louis.dartez@LIGO.ORG - posted 18:14, Monday 17 August 2026 - last comment - 10:47, Tuesday 18 August 2026(91576)
trying to get PRMI
E. Capote, L. Dartez, K. Kawabe, C. Compton Today we've tried to get PRMI locked without success. Here are some brief notes about what we tried since both Elenna and I have to go for the day. This morning Elenna ran a full initial alignment, checked PRX and MICH BRIGHT, and aligned for PRMI. She noted that the flashes we were getting were very low and PRMI wouldn't lock. We checked the input alignment and noticed that the alignment that Keita and I set up in 91447 were kicked sometime early last week. It wasn't immediately clear to us why. We checked for drift in the MCs and IMs then reset the IMs to where Keita and I placed them in 91447. We then had to re-find PRX by moving IM4, PR2, and PRM. We made a point to avoid touching ITMX and PR3 since their positions were set when we had the arm during last week's peek. We also checked the BS alignment in MICH to make sure we were in a good place there. We also measured the MICH and PRX OLGs ... they look fine and match the reference measurements in the dtt templates in userapps. We also went over Masayuki's notes from when he got PRMI locked last week (91501) to make sure all of his changes stuck. They did for the most part but for some reason the REFLAIR_A_RF9 gain got reset to 2.5. We fixed that but there might somewhere in the guardian code that needs to be fixed to keep the 3.5 gain. We continue to struggle to lock PRMI as it doesn't seem to trigger for long enough to maintain the lock. Some more things that we'll try tomorrow: 1.) trend and compare today's flashes against last week's. 2.) re-check the POP clipping situation to determine whether we made things worse. 3.) go out to ISCT1 to check the REFLAIR diode and make sure it's still well aligned. 4.) run POPX centering. Elenna and/or I will circle back to provide some plots etc here later on as we both have to run now. We are handing things off to the JAC and SPI teams for evening/overnight tests.
Adding more:
Therefore, we find ourselves in a situation where it's likely that the input alignment work done last week with IM4 and PR2 will need to be adjusted to accommodate the new IM1,2,3 alignment.
Keita suggested these steps:
Louis and I tried a few times to work through this process, but we found that even while we continued to improve the alignment, PRMI would not lock. I will add that before trying to relock PRMI, I would lock MICH to ensure that the beamsplitter alignment was good.
However, we are getting POPAIR B RF18 flashes that are roughly 40-50 counts on the fast channel. I have in mind that they could be higher, but I do think these flashes are large enough that we should be able to lock. We are definitely triggering, and even got a few very short locks. Also, the AS AIR camera alignment looks decent.
This has left me wondering if there is something else wrong, such as phasing or gains that are making it not possible to grab and hold lock. However, the team was able to lock PRMI last week, and there were no issues reported.
Here are some checks:
At this point, we looked at Masayuki's logs from last week to understand what might have been changed. Louis notes some issues or confusion we found above. Namely, I want to point out that we expect our MICH OLG to have a 5 Hz UGF, but masayuki says that is too low in his alog. We also noticed that some of his changes are not taking, such as the matrix value for reflair 9.
The only explanation I can think of at this point is that now we have a new input alignment between IMs1-3, so we need a new input alignment on M4 and PR2 that might be exacerbating the clipping issue on POP that was noticed last week, 91549. This is just a guess though, which is why Louis and I will start trending the POP diodes.