I wanted to investigate the weak error signal of the GRMC and its somewhat sensitive response to acoustic noise. Actually I found the SHG has a bit of the same behaviour. I'm not sure what happened, but recently I noted I had to turn the control loop gain way down to avoid oscillation. I probably should have checked the open loop transfer function of both but I got sidetracked by some other thoughts. Anyway it's a bit mysterious why the ideal gain went down for the SHG and GRMC.

Since the SHG temperature was not optimized in a long time, I changed it from 3.163 (174 mW green) to 3.192 (318 mW green). Actually this seems like a bit more, it was supposed to be 280 mW max. Input IR is about 777 mW directly in front of the SHG, compared to 680 mW 3151. Overall SHG efficiency is still 41% though. However, I later noticed that the IR Sensor type power meter (Thorlabs S-121C) measures 655 mW IR at the SHG, while the stick type sensor (S-130C) measures 777 mW. I did a small profile of the temperature vs green.

The green error signal amplitude is a bit weak. However, the green lock PD is very sensitive. Since it locks and outputs stabilized green power to the OPO, maybe I should just keep going and see if the OPO works properly. It seems quite surprising that we can lock despite using 88 MHz IR sidebands that are going through three dichroic mirrors. The 88 MHz signal at GRMC_REFL_RF is about -14.7 dBm. The green lock PD has bandwidth 100 MHz so the frequency doubled green sidebands don't show up in the PD spectrum.