I used diagnostic breadboard (DBB), with ISS first loop on and DBBPMC (not real PMC) unlocked, to use the DBB WFS DC output as the jitter sensor.

Right column is with the frontend beam into DBB, and left is with high power laser beam. QPD_[12]Q[XY] means WFS[12] DC [YP], and QS is the DC sum. I pressed "pre-align" so the beam comes to the center of the sensors, and the measurements were made while the pre-align servo was on. The total power for both of the beams sampled are about the same.

As you can see, the frontend beam is much quieter than the HPL beam over the entire measurement band, it's a factor of 40 or 50 larger at 100Hz.

The coherence between the HPL jitter and intensity to the PMC transmission RIN is large for the entire band (left bottom). Blue and brown show PIT to intensity coupling. Pink and cyan show intensity to intensity coupling. The fact that the coherence is high for PIT (Y) but not for YAW (X) means either the alignment into PMC is off in PIT, or some other jitter-intensity conversion mechanism e.g. clipping or QE dependence on the beam position is worse/steeper in PIT.

Just to see how this is different at different measurement point, orange and black are the PIT to intensity and intensity to intensity coupling measured by (one of the?) IO diodes downstream of the main EOM. (I know nothing about the analog filtering of this diode.) There's some but not huge difference in PIT-to-intensity coherence. Probably PIT-to-intensity conversion mechanism is more or less common to these two, e.g. alignment into PMC.