sheila.dwyer@LIGO.ORG - posted 14:36, Thursday 16 June 2016 - last comment - 21:33, Thursday 16 June 2016(27778)
Isolating BS stage 2 is causing locklosses
Jim, Sheila, Haocun,
Turning on the BS stage2 isolation loops sometimes causes locklosses. These are the locklosses for which Hoacun found that there is a glitch in the BS side osem. (27739)
The first attached plot shows the Z isolation loop gain coming on at t=-34 seconds, Jim tells me this comes on with a 10 second ramp, and you can see a small glitch in the Z CPS at around t=-24 seconds when the ramp should be finishing. At just after t=-8 the gain for X, Y, and RZ isolation loops increase, which corresponds to a glitch visible in the CPS, GS13s, and top mass side osem, and optical lever. This is the glitch that breaks the lock.
For now I've rearranged the guardian graph so we won't be isolating the beamsplitter stage 2.
Images attached to this report
Comments related to this report
Weird. Please check that the isolation gain is going from 0 to 1 with 2 ramps. The first goes from 0 to (intermediate gain) and the second goes from (intermediate gain) to 1. the intermediate gain is ~ 0.01 (so that the open loop gain at DC is ~1, so that the ERROR SIGNAL of the loop goes to about 1/2 of its final value. both ramps are 5-10 seconds. It looks like the second ramp-time has dropped to 0 (immediate) which would give the system a hearty whack. Please check that there are 2 ramps to get the gain to 1.