search

Jashcraf / poke

Poke (pronounced poh-keh) is a Polarization Ray Tracing and Gaussian Beamlet module for Python
BSD 3-Clause "New" or "Revised" License
32 stars 6 forks source link

Polarization Impact on FIber Nuller #5

Open Jashcraf opened 2 years ago

Jashcraf commented 2 years ago

GLINT is an instrument on subaru that uses a fiber-nulling photonic integrated circuit for high-contrast imaging. Depending on where in the pupil is sampled, the fiber nuller will experience variable polarization aberration. This issue is to track the progress on assessing the impact on the null depth GLINT could theoretically achieve.

Jashcraf commented 2 years ago

The Jones Pupil for Subaru - includes the primary, secondary, and tertiary mirrors

image

Jashcraf commented 2 years ago

We send two electric fields through the system, equal to the coherent sum of the elements in each row. Put more simply...

E1 = J00 + J01 E2 = J10 + J11

image

This is what that looks like

Jashcraf commented 2 years ago

Let's just pick two points and see how we do, start with +/- 1/4 of the pupil in the Y direction image

Jashcraf commented 2 years ago

Now we step through OPD on the second channel of +/- a wave and record the irradiance at the center of the beam. Here are the results for $\lambda$ = 0.75 microns. Both curves are normalized to the maximum of the curve where the same polarization state is injected in both channels. I'm not 100% sure this is the best approach, because we get inequal power in the resultant field due to diattenuation. I suppose that makes this a comparison of both amplitude and phase errors that arise from different polarization states.

image

Notice that there appears to be a slight shift of the polarized fringes w.r.t. the "Single polarization", which assumes that both beams experience the same polarization in channel 1 = E1 = J00 + J01. Then both are normalized by the maximum in the single polarization fringes. The fringe visibility doesn't change too much, but I'm not confident on the normalization factor.

image

Similarly, the null depth has increased slightly for the polarized case (different) image

Jashcraf commented 2 years ago

Repeated the experiment for +/- X points in the pupil image

image

It doesn't actually seem significantly different, though we note that the null depth is slightly worse image

Jashcraf commented 2 years ago

It could be useful to test the actual baselines GLINT is probing - maybe poke Barnaby about this?