Currently our default BCs for everything are dirichlet(0). This includes the ion and neutral momentum in all cases when no other boundary was set (which is every application of the code I've seen).
This is not realistic for the radial boundaries because they don't correspond to physical walls. For this reason, SOLPS-ITER uses Neumann(0) as their default BC for those regions.
I'm not sure if this is very impactful - having a non-rotating core impacts the other core rings through anomalous viscosity, but surely the flows there pick up speed quickly anyway. I will investigate the impact on my ST40 case, and in the meantime we can discuss whether we should change the default to follow SOLPS.
Currently our default BCs for everything are dirichlet(0). This includes the ion and neutral momentum in all cases when no other boundary was set (which is every application of the code I've seen).
This is not realistic for the radial boundaries because they don't correspond to physical walls. For this reason, SOLPS-ITER uses Neumann(0) as their default BC for those regions.
I'm not sure if this is very impactful - having a non-rotating core impacts the other core rings through anomalous viscosity, but surely the flows there pick up speed quickly anyway. I will investigate the impact on my ST40 case, and in the meantime we can discuss whether we should change the default to follow SOLPS.