Grating Coupler design issue about diffraction

Hello, I would like to design a grating coupler, so, I run the example that you have at Lumerical. In that example, the light goes from the fiber to the grating coupler. Also, in that example, finally, a CML model is created with the S parameters. But, my purpose is to design a coupler that the light will go from the grating coupler to the fiber. Also, I put a gaussian source (not a mode source, not ports). By using a movie monitor, I want to see that the light goes to the fiber at the angle of 10 degrees. But, when I open the movie monitor file, there is not any light that is transferred to the fiber. Could you please help me?
grating_coupler_2D_coupling efficiency optimization_best_solution.fsp (6.5 MB)

I could not upload the result from the video monitor but if you run the simulation you will see that no light goes to the fiber.

Hello @xh19363,

Sometime it can be difficult to see the fields with the movie monitors, so I would recommend you use ports or mode expansion monitors to measure the power transmitted into the fiber modes.

Generally Gaussian sources are used to inject a beam into empty space, the beam will not be injected properly if it is inside a structure like this. Why would you like to use a Gaussian source instead of a mode source in this simulation?

@kjohnson
I changed the Gaussian source to a mode source. So, I run again the simulation by setting the mode source at port 2 (at the waveguide).
grating_coupler_2D_with mode source.fsp (1.6 MB)
I use a DFTMonitor because I want to see the light that is diffracted to a certain angle (in that example at 10 degrees). But from the DFT monitor, I can see that a very small amount of power goes to the direction of 10 degrees. What can I do so as the majority of the light to be diffracted in that direction? Here, you can see the result from the DFT monitor.

This is what I would like to design

@kjohnson
Only with the mode expansion monitor will I see the field? The DFT monitor is not suitable for that? I would like to find something like the last figure.

Hello @xh19363,

I don’t think a mode monitor or port will be useful for visualizing the fields in this simulation. They can tell you how much light was coupled into the fiber modes, however, which is usually the figure of merit you would like to optimize when designing a grating coupler like this.

If you would like to determine the angular distribution of the light exiting the grating, you can remove the fiber, place a DFT monitor horizontally above the grating and perform a far field projection with the results. This will give you the fields propagating away from the grating as a function of angle.

Let me know if you have any questions about this,