Photonics

Photonics

Variation in output for different Lengths in EME solver

    • Anish_Mondal
      Subscriber

      Hello Lumerical Community,

      I am simulating a coupler made up of SiN using the EME solver in Lumerical Mode. I simulate 10um of the coupler using 25 cells and then use the EME solver to extend it to larger lengths 250um, 500um, and 1000um. But what I'm finding is that there is a difference in the field plot for the 3 different lengths. Why is this happening?

    • Guilin Sun
      Ansys Employee
      The results seem reasonable to me. When you increased the length, have you increased the "x resolution" at the same time?
      If so, you can zoom in for longer length result. Please note that you plot the results in above the same image scale whereas their actual lengths are quite different. When you squeeze the short length result, you will see similar images.

    • Anish_Mondal
      Subscriber
      No, I did not know about increasing the resolution. Does the resolution have to be increased for every time I increase the length or do I just have to keep it sufficiently large so that it can accommodate larger lengths?
      Thank You
    • Guilin Sun
      Ansys Employee
      The mesh sizes in yz are given in EME. However, in X direction users need to specify it. BY default, it has only 100 points. if you do not modify it, 1000um long has only 100 points, that means it has about 10um sampling in x. If you want to show the fields for the whole device you should properly increase this value.
    • Guilin Sun
      Ansys Employee
      It is not mandatory but the sampling is very coarse! it is up to you for the resolution: if you want to get better,smoother result you should increase the resolution. EME is also a discrete meshing algorithm, and the resulting profile can only be as good as the mesh.
Viewing 4 reply threads
  • You must be logged in to reply to this topic.