For stable recycling cavities we will have multiple suspended benches next to each other with the beam travelling between them. A question is what will be the beam jitter and drifts added by the relative motion of the benches controlled using local controls. We have already a similar configuration with SDB1 and SDB2, and it provide some information of performance we can expect for O5.
Figure 1 show the B1 camera image in LN3. It is a nice round beam, the camer integration time is relatively low, so the image is far from saturating with the highest pixel count at 38 compared to the maximum of 255. That beam is filtered by the OMC, so it shows the mode of the OMC as attached to the SDB1 bench, in some sense it acts as an optical lever beam from SDB1 to SDB2. And the motion of that beam on the camera located on SDB2 shows the relative motion of SDB1 vs SDB2.
Figure 2 shows the the trend of this beam spot position over 2 days, it is +/-5um.
Figure 3, shows the beam and bench motion on a shorter time scale of a few minutes, the beam spot motion has a peak at ~200mHz, and higher motion below 20mHz. This is more clearly visible on the X direction than on the Y direction, as they Y direction seems to have 4 times more sensing noise. Increasing the B1 camera integration time to fully exploit the dynamic could improve the precision of that measurement. A similar structure is visible on the SDB1 angular control TX and TY.
Figure 4 shows the coherence of the horizontal beam spot motion on the camera and the local controls of SDB1 and SDB2. The 200mHz (micro-seismic) peak is present on all degrees of freedom and coherent between each other. But the motion below 100mHz is coherent with only SDB1 TY. SDB1 is controlled to follow the beam of the interferometer using dither control demodulated on B1, so that is not an actual limitation of local controls of the benches. The OMC is attached rigidly to the bench so to follow the interferometer beam the bench has to move more than it would if it was just fixed with local controls.
In O5 the design is to have galvos for doing the beam pointing from the interferometer to the OMC, this means that the bench control of SDB1 and SDB2 could be based on local controls only, and have a smaller motion that what is measured currently. +/-5um is an upper limit on the relative beam motion between two suspended benches controlled with LVDTs.
Analyzing data with the OMC locked on the single bounce beam may provide additional information, in case the intereferometer beam jitter is smaller in that case. This would require adjusting the camera gain as in single bounce there is more power and it saturates, and compare the result between the standard configuration with the B5 drift control turned on and one where the SDB1 bench angle is fixed with local control only.
