Simcenter STAR-CCM+ Introductory Guide to Overset Simulations
2020-12-14T17:17:49.000-0500
Simcenter STAR-CCM+
Simcenter STAR-CCM+ Virtual Reality
Teamcenter Share
Simcenter Cloud HPC
Simcenter STAR-CCM+ Viewer
Simcenter STAR-CCM+ Application Specific Solutions
Summary
The article outlines the key guidelines for setting up an overset mesh simulation in Simcenter STAR-CCM+.
Details
Introduction
Overset meshes are used to bring together different parts of a computational domain and move them relative to one another without having to re-mesh multiple times. This functionality allows users to easily simulate different moving parts easily. A common application of overset meshing is gearboxes. For a general overview of overset and other mesh motion methodologies, visit here.
This article walks through various considerations at each stage in the workflow of an overset simulation. Additionally, it directs users to overset resources and documentation amongst Simcenter’s provided materials.
Preparing Geometry and Defining Regions
The geometry preparation and pre-processing for an overset simulation does not differ from the general process of a STAR-CCM+ simulation. There must be separate parts for the volume where the fluid exists (background), each moving body, and a volume to surround each body (overset). The background and overset parts will be used to create regions, and each part surface will be used to create a boundary.In a simple example of a sphere moving translationally through a domain, you can see the spherical body within the background region.
The following image shows the part that represents the background region where the fluid flows.
The overset volume surrounds the spherical body so the flow can be solved around it as it moves.
By multi-selecting and right-clicking on the background and overset parts, you can select “Assign Parts to Regions” to create a region to each part and a boundary to each part surface as depicted below:
The “overset mesh” type is assigned to the boundary surfaces of the regions in motion. This part surface is the exterior of the overset volume that will be coupled with the background mesh type.
Interfaces
Interface Set-Up“Overset Mesh Interface” type is to be used for any overset mesh simulation except for a zero-gap configuration.
- For each overset region, an interface needs to be created between it and the background region.
- If there are multiple overset regions that will overlap, an additional interface will need to be created between each pair of overlapping overset regions.
“Overset Mesh Zero Gap Interface” is to be used if you expect a zero-gap configuration to occur in your simulation. This means that two regions approach each other very closely or make contact.
Methodology
The stationary background and an overlapping overset mesh exchange flow information at the interface. This process consists of the hole-cutting algorithm and the donor search. The hole cutting algorithm determines which cells are active, inactive, and acceptors. The donor search checks that each acceptor cell in one mesh has a matching donor cell in the other mesh. Further explanation on the overset methodology is located in the user manual by the following path: Simulating Physics-> Physics-> Using Overset Meshes-> Overset Mesh Methodology. The overset hierarchy is necessary for the hole-cutting algorithm because it determines which region gets priority for each of the overlapping meshes. This is especially important when multiple overset regions are present. The overset interface hierarchy can be checked in the overset interface properties.
- Region-0 defines the overset region with higher priority.
- The background region or the overset region with lower priority is identified by Region-1.
Small Gap
If there are moving objects approaching each other or a wall, the cells in the gap between the bodies are set inactive, causing a void. As the bodies move apart, the cells become active again. The thickness of the gap when the cells are inactive depends on cell size and the overset intersector. If the inactive cells are large, there might be a remaining void between the wall boundaries. Overset mesh modeling requires a minimum of two to three active cell layers to resolve a gap between two wall boundaries. In the case of a remaining void, prism layer shrinkage should be activated to morph the cells of the prism layers between the surface of the bodies, ensures that this requirement is met.
Mesh Considerations
All mesh types are compatible with overset meshing. Trimmed meshes have the advantage of being faster to solve than polyhedral cells, due to the relatively lower number of faces. However, extra care needs to be taken for generating this mesh to avoid problems such as sudden cell volume changes in areas of interest. The meshers chosen should be the same for the background and overset regions.The most important mesh consideration is that the mesh density in the background and overset regions must be the same for overlapping cells around the overset boundary. In other words, the size of the cells needs to be of the same order as the refined background one.
For the overset to cut the background, you should have at least 4-5 overlapping cells between the overset and background. This ensures there are enough cells across the overlapping zone between background and overset regions.
If prism layer shrinkage is activated, it is required to have a minimum of five prism layers on all boundaries for which you expect close contact with a neighbor body.
Physics
The physics selection process for an overset simulation does not differ from the general process for a STAR-CCM+ simulation. If the simulation involves moving parts in real time, the implicit unsteady solver should be selected. Additionally, if the user chooses to utilize the overset model-driven refinement to refine regions of lower priority in the overset hierarchy, Adaptive Mesh Refinement needs to be selected and Overset Mesh Refinement can be activated through this option.
Adaptive Mesh Refinement
In order to efficiently utilize overset meshing, Adaptive Mesh Refinement is a key enabler. AMR refers to the process of automatically refining the mesh where and when it is needed and leaving it coarse elsewhere. This tool allows for quicker simulation runtime as well as a reduced overall cell count. More information on AMR can be found here.