Naoki, Vicky, Nutsinee
We found that the SHG green power decreased from last night. As shown in the attached figure, this happened also 7 days and 14 days ago and the green power recovered by itself with no reason. We have not figured out what it happening, but we will see if it recovers tomorrow and happens again next week.
The pump ISS could not catch up the green power decrease. The setpoint of OPO TRANS for pump ISS was 90, but this was too high with decreased green power. We increased the SHG launch power by rotating HWP and decreased the pump ISS setpoint from 90 to 80. The pump ISS is working now.
We tried several things, but none of them helped the green power decrease.
- Relocked SHG
- Adjusted the alignment to SHG by pico and adjusted the SHG temperature
- Adjusted the HWP before EOM to optimize the polarization to EOM
- Since we suspected the mode hopping, we changed the temperature of laser. The laser head crystal frequency was a bit low recently. We touched the temperature very slightly and changed the laser head crystal frequency from -160 MHz to -43 MHz
- Installed a new PD for monitoring the SHG IR injection power
An alog from Sheila I found from 2019 made me think this behavior we're seeing in the SHG might be mode hopping. At the time they were losing nearly 50% of the SHG power. We aren't quite that bad yet but something to keep an eye on.
SHG would lock at high efficieny then loses it after a few minutes. This behavior has been observed before in the past. I'm not sure if we ever figured out what happened.
Hopefully this SHG power drop, which coincidentally has happened 2xTuesdays in a row, is not a case of Tuesdays. We installed a PD on the SHG-IR injection path towards monitoring SHG-IR injected power, but it is not yet cabled into the digital system.
We installed a PD (Thorlabs PDA100A2) in transmission of PBS1, which splits IR pump power going to SHG (see SQZT0 layout D1201210); this replaced the beam dump on PBS-TRANS. 3 ND filters (ND2 + ND1 + ND0.6) are installed on the PD. This PD is powered from the 12V supply in SQZT0's table-top interface box, and its output is cabled to an empty TNC on the patch panel between SQZT0/7, near cables ISC_SQ_380 and ISC_SQ_383 (tabletop interface box inputs were full). We can next setup for long-term monitoring power in this path.
The green power is recovered now as shown in the attached figure. During the power decrease, the injection IR power seems stable.
The green power decrease was long yesterday and two weeks ago while it was not so long a week ago. Vicky pointed out that we went to SQZ tables yesterday and two weeks ago while we did not go a week ago. We will not go to SQZ table and see the power decrease next week.
Vicky, Nutsinee
We didn't go on to the table this week. The green power didn't drop. Yay!
Most of our SHG power drift coincides with LVEA temperature. Most of the abnormal incursions so far were caused by us opening the table. During those times the input power were stable, including the new spare we installed that looked at the input power just before the SHG (Spare_A_DC). It's possible that something inside of the SHG is just very sensitive to table temperature -- something to think about if we want to redesign SHG for O5. SHG conversion efficiency has been stable for the past 70 days.
