Virgo Logbook

AdV-ISC (Commissioning up to first full interferometer lock)

mantovani, magazzu, lumaca, ruggi, melo, taranto, bersanetti, berni - 16:24 Sunday 05 March 2023 (59136)

ITF recovery

The relock of today has been spent on several topics:

SSFS / IB alignment coupling

The ITF was unlocking at CARM 0 engaging SSFS. The plot is visible in Figure 1. There were some disturbances at 6Hz visible on SSFS (no loop oscillation) and bench control siguals (especially IB tx).

The reason could be a not optimal working of IB AA.

(The work has been interrupted by a failure of INJ/PSL occurred. INJ on-call has been called, see dedicated entry. BPC correction is high (UH corr).)

This kind of coupling between the SSFS and the INJ AA loop was observed in the past and was killed by setting offset on the QPD centering which minimized the coupling of the SSFS line on the QPDs.

The measurement are:

FF H 1362049034 : optimal wp is -0.35, see figure 5. (this is particularly big)

FF V 1362050783 : optimal wp is 0.157 see figure 6

NF H 1362052369 : optimal wp is -0.02

NF V 1362052982 : optima wp is -0.08

these have been updated in ISYS_Acl.

This action has solved the mis-behavior of SSFS/IB.

B1p shape @ DRMI

An other issue is that It is visible aslo in CITF that the DF shows a misalignment mode in ty. see figure 2

I've tried to align by hand to check if this was a problem of the BS AA loop but I could only increase the fringe without changing the shape of the mode.

The Figure 3 and 4 are the B1p Camera with the same gain.

We have talked to Diana.

Claudia has moved the ty of CH+DAS of NI. The B1p Camera shape @ DRMI did not change shape after Claudia's work.

SRCL set @ ~12 during the lock acquisiton

Then I found an issue (that started few days ago) figure 7. It was not possible to discharge the filters (I asked Diego and he has removed and put back the controller).

SRCL SET issue has been solved.

After having solved all these issues was possible to lock at CARM 0 1f

We locked then at 1f with the target of having gentle unlock after 1 h.

Figure 9 shows the trend of the last lock.

Images attached to this report

Comments to this report:

ruggi - 18:12 Sunday 05 March 2023 (59140)

The attached plot shows the trend of the sidebands during the first hour of CARM null, comparing yesterday and today. The large transient with the typical time scale of the yag absorption, always present in the past days, is totally disappeared, saying that the latest TCS tuning has been very effective.

Images attached to this comment

mwas - 9:50 Monday 06 March 2023 (59144)

Figure 1 shows the CARM null lock from last night, while Figure 2 shows the same channels for a lock from Jan 28 (when we were doing low noise DC read-out tests).

Figure 3 is hard to read, it is just the supeposition of figure 1 and 2.

In between the two the input power has been decreased on purpose by 25%. The 12MHz and 112MHz on B4 are between 20% and 25% lower, so in line with the input power reduction. B1p power is 40% lower, so improved more than just the input power reduction, but that can misleading as SR might not have been well aligned.

The phase camera powers are different by over a factor 2, but this is just the effect of the reduction of the reference beam for the phase cameras.

Figure 4, shows instead the the imbalance between the sidebands. The 56MHz power imbalance on B4 improved by a factor 2, the 6MHz imbalance on B4 is worse by 50%, the 56MHz imbalance on B1p is similar and the 6MHz imbalance on B1p is lower by a factor 4 after 1h long trending decrease that was interupted by the intentional unlock.

Images attached to this comment

59144_1678091546_Screenshot from 2023-03-06 08-10-54.png

59144_1678091550_Screenshot from 2023-03-06 08-11-42.png

59144_1678091800_Screenshot from 2023-03-06 08-36-25.png

59144_1678092348_Screenshot from 2023-03-06 08-45-35.png

bersanetti - 18:03 Tuesday 07 March 2023 (59169)

About the SRCL_SET servo, for some reason it was not possible to discharge the filter neither by acting on the LATCH trigger or the RESET one, and it was solved by removing the "Integrator" filter altogether, then re-engaging it.

To be more robust against this behaviour, now the filter is reset to LSC_NONE after an unlock and engaged (like all loops filter) while going to CARM_NULL_1F. To be verified at the next relock.