In order to search for retroreflections of scattered light that might recombine and cause scattered light noise, I take pictures from the point of view of the beam spot on important optics. The flash on the camera is right next to its aperture so, when the camera is held close to the location of the beam spot, the path of the light from the flash is similar to that of light scattered from the beam spot during observation, and glints in the photograph are indicative of potentiial retroreflections of scattered light. The distribution of light from the flash is, by design, more uniform than that of most scattered light, so more off-axis retroreflections in the photos tend to be less important.

An old photo taken from near the beam spot on PR3 showed a glint from an optic on the POP path so I re-took beam spot photos last week since we are baffling HAM3. The figure shows that I did not find a retroreflection in beam spot photos from PR3, but I did in photos from MC1 and from close to the center of PRM. I took beam spot photos with both my IR camera and cell phone – the figure shows cell phone photos as they are eaisier to interpret, but photos from both cameras show the same issues. I could verify these retroreflections and track them to their source by eye using a head lamp near my eyes. The PRM retroreflection is from the last optic on the POP path and the MC1 retroreflection is from the second to the last optic on the POP path (see diagram on the second page of the figure).

Im a little surprised that there are two of these random alignments, but I guess it is due to how small the angular differences between the paths from HAM3 to HAM2 are and the number of mirrors on the tables. I am also concerned that my beam spot photo technique only picks up retroreflections and not reflections of scattered light from one optic to a different optic. I would also like to make us more aware of these cross talk issues in planning optic paths and suggest that it might be worthwhile to develop code that checked for this.

I don’t know how bad these retroreflections are, but, in my photos, they are brighter than the reflections from ballast masses and the table that we are trying to baffle, so I may try to baffle the PRM retroreflection when we return to baffling. Im not sure how I could baffle the MC1 retroreflection without blocking the POP path.