On Mar 31, 2008, at 12:05 PM, fanguofang wrote: > > Hi: > I want to know > the relationship of the unit of the > meep ,resolution ,the wavelength and the smallest size width in the > device? > For example: > The wavelength is 1.55micrometer, the smallest > width size is 100nm ,the largest size is 0.1mm in the device, > When I set the unit is 1 micrometer and the > resolution is 10,it is too large for my computer to calculate. > So I want to set the unit as 10micrometer and the > resolution is 10, So there is only 1.5point on the wavelength and > only 0.1 point on the smallest width. > . Will the result have a big fault?
If you have less than 2 pixels per wavelength, the results will probably be garbage. You have to have at least a few points per wavelength. As a rule of thumb, I usually would use at least 8 pixels per wavelength even for my lowest-resolution calculations. And that's not 8 pixels per vacuum wavelength, it's 8 pixels per wavelength in the highest-index material. If your structure is too big to calculate with a reasonable resolution, usually you can break it up into smaller pieces that you can analyze separately. e.g. if your device consists of a bunch of waveguides, filters, resonators, etc., you want to compute the transfer functions through each of these components (and the junctions between them), and then combine these transfer functions to get your whole device's performance. Usually there is a way to do something like this if you think about it. Regards, Steven G. Johnson _______________________________________________ meep-discuss mailing list [email protected] http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
