View Issue Details

Jump to NotesJump to History
IDProjectCategoryView StatusDate SubmittedLast Update
0003286OpenFOAMBugpublic2019-06-03 08:312019-06-03 09:52
Reporterhannes Assigned Towill  
PrioritynormalSeveritymajorReproducibilityalways
Status closedResolutionno change required 
PlatformUnixOSOtherOS VersionUbuntu 18.04
Summary0003286: cyclicRepeatAMI produces obviously wrong transformation for vector field
DescriptionI created a simple test case for a turbomachinery application. It includes mapping between a sector mesh and a full 360deg mesh. I wanted to use cyclicRepeatAMI for this mapping.

First, I ran potentialFoam and examined the resulting velocity field. I expected a distribution, which is inward-pointing everywhere (the inlet is the outer patch of the wedge and the inner surface of the cylinder is the outlet). Instead, the result is an odd-looking distribution which is not symmetric.

I suspect that something is wrong with the determination of the transformations.
Steps To Reproduce1. unpack attached archive
2. run potentialFoam
TagsNo tags attached.

Activities

hannes

2019-06-03 08:31

reporter  

Auswahl_208.png (160,104 bytes)   
Auswahl_208.png (160,104 bytes)   
repeat-ami-transform-test.tar.gz (146,365 bytes)

will

2019-06-03 09:50

manager   ~0010505

Last edited: 2019-06-03 09:52

 cyclicRepeatAMI can only couple geometrically similar patches, i.e., a 30-degree sector to another 30-degree sector, as in the incompressible/pimpleFoam/RAS/impeller tutorial. It can't couple a 30-degree sector to a 360-degree sector. It is not a general mixing plane.

Issue History

Date Modified Username Field Change
2019-06-03 08:31 hannes New Issue
2019-06-03 08:31 hannes File Added: Auswahl_208.png
2019-06-03 08:31 hannes File Added: repeat-ami-transform-test.tar.gz
2019-06-03 09:50 will Assigned To => will
2019-06-03 09:50 will Status new => closed
2019-06-03 09:50 will Resolution open => no change required
2019-06-03 09:50 will Note Added: 0010505
2019-06-03 09:52 will Note Edited: 0010505