Abstract
The need for reliable approaches in numerical simulations stands out as a critical issue for the development and optimization of cardiovascular biomedical devices. This led the United States Food and Drug Administration to undertake a program of validation of computational fluid dynamics methods for transitional and turbulent flows. In the current investigation, large-eddy simulation is used to simulate the flow in the first benchmark medical device and results are confronted to the existing laboratory experiments. This idealized medical device has the particularity to feature transition to turbulence after a sudden expansion. The effects of numerical parameters and low-level inlet perturbations are investigated. Results indicate a considerable impact of numerical aspects on the prediction of the location of the transition to turbulence. The study also demonstrates that injecting small perturbations at the inflow greatly improves the streamwise velocity estimation in the transition region and substantially contributes to the robustness of the flow statistical data.