gabriele.vajente@LIGO.ORG - posted 11:42, Friday 04 August 2023 - last comment - 15:11, Wednesday 13 September 2023(71961)
MICH and SRCL feedforward measured and fitted
This morning between 9:20am and 10:00am I injected some noise to retune the MICH and SRCL feedforward. New filters have been uploaded with name '8-4-23'. Unfortunately the IFO lost lock before I was ready to test them.
We shoudl wait for the IFO to thermalize after relock, and then test the two new filters
Images attached to this report
Comments related to this report
Tested the new feedforward fits, they work better than the old ones, so we'll leave them running. ISC_LOCK updated and reloaded.
Images attached to this comment
SDF diffs Accepted
Images attached to this comment
Interestingly, retuning the FF reduced thee 52 Hz peak in DARM.
Images attached to this comment
Much later, we've identified that this routine filter update was informed by a measurement of the IFO incorrectly taken while calibration lines were still on. This causes the measurement fitter to create a filter that tries to "compensate" for the high-Q feature with some equally high-Q zero:pole pairs LHO:72537. Once installed, the impulse response of this zero:pole pair causes hours-long ring-ups in the IFO sensitivity right at the calibration line frequency as the two features mix LHO:72064. The procedure for taking LSC FF measurements has been rectified now (see LHO:72572), to explicitly call out that calibration lines MUST be turned OFF if you're gathering a set of measurements to be used for FF filter creation.