Takano

As suggested by Ushiba-san in the commissioning strategy meeting, I locked OMC with the single bounce beam at ITMX, without arm locking. FIgure1 shows the comparison of RIN sensitivity with different configuration. The red curve is the sensitivity with the single bounce configuration, the blue curve is that with X arm locked, and the black is the sensitivity as of O4a. The RIN spectra out of- and in-loop are also shown.

Apparently, for single bounce measurement the floor noise which limits the sensitivity of the single arm configuration gets lower or disappear. This strongly supports that this noise floor comes from frequency noise and CARM offset improves the performance of the frequency stabilization. Between

How to solve this issue:

1. Decrease RF AM level by adjusting polarization of the laser injecting to two EOMs
2. Recover (PR)FPMI and improve the suppression level of the frequency noise by better sensing of the frequency fluctuation and common mode rejection at AS port.