NDOS CommissionMtng 20101213¶
Attendance: Jon, Susan, Andrew, Chad, Kurt, Art, Minerba, Rick T., Bill F., Qiming, Leon, Mark, Nate, Sue, Jarek, AthansAgenda:
- DAQ/Electronics Update & Plans
- DCS Status & Plans
- Shifters' report
- Since last meeting, new DCM firmware was found to greatly reduce data corruption
- (Sue) We were able to run with a lower threshold; no pop/drop problems nor "missing" u-slice
- Trigger/Timing: AD lent us a timing box which provides a TTL pulse synched to an AD signal
- used $A9 to drive signal into FEBs
- found data time to be ~0.7s off from the trigger ($A9) time.
- Ron now understands the problem and did a work-around Friday night to fix the problem (add 1 sec. to the TDU time and truncate sub-second part of the time)
- Unfortunately the Spillserver seems to be falling over; we're getting about 1/20 spills. The fail rate went way up when MB spills were included.
- Mark: suggested we lower the thresholds and stress-test the DAQ
- Andrew: when we dropped the threshold late last week with the new DCM firmware, we saw about 80% CPU usage
- Sue: there is still the timing marker synchronization issue on the DCMs that some a few FEBs are out-of-synch with the rest
- has created a number of receiver apps + 1 sender app and can see messages being sent back and forth. Some minor issues to be worked out, but he is almost ready to put code in to talk to real DCMs
- Andrew pointed out that there is an issue with reading the status bits from the FEBs during normal data taking. This appears to be killing the DAQ. Qiming will investigate.
- Chad: all seems to be ok.
- Friday night there was an interlock failure on rack 2; the smoke alarm was registering a false positive and was disabling the power every couple of minutes.
- Andrew & Mike Matulik investigated and found that during the power down, the smoke detection unit was blown (not clear exactly how that happened). They replace the unit with a spare. We should get a new spare today or tomorrow.
- Jon is concerned that the DCS monitor in the CR never registered a problem. Turns out that the rack protection systems were never wired up. Chad and Athans will work on this this week.
Shifter report: some minor DAQ issues over the weekend, and some over-optimism that we were seeing neutrinos was reported, but otherwise things went ok. Seems like the documentation is improving a bit.