KAGRA Logbook

[Yokozawa, Ushiba, Dan]

Summary

In order to make more phase margin for `ALS_DARM_LOCKED`, we changed some filters:
- ELP200 in `K1:LSC-ALS_DARM`: 200Hz -> 400Hz
- resGs in `K1:LSC-DARM2`: removed 7.44Hz resG
Check the cross over freq with `HANDOVER_ALS_TO_IR`
- Normalization factors, gain of TRAS FB, and calibration of TRAS.
- We changed the normalization factors: `K1:LSC-TR_X_NORM` and `K1:LSC-TR_Y_NORM`
- Also changed `K1:LSC-TR_IRX_NORM` and `K1:LSC-TR_IRY_NORM`

Details

Make more phase margin for `ALS_DARM_LOCKED` or reduce the rms.
- Feedback to ETMX TM actuator should be lower than 100,000. (`K1:VIS-ETMX_TM_DRIVEALIGN_LOUT_DQ`)
- State: LSC_LOCK = ALS_DARM_LOCKED.
- Change ELP200 in K1LSC-ALS_DARM to higher. Original: `ellip("LowPass", 2, 1, 100, 200)`
- We checked the filters in the past.
- We removed `resG_low` in `K1LSC-DARM1`. This was not ON in the past state.
  - Measured OLTF anagin, and then the lines was fitted on the past one.
- Change ELP200
  - `ellip("LowPass", 2, 1, 100, 250)` -> 1 deg changed at 20Hz which is the UGF of the DARM loop.
  - `ellip("LowPass", 2, 1, 100, 400)` -> 2 deg changed at 20Hz which is the UGF of the DARM loop.
- Removed FM5 = resGs in DARM2 FB
- Spectrum measurement: there is no peak at 7.44Hz.
- Removed resG at 7.44Hz in FM5 in DARM 2 FBs
- OLTF measurement: `OLTF_ALS_DARM_rmresglow_ELP400_rm7p44resG.xml`: Fig001
Check the cross over freq
- Requested `HANDOVER_ALS_TO_IR` with `-1` setting at counter == 4. Made this Guardian change back after our works today.
- Because the diff of S_{ALS} and S_{AS} is only F_{TRAS}, if the calibrations are correct and norimariation factors are correct, the S_{AS}/S_{ALS} should be equal to F_{TRAS}:
  - LSC_OVERVIEW -> TRX_A_DC, TRY_A_DC -> Right bottom `K1:LSC-AS_TRNORM_PDA1_RF17_Q`: Fig002
  - The filter shows G=0dB is at 3Hz.
- Measured OLTF of IN1/IN2 of `K1:LSC-AS_TRNORM_PDA1_RF17_Q`, which is -S_{AS}/S_{ALS}: Fig003
  - It means about 5dB difference between the F_{TRAS} and the measurement.
- Inject a sin into DARM to check the calibration diff between displacement signal of S_{ALS} and S_{AS}
  - Injection point: `K1:LSC-DARM1_EXC`
    - 4Hz
    - Gain = 1 <- too large
  - Injection point: `K1:LSC-DARM1_EXC`
    - 100Hz -> 98Hz
    - Gain = 1e-3 -> 2e-3
  - Check the signals: `K1:LSC-AS_PDA1_RF17_Q_NORM` and `K1:ALS-DARM_OUT_DQ`
    - Check the ratio of the peak: OK
- Make a good normarization factors
  - INITIAL_ALIGNMENT -> XARM -> From top(request IRY_LOCKED) to the end(offload)
    - Check `K1:LSC-TR_X_NORM` in `K1LSC_TRCARM.adl`
    - Average the input of the filter `cdsutils avg 10 K1:LSC-TR_X_NORM_IN1` -> 3.4813
    - Then, make 1 over the value: 1/3.4813 = 0.2872
    - Change the FB `K1:LSC-TR_X_NORM` to this value: 0.2872
    - Remove the offset of `K1:LSC-TR_X_LP`
  - INITIAL_ALIGNMENT -> YARM -> From top(request IRY_LOCKED) to the end(offload)
    - Check `K1:LSC-TR_Y_NORM` in `K1LSC_TRCARM.adl`
    - Average the input of the filter `cdsutils avg 10 K1:LSC-TR_Y_NORM_IN1` -> 17.6035
    - 1/17.6035 = 0.0568067 -> FB of `K1:LSC_TR_Y_NORM`
    - Remove the offset of `K1:LSC-TR_Y_LP`
  - Also we changed the norm factors at `K1:LSC_TRGR_TRIR.adl` for IR: `K1:LSC-TR_IRX_NORM` and `K1:LSC-TR_IRY_NORM`
- Change 0.840 at the gain of the TRAS FB to be 1
- Measure the -S_{AS}/S_{ALS} again: Fig004
  - Now the 0dB is about 2.1Hz, which is closer to 3Hz.
- Spectrum measurement with ALS_DARM_LOCKED: Fig005
  - Error and feedback signal measurement
  - Compare to that before our filter changes.
  - It seems ELP200Hz is better than ELP400Hz from the point view of rms.
  - So we made it back. (We forgot measuring the OLTF with this condition.)