Looking at the SRCL signals (error/control signals) from last night's low noise lock, there is a large narrow line at 1.09 Hz. This feature is known to be in the HAM3 ISI motion, and the HAM-ISI spectra (SUSPOINT) from the same time shows the line in HAM3, and NOT in HAM4 or HAM5. The signal is thus getting into SRCL via PRCL sensing, and imperfect decoupling of PRCL from SRCL in the LSC input matrix (even this matrix element was tuned earlier in the night -- see Stefan's entry). The calibrated SRCL control signal shows that SRCL is being moved at 1.1 Hz by about 200 nm pk-pk. In other words, since the 1.1 Hz feature is not a true SRCL motion, we're dithering SRCL length by 200 nm -- doesn't seem like a good idea. Jim W is working on figuring out where the HAM3 1.1 Hz is coming from. The other question is why the PRCL -> SRCL decoupling is not stable.