Hi, I would like to discuss one problem that I have encountered in FDTD.
Most simulations that I went through is to use a single mode source and a eigenmode expansion monitor at the output to analyze the power that has been allocated to all modes (or polarizations). However it seems to work terribly when multiple sources are injected at the same time, where power normalization becomes wrong.
The way that I solve the problem is to deploy an eigenmode expansion monitor immediately after mode sources (before actual devices) to find the power distribution as input reference then to use output mode expansion monitor power to get normalized with input reference. But this will assume that two channels are not converted during propagation through the device under evaluation, which may not always be true.
It’s also strange that for example on a 900nm by 220 nm SOI waveguide when two mode sources (TE0&TE1) are injected concurrently with 1:1 power, the input reference does not give uniform power ratio. Is this a bug of FDTD tool?
I have also uploaded my simulation file foe you to look into.
beating.fsp (7.5 MB)