Transmittance at through port of grating assisted ring resonator


#1

Hello Sir,
I want to study transmittance of ring resonator having grating structure in coupling region. xy
Transmittance at through port, T>1 with wavelength which is a strange response. While response at drop port is as it should be


The problem usually does not arise if grating is absent in coupling region.
In my simulation, early shutoff (min) was unchecked, dt stability factor = 0.4, broadband fundamental mode having 50 nm span was used. Log file of the simulation is here:grating coupled ORR.log (10.3 KB)
It will be very helpful for me if any comment to troubleshoot this problem I can find.

Thanks!!


#2

Hi @krishnd

I cannot comment much about why you are seeing these results without looking at your simulation file. Please share the simulation files for a review.

When I checked the log file, it looks like auto shutoff is not triggered. This can be the reason why you are seeing oscillations and transmission above 1 at the through port. Also, I am not sure why you are using a low dt stability factor. Please elaborate.

Thanks


#3

Hi @bkhanaliloo
Thanks for your response!!
The log file and lms files are shared here:
https://drive.google.com/drive/folders/1rvhIAZ4rlmdjOcMD0fDg56t-Qrdqdw0v

I have modified some parameters as auto shutoff (min = 1e-05) is checked, simulation time is increased to 20k fs to ensure that signal passes simulation domain completely and source is TE polarized now . Earlier, I had reduced dt stability factor to 0.4 because I was expecting for a stability problem but even it is changed to 0.99, the problem still persist.
PML boundary is enough far from source and monitor to avoid the field to be truncated. The problem may be related to auto shutoff only or some other parameter can also be responsible to avoid it ?
Kindly help!
Thanks


#4

Hi @krishnd

The simulation setup is fine and I expect that the oscillations are due to early shut off. You can increase the simulation time to make sure that it triggers auto shutoff. I also had a few comments to make debugging faster:

  1. Start your simulations by using a coarse mesh. You can use a refined mesh only for final simulations.
  2. Decrease the xy span of varFDTD simulation region. A gap of wavelength size should be enough between geometry and the simulation boundaries.
  3. You can start by simulating a smaller ring and then move to larger rings when the simulations are working fine. A ring with no grating geometry might be also useful.

Please keep me updated with your results.