Minutes of the Trigger Operation Meeting (, 2007)
Attendees: Nicolas Arnaud, Debbie Bard, Rainer Bartoldus, Gregory Dubois-Felsmann, Al Eisner, Fang Fang, Kim Hojeong, Rahmat Rahmt, Su Dong, Mark Tibbetts.
- Pager broke over the weekend, waiting for a replacement.
- Record shifts over the weekend!
- There were some ZPD/BLT errors at the start of run 70857 (see here).
- EMT data looks fine so far.
- Threshold study - last Ax configuration change was 18/12/03 so Mark will look at runs after that. There's an L1TEmtTools package the root macros could be put in.
- Talk to Gagen about the track matching. The Mini-style matching is best understood.
- ZPD threshold analysis is on-going.
- Investigation into GLT bug found the GLT online release was not up to date (see here). During the ROD tomorrow it'll be tested, and some cosmics runs taken. We should check whether the other GLT packages are up to date - EMT and DCT definitely are.
- Moose validation of using L3 tracking in reco has been signed off, need more stats for xtc sign-off.
- There used to be a failure mode to avoid losing L3 hbook files if collection failed, but one was lost last week! Had to fill in info from log files, looking into it.
- There was a problem last week when the DCH had a radiation upset, which made the TSF data bad, which went to L3 and caused large deadtime becasue of L3 writing out error messages. Ought to have an alarm in epic for this.
Data Quality and Monitoring
- Run 6 looks good so far - pi0 mass peak has shifted 1MeV to 125MeV.
- Spike in lumi ratios for around run 54780 (as shown in collab meeting talk by Hojeong) was because of new Bhabha constants being loaded (2nd June), which were subsequently removed (24th June). See old hypernews discussion here" and here. Those runs were re-processed, so Rahmat's stripcharts probably didn't pick up the correct files. It would be good to get the stripcharts re-made with the reprocessed data - ask Silvano?
- Still need to account for off-resonance data on the charts.
- PEP still plans to increase trickle frequency at some point. This should be brought to the attention of the DQG so the DQM plots can be prepared for when/if it happens - a non-trivial adaptation.
- Nicolas would like to add more deatime monitoring - break down the deatime into separate contributions rather than just "busy" (L1) and "inhibited" (trickle) to show in the weekly background reports. This will help better understand the jumps in deadtime.
- Also add L1 rate vs lumi plot to weekly background report.
- Information about deadtime due to physics window could be added to the logbook from L3. The L3 physics window is not exactly the same as the reprocessing window, but they could be linked.
This page is created by Debbie Bard (firstname.lastname@example.org); however, you are welcome to edit and/or correct above items, specially for those who attended the meeting. Thanks.