EME - force fundamental TE mode (Ey dominant) for cells

bibaaa12bibaaa12 Member Posts: 3


I was wondering if there is a way to set the EME solver, so that it uses the fundamental TE mode for the power coupling and transmission results calculation?

I've tried using anti-symmetric boundary conditions but it still seems to use the wrong mode.

Thanks in advance.


  • greg_baethgegreg_baethge Posts: 93Ansys Employee

    Hi @bibaaa12,

    Thank you for posting your question. The symmetric/anti-symmetric boundary conditions will affect which modes are allowed but not necessarily the port mode used. To change which mode(s) is(are) used for the S parameter calculation, you need to modify the ports' settings. By default, the fundamental mode is selected, but you can choose any mode. To do so, in the port property window, in "EME port" tab, select "user select" for "mode selection" to choose an arbitrary mode:

    Then you can calculate the modes and select the one of interest.

    You can also select the fundamental TE mode. You can then visualize the mode properties to make sure they correspond to the correct mode.

    I hope this will help.

  • bibaaa12bibaaa12 Posts: 8Member

    Hi Greg,

    thanks for getting back to me so quickly.

    I have my port set to fundamental TE mode, but the mode profiles I see in the individual cells do not default to the fundamental TE mode.

    Does setting the port to fundamental TE guarantee that only that TE mode is used for calculating the overall transmission through my waveguide?

    Is there a way to plot the mode profile for each cell that is actually used for the power coupling calculation? Ideally via the script, as I'm running everything via the Python API.

    Thanks again

  • greg_baethgegreg_baethge Posts: 93Ansys Employee

    You're most welcome, @bibaaa12. I don't think it would necessarily make sense to limit the modes used in the cells as this will affect the mode expansion. If you suppress modes fields would couple in while propagating, the final result (S parameters) will not be accurate and there will be issues with the energy conservation.

    You should be able to access information from individual cells using getdata or getresult. The path to the cells would be like "EME::Cells::cell_1", etc. So you can use script to access them.

  • bibaaa12bibaaa12 Posts: 8Member

    Thanks for the reply. Maybe I am approaching this problem wrong. The issue I am trying to solve is that if I put my structure into 3D FDTD, with the same mode source as in EME, the field I see inside the segmented waveguide, ends up not being the same as what EME gives as the mode in the cell. See screenshot. So, the transmission I get ends up being very different.

    Any advice on this would be much appreciated. Maybe EME is not the best choice here.

  • bibaaa12bibaaa12 Posts: 8Member

    To clarify: Light propagates in the y direction and the field in the XZ crosssection ends up looking very different.

  • greg_baethgegreg_baethge Posts: 93Ansys Employee

    Thanks for the additional information. To clarify, I assume you did rotate the structure in EME since it can only support propagation in the x direction.

    One thing to keep in mind: the modes stores for each cell will not represent the fields as they propagate through your structure. On the other hand, the fields that propagate are a weighted sum of all the modes supported on the cross section. In that sense, it's not a problem these modes are not the fields you get from a FDTD simulation. To get the fields, you need to place a monitor where you want the fields and they will be reconstructed when you click "propagate".

    I think the transmissions being different could probably have a different origin. You may want t to run some convergence testing for both the EME and FDTD simulation, just to make sure the settings are correct. That said, could you share a screen copy of the EME settings?

  • bibaaa12bibaaa12 Posts: 8Member

    Hi again,

    yes, I did rotate everything, so in EME the light is propagating in the x direction.

    Here is a screenshot of my EME settings.

    I will try the additional monitor at the position I am interested in. Would you expect the EME monitor results to be similar to the power field monitor results from the 3D FDTD?

    If you see anything wrong with my settings, please let me know.

Sign In or Register to comment.