Versions Compared

Old Version 27

changes.mady.by.user Jan T. Balewski

Saved on

compared with

New Version Current

changes.mady.by.user Jan T. Balewski

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: Corrected links that should have been relative instead of absolute.

...

Common selection, cuts will be applied to W, Z-analysis and not varied

( unless explicitly stated)

  1. Run selection: run list (423 runs) , muDst list (8551 files) , available # of events: input=1.30e6, L2wET=9.65e5
  2. use SL09b muDst after BHT3 'notch' problem was fixed, beginGMT= 2009-03-28 12:34:27, comprising ~80% of recorded data
    • RHIC fill range: F10448 - F10536
    • STAR run range: R10087001 R10086140 - R10103046
      • skip runs w/ badly functioning TPC (~no vertices found) : R10084013, R10088103, R10081024, R10095032, R10085022, R10088060, R10085103, R10087007, R10093049, R10085118
  3. run analysis on muDst's
    1. access TRG,TPC,BTOW,BSMD directly from muDst containers
    2. apply calibrations from DB 'by hand' in analysis code
  4. Trigger bits requirement
      unmigrated-wiki-markup
    1. BHT3 trig ID: 23054\[01\] or small IDs for some runs
    2. Wiki MarkupL2W trig ID: 23060\[01\] or small IDs
    3. L2-bit1 (on line value was ET>13 GeV)
    4. count L2-bit0 (random accept of BHT3) for absolute 'seen' cross section
  5. Reco prim vertex requirements
    1. rank>=0rank>0
    2. | z-vertex | <100 cm
  6. Reco TPC track requirements
    1. prim tracks belong to selected prim vertex
    2. (flag=301 for primary) or (flag=101 for global) component
    3. if primary is valid but its global pointer is NULL skip the primary track, continue with this but keep the rest of the event
    4. nFit/nPoss>0.51
    5. for high energy electron candidate require additional
      1. nFitPoints>15nFitPoints>=15
      2. if track PT>10 GeV/c assume it is 10 GeV/c
    6. do not use TPC dEdX information for SL09b events
  7. BTOW: use ideal gains, actual status tables & pedestals , 'ofl' flavor for all, no time stamp set in the user code
    1. assume tower is dead if status!=1
    2. assume tower ADC=0 if (rawADC-ped)< [3*sigPed or 8 counts]
  8. BSMD use Willi's 2009 calibration , no time stamp set in the user code
    1. use relative gains, DB flavor=Wbose
    2. use status bits from DB , flavor=ofl
    3. use pedestal residua from DB , flavor=oflunmigrated-wiki-markup
    4. assume strip is dead if status not in range \ [1,19]
    5. assume strip ADC=0 if (ZS ADC-pedRes)< 3* sigPed
    6. for the record: must use custom absolute calibration ( ADC--> energy), describe it

...