Since we've gone down in power, Anamaria reminded me that we ought to check and tune our LSC input matrix that subtracts the PRCL contribution from SRCL.  Sheila reminded me that first I should check and tune the POP phasing, and then do the input matrix tuning.

POP phasing

• Turned on the Not88s in PRCL2, MICH2, SRCL2.  Used awggui to inject into PRCL1_EXC at 88 Hz with A=0.003 and gain=300.
• POP45 looked great (plenty of signal in I, nothing to be seen in Q).
• POP9 needed rephasing.  Started at -16.25 deg, changed to new nominal of -28.25 deg.
• First attachment, references are as-found, the 'current' traces are the new nominal.

LSC input matrix for PRCL / SRCL subtraction

• Turned off the Not88s in PRCL2, MICH2, SRCL2.  Used awggui to inject into PRCL1_EXC with butter("BandPass",4,5,100)zpk([],[10],1,"n"), with A=0.1 and gain=15000 (fifteen thousand).
• In the past, we've done this with single lines.  However, with inspiration from LLO (LLO alog 65830) I did a broader band injection this time.  I started with the bandpass that we use for PRCL OLG measurements, then added a pole, and turned it up until I could see it clearly in DARM.
• Started with the POP_A_RF9_I -> SRCL LSC input matrix element at 0.095.  Better at 0.06.
• Second attachment, yellow is DARM with no PRCL injection, gray is with that PRCL injection and the original input matrix element, red is with the new nominal.