[Ushiba, Komori]

Abstract:

We successfully improved the DARM sensitivity around 20 Hz by reducing the PR3 pitch coupling noise.

Details:

We implemented the new filter discussed in klog:33504.
Before the trial, we evaluated the PR3 pitch witness signal using the new filter (blue) and compared it with that from the conventional filter (brown), as shown in Fig. 1.
The RMS of the PR3 pitch signal increased due to a larger peak at 0.7 Hz.
The problematic peaks at 17 Hz and 23 Hz remain and are likely caused by oplev noise rather than actual suspension resonances.

We proceeded with the lock acquisition using the new filter and successfully reached the observation state without any issues.
Notably, the DARM sensitivity improved around 20 Hz, particularly at 17 Hz and 23 Hz (red vs. blue in Fig. 2).
The coherence between PR3 pitch and DARM has now almost disappeared, while the coherence with PR2 pitch is larger.

We also observed improvement in the calibrated PRCL signal around 20 Hz (Fig. 3).
However, the calibrated MICH signal became noisier, so we should investigate whether this change is related to the new PR3 filter.

To evaluate stability and lock acquisition performance with the new filter, we will continue using it for the time being.
I noticed that the PRMI 3F locking was failing, likely due to misalignment caused by an earthquake, so I set the IFO guardian to be the down state.

Next steps:

• Improve PR2 pitch control

• Reduce the RMS of PR3 pitch while maintaining similar filter characteristics

• Update PRCL control to reduce RMS further