Home Forums FDTD PWE Band Solver Simulation

PWE Band Solver Simulation

Profile Photo
(joined November 2018)
Participant

Hello,
In the tutorials, I read that when you include a point defect in your photonic crystal and run the PWE band solver simulation, you should fix the supercell settings so that the supercell includes the defect and the resulting band diagram should include a flat defect band where the original band gap was. However, when I ran my simulation with the supercell including the defect, the resulting band diagram does not include the defect band. Can you help me with this?

The supercell parameters with the defect are saved, and the supercell parameters without the defect are the default parameters.

Thank you so much,
Malik

Responses (25):

  • #50432
    Profile Photo
    Malik Javaid
    Participant

    I found the transmission spectrum attached, but when I run the FDTD simulation with the peak power in the graph, 4.508 um, the results don’t match what I found in the graph.

  • #50438
    Profile Photo
    Scott Newman
    Participant

    I would need to see your reference as well as your design file to comment.

    Scott

  • #50439
    Profile Photo
    Malik Javaid
    Participant

    Oh I thought I had attached them, I guess they didn’t attach.

  • #50476
    Profile Photo
    Scott Newman
    Participant

    Is that transmission spectrum in fact obtained for the structure without the waveguides and resonator?

    To be entirely honest, looking at the structure I am not entirely certain I know what you are in fact trying to accomplish with this structure. Why the need for the resonator with the waveguides? They are so close you will get coupling through evanescent fields whether you have resonance or not.

    Do you have a reference for what you are trying to setup?

  • #50477
    Profile Photo
    Malik Javaid
    Participant

    https://www.photonics.com/a63226/All-Optical_Logic_Gates_Show_Promise_for_Optical

    This article is the reference I am using. I am trying to replicate what they have built, except instead of using one input waveguide, I am planning on using two with different intensities so that the waveguide with the greater intensity light is able to control the other waveguide’s light using nonlinear properties. This article is also supported by the book you had referred to me, Photonic Crystals: Modeling the Flow of Light, on pages 214-218 and by the screenshot I have attached to this post which I had posted before. This screenshot is on page 197 of the book.

    Here is the link to the book in case you need it:
    http://ab-initio.mit.edu/book/photonic-crystals-book.pdf

    Also, the transmission spectrum I have attached above is in fact for the structure with the resonator and waveguides.

  • #50479
    Profile Photo
    Malik Javaid
    Participant

    I apologize, I attached the wrong file. Here is the correct one.

  • #50483
    Profile Photo
    Scott Newman
    Participant

    If that transmission spectrum is for the structure with the waveguides then a source of 4.5um corresponds to a region of minimal transmission, you would need to use a wavelength that corresponds to one of the spikes in the transmission.

  • #50485
    Profile Photo
    Malik Javaid
    Participant

    I ran the simulation again with a source of 4.5um and I received the graph shown where the peak is 4.055um. I then ran the simulation once again with a source of 4.055um and I received a transmission graph different than the one I had gotten before. Why is this the case?

  • #50495
    Profile Photo
    Scott Newman
    Participant

    The mesh resolution in your design file is set to auto which sets it to 1/10 the central wavelength of your source in the highest refractive index material. You are running at a fairly low resolution so changes in the resolution could result in different simulation results.

  • #50545
    Profile Photo
    Malik Javaid
    Participant

    I ran the simulation at the lower mesh size, as you said, but the results I received were similar to the results shown above.

    I also read that increasing the amount of PMLs in the simulation could reduce error, however, that also did not yield accurate results for me. Can you help me find other ways to decrease the error in the graph and get accurate results?

  • #50580
    Profile Photo
    Scott Newman
    Participant

    This is a design issue that you need to work through. I will look at it if I get a chance but this is up to you.

LoginYou must be logged in to reply to this topic.