No special equipment was required to set up the stage, but the materials were transported on a 3.5-ton truck, which was parked in front of Technical Building No. 2, adjacent to the Main Building. (The entire route followed the left side of the entrance.)
The company's stage assembly and dismantling hours were not communicated directly to the Control Room that day, but yesterday the Communication Office informed me of the company's on-site hours, which are as follows (local time):
11:00 AM to 1:00 PM
3:30 PM to 1:10 AM
ITF found during locking acquisition.
During the shift the ITF unlocked multiple times from LOW_NOISE_3, with the longest lock lasting 40 minutes. The others round up to 15 minutes.
Here the timetable of the events:
21:38 UTC - Science Mode
21:53 UTC - Unlock
22:48 UTC - Science Mode
23:29 UTC - Unlock
00:19 UTC - Science Mode
00:36 UTC - Unlock
1:48 UTC - Science Mode
1:55 UTC - Unlock
2:50 UTC - Science Mode
3:02 UTC - Unlock
4:13 UTC - Science Mode
4:29 UTC - Unlock
Relock in progress.
ITF found in relocking phase, PREPARE_SCIENCE with Autoscience enabled.
At around 13:07 UTC, while the ITF was in CARM_NULL_1F, EIB started to oscillate; I was able to reclose the loop, then Camilla and Mathieu (present in control room) recovered the standard state for the IMC.
At 14:08 UTC ITF in Commissioning mode for Iterative tuning of subtraction filter gains; activity concluded at 18:43, ITF set in Science mode.
The ITF unlocked and relocked other two times; the lock acquisition is working properly while the lock reliability seems to have some problem. To be checked.
ISC
At 18:35 UTC under request of Maddalena I changed the etalon loop: 2025-07-08 18h35m53 UTC berni 'WI:WI Set point [WI_RH_SET=20.8,rampTime=36000]' sent to LSC_Etalon_Acl
Guard tours (time in UTC)
17:52-18:30
The goal of the shift was to test a new algorithm that can iteratively optimize the gain of a subtraction filter (and later estimate the filter itself) using only science data, so without using noise injections.
We started the shift by setting the gain of the MICH subtraction filter to 0.1 at 14:20:29 (the standard one is 1.02). We ran the algorithm we prepared in MATLAB which found a new gain of 0.1214 (which we set at 14:38:30). However the increments with which it updated the parameter were way too low. For a few hours we tried to find the solution why the increments are so small but we haven't been able to find the issue and we can't reproduce this in simulation, where it works fine. We gathered some additional data at different subtraction filter gains to test this after the shfit to see if we can find the solution in our algorithm. We set these gains at:
16:54:30 - 0.5
17:13:25 - 0.8
With this we have enough data to hopefully find the solution. In the meantime during the shift Diego also performed some work and we unlocked twice (more about this in separate post).
This is a comparison of data in LN3, for the analysis of the new DIFFpTX control.
Fig 1: this morning, in the latest lock before the change of DIFFpTX control, the microseism was a bit higer in the critical region around 0.75 Hz. The following comparisons should take into account this difference.
Fig 2: The locking correction of West Arm sees this difference: the amplitude of the payload resonances are lower by a bit less than a factor of 2.
Fig 3: Pitch on COMMp angular dof is lower too; here the difference is a bit larger.
Fig 4: On DIFFpTX the difference is more evident: up to a factor of 3. This should be due to the improvement of the control filter.
Fig 5: One should expect an improvement of dark fringe stability, but not so much.
Fig 6: In the morning data, the 2D graph DIFFpTX Vs B1p seems to say that B1p fluctuation is entirely due to DIFFp TX. The same analysis applied to the afternoon data does not agree with the previous one: it seems that the coupling has changed a lot.
Fig 7: There is a simple explanation: ASC_DIFFp_TX_CORR is the signal after the loop gain, but the gain has changed by a factor of 2. I had forgot about the automatic adjustment of the gain. In principle, this adjustment should guarantee a sort of constant calibration for ASC_DIFFp_TX_CORR: for this reason I prefer to use the signal after the gain to make quantitative comparisons. But in this case the assumption is clearly wrong.
Fig 8: The 2D anaysis done with DIFFp_TX before the gain. Here the calibration is more realistically the same. The large improvement of DIFFp_TX and B1p_DC is not only due to the change of control filter, but also to the increase of the gain.
We have learned that the gain was lower than the optimal one for a good control accuracy. It is not clear why with different controls the adjustment of the gain works in a different way. By the way, looking at older data, it seems that in the recent data the gain was lower than usual.
The last point: the new filter has a roll-off not well refined and the noise at 20-40 Hz could be too high. A new version has been downloaded. No difference is espected in the control band, but we will check in the next lock acquisition that everything works fine.
This morning we finished the installation of the PDU switches in the remaining towers (Injection Bench, North and West Input, and Beam Splitter). For each PDU, we tested the proper functioning of all stepper motors.
Operation Start Time: 10:45 (local time)
NI: 10:45 - 11:06
BS: 11:07 - 11:19
WI: 11:20 - 11:30
IB: 11:32 - 11:43
Operation End Time: 11:43 (local time)
From 11:43 to 11:53 we took the chance to remove the clogged air filters from IB, WI, BS, NI and SR. They will be cleaned and re-installed on the next maintenance.
Trying to explain the change in sensitivity during SR TX misalignment with a 1/f^{2/3} noise change corresponds to a 20% change in noise level.
Figure 1 shows the simplified noise budget during the night before the SR TX offset
Figure 2 shows the simplified noise budget during the SR TX misalignment, near the peak of sensitivity
Figure 3 compare the sensitivity between the two times.
The simplified noise budget are not perfect fits for the data, so there is room enough for errors, and it could be some other kind of noise that is changing. As noted before the frequency noise gets worse, when the BNS range is improving.
/users/mwas/detchar/toySensitivity_20250703
Yesterday morning, when the seismic condition started to be quite haevy, we noticed a very unstable dark fringe (fig 1). The fluctuation was entirely due to DIFFp_TX (fig 2), the spectrum of which was dominated by a large peak at 0.75 Hz (fig 3). The sensitivity of pitch to longitudinal motion is a well known problem: the peak at 0.75 Hz is the second resonance of the payload and its coupling to the angular motion is not strange, but yesterday it looked a bit too strong and it would be better to understand is everything is working properly at the level of suspension control.
So far, we didn't find anything broken, but we found that DIFFp_TX control filter has some margin of improvement concerning the gain at 0.75 Hz. We developed and implemented an improved control. We tested it this morning, during the first lock acquisitions after the maintenance. We have data in similar seismic condition (fig 4), confirming what expected from the model: there is a certain reduction of angular motion and a dark fringe a bit more stable (fig 5, fig 6). We will continue to keep the new control under observation during the next data taking.
This is a comparison of data in LN3, for the analysis of the new DIFFpTX control.
Fig 1: this morning, in the latest lock before the change of DIFFpTX control, the microseism was a bit higer in the critical region around 0.75 Hz. The following comparisons should take into account this difference.
Fig 2: The locking correction of West Arm sees this difference: the amplitude of the payload resonances are lower by a bit less than a factor of 2.
Fig 3: Pitch on COMMp angular dof is lower too; here the difference is a bit larger.
Fig 4: On DIFFpTX the difference is more evident: up to a factor of 3. This should be due to the improvement of the control filter.
Fig 5: One should expect an improvement of dark fringe stability, but not so much.
Fig 6: In the morning data, the 2D graph DIFFpTX Vs B1p seems to say that B1p fluctuation is entirely due to DIFFp TX. The same analysis applied to the afternoon data does not agree with the previous one: it seems that the coupling has changed a lot.
Fig 7: There is a simple explanation: ASC_DIFFp_TX_CORR is the signal after the loop gain, but the gain has changed by a factor of 2. I had forgot about the automatic adjustment of the gain. In principle, this adjustment should guarantee a sort of constant calibration for ASC_DIFFp_TX_CORR: for this reason I prefer to use the signal after the gain to make quantitative comparisons. But in this case the assumption is clearly wrong.
Fig 8: The 2D anaysis done with DIFFp_TX before the gain. Here the calibration is more realistically the same. The large improvement of DIFFp_TX and B1p_DC is not only due to the change of control filter, but also to the increase of the gain.
We have learned that the gain was lower than the optimal one for a good control accuracy. It is not clear why with different controls the adjustment of the gain works in a different way. By the way, looking at older data, it seems that in the recent data the gain was lower than usual.
The last point: the new filter has a roll-off not well refined and the noise at 20-40 Hz could be too high. A new version has been downloaded. No difference is espected in the control band, but we will check in the next lock acquisition that everything works fine.
Today, I found the ITF in SCIENCE. At 06:02 UTC, the ITF status was in MAINTENANCE.
Below is the list of maintenance activities communicated to the Control Room:
- 07:45 UTC: TCS chiller refill (M. Ciardelli)
- Standard CEB cleaning (M. Ciardelli and external firm)
- Liquid nitrogen refill (Vacuum team)
- Checks of UPS at NEB, WEB, and along the tube (M. D'Andrea and external firm)
- Neovan head chiller filter replacement
- Switch on of newtonian sensors at NEB (M. Tringali, F. Paoletti)
- Installation of switches (V. Boschi)
- Switched on a spare pump on PR (A. Pasqualetti)
Here is the list of maintenance tasks I performed:
- OMC scan POSTPONED as agreed with M. Was
- Captured thermocamera images
- Checked TCS CO2 powers (see the table below). M. Cifaldi rechecked the powers and made the necessary adjustments.
TCS Power Checks
| CH [W] | INNER DAS [W] | OUTER DAS [W] | ||
| W | on the pickoff | 0.305 | 0.026 | 0.295 |
| N | on the pickoff | 1.196 | 0.060 | 0.535 |
11:53 UTC ITF back in SCIENCE. 12:22 UTC ITF unlocked
Romain has noticed that the slow shutter of the OMC takes a longer time to close. In the sense that the motor is on for sevreal minutes instead of ~40 seconds while closing the shutter. Last year we had a similar issue when opening the shutter, which was that the automatic stop at the end of the range was stopping to work. We suspect the same issue here.
This morning around 9:40 UTC I have changed the automation so that it requests instead a given number of steps (-60'000) to close the shutter, instead of requesting to go to the end of the range. I have tested it several times. This is more steps than for opening the shutter (45'000) and from what I could see that works to reset the range so that we always arrive at the mechanical end of the range for closing the shutter. I may have mis-estimated this, in which case the open and closed position will slowly drift, and eventually the shutter will remain open, even after launching the command to close the shutter.
I will monitor the situation at the next few shutter opening / closing to see if there is an issue or not.
ITF found in Bad_Weather and LOCKED_ARMS_IR, due to the high wind activity it was not initially possible to proceed with the locking acquisition.
At 22:00 UTC I started to relock and, at 22:56 UTC, the ITF was back in Science Mode.
It unlocked at 23:23 UTC, I tried to relock but due to the strong gusts of winds it kept unlocking at ACQUIRE_LOW_NOISE_3.
At 2:59 UTC the wind calmed down and, after a couple of attempts, it was possible to reach LOW_NOISE_3 again at 4:04 UTC.
ITF left locked
SBE
SIB2 Position loop closed via VPM at 22:00, the horizontal loop opened again at 22:12 UTC during the locking acquisition.
The NCal cancelling line at 36.04 Hz was left by mistake. It was visible during the short look segments of the night. It has been removed at 4:00 UTC.
ITF found in Bad_Weather, high wind activity which was preventing the lock.
After many failed attempts the ITF could reach LN3 at 18:59 UTC; the planned calibrations have been skipped.
The wind activity increased again and the ITF unlocked at 19:24 UTC.
ITF left in Bad_weather and in locked arms.
CAL
SBE
There are problems with SIB2; today I have recovered the vertical position but thekeep opening. The issue has been already reported in the logbook and to the SBE on-call.
Same plot updated with one week more of data.
Again the same plot, including one more week of data.
Two new NCal have been started around 3 UTC: NEN at 18.02 Hz and NWN at 18.04 Hz (rotor rotation, half of the h(t) signal). The sign of the canceling line might be wrong. It will be checked later today (if the relock is possible). They have been stopped around 4:20 UTC.
The compputation of the NWN temperature has been fixed (it uses a +- 20 V ADC and was configured for a +- 10 V ADC)
In addition, the frequencies of the following lines have been adjusted: NNF: 18.04 -> 18.08, WNN: 18.26 Hz -> 18.22 Hz, WSN: 18.28 Hz -> 18.24 Hz
The NCal cancelling line at 36.04 Hz was left by mistake. It was visible during the short look segments of the night. It has been removed at 4:00 UTC.
Taking advantage that we are locked on the IR, we moved the PR vertically (F0_Y_SET by -60 um) to bring the beam on the TX from 14 urad to 13 urad.
The beam is still well centered on the ITMs. (See attached plot).
Today, I found ITF in locking CARM_NULL_1F. 5:22 UTC ITF in SCIENCE. 07:39 UTF ITF unlocked. The ITF status set in BAD_WEATHER due to strong wind and sea activity, which made locking impossible during this shift.
ITF found in Science, BNS Range ~51Mpc
Sea and Wind activity quite high during the shift
- 02:10UTC, ITF unlocked
SBE
SIB2 Position Loop properly closed after opening by the guardian several times:
23:11UTC, 23:20UTC, 23:33UTC, 23:41UTC, 00:23UTC, 00:33UTC, 00:38UTC, 00:49UTC, 01:24UTC, 01:50UTC, 03:53UTC, 04:04UTC, 04:18UTC
QPD
B2_QD2 galvo loop properly closed after opening, following the closiure of the SIB2 Position loop:
23:12UTC, 23:21UTC, 23:34UTC, 23:42UTC, 00:23UTC, 00:33UTC, 00:39UTC, 00:50UTC, 01:24UTC, 01:50UTC
03:33UTC, I rearmed vBias of B1p_QD1 and B5_QD2
04:10UTC, I rearmed vBias of B1p_QD1 and B1p_QD2
Guard Tours (UTC):
23:17 - 23:54
01:06 - 01:42
03:08 - 03:45
Since the power observed for the LED of the WEN NCal was again unstable, it has been switched off this morning at 4:03 UTC while the ITF was unlocked. The corresponding DMS flag has been shelved.
ITF found in LOW_NOISE_3 and in Science Mode.
It unlocked at 15:46 UTC, back in Science Mode back in 16:39 UTC.
At 19:08 UTC, similar to yesterday (see report #67221), there was a period of missing data with the same error from SUSP_Fb and FbmFFE.
ITF left locked.
No special equipment was required to set up the stage, but the materials were transported on a 3.5-ton truck, which was parked in front of Technical Building No. 2, adjacent to the Main Building. (The entire route followed the left side of the entrance.)
The company's stage assembly and dismantling hours were not communicated directly to the Control Room that day, but yesterday the Communication Office informed me of the company's on-site hours, which are as follows (local time):
11:00 AM to 1:00 PM
3:30 PM to 1:10 AM
I was able to close the WI ID loop without issues by applying the standard procedure. After that, I removed the Earthquake mode and realigned the cavities. Then I started the lock acquisition, but there was a problem with the north cavity: although it initially locked, the lock was lost after a short time.
At 05:27 UTC, I set Troubleshooting mode and began contacting the ISC on-call and the SUSP on-call. Around 07:00 UTC, I also contacted Paolo.
After some investigation, he discovered an incorrect value for the gname SAT_NI_MAR_ENBL. Once this issue was fixed, we were able to relock the cavities; unfurtunately the ITF keep unlocking in CARM_NULL_3F. The ISC on-call was still connected and continued investigating the problem.
We discovered that this gname can be modified by clicking the "Reallocation" button on the SUSP GUI. We also found that two other gnames were affected: SAT_NE_LOWN_ENABL and SAT_WE_LOWN_ENABL.
In the end, we were able to restore the correct values, and finally, at 10:50 UTC, the ITF was put into Science Mode.
Guard tours (time in UTC)
11:30-12:00
ITF found in LOW_NOISE_3 and SCIENCE mode (Autoscience ON); BNS Range ~53 Mpc.
ITF unlocked at 01:36UTC due to end TM (MIR or MAR) correction saturation (pdf 1). WI Guardian opened F0 loops, reflections caused by movement of WI visible on WE_CAM. ITF set to DOWN in EARTHQUAKE mode waiting for WI oscillatons to subside.
Attempted to close WI F0_DC_ENBL at 02:50UTC with no success. Second attempt at 04:02UTC unsuccesful again, ISC on call notified.
Recovered at 05:22UTC thanks to intervention by Berni, lock acquisition in progress.
Guard tour (UTC)
-> 21:38
23:02 -> 23:45
01:05 -> 01:55
03:02 -> 03:42
ISC
ViolinModes red flag on DMS excited at around 00:50 UTC (fig.1)
When I arrived, locking acquisition was in progress
- 13:31UTC, Science mode
- 19:30UTC, Calibration mode to perform the planned measurements
- 19:31UTC, Check Hrec
- 20:02UTC, ISC injections
- 20:21UTC, back to Science
Guard Tours (UTC):
12:47 - 13:20
15:47 - 16:24
18:12 - 18:47
20:57 -
Today (Jul 5) around 18:00 UTC I have set back the ASC_SR_TX_SET to 0.022 with a 2 hours long ramp. As it was reset a few hours before by the unlock, and the BNS range has been lower by a few Mpc since that relock.