I also met this problem before. If your structure is so large and calculation time is too long, you should activate non-averaging to avoid divergence issue just like; (set! eps-averaging? false)
Quoting dy zhao <[EMAIL PROTECTED]>: > Dear Dr. Steven and meep users, > > Recently, I have met the divergence problem several times. Here is just one > example. In order to get the more accurate transmission of a PC slab, I set > a relatively long run-time, but the field is divergent at about 800 > meep-time-step for r=0.25. When r=0.20, it is also divergent, but when > r=0.28,0.30 or other values, they are not divergent. I think the resolution > 40 is enough and the values of the other parameters are also reasonable. > Why is it divergent for some special PC structures? Please give me some > advice. > > Thanks! > > Best regards, > > Deyin > > The following is the ctl file: > > (define-param n1 3.48) > (define-param eps (* n1 n1)) > (define-param r 0.25) > (define-param thick 0.34) > (define-param supercellhight 6) > > (set! geometry-lattice (make lattice (size 1 1 supercellhight)));the > computional cell > ;(set! eps-averaging? false) > (set! geometry > (list > (make block > (center 0 0 -1.0) > (size 1 1 thick) > (material (make dielectric (epsilon > eps)))) > (make cylinder > (center 0 0 -1.0) > (radius r) > (height thick) > (material (make dielectric (epsilon 1)))))) > (define-param fcen 0.625) ; pulse center frequency > (define-param df 0.5) ; pulse width (in frequency) > > (set! sources > (list > (make source > (src (make gaussian-src (frequency fcen) (fwidth df))) > (component Ey) > (center 0 0 1.5) > (size 1 1 0)))) > > (set! k-point (vector3 0 0 0)) > (set! pml-layers (list (make pml (direction Z) (thickness 1.0)))) > (set-param! resolution 40) > > (define (Eoutput) > (print "Eoutput, " (meep-time) ", " > (get-field-point Ey (vector3 0 0 1.0)) ", " > (get-field-point Ey (vector3 0 0 -1.8)) "\n")) > > (run-until 1638.4 Eoutput) > _______________________________________________ meep-discuss mailing list [email protected] http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
