Sheila, Naoki
We increased the AS72 A/B whitening gain from 12dB to 21dB to reduce the ADC noise. We accepted it in safe.snap as shown in the first attached figure. To compensate the whitening gain increase, we decreased the SRC1 gain from 4 to 1.4 (4/1.4 = 9dB) by changing the following guardian.
line 3429, 3438 in ISC_LOCK guardian, ENGAGE_ASC_FOR_FULL_IFO state
line 1095, 1097 in ISC_DRMI guardian, ENGAGE_DRMI_ASC state
Then we found that the IFO lost lock twice after DRMI ASC is engaged. We found that the dark offset of AS72 was large and that caused too large SRC1 error signal. So we ran the dark offset script in userapps/isc/h1/scripts/dark_offsets/dark_offsets_exe.py. After that, the IFO could go to NLN.
We accepted a bunch of SDFs both in safe.snap and observe.snap as shown in the attached figures.
Since the BNS range was worse and the lock duration was only 1.5 hour with this whitening gain change. We reverted the whitening gain from 21dB to 12dB. We also reverted the whitening filter from 2 stage to 1 stage, which was done in 74231. Then we ran the dark offset script again and accepted a bunch of SDFs in safe.snap as shown in the attached figures. We need to accept them also in observe.snap. We also reverted the SRC1 gain in ISC_LOCK and ISC_DRMI guardian.
I accepted these same SDF diffs in the OBSERVE tables when we relocked. Screeshots attached.
