I had a look at the three SRC gouy phase measurements listed in 88155, this is related to commissoning modeling git issue 33

To do this, Evan Hall helped get me started using the current version of finesse (3.13.13) in the control room.  Anyone can use this conda environment by using conda activate finesse.  

In the first plot, the upper left subplot shows the cold gouy phase measurement, which is reasonably close to the finesse prediction using nominal parameters from the LHO_O4.yml.  I've added traces to show what happens when the SR3 ROC and SR2 ROC are changed by the rss error given by Garilyn, +/-6mm for SR3 and +/-3.7mm for SR2, and varied the distance between SR2 and SR3 around the value in the finesse yml.  The gray band shows the measurement result, indicating that the gouy phase measurement could be explained by the nominal ROCs if the SR3-2 distance is longer than the value used in finesse by 3mm, or if the distnace is nominal the SR3 ROC could be shorter by 6mm to explain the measurement.  

The next two panels show the two measurements made in October 2019, before the ITM replacement (52638 and 52658).  The ring heaters were on for both of these measurement, decreasing the expected gouy phase, and central heating was also on which should increase the gouy phase.  Using the TCS actuation strengths from 90004 we cannot easily explain these measurements as compatible with the cold state measurement.  

The last panel shows the impact of each indivdual TCS actuator on the gouy phase measurement.  The fact that all these lines are fairly close to parallel suggests that a series of measurements where we change only one actuator at a time could be a useful check of the actuation strengths.  

In the second attached plot, I plotted the difference in gouy phase between having the heater on and off, which we can get from the comparison of the two 2019 measurements, with various parameter changes.  I was wondering if we could use this difference to get information about the SR3-SR2 distance if we trust the HWS measurement of the SR3 heater strength.  The 1.8 degree systematic uncertainty on the gouy phase measurements is large enough to make this seem not very promising.  

The script used to make these plots is in the commissoning modeling repo here commit 8635b914