Minutes of the Trigger Operation Meeting (Aug 30, 2005)
Attendees: Rainer Bartoldus, Kris Chaisanguanthum, Al Eisner, Yury Kolomensky, Selina Li, Brian Lindquist, Minghui Lu, Stephen Sekula, Dong Su.
- PEP/BaBar is doing great this past week. Lumi is about 8x10^33 and background is reasonable.
- Last Saturday, there were two occasions when the DCZ crates had to be power cycled. The first time happened on Saturday owl shift after the master crate was rebooted (for unknown reasons). The second time was in the Saturday swing shift, but the details were not clear from the elog (ODF was paged for this instead). Short-term instruction for the pilot is updated in case situations like this will happen again.
- Today's MD was cancelled, but tomorrow's MD is scheduled to last about 2 shifts (day and swing shifts).
- Last Wednesday, Dan and Selina (with help from Jamie) tested reading the EMT prescale configuration from xtc file, but failed due to problems with vxworks with templates. We also tried to test this at the EMT test stand, but the ROM couldn't detect the TPB boards due to most likely a problem with the TCB. We will be testing this again tomorrow during the MD. Rainer suggested Fang to get involved in this as well.
- The DCZ crates were power cycled due to master crates reboot (see "Operation" above). Work is underway to investigate and find a solution to the digital clock manager in the TSF firmware.
- Approximately 0.01% of word corruptions still remains at GLT-DAQ data. Note that the low level corruptions were seen in the past as well. The GLT has now currently down from 80 microsecond to 10 microsecond in DAQ delay. In principle, we could have done 0 microsecond delay if there were no corruptions.
- Rainer suggested the we could have L3 checks the FCT and GLT comparison, and also to monitor the prescale rate.
- Selina handed over the L3 pager to Rainer, and perhaps he can convince Arik to take the pager.
- The recent node trims seemed to be due to event builder problems rather than L3 tracking.
Data Quality and Monitoring
- Kris pointed out that there were about 4 runs (e.g. run 12228) with SL5 problems in run 1, but L3 didn't see anything wrong. There was a A track efficiency dip while the B track efficiency didn't seem to get affect much (because A track doesn't allow both SL5 and SL6 missing). It looked like there runs had the BLT masked on for SL5. We decided that these runs are still considered to be okay.
- Last week's reprocessed run 26725 was rejected by Trigger because both BLT and L3 tracks see a big deficient in phi efficiency most likely caused by DCH/ROM problem.
- Brian showed some time history plots for the TSF segments, GLT objects, ODF damages, etc. for run 45402 (when DCH tripped) and run 53744 (ZPD frame bit errors). He will submit the L1TFMon codes with the time history plots. And we will ask Jim Hamilton to make a special build with these new plots so that we can test the fastmon tomorrow at IR2. Tomorrow will be Brian's last day working at SLAC.
- There is a special request for triggering on single photon (energy > 500 MeV). See the proposal talk given last Wednesday in the physics meeting and a brief note here.
- We discussed a few options. Al suggested to look at 1Y in old data. We can also look at 1M, but there is a hugh prescale factor in L1 and the another in L3. We can also take a special run triggering on 1E.
- To help study the trigger rates, we will request a trigger rate test during stable beam data-taking to see what the 1E trigger rate is. We decided to replace the 1Z with 1E for this purpose. Rainer will make a configuration alias for a physics L1PASSTHRU run. The plan is to take one L1PASSTHRU run with the existing trigger configuration (which might also be helpful for us in understanding the trigger rate better and how we can better reduce the trigger rate), and then to take another run with 1Z replaced by 1E (without prescale). In fact the last physics L1PASSTHRU run was taken in Run4, and we should take one for Run5 to study the trigger rate.
This page is created by Selina Li (firstname.lastname@example.org); however, you are welcome to edit and/or correct above items, specially for those who attended the meeting. Thanks.