Structure Diagram for ECA: Electronics Calibration
             ==================================================
 
         An Introduction to the Structure Diagrams is available.
        NB  These links will be broken until the HTML Version of the Code is installed.
 

Introduction

Two files are required:-

• ECA_READ.CMD SNOMAN cmd file, which basically defines the input ZDAB files
• ECA_CONTROL.DAT TITLES Bank which sets:-
  • Which of the three calibrations to run
  • Which crates to include in the calibration
  • Tolerances for various parameters
  • DAQ to pe/ns conversion constants
  • Type of summary file to produce
  • Whether to force QSlope fit through previously measured pedestal

Output Files

TITLES files

• PEDESTAL FILES with names of the form PECA_yyyymmdd_vv.DAT
• QSLOPE FILES with names of the form QECA_yyyymmdd_vv.DAT
• TSLOPE FILES with names of the form TECA_yyyymmdd_vv.DAT

SUMMARY HISTOGRAMS/NTUPLES

These files are of the form PDST_yyyymmdd_vv.NTP, QSLP_yyyymmdd_vv.NTP
and TSLP_yyyymmdd_vv.NTP for the 3 different calibrations, respectively.
Each contains a set of "summary histograms" which show the total number of
"good", "suspicious" and "bad" calibrations in each 1/2 crate. "Bad"
values have also been flagged in the titles files and by default will 
not be used in application of the calibration constants. "Suspicious"
values are for diagnostic purposes only and are defined by comparing
a given parameter value with a running average of "typical" values
accumulated during the ECA run. For QSLP, histograms of average
QSlope values are also given for all 3 charges and, for TSLP, a histogram
of average TSlope values (linear fit) is also shown.

If a "long summary" has been requested in ECA_CONTROL.DAT, then a much
more detailed ntuple is also included, containing information on individual
channels. Information has been packing into variables as much as possible
to reduce the size, but these files can still be quite large and may
take a while to interact with via PAW. Nevertheless, it affords one the
opportunity to examine exactly why a given channel may have been failed
by the software. To make it somewhat easier to use the ntuple, first
execute ECA_QTSLOPE.KUMAC which will define the variables card, crate,
chan, cell, qtyp (for choosing QHS, QHL or QLX) and qslope.

For PDST -> the ntuple simply contains the charge and time pedestals
            for each CCCC. Negative values indicate pedestals which
            are considered "suspicious" (the true value = the magnitude).
            There is currently no criterion for establishing a "bad" pedestal
            value, aside from one which is listed as zero.

For TSLP -> the ntuple actually contains the injected (TINP)and 
            digitized (TDIG) values used to obtain the calibration, 
            along with an average linear-fit slope (SLP), for each CCCC. 
            A negative TINP indicates a "suspicious" values and a
            negative TDIG indicates a value deemed to be "bad"
            (remember, only "bad" values are tagged in the actual
            titles file, "suspicious" values are only for diagnostics).
            As before, the true values of TINP and TDIG are given by
            their absolute values.

For QSLP -> this ntuple contains similar information the the TSLP ntuple,
            but the injected and digitized values (QINP and QDIG) are
            given for each of the 3 different charges. This can be selected
            by using qtyp=1 for QHS, qtyp=2 for QHL, or qtyp=3 for QLX.
            The slope must be examined here through the variable qslope
            defined by ECA_QTSLOPE.KUMAC, since the SLPQ parameter in
            the ntuple has been used to pack additional information as well.

Titles Banks Used
-----------------
 
 
Initialisation Routines
-----------------------
 
ECA_INI
 
 
Execution Routines
------------------
 
ECA_ELECTRONICS_CALIB
ECA_T_FN_DIG
ECA_T_RTSAFE 
 
 
Termination Routines
--------------------
 
ECA_TRM