Craig, Danny, Georgia

• Locked with DARM control split between ETMX (L3 and L1) and ETMY (L2), and then manually transitioned L2 darm control to EX once at NLN using Sheila's instructions, this worked. I added a new final step in LOWNOISE_ESD_ETMX guardian which does the L2 transition from ETMY to ETMX. With DARM control on EY and EX we have had few glitches or whistles for most of the day. Towards the end of the evening glitches became more frequent.
• Danny changed the PRCL gain in INCREASE_POWER from 16 to 12, based on Sheila's observations yesterday.
• Dither line move
  • I moved the remaining dither lines back to the 20 Hz band, and saw an improvement in the DARM noise and in the coherence with the DHARD P and Y. Compare blue and red traces in first attachment. The coherence with DSOFT was not improved.
  • In the next lock after moving the dither lines, we saw that the extra 10-80 Hz noise in DARM had returned, even with the dithers at their new frequencies. It turned out the dither loops had not yet converged, and as they did the noise went away. We might benefit from tuning the A2L gains to find an even better spot position?
• Danny and Craig tuned the MICH feedforward, nominal low noise is now back to nominal.
• We had an unexpected fast lockloss from NLN at 02:59:11 UTC. There was a glitch in ETMX L3 ~0.7 seconds before the lockloss, and then some extra noise in the DARM input and all the actuators. Unclear what the cause was. Second attachment shows ndscope of the lockloss, actuator signals on the bottom row.
• We were planning on adding an offset to the OMC PZT3 (PZT1) and seeing if this reduced the RF9 RIN coupling to DARM. Currently we lock the OMC with 46 V on PZT2, and 10V on PZT1, if we want to explore voltage combinations which reduce the overlap between the carrier and the 9th order 9MHz sideband (eg 50V on PZT1 and 50V on PZT2, shown in the 3rd plot of alog 46667) we need to lock the OMC on the next FSR with the PZT voltage (~90V). I also noted that in order to see broadband coherence between the OMC QPD and DARM we need to go to higher-than-nominal modulation depth, which we already knew based on the coherence in the bottom right plot of attachment 1 of alog 46831.