I observe sharp changes in the pressure which correspond to partition boundaries. What causes this and how can I correct it?



In general the location of partition boundaries does not affect the solution. However, there are at least 3 cases where this is not true:

1. When a partition boundary lies along a GGI interface
2. When a partition boundary lies along a free surface interface
3. When a partition boundary lies along high aspect ratio elements

Basically these are all places where the equations are more difficult to solve, and adding a partition boundary at these locations causes problem. There are a few possible solutions:
1. In the case of high aspect ratio elements, improve the mesh to reduce the aspect ratio.
2. Run in serial.
3. Set the Partition Type to User Specified Direction, picking a direction that will avoid the problem area.





Show Form
No comments yet. Be the first to add a comment!