Time Monitors and Source Position:FDTD

Hello All:

I am trying to simulate 2D Photonic Crystals using FullWave FDTD simulation tool by R-Soft. My first step is to get the transmission charateristics of the PC. To do this, I am plotting the |Ey|2 vs. the frequency (taking the FFT).
1] I would like to know more about the positioning of the time monitor.
2] In the reciprocal space we know about the 1st Brillouin Zone, hence the Γ-X and Γ-K directions of propogation. The bandgaps are different for these two directions. Now, what is the equivalent in the real space (FDTD problem space)?

Thanks in advance.

V

Dear Zeroedus,
1) In order to determine the transmission coefficients (that you see in experimental datas), put a set of time monitors one wavelength (or period) away from the output surface of the crystal. The monitors may be placed by lamda/10 distance apart from each other. And they shall be placed parallel to the outpt surface. In addition, they shall span a length of approximately equal to the aperture of an horn antenna. Note that when FFT is enabled you will get the field at only one point. After the simulation is performed, all you have to the is to average the transmission amplitudes of all monitors. That will give you quite successful results.

2) The best way to realize these directions is to rotate the crystal by an amount corresponding to the propagation direction

Irfan:

Thanks a lot for the reply. I'm trying your suggested method, and will contact you if I have furthur doubts (hope that's fine!). It seems you have worked a lot in this area.

About your second suggestion, for a triangular lattice, I have to rotate the crystal structure by 30 ° increments to test different directions of propagation. Now to compare it to the band diagrams, I have to consider the portions containing the considered k-points. Have I understood this right?

Thanks once again.
V

I have worked and am still working on photonic crystal In my opinion, a solid knowledge of "solid state physics" is quite helpfull in understanding photonic crystals.

Regarding your question, it is right. But dont forget to remove the extra rods. It is similar to Miller planes in solid state physics. If you have the book by Kittel, it may be quite usefull.

Hello Irfan and All:

A rule of thumb for horn antennas is that the length/width of the aperture should be 3 times the wavelength for E -fields and 4 times for H-fields. The free space wavelength I have considered to be 1. Hence, this will be too large, so I have considered 7 monitors of length 0.45 each(my grid size is 0.02). It is giving reasonable results.

I have considered a source width of twice the period. What size of the source do you choose usually?

Irfan: If you have used your FDTD simulation results from FullWave in any of your papers, please can you provide the titles. It'll be of immense help.

Thanks.
V

I usually choose a source width large enough to excite sufficient number of Bloch modes. Usually, that would be liken 6-8 periods. But you can find a good width through a simple optimization. You can start from small width and then increase the width step-by-step till you get no difference in the transmission.

Hi
How can get the average the transmission amplitudes of monitors?
and what is length of the aperture of an horn antenna?