Displaying report 1-1 of 1.
Reports until 15:09, Tuesday 07 March 2023
H1 ISC
jenne.driggers@LIGO.ORG - posted 15:09, Tuesday 07 March 2023 - last comment - 16:22, Tuesday 07 March 2023(67834)
Camera servo lookback time for offset-setting reduced

We tried today to engage the camera servo after the ADS had converged.  However the camera copy guardian had only just started working (some camera server work was still ongoing from maintenance day).  I didn't realize that the lookback time for averaging to determine the current camera position was set to be 10 minutes (!), so we averaged in a lot of garbage time when setting the offset, so when the camera servos engaged, it dragged the IFO way out of alignment.  We weren't able to catch it and turn off the servos before we lost lock.

I have now reduced the lookback time to be 10 seconds (which should already be more than we need), so hopefully we won't get caught by this again (although camera server upgrades are also hopefully going to be rare).

Comments related to this report
jenne.driggers@LIGO.ORG - 16:22, Tuesday 07 March 2023 (67837)

This worked fine using a shorter averaging time. 

However, ETMY camera had frozen, and the CAMERA_SERVO guardian does not check effectively if it's stuck.  TJ is going to help me figure out how to implement that checking.  For now, if folks want to use the camera servo, they should watch the camera servos closely, and if it looks like they've gone funny and / or one of the cameras is frozen, then the camera servos should be turned off.

Displaying report 1-1 of 1.