TITLE: 11/21 Eve Shift: 0030-0600 UTC (1630-2200 PST), all times posted in UTC
STATE of H1: Planned Engineering
INCOMING OPERATOR: None
SHIFT SUMMARY: PEM measurements were going well until a lockloss @ 01:27 UTC; PEM team was out in the LVEA at the time but were not doing anything particularly noisy. Ran an initial alignment then started relocking, but had a lockloss for some reason in LOWNOISE_ESD_ETMY; this didn't happen on the second relock attempt.
After relocking, I started testing Sheila's SQZ angle stepper script to make sure it's okay to run automatically tonight, but I ran into a few issues, so at this point it is not set to run overnight. After the last test run I did of the script, I accidentally just reverted the ZM4 PSAMs voltage from 15.2 back to its nominal 6.7 (I know now this is MUCH too quick of a change and could have caused a lockloss), so squeezing doesn't look great. Even still, H1 is observing and is set to relock automatically as usual overnight.
LOG:
The squeezing angle stepper script is now in /ligo/gitcommon/squeezing/sqzutils/automate_dataset/
I've set it to go taking the longer data sets overnight tonight. I modified the script this morning to check that the IFO is locked and thermalized before it starts, which is working with some debugging from Ryan and TJ during breaks in the PEM work today. It doesn't handle locklosses part way through the data set well, there are places where it will wait for the squeezing manager to go to the requested state and get hung up there if the IFO is unlocked.
I've set up the first script (SQZ_ANG_stepper.py) to start with no squeezing time, and scan the demod phase for FDS with nominal psams(6.2V ZM4, -0.4V ZM5), then move ZM4 to 4 V and scan with FIS, this is going now.
I've also set up a second script to start again at 1:30 pacific, this should only start if we are locked for at least an hour at that time. This one will scan FIS for two different values of ZM5 psams, and if all goes well will finish around 4 am.