Upon arrival, I found the ITF in CARM_NULL_1F and BAD_WEATHER mode.

The ITF reached LOW_NOISE_3 at 14:43 UTC. I then set the COMMISSIONING mode.

As scheduled for the shift, Nardecchia turned off the NE and WE RH at 15:21 UTC (see #66809).

At 16:56 UTC, the ITF unlocked. It was back in LOW_NOISE_3 at 17:52 UTC. It unlocked again at 20:58 UTC.

I left the ITF in LOCKED_ARMS_IR with the AUTORELOCK active.

The WEN and WWN NCal rotor frequencies have been shifted from 18.18 and 18.20 Hz to 42.18 and 42.20 Hz at 17:06 UTC while the ITF was unlocked.

Both NE & WE RHs were switched off at 15:20 UTC while the ITF was in LN3. 

After the first interferometer relock, we noticed that the 3 Hz comb (and related noise) was no longer present in hrec.

This is quite inexplicable, since there were no connection reversals during the last tower opening, either internally (according to Ettore personal communication) or externally. There was only one tower opening, one mirror cleaning, and one tower closing.

To check whether anything had changed, yesterday we repeated a magnetic far-field injection at WEB. The results will be discussed in a separate entry, but preliminarily it seems that the coupling has decreased since the last anomalous one, although it is still higher than the standard values measured in the past.

Then we went to WEB, and at about 09:35 UTC we exchanged the phase of one of the two RH connections.
Next, we repeated the same magnetic injections, and once again (by eye), it appears that the CF is the same as that performed before this phase swap.

At about 10:55 UTC we restored the correct connections that we had previously reversed and also moved the local magnetic probe ENV_WE_MAG_ to the RH power supply.

During all these operations the 3 Hz Comb never reappeared.

Also, there is no coherence between the new temporary magnetic probe (which clearly detects the 3 Hz Comb still generated by the RH power supplies) and hrec.

ITF found in LOCKED_ARMS_IR COMMISSIONING mode.

ITF very unstable due to the bad weather conditions (BAD_WEATHER mode set) and the unlocks due to the WE glitches.
According with commissioners and SAT experts we decide to cancel the planned TCS activity.

Parallel activities:
- OMC scans with SR aligned and misaligned
- Temporary installation of Optris thermocamera at NE

In order to compare with previous measurements before the WE replacement (https://logbook.virgo-gw.eu/virgo/?r=65254) I have made this morning scans of the OMC with SR aligned and misaligned (initially planned for yesterday evening). The data analysis and comparison with previous scans of the SDB1 OMC and the EDB OMC remains to be done.

Figure 1 One thing that I have noticed is that when misaligning SR in CARM NULL 1F around 07:40 UTC, the power on B4 (carrier recycling gain) has increased. Which would mean that the effect of the point absorber, even with the beam off-center on WE depends on signal recycling, and with signal recycling misaligned the losses due to the point absorber are smaller.

Measurements times

06:06 UTC - OMC scan start in CARM NULL 1F (20min after lock)
unlock near the end of the scan

relock, SR_TY steady around -225.0 wihle SR/BS TX moving (due to the
mis-centering on WE), starting to add offset on SR_TY while the other
transients are ongoing. But SR TY moved -225.4 just before that, so using that as a reference of aligned SR to add ~2urad of misalignment to.

7:25 UTC, SR_TY_SET = 0.3
7:28 UTC, SR_TY_SET = 0.5
7:30 UTC, SR_TY_SET = 0.15 to stabilize at SR TY = -224

7:33 UTC - OMC scan start with SR misaligned
7:40-7:44 increase SR_TY_SET to 0.25 and then reverted to 0.2 to recover DCP around 190Hz and lower B1p power
Overshoots to DCP of ~150Hz, continously adjusting the SR TY offset to try to stay around 190Hz.

08:04 restoring SR alignment

As the WE glitch issue remained uninvestigated (66797-66799), today's TCS shift was cancelled.

The thermocamera has been removed from the NE tower at ~11.15, in order to be reinstalled at WI this afternoon.

To be noted: the lamp near the viewport (see Fig. 1) was reinstalled avoiding to put it into contact with the viewport itself. 

ITF found in Commissioning mode with Suzanne working on WE and NE mirror thermal camera imagines.

The planned activities could not be performed because the ITF could not be locked in a stable way, see: WE glitches. Valerio has been informed about this problem.

At around 18:30 UTC, in agreement with Michal, we decide to keep the ITF locked on the IR.

Suspension

The WE COIL H3 is close to saturation; Valerio has been informed.

SBE

SPRB vertical position recovered with step motors at 18:36 UTC.

Vacuum

The DMS was reporting a problem with WE cryotrap: I informed Antonio and he made a patch for it.

There are many unlocks caused by glitches at west end. These are visible in many of the WE channel. The glitches on Sc_WE_MAR_TY look like data errors, dropping to zero. And they are very frequent, but not always causing unlocks. Some of them are associated with glitches on coil, WE_F0_COIL_H1, and these were present in the three unlocks I have looked at, but not present for one MAR TY glitch that did not cause an unlock.

Figure 1 and 2 show examples of unlocks

Figure 3 shows that The H3 coil is close to saturation at 9V. The H1 is also flagged but less critical at 5V.

My conclusion is that there is some data communication error showing up as glitches, and some of them are kicking the suspension. The high horizontal correction on F0 are unhealthy, it is less likely they are the cause of the problem, but at least this should be simple to solve acting on the corresponding motor.

Figure 4 shows how the situation has detoriarated over the past 4 days with glitches increasing from zero, to many times per minute for MAR TY. The coils have a daily oscillation, the glitches might be associated with H3, whenever the correction becomes very negative. Could be that the correction draws too much power and causes insufficient power to the digital system doing the communication and computations.

Here are the NE payload images of the thermocamera:

Fig. 1: ITF unlocked;

Figs. 2 and 3: ITF locked in CARM_NULL;

Compared with the WE payload, the spot seen on NE mirror is much less centered and seems to be fainter as well. A more precise estimation will follow.

This afternoon we remove the Optris 640i thermocamera from the WE viewport (see #66757) and installed it temporarily on the same viewport (ZnSe viewport of the HR surface) of the NE tower. 

Here are more pictures of the thermocamera taken today.

Fig. 1: ITF unlocked;

Fig. 2: ITF locked without offset;

Fig. 3: ITF locked with offset applied to WE_Y (12 mm).

No clear difference can be observed between the spots of the two images with the ITF locked. The analysis of the position remains the same of the previous log-entry (#66776).

Only to have the comparison between the OG in the  various configurations:

- 'Old WE'; 'new WE before cleaning';'new WE before cleaning Y offset';'new WE after cleaning'; 'new WE after cleaning Y offset'

The operators reported that from time to time some parts of the DMS became grey due to missing ENV channels  . Several investiagtions were performed

• By looking at the same SMS channels using dataDisplay, FbmFE level or directly on the FbsMoni server,  ENV SER builder, it appears that the channels are missing only after the FbmFE level and are correctly collected by the FbsMoni server
• The channels collected by the EnvServer1500W and EnvServerSqz were stored in dedicated SER channels . In these condiftions, there channels were not anymore missing, meaning the issue is related to the ENV SER collection
• The problems appeared the 2025/05/06 around 7H00 :
  • At the same time period, the SMS data collection of the ENVnoise server was set up for debugging purposes and its SMS channels collected by the FbsISC slow frame builder in a SER called "ENV" too . It was the mistake as a SER is uniq in all the DAQ area , and it 's currenctly built  by the FbsMoni server . The ENVServer SMS data collection from FbsISC has been disabled at  2025-05-15-08h12m54-UTC
  • To avoid this statement, the ENVnoise SMS channels are still collected by the FbsISC server but stored in the ENVnoise and the ENV_NOISE SER channels : operation performed at 2025-05-15-09h10m28-UTC

This should fix the issue

Comparing the sensitivity before and after WE mirror cleaning, see attachment with different curves:

a) red before cleaning, no shift on WE mirror

b) blue before cleaning, shift on WE mirror: optical gain is about 20% better than in a)

c) purple after cleaning, no shift on WE mirror: optical gain is about 5÷6% better than in a)

d) green after cleaning, shift on WE mirror: optical gain is similar as in b)

A quick conclusion is that:

1) the broadband ~1/f^2 noise between 50 and 70 Hz, that was present before cleaning and did not depend on optical gain, is not present anymore

2) the 3 Hz comb, that was present before cleaning and whose ampltude did not depend on optical gain, is not present anymore

3) the broadband (1/f^2/3?) noise between 100 Hz and 200 Hz is scaling with the optical gain; with similar optical gain, it's roughly the same before and after cleaning.

The 50 Hz to be removed from DARM via feed-forward is back to normal. The old startegy for FF parameters computation works again.

we post the angular noise budget computed after a set of measurements performed this morning.

working conditions of the ITF: decentered position of the beam on the WE (12mm offset on vertical).

DIFFp TX (blue curve) shows higher coupling.

This is a report on the hardware installations, performed on May 13th,  for  the  phase noise cancellation loop  of the fiber   between Inj and EQB1. The installations involved several laboratories

Atrium:

The fiber circuit of the phase camera and the squeezing in atrium has been modified  (figures 1,2)  in order to decouple the phase camera outputs from the one used to power the box “phase noise box” (fig3) containing the setup for the fiber phase noise cancellation loop . In this way, for the phase noise cancellation loop, we act on a dedicated AOM (operating at 100 MHz) independent from that of the phase camera.

Injection E-Lab:

The electronics for the phase noise cancellation have been installed in the Injection E-room. The installation includes a HP8656 RF generator equipped with a custom RF amplifier that serves to power the AOM at 100 MHz and a new 4-channel DDS generator that provide our low phase noise frequency reference (fig4)

DetLab (EQB1):

The optical fiber coming from the atrium to EQB1 has been split in two. One output goes to the main PLL of the squeezer as in the old set-up and the other branch to the fiber reflector (see fig 1 scheme).

Finally, 3 low attenuation RF cables (L=15 meters) were laid between the atrium and the E-room.

There was no time to test the loop operation, which will be done as soon as possible in agreement with the ITF planning. The feedback system is therefore left off.

The NE and NS NCal power supplies were left OFF at the end of the NCal work of yesterday. They have been turned back ON and the NE NCals restarted this morning at 5:42 UTC  

I updated virgoData/VirgoOnline/SNEB_dbox_rack.cfg to set the reference positions of NW rotors to 0. Attached image shows the position histograms after this operation.

ITF found in LOCKING_ARMS_BEAT_DRMI_1F, we were able to reach ACQUIRE_LOW_NOISE_3, but the ITF unlocked 13:41 UTC.
From 14:00 UTC to 17:35 UTC the Cal Team worked at the both terminals for the planned NCal Activity (see reports #66781), after that the recovery of the ITF, carried out by Mantovani and Bersanetti, started (see reports #66782, #66784).
Activity concluded at 21:35 UTC, ITF left with AUTORELOCK_FAILSAFE engaged at CARM_NULL_1F.

We spent the evening in trying to recover the interferometer, activity which, by the way, we could not do for most of the day.

Despite scheduled times for the different activities, two locks were killed at the end of the acquisition by two ill-advised actions which both impacted SNEB_Fb:

2025-05-14-18h10m23-UTC>ERROR..-[TfbSource::StoreData] producer SNEB_Quadrants_DAQ_10000Hz: overflow at GPS 1431281441.199935050 (received 2001 / expected 2000) (2 times)
2025-05-14-18h10m23-UTC>ERROR..-[TfbSource::StoreData] producer SNEB_LC_DAQ_10000Hz: overflow at GPS 1431281441.199935050 (received 2001 / expected 2000) (47 times)
2025-05-14-18h10m23-UTC>ERROR..-[TfbSource::StoreData] producer SNEB_PSD_sensing_DAQ_10000Hz: overflow at GPS 1431281441.199935050 (received 2001 / expected 2000) (15 times)
2025-05-14-18h10m23-UTC>ERROR..-[TfbSource::StoreData] producer SNEB_Photodiodes_DAQ_10000Hz: overflow at GPS 1431281441.199935050 (received 2001 / expected 2000) (51 times)

2025-05-14-20h27m37-UTC>ERROR..-[TfbSource::StoreData] producer SNEB_Quadrants_DAQ_10000Hz: overflow at GPS 1431289675.199935070 (received 2001 / expected 2000) (2 times)
2025-05-14-20h27m37-UTC>ERROR..-[TfbSource::StoreData] producer SNEB_LC_DAQ_10000Hz: overflow at GPS 1431289675.199935070 (received 2001 / expected 2000) (47 times)
2025-05-14-20h27m37-UTC>ERROR..-[TfbSource::StoreData] producer SNEB_PSD_sensing_DAQ_10000Hz: overflow at GPS 1431289675.199935070 (received 2001 / expected 2000) (15 times)
2025-05-14-20h27m37-UTC>ERROR..-[TfbSource::StoreData] producer SNEB_Photodiodes_DAQ_10000Hz: overflow at GPS 1431289675.199935070 (received 2001 / expected 2000) (51 times)

This sums up to almost two hours lost, of the only five we could work today for recovery.

It is not acceptable not being able to coordinate the work in a decent way. During a lock recovery no other actions should be done on the interferometer.