Virgo Logbook

Detector Characterisation (General)

verkindt - 0:38 Thursday 26 March 2020 (48810)

Detchar Hardware Injections

Using the frame file produced by N. Arnaud and copied here: /data/procdata/calibration/HInjection/input0/HI_20200325_v1.gwf
the following detchar injections have been done:
14:00 UTC : one set of injections, but with all amplitudes divided by 10
14:12 UTC: one set of injections, with the initial amplitudes
14:30 UTC: one set of injections (but not finished), with all amplitudes multiplied by 10
then, we tried to use the ITF out of earthquake mode, but ITF unlocked and it took some time to get it back to LN3.
16:40 UTC: one set of injections, with the initial amplitudes
16:50 UTC: one set of injections, with the initial amplitudes

A set of injections has a duration of 495 sec and is made of the following sine-gaussian glitches sampled at 10 kHz, with a width of 0.6 sec and a frequency f :
f=811 Hz : SNR=20,50,100,500 (4 injections, one every 15 sec)
f=409 Hz : SNR=20,50,100,500
f=211 Hz : SNR=20,50,100,500
f=129 Hz : SNR=20,50,100,500
f=73 Hz : SNR=20,50,100,500
f=47 Hz : SNR=20,50,100,500
f=31 Hz : SNR=20,50,100,500
f=19 Hz : SNR=20,50,100,500

Plot 1 shows the spectrogram of the input set of injections as read from the input frame file
Plot 2 shows the spectrogram of Hrec_hoft_raw_20000Hz (normalized by median value to better see the injected glitches) where can be seen the injections done with all amplitudes divide by 10
Plots 3 shows the spectrogram of Hrec_hoft_raw_20000Hz where can be seen the injections done at 14:12 UTC, with the initial amplitudes (plot 4 shows the same but normalized by median value).
Plot 5 shows the spectrogram of Hrec_hoft_raw_20000Hz where can be seen the injections done at 14:30 UTC with all amplitudes multiplied by 10. It confirms that too high SNR injections at high frequency induces a saturation (visible as a big glitch at low frequency).

Plots 6 and 7 shows the spectrograms (normalized by median) for the 16:40 UTC and 16:50 UTC injection sets. Plot 7 shows that the 16:50 injections were in noisier conditions.

Plot 8 shows the Omicron triggers seen over the time period of those detchar injections.

Images attached to this report

48810_1585178741_detchar_hi_2020_03_25_14_00_median.png

48810_1585178751_detchar_hi_2020_03_25_14_12_median.png

48810_1585178765_detchar_hi_2020_03_25_14_30_saturation.png

48810_1585179339_plot_V1-Hrec_hoft_16384Hz-1269179018-12800_freqtime.png

Comments to this report:

ducoin - 15:18 Thursday 26 March 2020 (48814)

The first run of the safety pipeline with few of those GPS time (high SNR, low SNR, high frequency, low frequency) looks fine. Knowing the LIGO analysis results (https://dcc.ligo.org/LIGO-T1900697) the amount of injection should be sufficient for the safety analysis.

ducoin - 14:23 Tuesday 02 March 2021 (50943)

The above post suggest, for "initial amplitude" injections, gps time systematically ~8s before the trigger visible in Omicron in HrecClean_hoft_raw. I note here the gps time I obtained from Omicron-print for each one of these "initial amplitude" injections. Please note that everything was done by hand, double check could be necessary.

injection gps max amp (SNR 500): freq

1269180790.5527,1269189670.5527,1269190270.5527 811
1269180850.6133,1269189730.6133,1269190330.6133 409
1269180910.6797,1269189790.6797,1269190390.6797 211
1269180970.7344,1269189850.7344,1269190450.7344 129
1269181030.7969,1269189910.7969,1269190510.7969 73
1269181090.8438,1269189970.8438,1269190570.8438 47
1269181150.9219,1269190030.9219,1269190630.9219 31
1269181210.9688,1269190090.9688,1269190690.9688 19

injection gps high amp (SNR 100):

1269180775.5215,1269189670.5527,1269190255.5371
1269180835.6133,1269189715.5977,1269190315.6133
1269180895.6641,1269189775.6641,1269190375.6641
1269180955.7344,1269189835.7344,1269190435.7344
1269181015.7969,1269189895.7969,1269190495.7969
1269181075.8438,1269189955.8438,1269190555.8438
1269181135.9219,1269190015.9219,1269190615.9219
1269181195.9688,1269190075.9688,1269190675.9688

injection gps low amp (SNR 50):

1269180760.5137,1269189640.4980,1269190240.4902
1269180820.5898,1269189700.5742,1269190300.5664
1269180880.6484,1269189760.6484,1269190360.6484
1269180940.7031,1269189820.7031,1269190420.7031
1269181000.7656,1269189880.7656,1269190480.7656
1269181060.8438,1269189940.8438,1269190540.8438
1269181120.8906,1269190000.8906,1269190600.8906
1269181180.9688,1269190060.9688,1269190660.9688

injection gps lowest amp (SNR 20):

1269180745.5527,1269189625.4941,1269190225.4707
1269180805.5898,1269189685.5898,1269190285.5352
1269180865.6328,1269189745.6484,1269190345.6328
1269180925.7031,1269189805.6719,1269190405.7031
1269180985.7656,1269189865.7656,1269190465.7656
1269181045.8438,1269189925.8438,1269190525.8281
1269181105.8906,1269189985.8906,1269190585.8594
1269181165.9688,1269190045.9219,1269190645.9688

narnaud - 18:32 Wednesday 03 March 2021 (50969)

In the original frame file I generated, the injections were separated by 15 seconds and centered, meaning that the Gaussian term of the first injection is maximum at t0=7.5 s from the beginning of the frame file. And so on and so forth every 15 seconds.

=> More information below [the actual GPS times do not matter: they got overwritten by Didier before making the injections] and in the attached plot:

Samping frequency = 10000 Hz => dt = 0.0001 s
Injection gap = 15 s
=> duration = 495 s
# of samples = 4950000
Injection 1 / 32: t0 = 1269091119.36 [2020-03-24 13:18:21.357850+00:00 UTC], f = 811.0 Hz & tau = 0.1 s (Q = 509.57), snr = 20 => a = 1.9074e-21
Injection 2 / 32: t0 = 1269091134.36 [2020-03-24 13:18:36.357850+00:00 UTC], f = 811.0 Hz & tau = 0.1 s (Q = 509.57), snr = 50 => a = 4.7685e-21
Injection 3 / 32: t0 = 1269091149.36 [2020-03-24 13:18:51.357850+00:00 UTC], f = 811.0 Hz & tau = 0.1 s (Q = 509.57), snr = 100 => a = 9.537e-21
Injection 4 / 32: t0 = 1269091164.36 [2020-03-24 13:19:06.357850+00:00 UTC], f = 811.0 Hz & tau = 0.1 s (Q = 509.57), snr = 500 => a = 4.7685e-20
Injection 5 / 32: t0 = 1269091179.36 [2020-03-24 13:19:21.357850+00:00 UTC], f = 409.0 Hz & tau = 0.1 s (Q = 256.98), snr = 20 => a = 1.043e-21
Injection 6 / 32: t0 = 1269091194.36 [2020-03-24 13:19:36.357850+00:00 UTC], f = 409.0 Hz & tau = 0.1 s (Q = 256.98), snr = 50 => a = 2.6075e-21
Injection 7 / 32: t0 = 1269091209.36 [2020-03-24 13:19:51.357850+00:00 UTC], f = 409.0 Hz & tau = 0.1 s (Q = 256.98), snr = 100 => a = 5.215e-21
Injection 8 / 32: t0 = 1269091224.36 [2020-03-24 13:20:06.357850+00:00 UTC], f = 409.0 Hz & tau = 0.1 s (Q = 256.98), snr = 500 => a = 2.6075e-20
Injection 9 / 32: t0 = 1269091239.36 [2020-03-24 13:20:21.357850+00:00 UTC], f = 211.0 Hz & tau = 0.1 s (Q = 132.58), snr = 20 => a = 6.888e-22
Injection 10 / 32: t0 = 1269091254.36 [2020-03-24 13:20:36.357850+00:00 UTC], f = 211.0 Hz & tau = 0.1 s (Q = 132.58), snr = 50 => a = 1.722e-21
Injection 11 / 32: t0 = 1269091269.36 [2020-03-24 13:20:51.357850+00:00 UTC], f = 211.0 Hz & tau = 0.1 s (Q = 132.58), snr = 100 => a = 3.444e-21
Injection 12 / 32: t0 = 1269091284.36 [2020-03-24 13:21:06.357850+00:00 UTC], f = 211.0 Hz & tau = 0.1 s (Q = 132.58), snr = 500 => a = 1.722e-20
Injection 13 / 32: t0 = 1269091299.36 [2020-03-24 13:21:21.357850+00:00 UTC], f = 129.0 Hz & tau = 0.1 s (Q = 81.05), snr = 20 => a = 6.022e-22
Injection 14 / 32: t0 = 1269091314.36 [2020-03-24 13:21:36.357850+00:00 UTC], f = 129.0 Hz & tau = 0.1 s (Q = 81.05), snr = 50 => a = 1.5055e-21
Injection 15 / 32: t0 = 1269091329.36 [2020-03-24 13:21:51.357850+00:00 UTC], f = 129.0 Hz & tau = 0.1 s (Q = 81.05), snr = 100 => a = 3.011e-21
Injection 16 / 32: t0 = 1269091344.36 [2020-03-24 13:22:06.357850+00:00 UTC], f = 129.0 Hz & tau = 0.1 s (Q = 81.05), snr = 500 => a = 1.5055e-20
Injection 17 / 32: t0 = 1269091359.36 [2020-03-24 13:22:21.357850+00:00 UTC], f = 73.0 Hz & tau = 0.1 s (Q = 45.87), snr = 20 => a = 6.438e-22
Injection 18 / 32: t0 = 1269091374.36 [2020-03-24 13:22:36.357850+00:00 UTC], f = 73.0 Hz & tau = 0.1 s (Q = 45.87), snr = 50 => a = 1.6095e-21
Injection 19 / 32: t0 = 1269091389.36 [2020-03-24 13:22:51.357850+00:00 UTC], f = 73.0 Hz & tau = 0.1 s (Q = 45.87), snr = 100 => a = 3.219e-21
Injection 20 / 32: t0 = 1269091404.36 [2020-03-24 13:23:06.357850+00:00 UTC], f = 73.0 Hz & tau = 0.1 s (Q = 45.87), snr = 500 => a = 1.6095e-20
Injection 21 / 32: t0 = 1269091419.36 [2020-03-24 13:23:21.357850+00:00 UTC], f = 47.0 Hz & tau = 0.1 s (Q = 29.53), snr = 20 => a = 1.3204e-21
Injection 22 / 32: t0 = 1269091434.36 [2020-03-24 13:23:36.357850+00:00 UTC], f = 47.0 Hz & tau = 0.1 s (Q = 29.53), snr = 50 => a = 3.301e-21
Injection 23 / 32: t0 = 1269091449.36 [2020-03-24 13:23:51.357850+00:00 UTC], f = 47.0 Hz & tau = 0.1 s (Q = 29.53), snr = 100 => a = 6.602e-21
Injection 24 / 32: t0 = 1269091464.36 [2020-03-24 13:24:06.357850+00:00 UTC], f = 47.0 Hz & tau = 0.1 s (Q = 29.53), snr = 500 => a = 3.301e-20
Injection 25 / 32: t0 = 1269091479.36 [2020-03-24 13:24:21.357850+00:00 UTC], f = 31.0 Hz & tau = 0.1 s (Q = 19.48), snr = 20 => a = 2.328e-21
Injection 26 / 32: t0 = 1269091494.36 [2020-03-24 13:24:36.357850+00:00 UTC], f = 31.0 Hz & tau = 0.1 s (Q = 19.48), snr = 50 => a = 5.82e-21
Injection 27 / 32: t0 = 1269091509.36 [2020-03-24 13:24:51.357850+00:00 UTC], f = 31.0 Hz & tau = 0.1 s (Q = 19.48), snr = 100 => a = 1.164e-20
Injection 28 / 32: t0 = 1269091524.36 [2020-03-24 13:25:06.357850+00:00 UTC], f = 31.0 Hz & tau = 0.1 s (Q = 19.48), snr = 500 => a = 5.82e-20
Injection 29 / 32: t0 = 1269091539.36 [2020-03-24 13:25:21.357850+00:00 UTC], f = 19.0 Hz & tau = 0.1 s (Q = 11.94), snr = 20 => a = 1.3694e-20
Injection 30 / 32: t0 = 1269091554.36 [2020-03-24 13:25:36.357850+00:00 UTC], f = 19.0 Hz & tau = 0.1 s (Q = 11.94), snr = 50 => a = 3.4235e-20
Injection 31 / 32: t0 = 1269091569.36 [2020-03-24 13:25:51.357850+00:00 UTC], f = 19.0 Hz & tau = 0.1 s (Q = 11.94), snr = 100 => a = 6.847e-20
Injection 32 / 32: t0 = 1269091584.36 [2020-03-24 13:26:06.357850+00:00 UTC], f = 19.0 Hz & tau = 0.1 s (Q = 11.94), snr = 500 => a = 3.4235e-19

Images attached to this comment