Keita Daniel
We found that the transimpedance gain of the ALS-X_FIBR_A_DC PD was wrong (changed it from 20k to 2k). In turn, this meant that 20mW of light was on this PD.
After looking at the beatnote amplitude directly at the PD and found it to be way too small, we decided to swap the PD with a spare (new PD S/N S1200248, old PD S/N S1200251). However, this did not improve the beatnote amplitude. (The removed PD was put back into the spares cabinet.)
We then looked for clipping and found that the beam on the first beam sampler after the fiber port was close to the side. We moved the sampler so the beam is closer to the center of the optics. We also found the beam on the polarizing cube in the fiber path to be low. We moved the cube downwards to center the beam. After aligning the beam back to the broadband PD, the beatnote amplitude improved drastically. This alignment seems very sensitive.
We had to turn the power from the laser in the beat note path down from 20mW to about 6mW on the broadband PD.
This required a recalibration of the ALS-X_LASER_IR_PD photodiode. The laser output power in IR is about 60mW.
The beatnote strength as read by the medm screens is now 4-7dBm. Still seems to vary.
To recap, fundamental problem was the alignment (probably it was close to clipping before, and started clipping over time due to temperature shift or whatever). Also, the PBS mount or maybe the mount post holder for the fiber beam is not really great, a gentle push by finger will flex something and change the alignment enough to change the beat note. We'll have to see for a while if the beat note will stay high enough.
Wrong transimpedance value in MEDM was not preventing PLL from locking but was annoying. H1:ALS-X_FIBR_A_DC_TRANSIMPEDANCE was 20000 though the interface box gain was 1. This kind of stuff confuses us and slows down the troubleshooting. Whenever you change the gain of the BBPD interface box, please don't forget to change the transimpedance value at the same time (gain 1= transimpedance 2k, gain 10= 20k).
We took a small plier from the EE shop and forgot to bring it back from the EX (sorry).
Everything else should be back to where it was. Thorlab powermeter box was put on Camilla's desk.
It's still good, right now it's +5 to +6 dBm.
Too early to tell but we might be diurnally going back and forth between +3-ish and +7-ish dBm. 4dB power variation is big (a factor of ~2.5).
If this is diurnal, it's probably explained by the alignment drift, i.e. we're not yet sitting close to the global maxima. It's not yet worth touching up the alignment unless this becomes a problem, but anyway, if we decide to make it better some time in the future, remember that we will have to touch both the PBS and the fiber launcher (or lens).