Exploiting the periods with shutter closed and open acquired during the night I tried to understand if backreflected light from ESQB decreases the ITF sensitivity.

Figure 1: Normalized DARM spectrogram between 0 and 30 Hz. If we observe the frequency band between 10 and 15 Hz we observe the following effects:

• between 18 and 23 the slow shutter was closed. From 23 Fabio started to open and close it every hour. 
• We can see that between 23:00 of the 5 Dec and 6:00 of 6 Dec the noise between 10 and 15 Hz increased when the shutter was open and decreased when the shutter was closed

Figure 2: FFT of hrec between 10 and 100 Hz. Blue lines are Data stream with shutter closed and red lines are Data streams with Shutter Open. (Used FFT parameters: FFT length 10s, overlap 50%, number of averages 200, i.e. 1000s of data for each spectrum). We can see the increase of noise between 10 and 15 Hz.

Figure 3: The average of spectra of Figure 2. Blue shutter closed, red Shuttter Open. In black the target sensitivity of O3 taken from VIM. From the plot we can see:

• between 10 and 15 Hz the sensitivity was spoiled when the SQZ slow shutter was open
• between 25 and 40 Hz is visible a quite negligible increase in noise with the SQZ shutter open.

Summary: the back reflected light from the squeezer spoils the sensitivity between 10 and 15 Hz and between 25 and 40 Hz. Now this effect is not worrying, but this night the weather conditons were good (we are below the target sensitivity thank to the weather) . Maybe, it is better to check the centering of the mirrors in order to avoid problems in the future.