Photonics

Photonics

Issues Exporting to GDS From Lumerical

    • graemebargen
      Subscriber

      Hello,

      I have run the 3D inverse design of grating coupler example from (https://support.lumerical.com/hc/en-us/articles/1500000306621-Inverse-design-of-grating-coupler-3D-). I have successully run everything with my own parameters, however I am running into issues with the resulting GDS file that is created.

      When i export my grating design with the example code to GDS the resultant geometry is off by very small amounts in measurements. for instance, I ran the example twice, my waveguide width is 1450nm. but as you can see in the following screenshots from KLayout GDS Viewer, the waveguides did not export to this exact dimension.

    • Kyle
      Ansys Employee

      This might be caused by some sort of rounding error, try reducing the database units (the third argument) in the "gdsopen" command of the "extract_countour_script" string to 0.1e-9:
    • graemebargen
      Subscriber

      I tried to change the database unit as you mentioned and it did not work. The error is the same. For a 1.5um waveguide at the end of the grating it exports to a size of 1.512. This is quite a big error.
      Any other suggestions for how to remedy this?
      Thanks
    • graemebargen
      Subscriber

      I found this footnote on the page https://support.lumerical.com/hc/en-us/articles/1500007228522-GDS-pattern-extraction-for-inverse-designed-devices-using-contours-method relating to the contour extraction method.
      It appears that this method is not suitable to straight shapes like waveguides.
      I have had similar problems extracting a fully etched grating coupler geometry where I am still seeking help as I posted in another forum entry here https://forum.ansys.com/index.php?p=/discussion/27253/help-with-exporting-to-gds-lumerical-grating-coupler-example#latest
      It seems that exporting to GDS either based on the contour method or based on geometric object height are both insufficient for the instance of a filly etched grating coupler based on the Lumerical example files for the 3D inverse grating coupler design.
      Can you think of any other way of extracting the proper geometry to a GDS file with no error?
      Thanks again


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