The goal of the shift was to implement a drift control om the BS angular control and continue the CARM offset reduction.

In yesterday's turn we already observed that, while closing in on the IR peak, the BS started moving until the beams reflected by the arms would be clearly distincted (Fig.1). We failed to resolve this issue with the local control, and this was probably the cause of the increase on the B1p_DC power that ultimately caused the unlock.
To prevent this from happening again, we decided to implement a drift control loop on the BS and add it to the metatron node that closes the drift control for the PR. Since we had already identified an error signal for the BS alignment (B1p_QD2_6MHz, it just needed a tuning of its gain and demodulation phase), the implementation of the drift control was simple: copying the same instructions already used for the PR was sufficient. We tested this new loop during a lock of the CITF (Fig.2), and it appears to be working as intended by reducing the error signal for the BS alignment and decreasing the DC power on B1p.

We added in the CITF_init.py script the functions BS_TX(), BS_TY() and BS_open() to respectively close the BS drift control on TX and TY, and to open it.

We added in the carm_darm_green_m.py script the function get_carm_darm_set() to compute the average on CARM_FREQ and DARM_FREQ.

We locked attempted a CARM offset reduction at UTC 19:24:00 and got 100 Hz away from the IR resonance. We were stopped again by the two spots appearing on B1p, despite the BS drift control. The error signals for the drift control become progressively worse while the CARM offset is reduced, until they stop behaving like error signals at all. Also, during this attempt we had to increase by a factor 4 the gain of SRCL at the moment of the hand off to the 3f.
This factor has been absorbed into the corresponding 1f/3f ratio in the .ini file of the DRMI_LOCK metatron node.

Since in the previous shifts the actuation on SRCL always tended to be dangerously close to its saturation, we modified its control filter in Acl (UTC: 19:26:00) by adding a new pole @ 65 Hz, Q=0.8, and a zero @ 90 Hz, Q=3 (Fig.3).

After these modifications we worked again on the CARM offset reduction (IR peak @ 19974 Hz).

We managed to arrive again at 100 Hz away from the IR peak (UTC: 20:09:00). During this phase, we only needed to adjust the MICH gain (Fig.4). The drift control implemented for the BS failed again to prevent from showing the same behaviour observed before (Fig.5). We tried to recover with the local controls for the beam splitter after opening its drift control, but regardless of how much we tried to move the mirror on TX, it kept drifting away until we ultimately provoked an unlock of the CITF.

In this occasion, as we tried to act on the BS local control to recombine the reflections of the arms, we noticed that we seemed to be crossing a minimum on B1p: the laser spots would get closer and then again further away from each other. This might suggest that the BS is not the only cause for this behaviour, but we couldn't find other explanations during thsi shift.

During the unlocks we occasionaly saw the photodiodes B7_PD1 and (much less frequently) B2_PD3 go into safe mode, and we had to rearm them and re-open their shutter.

We leave the arms locked on the IR, and all the ALS and DRMI nodes in down.
 

The shift was dedicated to the planned ISC activity, Carm Offset Reduction, carried out by Bersanetti and Boldrini. In parallel the planned SIN activity, SQB2 suspension, continued.
ISC Activity concluded at 21:45 UTC, infrared cavities left locked.

DET
At 14:30 UTC SDB1 was moved in safe position.

SBE
During the afternoon the vertical correction of SDB2 drifted above its safety threshold. From 20:20 UTC to 20:50 UTC I recovered its correction acting on the related Motor3.

The aim of the shift of this morning was to investigate about the CARM/DARM error signals phase noise between 10-100 Hz.

We performed some acoustic injections with a signal generator downloaded on the cell phone. We injected a line at 316 Hz at different points in the atrium. The most critical one seemed to be the rail where the fibers bring the 1064 pickoff from the LB and the 532 beams from the arms, in fact we lost the lock.

We also performed some tapping test. The idea was to listen to DARM signal with headphones in order to point out which part was most critical. However this trial was unsuccessful since we couldn't hear a clear sound when the green was locked. We then performed more usual tapping test. In this case as well the most criticial part appeared to be the fiber bringing the pickoff from the LB to the splitter box: either touching the fiber next to the LB, either on the upper rail or close to the splitter box the lock has been lost.

We then swap teh fiber with one with a metallic jacket (passing on the upper rail as well). We repeated the acoustic noise injections and the tapping and it appeared to be still critical but slightly less sensitive.

We finally performed the air conditioning switch off test. It is clear that it is the cause of the excess of noise in the frequency region 10-100 Hz (look at the 1st plot).

Here we report the GPS for each one of the actions.

A more detailed analysis and plots will be posted later. 

This morning shift has been dedicated to ALS noise investigation, the work has been carried out without any major problem.

Others activities carried out in the morning:

- FC-IM suspension control test;
- EQB1 installation of superpolished optics;
- SQB2 integration;

Air Conditioning
From 12:40UTC to 14:45UTC injection area air conditioning switch OFF test according with the shift activity.

The action seem to have solved, or at least greatly improve the SNEB TY control.

Figure 1. However the additional problem that SNEB and SWEB are moving much more in Z and TX than SPRB remains. The motion in Z is about 5 times larger.

Figure 2. Comparing the local controls of SNEB and NE, the motion of SNEB is at least 3 times larger. Which shows that this is not the motion of the ground with regard to a good inertial mass, but actually SNEB is moving more than the ground.

The goal of this shift was to continue the CARM offset reduction while keeping the CITF locked.

To lock the arms on CARM and DARM with the green, we implemented the streamlined procedure already mentioned in #51466.
In particular, we included in the change_carm_darm_test() function a direct hand off to the beating signals, therefore the algorithm is reduced to:

• lock the arms on the IR
• lock the green on the arms (without reallocation)
• (optional) run CARM_DARM_FREQ_OFFSET_v1_3.py and the contained functions CARM_FREQ_OFFSET() and DARM_FREQ_OFFSET() to match the averages of the respective *_REFL signals with the *_FREQ signals.
• run carm_darm_green_m.py and the function change_carm_darm_test(carm_set=x, darm_set=y) (giving as arguments the current values of the beating signals as setpoints for CARM and DARM) to enable CARM and DARM loops using the beating signals.
• immediately put the NARM and WARM metatron nodes to DOWN, to prevent them to attempt a re-lock the arms on the IR
• engage the boost. we noticed that applying first the ALS_Arm_control2 and then the ALS_Arm_control4 grants a smoother transition than just enabling the final filter at once.

At this point the arms should be controlled with the green and kept on the IR peak. The CARM and DARM SETs might need a fine tuning after the procedure. We managed to complete the procedure at UTC 15:10:00 and found the IR peak for CARM_SET = 19667 Hz. We set the CARM_SET 10kHz away from the IR peak and started working on the CITF.
We managed to lock the DRMI and hand-off to the 3f signals at UTC 17:02:00. To smoothen this step, we adjusted the 1f-3f signals ratios in the metatron node .ini file by considering that the magnitude in DC of the 1f/3f TF (Fig.2) is given by (1f_mag*cali)/3f_mag. We modified the ratios to match this value. The handoff to the 3f signals went smoothly.

We then started reducing the CARM offset while monitoring the CITF loops' UGFs.

We managed to keep the the CITF locked while reducing the CARM offset down to 450 Hz (UTC 17:31:00) and then to 400 Hz (UTC 17:32:00) away from the IR peak, only slightly adjusting the gains along the way (Fig.1). At 400 Hz the CITF unlocked, this was caused by an unlock of the green laser in the north arm: a possibly related large glitch can be seen in (Fig.3) on the green reflection and vco correction signals.

We recovered the arm lock at UTC 19:43:00, found the IR peak at CARM_SET= 19894 Hz and then subtracted 2000 Hz. We then relocked the CITF. The hand-off to the 3f signals was reached at UTC:20:33:00 and we started closing up on the IR peak with the CARM offset.
Only a small adjustement to the SRCL gain was necessary during this reduction phase, to mitigate the saturation of SRCL_CORR (Fig.4). A stable offset of 75 Hz was reached at ~20.58 UTC
The CITF unlocked at the end of the ramp towards 50 Hz, about 60 Hz away from the IR peak (Fig.5). This time, the DRMI unlocked first while the green laser remained locked until we brought its nodes in DOWN a few seconds later. Further investigation shows that the unlock was triggered by B1p_DC overshooting its max treshold (Fig.6).

We leave the ITF with the arms locked on the IR, the ALS nodes in down and the recycling mirrors misaligned.

We install the table enclosure for EQB2,. It has two accesses for the cables on the north side. (Figure 1)

When we check the position of the SQB2->EQB2 window, we find out that it is not in the expected position ( Figure 2). The beam is expected to be at the Iris position, the QD sensor is in the foreseen position. The present position of the viewport is not compatible with the optical design of EQB2.

In the next few days it will be decided whether to move the viewport or to modify the EQB2 optical design. Meanwhile the installation has been frozen.

Attachment 

Figure 1 Accesses for cables scheme

Figure 2 Viewport position within respect to the EQB2 table

The planned activity on ISC went on for all the shift without major troubles; still in progress.

Other activities: SIN - SQB2 and SQB2 suspension.

Nothing else to report.

In order to optimize the clean aperture of the BS, we have chosen the Thorlabs Polaris K1C4 holders that require to glue the optics on the holder itself.

On 14/04 we prepared the optics in the CB clean room and we let it dry for almost 20 hours,(Figure 1) but the glue has not become hard. The day after, we decided to shine the optics with a UV light ( Figure 2). 

We first try to hard the glue on a 70:30(T:R), after that we check the coating behavior. The BS did not show any discrepancy with the expected behavior so we proceeded to shine the other support. In particular we glued 3 50:50 BS and 3 70:30(T:R) BS. 

This morning we opened the chamber and found out that one mirror ear was touching the ring heater and that one side of the marionette was almost touching the cage.

We adjusted the vertical position of the chain by removing some weight (small metal bars) from the marionette.

We adjusted the yaw of the chain by slightly rotating the top plate and we could free the mirror. Then the marionette roll was adjusted by displacing some of the metal bars screwed on it.

Looking at the signals it seems that the marionette and the mirror are now free. The resonance frequencies also look reasonable.

After recentering oplevs and closing IP controls, we quickly tried to close the MAR yaw loop and we could damp the 0.033 Hz oscillation. 

We performed several electric checks but we couldn't fix the two motors of the marionette, that are still not able to move. We suspect that the issue is connected with the long cables going from the rack outside the clean room to the chamber patch panel. Further tests will be done tomorrow. 

On Friday 9/4 we did some tests on the automatic alignment of LO and CC beams on the homodyne detector using angular dither lines.

1) we checked the interplay of AA and CC loops. In particular, we observed that the angular motion of the tip-tilt actuator on EQB1_M6 mirror influences the CC phase. Fig. 1 shows the augular open-loop error signals with an angular ramp on the LO_M4 mirror on the LO path: the error signals show a ramp corresponding to the actuation ramp.  Fig 2 and Fig 3 show the same with a similar ramp applied to the horizontal and vertical angle of EQB1_M6 mirror on the CC path. In the case of M6 actuation, a fast oscillation in the angular error signals appears. The same osciullation can be observed on the CC error signal. We do not understand if this effect is due to an issue with actuator/driver cabling, of to some mechanical cross-coupling. Anyway in such conditions we cannot reliably engage the CC and AA loop at the same time.

2) we tried a different error signal, i.e. we demodulated the SQZ_EQB1_HD_DIFF_ RF_4MHz_I channel at dither lines frequencies instead of SQZ_EQB1_HD_DIFF_ RF_4MHz_mag. The angular error signals are similar in the two configurations when the CC loop is not engaged; in Fig 1, Fig 2 and Fig 3 the dither lines demodulation was swapped from magnitude to I-quadrature at mid time. However when CC is engaged the angular error signals derived from SQZ_EQB1_HD_DIFF_ RF_4MHz_I  become meaningless, see Fig 4 wth CC engaged: dither lines demodulation was swapped from SQZ_EQB1_HD_DIFF_ RF_4MHz_I  to SQZ_EQB1_HD_DIFF_ RF_4MHz_mag at mid time. We note a drift of the 4 MHz magnitude over the afternoon, that seems not connected to the status of the alignment, see Fig 5.

3) we moved the dither lines from current values (i.e. 412, 512, 612, and 712 Hz) to  frequencies above the CC UGF (i.e. 2012, 2212, 2412, and 2612 Hz). We observed the corresponding lines on the spectrum of SQZ_EQB1_HD_DIFF_ RF_4MHz_mag, however the angular error signals have  lower SNR than with the previous setting, see a locking sequence on Fig 6; this is reasonable as the ACL process to perform the demodulation is performed at 10 kHz.

This morning was dedicated to Maintenance, here a the list of activities communicated to the Control room:

• Standard Vacuum Refill (VAC Team);
• Cleaning of Central Building, FCIM Lab and Clean Rooms (Ciardelli and Menzione with External firm);
• Cm Cascina domain ports range expansion (see entry logbook #51470), carried out during the first hour of the shift.
• Replacement of PR Motor Driver (see entry # 51471);
• Work on the UTA of the WE Building (Soldani);
• Movement of Magnetometers at WE Buildings (see entry #51474);
• Maintenance of UPS (D'Andrea with External firm);

In parallel, the activities on Input Filter Cavities and preparation for SQB2 suspension proceeded.
Maintenance concluded at 10:42 UTC, it was possible to relock both infrared cavities at the first attempt.

This morning at 10:50 LT, we temporarily set only magnetometer ENV_WEB_MAG_W in chopper mode while, for the other two magnetometers, the chopper mode is off. 
Moreover, all three magnetometers are positioned in parallel (along W) as reported in Figure. 

This morning, during maintenance break, the defective high-power motor driver at PR has been replaced with a spare (scsmotor44, IP: 172.16.1.244). The functionalities of all channels (top screw, F0-F3 fishing rods) has been successfully verified.

Concerning the issue on SNEB_TY accuracy, pointed out by ISC crew last friday, the performed intervention on the suspension was quite effective, removing totally the spurious  seismic peak (fig 1).

Since we were hitting the limitation in number of ports for the Cascina domain (most probably we had already at least an event were port requests spilled over the allocated range), this morning we have expanded the range (as seen in the CmDomains file entry) from:
Cascina     cmns.virgo.infn.it     29000 29001 899 /virgoData/Cm
to:
Cascina     cmns.virgo.infn.it     28000 28001 1899 /virgoData/Cm
pushing down the start to 28000 and extending to 1899 ports.
This implyed a stop/restart of the NameServer wich went quite fast and smootly.
Checking the situation afterward with some processes restart / new processes start (including the restart of the VirgoOnline VPM instance) everything looked fine.

The hand off to 3f signals  doesnt succeeded due to  cm_commands/automation issue  but  to the fact that the DRMI_TRIGGER conditions went to zero .

As consequence the  PRLC_TRIG, MICH_TRIG and SRCL_TRIG went to zero which made the length error signals equal to zero in this case just before the completion of the 3f hand off

ITF State: Arms cavities locked on infrared.
ITF Mode: Maintenance.
Quick Summary: IMC and RFC locked; all Suspensions loops closed; All SBEs loops closed (SNEB tracking OFF and LC TY lowNoise disabled).
Activities ongoing since this morning: Standard Vacuum Refill, Cm Cascina domain ports range expansion (Masserot, Carbognani).

The main aim of the shift was to further study the CARM offset reduction process. Additionally, a new and faster sequence to lock the arms and to find the CARM offset has been succesfully tested.

We first aligned the green beam in the North arm cavity using the galvos. The corresponding BPC setpoints are:

• TILT X= -0.144, Y=-0.3
• SHIFT: X = -0.663, Y = -1.035

This allowed to lock both the North arm and West arm green lasers at 16.00 UTC.

We then unlocked the arms and misaligned the end mirrors to proceed to the prealignment of the CITF. Once reached a good alignment we saved the working points andput the CITF in misaliged state and realigned and relocked the arm green lasers.

After some attempts we successfully implemented a new green arm locking procedure, which allows to handoff the arm control directly from the single-arm IR loops to the green CARM and DARM loops. This procedure also allows to find the CARM and DARM setpoints correspondent to the IR resonance before handing off the arm control. 

The new procedure still needs some automation refinement but can be executed as follows:

• Lock the arms on the IR using metatron.
• Put the ALS nodes in locked state and lock the fast loop
• Open the CEB flip-mirror and check the superposition between thereflection LSC_ALS_C/DARM_REFL and the beating ALS_CEB_C/DARM_FREQ signals
• Set the LSC_CARM_SET and LSC_DARM_SET to the current values of the ALS_CEB_C/DARM_FREQ signals.
• run the new change_carm_darm_test() function in the carm_darm_green_m.py script and then IMMEDIATELY put the IR arm nodes in DOWN (to avoid metatron attempting a relock of the IR).
• handoff carm and darm to the beating signals and enable the boost as usual.
• fine tune the CARM and DARM setpoints if necessary.

A further change migh allow the handoff directly to the beating signals, skipping an additional step.

Since the lock of the green loops was still unstabled and unlocked avery few minutes, a finer tuning of the North arm alignment was required at 19.20 UTC. This allowed to mantain the lock for the rest of the shift.

At 19.43 UTC the new locking procedure was completed with success in 4 minutes (see figure 1). Then 10 kHz of CARM offsets were added to allow the lock of the CITF.

Since the changes to the arm lock procedure took more time than expected, a realignment of the CITF was done and we were able to lock the arms with DRMI around 20.40 UTC for 20 mins.

In order to monitor the UGFs, we adjusted the amplitude of the the lines. Then we tuned the gains of MICH, PRCL, and SRCL  to tune the UGF to the stable values(see fig 2).

We then proceeded handing off to 3f signals, we immediately unlocked the DRMI. A few seconds later the arm loop (that had been locked for more than one hour) also unlocked(see fig 3). It is not yet clear if this unlock was triggered by the CITF unlock or it happened by random coincidence.

At the end of the shift, we put the green lock down, PR and SR Parked ,and locked the arms on IR.

Friday morning we replaced the six commercial mirrors on EQB1_DL  with the superpolished and we realigned this line to overlapp the BAB and  LO on HD.

The wedge orientation is the following:

         DL_M1: 0.5 (looking in front at HR surf: thicker side at right)

         DL_M2: -0.5 (looking in front at HR surf: thicker side at left)

         DL_M3: 0.5 (looking in front at HR surf: thicker side at right)

         DL_M4: -0.5 (looking in front at HR surf: thicker side at left)

         DL_M5: -0.5 (looking in front at HR surf: thicker side at left)

         DL_M3: 0.5 (looking in front at HR surf: thicker side at right)

OVERLAPP between BAB from EQB1_DL and LO

The DL has been aligned on the LO aligned on the BAB coming on SQB1.

Without fine working we reached the a HD visibility similar to that one reached in the case of BAB coming from SQB1, HD_VIS~0.82 (GPS 13062698478)   and  in the case of BAB from SQb1 HD_VIS~0.85 (GPS 13062610699) (we switched to BAB from SQB1 at ~14.00 p.m., i.e. 12.00 utc).

All the telescope in the position reported in https://logbook.virgo-gw.eu/virgo/?r=51458