On 3rd July we performed a test in order to investigate the possibility of moving optics on SIB1 (e.g., mode-matching lens, IPC2, mirrors for beam alignment) and avoid to unlock the ITF or the RFC during the operation. We observed that there are strong oscillations on the Mode-Cleaner end mirror corrections that kill the RFC lock. The frequency of these oscillations vary according to the object that is moving.

For that matter, P. Ruggi came up with a new low-pass filter to avoid that a high-frequency oscillation get in the actuation band: in Fig. 1 we can see the usual filter in blue and the new LP filter in violet.

So, with this new filter we moved the L1 lens and IPC2 and observed the effect on the FFT signals. Fig. 2 shows the GPS of the movements performed.

• Fig. 3 shows the spectra of the interested signals while moving L1. What could be observed is that the movement possibly triggered something not only mechanical but also some electrical contamination, as can be noted by all the 100 Hz harmonic oscillations;
• Fig. 4 shows the spectra of the same signals while moving IPC2. In this case, it seems that a pure mechanical oscillation is produced at 665.6 Hz (and harmonics).

Therefore, for the second case, we decided to try the standard RFC filter with a notch at the oscillation frequency of IPC2 (and a few harmonics) in order to verify if this solution could avoid the RFC lock loss. As can be seen in Fig. 5, the INJ system and in particular the RFC remained locked. Although this solution actually worked, the MC correction signal was very high, and it is possible that the ITF won't be able to remain locked in this conditions since these oscillations are coupled to the laser frequency. So, this test should be repeated with a full locked ITF.