I am trying to conduct a mesh independence study on this model. The first mesh worked, however changing the refinement boxes has now caused a much longer mesh time and says that the instance has become unhealthy and therefore restarted. Can anyone provide any support/advice on what to do?
It’s exciting to see you dig deep into your CFD analysis and conduct a mesh independence study! This is a great way to truly get down to the most accurate solution possible with flow modeling.
I took a wanted to point out a few things that will help you be more successful, both with your meshing approach, and your overall study:
Meshing Tips & Troubleshooting
When trying to generate multiple meshes, it is best to not overwrite your original meshes so that you can have a record of each mesh you’ve generated. You can copy a mesh’s settings so that you can progress from the first iteration and apply some modifications/refinements, keeping all else equal. Here’s a documentation page that goes further into the subject: Meshing in SimScale | Simulation Setup | SimScale
When applying mesh refinements, you need to ensure your selections for region refinements is valid. Looking at your project, both of your refinement zones are applied to the entire Flow Region as well as one of the Cartesian Box primitives. To ensure you apply the refinements only to a subdomain of the entire geometry, you need to ensure you don’t select the entire Flow Region for your refinements (the Assigned Volumes list should be empty, with only the appropriate primitives toggled on to be selected). Since you are applying the mesh refinements to the entire Flow Region, it is very difficult to generate a successful mesh with the computing resources available to Community users.
Best Practices to Follow for External Aerodynamics Simulation
With your current geometry, I’d say the flow region is actually far too small to get truly valid results for aerodynamic flow. There needs to be significant clearance from the front surface of the Delta Wing body to ensure the boundary conditions actually represent far-field conditions and do not affect the aerodynamic flow behaviors around the body itself.
Check out this tutorial for external aerodynamics on a plane wing; it’s a very similar setup, but uses the Compressible analysis type (Pro only feature) since the flow speed is much higher (more than double yours, and above 30% the speed of sound). Compressible Flow Around a Wing | Tutorial | SimScale
When creating your geometry, there is no need to address the angle of attack the geometry is flying at to angle it in reference to the overall flow domain. The Velocity Inlet boundary condition can be applied to 2 boundary surfaces of the Flow Region and the angle of attack can be used to calculate the component velocities. For 40 degrees AoA & 50 m/s flow speed, that makes the velocity components about equal to to<38.3, 32.14> m/s for the drag and lift direction components of velocity. You can see this applied in the same Wing analysis tutorial. Compressible Flow Around a Wing | Tutorial | SimScale
I hope this advice helps you better understand how to succeed with generating the accurate results you desire from CFD for this mesh refinement study. If you need any further help, feel free to follow up here.
