After Richard fixed the issue with the ETMX ESD (see alog 11573) this morning, I briefly took an angular transfer function of pitch and yaw using the oplev. They don't look crazy -- 1/f^2 shape and non-measurable crosscoupling between pitch and yaw. However there was a descrepancy from the model predictions by a factor of 2-ish. I am not sure what makes the descrepancy for the moment. The transfer functions are shown below with the models for comparison.
Anyway, the ESD is functional and ready to go.
Note that I included Jeff K's arc-integral-factor which gave an extra factor of 2/pi for the torque in the model. For those who are interested in the details of the calculation, please find it in the attachment (ESD_pitandyaw.m). The raw data are attached as well.
The L1 ESD actuation measurements, applying 2 Hz pitch and yaw signals and looking at the optical lever signals, also show the reponse being a factor of 2 smaller than expected. There seems to be something systematic here that we are missing.
The original templates for this measurement can be found here: /ligo/home/kiwamu.izumi/Desktop/binary/20140424_ESD_ETMX/dtt/ swept_sine_long.xml (for L2L) swept_sine_pit_yaw.xml (for P2P and Y2Y) In addition, they've been copied and committed to the SUS repository here: /ligo/svncommon/SusSVN/sus/trunk/QUAD/H1/ETMX/SAGL3/Data/ 2014-04-28_H1SUSETMX_L3_L2LPY_SweptSine_1to20Hz.xml 2014-04-24_H1SUSETMX_L3_P2PandY2Y_SweptSine_1to20Hz.xml