Yesterday night and this morning some diagonalization measurements were taken with pum drive on etmx and itmx. The low coherence for the pitch drive described on the previous alog was due to the optic not centered enough on the oplev, so both ETMX p2p and y2y were remeasured. Those measurements can actually be really fast (15/20min each).
ITMX measurement has some calibration factor of 1.56 for Yaw and 1.83 in Pitch. T1100378 was the reference for calibrating the drive from cts to N.
Attached are the plots comparing the measurement and the model (wirerehang for itmx, fiber for etmx). Note that there is a sign error in ETMX Yaw, so we should double check if that comes from the Oplev sign convention.
- The data was processed (removed points with bad coherence) and calibrated in rad/N.m. It lives in :
ITMX : /ligo/svncommon/SusSVN/sus/trunk/QUAD/H1/ITMX/SAGL2/Results/quad_{p2p/y2y}.mat
ETMX : /ligo/svncommon/SusSVN/sus/trunk/QUAD/H1/ETMX/SAGL2/Results/quad_{p2p/y2y.mat
- Dtt templates were saved under :
ITMX : /ligo/svncommon/SusSVN/sus/trunk/QUAD/H1/ITMX/SAGL2/Data/2014-03-06_H1SUSITMX_{P2PY/Y2PY}_WhiteNoise.xml
ETMX : /ligo/svncommon/SusSVN/sus/trunk/QUAD/H1/ITMX/SAGL2/Data/2014-03-04_H1SUSETMX_{P2PY/Y2PY}_WhiteNoise.xml
- The data was processed with the scripts :
ITMX : /ligo/svncommon/SusSVN/sus/trunk/QUAD/H1/ITMX/SAGL2/Scripts/itmx_L2_diag.m
ETMX : /ligo/svncommon/SusSVN/sus/trunk/QUAD/H1/ETMX/SAGL2/Scripts/etmx_L2_diag.m
PUM drivealign P2P and Y2Y filters were installed.
In each of the attached, foton screen shot to the left shows the actual filter installed, matlab plot to the right shows the P2P or Y2Y measurement, fit, model, and the expected TF from PUM to the test mass after the new filter takes effect.
As you see I made the filters such that ITM and ETM looks as if they share the same low Q (Q=3) resonance at 0.51 Hz for PIT and 0.6Hz for YAW. There's also a roll of at 30Hz, which is not that aggressive. If necessary you can easily add more.
I ignored the DC calibration of the measurement. I'll handle them in the gain.
The filters relevant are FM1 of H1:SUS-ETMX_L2_DRIVEALIGN_P2P, Y2Y and corresponding ITMX filters.
Scripts used for the fit and inversion as well as the filter definition files generated by the scripts are:
~controls/keita.kawabe/fit/[EI]XPUM_[PY]2[PY]fit.m
~controls/keita.kawabe/fit/[EI]XPUM_[PY]2[PY]inversion_soscoeffs.txt
[Yuta Keita Arnaud]
ETMX ITMY ITMX top mass P2P drivealign filters installed
On friday, filters were designed the same way as Keita, in order to get a single pendulum transfer function from the top mass drive to the test mass displacement in pitch. The filters were installed in the top drivealign matrix in FM1 of the P2P bank filter. ITMX filter is a copy paste from ITMY, assuming they have the same P2P transfer function (both are wires). Attached are the plots showing the design of the filters and the foton version. The transfer function after filtering has its cut off frequency at 0.45Hz for ETMX and 0.55Hz for ITMY/ITMX.
ETMX Length to pitch decoupling was also designed and installed in L2P drivealign matrix. Plots of design pending.
