First attachment shows that the ISS array PDs suddenly lost the beam (bottom left) right after the IM1 LL OSEM jumped (top left) on Friday, meaning that the problem was not just the OSEM sensing, somehow the alignment was affected by this.
This morning Rahul physically inspected IM1 and checked cables and electronics but nothing was wrong. He pulled the LL OSEM and nothing was wrong either. He put it back on, centered the OSEM depth (only for LL) and everything looked normal.
We locked JAC and suddenly we got the flashes on ISS array back, see the second attachment.
Though this is good, it's also frustrating that we don't understand what happened. My theory is that somehow something (OSEM bobbin thing?) was charged and pulled the closest metal electrostatically, changing the DC alignment.
We'll see if we can center REFL ASC sensors as well as ISS QPD. If we can, we'll proceed to close out.
FYI, following is the IM1 OSEMINF before the jump, right after the jump and now. UR experienced 500 counts jump and clearly got back after Rahul's job. Changes for UL and LR were much smaller but they got closer to the pre-jump position, too. Don't pay much attention to the 1000 counts difference for LL now VS pre-jump because it was reset.
| Before the jump on Friday | After the jump on Friday | After Rahul reset the OSEM | |
| UL | 16647 | 16577 | 16695 |
| LL | 9032 | 2492 | 10709 |
| UR | 14790 | 15292 | 14798 |
| LR | 7154 | 7022 | 7169 |
To investigate the issue on IM1 (LL AOSEM inmons dropped to 2000 counts from 9k approximately) I did the following things,
1. In HAM2 chamber, at first I thoroughly inspected IM1 for any signs of rubbing - found none and the pendulum looked free. The suspension was set to SAFE state during this time.
2. Unplugged LL AOSEM and connected healthy UL AOSEM cable to check for inmons - was still reading 2k. Hence, this ruled out any issue with LL in vacuum cable.
3. Took the AOSEM out of the suspension, visually inspected the LED and PD - looked all healthy. I also inspected the magnet on the optics for any damage - found none, everything looked fine.
4. As a backup I had spare AOSEM in hand for replacement.
5. I took open light counts of the LL AOSEM (now that it was plugged out of IM1) - the reading shows 19k counts which was good. This meant that there was nothing wrong with the LL AOSEM or the electronics chain.
6. I plugged that LL AOSEM back into IM1 and centered the flag/magnet to around 10k counts (50% open light).
7. Took chamber side transfer function measurements - all three dof. looked healthy, peaks and magnitude were good. The coherence was not so great due to the noisy environment.
Then I declared that IM1 is healthy and back in action.
What could have happened? - I think the flag magnet on LL could have been faintly touching or stuck and then set free, thus leading to a sudden drop in the inmon values. The second theory is that the AOSEM itself moved very slightly - its only held by a single set screw and does not have rails for moving them in and out.