Dear meep-community, I want to use a tensor for the drude-lorentz polarizability factor sigma. So, in principle I have a gold-circle in the center of my quadratic 2D domain, but the instead of just putting a 1 on the diagonal in the material_function::sigma_row, I also put a 0.3 on the off-diagonal. Result: as soon as an incoming wave hits the sample, fields in pixels on the metal-air interface near the area, where the staircasing is strongest (i.e. the x=+-y straight) grow exponentially. Has anyone ever experienced this before?
I put a minimum example on github: https://github.com/fesc3555/meep_nondiagonal_sigma/blob/master/weird_broken_gold_sphere.cpp the evil lines are 24 and 25. If you comment out those, everything works like a charm. A bit of background: The transformation optics things I posted here before (use transformation optics for variable resolution) transform the sigma in the same way the transform the epsilon. So, for non-diagonal Jacobians, this results in non-digonal sigma. Now, everything works fine for dielectric media, but if I start to transform metals, everything blows up. Thank you for your time in advance, Felix -- Felix Schwarz Technische Universität Ilmenau FG Theoretische Physik I Tel: +49 3677 69 3644 _______________________________________________ meep-discuss mailing list meep-discuss@ab-initio.mit.edu http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
