Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Check orthogonality of fiber modes

Please login with a confirmed email address before reporting spam

I simulated a few mode fiber in COMSOL and use maximum mesh size of 0.15 um to all domains to ensure accuracy of mode profile. I exported the electric field of the modes in x and y directions with grid step of 0.1 um and calculate the dot product between two mode pair to check the orthogonality (Ex1.conj(Ex2))+(Ey1.conj(Ey2)) but the resulted modes are not orthogonal enough as you can see in the attached fig1, there is a cross-coupling of around -40db below the maximum value. Is this normal or what I have to adjust to optimize the accuracy of the solution? and in general how to improve COMSOL accuracy when fiber has a large number of modes (e.g. 50 modes)?

I have also tried to use adaptive refinement but the dot product didn't improve as in fig2.



1 Reply Last Post 2019年9月28日 GMT-4 23:18
Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 5 years ago 2019年9月28日 GMT-4 23:18
Updated: 5 years ago 2019年9月28日 GMT-4 19:18

Caveat: I've never tried to check orthogonality of computed modes before, and I haven't downloaded your .mph file to look at it... That said, I would be interested to know if you've tried different discretizations of the elements (e.g., comparing results using linear quartic, and cubic elements). If using higher-order elements gives noticeably better results (which it normally does, if errors are purely numerical artifacts) then that is the issue, rather than something more fundamentally-incorrect about the way you've configured your model. Have you tried doing that?

-------------------
Scientific Applications & Research Associates (SARA) Inc.
www.comsol.com/partners-consultants/certified-consultants/sara
Caveat: I've never tried to check orthogonality of computed modes before, and I haven't downloaded your .mph file to look at it... That said, I would be interested to know if you've tried different discretizations of the elements (e.g., comparing results using linear quartic, and cubic elements). If using higher-order elements gives noticeably better results (which it normally does, if errors are purely numerical artifacts) then that is the issue, rather than something more fundamentally-incorrect about the way you've configured your model. Have you tried doing that?

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.