Photonics

Photonics

FDTD 2D and 3D simulation results are too different

Tagged: 

    • pengzheng97
      Subscriber

      I optimized a structure in 2D light field as follow

    • Guilin Sun
      Ansys Employee
      I believe your waveguide structure has limited dimension in z. However, when you do 2D simulation, the software will assume there is no any variation, or confinement of the em fields. Basically it would be uniformly infinite in the 3rd dimension in 2D simulation. So they are not the same physical device.
      If you check the vector fields, you will see the difference.
      I would recommend you to us evarFDTD, which uses information in the 3rd dimension, yet the simulation is done is 2D, which is much fast and yet reasonable accuracy compared to 3D simulation.
    • pengzheng97
      Subscriber
      Thank you for answering my question.
      I have another question, is varFDTD MODE, can it be used in combination with 'LUMOPT' inverse design?
      Best wishes!
    • Guilin Sun
      Ansys Employee
      Yes, you can find examples here:
      https://support.lumerical.com/hc/en-us/articles/360042305314-Inverse-design-of-waveguide-crossing
      Inverse design of y-branch
      Please write a new post if you have questions not in the topic content.

Viewing 3 reply threads
  • You must be logged in to reply to this topic.