[Ushiba, Sugimoto, Yokozawa, Hirose]

1. OMC lock recovery

Initially, the OMC was unable to lock. An investigation revealed that the CLC_GAIN value for the OMC PZT (LSC) dither had been set to 0. After enabling it, the OMC was successfully locked.

2. MICH 3f lock commissioning

We first attempted to lock PRMI, but the lock was unsuccessful. Therefore, we switched to MICH locking.

Since MICH locking will also be required for future CAL measurements, we checked whether MICH could be locked by requesting the corresponding state through the VIRTEX Guardian.

Initially, MICH 3f could not be locked. We made the following adjustments:

1. Dark offset removal

   • Removed the dark offsets from REFL51 I/Q and POP17 I/Q.

2. Demodulation phase adjustment

   • Roughly adjusted the REFL51 demodulation phase while monitoring the signals with ndscope, so that the I-phase signal was minimized.

3. Filter adjustment

   • The feedback signal showed large fluctuations and high-frequency noise in ndscope. We changed the low-pass filter from ELP500Hz to ELP100Hz.(The lamp time for these filters was set to ‘Immediate’.)

4. Guardian adjustment

   • Removed the threshold decorator used in the MICH locking sequence.

5. Low-frequency gain adjustment

   • Although the MICH resonance could be caught, the lock was unstable and the loop gain appeared to be low. So we removed the Lowpass(z1p0.1) filter. After this change, the MICH lock became stable.

6. Optical gain recovery

   • Measured the OLTF and found that the loop gain was still low. We adjusted the HWP(IFO-REFL) angle to 140 deg, where the Xarm power was maximized. The loop gain recovered after this adjustment.

3. MICH 3f calibration and UGF adjustment

Using the response of the BS actuator measured in the vertical (L) direction of OPLEV, the transfer function of MICH OUT2IN1 was calibrated to nm/cts. The BS longitudinal displacement is defined along the 45-degree direction. Therefore, when the BS moves by ($l_{\mathrm{BS}}$), the corresponding MICH displacement is $l_{\mathrm{MICH}}=l_x-l_y=\sqrt{2}l_{\mathrm{BS}}$.

The BS OPLEV longitudinal actuator response at 10 Hz was -127.96 dB in units of µm/cts. To convert this response to the MICH displacement in nm/cts, we applied the factor ($\sqrt{2}\times1000$), corresponding to
$20\log_{10}(\sqrt{2}\times1000)$=62.99 dB .
Thus, the MICH actuator response at 10 Hz is -127.96+62.99=-64.97 dB (nm/cts).

We measured the MICH OUT to IN1 transfer function and applied the calibration correction based on the difference between the measured response and the expected actuator response.
After calibration, we readjusted the MICH loop gain to achieve a UGF of approximately 20 Hz(Fig1). 
Finally, we carried out a phasing check and confirmed that REFL51 was at its minimum on Q phase.(FIg2)

4. MICH 1f lock commissioning

We also tested MICH 1f locking.

Unlike MICH 3f, MICH 1f was successfully locked simply by requesting the corresponding Guardian state, without additional adjustments.

We measured the OLTF and confirmed that the OUT2IN1 transfer function was consistent with the expected calibration in nm/cts.

Finally, we adjusted the MICH loop gain to achieve a UGF of approximately 9 Hz(FIG3), following the previous settings reported in klog29949 and klog29950.
Following the same procedure as on the 3f, we carried out a final phasing check and confirmed that POP17 was at its minimum on Q.