Hi guys,
When I use MFIX-DEM, it is easy to understand that the grid size should be larger than the particle diameter. But, how to understand that the grid size should not be less than 2diarlm_factor when grid_based (cell-linked) particle neighbor search algorithm is used?
Grid-based neighbor search algorithm, why the desgrid size should not be less than 2*dia*rlm_factor?
the cfd grid need not be larger than the particle diameter. this is only the case for grid-based kernels. however the neighbor grid better be larger than the largest particle or you are going to miss collisions. the grid method bins up particles and only searches over local particles to detect collisions. if your des neighbor search grid is too small (or your search interval too large), you will miss collisions, either all together (which is benign but not a good representation of the physics we try to model w/ dem) or late (which may be crash your run if there is an excessive overlap causing particles to rebound too energetically).
1 Like
Hi William, thank you very much for the detailed explanation. I am wondering if there is a reason for 2diarlm_factor; I mean, the limitation could be, for example, 3dia, why it has to be 2dia*rlm_factor?
That is the bare minimum to guarantee 2 particles in contact are in the same des grid cell (rlm_factor is a small safety factor). By default, the des grid size is 3 times the max diameter. This can be modified by specifying the number of des grid cells in each directions (keywords DESGRIDSEARCH_IMAX, DESGRIDSEARCH_JMAX, DESGRIDSEARCH_KMAX) in the Solids>DEM pane, “search grid partitions (optional)” setting.
2 Likes
Thanks a lot, Jeff. 