There are a lot of glitches in the NE Pcal laser power since November 1st, around ~0h, see first plot (the bottom right plot is for WE for comparison).
The plots 2 and 3 show glitches in the time-domain, partially hidden by the permanent lines . They are mainly visible in NE_Rx_PD1 (integrating sphere in reflexion of NE), less on the Pd on the injection bench (PD1 in loop, PD2 out-of-loop).
Plot 4 is the FFT of the signals at the same minute that plot2, with a few glitches in the window: the noise level is increased on NE_Tx_PD2 (out of loop) and NE_Rx_PD1 sphere, not on the in-loop Pd.
Plot 5 is the spectrogram of the 3 NE channels during the same minute.
The glitches are all going towards higher power measured.
Looking only at the out-of-loop sensors, Tx_PD2 and Rx_PD1, the measured power goes higher on Rx_PD1, while they are both calibrated to have the same signal. Maybe the bandwidth of the sensors reduce the amplitude on Tx_PD2 if the glitch is at high frequency (seems to be a ~2 samples at 20 kHz).
It may be worth trying to at least stop/start the NE laser and or the power supplies as a first "reboot", before further investigations...
