DRAFT - SHIFT IN PROGRESS

Today, I found ITF in SCIENCE. 15:02 UTC COMMISSIONING started (C. Derossi, P. Spinicelli).

ITF found in Science mode.

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

• from 7:00 UTC to 8:00 UTC INJ tower grounding (D'Andrea);
• cleaning of the experimental area and cryotrap refill;
• finalization of NI camera installation (Mattia, 66169);
• passive RF filter fiber EOM correction;
• Update of the ITF_CONDITIONS node;
• periodic tasks performed by the operator:
  • lock and scan of the OMC in single bounce;
  • TCS chiller refill;
  • check of IMC working point and position;
  • TCS thermal camera reference;
  • TCS power checks (see table below)

All the activities concluded around 11:00 UTC; final check performe by F. Nenci.

The ITF relocked at the first attempt after a manual prealignment of PR/SR; Science mode at 11:57 UTC. It unlocked at 12:07 UTC.

Science mode at 12:53 UTC.

Here is the result of the angular noise injections performed on Feb 6: DIFFp (both TX and TY) look quite close to the sensitivity between 10 and 50 Hz, higher than I remember them to be; there is also a non-negligible contribution from PR_TX at low frequency.

I attach the results of the LSC noise injections performed on Feb 1 and 3; in Figures 1 and 2 the projections on DARM for the two days respectively, while in Figures 3 and 4 the same for Hrec.

The two projections on DARM look quite the same, with CARM being the worst offender from 10 to ~35 Hz.

The two projections on Hrec look a little different instead; here CARM is not dominating in either case, but the contributions from SRCL and especially PRCL are different, with PRCL clearly more limiting on Feb 1 (and SRCL below the coherence threshold), while on Feb 3 PRCL is lower, but SRCL is comparable in coupling.

This morning during the maintenance I updated the ITF_CONDITIONS node and restarted it; the three changes are:

• introduction of a proper decorator for unstalling the managed nodes (ITF_LOCK and ITF_STATUS); this is an adaptation of the one we use all over the place, but it is used only when either autorelock or autoscience are True, therefore I needed a new one instead of using the one in META_library;
• I remove all states from the dropdown menu of the node, with the exception of the only ones that should be asked manually: STOP, AUTORELOCK_FAILSAFE, AUTOSCIENCE_ON and AUTOSCIENCE_OFF; the new states layout is attached;
• I added as FDIN_TAG all the guardian channels for the SATs and the SBEs (V1:SAT_*GRD_TG V1:SBE_*loop_pos* V1:SBE_*loop_act* V1:*LC_TXYZ_enbl); the future development will be to read those channels to catch when any guardian of one suspension or bench is triggered; this will cause the immediate "failure" of the lock acquisition, overriding the usual conditions (lock acquisition maximum duration and number of trials), thus triggering the alarm on the DMS.

This poses two reminders:

• close and re-open any ITF_CONDITIONS window hanging around the sessions;
• always STOP Autorelock and Autoscience before driving ITF_LOCK or ITF_STATUS by hand (this was already the case but the node is now less lenient about it).

I went to the NI tower during the maintenance and tuned the camera focus. The image is now good (Fig.1).

ITF in LOW_NOISE_3 and in SCIENCE mode for the whole shift.

03:26 UTC - S250211aa
04:36 UTC - S250211be

Guard tour (UTC):
22:05 - 22:40
01:00 - 01:34
04:05 - 04:40

This afternoon science mode stopped from 16:00UTC to 18:45UTC for calibration, ISC and ENV noise injections; then the ITF unlocked at 20:26UTC because of this earthquake:

M 5.2 - 20 km SW of Amorgós, Greece
Time 2025-02-10 21:16:29 (UTC+01:00)
Location 36.731°N 25.711°E
Depth 10.0 km

ITF relocked at the first attempt and science mode started at 21:14UTC.

- guard tours (UTC):
19:10 --> 19:55

Calibration
- 16:01UTC: Measurement of checkHrec, NE,WE,BS,PR,SR optical responses, and sensitivity (CALIBRATED_DF_SENSITIVITY);
- 16:58UTC: Measurement of actuators response for NE,WE actuators and NI,WI,BS marionettes (CALIBRATED_DF_PCAL);

EnvMoni
- 18:01UTC magnetic and acoustic noise injections;

ISC
- 17:31UTC: LSC injections (inject_SSFS.py);

I found ITF in SCIENCE; It remained locked for the whole shift.

ITF in LOW_NOISE_3 in SCIENCE mode for the whole shift. 

DET
B5_QD1 Galvo loop found open. Properly closed via VPM.
SBE
SQB2 vertical position recovered.

Guard tour (UTC):
22:00 - 22:38
00:47 - 01:25
04:00 - 04:40

Today, I found ITF in SCIENCE. 10:01 UTC ITF unlocked, 10:48 UTC ITF back in SCIENCE. 

Guard Tours (UTC):
07:15 - 08:00
08:57 - 09:44
11:45 - 12:26
13:01 - 13:41

ITF found in LOW_NOISE_3 and in Science Mode. It unlocked at 00:39 UTC due to an earthquake (23:23 UTC, M 7.6 - 209 km SSW of George Town, Cayman Islands), which has shaken the suspensions and prevented the relock.
From around 1:35 UTC it was possible to start again and, after various attempts, the ITF was back in LOW_NOISE_3 and Science Mode from 3:19 UTC.

Guard Tour (UTC)
22:03 - 22:30
0:21 - 0:42
3:21 - 3:53
Due to a failure of the service car, some Tour were canceled.

ITF found in LOW_NOISE_3 in SCIENCE mode. 

At ~18:30 UTC the BNS range started to decrease due to the increasing of the wind activity. I exit from SCIENCE mode for few minutes.
At 19:00 UTC ITF unlocked. ITF relocked at first attempt in LN3 at 19:46 UTC.

19:47 UTC - CALIBRATION mode set for:
19:47 UTC - check hrec.
20:12 UTC - ISC injections (inject_SSFS.py).

Back in SCIENCE mode at 20:34 UTC.

At ~21:41 UTC the BNS range started again to decrease due to the increasing of the wind activity. I exit from SCIENCE mode for few minutes.

Guard tour (UTC):
16:14 - 16:40
19:40 - 20:09

You should not go out of science mode because of a large decrease of BNS range. State vector online or DQ flags offline will take care of that.
Plot1 shows that BNS range drop occured 1 minute before you go out of science mode and that DQ_ANALYSIS_STATE_VECTOR  has CAT1 bit set to 0 while glitches occured.
Plot2 shows that several large band glitches occured in h(t) and in DARM and produced the BNS range drop.
Going out of science mode has flagged as bad some data that may be ok for data analysis.

ITF found in LOW_NOISE_3 and in Science Mode. It kept the lock for the whole shift.

Guard Tour (UTC)
21:50 - 22:30
0:13 - 0:51
2:07 - 2:47
4:01 - 4:39

DRAFT - REPORT  IN PROGRESS - will be finalized at the end of the shift
ITF found in LOW_NOISE_3 in SCIENCE mode. 

At 14:23 UTC, for a minute, I exit from SCIENCE due to a big glich of the BNS Range, below 20 Mpc.

SBE
16:50 UTC - SQB2 vertical position recovered.

Guard tour (UTC):
18:40 - 19:20
21:50 - 

Since Feb 4th 2025, in addition to the 25' glitches, there are high SNR glitches present in DARM and in h(t) about 10 to 20 times per day. They cover a large frequency band from 10 to above 1000 Hz. The consequence is for each glitch a large decrease of BNS range (which triggers gating) and maybe sometimes a not proper online noise subtraction as described in elog#66149.

Plot1 and plot2 show the median-normalized spectrograms of h(t) on Feb 3 and on Feb 7, in the frequency band 200-1000 Hz, where we can see more numerous glitches on Feb 7.
Plot3 and plot4 show the SNR of glitches detected by Omicron on Feb 3 and on Feb 7, where we can see almost only the 25' glitches on Feb 3 while the higher SNR glitches are visible on Feb 7.

The typical strong glitch that occurs about once per month, and already described in logbook entries 64928 or 64804, has occured again on 2025-01-29 around 12h13 UTC.
This is a large glitch occuring in NE mirror control signal Sc_NE_MIR_Z_CORR that introduces a glitch in hoft_raw. This glitch pollutes the TF used to do the online noise subtraction. This TF is updated every 4mn. As can be seen on the attached plot, the noise subtraction then introduces wrongly some noise in the band 8-40 Hz during 4 mn. Such event has variable consequences on h(t) (sometimes it does not deserve a CAT1 flag offline) but it is seen by gating, then by StochMon monitor (see here for instance). Origin of those Sc_NE_MIR_Z_CORR rare glitches are not yet understood. A protection on the online h(t) noise subtraction is currently under study.

The reduction of COMMp dither loop gain produced the expected improvement of angular motion seen by the quadrants, and also an improvement of arm power stability. The coherence between dither and quadranrt is still very high: the residual motion is still dominated by the same noise. Optical levers and NE dither have some coherence with the dither in loop: it means that the signal in loop is noisier than it should be.

ITF found in Science mode with Autoscience activated; it kept Science mode for all the night.

No info from received from the guardian.

ITF found in LOW_NOISE_3 and in Science Mode.
At 16:30 UTC I set the ITF in CALIBRATION Mode and manually unlocked to perform the planned calibrations with ITF unlocked. Here's the timetable of the action performed:

From 16:33 UTC to 16:46 UTC - Measurement of the TF between CAL and Sc channels, and of PCal sensing delays 
From 16:47 UTC to 17:33 UTC - Measurement of actuators response for PR and BS mirrors 
From 17:33 UTC to 18:17 UTC - Relock for next measurement
From 18:18 UTC to 18:51 UTC - Measurement of actuators response for NI,WI mirrors
From 19:22 UTC to 19:41 UTC - ISC injections (inject_asc_test.py)

ITF back in Science Mode from 19:47 UTC. It unlocked at 20:40 UTC, back in Science and LOW_NOISE_3 from 21:28 UTC.
ITF left locked.