Dear Steven G. Johnson and MEEP users, I am doing a simulation of a line defect with air holes square 2D PhC. Using MPB I have found the frequency of a defect in the PhC band-gap, using a supercell. After that, I have use MEEP to find the quality factor of the 2D PhC. Really, I did first in MEEP the simulation to obtain the flux with the pure structure (without line defect) and normalize it with the flux obtained without structure as usual. From that I find the spectrum, where it is clearly visible that the band-gap is present in the same frequency range as in the case obtained with MPB.
Now the question: when I simulate in MEEP giving the fcen and df obtained for this band-gap (where the defect is located), I wait for the source to be off (that takes approx. 260 units), and also wait for other 1000 units. With that situation, I achieve a Q factor of 4.000.000. If I do that with 3000 time units, I achieve the same results, but if I do it with more that 5000 there is no results using Harminv. I have also inspected the field, and could see that in the first case the mode is clearly present, but in the case of 6000 time units, the last slices present a lot of pixels and it appears as it were a very high order (very high) mode, some kind of artifact (fast random field - even more, one can see how the field begins to degrade as the time passes). If the Q is so large, it has to remain unchanged for a lot of time, so what is the field pixeled? I am quite sure that the defect is real, because from other simulations I could also find with MPB two defects in the band-gap (also with high Q factors, 300000 and 900000), and after that doing the MEEP simulation I could see how they interacted with each other. Applying a lower df, I could observe only one mode without problems. Moreover, the error that comes with Harminv is very low (10e-12) in contrast with (10e-8) for the imag. frequency. Why is the field pixeled? Thank you very much, Gerardo.
_______________________________________________ meep-discuss mailing list meep-discuss@ab-initio.mit.edu http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss