DQ Shifter: Evan Goetz
DQ Shift Mentor: Beverly Berger
LSC Fellow(s): Laurence Datrier

DQ shift wiki page

Summary (highlights) of the shift:

   * LHO cluster problems crashed summary pages Friday and recovery is ongoing. This was due to the large number of files from summary pages/plots and other DetChar tools (~500 million combined). There are some missing or incomplete plots starting Thursday, but many more Friday/Saturday. Omicron and Hveto are incomplete or non-existant from Friday - Sunday.
   * Observing summary: (71.06 + 59.53 + 62.41+ 98.06 + 35.29 + 90.35 + 67.60) / 7 = 69.19%; more or less in line the mean observing time for H1 during O3 thus far.
   * Several earthquakes in the latter part of the week took away from observing time. Also, when H1 was locked or re-locked near earthquake times, the whistle glitches were much more problematic because of the increased ground motion.
   * Whistle glitches remain problematic for the early part of the week [Omicron problems prevented late week Hveto updates] as evidenced by round 1 or 2 channel winner LSC-REFL_A_LF_OUT_DQ. The other ongoing common round 1 or 2 winner remained LSC-POP_A_LF_OUT_DQ.
   * Low frequency narrow spectral artifacts in H1 remain problematic. They are quite dynamic (changing day by day) and very prevalent below 100 Hz compared with L1.
   * The 48 Hz feature is still present with varying amplitude, as reported by previous DQ shifts. I looked through some accelerometer spectra and I am unsure if I see correlation by eye. I will continue to look in more detail.
   * Several other features seem to have cropped up this weekend at the start of lock stretches, but then seem to reduce in amplutide during the course of a lock stretch (~550 Hz, ~850 Hz, possible violin harmonic near 1010 Hz) - see below.
   * PCALX showed some differences in the spectrum [likely not immediately problematic for h(t)], perhaps indicative of an unstable optical follower servo as resetting the offset has helped the stability.
   * Later in the week violin resonances seem more rung up at the start of locks than early in the week.

Note that I'll update the shift summary once some of the backfilled summary pages are updated.