Since the relock of this morning around 3:30 UTC, a huge line or bump around 150 Hz is present in h(t) and fluctuates a lot, producing thus large SNR glitches around 150 Hz. Plot1 shows the omicron triggers time-frequency distribution, where can be seen those 150 Hz glitches after the relock and the usual low frequency glitches due to bad weather before the unlock around 1:50 UTC. Plot2 shows the spectrogram of h(t) over one day where can be seen the 150 Hz fluctuating line but also the large frequency band glitches that are at the origin of the BNS range dips visible in plot 3 and mentioned in elog 45870
Fig 2 shows a zoom on one of the glitches, which shows that for whatever reason, there is a glitch on the local control signals of the BS (which are in loop if the full-bandwidth AA is switched off). A short moment later, this causes a drop in the sideband power and a brightening of the dark fringe, which is then followed by a glitch in DARM. To be understood by the experts if this instability is somehow caused by the bad working point of the interferometer (which prevents the AA loop to be closed in the first place), or that this is a problem with the local controls that can be resolved.
Plot1 shows that, during the day of 23th May 2019, the high SNR glitches around 150 Hz were present only between 00:00 and 08:00 UTC and between 18:00 and 21:00 UTC.
Plot2 shows the spectrograms of LSC_DARM where we can see larger fluctuations around 150 Hz during those two time periods.
Plot3 shows that the alignment control is noisier during those two time periods and that BS alignment is worse when those glitches are visible.
So, maybe BS alignment is at the origin of both the BNS range large drops AND the 150 Hz glitches.
