We were wondering about the beam spot position measurements we make.  There was a major change in our IMC spot positions between June 5th and July 3rd, which made us question the validity of our results.  The most likely cause of this change was the NPRO replacement, but as our PRMI/DRMI buildups are currently bad with suspicion of clipping, we need to know if we are fooling ourselves with spot position estimates.

Spot positions are found via Koji's coil actuation alpha method.  Jenne and Kiwamu found alpha values via A2L coil actuation gains.  
I ran Hang Yu's A2L minimization script on the IMC twenty times overnight, and looked at the calculated beam spots for each run.  Each run takes a little under five minutes, and the whole suite took 97 minutes.  This would tell us (1) if our A2L minimization method has high variance and (2) if the beam spot is moving on an hour time scale.

Results

MeasDOF   Mean Spot Position [mm] +- 1 Sigma Unc
MC1 P     0.188   +- 0.031
MC2 P     1.653   +- 0.044
MC3 P     0.405   +- 0.123
MC1 Y     0.164   +- 0.169
MC2 Y    -0.741   +- 0.069
MC3 Y     1.102   +- 0.109

Our IMC spot position standard deviation is ~0.1 mm, and it is not moving on the hour time scale.  Plot 1 shows each measurement result.

I also looked at Hang's A2L minimization method, and calculated uncertainty in the A2L gain from his linear fits.  Example A2L gain uncertainty and fit shown in Plot 2.  Uncertainties in spot positions here also corresponded to about ~0.1 mm using this method, with some times of higher uncertainty (~3 mm).