A Customer is running NX11.0.X, using Advanced Simulation and doing Thermal/Flow. He is trying to run the Onera 6 wing analysis to validate the use of NX Flow for aerodynamic analysis. The pressure map seems fine until the point where the boundary layer is supposed to separate from the wing.
He has tried every every combination of solution type (i.e., laminar, all of the turbulence models, with and without wall function, etc). He has also tried various meshes, including a boundary layer mesh. He needs some guidance on how to set up the model to get the boundary layer separation.
Solution
Flow over the Onera M6 wing is transonic and compressible. The wing flow experiences supersonic conditions, a shock and boundary layer separation. This class of compressible Open-Air Flow is difficult to simulate especially in terms of convergence by using a pressure-based approach. A density-based formulation has more advantage over the pressure-based solver for these classes of flows.
Our CFD code uses a pressure-based scheme. This approach was developed for low-speed incompressible flows and extended to solve and operate for a wide range of flow conditions beyond their traditional or original intent. Currently our pressure-based solver is used for incompressible and mildly compressible flows and it can be used to capture shocks. As soon as strong shocks appear, it becomes more difficult to use it, especially in terms of convergence.
For external Aerodynamic applications, the simulation domain is not defined as part of the model. The fluid domain height should be really large - at least 20 chords around the airfoil. The region around the airfoil needs to be meshed with a very fine mesh and a higher refined mesh must be applied near the stagnation point to capture the rapid pressure change. Walls with slip condition on the far-field boundary should be specified.
Notes
Reference: IR 9023697