[Adam, Keiko]
Just a quick post as we have no time today:
We think the P and Y mixing at the TMSX QPDs is because of a too small beam size at the QPDs. From my past experiences of WFS, smaller beam size than 300um doesn't work at least for WFSs (3mm diameter quadrant diode). We think something similar for DC QPDs (1cm square quadrant diode). We did quick tests to check the mixing ratio at a small and large beam spots, and confirmed the PY mixing ratio got better with a bigger beam. Not sure why TMSY or green QPDs work ok with the same design though. For TMSX, we will re-design a better telescope.
More detail will be posted tomorrow afternoon.
Thanks so much for your efforts!
Actually, as long as I remembered, I rememberd some trials to tweak the location of lens-2 (the smaller lens in the GPT) for TMSY, by taking several days (actually mid-nights!) to measure the beam profile evolutions several times and re-design the locations of the QPDs according to the actual measured beam profiles. So those "designs" at X- and Y-ends might not be said "the same"... Maybe the same thing can be said for both of Gr GPTs.
About those DC Goupy phase telescopes (DC GPTs),
- For TMSX Green, I could not find the log on meauring the beam profiles at Gouy phase telescope (Gr GPT). 6477 seems the day though. Does someone remember where the profile data is?
- For TMSX IR, some efforts at the very initial stage are found at 7502, 7527, and 7574.
- For TMSY Green, some trials are found in 10272, 10300, 10302, and 10316.
- For TMSY IR, they are 10321, 10329, 10337, and 10365.; as shown there, especially the beam profiles of the IR light at TMSY were not so good, but it seems working well as a TMS. (And that's why I was surprised when I heard that such strange couplings were observed at X but not Y.)
Activity logs in terms of TMSX or TMSY can be found here and here as references (but I know such dialies would be difficult for the others...)
[Adam, Keiko]
Report of Dec 2.
TMS QPDs mix PIT and YAW unwantedly, and the ratio of mixing depends on the beam size (or the Gouy phase position). See the attached figures. Each left plot ("Original position") is the QPD signals at their initial positions of QPDs with ITMX excited at 0.1Hz, 150 cts at MN. The beam size is something like 200-300um (which is pretty small w.r.t the diode size). Each right plot is where the beam is much larger (but somewhat random positions). You can see, for example, QPDA2 has 1:1 mixing between P and Y at the initial posiion, where as approx 1:0.1 with a large beam.
The beam shape was also checked by the beam profiler CCD. The shape look OK, a bit eliptic, before each QPDs (see, the attached screenshot). Some distortion were seen upstream of the QPDs, near green beam filter. We are not sure this distortion is critial or not.
Thanks so much for your great efforts. If the beam size would be matter, moving location of the 1-inch lens on the table (it is already set on a translation stage) can help very much, as I believe the beam waist will dramatically change w. r. t. the lens's location. (But I also know that doing so you need to do several set of beam profiling as I did at TMS"Y"). Depending on the priority, I would be able to come to the site next week to do so...
Thank you for the suggestion. We can try to move the 1-inch second lens to move this week, while looking for some other options.
I believe the priority is very high, as we want to use them for the arms' SOFT mode sensors.