FDTD: Port Eigensolver Issue

[mc5284]

Hello

I interface with lumerical using the Python API, and am having an issue in FDTD where the injected port mode for a straight waveguide has a negative effective index due to the mode solver using "let the solver choose" option for the solver algorithm, when I'd like it to use the "E and H" option. There is currently no method that I've found available to call that would let me change the solver algorithm option (listed under the "Advanced Options" menu after entering the select mode option for user selecting modes). I've confirmed that using the "E and H" solver gives me the correct sign of neff and the complex part, and that both the "E transverse" and "H transverse" options yield the incorrect neff. Is there any line of code available to help me, or is this something the Ansys team can add soon? It's critical for performing sweeps where the geometries change - to manually intervene in this way is not ideal in any way.

Image 1: In this case, the "Let the solver choose", "E transverse" and "H transverse" options all yield an incorrect mode

Image 2: In the "E and H" solver algorithm, the correct mode is yielded:

Thanks and all my best,

MC

[Kirill]

Dear Subscriber,

1. It seems that the 'solver algorithm' setting is not accessible via script. This could have been done intentionally, or it might be a 'missing feature.' I will need to conduct further exploration to clarify this. Unfortunately, I can't propose any workaround for this issue at the moment.
2. A separate question is why you're getting negative values with the default settings. I believe this should not happen in most cases. If you'd like to continue the discussion on this topic, please provide more details of your simulation, ideally to the extent that I could reproduce this issue.

Best regards,
Kirill