I am trying to create a mesh around this body, I have had several attempts at reducing the region refinement and the coarseness of the mesh, however it still won’t run due to the machine running out of memory or stating that the mesh refinements are too small. Please could anyone offer any support?
My first recommendation is to use symmetry boundary condition by splitting the flow domain with respect to XZ plane, unless you will have a side slip flow. This will help you save significant amount of resources. Second recommendation, start with a mesh that is as coarse as possible, then start adding each refinement one by one. This way it will be easier to troubleshoot the error. You can even perform a mesh independency study.
Some other remarks: you may want to consider enlarging your far field geometry, and this will help flow to develope properly. Looking at the vehicle, please also be aware that Incompressible simulation type is supported for Mach numbers that are up to 0.3. In case you need a higher speed, you may want to switch to compressible simulation type. Compressible simulations should also give stable results with Mach numbers that are up to 0.6-0.7. (before the shocks are generated on wing surfaces.)
Hope these are helpful,
Thank you for this! I will try these and see if they work.
My recommendation is always the same for aerodynamic studies. Never use the automatic meshing system because it will generate the same grid for the whole domain; in practice, it wastes resources.
Following what @kaany says, my recommendation is the same as always:
- Use parametric hex-dominant mesh
- Start with a coarse background mesh and add refinement boxes as you approach the wing.
- Use layers to connect the mesh of the refinement box with the mesh near the wing. Start with a few refinement boxes and a very coarse background grid.
Indeed @jairogut, the Hex-Dominant Parametric mesh approach is ideal for such cases but it may be a bit harder to understand and use as it exposes a lot of different settings.
I would just like to add that a quick way to get started with it is following our dedicated tutorial (Hex-Dominant Parametric Meshing of a Front Wing).
I think it is a good starting point for anyone interested in this approach, especially for external aero cases.