I am trying to simulate the powers scattered and absorbed by a particle when the eliptical or circular polarized electric field is incident upon it under incident angles from 0 to 45 deg. For this I am using TFSF source and a cross_section analysis group. I don’t understand how it will be normalized in the case of TFSF source. Can I use two sources in the same simulation that are both incident at the same angle but differ in their phases? If not,should I just do two separate simulations with weighting factors and sum the power absorbed?
It is a good question. I have personally not tried running a simulation like what you have described. But I think it should be okay to run a simulation to contain the two sources.
The transmission function has normalization to the sourcepower function by default. In the numerator, it is the simulation results (integration of poynting vector), so it has to be a coherent result. In the denominator, the sourcepower function actually returns an incoherent sum of the power of the multiple sources. In the case where you only have one source, there should be no problem at all. However, if you have multiple sources where they should interfere, the transmission command may need some correction to the normalization. For example, in a simple case when two plane waves with the same polarization, phase and amplitude, the transmission function will return 2, due to constructive interference.
Having that said, in your case, the two plane waves are going to be orthogonally polarised. The transmission function seems to be able to return 1. I think the reason is that the power is the same for the coherent and incoherent sum in this case. So you should be good to go, but I will definitely recommend doing some sanity checks to make sure it will work as expected, eg, testing the transmission of multiple plane waves, different polarization, phase, etc. Same to the sourcepower, and sourceintensity (since you will need it for cross section calculation).
Then you can try to move on and test the behavior for the TFSF, but I will not expect different behavior from a plane wave in this regard.
As far as angled incident goes, we will suggest running single frequency simulations due to this matter.