(Squeeze team)
The first pic shows the spectrum of the squeezer beat note between the main and squeezer laser. There is a prominent line at 163kHz that has about 35dBc amplitude (the 650kHz bumps disappear if the noise eater is turned on). This line showed up in many squeezer servos. We tracked it down to the main laser. There, it showed up in the IMC VCO spectrum. This was in full lock. When we tried to measure it again today with only the IMC locked, we didn't see it again!?
We moved to diagnose the problems we are having with the laser locking servo. In particular, the old O3 settings no longer work with us unable to engage the EOM path. After playing around for a while we found a configuration with rather low bandwidth and ugf around 170kHz (see second pic). We then found the fast/eom crossover to be low as well around 7kHz. However, increasing the fast gain did not significantly increase the crossover ugf. This turned out to be a line at 2.5MHz that was saturating the variable gain amplifier in the fast path (3rd pic). We tracked this line to the main laser again, where the line shows up prominently in the TTFSS error signal as well as the IMC error signal. This line moves naturally around by a few Hz. We were able to increase its frequency by ~500Hz by reducing the TTFSS common gain by 1dB.
Next things to do:
- Check if the 160kHz line comes back when we are in full lock.
- Figure out how to suppress the 2.5MHz line.
