Putting the CP ghost beams as close as possible to the B1 beam while remaining on the B1/B5 diaphragm on SDB1 following instructions in https://logbook.virgo-gw.eu/virgo/?r=65427

Figure 1. Starting point, only WI CP ghost beam is visible, as the other ghost beam is on the silicium part of the beam dump which scatters much less

16:49 UTC - starting changing NI and WI CP TY tilt
17:08 UTC - finished the changes

Figure 2. End point of the alignment of the two CPs

There is a significant difference between the two pictures in terms of where are bright patches of light on different parts of SDB1. For example the bright patch on the side of the close loop picomotor of MMT_M2 disappears. This is not just a transient, as the starting point picture looks very similar to the end point picture from end of October. So the position of these light patches is stable over long time scale, but changed completely by changes in CP tilt of ~1mrad.

17:44 UTC - starting moving beam on WE mirror using MIR_Y_SET in order to check the 120Hz bump on B8 DC.

The mis-centered states by 3mm in either direction are not stable, with the interferometer progressively getting in a worse state once the transition on the end mirrors is over. At one point it caused an unlock.

During the relock checked in CARM NULL 1F that adding offsets in MIR_Y_SET doesn't seem to be creating issues, and the 120Hz is visible in that state too.

I will post some figures tomorrow, there seem to be an impact of the beam spot position on the arm cavity mirrors on the 120Hz bump on B8 DC and in h(t).