3D simulation error about optic fiber

    • atfkerem76

      Hello! I'm trying to simulate an optical fiber in FDTD. Previously I was working with nanometer based structures, but now I have a different product which has micrometer specifiments. 

      While working with nanometers everything was fine. But now I had to change every component I use, to micrometers. And now lumerical gives me an error like this:


      After getting this, I reduced the min mesh step and mesh accuracy. It worked but results of E-field in "monitor_3" which is centered to fiber core, was completely zero. 

      Is there a way to run this simulation with reasonable results? Or any other things I can do? 


    • Guilin Sun
      Ansys Employee

      FDTD is a discretized method. Therefore larger devices such as thousands of wavelengths will need a lot of memory.  To get resonably accurate result, mesh accuracy 1 may be used. But coarser meth than mesh accuracy 1 is not recommended but you could try mesh size lambda/max(n) to be 4 at most.

      You may need to significantly reduce the simulaiton volume. Place the source  just 1 mesh away from the fiber, PML can be closer to a quarter of wavelength. If it still needs memory for thousands of GB, then you may not be able to simulate.

      What result you want to simulate? scattering?

    • atfkerem76

      I reduced the simulation volume, put the source very close to structure and changed the accuracy. Currently it can simulate, but sometimes it crashes before it finishes or after simualiton, in monitor it gives me zero as E field values like this: 

      And yes, I want to simulate the scattering properties. Maybe I can use TFSF source at this point? 


Viewing 2 reply threads
  • You must be logged in to reply to this topic.