[Haoyu, Carl]
Summary: We tried again to measure pitch jitter noise coupling. We found Yokozawa-sans measurements this morning still have CARM coupling, even though we should have reduced it by ~1000 fixing MC optics L2P and L2Y. We tried driving MCE pitch we found this had less CARM coupling by at least a factor 5. Projecting IMMT1 TRANS QPD1 pitch into DARM with MCE drive now looks reasonable in the regions of the spectrum where there is coherence. This first attached figure shows the QPD signals top left, the coherence between QPD and DARM bottom left, the passive transfer function top right and the projection in to darm bottom right. Note that in the regions ~58Hz, 129Hz and 225Hz pitch jitter almost completely explains DARM noise, while yaw measured with the MCO drive also has a significant contribution in the same frequency bands. Why the sensors overproject in regions where it appears they are not sensor noise limited is not clear.
The transfer function measurement was made by driving MCE in pitch with 40 count in a 17 point swept sin 50-150Hz. The measurement is not very good with lack of coherence between the QPDs and K1:CAL-CS_PROC_DARM_DISPLACEMENT_DQ. There is coherence withe the OMC DCPDs, peaks were clearly visible in both so with somework the OMC DCPDs could be used to make a better projection. The second attachment shows the transfer function measurement. A better measurement (larger drive for longer) over a larger frewquency range would improve the projection.
To completely project jitter noise in to DARM we need to project both beam pointing and beam position to DARM. MCE drive is a drive of beam position at the IMC waist (see hoayu's simulations). We therefore also tried to drive MCI and -MCO that drives beam pointing at the IMC waist. However, this drive has the same CARM coupling as MCO drive. We had to drive point by point as there was no spare matrix / filter combination to drive a swept sin. We have 8 points between 34 and 121Hz. However, for now we just report that the IMMT1 QPD1 P TF to DARM at 121Hz is 2*10^3. This is comparable to the MCE drive so should give a reasonable projection in to DARM, despite CARM coupling. With only one withness sensor that measures a combination of beam pointing and beam position I think the right way to project is to only project the maximum coupling to DARM (to not double count). In the case of around 121Hz the maximum arises from IMC beam position noise and hence the projection in the first figure is correct.
