however, its application in medical devices is less developed due to high regulatory standards for safety purposes. In this context, the development of a rigorous and standardized CFD methodology is essential in order to improve the accuracy and ensure the reliability of biomedical applications. To that end, the Food and Drug Administration (FDA) proposed a benchmark model of an idealized medical device to provide a common ground for verification and validation processes. Previous studies have evaluated the potential of conventional turbulence models to predict the relevant flow features in the FDA nozzle but have also been deemed inaccurate or exhibited high dependency on the numerical scheme. Furthermore, validation of computational results relied on previous experiments performed with particle image velocimetry (PIV), which also exhibited noticeable uncertainties. Here, we perform direct numerical simulations (DNSs) of the flow through the FDA nozzle configuration, at Reynolds numbers based on the throat diameter Ret = 500, 2000, 3500, and 5000, using the spectral-element code Nek5000. The predictive capabilities of the synthetic-eddy method and parabolic-inflow conditions at the inlet were tested, and the results were compared with PIV data. Our results highlight the very high sensitivity of this flow case to the inflow conditions and the disturbances at the throat, particularly when predicting the laminar–turbulent jet breakdown. Due to this extreme sensitivity, any benchmark data of this geometry need to include a very detailed characterization of both the conditions at the inflow and the throat, in order to enable relevant comparisons.