Trimmed Cell Mesher
The trimmed cell mesher provides a robust and efficient method of producing a high-quality grid for both simple and complex mesh generation problems.
It combines a number of highly desirable meshing attributes in a single meshing scheme:
- predominantly hexahedral mesh with minimal cell skewness
- refinement that is based upon surface mesh size and other user-defined refinement controls
- surface quality independence
- alignment with a user specified coordinate system.
By default, the trimmed cell meshing model utilizes a template mesh that is constructed from hexahedral cells at the target size from which it cuts or trims the core mesh using the starting input surface. The template mesh contains refinement that is based on the local surface mesh size and local refinement controls. Growth parameters can be used to control the transitioning of the mesh cell sizes from small to large both at the surface and far field. A maximum and/or minimum cell size can be supplied as well to control the upper and lower cell size bounds. The template can be aligned in any direction in a user specified Cartesian coordinate system.
The resulting mesh is composed predominantly of hexahedral cells with trimmed cells next to the surface. Trimmed cells are polyhedral cells but can usually be recognized as hexahedral cells with one or more corners and/or edges that are cut off.
An example trimmed cell core mesh is shown below:
An additional feature that is useful in modeling external aerodynamic flows is the ability to refine cells in a wake region. This region is generally the volume of fluid at the rear of a moving body.
The input values used for the trimmed cell mesher can be set on three different levels:
If you remove the Trimmed Cell Mesher from an Automated Mesh Operation, Simcenter STAR-CCM+ saves any user-defined settings and restores them if you reactivate the Trimmed Cell Mesher.
Volumetric controls can also be included to increase or decrease the mesh density locally in the template using a range of prescribed shapes to define the extents.
The current implementation of the Trimmed Cell Mesher model is restricted to one trimmed volume mesh per automated mesh operation. In other words, a different automated mesh operation must be used for each trimmed volume mesh that you wish to generate. Alternatively, the per part meshing option can be activated to mesh parts or regions independently. At part or region interfaces a non-conformal interface is generated. Additionally, interfaces can only be used with the trimmed cell mesher when the boundaries in the interface belong to the same region. For example, an in-place interface could be used to model a baffle in the flow domain, or a periodic interface for periodic flow condition. In all instances, a conformal trimmed mesh across the interface is not guaranteed.
The trimmed cell mesher also has a built-in ceiling to control the size of the template cells next to surfaces. This ceiling is to avoid grossly de-featuring the model and generating bad quality cells. At present, the maximum surface template cell size is fixed to 150% of the average triangle edge length size in the local surface. This factor does not affect cells away from the surface.
Trimmed Cell Mesher Guidelines
- The trimmed cell template mesh size is based on the default target surface size. The trimmed cell mesh only grows by multiples of 2 from the target surface size. You are advised to set all refinements for the trimmed cell mesh as multiples of 2 from the target surface size, for example: 25%, 50%, 200%, etc. Using a target surface size of 100% of the base size can be helpful.
- Activate parallel trimming on meshes with more than 100,000 cells per process. For meshes smaller than this threshold, serial trimming will often outperform parallel trimming due the extra overhead of parallel communication between processes. Adding more processors will improve performance at a diminishing rate of return. At some point adding more processors will result in a negligible increase or even a decrease in performance. The optimal number of processors depends on the case.