Sheila and I went out to the floor to plug in the cable to run frequency noise injections (AO2).
I switched us to REFL B (79037) only to run the injections. I first ran the frequency noise injection template. Then, I disenaged the 10 Hz pole and engaged the digital gain to run the ISS injections (low, medium, and high frequency). I was fiddling with the IMC WFS pitch injection template to get it to higher frequency for a jitter injection when we lost lock.
Next step is to at least use the intensity and frequency measurements to determine how the noises are coupling to DARM (would be nice to have jitter too, but we'll see).
Commit 464b5256 in aligoNB git repo.
These results show that above 3 kHz, some of the intensity noise couples through frequency noise. There is no measurable intensity noise coupling through frequency noise at mid or low frequency.
The frequency noise projection and coupling in this plot are divided by rt2 to account for the fact that frequency noise is measured on one sensor here, but usually is controlled on two REFL sensors. This may not be correct for high frequency (roughly above 4 kHz), where CARM is gain limited.
code lives in /ligo/gitcommon/NoiseBudget/simplepyNB/Laser_noise_projections.ipynb