All day today the 2-minute-period oscillation in the yaw ASC loops persisted. We tried turning off the ADS loops, this had no effect. We tried lowering the gain of the SRC1 loop, this did nothing.

We decided to re-phase sensor for the beamsplitter ASC (AS_A_RF36_Q), since:

• it was clearly affected by the SR3 heater change (looking at the I-phase during the heating)
• Keita looked at the spectrum of the I and Q, pit and yaw outputs at frequencies where the beamsplitter had real motion (known by looking at the oplev) and saw the the phasing wasn't good.

I added ~25 degrees to each quadrant, (now all around 225 degrees). This stopped our ASC oscillation, and increased our 9MHz build up (POPAIR_B_RF18) freom ~42 to ~45. There might be more improvements to be had by turning up the phase further, I just stopped here since the rest of the ASC was taking a while to catch up.