It seems like we have still range to gain from increasing the generated squeezing levels back up to nominal. This has been a challenge since the LVEA temperature drifts, which I think have affected some on-table SQZT0 alignments, so we can't run with nominal/typical generated sqz levels. And moreover, it's caused our ISS to rail several times recently, which we've had to drop out of observing to fix.

Looking at on-table power trends, we are having to launch more power to fiber for the same power out of the fiber. The Y2 cursors show that for 20mW fiber launched, we used to have 85uW opo trans, with lots of range on pump ISS. Now we have way less power launched, and no room on ISS. Specifically:
- >2 months ago, we launched 20 mW into the fiber, for 85uW opo trans (w/lots of room to spare on iss).
- 1 month ago, we launched 20mW into the fiber, for 75uW opo trans (some room to spare). Camilla aligned mid-May 69704, but then a bunch of LVEA temp changes happened.
- Today, we launch 22mW into the fiber, for 60uW opo trans (almost no room to spare on ISS).

Things we can try to get back:
- Given the LVEA temperature changes, we can check/optimize the pump fiber coupling efficiency on SQZT0. We can also check pump AOM throughput, but its diffraction effiency looks good still (~60%).
- Can also check the OPO transfer function, or revert the powers to OPO's RF PD. Since I rejected power to not saturate the REFL RF80 PD used for OPO PDH locking in mid-april (~50 days ago, LHO:69677), even though the reported RF power was similar for the lowered dc power, maybe I should have adjusted the opo servo gain. But, we've recovered the "normal" situation since this change, so I don't think this alone is responsible.