Following on with measurements made in the field by Jeff (see [1]), we wanted to repeat these measurements for a PUM AOSEM in the lab up to 100 kHz. In order to characterize the AOSEM, I used a spare AOSEM, PUM coil driver, satellite box, and a coil driver test box. These were connected in a similar manner as Jeff has done previously (see alog 24725). For these measurements, I only needed the test "bench" setup, and the full connected chain. The data is stored at

/ligo/svncommon/CalSVN/aligocalibration/trunk/Runs/PostO1/H1/Measurements/Electronics/BenchPUMDriver/2016-02-29

Dividing out the bench setup from the full connected chain, we can recover the AOSEM behavior without the coil driver and test box. LISO is used to fit this resulting transfer function to one zero and two poles (see the attached figure):

Best parameter estimates:
Zero: 760.627 Hz +/- 3.734 Hz (0.491%)
Pole: 22.053 kHz +/- 167.9 Hz (0.761%)
Pole: 661.6 kHz +/- 46.36 kHz (7.01%)
factor = 4.880 mA/V +/- 14.19 uA/V (0.291%)