Photonics

Photonics

Top Illuminated PIN photodetector- CHARGE

    • Maira Perez
      Subscriber

      Dear Ansys support,

      I am currently trying to get the dark current of a vertical PIN photodetector using a 2D simulation in CHARGE. However I have a question regarding the solver paramater ''norm length". Would it be correct to set a length that multiplied by the x length of my simulation area gives the same area of my circle or should li just give the circle diameter.

      Top illuminated PIN configuratio with simulation region in orange.

      Thanks in advance for the help!

    • Guilin Sun
      Ansys Employee

      The norm length is used to get correct current. The solver actually calculates the current density, which is invariant this is why we can simplify 3D device to be 2D.

      Since current is additive, yes, you can choose a proper norm length.

    • Mariah Ha
      Subscriber

      what is the correct norm legth in that case please ?

    • Guilin Sun
      Ansys Employee

      it can be 2*pi*r where r is the proper radius of curvature.

    • Maira Perez
      Subscriber

      I am  having some troubles with to run this simulation. When the generation rate data is disable I can run my simulation of forward and reverse bias with the following warning

      Then when the generation rate data in enable (under illumination simulation) I get a convergence error. I have tried to folllow the suggestion from : Convergence issues in CHARGE with no sucess

      Then I tried to increase the auto- refinement and I played around with the global mesh constraint. However,i am still now managin to get my simulation converge under illumination.

      I would appreaciate some insigth and help.

      Error from log file

       

       

      Current CHARGE settings

      Changing gradient mixing for the other options was also tried.

    • Guilin Sun
      Ansys Employee

      Not sure if you refer this link Troubleshooting convergence errors in CHARGE

      It is the primary guide to overcome divergencing issues. You may also try to reduce the triangle quality, use non-zero nr ratio etc.

       

    • Mariah Ha
      Subscriber

      I was stuck in the same issue before and I would suggest reducing the max global mesh size  as well.

Viewing 6 reply threads
  • You must be logged in to reply to this topic.