Minutes of the Trigger Operation Meeting (19th June, 2007)
Attendees: Debbie Bard, Rainer Bartoldus, Al Eisner,Kim Hojeong, Rahmat Rahmat, Mark Tibbetts.
- On June 15th there was a synch request from DCZ that wasn't fixed by starting a new run. The odf log file shows the MPIC spurious interrupt messages - a power cycle of the TSF crates solved the problem as usual.
- A curiosity: This same thing happened on 15th March and 15th April, and hasn't happened any other time (15th May was a ROD and Fang was working on the GLT libraries. If anything happened then we wouldn't have noticed it). Is there something that happens on the 15th of every month that maybe disrupts the timing signal from PEP? Patches installed, backups taken, reboot of something? Could be a coincidence but would be interesting to find out.
- Tomorrow is the monthly trigger status report at the 3.45 meeting. Please have a look at the slides here and let Debbie know if there are any comments.
- TPB #22 was taken to Lupe to re-attach the chips that were knocked off. These are involved in the GLT input/output circuit which explains the errors seen when the board was tested.
- Still some MC validation pending for the L3 tracking study. Arik is unavailable for a few weeks still so maybe someone could take over the task?
- It would be a good idea to increase the size of background collections. The how of it needs a bit of study, but using the cyclic 10 trigger is the obvious choice.
Data Quality and Monitoring
- Run 6 data looks fine so far for trigger.
- Rahmat will make a link on the trigger data quality webpage to the abbreviated DQG plots which show just the trigger plots.
- He will also attempt to run the stripcharts at Padova himself, to see whether the problem with the lines connecting the points in the trigger plots is local to Padova setup or is something else.
Trickle injection monitoring
- Meeting last week to discuss the reverse trickle window project. There are two choices - either process trickle window events parasitically or run over those events seperately. The second option is the most straightforward (if one can perhaps define a new tag bit to be set independent of recoBGFilter) but computing resources are limited. It's also not clear whether the trickle window events can be processed parasitically without getting into the monitoring.
- Also the question of whether damaged events get into the monitoring or whether they get filtered out beforehand. It's not clear.
- Errata - see hypernews posting.
- SP9 MC is here, but the lack of pressure to implement a tightened trigger configuration has meant the study has languished of late. It should be resuscitated!
- Discussion regarding the 1E trigger line - taking a new run with the line is not much work but does not seem worth it until we can take a run with much higher luminosity. There is no clear position on whether we want to add a line that could add ~8% to trigger rate. This may be discussed further in Friday's Trigger/Filter/Lumi AWG meeting.
This page is created by Debbie Bard (email@example.com); however, you are welcome to edit and/or correct above items, specially for those who attended the meeting. Thanks.