I took a PCAL to DARM broadband injection earlier this evening, and when I did I noticed that the front end CAL-DELTAL_EXTERNAL calibration was not great in the bucket. We were underestimating DARM meters by 10% at 70 Hz.
I spent a long time playing around with front end calibration gains in an effort to improve the front end calibration. I was largely unsuccessful, as we lost lock before I could find a reasonable calibration.
Things are mostly within 10% now, but still underestimating DARM meters at 70 Hz by about 8%.
The calibration from this afternoon (after Stefan played with these gains) is in red in the plot below, while the one now is in blue.
Cal settings This afternoon Now
----------------------------------------------------------------
H1:CAL-CS_DARM_ANALOG_ETMX_L1_GAIN 1 3
H1:CAL-CS_DARM_ANALOG_ETMX_L2_GAIN 1.65 1.3
H1:CAL-CS_DARM_ANALOG_ETMX_L3_GAIN 1.24 1.2
H1:CAL-CS_DARM_ERR_GAIN 1.175 1.175
Here are the afternoon values, including a PCAL sweep:
Calibration gains at GPS 1235532403, Mar 02 2019 03:26:25 UTC :
H1:CAL-CS_DARM_ERR_GAIN = 1.11
H1:CAL-CS_DARM_ANALOG_ETMX_L3_GAIN = 1.073
H1:CAL-CS_DARM_ANALOG_ETMX_L2_GAIN = 1.0
H1:CAL-CS_DARM_ANALOG_ETMX_L1_GAIN = 1.0
Attached is the PCAL sweep done with these values. It lives here:
/ligo/svncommon/CalSVN/aligocalibration/trunk/Runs/O3/H1/Measurements/FullIFOSensingTFs/2019-03-01_H1_PCAL2DARM_TF_5t1100Hz_8min_AFTER_CHANGE.xml
Note that is was within 5%, underestimating the range contribution at 80Hz, but overestimating it at 200Hz. Usually 80Hz is more critical for the range, so I would expect the actual range larger than the reported range. The reported range for this was around 104 Mpc.
The values Craig reported were an attempt to even out the frequency-dependent response, but we lost lock during the process. Since these values make more sense that what Craig mentions above, and gave <5% errors, I set them back to their values from GPS 1235532403, Mar 02 2019 03:26:25 UTC.
I should also say that this measurement was done after Rick's work (alog 47215)
