Separation capture from curved surfaces using various turbulence models

The accurate modelling of flow separation from a curved surface using Computational Fluid Dynamics (CFD) is computationally expensive. In the work presented here, we use a curved backward facing step and report separation length predictions using varying curvatures and four different turbulence models. The Spalart-Allmaras model results showed the closest agreement with experiment on the separation length, which is shown to vary proportionally with the amount of curvature of the step.