InP MMI reflector

Stefan_1029Stefan_1029 Member Posts: 1

Dear everyone,

I'm trying to replicate the results of a symmetric InP MMI reflector based on ( I've attached my model for this (, based on an EME simulation. I'm getting similar peaks when sweeping the multimode region around the lengths mentioned in the paper (40um), but the efficiency is significantly lower than the one presented (around 15% reflection). I'm not sure how could the simulation been improved or if my setup is not suitable for this type of simulation.


  • greg_baethgegreg_baethge Posts: 30Ansys Employee

    Dear @Stefan_1029,

    Thank you for posting your question. Unfortunately, I'm not allowed to download and look at your file (see this post for more information why), but I can try to provide some guidance. Typically, the EME solver can be quite sensitive to the number of mode used for the mode expansion. You can check using the mode convergence sweep. More generally, you can run some convergence testing (see Convergence testing process for EME simulations) to see if the simulation settings are correct.

    Feel free to post screen captures of your simulation settings, such as the EME solver settings, etc. with a description of the device and the results you get and expect. This can give some information as well.

    I hope this will help.

Sign In or Register to comment.