TITLE: 07/25 Owl Shift: 07:00-15:00 UTC (00:00-08:00 PST), all times posted in UTC
STATE of H1: Preventive Maintenance
INCOMING OPERATOR: TJ
SHIFT SUMMARY: After the initial lockloss of the shift, I have been unable to get past REDUCE_RF45_MODULATION_DEPTH without OMC DCPD saturations. After the first couple of locklosses at this point, I went stepwise through each guardian state from DC_READOUT_TRANSITION. While watching the ISC_LOCK log at REDUCE_RF45_MODULATION_DEPTH, it seems that the CHARD P loop never really converges. It vacillates between converged and waiting for convergence. If you select any higher guardian state, it will immediately move on when it registers converged and OMC DCPD saturation alarms start. SRC1_Y error signal on the StripTool starts running away from zero once the DCPD saturations start. Performed IA after 3rd lockloss during relocking; did not help. ETMy violin mode 5 has been an unresponsive nuisance during this saga, but I can't say for sure that it is entirely responsible.
On the bright side, PCAL X RX PD seems to have fixed itself at the lockloss from the step at another lockloss reported in alog 37726. See attached screenshot.
LOG:
Prior to 13:20, see Shift Summary above.
13:20 Called Keita for help
14:25 Lockloss while Keita was working to remove OMC whitening. We decide to forego trying to relock since maintenance day is starting in 35 minutes.
14:51 JeffB to diode room, then to cleaning area
14:59 JeffK starting TFs on BS and PR3
To operators:
The reason why I wanted to remove whitening was because OMC DCPDs were railing for whatever reason (violin?). As soon as DCPD starts constantly railing it becomes hard to damp violins and PI (Travis had a hard time taking care of PI).
I tried to manually punch in 0 to PDB gain and 2000 to PDA gain quickly enough but it took me longer than I wanted, IFO didn't like it.
I could have written a script to do it quicker, but JeffK later showed me that you can do it in the guardian.
Open OMC_LOCK, then ALL, then MANUAL, then REMOVE_WHITENING (or ADD_WHITENING), and get back to MANAGED.
Another note (July14th) for REMOVE_WHITENING steps when violins rung up (with note about Manager of OMC_LOCK node): here.