BRAT-2-1-10

From: Peter H. L. Christiansen (pchristi@nbi.dk)
Date: Fri Oct 12 2001 - 07:56:54 EDT

  • Next message: Stephen J. Sanders: "How do I get brag configure to work with pre 2001 cernlib"

    Hi developers
    
    I added the new tpc calibration classes to BRAT that I have been working
    on for the past 2 weeks. Here is a short description of the main classes
    followed by the transcript of the ChangeLog. The whole thing should be
    transparent meaning if you do what you have always done you should not
    feel a thing. There are 2 non-transparent things. 1) New histogram with
    the confidence levels in BrTpcTrackModule. 2) Added BrTpcRdoModule to
    BrTpcTrackPackage so now there is no excuse for not using it. It removes
    all the tables created except the track table and also adds the rdo table!
    
    The whole scheme is inspired/stolen from Djamels tof stuff.
    
    data/calib/BrTpcCalibration :
    This is the base class for the TPC calibrations where the different
    calibrations are defined and getters and setters.
    
    modules/calib/tpc/BrTpcCalModule :
    Base class
    modules/calib/tpc/BrTpcPadStatusCalModule :
    This module is inspired by the JIJ online pad monitor and tries to
    identify non-instrumented, cold and hot pads in the Tpc and remove those.
    It does so by calculating the averag adc per timebin for a given pad,
    average = 0=> non-instrumented, normalise the rest of the pads to the tpc
    average (for non-instrumented) and make cuts (default: <0.2 = cold, > 3 =
    hot).
    
    modules/calib/tpc/BrTpcTimeCalModule :
    This module tries to correct for the time row offsets seen in the data. It
    follows most of the ideas described on :
    http://www.nbi.dk/~pchristi/BRAHMS/TPC/timerow.html
    
    modules/calib/tpc/BrTpcHitWidth :
    It makes a doublegaus fit to the deviations seen in x and y for all tracks
    and all rows. It uses the minmum sigma as sigma. There is no
    specials reason for using a double gaus, except that the data are clearly
    not gausian.
    
    This is the first iteration of the mdoules and they probably all could use
    some fine tuning so feel free to touch and comment. Script for using the
    modules can be found in ~brahms_app/pc_app/tpc/calTPC but uses
    hilux12.nbi.dk as the test database (if you are bright you might be able
    to guess the passwords;). There is no commit stuff implemented yet so the
    calibrations are just ascii files at this time.
    
    I am gone next week but I will try to post some nice pictures before.
    
    Cheers
       Peter
    
    2001-10-12  Peter H Christiansen  <pchristi@rcas0008.rcf.bnl.gov>
    
            * doc/guide/bratguide.tex:
            Added another missing part in how to add a library.
    
            * configure.in:
            Added new directory modules/calib/tpc. Bumberd version to 2.1.10.
    
            * modules/calib/tpc/BrTpcTimeCalModule.h,
    modules/calib/tpc/Include.h, \
    modules/calib/tpc/LinkDef.h, modules/calib/tpc/Makefile.am,
    modules/calib/tpc/B\
    rTpcCalModule.cxx, modules/calib/tpc/BrTpcCalModule.h,
    modules/calib/tpc/BrTpcH\
    itWidthCalModule.cxx, modules/calib/tpc/BrTpcHitWidthCalModule.h,
    modules/calib\
    /tpc/BrTpcPadStatusCalModule.cxx,
    modules/calib/tpc/BrTpcPadStatusCalModule.h, \
    modules/calib/tpc/BrTpcTimeCalModule.cxx, modules/calib/tpc/.cvsignore:
            Added new directory tpc. Added the first calibration modules. They
            have all been tested and found to work. The algorithms might not
    be
            optimal, but are at least fully automatic and to some extent
            documented.  CVS:
    
            * modules/calib/Makefile.am: Added new directory tpc.
    
            * data/params/BrDetectorParamsTPC.cxx,
    data/params/BrDetectorParamsTPC.\
    h:
            Added new variables fHitWidthX and fHitWidthY and setters and
    getters
            plus a new method ListPadStatus() for printing out the status of
    all
            pads in tpc.  CVS:
    
            * packages/BrTpcTrackPackage.cxx, packages/BrTpcTrackPackage.h:
            Added BrTpcRdoModule to the track package.
    
            * data/calib/BrTpcCalibration.cxx, data/calib/BrTpcCalibration.h,
    data/\
    calib/Include.h, data/calib/LinkDef.h, data/calib/Makefile.am:
            Added a BrTpcCalibration base class for the TPC calibrations.
    
            * modules/track/tpc/BrTpcSequencePPModule.cxx:
            Instead of just checking for dead pads the PP module now instead
    checks\
     that
            the pad is active. This makes a difference for simulated data
    only.
    
            * modules/track/tpc/BrTpcTrackModule.cxx,
    modules/track/tpc/BrTpcTrackM\
    odule.h:
            Added ned histogram with confidence levels of best track. I also
            added option to get CL as quality in best track instead of
    chisquared.
            To use you have to call the method SetUseCL(kTRUE).
    
            * modules/track/tpc/BrTpcHitModule.cxx,
    modules/track/tpc/BrTpcHitModul\
    e.h:
            Added option to use the calibrated hit widths from the database.
            You have to do SetUseCalibratedWidths(kTRUE) to and have them
            read in.
    
            * config/config.mk: Added new brat directory modules/calib/tpc
    
    
    -- 
    :-) --------------------------------- )-:
    |Peter H L Christiansen aka PAN @ NBI	|
    |EMAIL  : pchristi@nbi.dk		|
    |OFFICE : Tb1 @ NBI			|
    |PHONE  : 353 25269			|
    |SNAIL  : Hans Tavsensgade 35, 4th	|
    |PHONE  : 35 349336			|
    :-D --------------------------------- \-:
    



    This archive was generated by hypermail 2b30 : Fri Oct 12 2001 - 07:57:40 EDT