The attached plot shows the typical motion of the BAB center position on the HD cameras in two configurations: retroreflection from FCIM (magenta), retro-reflection from SQB1 (blue). The angular jitter with reflection from filter cavity is mugh larger both because of the double pass in the MM telescope on suspended benches, and due to the relatively larger motion of SQB2 wrt SQB1 (in particular, oscillations on SQB2_LC_TX can reach 10 urad p-p). In the attached plot the horizontal motion of the spot on HD cameras reaches 30% of the beam radius.
The time you choose in the plot is just before I tuned the TXZ loop of SQB2 as reported in logbook entry 53593.
The plot attached shows the RMS and the trend (max/mean/min) of the motion of SQB2 in TX which is much reduced after the tuning, the moment you choosed is just at the beginning of the plot.
It would be good to redo the plot with data after the tuning to see if the reduction is enough or if further tuning should be done.
Indeed with 20x less motion on SQB1 the position of reflected BAB on HD cameras is very stable, see attached plot.
During the weekend I left the BAB alone on the HD cameras to observe the long term angular drifts. In 18 hours, until the FC unlocked, the beam position moved by less than 10% of beam radius horizontally and about 25% of beam radius vertically.
This is clearly visible in the 2 plots attached.
- Plot 1 is a trend data of the motion of SQB2_TX d.o.f where the SBE loop is switched Off in the middle, the noise is clearly decreasing.
- Plot 2 shows a comparison of the spectra of the correction and error signal of SQB2_LC_TX in purple when the SBE loop is ON and in blue when it is OFF. The noise is reduced by a factor 3 at the 0.4 Hz bump and around that frquency (0.3 Hz to 1.5 Hz).
A better tuning of the SBE loop would be profitable with tracking of the NI SAT if necessary.
