ITF found in LOCKED_ARMS_IR in PREPARE_SCIENCE mode.
In order to have warm ITF for the planned activity on "FIS/FDS injection", I started to lock but unfortunately the NE ALS was in FAULT.
Thanks to Casanueva we recovered the standard state of ALS_NE and I started to relock ITF, after the usual cross-alignment in ACQUIRE_DRMI, up to LN3_ALIGNED.

After few attempts and one unlock event, ITF remained stable locked at LN3_ALIGNED till the unlock occurred at 13:52 UTC when the PR Guardians opened LC, F7 and ID loops. I tried to close them but the signals was very out of standard values. I asked Ruggi to check on it.

Realigning in progress after the recovery of PR suspension.

SUSP
- At 08:25 UTC - under suggestion of Boschi, I disabled PduServer / Set Outlet State
3, 4, 5 of PR_Electr via VPM.
- At 09:30 UTC Boschi chanced frequency modulation of NI

This morning we went to the DAQ room and we re-positioned the fiber that goes to the North Arm until we recovered enough power in the terminal building. The system should be working now.

One remark: the clamp used is a Tektronix Current Probe A621 (datasheet).
The model indicated in entry #68265 was A622 which was wrong, the max range set 0-10A, 100mV/A was correct.

This morning, the NCal and associated PCal lines have been moved from near 18Hz to near 86 Hz.

Lately the monitoring channel for the North ALS has shown some degradation; tonight it went below the threshold which triggers the fault state twice, and the rising wind is not helping. It also shows a very slow oscillation around 13 mHz.

I stopped the lock acquisition at LOCKED_ARMS_IR.

Tonight I found the ITF stuck in DOWN, as the bug impacting the autorelock failsafe of the ITF_CONDITIONS node is still present. I put that node in STOP and the lock acquisition restarted automatically.

Please do not engage the autorelock failsafe until the bug is found and dealt with.

Today we made progress on the work to minimize polarization noise in the fiber optic circuit for the squeezer installed in the rack in Figure 1 in the atrium

1. We have improved the alignment of the polarization to the fast axis of the PM fiber that carries the pick of from the detection to the atrium (acting on the  l/2 and l/4 plates as described  in entry #67833).
2. We have minimized the ellipticity of the polarization exiting the squeezing fiber during heating with a heat gun. This required optimizing the mating connections in the first junction box (figure 2) which is located at the bottom of the rack  This required testing several different mating  and two different fiber splitters. We also had a burn at the output of fiber exit from the injection. This required polishing the fiber with polishing sheets.
3. We replaced the 10 meter long Thorlabs PM fiber between the first ("out0" in figure 2) and the second ("SQZ and phase cam" in figure 3) splitting boxbox with a 2 meter long OZ PM fiber

In the end we achieved a good result up to the fiber output connected in the  ''split box' (see entry #67833 and figure3). The optimization of the alignments beyond this box will be continue in a future shift.

Last Friday (March 13th) we performed a quick test of slowing down the INJ HVAC. The SUPPLY fan was moved from 23 Hz to 17 Hz (-35%), and the RETURN fan from 20.4 Hz to 17Hz (-20%). The test lasted about 15 minutes. This is similar to the test performed in https://logbook.virgo-gw.eu/virgo/?r=60553. 

Figure 1 shows the FREQ,  FLUX and overpressure (PRES) signals. Note that FLUX_EXT and FLUX_IN and FLUX_OUT sensors are evidentely faulty (Davide is notified for replacing those). Sensors measure a reduction of the air flux into the clean areas of: SAS -39%; Atrium(LAR) -34%; LaserLab (LLR) -45%; Minitower(LMR) -55%. And reduction of overpressure of approx 50%. However the duration of the tast has been a bit to short to completely trust these values.

Robert, who was nearby the AHU enclosure, noticed a significant reduction of acoustic and vibration from the fan box itself. 

Microphones in the laser room (LLR), Laser bench (LB) and EIB bench see some reduction:

• approx. -80% in the range 500 Hz -- 2kHz
• approx -20% in the range 1 -- 30 Hz
• not significantly in the range 30 -- 300Hz

We also notice acoustic lines at the fan SUPPLY frequency (23Hz) and third harmonic (69Hz) which are not present when fans are set at 17Hz.

The source has been identified being one vacuum pump (dry pump) in operation at 2400W. The pump was powered by one small dedicated UPS in turn powered from 300W IPS. The pump was switched off on Nov. 3rd at 9:19 UTC. It will be normally kept off.

As a cross check an on/off test can be easily done.

Friday Federico recovered the ENV_CEB_ELECTRIC sensor. However the gain looks a factor 10 larger than before the shut down. The sensor is not saturating, but it might for example during grounding injections. To be adjusted at next access.

ITF found in locked_arms.

At 7:00 UTC ITF in Maintenance mode, below the list of the activity communicated in control room:

• cleaning of the experimental areas.
• tasks performed by the operator:
  • lock of the OMC; by mistake I performed the lock in single_bounce_wi. The scan was skipped in agreement with Michal.
  • TCS thermo camera reference.
  • TCS CO2 power checks:

• from 9:00 UTC the SQZ team started working in FIS/FDS injection activity in the new ITF configuration SINGLE_BOUNCE_SR; the activity will continue in the afternoon.
• from 9:00 Jean Pierre started working in Laser Lab;  the activity will continue in the afternoon.

ACS

This morning we had a failure of the DET ACS; this had a big impact in the humidity of the DET LAB. The problem was fixed by an external firm contacted by HVAC team.

ITF found in SINGLE_BOUNCE_SR in COMMISSIONING mode, with SQZ team at work on FIS/FDS injection activity.

Activity concluded at 19:30 UTC. With Spinicelli and Was we recovered the standard state of INJ and DET.
After a realignment of both cavities, I finally acheived LOCKED_ARMS_IR.

During the test injection, the sensor with the best Signal to Noise Ratio is ENV_CEB_ELECTRIC, followed by ENV_CEB_GROUND_CURR_MON and ENV_CEB_GROUND_MAG.

The magnetometers near the towers all detect the Comb signal, in some cases up to three orders of magnitude above the noise floor.

This is expected: the cable running from the TIRA BAA-120W power amplifier output toward the NI suspension rack and, in parallel, toward the SR tower forms a large loop, which generates a significant magnetic field.

This must be considered when evaluating the effect in Hrec: coupling may arise from the magnetic field rather than from currents in the grounding connections, as the field can act directly on the mirror magnets.

When only real ground-current flow is present, this effect should be absent or strongly suppressed, since no large loop is formed.

Last Friday (13th March) we installed the ground current injection setup (pictures 1-6) and we tested it without ITF. Figure 7 shows a scheme. The injection is performed between the electrical ground of the NI SAT crate and the SR vacuum chamber. The driving signal was one 11Hz comb (saw-tooth by the Hameg generator) amplified to approx 0.5 A and 5 V  (as read in the display of the TIRA BAA-120W amplifier, and also with one ammeter in series with the resistor). Tips by Robert and Federico: do not exceed 1A, the resistor should not become "hot". Be careful not to swap '+' and '-'. The two ends of the cable out of the amplifier (both black) have been labelled. 

The current clamp sensor (ENV_CEB_GROUND_CURR_MON) is described in this entry: https://logbook.virgo-gw.eu/virgo/?r=68265. It was placed around the cable connecting the ground of all platform racks to the CEB hall main ground copper bar. In turn this copper bar is connected to the CEB main copper bar located in the service room next to the main entrance. Finally the latter bar is connected to the safaty earth, which consists of one external ring with ground rods placed all around CEB.

The comb is sensed widespread: by the current clamp (above) and by the ENV_CEB_ELECTRIC probe which is measuring the voltage difference between the WI SAT rack and the WI vacuum chamber.

Figure 8 shows the injection (purple). In the blue curve the current clamp was disconnected from the grounding cable.

After open/closing the SDB1 angular could work again. But then in DRMI 3F the correction diverged, and the loop didn't want to close anymore. After some time I have realized that the floating set point have saved a very misaligned position. Reverting to fixed set point the lock acquisition went to LN3 ALIGNED on the first attempt. I have restored floating set point for SDB1 so that they learn again the correctly aligned position for SDB1.

This afternoon, the NCal and associated PCal lines have been moved from near 102Hz to near 18 Hz.

Indeed, the ITF unlocked at 7.40.55UTC because of another glitch of the NI_F0_ACC (see fig. 1).

After that, the MAR controls of SDB1 opened (and so they are yet), see fig. 2.

Maybe, it is better to leave the ITF in Locked_arms.

The interferometer has been systematically unlocking in LOCKING_DARM_IR this morning, with ~10 failed attempts. I have changed the lock acquisition request to LOCKED ARMS BEAT DRMI 3F.

ITF found in LOCKING_ARMS_IR State and COMMISSIONING Mode.
All time are UTC.
15:00 Activity on Attempt to inject SQZ into the interferometer finished (Vardaro, #68877).
15:20 NEB platform activity (Tringali, FIori).
16:16 Recentered MAIN_PLL_dev DMS flag.
ITF struggled to go over CARM_NULL_1F and/or locking the OMC even after the ending of the activity in the experimental buildings.
18:49 ITF reached LOW_NOISE_3_ALIGNED.
ITF left in LOW_NOISE_3_ALIGNED and COMMISSIONING Mode.

This morning I went in DET e room and i realized that the Frequency generator used as actuator of the CC loop was not setted with DC External modulation. I adjusted it Fig1. I also measured the frequency of the Generator in DET E room and the one in INJ Eroom to check if their frequency where similar

After the first lock I setted the CC DAC offset at 0.045 and I managed to close the CC loop and the AA loop. The magnitude of the CC is still 2mV and before the shatdown was higher than 2.7mV 

We need to realign everything more carefully. Next week we will continue 

ITF found down with the automation blocked: the shutter on B1 was open and DET_MAIN was attempting to close it. 

I manually restored the automation, realigned the cavities, and then in CITF I prealigned the SR and PR; the ITF could not pass CARM_NULL_1F.

At around 8:00 UTC we stopped the lock acquisition because not compatible with the other planned activities:

• grounding injection tests (ENV team);
• site visits associated with the Pisa LVK meeting

At around 13:00 UTC we relaunched the lock acquisition and again we had troubles in locking the OMC; I contacted the DET on-call and he fixed the issue (see OMC demodulation phases adjusted ). 

In parallel Marco worked at FDS recovery.

We noticed that the demodulation phase of the OMC locking error signal B1_PD3 was very mistuned (with the Q signal higher than the I signal). We corrected it by changing the demodulation phase from 0.5 to 1.5 rad. The external configuration file was reloaded at 13h07 utc.

While the ITF was in Low Noise 1, we also adjusted finely the demodulation phase of the B1 error signal, by chaging the phase from -1.3 to -0.7 rad. External configuration file reloaded at 13h17 utc.

As we are considering running with a detuned signal recycling. I went back to this simple parametric model of the DARM response with SR detuned, which has two free parameters, the pole of the response when SR is tuned (in Hz), and the angle of detuning on the circle in the complex parameter space of the pole (which monotonic but not directly related to the SR detuning phase). A question is how to use that model for Hrec, which currently relies on two lines, one at 70Hz and at 360Hz.

Figure 1 shows the changes in the amplitude of the response at 70Hz, for a reasonable range we would explore in both parameters of the model. The changes are very small, at most 1.5%, and depend mostly on the SR detuning angle. The amplitude of that line should remain a good parameter for measuring the absolute scaling of the optical response, that can be affected by the power in the arms, the maching with the OMC, or other type of detection losses.

Figure 2 shows the changes in the amplitude of the response at 360Hz. In that case the amplitude is mostly affected by the pole frequency of the model. In principle it should remain around ~430Hz when detuning, but that gives the freedom for the calibration to follow any changes, for example due to SR misalignment.

Figure 3 shows the changes in the phase of the response at 360Hz. The phase is mostly affected by the SR detuning. So that will be a good measure of the SR detuning for the optical model.

Figure 4 looking at the line at 491.3Hz the phase change of 0.2 rad that was measured during the experiment of Feb 24 corresponds to roughly ~1 of detuning in this parametric model.

This model doesn't include the optical spring at low frequency. In principle the parameters should be related with the SR detuning, in practice I doubt it will be reliable, as the optical spring frequency also depends on the power in the arms, and simulations show it is also affected by the PR and SR RoC mistuning. So there the only solution that seems reasonable to me is to measure the frequency and Q factor of that pole using additional lines at low frequency.

/users/mwas/calib/detunedSRfit/SR_TF_parameter_scan.m