You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As our random ray solver currently uses the flat source approximation, cells (a.k.a. flat source regions (FSRs)) usually need to be sized at or below the scale of typical mean free paths in order to maintain accuracy. This is presently left to the user to do by manually subdividing cells.
More ideally, we would be able to use some of the abstractions in OpenMC for mesh tallying to overlay a virtual FSR mesh over the grid. This would allow for automatic subdivision of physical cells into smaller FSRs. While the FSR shapes may not be as ideal as if manually divided by a user, in many cases manual subdivision is impractical, so this offers a good path forward in many cases. There are expected to be a few issues that pop up with this (e.g., infinitesimally small FSRs getting generated), but previous work has shown that there are reasonable solutions to these issues and the general method should work.
The text was updated successfully, but these errors were encountered:
As our random ray solver currently uses the flat source approximation, cells (a.k.a. flat source regions (FSRs)) usually need to be sized at or below the scale of typical mean free paths in order to maintain accuracy. This is presently left to the user to do by manually subdividing cells.
More ideally, we would be able to use some of the abstractions in OpenMC for mesh tallying to overlay a virtual FSR mesh over the grid. This would allow for automatic subdivision of physical cells into smaller FSRs. While the FSR shapes may not be as ideal as if manually divided by a user, in many cases manual subdivision is impractical, so this offers a good path forward in many cases. There are expected to be a few issues that pop up with this (e.g., infinitesimally small FSRs getting generated), but previous work has shown that there are reasonable solutions to these issues and the general method should work.
The text was updated successfully, but these errors were encountered: