For OFC2, the laser lock quality seems poor. This and this show the 775 error signal used to lock the cavity in red, and blue shows the TRANS PD for 1550. I am not sure how to quantify a "good" lock in these cavities. For any diagnostics, this is a poor lock quality as the total 1550 transmission is fluctuating by alot. But in GQuest operation mode we will have it detuned by 17.6 MHz, thus supressing 1550 light. If it jitters around 17.6 MHz, still suppressing the carrier and allowing signal photons through, is this still an issue? We cannot suppress those fluctuations further, even though increasing the UGF of the loop squashes these fluctuations. At a certain point, something rings up (in the laser dc modulation port? elsewhere?) and doesn't allow further supression.

I am attempting to look at this issue on OFC1. It is largely the same. One major problem I diagnosed is having both EOM phase modulations on in one cavity at the same time. The clock sync that we have between the 4 mokus seems to not be sufficient. The signal is fluctuating multiple times a second. One moment and the next. This may not seem like a big deal but since the two wavelengths are tracking eachother, having a 1mV difference in the average level of the error point has the effect of detuning the 775 light away from the 1550 peak about a FWHM distance (read: when the average error point is 0, they are coaligned, when the average is lower, the 1550 TRANS light jumps to about half the value). Turning off the 1550 EOM signal gets rid of it. This is when the two EOMs are at the same frequency. When they aren't there is a large beat note in the error signal.