We made a measurement of the power stability after the green mode-cleaner with both mode-cleaner and MZ locked.

The mode-cleaner was locked with the Stanford preamplifier (first order low pass, cut off frequency 3 Hz, gain = 5). The unity gain frequency should be measured, but the lock was very stable.

The MZ was locked with the Emil board at a position of the frange intermediate between the half fringe and the bright fringe (bright fringe correspond to 1,3 V and MZ was locked at about 1 V, corresponding to 11.3 mW), using the transmisison signal after the mode-cleaner. The measurement of the unity gain frequency gave a 3 kHz ugf (see picture 2), compatible with the ugf measured by Emil. (gain of the potentimeter 9.9).

Figure 1 shows the power transmitted with no lock, then with the 3 kHz ugf lock and with a gain reduced  (from the position 9.9 to the position 3.5, corresponding to a measured ugf less than 100 Hz). We remark that, during the transition between high gain and low gain the loop experiences an instability at about 600 Hz (as altready observed by Emil) and it is coming back to a stable position.

The power transmitted by the mode-cleaner was 11.3 mW (corresponding to about 1 V). The fluctuations of the transmitted power on several tens of seconds timescales have an amplitude less than 10 mV (which corresponds to 1%)  with the high gain. With the reduced gain, the fluctuations increase by roughly 1.5 and they remain less than 2%.