Ack, ignore my response; I just noticed that -0 recenters the origin to
the center of the dataset.
On Tue, 15 Jul 2008, matt wrote:
>
>
>
> I haven't looked at your code closely, but I have an idea about what
> could be wrong.
>
> when you ask h5topng to give you a "y 0" slice, that's not meep
> coordinates y=0; that's the first y slice of h5 data. try using h5ls to
> figure out how many slices there are in the y dimension, and pick a
> different slice in that range, corresponding to the middle of your
> waveguide (e.g., y 18).
>
> best,
> matt
>
>
>
> On Tue, 15 Jul 2008, Tobias Glahn wrote:
>
>> Hello meep-discussion list,
>>
>> I'm trying to simulate a 3D rectangular waveguide (one could say I
>> should do it in 2D but it's just some steps from what I really want to
>> do...). My .ctl - file looks a follows (I'm not sure if I messed with
>> the meep units and the cutoff freqeuncy):
>>
>> ------------------snip----------------------------------------------------------
>>
>> ; define some constants and measures
>> (define-param length_z 24) ; already added some space for pml
>> (define-param width_x 7)
>> (define-param height_y 1)
>> (define-param pml-size 2.0)
>> (define-param inner_block 5)
>> (define-param inner_eps 4)
>> (define-param center_freq 0.5) ; with a = 1m and c = 1 I get w_c =
>> pi*c/width_x < 0.5
>> (define-param freq-samples 25)
>> (define-param df 0.1)
>>
>> ; create a lattice with proper size
>> (set! geometry-lattice (make lattice (size width_x height_y length_z)))
>>
>> (define-param no-block? false) ; to get transmission spectra one can set
>> this variable from the command-line
>>
>> (set! geometry
>> (if no-block? ;if the variable is set from outside there are two
>> implementations!
>> (list
>> (make block
>> (center 0 0)
>> (size width_x height_y length_z)
>> (material (make dielectric (epsilon 1)))))
>> (list
>> (make block
>> (center 0 0)
>> (size width_x height_y length_z)
>> (material (make dielectric (epsilon 1))))
>> (make block
>> (center 0 0)
>> (size width_x height_y inner_block)
>> (material (make dielectric (epsilon inner_eps)))))))
>>
>> (define-param absolute-flux-position (-(- (/ length_z 2) 1) pml-size) )
>>
>> ; create a source
>> (set! sources (list (make source
>> (src (make gaussian-src
>> (frequency center_freq)(width 6)))
>> (component Ey)
>> (center (+ absolute-flux-position 1) 0)
>> (size 0 (* height_y 0.95) 0)))); this source should
>> not touch the boundaries
>>
>> ; apply the pml in z - direction (the others will be handled later)
>> (set! pml-layers
>> (list
>> (make pml (thickness pml-size) (direction Z))
>> (make pml (thickness 0) (direction X)); electric boundary in X direction
>> (make pml (thickness 0) (direction Y)))) ; electric boundary in Y
>> direction
>>
>> ; create Flux planes (like ports, but no input!)
>> (define trans ; transmitted flux
>> (add-flux center_freq df freq-samples
>> (make flux-region
>> (center 0 0 absolute-flux-position)
>> (size width_x height_y 0))))
>> (define refl ; reflected flux
>> (add-flux center_freq df freq-samples
>> (make flux-region
>> (center 0 0 (- absolute-flux-position 0))
>> (size width_x height_y 0))))
>>
>> ; boundary conditions for hollow guide
>> ;(init-fields) ; fields need no init if fluxes added before; not sure if
>> those are needed (all metallic)
>> ;(meep-fields-set-boundary fields High Y Metallic)
>> ;(meep-fields-set-boundary fields Low Y Metallic)
>> ;(meep-fields-set-boundary fields High X Metallic)
>> ;(meep-fields-set-boundary fields Low X Metallic)
>>
>> ; set the resolution to some value
>> (set! resolution 20)
>>
>> ; if there is an older computation load the spectrum for the input
>> (produces bad error when file not there!)
>> (if (not no-block?) (load-minus-flux "refl-flux" refl))
>>
>> ; really start the simulation
>> (run-sources+ 100
>> (at-beginning output-epsilon)
>> (to-appended "ey" (at-every 0.6 output-efield-y)))
>>
>> ; save the flux for later transmission tricks
>> (if no-block? (save-flux "refl-flux" refl))
>>
>> ; show the values to get them into matlab
>> (display-fluxes trans refl)
>>
>> --------------------snap-----------------------------------------------
>>
>> so far this script is partly adopted from the tutorial's page. Now
>> running this script with:
>>
>> ~> meep no-block?=true waveguide.ctl | tee out.log
>>
>> and extracting some graphics via
>>
>> ~> h5topng -0y0 -t 100 -Zc bluered waveguide-ey.h5 -o Pic_y.png
>>
>> I only get black bricks (the same for x and z direction) in
>> Pic_x,y,z.png. Whats my mistake? Is the source somehow not working
>> right? I think putting in a single e_y component sould somehow exite the
>> TE_10 - Mode. Any Ideas? Thanks in advance for your help,
>>
>> greetings
>>
>> Tobias
>>
>> _______________________________________________
>> meep-discuss mailing list
>> [email protected]
>> http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
>>
>
> _______________________________________________
> meep-discuss mailing list
> [email protected]
> http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
>
_______________________________________________
meep-discuss mailing list
[email protected]
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
