Minutes of the Trigger Operation Meeting (July 26, 2005)
Attendees: Rainer Bartoldus, Yaw Ming Chia, Al Eisner, Fang Fang, Larry Gladney, Walter Innes, Selina Li, Brian Lindquist, Dong Su, Dan Walker.
- Dan said that the time scale for getting the EMT prescale working is this week, and should be ready to test at IR2 next week if opportunity arises.
- TSF SL9 had a dip in layer 35 wire 5 (see plots in for DCH hits and TSF segments in that region) last week. But the dip has been gone on its own since swing shift on Monday July 25.
- The old TSF-X8 had mismatches in occupancy plots in run 56505, as reported by the DQM shifter.
- Ming is performing some preliminary testing of the power-cycle problem with the TSF crates in the test stand. He has got some instruction for this from Manchester. He worked out a schedule to share the test stand with the DCH group who is testing the trigger as part of their testing for the new electronics.
- There is a discussion for the schedule of turning-off the old DCT system. Some concerns for keeping the old DCT running parasitically are:
Advantages for having both the new and old DCT on:
- contribution to DAQ deadtime (~0.1% caused by the old DCT?)
- maintenance problem.
- extra redundant DCT data being written to disk.
- ROMs that could have been used as spares.
Walt would like to see the old DCT goes away by Aug.1 as promised; however given the first two reasons in the advantage list above, we all agreed that it makes sense to keep it till after the new DCH electronics installed. One possible scenario is to leave the old DCT out of configuration for now, and then turn it back on in November.
- The new DCH frontend electrionics will be installed in October, so running with both the new and old DCT will give the DCH group (and ODF group) a confirmation or confidence for the front-end electronics. This would help avoid a lot of confusion and also reduce the time for debugging right after a hardware change like this.
- The TSF firmware might also be updated after a solution is found for the power-cycle problem. If that's the case, then it would be beneficial to have the old DCT system to cross check with the new DCZ. Ming mentioned that the MC test doesn't catch some problems that would only appear when running the real system at IR2.
- There has been one case of the new TSF had out of frame error while the old TSF didn't. We still didn't fully understand why it happened.
- One of the TSF-i board in the test stand is a prototype board. Two spare boards at IR2, and no spares at the test stand. It seems like the Glink ready LED was not working properly at the test stand.
- Su Dong is happier with the current version than the last one. It should be ready to be tested at IR2 early next week.
- Rainer is working on value reset from the PCA server that sees large rate during the begin run transition.
- trickle injection plots were discussed in HN. It will be interesting to look at histogram for the frequency of injection.
- The possibility of lossing markers is not verified yet for injection trickle. Perhaps we need a way to monitor the markers?
Data Quality and Monitoring
- Rainer reproduced the L3 archieve plots.
- We need someone to keep the livefastmon for L3 up to date.
This page is created by Selina Li (email@example.com); however, you are welcome to edit and/or correct above items, specially for those who attended the meeting. Thanks.