>> but z coordinate of objects was changed to any nonzero value, values >> of >> propagated EM field and following spectrum were diverged, >> >> thinking theoretically, propagation and follwing results should not be >> affected by parallel transfomation along any z direction. > > By "diverged" do you mean "going to infinity" or just "larger than you > expect"? I'm guessing the latter. > > If the source is too close to the spheres, near-field coupling to > resonant modes will increase the power output by the current source > relative to vacuum (see also the Meep FAQ). (This is essentially the > definition of a resonant mode: in increase in the local density of > states, which is directly related to the power radiated by a localized > current source.) This is especially problematic since you are looking at > such long wavelengths, longer than the separation between your source and > the spheres). Look only at shorter wavelengths, or move your source > farther from your objects. >
Thanks a lot for your comment, dear prof. Steven. As you could see my object model put in the input, my plane wave source is located at z=10, and objects were in origin of z coord. at first attempt. I think that distance is reasonably far enough because when I tried FDTD calculation before with some simple multilayer models located at from z=0 to about z=-8, and same source at z=10, I could get very nice spectrum results. And thickness of object layer was just square of 2, I also think the distance compared to layer thickness was far enough. I'm afraid to say that, in the concrete, trying to calculation with object whose z coordinate was shifted even small value, +0.1 or -0.1, the values of wave stored in the output-efield-y h5 files, and corresponding flux, were diverged - going to infinity! I tried to perform with z= -2, -1, 1, 2, and above. And that phenomenon seemed to no relation with a sign - + or - - of the z coordinate. Only with the object located exactly at z=0 I could obtain somewhat reasonable spectrum. I questioned and guessed that some symmetry breakdown for z coordinate makes results ruin, I cannot understand or even guess what is the essential of that "symmetry" at all. Could you guide me once more? Thanks again for your care, sincerely. _______________________________________________ meep-discuss mailing list [email protected] http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
