The ASC REFL sensors at both Livingston and Hanford have usually shown a strong 6 Hz peak that comes from a suspension resonance of the RM blades transducing vertial HAM1 motion into pitch motion.  At Livingston they recently (LLO:59677) locked down their RM blades to remove this resonance. This not only eliminated the resonance, but also gave a broadband reduction in the coupling of vertical HAM1 motion to WFS pitch motion, which could mean a broadband reduction in cHard noise.

The purpose of this alog is to document that a similar noise situation exists at Hanford, and RM blade locking should be considered for the next HAM1 incursion. The attached plots compare Livingston and Hanford's cHard pitch error signals (calibrated roughly into watts), the appearance of the 6 Hz feature in DARM, and the pitch damping error signals of the RMs (interferometer unlocked). The 6 Hz feature produces similar rms in cHard for both interferometers, as well as similar rms in DARM. The rms seen in the suspension local damping is similar for both RM1s, but Hanford's RM2 has about 10x more rms than Livingston's.

It would be interesting to measure the transfer function of HAM1 vertical motion to WFS pitch as currently configured, to compare to the Livingston measurement.

Calibration note: cHard error signals were calibrated by matching the ac noise level to the predicted shot noise estimated by the dc power on the ASC REFL diodes. each ASC REFL diode (summed over all quadrants) gets about 40% of the light as the LSC REFL diode, and during these locks Hanford’s LSC REFL A had about 19 mW while Livingston’s had about 27 mW (see LHO:44947 and LLO:41503 back from 2018). This is also roughly consistent with Elenna’s recent ASC calibrations (LHO:62976).