To the right is shown a plot of TDC value versus E-counter ID(1-384). This a typical distribution for uncalibrated tagger data. This histogram was used to determine the offsets needed by each E-counter to align them all in time.

Here is shown the same distribution after calibration. This was generated using only some of the Pb-glassII data. This looks as expected now allowing for a narrow cut to easily be placed around the band.

Here is a plot made from events at the begining of the Pb-glassII data set. The calibration used for the above plot does make a smooth band. But now, there is a second band emerging and an obvious slope as you go from low to high photon energies (left to right). It's important to remember that all of this data came from one run which basically swept the beam through the whole calorimeter before eventually centering it on the center of the array for the last part of the run. This seems to indicate two things: The calorimeter was determining the trigger time for at least some of the data rather than the Master OR from the tagger. The appearent timing correlation with photon energy probably means significant timewalk on the calorimeter sum signal. Assuming these both are true, then the ADC gate would have been moving around with respect to the signals in time. This would mean any tail fraction outside of the ADC gate would not be a constant and would therefore, need to be corrected for if possible.