sheila.dwyer@LIGO.ORG - posted 15:44, Monday 28 October 2019 (52733)
locking today
- Added 20dB of gain to the MICH ASC loops in PRMI.
- Since Jenne reported trouble with the CARM offset reduction yesterday, I tried to do the sequence the way it was before we modified it for the low recycling gain (REDUCE_CARM_OFFSET sets TR_CARM to -40, and then use the CARM 5 picometers state to calculate an offset st which we make the switch to REFL. This worked once so I've set the guardian to do it next time.
- Injecting squeezing broke the IFO lock, this is the first time I've seen this since we switched away from the OPO reference squeezer control scheme. This lockloss is 1256308786, the attached screenshot shows that the LO loop seems to break lock and saturate in the seconds before the lockloss. (Yesterday the squeezer fiber distribution AOM drive was changed to a fixed frequency 52728, which requires us to update the guardian so that the code no longer tried to adjust the squeezer VCO frequency before attempting to close the LO loop. Since this hadn't been loaded this morning, the squeezer laser was probably very far off in frequency when we injected it and as the LO loop tried to lock we might have crossed a frequency that disturbed the IFO.)
- In the next lock Robert and I did a few measurements with the POP beam diverter. The pop beam diverter was closed from 15:39:03 UTC to 15:46:01 UTC.
- I ran some calibration sweeps during and just after the morning meeting. We got the sensing function, L3 and L2 sweeps, and we nearly got the L1 sweep but the interferometer lost lock for unknown reasons part way through this. Timesh has offered to start processing these sweeps.
- We weren't able to relock because the COMM beatnote was low, so Niko and TJ started an initial alignment. I moved the IM3 yaw slider back to 4870, the value it was at before the vent.
- Spent some time thinking about the X arm/PR3 alignment, here is a summary of the story:
- Looking back at the last locks of O3a (second attachment), the comm beatnote was around -4dBm. In those locks the ITMX alignment changes by 3 to 4 microrad when the spots get centered, meaning that our inital alignment references at the end of O3a weren't actually set well for centering the beam on the optics. We came into lock in September with the reccling gain monitor reading around 46 before any ASC was engaged.
- We believe that we lost the ITMX green camera reference durring the vent.
- We reset this once when we had partially recovered good recycling gain, but not yet been able to center the spots on the optics. After that reset of the camera and green QPD references, we had a good COMM beatnote. 52692
- After we pico'd AS_C we were able to close SRC ASC, and that allowed us to close the ADS and center the spots on the optics, we then reset the green camera and QPD offsets for this alignment. 52700 Wit these alignment settings we have been coming into resonance with a good recycling gain, around 47 according to our monitor. This change also resulted in a lower comm beatnote, which caused some difficulty with locking for Jenne and Varun yesterday (52725), they moved PR3 to get that back.
- We thought that PR3 had not been moved during the vent, but now that I am looking at the trend I see that it was actually moved a few microradians, so I've reverted both PR3 sliders to where they were in September. Bringing PR3 back to its September slider values restored the COMM beatnote to -5 to -6dBm, which is good enough for us to relock.
- Patrick is writing a script that can compare our witness sensors now to O3a for different states.
Images attached to this report
