This morning we swap the rtpc7 with the rtpc10 :.

• the RFC loop was opened
• the optiical tranceiver and the fiber were moved from the rtpc7 Tolm link 1 to the rtpc10 Tolm link1 .
• The SIB2_rtpc servers were stopped ans restarted to run on the rtpc10
  • the SIB2 Daqbox were not reconfigured as nothing were updated
• The Tolm packet paths were upgraded for the following Tolm packets: "LNFStoSIB2", "INJTproToSIB2", "ENVnoiseToDaqRoom" and "EER_ADC_INJ0_1" . As consequence the related server was restarted or reloaded  : LNFS_demod. INJ_Tpro, ENVnoise and  ERR_ADC7674

After these operations, the RFC was relocked successfully  and the arms  too

ITF State: RFC Unlocked.
ITF Mode: Maintenance.
Quick Summary: IMC locked, RFC loop manually opened by Chiummo this morning at 6:50 UTC to allow the swap of rtpc; all Suspension loops closed; all SBE loops closed; 
Activities ongoing since this morning: Measurement on mSAS on SDB2 (Bulten), Vacuum Refill (VAC Team), ENV sensor activity at NE Building (De Rosa).

The objective of the shift was to prepare the swap of the IMC control loop  from the RFC error signal to the CARM_FREQ beating signal.

We also wanted to characterize the beating to CARM/DARM loops by injecting noise. The shift was therefore divided in two phases.

We started the shift by locking the ALS loops with reallocation and drift control engaged, and then switched to the CARM and DARM basis using the beating signals.
After engaging reallocation and drift control on the arms, we implemented the updated control filter prepared by Ruggi for the reallocation and CARM DARM loops (switch performed at 16.30 UTC, see figure 6). The filter change involved moving a zerofrom 0.15 Hz to 2 Hz  and allowed increasing the gain from the initial +/-50 for NARM/WARM  to 240.  This seemed to reduce the low frequency (<1Hz) oscillations of these signals and increase their accuracy but figure 5 shows an increase in the high frequency components of the correction signals when the filter was swapped. A further analysis of this swap might be necessary.

To prepare the handoff to the IMC loop, we performed a scan of the setpoints of the control loops for CARM and DARM to find the IR resonance peaks on B7 and B8, thus obtaining a condition analogous to locking the arms on the IR before enganging CARM to SSFS.
At this point, a sudden unlock of the RFC (17:43:30 UTC, Fig.1) provoked the unlock of the green. We tried to lock it again, but the procedure kept failing at the moment of engaging the reallocation.
We concluded that the new filter, while being effective at reducing the noise after the reallocation loop was closed, made the lock more difficoult.
We decided to phase back to the previous filter to engage the reallocation, and let the CARM and DARM loops change the filter, with the new gain, at the moment of their activation before swapping their input signals.
This proved successful and we obtained a stable lock of the green with reallocation which lasted, with few interruptions, for the rest of the shift. Notice that we opted not to engage the drift control.

We then completed the tuning of the CARM/DARM setpoints (around 19:10 UTC, Fig.2) in order to find the IR resonance. We obtained a tramsission for the IR of around 0.4W on B7_DC and B8_DC, later (around 19:24 UTC, Fig.2) increased to 0.6/0.7W with:

• DARM_SET = -118.4685 (corresponds to -36837 Hz on the LSC_DARM_FREQ_ZEROED channel when divided by the calibration factor)
• CARM_SET = -36.0273 (corresponds to -11202 Hz when divided by the calibration factor)

The calibration factor is 6.7e-4*4.8, where 6.7e-4 is the gain found as explained in the logbook entry #50897, and 4.8 is the new gain factor related to the new filter.

At this point Ruggi started injecting noise on the RFC at 19:13 UTC for 200s to obtain a calibration for the signal swap. The swap itself requires to understand the logic of INJ loops in Acl, and for the moment we decided to postpone it and proceed with the characterization of the CARM/DARM loops in the second part of our shift.

We performed the following noise injections. Due to an error in the command sent to ACL to set the noise filters, we used the "flt_NONE" to filter white noise, which has a pole at 10 Hz and a gain of 1e-20. The very low gain of the flt_NONE explains the huge coefficients that we used.

• 20.48.00 UTC: START test real noise injection on CARM.
• 21:12:20 UTC, GPS:1298668358: Amplitude: 2e18 (5 min)
• 21:18:00 UTC, GPS:1298668698: Amplitude: 2.5e18 (1 min)
• 21:19:18 UTC, GPS:1298668776: Amplitude: 3e18 (5+ min)
• 21:27:00 UTC, GPS:1298669238: Noise OFF

In Figure 3 are reported the transfer functions of the two injections of duration 5 minutes (respectively 21:12:20 UTC and  21:19:18 UTC) and in Figure 4 the correspondent FFT of the noise injected.

Due to an error in the DARM noise injections, the NI local controls opened, compromising this operation, that will be planned for the next one.

We left the interferometer with the ALS loops in DOWN and the arms locked on the IR.

This afternoon was dedicated to ISC green lock, the activity went on all the shift without any major problem; we left the cavities locked on the infrared.

SUSP
At 21:52UTC the NI local control opened by the guardian probably because of the activity, properly closed.

This morning we started trying to solve issues related with DAQ together with Alain, as reported in logbook entry #50920. The Tolm Packet size was increased and then we could send to the DAQ all the channels we needed.

We used this time to add and implement a new configuration file for SRCL (/virgoData/VirgoOnline/LSC_SRCL.cfg) in the LSC_Acl process. As of now, the file contains the usual loop structure, recycled from PRCL, with the needed changes for the sensing matrix and possible trigger signals that SRCL will use for its single lock acquisition, as it is done for PRCL. This is for the single recycling cavity lock and auxiliary commissioning configurations. The integration with the new trigger logic will be done soon.

Then, together with Paolo, we started with the CITF alignment in a safely way. With the arms locked on IR, SR parked and PR parked. We first misaligned NE, WE and WI. B1p was very quiet, so we scan few urad for PR and NI to see a clear movement on B1p camera. Next, leaving PR parked and NI in its good position, we started moving SR to get power built between NI and SR. Finally we fine adjust SR in Yaw and Pitch to have multiple reflections centered (GPS: 1298629178 + 300 sec)

•  UTC 10:19:20 PR misalinged and scans of SR. See figure 1. CITF_SR_scans_20210301.png

While doing all this, we implemented the new code for the Metatron Nodes for the Suspensions, in order to use and also detect the new safety. The tests went fine, here some more remarks:

•  MAR_[TX,TY]_SET setpoint MUST NOT be used for the parking of the mirrors; instead, the new gNAME MAR_TY_ALI should be used for this; this is what Metatron does now, the "delta" feature has been decommissioned.
• All SUS nodes support as usual the SAVE_ALIGNED feature, but please do not use it for the PR just yet, we have to understand a bug, possibly related to a too fast read/write sequence which happens only in the PR case; given the new structure, SAVE_MISALIGNED will do nothing for PR and SR.
• UTC 10:28:05  SR MAR TX was misaligned in order to trigger PR to move to a safe position.  See figure 2. CITF_PR_security_test_20210301.png
• SUS_PR was able to detect the safety tripping; the procedure is the following: it detects the mirror being parked and it self-requests the MISALIGNED state, forcing the new gNAME accordingly; from there onwards the suspension is simply misaligned, but with a notification message explaining the tripping of the safety; one can realign the mirror, but in this case it is best to check things by hand. The same will be replicated for the SR mirror, but this is not online yet.
• UTC 10:48:30 start misaligning SR and PR, then lock again arms on IR. See figure 3. CITF_SR_PR_parked_arms_locked_20210301.png

At around 11:30 UTC we waited for DAQ/Computing to recover from the crash of rtpc7, and the check of the Injection.

At 12:10 UTC we locked the arms again on IR, but we could see that the power in transmission in B7 and B8 was decreasing over time. Analizing the FFTs of the transmitted power, we could see peaks of noise at around 7 Hz and 9 Hz (one per arm), which are close to the dithering lines frequencies for NArm and WArm. After noticing this we unlocked the arms. We cheked if the drift control was engaged, when the comand was sent using the corresponding Metatron node, but it was not the case, but only for the Acl side. Therefore we rebooted the ASC_Acl process and we tried again to turn the boost on for NArm and WArm, and we recovered the cavities.

Moving forward, at 14:20 UTC we started the CITF alignment in order to collect more data for the search of triggers. For clarity, here we describe the steps of the alignment.

• WE, WI and NI misaligned using NARM and WARM metatron nodes.
•  PR aligned from Tx = -43 urad Ty = -49.9 urad to Tx = -43 urad Ty = 0.4 urad and then was fine-tuned to Tx = -38.5 urad Ty = -0.7 urad looking at B1p camera. Then we saved these values using metatron's SAVE_ALIGNED state of SUS_PR.
•  PR misaligned to Tx = -38.5 urad Ty = -51 urad using PR metatron node (MISALIGNED state).
•  SR aligned from Tx = -84 urad Ty = -671.5 urad to Tx = 84 urad Ty = -193.5 urad and then was fine-tuned to Tx = 83.9 urad Ty = -192.3 urad looking at B1p camera. Then we saved these values using metatron's SAVE_ALIGNED state of SUS_SR.
• PR realigned using PR metatron node (ALIGNED state), which uses previously saved values in step 2.
• WI aligned from Tx = 67 urad and Ty = 100 urad to Tx = 13.6 urad and Ty = -8.5 rad.

We left the Interferometer alone for 10 min at UTC 15:04. All this process can be seen in figure 4.

At the end of the shift we restarted LSC_Acl, after implementing many changes:

• we acquired some new channels, all from SIB2_Photodiodes; as they were already present in the Tolm packet, there was no issue about it;
• we added the demodulation phase monitor for B2_56MHz;
• we generated a few normalized signals for sensing and triggers for the CITF, and sent them to the DAQ;
• we integrated the new LSC_SRCL configuration file, and added a first draft for the lock flags, and sent all the new relevant channels to the DAQ;

Please remember that as of now there is no automation about SRCL, so any change done by hand has to be reverted by hand as well.

This morning we performed measurements to characterize the acoustic isolation of the ESQB1 enclosure.

We used two acoustic sensors: the microphone already placed on the bench (ENV_ESQB1_MIC) and one additional microphone (ENV_DET_MIC) we placed in the SQZ bench area. The ENV_ESQB1_MIC is a B&K mod.4190 (stardard microphone), the  ENV_DET_MIC is a B&K mod. 4193 (extended infrasound microphone).

First we performed an Huddle test (microphones placed side by side, a 1cm away) to cross-calibrate the two probes. Figure 1 shows the measured transfer function. The modulus of the transfer function  ENV_DET_MIC over ENV_ESQB1_MIC is flat and equal to approx 0.9 between 4Hz and few 100 Hz: thiscan be attributed to a miscalibration of one of the two probes, and needs to be corrected. Below about 4Hz, the transfer function grows becase of the difference in response between the infrasound and the standard microphone (see also https://logbook.virgo-gw.eu/virgo/?r=21732 ). We can decide to analyze data only above 4Hz, or compensate for this difference in response below 4Hz.

Then, we measured the ambient noise in different positions of the external (5 positions) and internal microphones (4 positions), as shown in Figure 2 (internal mic positions) and Figure 3 (external mic positions).

We used two noise sources: (1) DET laboratory air conditioning, kept on in FULL power during the whole test; (2) the generator of laminar air flux  in the ESQB1 area

For each tested position we have taken 2 minutes data with the laminar air flux on  (Figure 4), and 2 minutes data with the laminar air flux off (Figure 5).

The attached text file contains GPS time of single actions.

Between about 10Hz and 150Hz we noticed a slight amplification of the noise measured by the bench microphone. We suspect this might be associated to resonance modes of the enclosure. We performed a tapping on the enclosure, East side. The result is shown in Figure 6. 

Analysis of the data will follow.

The shift was dedicated to the planned ISC Central Interferometer and Automation activity, carried out by Bersanetti, Tapia, and Ruggi. 

Other activities carried out during the shift:

• During the morning Fiori, Tringali and Sorrentino performed acoustic noise measurements in the DET Lab for Squeezing;
• During the weekend the rtpc7 became unreachable (see report #50924). At 11:00 UTC Masserot and Cortese completed the inspection of the machine and, as a safety measure, the RFC loop was opened by Spinicelli before the reboot.

I had a closer look on the spikes on Feb 27 and 22.

Figure 1 shows an example of a few spikes from Feb 27. One can note that they only happend when B4 is large (and conversely B1p is small), that is when the BS moves to a dark fringe position for MICH.

Figure 2 is a zoom. The glitches are visible on B7, B4 and B5 and less clearly on B1p. On B7, B4 and B5 there have the same sign, the power goes up and down at the same time. For B5 the interference is only constructive, the power goes only up compared to the level far from the glitch, for B7 the interference can be constructive or destructive. For B4 it is hard to say as the base level fluctuates much more than the glitches due to the MICH fringes, but they give me the impression of being both constructive and destructive intereferences.

The glitches seem to be also correlated with the regular 1.7Hz oscillation visible on B2. The glitches only happen when the B2 power is high. I don't know what is the light on B2, I expect PR is misaligned in this configuration so there is no real "B2" beam reaching the B2 photodiodes.

Figure 3 is an exemple from Feb 22. Again the glitches happend when the B4 power is high and B1p power is low (MICH dark fringe), and at the same time the B2 power is high.

Figure 4 and 5 show a spectrum of B2 at that time. The oscillation on B2 is very regular and mostly at 1.7Hz. I guess this a dither line. Which mirror is being modulated at 1.7Hz? This dither seems to be putting some mirror in an interference position at each cycle.

One thing to check if this not due to a combined effect of PR and SR misalignment. That is that SR compensates the misalignment of PR, and causes the beam to be realigned in the CITF, but with a lateral shift. This could explain why there is a beam reaching the B2 photodiode. One way to check would be to move PR in TX. If that is indeed the explanation, then PR probably needs a TX component in its parking position in addition to the TY misalingment.

Note that simulations by Michele Valentini show that up to 17W could be sent to detection with 25W of input power, see figure 1. So the current situation may be understimating the powers we will obtained once CITF is fine aligned and matched, and the powers could increase by a factor almost 3. So it could correspond to up to 10W on the diaphragm.

This week one more the rtpc7, the one devoted to the SIB2, became unreachable and all its data lost  from the  2021-02-28-10h19m36-UTC. 

Some new monitors were setup and were running before the crash but we have learn no new informations .

The SIB2_rtpc was reboot at 2021-03-01-11h05m25-UTC and all the servers restarted

I took a look at the glitches and unlocks over the weekend (plot 1).
There were three unlocks on friday afternoon. There are all due to an unlock of the RFC that seems to be due to a sudden loss of correction going to the ML (plot 2).
The rest of the time the IMC remained locked.

I took a look two glitches that occured on EOM_CORR but which did not make the IMC unlock:
- The fisrt one on the 28/02 at 00:51 (plot 3). EOM_CORR seems to start by its own way before the other signals, even the error signal of the IMC. Nothing visible in the slow signals (ENV, MAG, HVAC...)
- The second one on the 28/02 at 14:57 (plot 4). The frequency loop oscillates at about 23 kHz for a brief moment without making the IMC unlock. This is a quite unsual frequency for oscillation on the frequency loop.

The fact that we had no fast unlocks over the weekend might be explained by the increase of the UGF of the IMC loop on friday (log entry https://logbook.virgo-gw.eu/virgo/?r=50909 ). It has been increased from 50 kHz to 120 kHz. Lately it was quite usual to see an oscilaltion at 60 kHz during the fast unlock.
We should keep on monitoring the system in that configuration to confirm its good behavior.

The misfunctionning related to new DAQ channels was fixed by increasing the channel buffer size of the ISC_rtpc(rtpc20) by

• adding the line "TOLM_PROCESSOR_CHANNEL_BUFFER_SIZE    65536" in the configuration file of the ISC_Tpro server
• and restarting all the  rtpc servers

ITF State: Arm Cavities Locked.
ITF Mode: Commissioning.
Quick Summary: IMC and RFC locked; all Suspension loops closed; all SBE loops closed; rtpc7 not responding to ping and the related process on VPM are down.
Activities ongoing since this morning: Preparation work for ISC activity.

Just a remark: the worsening of PR and BS local signals in the trend data likely shows that the noise on the arm powers is affecting the dither signals, built demodulating the power itself. The dither signals are the ones in loop if one look at the very low frequency. The local signals are out of loop, and they register the increased noise on the in-loop signals.

As reported in #50914, the spikes were present again tonight, then they disappeared during the activity, then they came back without any intervention on the ITF:

• what has happened in the time frame of the disappearance is the following:
  • many optics were aligned/misaligned, but aside a very few urads (or fractions of) for PR and SR and the usual alignment by the drift control for the arm cavities, no macroscopic changes were done; also, the translation of the PR did not change a lot in absolute value (Figure 1, more info below);
  • we suffered three unlocks of the INJ system; the first one in particular shows the biggest change in the working point (for example, the Z position of the MC) (Figure 2, more info below);
  • we relocked the ALS on both arms with the reallocation on the mirrors; nothing was changed on the North Arm, while the BPC on the West Arm was open at the beginning, so it was closed before relocking the arm; the drift control for the beam direction was engaged during the time of the ALS lock; as a test, I disabled manually the BPC on the West Arm while the IR was locked without the spikes, but they did not come back;
  • the spikes did come back eventually, very slowly, during a single, non perturbed lock (still ongoing); they showed up at the same time on both arms, but initially they were a bit more visible on the North Arm, possibly because the SWEB alignment is a bit worse w.r.t. the one of SNEB;
• in Figure 1 there is the overall trend of the phenomenon, alongside the position of all the optics: it can be seen that when the phenomenon worsens, the local controls angular signals do not get particularly worse with the exception of the ones for the PR translation stage (which controls the overall beam direction) and the BS angular controls (which do the same, but only for the West Arm);
• in Figure 2 there is the trend w.r.t. the INJ system: here a couple of things can be noted:
  • the strange behaviour of PSTAB_AMP_CORR, which was stable during the spikes' disappearance, then had a jump around 20:00 UTC, followed by a slow trend; I am no expert of the system, so I don't know the possible meaning of this; a longer trend of the week is in Figure 3, maybe it can be helpful; the week before (Figure 4) seems much more quiet on this matter, but still the correlation is unclear (no spikes on the 16th Feb, for example);
  • the three jumps of Sa_MC_F0_Z are due to the three unlocks, followed by the scan of the RFC resonance; no correlation is evident, and even some mistuning of FmodErr never had this kind of effect on the arms powers, to my memory;
  • the corrections of the BPC are slowly diverging; the DMS pointed me to look at _UV in particular, which went above the 4.5 V threshold (Figure 5), but even much higher values did not cause this effect in the past;
• in Figure 6 there is a snapshot of the arms loop: it can be seen that the accuracy worsens, but the corrections do not change;
• in Figure 7 and 8 there is the trend of the powers of the green beams (transmitted and reflected) and the local controls used by the BPCs; there is some trend on the NEB_MONI signals, but not on the WEB ones.

In conclusion, the major effect was seen on the PR translation stage (and the BS local controls) which could point to something related to the pointing of the main beam, but correlation to INJ signals was not straightforward to me; beside this, it does not look like it is related to the position of some optics, or to the arm control loops; at a first look I could not find a correlation to the green beams either, but a switch-off test could be definitely useful, also in view of later stages of lock acquisition where ALS will not be needed anymore.

Today we worked on the pre-alignment of the CITF and the test with the lock of the ALS:

• we started by relocking the arm cavities, in order to relign the optics and the input beam; we could see that the spikes on B7 and B8 were present again (see #50856 and #50910); we spent some time looking into this without finding any conclusive cause, but we could notice some secondary flashing appearing on the B1p camera;
• while looking into that we re-enabled the B5 drift control on SDB1, with no apparent change; we checked the position of the SNEB/SWEB benches (see #50913 for more details); while the power on B8 improved in its steady value, no changes were observed concerning the spikes; given that the new setpoints degraded a lot the alignment of the green beams, which we wanted to use later on in the shift, we decided to restore the old positions; the re-alignment will be done on Monday profiting on the presence of both crews;
• then we tested the logic for the new protection for SDB1 (see #50911); the logic is working fine and now we decided the correct implementation which will be used by the Automation; the code is being written offline so, in addition to what Paolo wrote in the aforementioned entry, PLEASE DO NOT USE METATRON TO STEER PR AND SR; the new version of the code should be deployed on Monday;
• then we used the new procedure for the CITF alignment, checking that everything was fine on SDB1/2 and that the safety between PR and SR was working (Figure 1):
  • we misaligned NE, WE and WI;
  • we realigned PR, adjusting its position finely looking at fringes and the B1p camera; SR was correctly parked so the new safety allowed this;
  • we misaligned PR, then we realigned SR; with PR parked, there is no issue about SDB*; we proceeded with the fine alignment for SR;
  • keeping SR aligned, we realigned PR and WI, so we got to have the CITF aligned;
  • free swinging data has been acquired starting from 18:45:00 UTC, duration 15 minutes; these data will be analyzed offline;
• we started to write some test triggers in LSC_Acl for the CITF; however, while the new channels themselves did not cause errors, putting them in the DAQ did, throwing an error about the Tolm Packet, but pointing to a completely unrelated channel (see the log attachment); we will contact experts and figure this out (and add a couple of channels to LSC_Acl); in any case the trigger signals can be reconstructed offline for this first analysis;
• then we went back to standard conditions, and relocked the arms on the IR: optics realigned, same power levels as before, spikes still there;
• we locked the ALS on both arms, reallocating the correction on the mirrors, in the single arm basis; this was smooth and without issues; then we realigned the recycling mirrors (see Figure 2), first SR (at 19:09 UTC) then PR (at 19:11 UTC); the green lock was undisturbed and everything looked fine on SDB2: the DC levels of B1p and B5 were only slightly higher with respect to the pure CITF configuration we had before;
• after this we misaligned the recycling mirrors and relocked the arms on the IR beam, we found out that the spikes were gone from both B7 and B8; since the very next step involved less use of online data, we kept the arms locked trying to address also the spikes issue; more details will be posted in the proper thread.

Yesterday afternoon the AEI source was turned from BAB mode into squeezed vacuum mode, for the first time after the installation in late October. On first the green locking/pump beam was found a bit misaligned on the OPA cavity, and a bit of tweaking on the steering mirrors in front of the OPA was needed before locking. Given the OPA locking error signal, we expect a level of produced (anti)squeezing between 5 and 6 dB.

After setting up the squeezer we performed a few tests on the homodyne detector, which was previously aligned on BAB with 96% visibility.

1) Slow coherent control loop: we checked the SNR of the CC sensing without the BAB, using the same parameters as in the previous test with the BAB, see fig. 1. To close the loop on HD_M4_Z mirror actuator we had to lower the gain below the value of 20 to avoid oscillation.

2) Phase scans: we performed some slow scans of the CC phase, observing the changes in the LO shot noise from the rms of the HD differential audio channel. We find that the maximum (anti)squeezing corresponds roughly to the minimum (maximum) of the Q quadrature of the SQZ_EQB1_HD_DIFF_RF_4MHz demodulation (the I quadrature is the CC error signal), see fig. 2.

3) Squeezing level: from the spectrum of HD differential audio channel we measure a squeezing level of about 2 dB and an anti-squeezing level of about 5.2 dB at high frequency, see fig. 3. We observe some excess noise below 200 Hz with squeezing injection.

4) Impact of subcarrier: we opened the shutter of the subcarrier beam, and we slightly tweaked the SC_M1 and SC_M2 mirrors to optimize the alignment (just looking at the HD DC signals); we did not observe any change in the squeezer control signals; we set the OPA in scan mode and tried to observe a transmission peak for the SC laser on the SQZ_PumpPhase_PD_DC channel, but we did not see any. With the OPA locked again, we repeated the phase scan with the SC laser on, see fig. 4: the noise level on HD differential audio channel is higher bot for the extra shot noise and for additional low frequency technical noise

5) automatic alignment on HD: error signals for AA loop with dither lines are rather clean; however we did not manage to close the loop, and we did not spend time do debug it.

Further tests of the CC loop will be done starting from next Wednesday. This afternoon the source was turned into BAB again. The OPA transmission is diverted to the SQZ_Homodyne photodiode in order to monitor the BAB power over the weekend.

The afternoon we monitored the powers sent to the detection while the CITF was being aligned. This corresponds to the situation shown on Fig.1  between 18:44 and 19:02 utc, with the end mirrors misaligned. One can see that the maximum power on B1p_PD2 is 1.25 mW, corresponding to a power of 5.2W on the SDB1 bench.

Then the CITF was aligned with the arm cavities locked on the green beam. This corresponds to the situation between 19:11 and 19:15 utc. In this configuration the maximum power reached on B1p_PD2 is 1.5 mW, corresponding to 6.2 W on SDB1.

In case of accidental misalignment of the SR mirror (which should not happen with the new safety), the ghost beam that could hit the B1/B5 diaphragm would be lower than 4W, and could therefore only marginally exceed the safe threshold value (~3W). This statement is valid with 25W of injected power.

It is however crucial to protect the OMC and the B1 photodiodes during the CITF alignment. The procedure should to be to always keep the OMC slow shutter closed during the CITF alignment and the lock acquisition until the ITF is locked at dark fringe.

Fig.2 shows the fluctuations of the SDB1 B5 QD2 signals and of the angular corrections sent to the marionetta. One can see that at the beginning of the CITF alignment we had left the SDB1 drift control closed and large fluctuations were observed in the TY correction. Therefore, we decided to open the drift control (after updating the set points to the current values) before aligning the PR mirror. The procedure to follow concerning the bench angular control remains to be refined.

Upon arrival I found both cavities unlocked.
The planned ISC activity on CITF (carried out by Bersanetti, Ruggi from remote) went on for the whole shift without major problems. ITF left with both cavities locked.
Commissioning Mode stopped at 21:56 UTC.

Other parallel activity:

• Tacca, Vardaro: FCEM and FCIM.

This afternoon, we checked the alignment working point of the SNEB and SWEB benches while the arms were locked on the infrared beam.

We found that the optimal angular positions to have the B7/B8 beams well centered on the photodiodes are the following:

• SWEB: TX = +144 urad, TY = +1778.5 urad
• SNEB: TX = -575 urad, TY = -131 urad

As shown in the attached figure, the power on the B8 photodiode was significantly improved with these angular setpoints.

However after this check we restored the old setpoints of the bench which correspond to a better alignment of the green beams.

In order to prevent damages on SDB1 due to powerful beams coming from the aligned CITF, a protection has been implemented in PR and SR DSPs. First of all, the  alignment procedure for PR and SR has been changed, and it is mandatory to use only the new one, otherwise the protection will not work. PLEASE, DO NOT ALIGN PR AND SR IF YOU ARE NOT SURE OF WHAT YOU ARE DOING.

The main change is the introduction of a switch (gNAME: MAR_TY_ALI), which is supposed to have value 0 or 1. If 0, PR/SR is moved to the supposed parked position. If 1, PR/SR is moved to the supposed aligned position. The true aligned position can be found in the usual way, only after setting the switch to 1. At this point, one can find that a small refinement has to be done, and he can change the usual gNAMEs MAR_TX_SET, MAR_TY_SET. At the end, the good setting must be kept up to the next refinement, and not changed for any other reasons. If the mirror has to be parked, it must be done using the switch. No other condition is allowed, but the aligned and the parked ones. In the procedure used to refine the alignment, one can temporarily misalign as usual the mirrors by 10 urad, and this is not dangerous.

The protection consists in the fact that, when PR is in the aligned state, if SR crosses for any reason a 'forbidden region', PR is quickly sent to the parked state, avoiding flashes in the central cavity. Currently the forbidden misalignement for SR, both TX and TY, is from 15 urad to 400 urad. The protection acts on the switch, putting it at zero, and obviously it does nothing if the switch is already at zero.

As it was noticed from the IMC TF plot attached to the original entry, we suffered a substantial decrease of UGF in the transfer function of the IMC. This morning, taking advantage of a break during the ISC activities, we went to the piscina to investigate a bit and found out that it was possible to restore the UGF to reasonable value by decreasing the attenuation of the error signal fed to the rampeauto by 6dB (from 6dB to 0dB).

The UGF was below 50KHz before the intervention, it is now at around 120KHz (see attached plot: purple before intervention, blue after intervention).

This is most likely due to the fact that we did not compensate the power on IMC_REFL when we decreased the power sent to the ITF. This must be done by acting on the IPC8 via the axis 2 channel 1 in the Electronic Injection Room. The target power measured by IMC REFL when the IMC is locked should be of around 0.25V (as it was during O3b), currently it is around 0.14V. We will do it as soon as a window of opportunity opens up.

During today's shift we injected noise into North and West arms' control loop to later calculate the transfer functions and check the stability of the loop (relative to the new gains put in place last Tuesday # 50879).

The amplitudes used for the noise are the followings:

• Amplitude: 6700 - GPS time:1298372960 + 120
• Amplitude: 7000 - GPS time: 1298373100 + 120
• Amplitude: 7300 - GPStime: 1298373602 + 120
• Amplitude: 7600 - GPStime: 1298373772 + 120
• Clean data: GPStime: 1298375000 + 120

Noise injection WArm:

• Amplitude: 7100 - GPS time:1298377162 + 120 sec
• Amplitude: 7400 - GPS time:1298377292 + 120 sec
• Amplitude: 7700 - GPS time:1298377424 + 120 sec
• Amplitude: 8200 - GPS time:1298377555 + 120 sec
• Clean data GPS time: 1298377800 + 120 sec

We created a new function meant for noise injections located in /olusers/virgorun/ISC/LSC/. However 'Injection_noise_LSC(ampl)', was not producing any desirable result. That function is in the narm_lock.py and warm_look.py scripts. It turns out that the command CAL.LSC_filter_enbl = 1 is essential to turn on the noise injection and therefore we added it in Injection_noise_LSC(ampl). Therefore, the whole process ended up being longer than expected. TF yet to be plotted and posted.

Later we scan SRM in yaw and pitch until we could see multiple reflections in B1p camera. We moved SR TY from -669urad to 510urad. When we were around ~-265urad we were able to see multiple reflections on B1p camera. This happened twice in the scan. See figure 1.

Zooming-in for the first peak and studying the demodulated signals from B1p and B4 we can see an error signal of the SR cavity. See figure 2. This is not the signal we will use to lock SRCL, however it give us an idea of how the signal should look like on B2_I_56MHz, once PR is aligned.

Finally, we moved SR also in pitch from -200 urad to 200urm, but we could not distinguish peak in B1p nor B4 PDs. See Figure 3.