On Fri, 9 Mar 2007, Lu Hu wrote:
I tried to test meep by calculating reflection from a dielectric thin film.
It looks like the results never converge to the analytical solutions no
matter how fine the grids are. Could anybody point out the mistakes
for me? The ctl file and plot are attached.
It works for me. Perhaps there is a units mixup?
For example, at your center frequency fcen = 0.20833, the analytical
transmission and reflection through a slab of epsilon=9 with thickness d=5
at normal incidence are T = 0.52941 and R = 0.47059, respectively. I
reproduce this to within 9% at a resolution of 10, and to within < 1% at a
resolution of 40.
I took your input file, and made a couple of small modifications to make
it run more quickly.
* I commmented out the (run-until 2000) line and uncommented the
(run-sources+ (stop-when-fields-decayed ...)) line, which is more
appropriate here. However, you need to change Ex to Ez in that line since
you are using an Ez source.
* I changed the cell size to no-size along the x direction - there is no
point in having more than 1 pixel along the x direction for a structure
that is uniform along the x direction.
* I changed nfreq to 1 and df1 to 0 so that it only computes the flux at
the single frequency fcen (since I just wanted one frequency as a quick
check).
Cordially,
Steven G. Johnson
_______________________________________________
meep-discuss mailing list
[email protected]
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
