Nevertheless, we have some extra noise in the band 1-10 Hz. I checked all raw signals (voice coils, geophones, LVDTs) and they seem all correct. I switched off the MotorSAS process in the vpm, since the driver coils can make electronic noise, but that did not change anything. In the evening when Romain gave the green light, I switched off controls, switched on viscous damping, and switched on the full pid. With the full pid the ground motion is damped well below 1 Hz but there is an extra shoulder between 1-10 Hz. I think it arises in Tx and Tz and is coupled to Ty,x,z via the tilt stabilization but this is just conjecture. Suzanne did not find anything rubbing against the bench, but maybe there is some small detail. I did not see much difference in acoustic noise (such an effect on the bench, bypassing the floor, could maybe cause such a shoulder) and I checked that the actuators and pid have the same transfer functions as last week (plotting the transfer function from actuator current to bench motion) so it is difficult to find the cause of the extra noise now; it seems all sensors and actuators are performing as well as last week,
The only "large" difference is that there is now some 10 times more noise at the secondary microseismic peak (0.3 Hz); this peak is damped well. It should not spoil the pid at 3Hz, but we may hope for improvement on a quieter day.
Included are the files of the d.o.f. and time signals of the actuators with actuation off/only viscous (first attachment), actuation off/pid (second attachment), a comparison to last week (3rd file), and the raw signals/seismic motion for today and last week (4th and 5th file). The excess noise shoulder is obvious, but the damping is good at 0-1 Hz and similar at +10 Hz.
Today I reconfigured the 4 DAQ boxes of SDB2 in order to solve timing issues resulting from our intervention on the bench.
The reconfiguration took place between 21h59 and 22h03 utc.
The Timing flag in the DMS is back to green, however, since the reconfiguration the processes SDB2_Photodiodes, SDB2_Quadrants and SDB2_Readout seem to be misbehaving. A DAQ expert has been informed;
As agreed during the daily meeting, we post this entry to inform that we have completed our activities on SDB2, and we do not need the NI single bounce beam any more until the SDB1 tower venting is completed (tomorrow morning).
Today we continue the work on SDB2 bench.
We open the OMC shutter at 07h52 utc.
We decide to use B5_QD2 instead of B5_QD1 in order to shorten the optical path up to the galvo and increase the distance between the galvo and the QPD of B1s.
We displace the B5_QD2 quadrant by -150 mm in X. Its galvo was also translated by -125 mm along X. The distance between the galvo and the B5_QD2 oblong hole is now 275 mm.
With this reshuffling we should have preserved the Gouy phase tuning performed yesterday with B5_QD1. In other words the optical path length from the lens to the B5_QD1 should now be the same as the optical path length that we had adjusted yesterday between the lens and othe B5_QD2.
After this change we align the B1s beam on the B5_QD2 quadrant by acting on the picomotors of B5_M1 (-4200 steps in H, -14000 in V) and we managed to close the galvo loops as shown on Fig.1.
We placed a beam dump for the reflection from the quadrant.
We realigned the galvo and QPD B5_QD1 on the B5 beam. We acted on B5_M1 picomotors.
We realigned the beam on B5P_PD1 using the motors of B5_M3. Then we realigned the beam on B5_PD2 by adjusting laterally the photodiode position and acting with B5_M4 (see Fig.2).
Finally we realigned the B5 beam on the camera (see Fig.3) by displacing the camera by 5mm towards east and by vertically adjusting the beam position with the last folding mirror.
To be noted: we had difficulties in seeing a beam on B5_PD1. we suspect that its shutter was stuck in the close position. When it opened we could rapidly realign it.
During this intervention, the elements we have removed and/or added to the bench:
- Added: 6g for cable ties
- Added: 175g for the glass beam dump to block the beam going towards B5_QD1
- Removed: 11g for two screws and their washers
After summing all the added and removed weights on the bench, since yesterday, we conclude that we have added a total of 329.5 g.
To compensate for the extra weight we have removed 332g of counterweights.
We added some temporary weights in order to bring the bench height around +500 um.
We then balanced the bench and were able to close both the angular and position loops (Fig.4).
We performed alignment checks with the bench suspended and controlled in air:
* the beam position on the B1p, B1s, B5 camera did not change at first order, compared to the reference taken last monday before the intervention: Fig. 5,6,7.
* we checked that we obtain the expected power on the B1s and B5 photodiodes and quadrants, compared to what we had before the intervention. The only doubt is on B5-PD2 where we have now more power than we used to.
* we were able to close the galvo loops of B1s and B5 as earlier (Fig. 8).
* we checked the alignment of the B1s beam on the EDB bench. The beam was still passing through the two irises installed last monday and reaching the OMC. We checked that, when opening widely the irises aperture, we have a power of about 2.1 mW reaching the photodiode EDB_B1s in reflection of the OMC, which corresponds roughly to our expectations (Fig.9).
Finally, we added a beam dump in EDB, right after the telescope placed before the OMC. We oriented this beam dump in such a way that its reflection reaches another beam dump.
At the end of the activity on SDB2, we performed the following actions:
* close the doors of SDB2 with only one screw, keeping the temporary weights inside, and keeping the clips holding the door rings.
* close the OMC shutter
* reopen the SDB2 position loops (in preparation of the SDB1 tower venting).
* Open SDB1 angular controls
* Open OB suspension controls
* Open OMC temperature stabilization loop
ITF found DOWN in COMMISSIONING mode. Still in SINGLE_BOUNCE_NI with PR misalignet (30urad ty).
planned activity:
- B1s QD path installation on SDB2 (Gouaty, Romero)
- 15:00 UTC - start the SDB1 tower venting (Vacuum team)
- intervention on SDB1 postponed to tomorrow morning.
SBE
07:30 UTC - SQB1 vertical position recovered via stepping motors.
The beam parameters obtained from fitting the measured data can be seen in the attached Figure. It must be noted that the sagital and tangential data points have been fitted independently.
This morning we resumed our activities on the SDB2 bench. We reopen the OMC shutter at 07h34 utc. We checked that the bench position (blocked) and the beam position on the B1p, B1s and B5 cameras did not change during the night.
We install one iris on the B1s beam path next to the SDB2-EDB viewport and another iris in front of the OMC to have alignment references.
We start the process of displacing SDB2_B5_M1 to its new position: (x, y) = ( -0.225, -0.331) m (using the optocad reference frame).
We realigned the B1s beam on the first iris by acting on B1s_M2 picomotor:
- B1s_M2_V: 2700
- B1s_M2_H: 13000
We also realign the B1s beam on the iris placed in front of the external OMC by acting on the last folding mirror before the EDB telescope.
We acted on the picomotors of B5_M1 in order to center the beam on B5_QD1.
- B5_M1_H: -3800
- B5_M1_V: -6000
To characterize the beam after the lens, we use the Smartek camera, its sensor is placed at an initial position (z=0) which corresponds to the bench coordinate X=-450+22.5-12 = -439.5mm along the beam axis between the galvo and the quadrant.
We take data points at:
- z=0 at 14h37 utc (integration time is set to 8000 us)
- z=25 at 14h42 utc
- z=50 at 14h46 utc
- z=75 at 14h51 utc
- z=100 at 14h52 utc
- z=125 at 14h55 utc
##### Error in measurement: height of beam on the lens was mistakenly changed
- z=275 at 15h13 utc (we change the integration time to 16000 us)
- z=300 at 15h16 utc
- z=325 at 15h19 utc
- z=350 at 15h22 utc
- z=375 at 15h24 utc
- z=400 at 15h26 utc
##### New measurements with corrected height of beam on the lens
- z=275 at 15h37 utc (we change the integration time to 16000 us)
- z=300 at 15h43 utc
- z=325 at 15h46 utc
- z=350 at 15h49 utc
- z=375 at 15h52 utc
- z=400 at 15h55 utc
- z=425 at 15h58 utc
- z=450 at 16h01 utc
- z=475 at 16h05 utc
The analysis of these data points gives the following parameters for the B1s beam going towards the quadrant:
* Sagital beam : w0s = 322.3 +/-2.3 um, z0s = -116.8 +/-2.3 mm, zRs = 306.7 mm
* Tangential beam : w0t = 256.7 +/-1 um, z0t = -68.9 +/-1.3 mm, zRt = 194.6 mm
This gives the following coordinates for the Gouy phase at 45 deg:
* For the sagital axis (vertical), X = -439.5 + z0S +zRS = -249.6mm
* For the tangential axis (horizontal), X = -439.5 + z0t +zRt = - 313.8 mm
Therefore the QPD sensor should be placed around X = -281.7 mm. This means that the QPD oblong hole should be around X = -258.2. Since this position is too close from the galvo, we displaced the galvo by -75 mm in Y in order to shorten the optical path. And we finally placed the QPD sensor at X=-350-23.5 = -373.5 mm, and Y = 450 mm.
We then aligned the beam on the QPD by acting on the B5_M1 picomotors with -3800 steps in H and +2800 steps in V. We then were able to close the galvo loops as shown on Fig.1. We notice that the corrections in vertical have large fluctuations. This is probably due to the limited distance between the galvo and the quadrant (only ~123 mm).
We added a beam dump in reflection of the QPD.
During this process we had to remove and add some components, namely:
- Removed: Beam dump and screws weighing 169g.
- Added: clamp weighing 20g
- Added: 5 cable ties weighing 6.5g
- Added: cable extension weighing 42g
- Added: clamp and screws weighing 23g
- Added: new lens (f = 125 mm) mounted on column and clamp weighing 267g
- Removed: clamp plus screws weighing 15 g.
- Removed: beam dump in reflection of B5_M1 with clamp and screws weighing 169 g.
- Removed: clamp plus screws weighing 16 g.
- Added: Beam dump and screws weighing 170g (in reflection of the B5_QD1 quadrant).
We stop here for the day. Tomorrow morning we will rebalance the bench and check the alignment with the bench suspended in air.
The beam parameters obtained from fitting the measured data can be seen in the attached Figure. It must be noted that the sagital and tangential data points have been fitted independently.
ITF found in COMMISSIONING Mode and DOWN state, with MISALIGNED_ALL_BUT_NI state on ARMS_LOCK node (equal to SINGLE_BOUNCE_NI state from ITF_LOCK).
The activities of the shift were:
- from 6:00 UTC to 10:00 UTC Maintenance
- from 7:00 UTC B1s QD path installation on SDB2 (Gouaty, Romero); still in progress.
ITF left as it was found: COMMISSIONING Mode and DOWN state, with MISALIGNED_ALL_BUT_NI state on ARMS_LOCK node.
After restoring the Injection loops yesterday, we notice that the system was not completely stable. The issue was initially identified through an alarm on the RMS value of the BsX_TX signal (Fig. 1). Further investigation showed that the excess noise originates from the EIB bench actuators (Fig. 2). All actuators show increased noise starting from the EIB rebalancing performed yesterday morning. Around 17:30h LT, we went into the Laser Lab to do a visual inspection on the bench, but we coudn't identify anything suspect (nothing touching or blocking the bench somehow). We slightly tightened the screw on the plexiglass panel that had been opened to move some weights on the bench yesterday. This appeared to reduce the excess noise somewhat, but the noise is still present this morning (see FFT on Fig. 3 of the broadband noise before and after).
Investigations are ongoing.
Entry of August 17, 2026 remained in draft by mistake.
ITF found in TROUBLESHOOTING Mode and DOWN State.
All times are UTC.
07:20 - 08:14 INJ: EIB restore (Melo, Lagabbe, #69586, #69587).
08:06 - 08:56 VAC: Activity on minilinks near to BS tower (Macchia, Francescon).
08:38 ISC: Etalon loops turned off (operator).
11:35 DET: Start of SDB2 intervention preparation (see attached document):
- ITF_LOCK node in DOWN state, and PAUSE operation;
- INJ_MAIN node in AUTO mode, requesting IMC_RESTORED state;
- ARMS_LOCK node in AUTO mode, requesting DOWN state;
- DRMI_LOCK node in AUTO mode, requesting MISALIGNED_PR_SR state;
- DET_MAIN node SHUTTER_CLOSED state, and PAUSE operation;
- SDB2 LC Angular Loops Open and SBE Position Loops Open;
- SDB1 Drift Control on B5 enabled (SDB1_LC.LC_B5_QD2_enbl = 1);
- SDB2 venting started (#69584);
- DET HVAC set to full speed (Portata Nominale);
13:10 VAC: NI/WI Large Valves closed (#69584).
13:35 DAQ: Virgo Process Monitoring update to version v10p4r2 (Pacaud, #69588).
14:46: Set MISALIGNED_ALL_BUT_NI state on ARMS_LOCK node.
14:28: Added -20 urad PR TY to avoid ghost beams.
14:54 DET: Offsets for SDB1_Quadrants removed (as done performing OMC Lock/Scan during periodic maintenance).
15:00 DET: Begin of SDB2 activity (Gouaty, Romero).
ITF left in COMMISSIONING Mode and DOWN state, with MISALIGNED_ALL_BUT_NI state on ARMS_LOCK node (equal to SINGLE_BOUNCE_NI state from ITF_LOCK).
The activity in DET LAB is still ongoing.
This afternoon, we entered the Detection Lab between 15h00 and 15h30 utc. We opened the door of SDB2 vacuum chamber.
Then we blocked the bench and adjusted the blocked position in order to have the bench as close as possible from its nominal setpoints.
Nominal setpoints are the following : TX = -280, TY = +880, TZ = +24, X = +10, Y = +500, Z = -190
The blocked position we achieved is the following (see Fig.1): TX = -239, TY = +819, TZ = +24, X = +82, Y = -152, Z = 154.
In these conditions, we can observe a beam on the B1p camera as shown on Fig.2 (the beam is miscentered by only +0.4 mm both in X and Y). The NI single bounce beam is set, with the SDB1 bench under B5 drift control (with B5 QPD offsets at 0).
We opened the OMC shutter and checked that the B1s beam is more or less centered on its camera (see Fig.3). The B5 beam is also visible on the camera (Fig.4).
After this check we closed the OMC shutter.
We then tested the camera that we intend to use tomorrow to characterize the beam properties on B1s. To this purpose we plugged the power cable of the camera on mezzanine camera of DAQ box 51 using the second slot (slot 1), that is normally used by B1t_Cam. We unplugged the ethernet cable (green) from B1t_Cam and plug it on our camera. We updated the serial number of SDB2_B1t_cam in process DET_Img_CEB. We also took this opportunity to correct the pixel size of the B1s camera (see entry https://logbook.virgo-gw.eu/virgo/?r=69551 ). Process stop and started at 18h22m29 utc.
We leave the ITF in NI single bounce for the night.
.png)
.png)
.png)
.png)
Xenhost160 has been replaced with another PowerEdge M640 as the broken one.
- olserver403
- olserver404
recovered.
After properly recovering EIB after the incident of Friday afternoon (#69582), we attempted to close the BPC loop, but the BsX_PZT_UH correction was too large and the loop was unable to close, as the control action was continuously driven away. We therefore discharged the DSP-card integrators and adjusted the BPC setpoints to keep the beam approximately aligned on the Mode Cleaner.
Since the resulting alignment was not optimal, the following actions were performed, with only the BPC loop closed:
- The corrections of the BPL loop on ACL were adjusted to maintain the rough beam position.
- The Automatic Alignment loop was closed.
- The SIB1 bench
TYsetpoint was slightly adjusted to compensate for the changes introduced by the BPC loop, particularly onTY. - The BPC setpoints were gradually brought back towards their original values while monitoring the corrections to prevent saturation.
- We attempted to bring all BPL setpoints back to 0. However, the
Xsetpoint was kept at-100 µm, as setting it to 0 resulted in excessively large BPCUHcorrections (approximately-6 V). - The
MAR_TYbench setpoint was returned to its original value, as otherwise the BPCXandTYcorrections were not properly centered.
After these adjustments, all loops could be successfully closed at full bandwidth, including the RFC loop.
Some notes:
1. while steering the corrections of the BPL loop we noticed that changes we made on the TX and TY were making effect on X and Y because of a copy-paste error on the ACL code and we fixed it. Now all DoFs correspond to their respective corrections.
2. The BPL loop was closed after the BPC, this was probably the cause of the bad alignment we had on the BPC and probably the reason why it could not close on the nominal setpoint to begin with. Therefore, it should be the first loop to be closed (for future reference).
I have released a new version v10r4p2 of VirgoProcessMonitoring, that fixes some issues in the WEB interface for browsing configuration file history, and add some facilities for the defintion of the directories used for log files and application executables. All instances have been restarted to use this new version around 15:30 today.
15h10 LT closed both arms: valves Cryo-NI and Cryo-WI
This morning, staring from 7:30h UTC we continued the work on EIB. In order to close again the loop, we had to move around some weights on EIB, and touch a bit the water tubes and then the loop could close correctly. Fortunately no weights had to be added or removed.
At around 14:30 LT the EIB control actuator (ACT5) went out of range without any apparent reason. Attempts to open and close the EIB loops were unsuccessful.
I went to the LL and checked/rearranged the water tubes, which are a known weak point and a common cause of EIB going out of range. Although I was able to bring the TY closer to its nominal position, the loop still could not be closed.
I then contacted Henk Jan, and we remotely adjusted some microsprings to bring the EIB back towards its nominal position. This was not sufficient, however, as the microsprings were approaching their maximum dynamic range.
The current solution is therefore to adjust the weights on the bench to bring the EIB closer to its nominal position and allow the loops to be closed. However, this action is very time consuming and we decided to postpone it to Monday morning since the ITF is not needed during the weekend. Therefore, I left the system in safe mode: all the loops opened (BPC, AA, RFC), put the PMC in scan and blocked the beam on the output of Laser Bench.
This morning, staring from 7:30h UTC we continued the work on EIB. In order to close again the loop, we had to move around some weights on EIB, and touch a bit the water tubes and then the loop could close correctly. Fortunately no weights had to be added or removed.
ITF found in LOCKED_MICH_HF_DC in COMMISSIONING mode.
At 12:00 UTC we collected several crisis of ISYS. We were not able to close EIB loops. Melo in Laser Lab to check on it with Bulten from remote. After a deep investigation we decided to postpone the intervention on EIB for Monday morning.
ISYS left in safe conditions.
TROUBLESHOOTING mode set.
We plan to resume operations on 1st September
ITF found in COMMISSIONING Mode and LOCKED_ARMS_IR State.
All times are UTC.
07:55 - 08:35 INJ: Damper removal to allow refence beam reaching phase cameras (Melo, Tacca, #69570).
08:54 - 09:55 ISC: B4/B1p PC realignment, works on EPRB (Tacca, Guo, Bothra).
10:05 - 10:39 INJ: PC pickoff alignment (Melo, #69572).
12:09 - 13:10 ISC: Installation of PC enclosured on EPRB (Tacca, Guo, Bothra).
13:44 - 14:53 DET: Checks in DET Lab (Tacca, Guo, Bothra).
The work during the shift was the Check of Phase Camera alignment carried out by Bothra, Guo, Melo, and Tacca (see #69578).
ITF left in COMMISSIONING Mode and LOCKED_MICH_HF_DC State.
We checked the alignment of both B4 and B1p phase cameras profiting of the interferometer locked with MICH at half fringe.
Work for B4 phase camera:
- at first we removed the scanning FP and its auxiliary optics since it has not benn used for a long time (scanning FP and photodiode have been stored by Suzanne);
- then we checked the superposition of the reference and the B4 beams: even if the power on B4 is not too high, we were able to optimize the superposition between the two beams. We had to remove the neutral densities on the reference beam path since the coupling of the pick-off beam into the oprical fiber in the laser bench is good but not yet optimized (see entry 69572);
- we barely centered the photodiode and we are happy for the time being;
- we added a new cover on the bench to sheld the laser beam (see pictures 1 and 2).
Work for B1p phase camera:
- we checked the superposition of the reference and the B1p beams and it was already almost good;
- as requested by Romain, we add a beam dump on the reference beam path for the SDB1 intervention scheduled for next week (see picture 3).
Figure 4 shows the beams acquired by the phase cameras after the intervention: B4 is not perfectly centered, we will optimize its alignment once all the alignment loops are engaged in CARM NULL; B1p is showing fringes, we checkd the path of the beam on the bench and we realized that the fringes are coming from the ITF, probably due to some clipping on the suspended benches not perfectly aligned at this stages.
Next steps:
- Remove the beam dump fon the B1p path after the SDB1 intervention;
- Together with INJ people, optimize the coupling into the fiber on the laser bench for the reference beam. Don't do it before installing back neutral densities on EPRB to avoid to burn the photodiode;
- Optimize the superposition of the beams and the alignment of the photodiodes once the alignment loops are engaged in CARM NULL.
Some issues occured last week in the RDS stream on the archive_50Hz directory : as consequence the rds Frame File list was built with some frame files not correctly handle by the Fd library .
To avoid this kind of issue, the Fm package has been upgraded in the v5r00 release
- to check the expected frame file suffixes
- to choose the correct frame file location as the same file could be in the online frame disk buffer or in the archive one
After some tests, this release has been put in operation on the VirgoOnline Fm servers :
- operations performed between 2026-08-13-10h12m58-UTC and 2026-08-13-10h32m24-UTC
Since the power on the phase camera was very low, M. Tacca asked me to align the reference beam. So, I went to the LL and with the two mirrors in front of the phase camera pick-off in fiber I tried to align the beam around 12.30h LT. For the alignment, I took as reference the channel EPRB_B4_PC_CAR_PWR and the corresponding camera image. The alignment is not easily done since the mirrors are difficult to reach by hand, but some improvement could be done as shown in the attached image.
.png){kind=link}
.png){kind=link}
.png){kind=link}
.png){kind=link}
