Discussion Forum

Hi all, I'm working with Comsol 5.2 + Wave Optics module. I need to use the scattered field formulation in order to access the relative electric (relE) and magnetic (relH) fields. To understand how the modelling works, I simulated the propagation of a plane wave in vacuum (E field along Y). So, my geometry is a rectangular box extending along the z-axis, laterally bounded by PEC (perpendicular to the E field direction) and PMC (parallel to the E field direction) boundaries and with PMLs on the top and on the bottom of the geometry (see image1). Moreover, in image2 you can see how I defined the background wave. In the simulation volume there is nothing else than vacuum, so I expected to have no "rel" fields, since my background is the analytical solution of my model.

However, image3 shows the plot of relEy, relDy, relBx and relHx (normalized to the amplitude of the bakground fields). They are calculated at the red point in image1 and the x-axis represents the phase of the solution (calculated with a Frequency Domain study). E and D behave as expected (no "rel" fields), but I get a "wrong behaviour" for B and H: relBx and relHx should also be zero. Assuming they were not zero, at least I expect them to be proportional (in vacuum), but also this is not true.

I don't understand what I'm doing wrong. Could you help me? Many thanks!