Naoki, Daniel, Vicky
Today we re-installed the mini-circuits amplifier, a ZHL-500HLN+, onto the CLF RF6 PD, cable ISC_SQ_222. This reverts the change from Jan. 27, 2023 (LHO:67038), when we went to a high CLF configuration to study what the problem of high CLF powers. Now for O4, we want to nominally operate with very low CLF powers, so we have reverted to dumping 95% of the optical power into the fiber as designed on SQZT0 (69457), and today restored this RF amplifier to the RF6 path. I think this now reverts up the major CLF path changes we made to study high-CLF issues prior to starting O4.
With the same 6uW of optical power incident on the RF6 PD, we now have CLF_REFL_RF6_DEMOD_RFMON = -26 (was -47) on a noise floor of -52 (was -55), see the trends here. I think we can now afford to turn down the CLF powers a bit more.
With 6uW on CLF_REFL and this added +21dB RF6 power, we re-measured the CLF transfer function. Here, we measured UGF ~ 2.4 kHz and phase -159, with CLF CMB gains set as: common gain = 15 (was 22), and fast gain = 6 (was 31), consistent with when we previously used this exact amplifer. The CLF path should be back to its nominal low-power configuration now for O4, and we explore lower-power CLF operation when IFO relocks.
To re-lock the OPO on CLF+GREEN dual resonance, now that the amplifier is adding +21dB gain to the CLF RF6 signal, I updated the nominal CLF RF trigger level (H1:SQZ-OPO_IR_RESONANCE_CLF_NOM = 0.005 (up from 0.0009)) to re-lock the OPO on dual resonance, state "LOCKED_CLF_DUAL". This setting is pretty sneaky, but can be found in medms from SQZ7 >> OPO IR (bottom right corner box), see screenshot. The updated CLF RF Nominal value is saved in SDF.
I noticed and fixed this as I relocked the squeezer just before Observing tonight, to pre-empt the OPO PZT (H1:SQZ-OPO_PZT_1_VOLTS) bottoming out in a couple hours (it was at 45V, it loses lock usually below 40V; this PZT is nominally between 70-110V).
