We closed the temperature and bEOM loop of the FSS. Also FSS loop gain was optimized. It seems very stable such that the reference cavity continued to lock for over an hour.
Each parameters are as below
UGF of total loop : 100kHz
cross over frequency between bEOM and laser PZT passes : 24kHz
Fast gain nob : 270 (with ×5 gain on SR560)
Common gain nob : 0
Slow servo switch : inv
What we did
Yesterday the FSS was locked with only PZT as actuator. First of today I tried to activate the laser temperature loop and it was done with turning the servo/off/inv switch to inv. (that means to turn on the temperature loop and the sign is opposite to the PZT loop)
Then we closed the bEOM loop. It was also done very easily. I just connected the EOM and TTFSS servo with the estimated optimal gain.
We tried to measure the open loop TF but the feedback signal to EOM was kicked and saturated with tiny excitation signal such as -45dBm. The bEOM seemed to have too large load, so we tried to increase the cross over frequency between the laser PZT and bEOM. The TTFSS servo gain was not large enough, so SR560 was placed between the TTFSS servo and the laser PZT with gain 5. Then the FSS loop seems very stable. Now the crossover frequency was 24kHz. (mesured Fast OUT1/OUT2 with exciting Fast EXC. This measurement measured Gpzt/Geom where Gpzt is OLG of PZT loop and Geom is OLG of EOM loop. The result is attached as picture).
The OLTF could be measured with -17dBm excitation signal, and the measured UGF was 100kHz and gain margin was about 30 degree (see the attached figure)
Our target UGF is 200kHz-500kHz, but the TTFSS tunable gain was already maximum. So we should improve TTFSS servo to have more gain, and then we would achieved this UGF. The gain margin is enough even at 500kHz.
We leaved the FSS loop locked, let's see how stable the loop is.
