Plane Wave and Open Boundaries Cavity Resonance Problem

[sn_london]

I am simulating a sample mounted on a substrate (high-frequency time domain solver CST) and am trying to look at the resonances of the sample. However, I am having an issue with the Fabry Perot resonances produced from the modeled thickness of the substrate I am mounting my sample on.

I currently have a plane wave propagating through the substrate to the sample and have the boundary conditions set to open. I thought this would make the substrate's edge touching the plane wave act as a perfect absorber and have minimal reflections therefore effectively being as if the plane wave starts from inside the substrate. However, the width of the modeled substrate really affects the results, e.g., if I model it with 50um or 100um then the resonances change drastically.

Any idea what I am doing wrong?

--- Updated Aug 21, 2023 ---

This is my result, rz is just changing the thickness of the mount.

 

[PlanarMetamaterials]

In many simulators, boundary conditions set to "open" implies some given background material, e.g., air/vacuum. Therefore, I wouldn't expect it to work for your substrate without specifying this material.

It may also be the case that your simulation suffers from convergence issues, which are exacerbated when the mesh changes. I would rule these out first by increasing mesh density or applying more stringent convergence criteria.