Yesterday with the commioning team, we locked the PRMI but it is very unstable. We localized the problem to the BS the OL signal suggested that the motion of the BS was not suppressed. In particular, it was fluctuating in yaw at a constant ~ 0.16 Hz with a peak-to-peak amplitude of around 2 µrad and theres fast transient at times with a similiar magnitude. This keeps misaligning the interferometer hence making the lock very unstable.
The 0.16 Hz motion comes from the IP YAW and we can see it clearly in the IP signal. Therefore, we increased the Y damping gain by a factor of 5. This seemed to help suppressing the motion at that frequency but then the optic's motion is still too high for a stable lock. I checked that the BS yaw motion has high coherence with the seismic Y motion and the peak is extremely high compared to that of SR2 (Almost 2 order higher). I attached a comparison of the amplitude spectrum. I will try a lower blending frequency for the inertial damping.
Stefan suspects that the BS might be touching somewhere because the fast transients are definitely not harmonic motions. I measured TFs in IM R and TM Y but the resonance peaks stayed where they were except now theres huge amount of Y2L coupling but we still don't know why. I hope to improve the OL quality and so to improve the OL damping performance.