Reports 1-1 of 1 Clear search Modify search
Detector Characterisation (Broadband noise)
mwas - 7:48 Saturday 01 June 2019 (46008) Print this report
Additional broadband noises in h(t) compared to April

Figure 1. Shows the h(t) spectrum at 3 different times. On April 21 in blue, when the BNS range was at 50Mpc, on May 26 in red (before the earthquake), and on May 31 in yellow after the TCS/ISC work on the ITF working point. Between April 21 and May 26 the noise between 50Hz and 100Hz has increased. After the earthquake the shot noise has increased and several structures at a few hundred Hz have increased.

Figure 2 Shows the h(t) spectrum on April 21 in blue and on May 31 in red, adding to the Apr 21 data 1/f noise (yellow line) and a f noise (shot noise, in purple), yields the green curve which matches quite well the May 31 data. So it seems that since Apr 21 the flat noise has not changed, however an additoinal 1/f noise is present (and was present already before the May 26 earthquake) and the shot noise increased after the earthquake. The shot noise increase might be cured by realigning the OMC and/or squeezing alignment and phase tuning. The origin of the 1/f is to be found.

Figure 3. Compares the B1_PD1+B1_PD2 spectrum, and the B1_PD1-B1_PD2 (NULL) spectrum. The NULL spectrum has not changed between Apr 21 and May 31, so the new 1/f noise is not coming from an issue with the B1 photodiodes, as it is a noise that is common to both photodiodes.

 

Images attached to this report
Comments to this report:
mwas - 15:41 Friday 14 June 2019 (46119) Print this report

Figure 1 shows the spectrum of h(t) for each week starting from April 10, selecting each time the 30min with the best average BNS range (using harmonic average to select out range drops). The range was the best in the first two weeks, the noise between 70Hz and 140Hz has increased after that, being the worst in the last two weeks.

Figure 2 and Figure 5, shows the same but with a higher resolution PSD, it is hard to draw a conclusion as the figures become crowded.

Figure 3 compares April 10 to June 2, and shows a large increased noise that can be fitted with f additional 1/F noise and increased shot noise

Figure 4 compares April 10 to May 25, and shows that the increased noise contributions were smaller

Figure 6. However, if one looks one week later, for June 9, adding a flat noise instead of a 1/F noise is a better fit to explain the increased noise.

 

Code for these plots can be found in /users/mwas/detchar/hCompare_20190614

Images attached to this comment
Search Help
×

Warning

×