After performing an analysis of the WI thermocamera images (see #69785), a similar analysis for the NI was done.

• Reference Image: The analysis utilizes the thermal image containing the mirror outline and the identified point absorber location (marked at pixel coordinates [X, Y] = [310, 199]), see Fig. 1.
• Geometric Reconstruction: The mirror boundary (D = 35 cm) was fitted to extract the center in pixel coordinates, yielding C = [346, 204].
• Perspective Correction: Taking into account the viewport entrance angle of 18.2 degrees from the left (along the decreasing columns axis), the horizontal axis pixel scale is corrected for perspective compression.

Geometric Mapping:

• X-Axis (Columns): The mirror center is located at column 346, whereas the point absorber is at column 310. This places the point absorber to the left of the center (an offset of -36 pixels in the image, corresponding to approximately -5.7 cm in physical space).
• Y-Axis (Rows): The mirror center is located at row 204, whereas the point absorber is at row 199. This places the point absorber slightly above the center (an offset of -5 pixels in the image, corresponding to approximately -0.8 cm in physical space).
• Radial Distance: The total estimated physical distance of the point absorber from the mirror center is approximately 5.8 cm.

Figs. 2 and 3 display the MATLAB output, showing a differential temperature of approximately 0.52 degrees for the point absorber (calculating by subtracting the image of the mirror in CARM_NULL and the image with the ITF in DOWN status). No other strong PA has been spotted. 

The position of the PA from the HWS-INJ measurement was evaluated.

Figure 1 shows the position of the OPL maximum as a function of the wavefront number. Figure 2 shows an example of the HWS map, where both the PA position and the mirror center, previously identified using the RH/CH, are indicated.

The PA position relative to the mirror center is:

X_PA = [14 +/- 1] mm
Y_PA = [4 +/- 1] mm

This result is not consistent with LMA and the thermal camera image. In particular, according to the HWS measurement, the PA is displaced from the mirror center mainly along the horizontal direction, and not in the vertical one

One possible explanation is a change in the SLED beam position on the mirror between the two configurations used for the measurements: the PA position was evaluated at CARM NULL, while the mirror center was identified in WI single bounce . In Fig. 3, the pink curve (04/09/2026) represents the position of the main mirrors in the WI single-bounce configuration when we performed the measurement to identify the mirror center, while the blue curves correspond to 18/09/2026 and show the mirror behavior during the gentle unlock performed for the PA measurement. It is difficult to assess whether these variations are sufficient to explain the disagreement with the thermal camera measurements.
To verify this hypothesis, both measurements (TM center identification and PA position) could be performed within the same acquisition run. The interferometer could first be locked at CARM NULL to measure the PA position, followed by a gentle unlock. After waiting about 1/2 hour, the CH could be switched on to identify the mirror center without changing the HWS acquisition reference.

 The ITF lock recovery went on in the afternoon trying to have a stable CARM_NULL lock (#69804), actvity stopped at around 19:30UTC; we left the cavities locked on the infrared.

SUSP
at 15:11UTC the WI local controls opened by the guardian following the unlock, properly closed.

We obtained a reliable lock acquisition to reach CARM_NULL_1F with noticeable smoothness, but the locks achieved were nevertheless unstable and had an average life expectancy of a few minutes each.

We re-engaged COMMp, soft modes and diagonalization in the automation at CARM_NULL_1F.

The unlocks observed were different in nature and not easily explainable with loop instabilities. In particular, one of them was due to a glitch in the NI suspension that shot the F0 away by about a hundred microns (Fig.1).

We decided to decrease the CHRoCC voltage by another 10% before tomorrow's morning shift.

The reduced gain of PR_Y damper seems to be not enough, because the 11.6 Hz instability raised again during the latest lock in CARM NULL. I restored the old gain, three time larger than the one which has not worked.

ITF found in COMMISSIONING Mode and LOCKED_ARMS_IR State.
All times are UTC.
06:22 ITF in LOCKED_ARMS_BEAT_DRMI_1F State.
07:44 - 08:06 CLEAN OP: HYMO checks at WEB Clean Rooms (Menzione).
08:40 - 09:13 VAC: Vacuum pump intervention at NEB (Macchia, Pasqualetti).
10:51 SUSP: WE MAR loops opened by the guardian. Properly closed (Operator).
14:00 SUSP: WE MAR loops opened by the guardian again. Properly closed (Operator).

The shift was dedicated to the planned activity on CARM NULL locking / DAS adjustment, the ITF ulocked sveral times due to ALS, issue investigation in progress.
ITF left in COMMISSIONING Mode and LOCKING_ARMS_BEAT_DRMI_1F State.

After the ~10% increase of the ChRoCC voltage of Wednesday morning, the rest of the day was devoted to the relock and the characterization (ISC-wise) of the interferometer. The idea was to reproduce a configuration which is more similar to the one we were using during the run, i.e. with a lower sideband gain and possibly an improved stability of the interferometer. The 10% increase of the actuator's voltage did non correspond exactly to 10% in the working point, something less around 7.5% instead. To be confirmed if this is due to a slight non-linearity of the actuation or if the basal temperature needs to be subracted as a static offset (after which the 10% is present also in terms of working point).

From the lock acquisition point of view, we had to tune again several parameters in many steps. We could complete the recovery up to CARM_NULL_3F, with a half-baked recovery of the _1F one. Even in steps considered recovered, the lock could not last much. The sheer variation of the numbers differs from DoF to DoF, but we could see some patterns:

• demodulation phases and gains changed more in the DRMI and CARM_Null configurations than in the intermediate steps; differently from the standard, the changes in the intermediate steps of the CARM offset reduction phase were relatively small from one to the next one, while to lock the DRMI and to go to CARM offset zero the changes were bigger;
• alignment loops were quite impacted, while usually the angular sensors are less prone to changes than the longitudinal ones (MICH/SRCL being the worst among those); DIFFp in particular had to be changed a lot, both in phases and gains, the latter needing to be increased substantially;
• again on the alignment loops: demodulation phases for TX and TY were usually different from one another, and again this happened in particular for DIFFp. When we are in a stable configuration, the phases for the two DoFs are the same.

A different issue affected the SRC: much more than normal, while locking the DRMI SRCL would not lock properly, with the longitudinal correction rapidly diverging out of range (always in the negative direction) but the interferometer staying "locked" until going too much forward or killing the lock ourselves. This was seemingly unrelated to the SR initial alignment, which in principle never changed if not while pre-aligning the DRMI, as we didn't reach any step where a ASC loop could be engaged. The consequence, though, was for the SRM to occasionally resulting quite misaligned. We suspect some kind of kick and/or lenght-to-angle coupling (see Figure), but this needs some verification first. In at least one occasion, we could see a clear difference also between the global alignment in the later stages of the acquisition and the initial alignment of long cavities+beam.

Given the global outcome, we decided at the end of the shift to move again the ChRoCC, from +10% with respect to Tuesday to -10% (V_DAC = 4.28 -> 3.49 V).

The activity CARM NULL locking / DAS adjustment went on through the afternoon until 19:30 UTC (Bersanetti, Boldrini, Bossilkov). I had troubles multiple times aligning SR ty because B1p camera showed a strongly unstable and oscillating signal in the direction of ty-, and the last saved SR ty position worked for the following CITF locks but might have to be corrected tomorrow.

When the commissioning activity ended, the interferometer reached CARM_NULL_3F and unlocked at engaging DIFFp in LOCKING_CARM_NULL_1F at 19:48 UTC . The second and the third time, at 19:58 and 20:12 UTC, it unlocked in LOCKING_CARM_NULL_1F after engaging SSFS ("... and update UGF" USRMSG)

At 20:16 UTC I set Chrocc_PR DAC_V to 3.49 with a ramp of 600s

ITF left in LOCKED_ARMS_IR

Guard Tour

19:51

ITF found in LOCKED_ARMS_IR in COMMISSIONING mode.
planned activity:
CARM NULL locking / DAS adjustment (Spinicelli)

Relocked up to CARM_NULL3F after the usual cross-alignment in ACQUIRE_DRMI.
Activity in progress...

The vertical mode of PR mirror (11.6 Hz) gets randomly excited during the lock acquisition; the lock can even become unstable on that (orthogonal) resonance because of variable and not predictable coupling of PRCL sensing to PR vertical motion.

A possible way to control the instability is to send a fraction of the longitudinal correction to the marionette vertical actuation. That sort of damping works for the arm test masses and is used to reduce a similar instability occurring on WI during the lock acquisition. Unfortunately the same strategy doesn't work with PR, because the coupling seems to have different sign at different steps of the lock acquisition.

For that reason we decided to implement a true local damper of PR mirror vertical mode, using as error signal the vertical LVDT which measures the relative motion of filter7 and its crossbar. F7 crossbar is part of the 3 bodies system (mirror-marionette-crossbar) for which the 11.6 Hz mode is one normal mode of resonance. The LVDT signal is sensitive enough to see the peak when the instability becomes large and a loop can be effective in reducing the Q of the mode, but we have to pay attention on the possible noise injected on the payload by a local control.

A measurement of F7 LVDT response to MAR_Y_CORR has been measured and fitted (fig 1). The fit requires a high quality factor, at least 1000. A narrow band loop around 11.6 Hz has been developed (fig 2, fig 3). Repeating the TF measurement with the loop closed, the Q appeared to be lowered down to 300 (fig 4).

The damper has been tested during a lock acquisition, when the 11.6 oscillation had reached a very high level, well visible everywhere (fig 5 - the probe Sa_PR_MAR_Y_CORR is the output of the damper, which was actually not sent to the actuator when the resonance was growing). The damping have been quite effective and fast, but during its action some low frequency motion of the payload was induced. This is visible on F7 LVDT and also on the quadrant signal used in loop for the beam control (ASC_PR_Y), saying that the action induced a small excess of PR vertical motion and beam angular fluctuation. The excess noise disappeared when the steady state was reached, so I would assume that the noise of the damper is negligible, but the lock did not last enough to check better. Anyway, given the effectiveness of the damper, I decided to reduce its gain by a factor of 3 and leave it permanently in operation. A long lock in CARM NULL is needed to see if this gain is enough to guarantee the stability at 11.6 Hz.

A possible improvement can come from a swap between the local sensor and the quadrant in CARM NULL: the SNR of the quadrant is quite better, but that signal is available only in CARM NULL, while the 11.6 Hz can affect the lock also in the previuos steps.

This morning at 7.25 UTC we set the PR CHROCC (V_DAC) from 3.88 V to 4.28 V with a ramp of 300 s.

The ITF lock recovery went on all the afternoon without succeeding to achieve a stable lock at CARM_NULL_1F (#69796); activity stopped at around 19:20UTC. I left the cavities locked on the infrared.

SUSP
at 18:30UTC the WI local controls opened by the guardian following the unlock, properly closed; I wasn't able to use the suspension GUI.
at 19:09UTC the WE local controls opened by the guardian following the unlock, properly closed; again I wasn't able to use the suspension GUI.

- we reverted all the locking parameters to friday 18th at 12 UTC

the carm null 1f was not reachable (no loop oscillations), so we started to remove part to try to survive

- removed COMMp, CARM slow, alpha and Soft

the lock was killed by instabilities on the dark fringe, so Michal raised the shutter treshold to 0.027

we then tryed to lock with SR misaligned (-2urad ty) and we increased the jumps rate, see Figure 1.

The lock behavior is not predictable and no loop oscillations are present. It could be that we are too close to the cavity instability. 

In agreement with the commissioning crew, I reverted all the steps performed over the weekend to return to the last configuration we had on Friday (STEP 8, entry 69755) when the lock acquisition seemed to be reliable. The main goal was to try HWS-DET acqusitions in order to veirfy the situation of PAs on the NI mirror. 

Activity completed at 10.30 UTC.

Here the TCS settings:

At the end of the transient, during the afternoon shift, we resumed the lock acquisition attempts, but the same issues observed yesterday persisted.

ITF found in locked arms.

At 6:00 UTC started the planned maintenance, below the list of the activities communicated in control room:

• cleaning of the experimental areas;
• TCS chiller refill;
• TCS thermo camera reference (operator);
• LNFS 22MHz replacement with a DDS
• tuning of the TCS CH (Ilaria from remote)

All the maintenance activities completed around 11:00 UTC; then started the planned commissioning activity of ISC.

R. Cavalieri, F. Nocera, P. Spinicelli

This is to detail the RF Modulation configuration we are likely to use for IR1.

Starting today, the 22 MHz sideband is generated with a single-output DDS, as per AdV baseline design.

In an attempt to have a fallback plan for the remaining interferometer RF sidebands generation heading towards IR1, we have tried to verify the phase noise of the LNFS referred to an Synth4 in https://logbook.virgo-gw.eu/virgo/?r=69715, i.e. the one which did not turn on, after hacking our way around the power-on issue (we've tracked down the problem to a failure of one of the two AC-DC converter, the one for the +5Vdc, that we bypassed using a bench power supply in its place).
We verified its phase noise performance are not in spec but they are comparable or better than the one currently in use for 6, 8, and 56 MHz. In addition channel three seems to be working as well, contrary to what has been observed in the last few days in the Lab.

The serial communication works ok.

We have left it on but, at this time, not connected to the modulation chain.

The situation therefore is shaky (the "spare" relies on an external power supply, channel 3 reliability is uncertain) but not as bad as it looked yesterday.
 

This morning we swapped the LNFS that generates the 22.3 MHz used for IMC longitudinal and angular controls with a DDS. 
This LNFS is the one with the communication problem. It will be sent for repair. 

When resetting the output frequency and amplitude of the DDS to nominal values the IMC relocked and so did the RFC. 

We checked some INJ signals : IMC and RFC error signals, 22 dfreq, IMC OLTF... and everything seems to be ok (figures attached)

To attempt to help with COMMp, yesterday at 14h53 UTC I have increased the SNEB drift control loop gain by a factor 2. The loop error signal is power normalized, so this is not intended to compensate a loss of loop gain due to low power, but to actually increase the loop gain.

Another direction to explore is what is happening with PRCL.

Figure 1 shows an example of an unlock of yesterday evening, B2 6MHz I is in loop for PRCL, but the out of loop signals B2 18MHz I and B2 8MHz I start to diverge at the same time as the power in the arm drops. B2 18MHz is independent of the carrier, so the drop in carrier power is not sufficient to explain the change in B2 18MHz I. There is also a sharp drop in B4 12MHz mag that starts at the same time, which is also independent from the carrier.

Figure 2 shows an example of a longer lock from last Friday, which shows that the offset on B2 6MHz I  is comparable to what it used to be, and that the change in sign of the offset is just due to the adjustment of the demodulation phase of  B2 6MHz before the signal is used. B2 18MHz and B2 8MHz have large fluctuation that start at that time, which makes the running away of these signals on Figure 1 less convincing. However, there is no simulataneous decrease in B4 12MHz mag in that case.

During the afternoon two activities went on: #69785 TCS tuning and #69787 CARM_NULL locking

Activities ended at 20:00 UTC

ITF left in LOCKED_ARMS_IR

In the afternoon shift we noticed a difficulty in locking the DRMI; once we could lock we checked the different steps, and yet again the MICH/SRCL demodulation phase had changed, especially for the 56MHz (and less so for the 169MHz); we checked a few things along the following steps:

• gains were mostly ok, we increased ASC_PR given the sidebands were lower today; both DARM and MICH had also to be increased (the former especially);
• the demodulation phase for DARM was also retuned, although to a lesser extent;
• a first check of the calibration between PD2 and PD1 for B7 was attempted, as we could see that during the swap the level of B7_DC changes; it doesn't look a lot when the change happens (CARM offset 150Hz), but it clearly escalates with the intracavity power; this was done by heart as I couldn't compute the ratio with dataDisplay (it returned a plain zero) nor did I dare to open Matlab during the shift; this will be checked again offline;
• when we managed to get to CARM Null, the PR vertical resonance at 11.6 Hz killed the lock after a while;
• later on, the acquisition up to STEP3 was quite reproducible, but we then faced issues going to CARM Null; a general lack of gains (PR, DARM, DIFFp) that we increased, but the gain of SSFS was quite high and we reduced it (again, to be better checked offline);
• what killed the lock consistently was a quick drift of the carrrier, that we could not explain so far; it happened most of the time in CARM_NULL_3F or while doing the transition to 1F. Sometime we could link it to DIFFp/WE, but not all times so this is not conclusive.

At around 13:30 UTC, we prepared the DET HWS measurement to sense the NI mirror: the INJ HWS beam was turned OFF and the DET HWS beam was set to 1.4V.

After various attempts of lock acquisition up to CARM NULL 1F, we agreed with the ISC team to perform a CH common step. Specifically, at 14:11 UTC both the CH powers have been increased by 20% (STEP 10), as detailed in the table below.

No sufficiently long locks have occurred to allow a measurement with the HWS DET, so the activity will be postponed to the upcoming shifts.

Today, an analysis of the WI point absorber was performed using thermocamera images.

• Fig. 1 shows the reference thermal image selected for the analysis.

• Fig. 2 displays the MATLAB analysis output, highlighting a differential temperature of approximately 0.65 degrees for the point absorber (calculated by subtracting the mirror image in CARM_NULL from the image taken with the ITF in DOWN status).

• Fig. 3 illustrates the reconstructed mirror geometry, featuring the identified mirror center and the point absorber location based on pixel-to-distance mapping.

Geometric Mapping:

• X-Axis (Columns): The mirror center is located at column 388, whereas the point absorber is at column 384. This places the point absorber slightly to the left of the center (a offset of 3.5 pixels in the image, corresponding to approximately 0.5 cm in physical space).

• Y-Axis (Rows): The mirror center is located at row 241, whereas the point absorber is at row 221. This places the point absorber above the center (an offset of 19.5 pixels in the image, corresponding to approximately 3.0 cm in physical space).

• Radial Distance: The total estimated physical distance of the point absorber from the mirror center is approximately 3.1 cm.

Note that because this spatial mapping is reconstructed manually from off-axis thermocamera images, it is subject to geometric projection and pixel-resolution uncertainties. Consequently, this estimate should be considered less precise than the HWS measurement and serves primarily as a qualitative/cross-checking reference.

I will soon perform a similar analysis of the NI PAs.

Today we attempted to recover the lock acquisition after this Sunday's DAS step.

We noticed that sidebands power has decreased at the DRMI step (Fig.1), but despite this the lock acquisition is somewhat workable without the need of any major change, although it tends to fail in STEP_3_OF_3 for reasons unclear.

The lock in CARM_NULL_1F is achievable, but not particularly stable. We noticed that a frequent cause of unlock is COMMp, who attempts to engage far away from its working point, possibly due to a slight misalignment of the SNEB that introduces a large DC offset in the error signal. For the moment, COMMp and the soft loops have been commented in ITF_LOCK.py (lines 4544-4545 for COMMp and 4458-4461 for soft modes)

ITF found in LOCKED_ARMS_IR in COMMISSIONING mode.
Locking in progress up to CARM_NULL_1F after the usual cross-alignment in ACQUIRE_DRMI.

Planned activity:
CARM NULL locking / DAS adjustment (Boldrini)

DET
Both SDB2_B1p_QD1 and SDB2_B1p_QD2 VBias' rearmed and shutter opened several times during the locking attempts.