Looking at the current glitch rate, it is elevated compared to previous locks in the last days. Figure 1: current glitch rates (Nov 30) Figure 2: Nov 28 glitch rates Note that the current glitch rates are all elevated. It's easier if you open these in windows you can swap back and forth, or just stare at them side-by-side. Looking at the SNR distribution, there is a large population of SNR<30 glitches (note large hump instead of linear decay on the log-log plot) Figure 3: current SNR distribution (Nov 30) Figure 4: Nov 28 SNR distribution Now looking at the histogram of glitch SNR versus frequency, it's clear there are more numerous higher SNR glitches at low frequencies, but the higher glitch rate for low SNR glitches seems to be coming mostly from the (new?) 2 kHz and 3 kHz glitches. Figure 5: current SNR versus frequency (Nov 30) Figure 6: Nov 28 SNR versus frequency
Our violin mode second harmonics are rung up, which is most likely to be the problem here. We had a rough lockloss late monday night in which things got rung up. For now Ed, Jeff B and Jeff K are working on damping some of the rung up first harmonics, that is why we are not in observing mode right now.
The guardian automatically damps the first harmonic violin modes, so they are normally small after we have had some long lock stretches, but the second harmonics will only get damped if operators actively work on them. It would be a good idea for operators to try to watch these and try to damp them as well as we can. Allowing for operators to damp these and change settings while we are in observing mode would facilitate getting these modes damped.
We have been having ISI trips on locklosses recently which is probably how these are getting rung up. We are hoping that the tidal triggering change described in alog 31980 will prevent the trips, so that harmonics will not get rung up as often.