Lines at integer and nearly integer frequencies
By looking at high resolution spectra we have identified the presence of two families of lines at frequencies below ~300 Hz. One consists of several lines with integer frequency, see examples in Fig. 1 and Fig. 2. These lines are very narrow (by further increasing the FFT length we observe df~1e-4 Hz). A few of those at higher frequency have been already reported in https://wiki.virgo-gw.eu/DetChar/DetcharShift2019_Feb26_Mar05?validation_key=d833012f027a8496d7fc84f3a2d8b439.
The second family consists of lines with frequency slightly smaller than a multiple of 10 Hz (e.g. 159.863 Hz, 169.86 Hz, 179.852 Hz, 199.835 Hz, 209.825 Hz, 219.82 Hz, 279.77 Hz etc.), see e.g. Fig.3. These lines have a width of ~5 mHz and we have found the following approximate empirical relation to predict their frequency (in Hertz): f ~ floor(f)+0.863-k*0.0075, where k=(floor(f)-159)/10.
A similar family, with frequency near integer 5 Hz is also present (154.87Hz, 164.863 Hz, 184.845 Hz, 194.84 Hz etc.).
While most of these lines have relatively small amplitude, they may have an impact on wide parameter space searches of CWs. Understanding their origin (non-linear coupling of other lines?) and reducing their strength would be useful.
Images attached to this report
palomba - 10:07 Tuesday 12 March 2019 (45230)
Print this report
In fact, this family of lines can be explained as a comb with a fundamental frequency of 9.9914 Hz. There were combs with similar frequency (9.999xx Hz) also in O2. The lines have a structure, as can be seen in Fig.1 and Fig.2
Images attached to this comment