Bragg grating and Bragg cavity




I am trying to use FDTD to simulate a cavity surrounded by two bragg gratings. The design is basically a core of a silica fiber ( diameter 2 micron) in which the grating and the cavity are etched.

First I wanted to check the behavior of the grating, so I simulated 20 periods of 1 micron depth, 808.3 nm width, the period is set to 202 nm with a silica length of 127nm.

The results I get strange results such as a transmission above 1, and the reflection spectrum does not fit a Bragg mirror spectrum, as you can see on the following picture:

Here is the file I made :
BraggGrating - 20 periods.fsp (6.2 MB)

I tried different meshes and several parameters in fdtd but I still get similar results.
Can somebody advice me ?


Dear @jeansc

I found three main problem in your simulation:

  1. you were using custom PML. I think this was causing transmission above 1. We strongly recommend to start with standard PML. I personally never have used custom PML.

  2. grating does not seem to be centred, and as you can see in the perspective view, they are sticking out of the fiber. This might explain why reflection is not as you are expecting.

  3. Once you have defined port objects, you do not need to define another source.

Transmission and reflection plots are shown below:

Here is the modified simulation file for your review:
BraggGrating - Circular - 20 periods_modified.fsp (6.2 MB)

Please go ahead and modify the geometry and let me know if you had further questions.


Bragg reflection of nanohelix

Thanks for you help @bkhanaliloo

The grating is not centered because it is actually etched directly on the fiber. I used index monitors to ensure that the index profile fit the aimed design. If you have any suggestion to make it better please tell me.


Dear @jeansc

I used a finer mesh but results were still the same. Can you please explain what you expect for the results? Do you have any references for your simulations?



@bkhanaliloo I am actually doing this to have some simulation data to fit some measurements already done.
It seems that my colleagues are happy with the result from those simulation, so the result must be accurate :slight_smile:.

Thanks again for you help !


Dear @jeansc

I am glad to hear that results make sense.

Best of luck in your research.