camilla.compton@LIGO.ORG - posted 11:22, Sunday 15 March 2020 - last comment - 15:09, Sunday 22 March 2020(55601)
H1 Lockloss 17:53:18 UTC
Lost lock, after 29h30m lock. No obvious reason why. Environmental conditions are all fine.
No green light on the y-arm yet so it will be interesting to see how increase flashes copes with that, I may have to intervene.
We saw this before after a long 58 hour lock with dramatic temperature change (alog 54580). Though I'm not sure we can blame temperature here as the outside temurature has only changed +3°C since the last lock. -16°C change from a few days go it if takes the equipment a long time to change temperature.
Had to move both ETMY and TMSY a decent way to get light onto the camera (20 counts on the slider for ETMX Pit and 15 for TMSY Pit).
Had dragged TMSY too far away from it's normal position so that we were not getting enough light on the QPD's and had ALS_Y in fault with message "PZT trig servo not on". I think this is a different problem to the one disrupted by TJ in alog alog 54526 but it may be related.
Keita helped me to get light back on the QPDs:
From sitemap>ALS> End Y overview > QPD_A (FE) and QPD_B (FE), the NSUM should be ~1 for green arms locking, was closer to 0.05. after deciding that the suspensions had bee moved too much, we the QPD NSUMs and the alignment sliders and reverted them to their values at the time of last green arms locking, then did the same for the BIAS on the PZT1 and PZT2 PIT and YAW filters. NSUMS then came up to 1. Rs-adjusted the PZT filter bias numbers to be closer to their outputs and then continued locking.
We have just lost PRMI lock (not total lock) 3 times after I noticed that the FIND_IR TR_Y trace was way higher than it should be attached photo (20 rather than 1). Something strange must have happened so I will try and initial alignment.
DRMI locked straight away after a smooth initial alignment. Unsure what went on before!
At the start of this locking acquisition, there was no light on ALS-Y_QPD_B (NSUM ~ 0.05). ALSY PZT1 and PZT2 were then not adjusted. It can be seen from the top 2 plots of the attached image that they were being significantly adjusted at the start of each green arms locking to maximize QPD light. Keita helped me move the PZT1 and PZT2 values closer to the values which maximized NSUM in the past so in future they should not be so far away.
It's still not clear why there was no light on QPD_B to start with. The values of PZT1 and PZT2 do drift over time to give the maximum light on the QPDs so it's possible that to was just out of range. Maybe these biases should be checked and adjusted more regularly. Tagging ISC