Reports until 18:00, Monday 12 December 2022
H1 SQZ
dhruva.ganapathy@LIGO.ORG - posted 18:00, Monday 12 December 2022 - last comment - 18:33, Monday 12 December 2022(66312)
Tests of the FC with the beam diverter open (and some frequency dependent 'squeezing'!!!)

Vicky, Naoki, Nutsinee, Daniel, Dhruva

With the IFO locked in nominal low noise today, we decided to check whether the filter cavity locking loops inject any length noise that turns into backscatter in DARM. In order to take this measurement, we first moved the RLF frequency to what we thought was carrier resonance in the FC. This frequency ended up being 104.66 kHz, about 2 kHz away from the frequency that we found on Friday (alog 66281). Not sure why this frequency moved so much. Before making any measurments, I used the ADF to try and align better to the IFO and we think the alignment isn't too bad right now. We have yet to optimize PSAMS.

We measured DARM with the FC locked in green near carrier. We saw some excess noise at low frequencies (could be the backscatter we were expecting to see). We iteratively tried to update our best estimate of where carrier resonance was by looking at DARM. We locked the FC in IR and saw this noise come down. At this point, it's still a bit hard to decouple what how much excess noise is coming from scatter and how much is related to squeezing rotation. This excess isn't very clearly visible when we're anti-squeezing either. This is the dtt screenshot with the aforementioned traces. Daniel also took a scatter measurement by dithering FC2 L at 40Hz while FC was locked on IR. The scatter was clearly visible in DARM, alog to follow..

Speaking of squeezing rotation, however, we think we have seen some solid evidence of frequency dependent rotation while we're injecting squeezing into the IFO!!!!! We moved the RLF up in frequency by 1 and 2 kHz to see whether we could see FDS and we did! The main issue, however, was that our non-linear gain and generated squeezing is way too low to actually see any real squeezing. But the frequency dependent rotation was super clear when we were antisqueezing, with the quantum noise dropping back to shot noise in the FC bandwidth. 

Images attached to this report
Comments related to this report
daniel.sigg@LIGO.ORG - 18:33, Monday 12 December 2022 (66315)

With FC locked on red, we injected a 40Hz line into the FC2 lengt path (see attached plot). The line was visible in the displacement spectrum and became about 2Hz wide. Looking at the rms in a 2.5Hz bandwidth, the line in DARM was about a factor of 5 above background, whereas the line in SUS-FC2_M3_ISBINF_L was about a factor 125 above background.

Non-image files attached to this comment