Learning Forum

[dubey]

Hi, I am trying to simulate a SiN waveguide with SiO2 as BOX (Bottom oxide). When I change the y-span from 4 microns to 3 microns leaving all Oher parameters as it is, the losses is changing drastically. I am confused which is the correct value of the losses. Can you explain the problem and perhaps suggest the solution so that I can avoid such problem? Thank you. (Detail description in the picture, please have a look.)

[dubey]

I recall that I have asked the similar question earlier where decreasing a simulation span was causing an increase in loss which may be due to PML, but here it is quite opposite increasing the simulation span is increasing the losses.

[Guilin Sun]

It seems the top has yspan 3um as the mode profile looks larger. 

when you change spans, please make sure that you use fixed mesh size instead of number of meshes.

Not sure what the wavelength is. If SiO2 has very small loss, you can use simple dielectric refractive index. 

Loss change can be due to PML distance and materials.