Hirose, Yokozawa, Saito, Kawakami, Tanaka
We calibrated the 1f signals to nano-meter unit by comparing TFs from actuators to 1f sensors with ones from actuators to LEN OPLEVs. We stopped today's commissioning due to the large earthquakes at Kumamoto, unfortunately.
## What we did
- We tried to calibrate by comparing the response from {SRCL,PRCL] feedback to 1f sensor (K1:LSC-{SRCL,PRCL}1_IN1/K1:LSC-{SRCL,PRCL}2_OUT) with the response from {SRM, PRM} TEST to {SRM, PRM} LEN oplev (K1:VIS-{SRM, PRM}_TM_OLDAMP_IN1_DQ/K1:VIS-{SRM, PRM}_TM_TEST_L_IN2), which was measured in Health check (PRM in klog33132, SRM in klog36856).
- The measured frequency range of the 1f sensor response (10-100 Hz) differs from the range of the OPLEV response (0.01 - 7 Hz). But above the resonant frequency, which is less than 1 Hz, the frequency response is f^-2. So we estimated the Oplev response value around several tens Hz from the value around several Hz.
- PRM: Oplev response at 3 Hz = -80 dB, that is, the response at 30 Hz = -120 dB. On the other hands, PRCL 1f sensor response at 30 Hz = -80 dB (green line in right upper panel in fig.1). So we obtained the calibration factor of the 1f sensor, -120 - (-80) = -40 dB = 0.01 um/cnts. Also, we converted from um to nm, 0.01*1000 = 10 nm/cnts. The relative phase from oplev to sensor is 180 deg. However, we found that the PRCL -> PRM element in LSC_OUTMTRX is -1. In this case, The actual direction of PRM Longitudinal direction is inverse that the we assumed. So the sign of the input element value should be positive in terms of the "VIS" geometry.
- SRM: Oplev response at 4 Hz = -101 dB, that is, the response at 40 Hz = -141 dB. On the other hands, SRCL 1f sensor response at 40 Hz = -92 dB (green line in right upper panel in fig.2). So we obtained the calibration factor of the 1f sensor, -141 - (-92) = -49 dB Also, we converted from um to nm, -49dB + 60dB = 11 dB = 3.5 nm/cnts. The relarive phase is 0 deg. So we applied the sign of the input value to positive.
- Then, we modified the LOCKING_DRMI_1F state in the VERTEX guardian to input these calibration factors into the LSC INPUT MATRIX.
- Moreover, we adjusted the each setpoint of MICH, PRCL, and SRCL and the filter gains in FM4 of {MICH, PRCL, SRCL}2 because we changed each unit.
- The measured frequency range of the 1f sensor response (10-100 Hz) differs from the range of the OPLEV response (0.01 - 7 Hz). But above the resonant frequency, which is less than 1 Hz, the frequency response is f^-2. So we estimated the Oplev response value around several tens Hz from the value around several Hz.
- After that, we measured the OLTFs and we confirmed the overall gains were not changed (PRCL:fig.1, SRCL:fig.2).
- Then, we tried to measure the sensing matrix of 3f signals but the large earthquake occured at Kumamoto. So we gave up today's commissioning.
## Note
After today's work, I heard the reason why the sign of the value in OUTMTRX from PRCL to PRM is -1 from Yamamoto-san. It is because the definition of the sign of the cavity length change is that the direction which the cavity length expands is postive. So, the positive direction in terms of "LSC" is inverse from the VIS geometry. Therefore, we should change the sign of PRCL, SRCL factors from positive to negative.
