Virgo Logbook

AdV-COM (AdV commissioning (1st part) )

ruggi, izumi, calloni - 14:21 Tuesday 03 January 2017 (35962)

SSFS noise injections

Yesterday we made again some noise injection on MC loop and SSFS loop, changing the shape of the noise in order to measure the TF below 100 Hz.

Fig.1 (gps 1167380016,200) MC locked on RFC with a smooth control filter

Sc_MC_RFC_Err_Pre over Sc_Mc_RFC_Err_Post (MC open loop TF)

Fig. 2 (gps 1167381616,200) MC locked on RFC with a smooth control filter

SIB2_RFC_6MHz_I over SSFS_corr_RA_10KHz

Fig.3

zoom of Meas #2 around 30 Hz --- not expected features

Fig. 4 (the same injection as #2)

Meas #2 multplied by (1-Meas#1) (SSFS open loop TF)

Fig. 5 (gps 1167392235,200) MC locked on B4 with the standard control filter

Sc_MC_RFC_Err_Pre over Sc_Mc_RFC_Err_Post (MC open loop TF)

Fig. 6 (gps 1167388616,200) MC locked on B4 with the standard control filter

LSC_B4_6MHz_I over SSFS_corr_RA_10KHz

Fig. 7 (gps 1167407086,160) MC locked on B4 with the standard control filter - noise at high frequency

SSFS open loop TF

Comments to this report:

ruggi - 14:24 Tuesday 03 January 2017 (35963)

missing plots

Images attached to this comment

ruggi, calloni, izumi - 17:08 Wednesday 04 January 2017 (35971)

Yesterday we made an SSFS noise injection after the tuning of fModErr (gps 1167490266,200).

All the strange structures in the TF were still there, with significant differences in amplitude and phase.

Images attached to this comment

barsuglia - 17:14 Wednesday 04 January 2017 (35970)

In order to study the origin of the the double peak around 30 Hz, of the peak slightly below 40 Hz and of the peak slightly above 60 Hz (see plot 2, but also entry 35967), I have looked at the INJ system noise budget, as described in the documents VIR-0369A-15 and VIR-0212A-16 (Capocasa et al.). I have noticed that there are similar structures in the TF between the IB_tz control signal and the frequency noise (see for instance the attached plot, from VIR-0369A-15, but also the figure at pag.10 of VIR-0212A-16). In particular, the double peak around 30 Hz is clearly visible in the IB angular TFs. Of course this can be a coincidence, and moreover I don't know which is the mechanism allowing these structure to be present in the SSFS TF, but I think it is interesting to investigate better this point.

Images attached to this comment

ruggi, barsuglia, capocasa, izumi, calloni, pillant, bersanetti - 16:14 Thursday 05 January 2017 (35975)

As suggested by Matteo, we measured the SSFS transfer function with the IB linear alignment turned off. The structures below 100 Hz disappeared. It means that the IB angular control makes a relevant loop on the MC length through the frequency noise seen by the quadrants, amplified by the bench mechanical resonances.

The structure at 156 Hz does not change. In my opinion also that could be related to some resonance at the level of suspended IB, but I have not a guess about the mechanism of coupling.

Images attached to this comment

ruggi, majorana, izumi - 21:25 Thursday 05 January 2017 (35979)

The transfer function between SSFS_corr and one error signal of the bench automatic alignment is shown in Fig. 1. After few seconds of amazement, kiwamu suggested that such a perfect response could be due to some electronic coupling.

The second plot shows the different effect on the alignment signal if some frequency noise is produced moving the MC mirror (blue curves) or injecting some SSFS_corr.

Images attached to this comment