Photonics

Photonics

FDTMonitors T and R for Multiple Sources and Oblique Incidence

    • joaonizer
      Subscriber

      Hi, I have a simulation setup where two Gaussian beam sources are used to launch a Probe and a Pump signal in a thin slab of nonlinear material (but you can think of it as being linear, for simplicity I've used a Drude model for the upcoming results). The probe has a bandwidth of 1000-1650nm, 1 MV/m amplitude, normal incidence. The Pump signal has a narrower bandwidth (1200-1300 nm) amplitude 1 GV/m, TM Polarization and 60 degree oblique incidence.

    • Taylor Robertson
      Ansys Employee
      Hello Using override global monitor settings is a good practice here, and I would first ensure you are observing the results you expect from each beam separately before trying to combine them. Since your pump beam will be 1000x greater amplitude if they are not interacting, then I would expect this response to dominate
      I think one problem is that the simulation span is not wide enough, and your pump beam is being directed into the PML. Are the pump and probe simulation both running to completion? Maybe check the logfile to see how the simulation is ending. It may be helpful to verify the set-up by comparing with the Fresnel coefficients from STACK.
      https://support.lumerical.com/hc/en-us/articles/360034914653-Stack-optical-solver-overview
      As for the normalization state I don't know exactly what will work for you, but given that this will be nonlinear I would avoid cwnorm. It may be that outside of the pump bandwidth you can multiply by x1000 to bring the signal up to similar levels? To begin with I would keep the amplitudes equal, just so that it is easier to compare. In the end you will likely want to use a custom time signal for the pump and probe, and if these match up your system then nonorm would be most directly comparable to what you would measure.
      I hope this helps. Best



Viewing 1 reply thread
  • You must be logged in to reply to this topic.