Hello all,

I face a problem simulating the transition from one dielectric waveguide to another one of the same type but 90 degrees rotated along its axis. My setup is very similar to the "Straight Section" example. The difference is that the cladding in my model is a dielectric in one dimension and vacuum in the other:

I get this warnings:

and indeed the balance is above 1 leading to unrealistic S-parameters. The same warning appears even in the examples provided in the component library. How can I improve my setup to get more accurate and physically consistent results?

I am happy about any recommendation, thanks for taking your time.