Stefan, Yuta, Kiwamu,
We could lock the sideband-resonant PRMI for more than 8 minutes tonight. The angular fluctuation seemed better because of the L2P(Y) decoupling filters.
The locking is becoming reliable. We still need to investigate what causes the lock loss events.
Here are what we did:
-
We changed the demodulation phase of POPAIR_B_RF by 180 deg such that it goes up when the sidebands are resonant.
-
H1:LSC-POPAIR_B_RF18_PHASE_R = -100
-
We decreased the whitening gains on REFLAIR_RF9 and RF45 to 0 dB to compensate the recent power increment.
-
We reset the threshold for triggering. See the attached screenshot for the detail.
-
We flipped the sign on the BS oplev gain because we had a sign flip in the coil gains recently (see alog 9415).
-
We also flipped the sign of the output matrix for BS for the same reason.
-
H1:LSC-OUTPUT_MTRX_7_2 = 0.05
-
The MICH servo gain was increased from 30 to 150
-
The PRCL gain was also increased from -2 to -6. Plus FM9, +6dB, was engaged. FM3 is newly added as a triggered filter to give some more gains.
-
REFLAIR_RF45_Q had a non-negligible RFAM. We had to add -200 counts in the MICH input to have a nice zero crossing. Note that the Q signal showed 1000 counts pp when it is unlocked..
-
The PRCL UGF was measured to be about 35-ish Hz. See the attached.
-
We didn't get a resonable-looking MICH OLTF for some reason. We may need PRM for the MICH actuator for better decoupling.