On Wed, 23 May 2007, [EMAIL PROTECTED] wrote: > in one of my previous posts I try to use a source in (111) direction. I > found a file written by Steven G. Johnson, modified by Ashifi Gogo (Jan > 2007) > with help from Markus Testorf ( > http://article.gmane.org/gmane.comp.science.electromagnetism.meep.general/706 > ). I have rewritten it to make it 3d (file is attached). Unfortunately > this source is not stable, at timestep 100 something goes wrong (see the > attached png, y-slices). I tried continuous-src, gaussian-src and a > broadband source, but only the continuous-src seems to be stable.
I'm not sure what you mean by "stable" but from your plots I guess you mean "produces a wave in the expected direction?" You are producing a source at an angle by specifying a planewave source with a given wavevector k along the source plane. This produces a source with a fixed angle that depends on frequency. So, if you use a continuous-src (fixed frequency) it will settle down into a fixed angle, but if you use a Gaussian pulse source you will get a superposition of many frequencies (unless it is very narrow-band) and thus a superposition of many angles. Not that the latter is a problem; as I've explained in several previous postings, you can still use the flux spectrum to get the transmission at a particular frequency, and thus a particular angle, from a short pulse. In fact, you can get the transmission spectrum from multiple angles/frequencies simultaneously ... this is a big advantage of a time-domain simulation. See my previous postings on the subject. Cordially, Steven G. Johnson _______________________________________________ meep-discuss mailing list [email protected] http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
