[Lucia, Maria, Diego, Yuta]
We have measured the angular actuation transfer functions of IMC mirrors.
What we did:
1. Measured transfer fuctions from K1:VIS-MC(I|O|E)_TM_DITHER_(P|Y)_EXC to K1:VIS-MC(I|O|E)_TM_OPLEV_(PIT|YAW)_OUT.
2. Calibrated the transfer function with 20 V/ 2^16 to convert urad/counts into urad/volts.
Results and discussion:
Attached figure is the measured transfer function. The coupling between pitch and yaw (smaller dots) is more than roughly an order of magnitude smaller except for MCe YAW.
Blue line shows the model of the similar suspension for TAMA. The model gives rad/Nm, so it is calibrated into rad/V with 4*0.0014N.A*20mA/V*33mm (see this wiki). We can see that the measured transfer function qualitatively matches the model.
However, we see an order of magnitude variety between suspensions. We suspect it is from oplev calibration error (see klog #3175), badly aligned coil-magnet position, and/or actuator polarization error (JGW-T1707012). Could be related to klog #3295?
To do:
1. Fit the measured transfer functions with zpk.
2. Look into oplev.
3. Make a NoiseBudget model for ASC.
We updated the plot with corrected oplev calibration factor (see klog #3721).
Also, yaw to pitch coupling plot was wrong. The coupling between pitch and yaw (smaller dots) is more than roughly an order of magnitude smaller also for MCe YAW.
To check the consistency with klog #3611, I made an uncalibrated angular actuation transfer function plot (in urad/counts) for MCo (see attached). The measured transfer function is divided by 4 to make the actuation TF per coil. Our measurement is consistent with the measurement in klog #3611.
There is a mystery that DC responses vary by a factor of 4 or so between the coils of MCo (klog #3295, klog #3600) and we think that something is wrong with MCo. Actually, pitch to yaw coupling is quite high below ~4 Hz. However, in terms of angular actuation efficiency, MCo is in between MCe and MCi.
