Tanner: DQ shift this week. Still having problems with Ross, masses are misbehaving and also having data dropouts. The data is not good. Should check on the power levels at this station. Lost contact with 4850C station yesterday, down for more than 12 hours now (it is still not back). This is probably due to the test blasts going on there this week. Did not see any sign of battery draining prior to the drop out. Should follow up with Jaret - Tanner will do it.
Gary: Noticed that Yates is acting up (Guralp), showing spikes in all 3 components. Got advice to lock and unlock the seismometer, and this cleared it up. The data was probably usable before the fix, but noisier.
Gary: should recenter 300, another Guralp.
Tanner: Also E2000 and WTP should be re-centered.
Gary: WTP has little dirt on the top, this could be the problem.
Gary: the new 2000E location seems to be perfect, don't see many tilt events at all.
Gary: Calibration procedure worked mostly. Intended to post the results, but did not have a chance to do it yet. Results are in the form of figures, shown as the phase and amplitude spectra, measured vs nominal for each sensor. All STS-2 sensors are indistinguishable from the specs, so the calibrations are within the error from the nominal for all STS-2 stations. But there were problems with Guralps: We were running an older version of the software which had a bug when dealing with Guralps. So the calibration did not work for Guralps. This put Ross into an unusable state - this has happened before in other projects, and the fix that sometimes worked was to cycle the power at the site. Could be done from the doghouse, we don't have to open the vault.
The Guralps are at Yates, Ross, Deadwood, and 300 - we don't have the calibration for these. We should calibrate these when we visit next, to avoid stalling other stations.
Vuk: We have only one issue with the array, the Ross, which is not a critical station. Do we need to plan a visit to fix it?
Daniel/Gary: Maybe someone at Homestake could do this for us.
Vuk will follow up on this with Jaret.
Gary: Another issue is that the balers may be filling up. Can we find out how much data is on them? Tanner will try to follow up on this.
Gary: we'll probably be ok for 1.5 years, which means the first (oldest) stations could be filling in the late spring.
Vuk: contacted Caltech about copying frames to the Caltech cluster. Hope to get an answer within a week or so, so that we can start copying the frames to the LIGO cluster.
Tanner: Clock time percentage seems to be recorded differently in the baler and real-time data, sometimes not exactly consistent. Is this a concern?
Gary: don't know. The real-time data does repackaging, so maybe that's the reason. Probably not a significant problem anyway. Baler data is definitive, a direct dump of the data from the machine to memory, so we should default to it.
Tanner: working on consolidating the real-time and online data to fill the gaps. Trying to do consistency checks now, should be close to finalizing it, after which plan to produce the new frames.
Gary: have an old machine that can be tapped into the data stream. Currently not getting real-time data. Need sysadmins to set up the firewall correctly.
Daniel: Caltech real-time data recorded well.