CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Pointwise & Gridgen (https://www.cfd-online.com/Forums/pointwise/)
-   -   Bad elements after generating unstructured volume mesh (https://www.cfd-online.com/Forums/pointwise/194774-bad-elements-after-generating-unstructured-volume-mesh.html)

max_beetle October 24, 2017 01:52

Bad elements after generating unstructured volume mesh
 
5 Attachment(s)
Hello, I'm trying to mesh a ground vehicle model for external aero analysis using unstructured mesh. I watched this video as basic steps https://www.youtube.com/watch?v=I7PmmVfoGSg. The volume mesh near the vehicle has very high aspect ratio (>80) and max. included angle. I'm trying to figure out why. My steps for generating the unstructured mesh are follows:

1. Set defaults mesh settings for connectors and domains. I set the average deltaS for connectors and leave anything by defaults. For unstructured domains isotropic option; algorithm: advancing front ortho with triangle and quads cell types, quad max. included angle: 165 deg, boundary decay 0.85 and leave anything else to default.

2. Assemble model, join some quilts, generate the mesh and examine aspect ratio and max. included angle and it's quite good. Figures attached.


3. Build the computational domain using draw shapes. Generate the surface mesh with advancing front algorithm and triangle cell type. The deltaS is equal to 1.5% of the computational domain length size. Surface mesh aspect ratio and max. included angle are great.

4. Assemble 2 models together using assemble special --> blocks --> automatic (create baffle faces)

5. Create sources around the vehicle model to make smaller tetra mesh size with constant sizing

6. Defaults for unstructured block are: boundary decay for isotropic is 0.85 and leave anything else to defauls settings. For T-rex, max. layers: 50, growth rate: 1.2, push attributes checked, cell types: all-tets, pyramids, prisms, hexes, collision buffer: 5, max. angle 165, centroid 0.5, wall deltaS is set to make y+ ~ 1. Set BC for the vehicle and ground to wall type and the domains adjacent to the ground (inlet, outlet, sides) to match type. Initialize.

7. Examine the mesh and found that the meshes have high aspect ratio and max. included angle near the vehicle. Figure attached

8. For the gap region (between the vehicle and ground), it is expected that the mesh around that area have high angle, but I hoped that by creating sources, I can reduced the mesh size on that area in order to prevent that kind of problem.

9. But why a lot of elements with very high aspect ratio generated around the vehicle? I tried to reduce the mesh size for the computational domain to make the transition smoother, but don't work.:confused:

Thank you :).


All times are GMT -4. The time now is 16:20.