Jason, Kiwamu (WP 5199)
PR3 oplev is now back to functional.
As reported in the last week (alog 18246), PR3 was found to be not functional. This morning we took a look at the PR3 oplev and found that the beam had been clipped. We fixed it by steering the launcher telescope and recalibrtated both pit and yaw signals.
[Unclipping]
Looking at the returning beam, we found that the beam was largely clipped on the west side and its bottom side. Apparently this was causing the nonideal cross-coupled behavior as reported in alog 18246. We could not identify what object was occulting the returning beam, but according to Jason, it could be somewhere around the viewport for the return beam. We decided to steer the launcher telescope to unclip the beam. The unclipping went smooth and no major issue was found. Since we moved the beam toward the East, we had to recenter the QPD stage afterwards. The beam is now about 1 cm away from the clipping edge, meaning we can move PR3 by several 100 urads until clipped. This is good enough.
After the unclipping, Jason checked the beam shape and confirmed that there was no sign of clipping. We also briefly checked the responsivity of each segment by placing the beam on each segment by steering PR3. This convinced us that the responsivity is almost the same among the four segments. We physicallyt locked the launcher telescope as a part of the normal procedure so that it is not going to move.
[Recalibration]
After the unclipping, I recalibrated the oplev. Here is a summary of the new calibration:
| old calibration [urad/cnts] | new calibration [urad/cnts] | |
| PR3 pit | 38.86 | 26.90 |
| PR3 yaw | 38.53 | 29.60 |
I also updated the SDF accordingly. The data and fitting code are attached as a zip file.
One thing I found during the calibration is that there is still some cross-coupling between pitch and yaw at 10% level. I don't think this is a big issue, but those who use PR3 oplev should be aware of it. I am not sure what part of it is causing the cross-coupling.
