Learning Forum

[Kim Bui]

Hi all,

I am simulating resonator structures with periodic boundary conditions (and normal source incident angle) in FDTD where I have increased the simulation time such that the standard autoshutoff level of 1e-5 is reached.

However, for some resonators, the transmission result still has "ripples". Following the instructions on Transmission Results Greater Than One – Ansys Optics, I have set the PML boundaries to "steep angle" instead of "standard". But the ripples are still there:

When I decrease the "auto shutoff min" to 1e-6 the ripples of the result above are gone:

In Auto-shutoff level (ansys.com), it was stated that the default threshold level of 1e-5 is low enough for most simulations, but is there a way to determine what the necessary threshold is?

Kind regards,

Kim

 

[Amrita Pati]

Hi Kim,

That is true, an auto shutoff min of 1e-05 works for most cases, but typically for resonators we have seen that you might need lower values (1e-06, 1e-07). Unfortunately, there is no way to predict that. How the energy decays over the simulation time is very spefic to the geometry and the source conditions, and requires a bit of trial and error to get to the right value.

Regards,
Amrita