More PSTAB glitches are being produced. The first plot shows these glitches in h(t).
These glitches are monitored by Omicron. Plot 2 shows the PSTAB monitoring channels where we see the glitch rate in PSTAB_PD1_AC and PSTAB_PD2_AC. As explained in https://logbook.virgo-gw.eu/virgo/?r=46436, These glitches are now best monitored by PSTAB_PD2_AC.
We will continue to produce an offline veto for GW searches.
# looking at one one the mentionned glitches: 1/08/19 - 16h33mn19 UTC / rough analysis
it seems Pstab goes into saturation and triggers the glitch
Pstab goes into saturation because of a too large demand on HF_CORR
Since PStab makes the job (up to the saturation) PMC_TRA is kept stable. But we see that PMC_REFL_DC is still growing accordingly to HF_CORR. Time scale is few seconds It would mean that either the PMC gets misaligned or some intra-cav losses or some mismatching upstream.
Conclusion: Pstab is ok, PMC reveals some issues (either PMC connected or seed connected)
# understand why PSL require more corrrection from PStab, understand how PMC intra cavity loss evolve
see plot1 (plot2 gives the legend)
- it is visible that the PMC REFL is worsened at each PSL stops (ENEL on 9/06/19, ML fibering on 30/07/19), see the yellow line at ~ 15 W
- it is visible that PMC REFL is more unstable after the ML fibering, likely due to a residual misalignement (although the fibering should have not affected it)
- it is visible that the PC throughput is worsening since the ENEL shutdown, no idea why. It can be fake and connected to some drifts, since the PMC_temperature (brown red) does not seem to be impacted like it used to be Oct the 28 (see plot 3 & 4, temperature increases by 5 units)
- the extra demand for PSTAB_HF_CORR around the 1/08/18 16H UTC does not seem related to anything specific, looking at the trend of the signals of interest in PSL
- the drift of the noeVAN beam (green curve) continues. We have guessed that this might be connected to the pump diodes ageing. This can induce the noticeable drift in the beam position at the neoVAN output.
# Conclusion
- worth to realign properly the PMC to reduce the overall associated instabilities, might improve the HF_CORR too
- PSL overall alignement needs to be done some time (oct. break). Also to make a direct meas. of the PMC throughput
- one more event where Pstab is forced to go into sturation because too much demand from the system: plot1
- from plot 2 we see that PMC_TRA instability has increased after the maintenance 30/07 and the PSL stop (Slave laser), but it is not 100% sure that this is rsponsible for the whole instability since it was present a bit before that period
