patrick.thomas@LIGO.ORG - posted 18:17, Sunday 28 April 2019 - last comment - 21:14, Sunday 28 April 2019(48834)
Can not get SRC to stay locked
Currently on "Align SRC" in initial alignment sequence. No issues with previous steps. The ALIGN_IFO guardian had gotten to SRC_ALIGN and stayed locked there earlier, but ASC-SRC1_Y would not converge. Tried misaligning SRM and centering spot on AS-AS_C photodetector medm with SR2. Tried aligning SRM to make H1:ASC-AS_A_DC_NSUM_OUT16 flashes high. Tried trending back witness sensors. Now it has trouble getting to SRC_ALIGN.
Patrick, Sheila
It looks like this particular problem was that the outputs of th elock filter banks in the SRM were turned off. The suspension guardian turns off these output when the watchdogs trip (or in INIT) and they get turned back on as the guardian moves through the path to "ALIGNED".
Looking through the log, it seems that for some reason the guardian state was stck in RESET for about 20 minutes after the last watchdog trip filling the logs. (As an aside, it would be nice to rewrite that state so it doesn't fill up the log). I'm not sure why it would be stuck in RESET, possibly it was stalled after the trip. This situation is normally resolved by re-requesting "ALIGNED" or "MISALIGNED".
It looks like someone went to INIT, and then set the guardian into MANUAL mode (screen turns pink), and went to MASTERSWITCH on. Apparently this skips steps that would turn on the damping loops, which is why Keita and Niko had to turn those on manually, and also the outputs of the lock filters which is why Patrick ran into difficulties locking SRY.
This is one of the reasons to be careful about using MANUAL mode. It shouldn't be necessary in the normal locking sequence or to recover from a trip,