Minutes of the Trigger Operation Meeting (Sept 20, 2005)
Attendees: Rainer Bartoldus, Kris Chaisanguanthum, Selina Li, Dong Su
- We are reminded of the safety review which will happen during Oct.3-11. Please review the talk for Integrated Safety Management System (ISMS) at SLAC.
- We discussed possible October shutdown activities. It looks like the only potential work for TRG is the TSFfirmware upgrade to solve the problem associated with the special power-up sequence. We need to find out where the man power is going to come from (most likely UK).
- With help from Jamie, Dan was finally able to successfully read the xtc prescale configuration at the test stand. He is now ready to test the EMT at IR2. Today there was an opportunity to test this at IR2, but Dan was busy with commitment. The plan is to test this with 3 cosmics runs with different prescale factors (ie. 1, 2, 2000, etc).
- Rainer mentioned that he hasn't seen Dan's code yet. It's preferable to have either Rainer or Andy to look over the related configuration code as soon as possible before submitting. We need a SRT release for this code.
- Su Dong reminds us that OPR needs to look at the bit for the prescale.
- The old DCT crates were powered off last week. See HN posted.
- We discussed things we will need to do when we turn the old DCT crates back on for data-taking again after October's down. Here is the 3 items:
- modify the scripts for ps archive maker and html for fastmon with JAS. Kris will make a tag with a version of L1TFMon for running with the DCT;
- include the old DCT crates in the alarm handler again;
- change the default crate mask (if running with DCT for at least a week or a month).
- Matt Weaver pointed out to us that the TSF ticks appeared to be off in TRG pattern calibration. It sounded like the sequence pulsing was too late, and the hits come out at the edge of the timing windowm. The suggestions were to either (1) change the TSF DAQ window (as Jamie suggested, but Su Dong doesn't think it's a good idea since it would have to be maintained separately for calibration and colliding-beam data); or (2) tweak L1 Accept delay which is part of the calibration setup that wouldn't affect colliding-beam data.
- There are some TSF frame bit errors seen in TSF-Y2 and TSF-Y3 in run 58115. See plots here. The location of these boards corresponds to the GLINK error seen in the DCH side. See HN posted and its follow-up postings.
- The number of TSF segments suddently increased in run 58117 as shown in the time history plots, which correlates very well with the DCH current plot, as a result of beam-gas background.
- OPR checks on the 2 L1passthru runs take two weekends ago.
- We got paged last Wednesday (Sept.14) owl shift due to L3 node crash. It turned out that one of the farm node died and cannot reconfigured to take data due to the minimum number (29) of available nodes being reached. OEP fixed this problem by reduce the limit for this to be 20.
- We talked about combining the Friday L3 meeting with the operation meeting, and could keep the L3 meeting time block if extensive discussions needed.
Data Quality and Monitoring
- We need to work on getting the entire run's time range in the time history plots. Rainer will take a look at this to see if he can fix this.
- The .ps archive wasn't updated earlier last week due to some soft links with the /afs/ files that are not allowed for references from IR2. This problem was fixed. Note that our trigger files are mounted to /nfs/ disks.
- We went over the list of our triggger monitoring tasks. Most items in the list are done.
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.