Jennie W, Sheila D, Jeff K
In order to get an understanding of output losses to understand why we are not getting the level of squeezing we expect I ran craig's scripts to step the DARM offset. Instead of measuring PCAL line heights on DCPD_SUM at each step we will look at these line heights on OMC REFL.
As this PD is much noisier than DCPD I had to do some tuning of the PCAL lines we use. Since we have limited measurement time and the IFO was still at MAX POWER, Sheila suggested we spend some time tuning the line heights to use in the measurement.
By this point we were waiting for violin modes to damp while in the OMC_WHITENING STATE.
original PCAL EY and EX values are first image.
I ran set_up_pcal_for_darm_offset_test.py. Originally this uses:
PCALX 255Hz 40000 counts
PCALY 410.3Hz 40000 counts
but this did not give us the correct resolution on OMC REFL for the higher frequency of the 2. Neither OFS PD was saturated.
See Ref0 (4th image attached) for the OMC REFL spectrum with these settings.
I manually stepped up the 410.3Hz line in ampltiude (on PCALY PCAL_END screen to 48 000 counts at which point the OFS PD saturated. Following Jeff's instructions I turned off the loop_enable switch on the PCAL_END screen for PCALY, changed the amplitude down to 46000 and swtiched the loop back on. OFS no longer saturated.
see Ref1 (5th image attached) for new OMC REFL spectrum. This still does not give a good SNR.
I then switched off the OSC_SUM_MATRIX (ie. the output) and changed this line to 11.5Hz at an amplitude of 40000 counts. This does not saturate the OFS PD.
See Ref 2 (6th image attached) for the OMC REFL spectrum.
The IFO moved to NLN and this did not change the PCAL settings I had on.
The third image attached shows the PCALEX and EY settings we used for the DARM offset measurement.
I then ran DARMOffsetStep.py starting at GPS 1368922456 to step the DARM offset.
Jeff reset the PCALEY and EX values using sdf.
xml from tuning measurements is in /ligo/home/jennifer.wright/git/OMC_mode_matching/2023-05-23_DARM_offset_meas.xml
Code is in /ligo/gitcommon/labutils/darm_offset_step
DARM offset results are in data/darm_offset_steps_2023_May_24_00_13_58_UTC.txt
Stay tuned for processed measurements.
Jennie W., Sheila D.
After using an adapted version of Craig and Dan's code to do the analysis I have plotted optical gain (as measured at the OMC REFL PD) while DARM offset is stepped vs. power on the OMC DCPDs, shown in first plot.
The optical gain is determined from the line height of two PCAL lines as measured on the OMC REFL PD, 11.5Hz and 255Hz, unfortunately we were trying to get the measurement done quickly and so did not notch the lower frequency one out of the DARM fb loop.
This plot can be thought of as light rejected by the OMC vs. light passing through the OMC.
We did not get the parabola we expected from this (ignore the fit line).
The second plot is the same optical gain against DARM offset in counts.
Looking at the ndscope of the time series during this measurement (first screenshot), it can be seen that there is a marked upward trend on the PD, presumably due to the fact we started this measurement during OMC whitening so we had not been locked for long and the light on the PD increases with thermalisation in the arms.
The second screenshot shows the noise spectra after the measurement.
Plots are in /ligo/gitcommon/labutils/darm_offset_step/figures/plot_OMC_REFL_vs_dcpd_sum/
data is in /ligo/gitcommon/labutils/darm_offset_step/data/darm_offset_steps_2023_May_24_00_13_58_UTC.txt & darm_offset_steps_2023_May_24_00_13_58_UTC.pkl,
code is in /ligo/gitcommon/labutils/darm_offset_step/plot_OMC_REFL_vs_dcpd_sum.py,
Screenshots are in jennie.wright/git/OMC_mode_matching.
Summary: We can't get anything useful out of this measurement without a quieter PD. Maybe we can get the same info a different method. Also why is the OMC REFL PD so noisy?
