Hi guys, I have another question, so bear with me?

What are the structured grid-types I can use in Fluent?

For face meshing, I can use Quad, and for Volume, I can use Hex, but what about the other elements like triangles?

Thanks -Riaan

For simple geometries, quad/hex meshes can provide high-quality solutions with fewer cells than a comparable tri/tet mesh and for complex geometries, quad/hex meshes show no numerical advantage, and you can save meshing effort by using a tri/tet mesh.

-Prakash

The only structured grid type is a mapped quad mesh. The tri-primitive and submap are also structured meshes for faces, but it's simply an algorithm they use that breaks the faces into quadrilaterals that can be meshed using a mapped scheme. I've heard the wedge-primitive meshing scheme described as either structured or unstructured (because it uses tri elements at the tip and then a mapped mesh out from that). A quad mesh is unstructured if you use a pave meshing scheme, and a tri mesh is always unstructured. A volume mesh is structured if all of the faces are structured and either a mapped or cooper meshing scheme is used for the volume(a cooper mesh is ONLY structured if ALL of the face meshes are structured!).

All of that is simply semantics though. Prakash was right on when he said that for complicated geometries an unstructured mesh is just as good. In fact, I've seen people create very poor meshes in the hope that the fact that it was structured would make up for the fact that the mesh was close to useless. They waste hours or days building the mesh, and then the results are no good because of the mesh. For simple geometries a quad mesh will use less elements and therefore save computation memory and time. It's a trade off deciding if it's worth your time to spend it meshing or or letting the computer do the computing. 9 times out of 10 I would say spend it computing though. Besides if it's a big mesh, you can do other things while the computer thinks!

Hope this helps, Jason

Thanks for the help! The reason I am going structured is that I will be doing a parametric study of the delta wing and that I will have more control over the grid sizing and locations.

Right now my delta wing will have about 500k-600k cells, so thats a good start for a grid-dependance study.

Regards Riaan