Started making a new HAM3 noise budget with the CRS included
- Modified pre-exisiting NB: /ligo/svncommon/SeiSVN/seismic/HAM-ISI/Common/HAM_ISI_Model/HAM_simple/ham_isi_nb_clean.m
- Added new function get_blnd_filters_BP.m which is the same as get_blnd_filters.m but allows for HAMs to have BP filters
CRS out of loop measurement (1470454819) : Witnesses the table rotation. Looks how we expect it, with the CRS tilt measured more or less matching the predicted noise cure below 1Hz and matching the GS13. (Figure 2)
CRS in loop measurement (1470458418): Figure 3 shows the RY blends on HAM3 with the CRS in loop (i.e. with a bandpass filter),
Figure 4 shows the Noise Budget for the platform, with the GS13 and CRS tilt measurements, the predicted tilt, and the noise contribution from each sensor. The CRS measured tilt falls a bit below the predicted tilt, but this might be from the various sensor noise estimates (most likely CRS) being a bit higher than realitly, or the ground tilt estimate being inaccurate at low frequencies.
Figure 5 and 6 shows the CRS noise contributions to the CPS and GS13 respectfully
Figure 7 shows the CRS noise budget
These are still a little rough and I'll keep working on them. In particular, I'm slightly confused on why the CRS noise contribution is so dominate in for the GS13.
Thanks for adding the in-loop crs functionality to the noise budget, this looks great. Note, the base infrastructure for this noise budget is described here: https://dcc.ligo.org/LIGO-G2100779.
Since the orange curve on Figure 4 is the predicted residulal motion, we don't necessarily expect the measured in-loop CRS to match it. The in-loop sensor signal might be low, but still impress some of it's own (or other blended sensors) noise into real platform motion.
This new predicted residual platform tilt noise (crs in loop) is probably a good enough proxy to be used as an input to the horizontal loop to help design new X blend filters.