ring modulator

Lizzie_LeonLizzie_Leon Member Posts: 8

Dear all,

I have been trying to repeat this example -> (https://support.lumerical.com/hc/en-us/articles/360042322794-Ring-modulator) with my parameters. But the thing is I get too high coupling coefficients in FDTD. So when I import the results into Interconnect there are very poor results.

Also I have noticed that the results depend on the numbers given to core width, core thickness and slab thickness.

In the example, the ring radius is 10 um and in my work it is twice smaller.

My questions:

1) Why it is happening?

2)How those core width, thickness and slab thickness are affecting the results and how to know what are the proper values for those parameters?

Below there are screenshots from the results which I get:


  • gsungsun Posts: 751Ansys Employee

    The simulation result in INTERCONNECT definitely depends on FDTD result where is originated from the device configuration, eg, the design parameters. Usually you will need to optimize the coupling length, or the gap, for a given waveguide geometry/material. If the coupling is strong, you can either increase the gap, or decrease the length.

    In order to get knowledge on how do the core width, thickness and slab thickness affects the coupling you can sweep them (looped sweep): https://support.lumerical.com/hc/en-us/articles/360034922913-Creating-nested-parameter-sweeps but it will take long time since it is looped. This is your design process.

  • omniaomnia Posts: 7Member

    I have a question on the same example please.

    Taking the coupling coefficient as abs(S)^2 makes it the same value as T taken from power monitor at the same place. right? In this case coupling coefficient (expressed most of the time as kappa) is dimensionless. In reality it should be expressed in units of m^-1.

    How this can be interpreted and how we can get kappa in units of m^-1 to compare to other published papers.

    Thank you.

  • gsungsun Posts: 751Ansys Employee
    edited June 28

    Personally, I think people use different definitions for "coupling coefficient": some use kappa as the field coupling (complex value), and others use power quantity such as in this example. They are quantify different properties or different devices. I do not think those two are interchangeable. So please use your own definition to quantify the property you want. You may refer Yariv's book for the field coupling coefficient.

Sign In or Register to comment.