Automatically adjusting EME span

crossingsymmetrycrossingsymmetry Member Posts: 3


I am going through your Lumerical mode example on 1x2 MMI splitter. I would like to do parameter scan to extract the optimal MMI length. I noticed that when I change the MMI device length, the EME setup parameters remain fixed. As a result, the simulation region is no longer optimal (ex: if you increase the MMI core length from 32um to 50um during the parameter sweep, the x min will no longer be at the end of the tapered input waveguide and the simulation results are meaningless.) Is there a way to set the group spans and x min as variables instead of having them fixed?

Following is an illustration of it, as I run parameter sweep, when the MMI length increases to 50 micron, the input ports and cell group 2 length remain fixed, resulting in unphysical results.

Thank you for your help.



  • kjohnsonkjohnson Posts: 175Ansys Employee

    Hello @crossingsymmetry ,

    The cell group spans of the EME solver object itself (like in your screenshot) are used to define the cross-section and initial length of each cell group, if you sweep them without changing the geometry as well then they will not line up with the geometry properly, as you have seen. To perform a sweep of the length of a cell group with the EME solver you should use the propagation sweep feature in the EME analysis window:

    This will properly sweep the length of the section of the geometry in that cell group, without having to recalculate the modes.

Sign In or Register to comment.