Sheila, Jennie W, Ryan S

Summary: The camera servos got turned off accidentally last time we moved PR3. Worth another try at this measurement

Analysis of why we lost lock the other day while doing the PR2 spot move in lock by moving PR3 yaw alignment and pico-ing to stay on the POP and POPAIR PDs, see image.

When we first started alktering the yaw of PR3 at the first cursor the circulating power started to get higher in the arms and around 17:22:02 UTC the circulating power began to go down as did LSC-POP_A. About 30 mins after this the circulating power began to recover as we stopped changing PR3 position and the pic-motor position. We are not sure why this happened. After this preiod we started moving PR3 yaw down again and the circulating power and POP-A power decresed and then we lost lock.

Over this period when wer were not actively chnaging the alignment, PR2 was still moving. So we checked the camera servos to see if they move PR2 (they don't) but we discovered that the camera serrvos were switched off by the camera guardian, see image.

We realised that because the PR2_SPOT_MOVE guardian state that we had ISC-LOCK in is less than 577 which tripped this condition in the CAMERA_SERVO guardian.

The CAMERA-SERVO guardian went to state 500 as shown in the  ndscope final row at the first cursor. This guardian node then stalled here as the PR2_SPOT_MOVE state does not contain a call to the unstall nodes function in ISC_LOCK, instead of switching on ADS servos and then trying to get back to the CAMERA_SERVO_ON state as in its state graph.

We altered the CAMERA SERVO guardian to eliminate the turning off of camera servo if it thinks the IFO is unlocked (ie. in a low number state) as this should be handled by ISC_LOCK which manages it.

Still need to think about why our overall circulating power got better then worse several times during these changes and why precisely we lost lock.