Tower Gain Matching ... Continues with LEDs

The eta dependence of towers using the latest HV (set q*_lG). The gold bars are the ideal slope values taken from select subsectors using original HV (set q*) in pp running.


The data is the same as above (HV set q*_lG) but the gold bars are the ideal slope values taken from select subsectors using original HV (set q*) in Au-D running.


SUMMARY of LED calibration follows:

The source scans from a select group of subsectors are used to define the desired eta dependence for all towers. The application of the isolated tower technique with the Au-D data gives the ideal eta dependence in terms of slopes. The inverse slope of the tower spectrum is proportional to the gain of the corresponding PMT. Therefore:

gainA/gainB = slopeB/slopeA

The LED spectrum gives a gaussian peak in the tower spectrum whose centroid is also proportional to the gain. Therefore:

gainA/gainB = centA/centB

It is possible then to use the LED's to finish the calibration started with the Au-D data. Specifically:

ideal gain/gainE = slopeE/ideal slope = ideal cent/centE

The ratio of the slopes are known and the centoids using q*_E HV files give centE for each tube. From this data the ideal centroid for each tube can be deduced. The HV is adjusted until the ideal centroid is achieved.
  1. The ratio (slopeE/ideal slope) for each tower/tube ( ratio )
  2. The goal centroid for sectors 5&6 for each tower/tube ( ideal centroid )
  3. The goal centroid for sectors 7&8 for each tower/tube ( ideal centroid )

    used HV file used data file output gains &
    new HV
    plot of
    gain factors (R)
    q#_E Apr18_led/run01006 for sec 5/6
    Apr21_led/run01012 for sec 7/8
    ledE_56.dat
    ledE_78.dat
    sec05 , sec06 ,
    sec07 , sec08 ,
    q#_lF Apr21_led/run01014 for sec 5/6
    Apr21_led/run01013 for sec 7/8
    ledF_56.dat
    ledF_78.dat
    sec05 , sec06 ,
    sec07 , sec08 ,
    q#_lG Apr21_led/run01015 for sec 5/6
    Apr21_led/run01016 for sec 7/8
    ledG_56.dat
    ledG_78.dat
    sec05 , sec06 ,
    sec07 , sec08 ,

    The crate settings for LED data for Sectors 5 and 6:
    CRATE RHIC LEDlat LEDamp LED1mask LEDdelay FIFO
    3 10 4 2000 62 70 5
    4 0 4 2000 62 50 5
    5 0 4 2000 0 55 (30for1006) 5

    The crate settings for LED data for Sectors 7 and 8:
    CRATE RHIC LEDlat LEDamp LED1mask LEDdelay FIFO
    3 10 4 2000 0 70 5
    4 35 4 2700 62 20 5
    5 0 4 1800 62 55 5

    13 towers that failed this calibration process.
    tower reason
    05TC12 no LED spectra -- marked bad tube in DB
    05TD04 no LED spectra -- marked bad tube in DB
    Why does slope in first setE.dat files looks fine?
    05TD10 no LED spectra -- marked bad tube in DB
    06TA07 marked at bad tube in DB but showed LED response
    as well as slope in first setE.dat file. HV set
    by calculation from first setE slope.
    06TA12 no LED spectra -- marked bad tube in DB
    06TB04 no LED spectra -- marked bad tube in DB
    07TB09 no LED spectra -- WAS FINE in Au-D running
    07TD08 no LED spectra -- marked bad tube in DB
    08TB02 Good LED spectra but marked bad tube in DB
    and shows bad slope in both setE files.
    08TB07 Good LED spectra and reasonable slope in
    first setE.dat file. HV set using slope from
    first setE.dat file.
    08TD04 Good LED spectra but shows bad slopes in each
    setE.dat files but not marked as bad tube in DB
    08TD08 Good LED spectra but not responsive to HV change.
    Set back to HV in original source q files.
    08TE12 Good LED spectra and ok slope in first setE.dat
    file. Bad slope in second setE.dat files. Used
    first slope to set HV.