Status Report of the ROD Testing at BNL

1
Status Report of the ROD Testing at BNL

• Activity update of the ROD system at BNL
• DAQ-1 with trigger controller
• Data corruption testing
• calibration etc.

• Kin Yip

2
DAQ at BNL, too
1

• Started from the setup of L. Monetas, and with
  additional help from M. Caprini, K. Serguei and
  R. Jones
• Quite a bit of work to install and set up in a
  non-CERN software environment ( slackware etc.)
• I have also tackled some synchronization problem
  which may hang the system occassionally, with
  help from K. Serguei et al.
• NEW(!) Trigger Controller, according to M.
  Caprini, the first controller to be used for
  calibration and trigger control.
• Demonstrated to people in Atlas week at BNL and
  US Atlas review

3
Two controllers in two different crates
4
(No Transcript)
5
Controlling trigger rate
6
Issues related to data integrity
2

• SDRAM After a couple newer versions of OC FPGA
  firmware from Daniel, we can now write to 8
  Mbytes without data corruption
• EOE flag doesnt seem to come up at trigger rate
  gt11.1 kHz
• VME read errors at high trigger rate are still
  there

• A recent and different kind of data corruption
  (eg. the error bits are wrong)
• If rebooted, the data read from the FEB through
  the optical link seemed to be OK for a little
  while and then went crazy again
• This probably didnt happen so regularly before
  in the last several months

7
Issues related to data integrity (cont.)
2

• There is a jumper in the GLink Receiver card
  which be used to control the phase between the
  data and clock
• But adjusting the jumper position didnt solve
  the problem
• After a lot of detective work, the problem is
  found to be due to the fact that the temperature
  at the back of the VME crate is too high
• Temperature gt 60o C ( with poor A/C in the lab.)
  for the Glink chip
• There didnt seem to have this problem in the
  winter
• GLink chips have malfunctioned such that the
  phase between the data and clock would shift ?
  corrupted data (including wrong error codes)
• Solution
• I put a small fan blowing air right towards the
  Glink receiver card
• Temperature drops and the phase doesnt change
  any more, which one can see even in a scope !

8
Pulse from the trigger analyzer board
Thermometer
9
Calibration and miscellaneous
3

• Regularly using a calibration board (LAPP) to
  inject signals to FEB ? ROD, as part of the data
  chain
• See the plot
• Making a small board to veto trigger on BUSY
• Progress in (DSP) calibration codes in SMU, Texas
• Installed TI's code composer studio v1.20
  (Windows) but couldnt produce an executable to
  run DSP correctly
• Technology (software) transferred to Jose M.
  Castelo to set up the ROD for TileCal

10
Pulse from the trigger analyzer board
Data seen from the ROD using 1 calibration channel
11
Plans

• Further optimize the DAQ system and extend the
  trigger controller with more functionalities (?)
• To be really used for doing calibration with the
  read-out chain of the BNL LAr electronics mock-up
• Hope to have and use more sophisticated
  calibration codes in the DSP from SMU soon