ubermag / oommfc

OOMMF calculator.
http://ubermag.github.io
BSD 3-Clause "New" or "Revised" License
50 stars 17 forks source link

Question about open boundary conditions seem to not working? #162

Closed dovod91 closed 2 weeks ago

dovod91 commented 2 weeks ago

Hello team,

I have a question about the open boundary conditions. I naturally assume these are 'perfect' absorbers, however, in the following picture at timestep 112, I can see reflections at the borders of my simulated box (see x = 0). The source is a localized region at x = 50. The mesh settings are:

mesh = df.Mesh(region=region, cell=cell, subregions=subregions, bc='')

I am assuming open boundaries are set correctly, so I cannot see what might be wrong with my mesh. Subregions are shown here too, but I don't think they are an issue... many thanks for any insight of what I could check next!

image image

fangohr commented 2 weeks ago

I have a question about the open boundary conditions. I naturally assume these are 'perfect' absorbers, however, in the following picture at timestep 112, I can see reflections at the borders of my simulated box (see x = 0).

It is not clear to me why open boundary conditions should be perfect absorbers.

There is a paper suggesting change of damping at the edge to suppress reflections: https://doi.org/10.1016/j.jmmm.2017.06.057