As planned, at 07.51 UTC the SR RH power has been increased up to 9.8 W (V=27.37 V) in order to have the RoC around 1426 m (-5 m wrt 1431 m).

The shift was dedicated to the ISC/OSD/DET (PRC matching with CHRoCC fine tuning, scan of arm cavities and OMC in CITF), carried out by Boldrini (see report #56753).
At the beginning of the shift, we performed the measurement with the PR CHRoCC V_DAC set yesterday afternoon (0.43 V). Here is the Timetable of the performed measurements:

7:11 UTC - Free Spectral Range Scan while in LOCKED_ARMS_BEATING, command CARM_SET_scan_FSR(6);
7:17 UTC - Cavities Scan Matching, command SCAN_TEMs_TOT(1) (see attached plot #1);
8:34 UTC - 5 minutes of CITF locked as a reference;
8:56 UTC - OMC Scan (see attached plot #2);
9:20 UTC - CITF Scan;

At 9:29 UTC we changed the V_DAC of the PR CHRoCC to the second value agreed (0.41 V). 
We waited 90 minutes and then performed the second set of measurements:

11:14 UTC - 5 minutes of CITF locked as a reference;
11:21 UTC - OMC Scan (see attached plot #3);
11:43 UTC - CITF Scan;
11:54 UTC - CARM_SET_scan_FSR(6);
11:59 UTC - SCAN_TEMs_TOT(1) (see attached plot #4)

At 12:12 UTC we changed the PR CHRoCC to the last agreed value (0.39 V). We waited again 90 minutes and then performed with the last sets of measurements:

13:44 UTC - CARM_SET_scan_FSR(6);
13:49 UTC - SCAN_TEMs_TOT(1) (see attached plot #5);
14:14 UTC - 5 minutes of CITF;
14:32 UTC - CITF Scan;
14:49 UTC - OMC Scan (see attached plot #6); The ITF unlocked during the last part of the scan at 15:02 UTC. The unlock was possibly caused by an Earthquake (Magnitude 6.6, South of the Kermadec Islands, 13:44 UTC, see attached image #7), which also prevented the lock of the CITF for around 30 minutes.
15:45 UTC - OMC Scan (see attached plot #8);

After the measurements were completed I set back the V_DAC of the CHRoCC back at 0.43 V at 16:09 UTC.
Activity concluded at 16:10 UTC, infrared cavities left locked.

No DMS alarms received during the shift.

We repeated the study of the matching for different values of the CHRoCC power, for the new value of the SR RH power. Since reducing the CHRoCC power seemed to be the most promising direction in the past few days, we took data for 0.43 V -> 0.41 V -> 0.39 V of CHRoCC power. The following table recaps the GPSs of each data gathering:

Before taking the first reference of the DRMI, we verified that phases and gains of the DRMI were correct (PR_AA, BS_AA, PRCL,MICH,SRCL in 3f). Only the phase of the 18MHz signal for PR_AA had to be tuned from -1.05 rad to -0.7 rad. After taking the first reference for the DRMI, we also took 120s of data with the LF line of the NI pumped to ampl: 6e-4, to be used for further comparisons between the simulated and actual alignment sensing. The GPS for this measurment is:

UTC:8:47:30 +120s

Notice that we relocked the arms to reset the alignment of the mirrors once per 30-40 minutes on average. The SR was praligned at each lock of the DRMI. The first and third points of the measurements were timed with two distinct earthquakes, one in China and one in the Pacific Oceans. They were not strong enough to cause any significative distruptions, but the CITF was more fragile during the CARM_SET ramp and we could only perform 3 ramps (in particular, the DRMI unlocked during a ramp up but relocked on its own at CARM_SET = 32250 Hz, the following three ramps down-up-down were smooth, the third ramp up unlocked the DMRI again, Fig.1). The unlocks of the DRMI were very sudden, and do not offer clear explanations as why some ramps successfully arrive at the end while some do not.

Please refer to the operator entry for the plots relative to today's scans.
 

Figure 1 show the phace camera 56MHz powers during the three steps of 0.43V, 0.41V and 0.39V on the CHRoCC. The situation is consistent with what is observed with the OMC. The 56MHz LSB power is decreasing, and the OMC sees the reduction in power in HOM of the 56MHz LSB. The 56MHz USB power is increasing, and the OMC sees the increase of the TEM00 power with lower CHRoCC voltage.  The first step in decrease of CHRoCC voltage also improves the 56MHz LSB vs USB overlap, both on B1p and on B4.

Figure 2, 3 and 4 shows the B4/B1p PC images for respectively 0.43V, 0.41V and 0.39V. On B4 the beam is getting clearly more round with lower voltage on the CHRoCC. But is hard to draw a conclusion from the B1p image shape, it doesn't significantly get better or worse. So the OMC provides some information here by doing a mode decompostion, but that mode decomposition could in prinicple be done directly on the B1p PC image, and skip the 20min OMC scan, and 30min of analyzing the scan to figure which mode is which.

A question is what we expect for the 56MHz on the anti-symmetric (B1p) port. It is not clear to me if we expect the 56MHz to be on the dark fringe, or to have a bright TEM00 due to the Schnupp asymmetry. And whether this depends on which sideband is used to control the MICH degree of freedom (if 56MHz is used would we expect the 56MHz to be on the dark fringe, or that remains of dark fringe due to the Schnupp asymmetry, as the error signal still corresponds to a beat with the carrier light). This could actually be different for the 1f and 3f locks. I don't have a clear view here.

Overall the situation seems to improve with lower CHRoCC correction. The checks of arm cavities shows a worse matching between the CITF mode and the arms. But could it be due to the input beam matching? As the PR CHRoCC power is changed, the input beam is no longer well matched with the arms as the POP lens is changing, and to have a true view whether the CHRoCC tuning is improving things one would need to retune the input telescope to match well the arms on the carrier input beam (CITF not locked)

ITF State: LOCKED_ARMS_BEATING
ITF Mode: Commissioning.
Quick Summary: IMC and RFC locked; all Suspensions loops closed, all SBE loops closed; Meteo Station not working;
Activities ongoing since this morning: Cavities Matching Scan in progress.

I turned off the SC_PLL device server process on the phasecam1 machine. It is concurrent to the currently running Py_SC_PLL thus it was causing errors.

I have looked at these two OMC scans at 0.45V and 0.47V on the CHRoCC and compared to the 0.43V case from yesterday as a reference value. This complements the analysis of the OMC scans from yesterday

Figure 1 shows the overview. Figure 2, 3 and 4 shows the USB order 0, 1 and 2 modes. While Figure 5, 6 and 7, show the LSB order 0, 1 and 2 modes.

There is the same pattern as observed yesterday, increasing the CHRoCC power reduces the USB TEM00 power, by a factor 5. While the order 1 and 2 modes stay about the same within 10%. For the LSB the order 0 and 1 mode remain the same, while the order 2 mode keeps increasing by almost a factor 2 when reaching 0.47V.

Summary

• Lowering the CHRoCC power seems to improve the 56MHz mode matching by 40% for the USB and LSB from the point of view of the OMC
• The mode shape is very bad in any case with ~30% of order 1 mode and ~50% of order 2 mode, and many bright higher order modes

I have looked at the scans for 3 powers of the PR CHRoCC. Note that the USB (as described by the phase camera) is at lower OMC temperature, so that is why it appears first and is marked at negative frequency in the figures below.

Figure 7 is the overview, with the time roughly rescaled into MHz. The black line is for 0.43V, purple for 0.41V and grey for 0.39V. Also the position of the LSB/USB order 0, 1 and 2 are marked as a guide for the eye. The order mode has roughly 1/3 of the power of the TEM00, and the order 2 mode has roughly 1/2 of the power of the TEM00. This is a lot of misalignment and mode-mismatch. But I don't know what our expectation should be. Note that there is quite of higher order modes to. For example the USB order 3 mode has also 50% of the power of the TEM00.

Figure 1, 2 and 3 show respectively the oder 0, 1 and 2 mode for the 56MHz USB. Figure 4, 5 and 6 show the order 0, 1 and 2 mode for the 56MHz LSB.

The USB TEM00 seems to increase in power by ~20% for each step of reduction in CHRoCC power. The order 1 and 2 modes stay roughly the same, changing by less than 10%.

The LSB TEM00 stays the same within 10%. While the order 1 mode is reduced by 30%, and the order 2 mode is reduced by 40%.

So for both the 56MHz LSB and USB the mode shape improves as the CHRoCC power is reduced, as the ratio of TEM00 to order 1 and 2 modes is improving by 40%. The balancing of the sideband TEM00 remains roughly the same but changes signs. For 0.43V, the USB TEM00 was 30% lower than the LSB TEM00, while at 0.39V the USB TEM00 is 30% more powerful than the LSB TEM00.

When I look at the mode shapes during the scan at 0.43V the USB order 1 mode is ~80% a vertical misalignment, and the order 2 mode is completely vertical. For the LSB the order 1 mode is also ~80% vertical, while the order 2 mode is unclear as the camera shows a vertical mode, then a cross shaped mode, then a bull-eye mode, and there is only one image per second, so it is unclear what is corresponding to the peak, the order 2 mode in the time series does look like it is double with one peak half the size of the other, so it is reasonable to think that the horizontal and vertical order 2 mode contribute, with power ratio of a factor 2 between them.

When I look at the scan at 0.39V. The USB shows the same pattern, and the LSB as well, with also an ambiguous order 2 mode shape, that happens to be the brightesst for the horizontal order 2 mode, but this is somewhat random due to the sampling of 1Hz for the camera image.

Code used for these figures is in /users/mwas/OMC/OMC_scan_20220812

So far, we applied 6 W to the SR RH and the RoC should be 1431.4 m according to the simulated dynamics of the SR RH [RoC=-1.43*Pappl+1440 m (VIR-1166A-19)].

As planned, at 15.45 UTC the SR RH power has been decreased down to 2.8 W (V=14.63 V) in order to have the RoC around 1436 m.

The shift was dedicated to the planned ISC/OSD/DET activity (PRC matching with CHRoCC fine-tuning, scan of arm cavities and OMC in CITF), carried out by Casanueva, Spinicelli, Guo.
During the first part of the shift, we finished the measurements started Yesterday morning, exploring the previous PR CHoRCC value (V_DAC at 0.39 V). Here is the timetable of the measurements performed;

From 7:00 UTC to 7:25 UTC - CITF Scans;
7:28 UTC - 5 minutes of CITF locked;
7:45 UTC - Cavities Scan Matching, command SCAN_TEMs_TOT(1) (see attached plot #1);

At 7:59 UTC we changed the DAC_V of PR CHRoCC in the opposite direction, from 0.39 to 0.45 V. We waited 90 minutes and then proceed with the measurements:

9:39 UTC - CITF Scan;
10:35 UTC - Attempted OMC Scan, it was followed by an unlock a few seconds after starting;
10:58 UTC - OMC Scan (see plot #2). At 11:11:50 we noticed a spike in the Powers of the CITF and a change in its behavior.
11:33 UTC - Free Spectral Range Scan while in LOCKED_ARMS_BEATING, command CARM_SET_scan_FSR(6);
11:38 UTC - SCAN_TEMs(1) (see plot #3);

At 11:51 UTC we changed the V_DAC of PR CHoRCC, from 0.45 to 0.47 V. We waited 90 minutes and then performed the last sets of measurements.

13:25 UTC - CITF Scan;
13:43 UTC - OMC Scan (see plot #4);
14:14 UTC - CARM_SET_scan_FSR(6);
14:19 UTC - SCAN_TEMs_TOT(1) (see attached plot #5);

At 14:33 UTC, as agreed with the Weekly Coordinator, I set back the Voltage to 0.43 V as a starting point for tomorrow's shift.

Infrared cavities left locked.

Vacuum
At 8:45 UTC the value of the flag mean_VAC_TOWERWE__MG2_CH1_60 suddenly increased (see attached plot #6). Francescon came on site and, at around 12:30 UTC, switched on a supplementary P61 Pump in the WE Building. The value returned to its threshold, further investigation will be performed next week.

Today, we have explored the two remaining points of the CHRoCC power (0.45 and 0.47 V), going in the opposite direction as yesterday (towards higher values). For each point several measurements have been done. Here we report about the scans done with the CARM offset, while locked on the CITF.

Figures 1 and 2 show a trend of these scans and the zoom of one of them for the first point, 0.45V. Notice that the same offset that we used yesterday as final point (32250Hz) did not work every time today. For this reason, in order to make sure that enough data was available, we have tried also to reduce it slightly, to ensure more sets of scans (31700 Hz). FIgures 3 and 4 show the same information but for the last point (0.47V).

We noticed a curious thing: for most of the scans the peak corresponding to the lower HOM2 of the 56MHz (as seen by the North arm) was located around 30kHz. However, for some of them we could not see it, even scanning larger ranges (up to 31700Hz since already 31900 was causing the unlock of the ITF).

We leave arms locked on IR and the CHRoCC power back to 0.43V. TCS crew will now change the SR RH power for the tomorrow activity.

ITF State: LOCKING_ARMS_BEAT_DRMI_1F;
ITF Mode: Commissioning.
Quick Summary: IMC and RFC locked; all Suspensions loops closed, all SBE loops closed; Meteo Station not working;
Activities ongoing since this morning: Relock of CITF in progress.

The goal of the activity was to measure the induced curvature of the NI CO2 actuators, carrying on the activity started yesterday (56729). The nominal values, from OSCAR simulations, are reported in the following table:

NB: The NI DASes were already switched on (IN @ 0.075 W and OUT @ 1.7 W) in order to compensate for the NI CP cold lens.

Since the NI CH induced curvature observed yesterday was not large as expected, today we tried giving double the value, so the today value for N CO2 actuators were:

NB: The NI DASes were already switched on (IN @ 0.075 W and OUT @ 1.7 W) in order to compensate for the NI CP cold lens.

To consolidate the knowledge of the ITF initial status, it was locked in CITF (15.03-15.30 UTC) and then a matching scan was performed (see entry 56737).

At 15.51 UTC the ITF was put in NI Single Bounce and at 16.10 UTC an HWS-DET acquisition was started.

At 16.20 UTC the NI CH was switched ON  at 0.282 W (double the value of yesterday). Once the curvature reached the steady state, the NI DAS powers were increased of the amount reported in the previous tables: up to 0.4 W for the inner ring and 4.1 W for the outer one.

The entire behavior of the curvature is shown in figure 1. There is still some overcompensation by NI DAS powers, with respect to NI CH power.

The CH curvature steady state still does not fit with the calibration (VIR-0017A-22): further investigation are needed.

At the end of the measurement, a new matching scan was performed (see entry 56740); unfortunately, it was not possible to lock the CITF.

At 21.30 UTC the ITF was locked on IR and we left the CO2 TCS actuator’s powers in the following condition:

Updated version of the Tolm network including the WAB Tolm one

Explore different CHRoCC settings and to study the matching of the sidebands in the CITF.

Most of the data analysis has already been explained in the entry by Cabrita (56736).

Additionally, at 14:00 UTC the thermal transient for a third point finished, and we repeated the procedure for the matching scan. The following table shows the UTC of each scan over the course of the shift.

For the CITF scan, the CARM offset was swept between 3000 Hz and 32250 Hz with the CITF locked (56737).

While the CITF unlocked repeatedly during the "return" part of each ramp for 0.41 V, we decided that the data gathered during the first half of each ramp were sufficient and, given the long time taken by each measurement, it was more important to move on the the next point than to try and make the CITF stay locked. With this idea in mind, from the ISC point of view, no significant change to the lock parameters were made over the course of the shift.

For the third CITF scan, we were able to complete one scan at 0.39 V but lost lock ramping up on the second. This point will be studied further during tomorrow's shift.

We left the ITF in single bounce.

The shift was dedicated to the planned ISC/OSD activity (Arms cavities analysis of 6 MHz SB in CITF, study of asymmetry between Warm and Narm measurement, OMC scan), carried out by Cabrita, Casanueva, Boldrini, Spinicelli.
During the first part of the shift we performed the measurements with the initial CHRoCC DAC_V value (0.43). Here is the list of the performed measurements:

6:58 UTC - Free Specral Range Scan while in LOCKED_ARMS_BEATING, command CARM_SET_scan_FSR(6);
7:04 UTC - Cavities Scan Matching, command SCAN_TEMs_TOT(1) (see attached plot #1);
7:39 UTC - 5 minutes of CITF Locked;
7:52 UTC - CARM SET Scan in CITF;
From 8:26 UTC to 8:47 UTC - OMC Scan with the remote support of Gouaty (see attached plot #2);

After this last measurement, at 9:12 UTC we changed the DAC_V of the CHRoCC PR from 0.43 to 0.41. After waiting 90 minutes we attempted again the measurements, starting from the CARM SET Scan in CITF, but the ITF unlocked after a few minutes. Here the timetable of the other measurements carried out with this configuration:

11:23 UTC - 5 minutes of CITF Locked; 
From 11:33 UTC to 11:55 UTC - OMC Scan; (see attached plot #3).
12:04 UTC - CARM_SET_scan_FSR(6) 
12:14 UTC - SCAN_TEMs_TOT(1) (see attached plot #4).

Completed this second set of measurements, we adjusted again the CHRoCC DAC_V, from 0.41 to 0.39. We waited 90 minutes and started the third set of measurements:

14:11 UTC - CARM_SET_scan_FSR(6);
14:16 UTC - SCAN_TEMs_TOT(1) (see attached plot #5).
From 15:06 UTC - OMC Scan.

OMC Scan in progress.

Yesterday evening and this morning we implemented a DISCHARGE_GR state. Its purpose is to move the values accumulated into the integrator of the AA actuator loops to the new setpoints. It is done by continuously monitoring the corrections done by the loops and saving it into the SQZ_FLT.set file under the constraint that the cavity transmission is satisfying. The discharge process can be safely executed when the cavity relocks.

The new DISCHARGE_GR state substitutes the previous discharge mechanism which caused a rapid jump of the actuator. The jump was as high as the the accumulated correction. It was frequently exciting a short FLT cavity mirrors oscillation.

In the upgrade process, the two existing functions restore_FCIM_FCEM_from_file() and restore_GR_M5_M7_from_file() were modified and the new state was added.

Detailed modifications log is present in the automation git in the dev-discharge branch.

The new code was successfully tested and will be left for the long run during the weekend.

With the interferometer locked in CITF, we scanned the arm cavities around the second order modes of the 56 and 6 MHz sidebands. The scan in the west arm looks very weird, with very low power and weird shapes (Fig.1). From the north arm scan (Fig. 2) the peak heights for each SB HOM2 are:

LSB56 HOM2: 0.12

USB56 HOM2: 0.09

LSB6 HOM2: 0.255

USB6 HOM2: 0.273

Aftwerwards the CHRoCC power was decreased by 5% and after waiting for ~1.5hrs, re-doing the arm scans show that the power in the west arm scan increased significantly (Fig. 3) and the shape improved a bit, although the shape is still dubious. From the north arm (Fig. 4) the peak heights for each SB HOM2 are:

LSB56 HOM2: 0.44

USB56 HOM2: 0.21

LSB6 HOM2: 0.37

USB6 HOM2: 0.53

So, the mismatch increased for all sideband modes (that is expected from simulations, for CHRoCC changes all SB order 2 modes should follow the same trend). LSB56 mode 2 increased by 3.7x, USB56 mode 2 increased by 2.3x,  LSB6 mode 2 increased by 1.4x and USB6 mode 2 increased by 1.93x .

We can see that the mode imbalance for the 56 MHz increased, with the LSB56 having a higher mode 2 content - according to simulation this is expected for a positive increase in PRM RoC around the nominal case, and indeed by decreasing the CHRoCC we increased the RoC of the PRM, so that is consistent with simulations. Simulations do not predict a mode imbalance for LSB6, so it is interesting to try to reproduce this effect in simulation.

A full analysis with all data points taken and using the B4 phase camera to normalise the sideband power during the scans will be made soon.

In order to allow the check of IR and GR AA during long term measurement I left the system in IR_AA state, I updated all the values of the camera integration time to avoid if possible saturation.

I also put HDMIR from -3000 to 2000 and I punt the IR filter toward HD Cameras. In this way we could also check the position of the IR reflected by FLT on EQB1.

I left the automation in the branch dev-discharge. An entry which describes this brench will follow.

N.B. I don't save these integration time in the config file thus take care if you restart the process

Two CITF locks are taken in account here:

1 - blue - gps=1344248818 - PR parking position and pre-alignment as bad as usual

2 - red - gps=1344248818 - first CITF lock after the attempt to optimize PR parking and pre-alignment.

Fig 1: error signals of cavity alignment in the conditions supposed to be referable to the respective CITF locks

Fig 2, Fig 3: figures of merit supposed to give information about the quality of CITF alignment.

The good pre-alignment appears to produce a CITF in a better shape. Quantitatively, the difference does not seem very important, but it is better to have an opinion from the experts.

To be taken into account that the second CITF lock occurred 6 hours after the first. The figures of merit have shown to be sensitive to other parameters, which stability in time is not perfectly known. Part of the changes we are showing could be due to drifts of those parameters.

New data can be added to the analysis (yellow plot in the attached figure). The alignment of the cavities done yesterday evening in the new parking position went on without big drifts, even if something was still visible, especially on north side.

Olserver124 recovered. All nodes in Cluster Mysql on line and working:

[root@olserver124 ~]# ndb_mgm
-- NDB Cluster -- Management Client --
ndb_mgm> show
Connected to Management Server at: localhost:1186
Cluster Configuration
---------------------
[ndbd(NDB)]    2 node(s)
id=3    @90.147.137.26  (mysql-5.6.21 ndb-7.3.7, Nodegroup: 0, *)
id=4    @90.147.137.27  (mysql-5.6.21 ndb-7.3.7, Nodegroup: 0)

[ndb_mgmd(MGM)]    2 node(s)
id=1    @90.147.137.24  (mysql-5.6.21 ndb-7.3.7)
id=2    @90.147.137.25  (mysql-5.6.21 ndb-7.3.7)

[mysqld(API)]    15 node(s)
id=5    @90.147.137.25  (mysql-5.6.21 ndb-7.3.7)
id=6    @90.147.137.25  (mysql-5.6.21 ndb-7.3.7)
id=7    @90.147.137.25  (mysql-5.6.21 ndb-7.3.7)
id=8    @90.147.137.24  (mysql-5.6.21 ndb-7.3.7)
id=9    @90.147.137.24  (mysql-5.6.21 ndb-7.3.7)
id=10 (not connected, accepting connect from any host)
id=11 (not connected, accepting connect from any host)
id=12 (not connected, accepting connect from any host)
id=13 (not connected, accepting connect from any host)
id=14 (not connected, accepting connect from any host)
id=15 (not connected, accepting connect from any host)
id=16 (not connected, accepting connect from any host)
id=17    @90.147.137.24  (mysql-5.6.21 ndb-7.3.7)
id=18    @90.147.137.25  (mysql-5.6.21 ndb-7.3.7)
id=19    @90.147.137.24  (mysql-5.6.21 ndb-7.3.7)

ITF State: ITF Locked on IR.
ITF Mode: Commissioning.
Quick Summary: IMC and RFC locked; all Suspensions loops closed, all SBE loops closed; Meteo Station not working; WE Vacuum Tank pressure above threshold.
Activities ongoing since this morning: No activities known so far in the Control Room.