View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0004210 | OpenFOAM | Bug | public | 2025-01-31 16:43 | 2025-01-31 20:58 |
| Reporter | peth | Assigned To | will | ||
| Priority | normal | Severity | minor | Reproducibility | always |
| Status | closed | Resolution | suspended | ||
| OS | arch linux | ||||
| Product Version | 12 | ||||
| Summary | 0004210: foamPostProcess patchIntegrate problem on NCC boundary | ||||
| Description | In $FOAM_TUTORIALS/multiRegion/CHT/misalignedDuct tutorial foamPostProcess's "patchIntegrate" function on non-conformal mesh with NCC coupling seems to be problematic on fluid1Solid2_on_fluidNonCouple1, as the coupled boundary net surface size is calculated to be close to 0 in the integration. However, by inspecting the mesh boundaries fluidNonCouple1 and SolidNonCouple2 (these are the two selected base patches of the "fluid1Solid2_on_fluidNonCouple1" coupled boundary), I think the two patches have a non-zero overlapping area. This problem also happens on the "fluid2Solid1_on_fluidNonCouple2" coupled boundary too, however on the "fluid1Fluid2_on_fluidNonCouple1" patch the overlapping area is calculated correctly (0.0004 m^2). | ||||
| Steps To Reproduce | #create new sample simulation from tutorial cp -r $FOAM_TUTORIALS/multiRegion/CHT/misalignedDuct sim #navigate to sim_source and run it for a few iterations cd sim_source ./Allrun #run foamPostProcess to integrate some quanitiy (integrating T is meaningless in this case, it is selected only to show the problem) foamPostProcess -region fluid -latestTime -func 'patchIntegrate(patch=fluid1Solid2_on_fluidNonCouple1,T)' | ||||
| Tags | No tags attached. | ||||
|
|
edit: in "Steps to reproduce" part the second command should be: cd sim |
|
|
fluid1Solid2_on_fluidNonCouple1 has (near) zero size because it is a inter-region nonConformalMappedWall interface and foamPostProcess has only loaded one region. Without both regions the intersection cannot be calculated and the interface is therefore left empty. fluid1Fluid2_on_fluidNonCouple1 has the correct size because it is an intra-region nonConformalCyclic interface and hence it only needs one region to be loaded for the intersection to be calculated and the interface to be propertly constructed. The solution here is to extend foamPostProcess so that it has (optional) multi-region support. Closing pending funding for such a development. |
