J. Kissel

I've driven a few PCAL excitations into the IFO during its current status (LHO:63284), without changing *anything* in the front-end calibration pipeline since I last made a change in April 2021 (LHO:58691).

Please -- use all of our might -- to not take away a "final" message from these plots:
   - understand that things that affect the DARM loop are very much in flux, 
   - I'm rusty, caught between a state of 
       - "shake off the dust from what we did last time" 
   and 
       - "there's brand new DARM model analysis software that is still in its baby, prototype, barely-exercised, form *and* I'm still learning how to use it."

But I can at least comfortably say that "yes, the calibration is different than it was before, and there is quite a bit of systematic error since we haven't changed anything to follow it."

First result:
    2022-05-23_H1_PCAL2DARMTF_LF_BB_3min.png
    This covers the *first* lock we got right as we opened up the HAM5-HAM7 gate valve, after the IFO had been at 50 W input for ~30 minutes.

Second result:
    2022-05-23_H1_PCAL2DARMTF_LF_SS_5to1100Hz_10min.png
    This covers the *second* lock we got after the gate valve, also with the IFO "cooked" at 50W for 30 minutes.
    However, we switch the ETMY UIM driver from state 2 to state 1
    We lost lock after this, while team SQZer was trying to SQZ in some SQZ (very doubtful it was their fault.)
    
I cannot say what the systematic error is from, but I can make some educated guesses:
   - It is definitely *not* just one thing.
   - One can easily see that these two measurements -- nominally the same transfer function -- show *different* results, especially in phase at low frequency. For this, my first guess as to the *difference* would be UIM coil driver state change between the measurements, but it could also be some time-depenendent SRC detuning and/or L2A2L coupling because our spot positions are still. I can make no claims about the static, large error.
   - Our optical gain *should* be roughly the same now as it was in 2021 given that the power circulating in the arms is about the same; we've got a bit lower recycling gain, but we've made up for it with more input power. So, I'm surprised to see the ~10% lower DeltaL / PCAL ratio. But, that's likely the bulk of the error at high frequency
   - Craig's loss assessment from power build-ups within the cavities during power up leads him to believe that the DARM cavity pole may be a *bit* lower than the 450 Hz we had in 2021.
   - We have an new analog electronics chain for the OMC DCPDs, and it has some ~few % wiggles all over the frequency band LHO:62916
   - It's been a year our more since we assessed the charge situation of ETMX's ESD, so that actuator strength could easily have changes at the several % level

   Finally -- even the templates may have some bogus corrections for super-Nyquist elements of the DARM loop, in a lame attempt to make these plots *look* like a GDS-CALIB_STRAIN/PCAL plot. And this "high-frequency" corrections have demonstrated wiggle-causing effects near the DARM unity gain frequency around ~100 Hz.

So -- yeah -- please take these results with lots of grains of salt, but it's a start.

As always, more to come.