Reply To: For bfast light source, the reflectivity swings back and forth

[Amrita Pati]

Hi,

The BFAST algorithm is known to be more numerically unstable compared to the FDTD, especially for steeper incidence angles. I will recommend you to have a look at this article: Broadband Fixed Angle Source Technique (BFAST). The section on numerical stability talks about possible causes of instability and the next section talks about the best tips for using such a source. I will also recommend you to try RCWA for the same setup at some point. The BFAST algorithm is slower than FDTD as it uses smaller time step, and as you are experiencing could be unstable. RCWA on the other hand will be much faster as it is semi-analytic and also doesn't required any special measures for simulating angled broadband source. You can add an RCWA object to the same simulation region and can also use it for comparison with FDTD result.

Regards,
Amrita