Reports 1-1 of 1 Clear search Modify search
Detector Characterisation (Glitches)
direnzo - 18:30 Monday 09 December 2024 (65756) Print this report
Analysis of glitch families of Decembr 8

To verify that the glitch issues of the past few weeks had passed, I analyzed the high glitch-rate period between December 7 at 18:00 UTC and 8 at 6 UTC. In this, I identified two main glitch families, in addition to the usual 25-minute glitches.

Figure 1 shows the spectrogram of the strain data of December 8 from 2 to 5 UTC. The usual Scattered Light glitches are visible as excess energy between about 20 and 50 Hz. Further down, between 10 and 20 Hz, a second (new?) family of glitches is particularly visible as "bursts" of noise at about 2:40, 2:50, 3:25, and 3:40 UTC. Additionally, a very loud glitch at 3:24:02.6 UTC caused a two-minute long excitation of the 48 Hz lines, with a beat every about 3.5 seconds, compatible with the frequency of the crossbar resonances of filter 7 of the terminal suspensions: #48183.

Figure 2: detail of some scattered light glitches with superimposed the frequency of the most correlated predictor channel: V1:Sc_WE_MIR_Z. To study these glitches, I applied the same technique described in many other logbook entries during O4 (#63761#64022 , ...). The list of most correlated channels is the same as in other periods with high microseismicity, such as the one described in #65142,  #64022 and #64026. This is the file with the results.

Figures 3 and 4: spectrograms, over different time windows, of the other glitch family at low frequency. As can be seen from the second image, these glitches are centered between about 10 and 20 Hz, and appear, almost like a beating, with an almost constant rate of 9 glitches every 10 seconds. I tried to investigate repeating the same analysis as above, as if they were Scattered Light, but without success; no correlation emerged with the velocity derived from all the position channels of Virgo. Further investigations should be done to understand their origin, however, they do not seem to be linked to the glitches between 20 and 80 Hz present until a week ago: #65625.

Figure 5: spectrogram of the very loud glitch and the resulting ringing of resonance at 48 Hz for about two minutes. I have not had time to investigate this specific glitch in detail, but it would be interesting to do so, to understand what is causing it and the coupling with the frequencies of the Filter 7 crossbars at the North End and West End.

Images attached to this report
Non-image files attached to this report
Comments to this report:
Paoletti - 11:17 Tuesday 10 December 2024 (65764) Print this report

The "very loud glitch" that excited many crossbars around 48 Hz appears to be due to an IPS mains line voltage drop that occured at 03:24:02,600 + 100ms.

The associated magnetic glitch is visible in all Virgo buildings magnetometers.

Images attached to this comment
direnzo - 14:03 Wednesday 11 December 2024 (65782) Print this report

I add to Federico's excellent catch on the origin of the glitch and the 48 Hz resonance the pointer to a past entry reporting the same kind of noise: #48619.

Search Help
×

Warning

×