On Tue, 19 Dec 2006, Christine Kreuzer wrote:
So I can treat these two results from (harminv (component Ex) position) as
equal and say that the Q from one mode is three times the other one?
(harminv0:, frequency, imag.freq., Q, |amp|, amplitude, error)
harminv0:, 0.398810428143646, 1.65113746462119e-5, -12076.8390485024,
0.0108854479386194, -0.00326644571009244-0.01038380032779i,
2.44300222652076e-6
harminv0:, 0.396857014202571, -4.64781040806467e-5, 4269.2900458457,
0.917322813922249, 0.564914833787812-0.722739493530499i,
7.30449030154138e-8
A Q=-12000 should certainly not be treated as the same as a Q of +12000,
if that is what you are asking. It can mean one of three things:
1) you didn't run for a long enough time, and the decay rate is so small
that you are just looking at noise.
2) You forgot to wrap your harminv call in (after-sources ...) ... if
harminv overlaps your sources it will get confused because the sources are
not exponentially decaying fields.
3) You are encountering some instability and the fields are actually
blowing up slowly, hence Q < 0.
When in doubt, the best thing to do is always to run with a narrower
bandwidth source around your frequency of interest (here, 0.398 or so),
and/or for a longer time.
Steven
_______________________________________________
meep-discuss mailing list
[email protected]
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
