Wen, Nutsinee, Daniel, Vicky
Summary: Recently we have noticed the OPO crystal get lossy, so Daniel made a cable, and today we moved the crystal translation stage slightly. We are back to a reasonable spot to continue sqzing!
Daniel made a cable to move the OPO crystal translation stage at the ham7 flange; cable worked beautifully! So, following some notice of crystal degredation we have observed recently (less NLG for same OPO pump powers, reduced OPO_GR_TRANS), we decided to move the crystal position slightly to re-optimize nlg.
Before moving OPO crystal (after recent Aug 2022 move, LHO:64272), here are some observed parameters. More alog and forensics on this time's degredation to come; see Haocun's OPO loss study for some previous measurements of crystal degredation based on green refl/trans.
- SHG_LAUNCH_DC_POWERMON (sqzt0) = 6.7mW
- OPO_REFL_DC_POWER = 0.8/1.4 (on/off res)
- 2x more reflected green from opo b/c 50:50 SK:BS1
- OPO_TRANS_DC_POWER = 24 uW
- SEED_LAUNCH_DC_POWERMON = 3 mW
- nominal_IR = 40 uW (scan opo pzt w/green blocked for no nlg, and seed light on. Record max OPO_IR_PD_LF_OUT_DQ.)
Procedure for optimizing crystal position when we are very close (following on from LHO:43292):
- Staying near our recent nominal temperature ~ 32.3 C.
- Move the crystal translation stage in steps of 50 or 100 counts (10 for fine adjustments).
- small moves ~10 steps typically don't require re-locking opo cavity, only crystal temperature adjustments to optimize NLG. 10 step moves are noticeable on opo lock signals.
- bigger moves ~100 steps typically required re-locking in green (OPO GRD "LOCKED") then changing tec temperature to find+maximize OPO_IR_TRANS
- Adjust OPO TEC temperature setpoint to maximize non-linear gain of seed power (aka power in OPO_IR_TRANS). Record max_ir/nominal_ir.
This screenshot shows how, with constant input green pump and input seed (red) power, changing the opo crystal position increases non-linear gain to increase the max seed power transmitted.
| nonlinear gain | co-resonance temp (C) | max_IR / nominal_IR | |
| start position | 1.5 | 32.33 | (62/40) |
| +10 steps to "left" | |||
| +100 more steps to left | 2.5 | 32.345 | 103/40 |
| +50 more to left | |||
| -300 right | |||
| +100 left | 3.1 *underestimated | 32.03 | 124/40 |
| final position (-40 to right?) | 32.03 |
Ran out of laptop battery on the windows computer, so leaving crystal translation stage here. As told, I left Daniel's crystal stage connector cable in the ameristat bag he gave us, and left that by the HAM7 connector flange. All is un-powered. We will clean up more tomorrow.
The NLG=3 is underestimated due to pump depletion, we are using too much seed power to look at this (moved crystal w/3mW seed launched, with pico-waveplate we can get down to 0.5mW seed launch, but we really want < 0.1mW launched..). At our minimum of 0.5mW seed launch, we measure NLG:
- for pump launched 2.5mW, OPO @ 32.033C, max_IR/nom_IR = (1.77)/(0.91). NLG = 1.95
- for pump launched 6.6mW, OPO @ 32.028C, max_IR/nom_IR = (3.14)/(0.91). NLG = 3.45
Based on these NLG and launched green powers to the fiber, an extremely quick (over)estimate of threshold is ~ 31mW into the fiber. (using x = (green/threshold)**0.5 = 1 - NLG**-0.5), a little higher but similar to before (LHO:64464 Dhruva carefully calculates/measures threshold ~25mW into sqzt0). Especially, given our 0.5mW of seed launch we see noticeable pump depletion, so here we underestimate NLG, and overestimate threshold.
Great! Looks comparable (only slightly less good) for NLG vs OPO_TRANS data vs the previous position, LHO:64464.
Here are some plots that underscore the OPO's degradation in the last few months. It is clear that the 532 trans and NLG are much lower than they were 4 months ago for the same value of fiber launch and OPO refl.
