Thanks Gursoy, I followed your suggestion. I tried Ey component of source, but again normalized transmission is very large. and I can't understand field pattern. I don't know how to calculate number of pixels per wavelength , so I changed resolution to 10 and decreased the length of waveguide, but still normalized is more than 1 and not matching to expected one.
I didn't understand this point *"**normalization curve you plot has little meaning since that is the part you subtract from the second run" *. Could you please explain this. I feel there is some problem in normalization process. It may help. I would like to understand some points: *1.* In field pattern, the flux is not through whole waveguide (field pattern is appended below). But when I use continuous source for field pattern, flux is through whole waveguide. should I use continuous source for flux spectrum also. *2.* I am expecting a Gaussian profile as normalized transmission, but due to *very low incident flux* (which has gaussian profile and transmitted one has also gausian profile) under this normalization process, this unexpected and large normalized transmission is coming. Is this normalization process is right? I would like to request to meep user community, could you please have a notice on this problem. Thanks in advance. Prem Prakash Field pattern with gaussian source: [image: Inline image 1] On Thu, Dec 6, 2012 at 5:14 PM, gursoy akguc <[email protected]> wrote: > hello, > Could it be due to the source in z direction but you meant to excite > waveguide mode in y direction? > I believe you need to change source component to E_y and plot E_y > component of electric field. I know this is inprincipal same for a > rectangular geometry but your simulation is in 2d , it has no knowledge in > z direction. And the other 2 things I may think of; one is normalization > curve you plot has little meaning since that is the part you subtract from > the second run and the second is you need to calculate the number of pixel > per wavelength for your system, should be around 8 at the least, since you > are using long waveguide made up of silicon the pixel number per > wavelength may not be enough for that resolution you use. > I hope this helps . > cheers! > > gursoy akguc > > *From:* prem prakash singh <[email protected]> > *To:* [email protected] > *Sent:* Wednesday, December 5, 2012 7:18 PM > *Subject:* [Meep-discuss] Normalized transmission across rectangular > waveguide. > > Dear Meep users, > I am trying to get transmission across grated waveguide ( grating > inscribed in waveguide). I used MEEP tutorial for holey waveguide as a > reference. Problem is, I am not getting expected transmission spectrum. So > I went to check transmission spectrum across rectangular waveguide. The > result is more shocking. I think may be normalization process is wrong. > for normalization, I used (set! geometry (if norm? (list) > I would like to request , could any one comment on my code. > Transmission spectrum and field pattern is appended below. > > > Regards > Prem Prakash > > > > *********************************************************************************** > (reset-meep) > (define-param eps 12) ; dielectric constant of waveguide > > (define-param w 0.23) ; thickness of waveguide > (define-param sx 34) ; size of cell in x direction > (define-param sy 4.23) ; size of cell in y direction > (define-param dpml 2) ; PML thickness > (define-param norm? false) > (set! geometry-lattice (make lattice (size sx sy no-size))) > (set! geometry (if norm? > (list) > (list > (make block (center 0 0)(size infinity w infinity)(material (make > dielectric (epsilon eps)))))));device layer Si > > (set! pml-layers (list (make pml (thickness dpml)))); PMl layer > (set-param! resolution 40) > (define-param fcen 0.645) ; pulse center frequency > (define-param df .2) ; pulse width (in frequency) > (define-param nfreq 500) ; number of frequencies at which to compute flux > (set! sources (list (make source (src (make gaussian-src (frequency fcen) > (fwidth df)))(component Ez)(center -15 0)(size 0 w)))); Source placed just > after PML > (define trans (add-flux fcen df nfreq > (make flux-region (center 14.5 0) (size 0 (* w 2))))); collect flux before > PML > (use-output-directory) > (run-sources+ (stop-when-fields-decayed 50 Ez (vector3 14.5 0) 1e-3) > (at-beginning output-epsilon) > (at-end output-efield-z)) > (display-fluxes trans) > > > ***************************************************************************************** > > for normalisation - norm?= true > Transmission spectrum: > > [image: Inline image 1] > > Ez pattern: > [image: Inline image 2] > > > > > _______________________________________________ > meep-discuss mailing list > [email protected] > http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss > >
<<ez-000250.20.png>>
_______________________________________________ meep-discuss mailing list [email protected] http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
