Multi-Region Surface Requirements

For STAR-CCM+ to successfully surface remesh and/or volume mesh a multi-region case with a conformal mesh between regions, the surface has to be prepared to adhere to the following conditions:

each region should be closed, manifold and non-intersecting;

common boundaries that exist between regions should be grouped together and flagged as interfaces; and

in instances where the interfaces between regions have different areas, imprinted mesh lines have to exist to define the common area

In instances when a conformal volume mesh is not required between each region, then the requirement to have interfaces and imprints defined is relaxed. In this case, each region is meshed independently of each other using a different mesh continua and can then be joined back together later either by using an in-place interface or by fusing the boundaries together. In some respects, this is a far easier approach to going to all the trouble of making sure the region interfaces and imprints are properly defined. However, the process will result in an additional overhead at the simulation stage, resulting from the extra interface information stored or the extra face divisions resulting from the boundary fusing process.

When dealing with multi-region topologies, there are many variables that affect the definition of the CAD data with respect to the requirements for STAR-CCM+. These include:

the format used to export the CAD data e.g. STL, IGES, VDA etc;

whether each part in the assembly is exported to the same file or to separate files;

whether duplicate surfaces exist at inter-region boundaries;

whether inter-region surfaces have been imprinted or not; and

whether the assembly contains all relevant fluid/solid parts or only the solid material portion of the geometry

For example, certain CAD packages such as SolidWorks allow you to save an assembly as a single STL file or each part as separate STL files. Similarly, in most CAD packages, you can elect to open each part one at a time and decide to save them individually as an IGES file or as other formats. Alternatively, the entire assembly can be saved to one file, which is usually the default (or only option allowed) in most cases. Depending on which of the above methods are used to save the import file(s), you may end up with either single surfaces at inter-region boundaries or duplicate surfaces, which may or may not share the same perimeter triangulation. Additionally, some CAD packages allow imprinting (whereby one part can be "stamped" onto another, leaving an edge definition representing the common area) while others do not have this feature. Finally, as is traditional in the design industry, the assembly may represent the solid material component and may not contain the opposite fluid volume which will be required for the simulation if performing a fluids analysis in conjunction with the solid. Some CAD packages easily allow the fluid portion to be extracted and combined back into the assembly while others do not have this capability (or it cannot be easily achieved, depending on the complexity of the model). So, as we can see, a number of different factors can influence how easily or not we are able to prepare the import geometry ready for STAR-CCM+.

In general, regardless of the factors that affect the multi-region geometry definition, STAR-CCM+ contains the required tools to properly prepare the surface ready for meshing most cases. In particular, two operations specific to multi-region meshing are essential for improperly prepared surfaces to be manipulated in order to conform to the above requirements. They are:

create a (duplicate) interface boundary; and

create a new region from a collection of boundaries;

Being able to create a duplicate boundary at an inter-region interface allows two regions that were previously "joined" to be separated out, thereby allowing a new region to be created from individual boundaries. The surface boundary that was duplicated maintains an in-place interface definition as part of its make up so no further action is required once the region has been redefined.

The use of the in-place interface option is also important for defining the surface correctly in cases where the regions are already defined at import but no connectivity information relating one region to another is included. By including an in-plane interface, conformal meshes at interface boundaries can be created for the surface and volume meshers.

Two different starting scenarios will be discussed here showing how the starting CAD surface data can be manipulated in STAR-CCM+ so that a conformal volume mesh can be constructed. The two different scenarios are:

the CAD import results in a single region; and

the CAD import results in a multiple region

Other alternative instances (for example, a combination of the above two) can arise but in general the two approaches outlined should provide a sufficient enough insight as to be able to deal with all other cases that you are likely to come across.