This morning Manuel recovered the NE suspension; after that relocking the arm was very easy.

I restored the normal behaviour of the NARM_LOCK node, then relocked up to LOW_NOISE_3 without any manual operation. AutoScience engaged again.

This morning I restored the threshold value of 0.04 for B1p_PD2_DC in DET_MAIN.ini.

Following the 7.0 earthquake in Canada, the ITF unlocked and, after several failed attempts, the NE controls opened.

I disabled the AutoScience and, in order to keep the automation operational for the West Arm and the IMC, I left everything running but I disabled online the lock enabling flag for the North Arm, and commented line 475 of the NARM_LOCK node in case of future attempts. After the recovery of NE, it should be uncommented, the node reloaded and then the lock acquisition should be started again from scratch.

As a side note, the AutoScience still suffers from the same bug discovered several days ago: when the safety trips. AutoScience is disabled and the fallback state is requested, but both are then reverted and the lock acquisition starts again as usual.

The NCal lines have been set in the 18-18.32 Hz range around 22:45 UTC, while the ITF was unlocked.

Yesterday, similarly to the shift on Dec 3rd, we synchronized the HWS-INJ acquisitions with the CO2 PA switched on, aiming to identify the PA position on the WI HR surface.
An example of the wavefront recorded with the PA projected on the WI HR surface for each tested CO2 PA position, is reported in Fig. 1.

Figure 2 shows a summary map of the CO2 PA positions for both the December 3 and December 5 shifts. The origin (0,0) corresponds to the maximum YAG-induced Delta OPL, which is assumed to represent the mirror center.

The X and Y coordinates of the CO2 PA were evaluated from the HWS maps by identifying the minima of the Delta OPL.
Only the Y coordinates of the CO2 PA positions, evaluated from the HWS maps, were flipped to be consistent with the thermal camera installed in front of the viewport which observes the HR surface at a horizontal angle of 33° but at the same height as the mirror center (i.e., the ZnSe viewport installed to monitor inside the tower, not the one used to inject the heating profile generated by the PA mitigation system). By flipping the Y coordinates of the HWS maps, when the CO2 PA moves from top to bottom on the thermal camera (and consequently on the mirror), the same movement is observed in the HWS maps. Regarding horizontal displacements, when the CO2 PA moves from left to right on the thermal camera, the same movement is observed on the HWS maps without the need to flip the X coordinates.

ITF found DOWN in COMMISSIOING Mode.
All times are UTC.
ITF failed to relock at first attempt at LOCKING_CARM_MC_IR. During the second attempt I added 2 urad to SR TY in 120 s at the end of CARM_NULL_1F state. LOW_NOISE_2 reached at 15:42. Unlocked at 16:05.
ITF failed to relock at first attempt at LOCKING_CARM_NULL_1F. During the second attempt I added 2 urad to SR TY in 120 s at the end of CARM_NULL_1F state. LOW_NOISE_2 reached at 16:48. Unlocked at 17:19.
ITF relocked at first attempt in LOW_NOISE_2 at  17:51, I manually added 2 urad to SR TY in 120 s at the end of CARM_NULL_1F state. Unlocked at 18:09.
Commissioing activity ended at 18:20 when Melo and Spinicelli went in CEB to turn off IPATSiA setup. 
ITF failed to relock at first attempt at LOCKING_CARM_MC_IR. During the second attempt I added 2 urad to SR TY in 120 s at the end of CARM_NULL_1F state. LOW_NOISE_3 reached at 19:07 and SCIENCE mode set.

At 14:56 recovered SQB1_SBE vertical.

Today we continued to investigate the IPATSiA effects on the ITF through the PA injected on the WI HR surface mirror. The last IPATSiA shift was reported here #68304.

In the morning, at 9 UTC we switched on the chiller, the laser and started shining the mirror with 32 mW of CO2 power at 10:35 UTC when the ITF was stabilized at LN2. 

In Fig.1 we can see the new position of the PA (POS 1). To be noted that this position is different from all positions from the other shifts. We moved in the horizontal direction to the left and the idea is to scan some points in the vertical direction. Also to be noted that the thermocamera used is a different one w.r.t the other shifts. Since the thermocamera used before had an issue and we had to restart it every few seconds, we installed a new one, but this time facing the HR mirror with the same height and therefore there is no vertical angle of view. Moreover, this camera is not flipped upside down as the other one (this was necessary to install it in the setup), so to understand the relative position of the PA it is necessary to keep this in mind (see Fig. 2: here we have the old thermocamera image of the POS1, i.e., the same position of the new thermocamera shown in Fig. 1). 

When we started shining the mirror a big oscillation in all signals of the ITF (PR, BS, arms signals, ect) unlocked the ITF. After locking again in LN2, we tried again (started shining the mirror at 11:27 UTC) but we had the same effect as before so we decided to stop shining the mirror (11:32:37 UTC) and change position (+10k steps of the picomotor, corresponding roughly to 8 mm dwon w.r.t the previous position). 

So, here below a list of the positions and corresponding UTC time. Each position is 8 mm down w.r.t the previous one:

- At 11.45 UTC we started shining the mirror in POS2 (Fig. 3);

- At 12:52 UTC we moved to POS3 (Fig. 4);

- At 13:53 UTC we moved to POS4 (Fig. 5);

Then, at 14.55 UTC we moved the PA horizontally in the mirror towards the right direction - POS5 (Fig. 6). However the ITF unlocked and we stopped shinning the mirror at 15 UTC. After some atempts to shine the mirror with the PA, the ITF did not manage to remain locked in LN2 in this position. At 17.36 UTC we moved again the PA upwards (8mm up) to try in this new position. The ITF relocked in LN2 and we tried to shine the mirror at 17.59 UTC but the ITF unlocked again at 18.09.15 and we stopped the shift here.

Around 18.15 we switched off the CO2 laser and the chiller.

On Fig. 7 we show a summary of the main signals during the whole shift. 

The NCal lines have been set in the 92-192.32 Hz range around 17:25 UTC, while the ITF was unlocked. NEN and NWN are left at 36.04 and 36.08 Hz

**** ENV noise injection files - Saturday, 27 November 2025 ******

Sweep magnetic injections /virgoData/NoiseInjections/MagneticInjectionsO4/output:

• CEB: MagneticSweep_NOISE_MAG_CEB-1448451700.txt
• NEB: MagneticSweep_NOISE_MAG_NEB-1448452630.txt
• WEB: MagneticSweep_NOISE_MAG_WEB-1448453280.txt
   

Acoustic injections /virgoData/NoiseInjections/AcousticInjectionsO4/output:

• DET Lab (value sent to DAC= 0.08):  AcousticColored_NOISE_CEB_DetLab_speaker_1-1448454972.txt
• CEB hall (value sent to DAC= 0.04): AcousticColored_NOISE_CEB_DetTerrace-1448453935.txt
• Laser Lab (value sent to DAC= 0.01): AcousticColored_NOISE_CEB_EEroom-1448455319.txt
• NEB (value sent to DAC= 0.1): AcousticColored_NOISE_NEB-1448454278.txt
• WEB (value sent to DAC= 0.01): AcousticColored_NOISE_WEB-1448454623.txt

One needs to be wary of using the phase camera for quantative analysis as the phase camera status is not being monitored.

Figure 1. After the EDB OMC installation in April the B1p phase camera has been realigned, but then the alignment has drifted over the following few months, with the power on the phase camera dropping by a factor 3, likely the reference beam is not well centered on the photodiode anymore. The normalized channels remove some of that effect, but it may biased for large changes in power by a factor few, and the normalized channels may not be the ones used in all analysis. It is probably the explanation for the trend in carrier contrast defect reported by Fiodor in VIR-1178A-25.

The conclusion that in LN2 a significant fraction of the power on the dark fringe is due to the 56MHz sidebands is in disagreement with the EDB OMC measurements.

Figure 2 shows the EDB OMC transmitted power when locked on the 56MHz sidebands and on the higher order modes. The OMC alignment was performed in 4 degrees of freedom by maximizing the 56MHz USB transmission, and minimizing the beam jitter coupling when locked on that mode.

Figure 3 shows that during that measurement the total power arriving on the EDB OMC was ~12.5mW, while there is ~0.5mW in transmission of the OMC when it is locked on the 56MHz LSB or USB. This would mean that the total power of the sidebands is ~1mW, or 10% of the total power on the dark fringe.

Figure 4. During the EDB OMC installation it was locked in single bounce, the alignment was not perfect, but it shows that the cross-calibration of B1s and B1t photodiodes is roughly correct, when 2.2mW of power disappears from the OMC reflection (B1s) it appears instead on the OMC transmission (B1t).

Some of the 56MHz sideband power can be in higher order modes instead of the TEM00, but that should not be significant as the other higher order modes shown on Figure 2 have power less than 0.1mW, so less than 1% of the total power arriving on the OMC. Note that the unlabelled mode just right from the 56MHz LSB is the carrier order 1 mode.

Piernicola has investigated an unlock that happened during a previous IPATSIA measurement on Dec 03, at  ~10h06m05 utc. It turns out that the unlock was triggered by DET_MAIN who detected too much power on the B1p_PD2 photodiode according to its log file:

2025-12-03-10h06m04-UTC>WARNING-[SHUTTER_OPEN.run] USERMSG 0: Too much power on B1p PD2! unlock ITF and close fast shutter

Indeed, the power on B1_PD2_DC exceeded the threshold of 0.04 set in the DET_MAIN.ini file as shown on the attached figure.

In order to give a bit more of flexibility for the on-going shift, we have increased the threshold to 0.045 today at 11h03 utc. With such a threshold, the power on B1_PD2 will trigger the unlock at approximatively the same time as the power on B1s.

We agreed that the threshold will be lowered back to 0.04 at the end of the shift.

ITF Found in LOCKING_DARM_IR and COMMISSIONING Mode.
During the relock added 2 urad SR TY in 120 seconds to lower SDB2_B1p fringe. Requested LOW_NOISE_2 achieved at 15:25 UTC. 
RAMS noise injections activity ongoing (De Rossi, Gosselin, Spinicelli, Nocera, #68313).
At 16:52 UTC Requested LOW_NOISE_3. Request achieved at 17:02 UTC.
At 17:25 UTC I manually unlocked the ITF and put INJ_MAIN node DOWN to allow Spinicelli and De Rossi to go in CEB to disconnect RAMS noise injection setup.
Relock started at 17:40 UTC, LOW_NOISE_3 achieved at 18:23 UTC. SCIENCE Mode set at 18:24 UTC.
ITF left in LOW_NOISE_3 and SCIENCE Mode with Autoscience ON.

Guard tours:
UTC 21:05 -> 21:38

After completing the realignment activities and confirming that we could lock the ITF until LN2, we went in piscina to install the necessary hardware for the RAM noise injection. We set it up for 6 MHz and corrected the phase shift on the LNFS as explained in entry 62346.

Between 12:30 and 15:30 UTC, we attempted to relock the ITF. Initially, there were telecommunication issues with the LNFS server and Metatron. The ITF unlocked twice—first at 3f and then at 1f—before eventually locking at LN2 around 15:30 UTC.

The DAC dedicated to the RAM noise injection was used to do the noise injection. We injected noise in two phases: first, a line at 111.1 Hz, followed by colored noise both in LN2 and in LN3. There is a 9 V offset on the DAC to ensure proper operation of the input mixer of the electronis for noise injection However, since the DAC is limited to 10 V, we were unable to observe any noticeable effects on the ITF signals for the line injection when the amplitude was less than 1 V or for colored noise that did not exceed 10 V.

Instead when we exceeded the DAC’s 10 V limit, it began to saturate, and as a result, we were injecting not only the 111.1 Hz line but also its harmonics, due to signal distortion. In this scenario, we were able to observe clear effects on the ITF signals. I’ve attached plots of these effects. The times of the most relevant injection are at the bottom of the entry. Additionally, I am attaching the log file from VPM with the GPS time and noise amplitude data.

At this point, it's difficult to draw any conclusions. We will better analyze the data to see if there is something to understand, maybe upper limits by using the INJ_EOM_6MHz_mag signals.

We would need to repeat these measurements. One potential solution could be to use a dedicated waveform/noise generator at the input of the injection electronics and acquire the data on an ADC for reference.

#### LN2 ####
15:48:40 line 111Hz ampl 3 dur 1 min
16:50:20 ampl 2.8 dur 2 min
16:52:40 ampl 2.5 

#### LN3 ####
17h11 linea 111Hz LN3 5 min
18:17:10 bandpass  111-333 Hz 1e-2 2 min
18:19 bandpass  8e-3 2 min
18:21 bandpass 6e-3

This morning, we performed a general realignment on the laser bench.

We first attempted to improve the beam alignment into the Neovan head using the upstream mirror, in order to increase the DC output power and possibly reduce the temperature, as previously observed in entry 67117. We tested both horizontal and vertical alignment directions but could not obtain any improvement (see Fig. 1).. Achieving a better alignment—if required—would take more time, using both mirrors and additional figures of merit such as the beam shape.

We then realigned the beam inside the AOM to try to recover a better situation after the slight degradation noted following the last alignment (entry 68148). We placed a mirror in front of the water-cooled beam dump, which receives both the PMC reflection and the AOM first-order beam. This mirror redirected the beam onto a high-power meter. The goal was to maximize the AOM first-order mode. Starting from about 3 W, we first adjusted the horizontal axis: as we moved toward the edge, the first-order power increased to about 4 W. Before reaching a clear maximum horizontally, we also adjusted the vertical axis. Larger vertical corrections pushed the power up to 5 W without indicating proximity to a maximum.

At that point, we realized that the PMC had unlocked due to misalignment. The observed power increase could therefore have been partly due to the PMC reflected beam hitting the power meter, as separating the two beams was difficult under these conditions.

We therefore stopped the AOM alignment and focused on relocking the PMC. We acted on both mirrors to correct for both shift and tilt misalignments, and eventually reduced the PMC first-order misalignment mode to below 1%.

We then relocked the injection to evaluate the effects of the work performed. The main result is an improved PMC transmission and reduced low-frequency noise in the PMC TRA and REFL signals, as well as in the PSTAB signals (see Fig. 2 and Fig.3). Given this improvement, we did not resume the AOM alignment. We relocked the interferometer up to LN2 and prepared the remaining activities of the day related to RAMs noise injection (see dedicated entry).

Note for the next AOM alignment:
- It is possible to separate the AOM first-order beam from the PMC reflection by using the mirror located in front of the prism.
- There is a translation stage below the AOM in order to perform a pure shif

The NCal lines have been set in the 104-104.32 Hz range around 14:30 UTC, while the ITF was unlocked. NEN and NWN are left at 36.04 and 36.08 Hz

ITF found locked at LN3 in SCIENCE mode. (AUTOSCIENCE ON).
At 07:33 UTC the planned acrivity on Neovan + AOM + PMC realignment (Gosselin, DeRossi, Melo) started.
COMMISSIONING mode set.
I manually unlocked ITF, INJ_MAIN in pause.
At 10:00 UTC ITF back in operation. After the usual cross-alignment in ACQUIRE_DRMI, ITF acheived LN2 at 1st attempt at 10:33 UTC.
Upon concluded the checks on INJ we manual unlocked in order to allow the second part of the shift on: RAMS noise injection (Gosselin, DeRossi, Melo).
Activity concluded at 13:00 UTC.
Before relocking we noticed a failure in the communication LNFS / INJ_MAIN node (No collect data). We acted on VPM:
- Remove SMS [SMS server=Lnfs100]' sent to FbsISC
- Reload Configuration' sent to FbsISC

During the various relocking attempts, I had to open shutters and re-anabled Vbias' for B1p_QD1, B1p_QD2, B5_QD2, for several times.
Relocking in progress...

Parallel activity:
from 10:30 to 11:30 UTC, Dattilo and Passaquieti done the planned activity on Ground current measurement.

I ve changed the WI controller, which was oscillating, to be equal to the NI to better understand also the difference between the two plants

ITF found locking to LN2 with IPATSiA activity ongoiong. Since SR had just been aligned, it required some misaligning (+2 urad ty over 120 s) during CARM_NULL_1F timer.
15:30 UTC ITF in LN2, 30 Mpc
19:08 UTC end of IPATSia activity
19:21 UTC ITF in SCIENCE, autoscience ON
22:00 UTC ITF left in SCIENCE

Since the BS Voltage Monitor were not connected when the board was replaced a few days ago, the DMS flags/emails associated to the BS Voltage Monitoring have been shelved, until January13th (under Calib_Hrec / CalBS ).

On site: Suzanne, Piernicola, Manuel

Remote: Luciano, Ilaria

This morning, we performed HWS-INJ acquisitions to measure the optical path variation induced by the CO2 PA.

The HWS-RC acquisitions were done for different CO2 PA positions. 
Before starting, we moved the CO2 PA in the opposite direction wrt the Position 2 described in the entry 68291, we will called it Position 4 (see fig. 1).  

We started by applying approximately 40 mW, instead of the 50 mW used in 68291.

Position 4   

The HWS-INJ SLED ON at 1.63 V.

The CCD illumination is shown in Figure 2. 

ITF configuration: LN2 achieved at 09.20 UTC

Start acquisition:  09.56 UTC (Folder 20251203T1056)

IPATSiA ON: 10.01 UTC (WF 13)

Unlock at 10.07 UTC (WF 26)

IPATSiA OFF: immediately after the unlock

HWS-INJ acquisition stopped at WF 27.

The CO2 PA immediately appears in the HWS-INJ map (see Fig. 3). For convenience, the same figure also shows the HWS-INJ image with the YAG on the WI mirror.

%%%%%%%%%%%%%%%%%%%%%%%%

Since the impact of the CO2 PA was destructive and the ITF unlocked very quickly, we decided to move the CO2 PA to Position 1 described in entry 68291, as in this position the ITF was able to maintain the lock.

%%%%%%%%%%%%%%%%%%%%%%%%

Position 1-HP (the same Position 1 as during Monday’s shift)

ITF configuration: LN2 AT 10.39  UTC

Start acquisition:  10.50 UTC (Folder 20251203T1150)

IPATSiA ON: 11.01 UTC (WF 29)

Unlock at UTC 11.05 (WF 41)

HWS-INJ acquisition stopped at WF 42.

The CO2 PA immediately appears in the HWS-INJ map (see Fig. 4). 

%%%%%%%%%%%%%%%%%%%%%%%%

Again, the ITF unlocked very soon. So we decided to decrease the power injected up to 10 mW.

%%%%%%%%%%%%%%%%%%%%%%%%

Position 1-LP (same as before, but with 10 mW of injected power)

ITF configuration: LN2 achieved at 15.30  UTC

Start acquisition:  15.44 UTC (Folder 20251203T1643)

IPATSiA ON:  15.48 UTC (WF 12)

The CO2 PA immediately appears in the HWS-INJ map (see Fig. 5).

%%%%%%%%%%%%%%%%%%%%%%%%

Without closing the flip mirror, at 16.32 UTC (WF 129), we moved the CO2 PA in Position 4-LP.

After half an hour (at 17:03 UTC, WF 212), the CO2 PA was moved in the same direction as before by the same number of steps (10000) [POSITION 5-LP].

After another half hour (at 17:36 UTC, WF 300), the CO2 PA was moved again in the same direction by the same number of steps (10000) [POSITION 6-LP].

%%%%%%%%%%%%%%%%%%%%%%%%%
IPATSiA OFF: 18.07 UTC (WF 383) [FLIP MIRROR CLOSED]

The HWS maps for each data set are summarized in Figure 6.
Then, Piernicola went to the CEB to switch off the entire IPATSiA setup (laser+CH).
He came back at 19.08 UTC / activity concluded.

In this log entry, we use the following naming convention for surveyed points (three per platines): 

Each platine is named after the first two letters of the near rotor + 1 character which is "T" for the point above the rotor, "N" for the surveyed point nearest to the Near rotor, "F" for the last surveyed platine point (further away fro the near rotor)

The "transation" between  this convention and the one from the log entry 68205 is as follows:

WNT = WE_Ncal_NB_06-T
WNN = WE_Ncal_NB_03-SW
WNF = WE_Ncal_NB_04-NW

WET = WE_Ncal_NB_05T
WEN = WE_Ncal_NB_02-SE
WEF = WE_Ncal_NB_01-NE

WWT: WE_Ncal_SB_05-T
WWN: WE_Ncal_SB_04-NW
WWT: WE_Ncal_SB_03-SW

WST = WE_Ncal_SB_06-T
WSN = WE_Ncal_EB_01-NE
WSF = WE_Ncal_EB_02-SE

In this log entry, we use the following naming convention for surveyed points (three per platines): 

Each platine is named after the first two letters of the near rotor + 1 character which is "T" for the point above the rotor, "N" for the surveyed point nearest to the Near rotor, "F" for the last surveyed platine point (further away fro the near rotor)

The "transation" between  this convention and the one from the log entry 68206 is as follows:

NET = NE_Ncal_EB_05T
NEN = NE_Ncal_EB_03-SW
NEF = NE_Ncal_EB_02-SE

NWT: NE_Ncal_WB_05-T
NWN: NE_Ncal_WB_01-NE
NWT: NE_Ncal_WB_04-NW

ITF found relocking arms but N cavity was misaligned. I realigned it and restarted the lock acquistition up to LN2 for the planned LN2 IPATSiA activity (Melo, Spinicelli).
After the usual cross-alignment in ACQUIRE_DRMI, I misaligned SR about 2 urad in ty+ at CARM_NULL_1F.
COMMISSIONING mode set. LN2 acheived at 07:41 UTC at 1st attempt.
08:00 UTC started the activity.

Unfortunately ITF is particulary unstable.
After several unlocks and relocks, at 13:07 UTC the unlock opened the WI LC loop. Properly closed.
Relocked at LN3. Activity in progress...

Since few days, the BPC TY corrections were quite high, very close to their limit. 

In order to reduce them, at around 11.20UTC we moved the PR_X by +40µrad, centering at the same time the beam onto the ITM.