B Non-parallel double slit experiment

[Devin-M]

Is it possible to tell which slit some of the light went through when the slits aren’t parallel (green & red boxes in option 3)?

I used a Bahtinov mask to cover the aperture of my 600mm f/9 lens while observing the star Polaris. Various open slits were tested while others were covered with masking tape. -
All exposures were 5 minutes with a 656nm narrowband filter on a Nikon D800 @ 6400iso. Only option 4 appears to show a double slit interference pattern while options 1, 2 & 3 appear to show a single slit interference pattern.

RAW Files: https://u.pcloud.link/publink/show?code=kZcjE2VZR1eBfOCgGAYfuGOv9EMgPH3KIR07

A bit of prior discussion…
Post in thread 'Diffraction Effects and Artifacts in Telescopes like the JWST'
https://www.physicsforums.com/threa...telescopes-like-the-jwst.1047305/post-6823238

 

[Devin-M]

I did find this which seems related… the “Atlas of Optical Transforms”

https://iubemcenter.indiana.edu/doc/atlas-of-optical-transforms.pdf

 

[Drakkith]

Is it possible to tell which slit some of the light went through when the slits aren’t parallel

I was going to say what I said in the thread you linked to, that you can't tell which slit the light went through. But after thinking about the problem more, I have to confess that I don't actually know. I'm assuming the more your rotate the slits relative to each other, the less correlation their patterns have, as image #3 appears to show.

 

[vanhees71]

The point is that if you make it possible to observe, through which slit a photon comes with certainty, you completely loose the interference pattern. On the other hand, if you want an interference pattern with the best contrast you can get you need perfect ignorance of the which-way information. You can of course have everything between these two extremes, i.e., you can have some probabilistic information through which slit the photon came (i.e., with some probability larger than 1/2 a photon came through one of the slits instead of the other) at the cost of lower contrast of the interference pattern and vice versa. This, of course, is a typical wave phenomenon applying to both classical waves as well as single quanta (photons) of the quantized field.

 

[Devin-M]

On the other hand, if you want an interference pattern with the best contrast you can get you need perfect ignorance of the which-way information.

The picture below seems a strange case… on the one hand, for the circled light (vertical spikes) we see a strong interference pattern and don’t know exactly which slit the light went through, on the other hand it seems like we also can deduce with high probability the circled light didn’t go through the diagonal slits…