This is an update (second) on the current progress of studying shower shapes in Run11 data. Some detail of how the study is being done (i.e. by looking at high towers in data and comparing that to high towers created by formula) can be found in my previous post. This week, after getting Minuit to work (at some level), I have tried to get shower shape parameters (a's and b's) by calling Minuit with a set of different starting parameters, and storing the fit parameters, along with the chi square values. Some of the plots for these fits are attached.

In general, the fits for the first 4 high towers are very good; however, the fits for the next 4 do not look as good. Prof. Heppelmann suggested, on Friday, that one of the reasons for this could be due to digitization of energy in detectors. Details of this has not been looked at yet. The cell that was being looked at as of now is r3_c3_3 (note the last number is the detector number that runs from 0 to 3). The recent best fit for this gave the following fit parameters: {.80, .30, -0.10, 0.68, .20, 5.86}. The CHI2/ndf for each high towers are:

chi2: 0.837618 

chi2: 2.196859 

chi2: 1.256642 

chi2: 1.403731 

chi2: 8.928674 

chi2: 10.778306 

chi2: 30.431694  

chi2: 32.115559

The plots can be found here.  

This set of a's and b's was used to look at the average shower shape; the hightower data were looked at once (not including cells that are less than two cells closer to the edge). The resulting plot and chi2/ndf can be found here for det=2 and here for det=3.