Half Planar Green's Function vs FEM

pmcardle
pmcardle Altair Community Member
edited October 2020 in Community Q&A

  I am trying to simulate the propagation of electromagnetic surface waves at the interface of free space, and a material. These surface waves are excited by a plane wave incident on a sharp metallic object near the material/air interface. To inspect the surface waves,  I am requesting the electric near-field in and out of the plane of sample, and I expect to see something resembling a spherical wave in the electric field distribution. To simulate the sample. I initially tried using the Half-Planar Green's function method, but I was not able to see any surface waves. The field distributions exponential decayed away from the metallic object with no oscillatory behavior. I then switched to the FEM with the same material and got fields that resemble a surface wave. Any thoughts on why the Green's function method is not producing the same results as the FEM? My initial thought was the discretization of the material in the FEM vs Half-Planar greens function. But this is a hypothesis that I cannot prove which is why I am consulting this forum.

Tagged:

Answers

  • Mel
    Mel Altair Community Member
    edited July 2020

    Near field calculations are supported up to 100 free space wavelengths from the closest mesh element. How far have you calculated them?


    Please post some images or better even the models if you can.

     

     

Categories