ArmenVartapetianTileCalMonitoringDataQualityHistograms

Introduction

Below is presented a general list of monitoring/data quality histograms which can be used with the final setup of TileCal. For near term use (cosmic runs) we can tailor a smaller set of histograms including limited modules and channels. If cpu resources are there and the monitoring system performance is not compromised, pretty much all the histograms can be used for the final monitoring and data quality evaluation purposes. Some of the proposed distributions, particularly the ones over phi and eta, are carrying duplicated information. The solution could be to go with eta distributions while we have limited number of modules, then switch mainly to phi when full angular coverage is in place (the main advantage of phi is that the distributions at the collision mode are expected to be flat - easy to monitor).

Histograms For General Use

1. E_total spectrum
2. Etmiss spectrum (only for final setup with full angular coverage)
3. E_total spectrum per module
4. E_total per module, phi distribution for CB and EB modules
5. E spectrum per tower (grouped in phi)
6. E flux per tower & sampling (on the same plot), vs. phi and/or eta
7. Relative energy in samplings within tower (on the same plot), vs. phi 
       and/or eta   
8. E spectrum per channel
9. Correlation/asymmetry of the channels within the cell: (L-R)/(L+R) 
   (to limit the non-correlated contribution of the noise, require signal 
    E > E_min in either of 2 channels)  
  9a. Distribution of the correlation value for each cell   
  9b. Mean value distribution vs. cells (within each module)   
  9c. RMS value distribution vs. cells (within each module)
10. Number of channels N_ch (per module) with E_ch > E_threshold (pick a 
       sensitive E_threshold), phi distribution for CB and EB modules 
11. Eta-Phi plot of energy flux in towers (color defined scale)
12. Eta-Phi plot of energy flux in cells, for each sampling/depth (color 
       defined scale)
13. Timing of the signal
  13a. Timing distribution, for each channel   
  13b. Mean value distribution vs. channels   
  13c. RMS value distribution vs. channels  
14. Pedestal 
  14a. Pedestal distribution, for each channel   
  14b. Mean value distribution vs. channels   
  14c. RMS value distribution vs. channels
15. Noise value over time
  15a. Mean value per channel vs. time   
  15b. RMS value per channel vs. time   

Comments on Monitoring Strategy

As you have probably noticed the histogram list is arranged in the order of decreasing size of the monitored object - from detector, modules, towers to cells and channels. The idea is that one can monitor much easier and more often the small amount of histograms showing the general status of the detector and the modules. Presumably the significant/big scale problems in the majority of cases will show up in those histograms, letting us take immediate actions. Monitoring and running data quality algorithms on the numerous histograms at the level of channels, though being even more important, inevitably will take more time and resources to perform, and will be carried out with relatively less frequency.

Dedicated Histograms For Cosmic Runs

For the cosmic runs, in addition to the list of general histograms, it is extremely useful to have a set of dedicated histograms, which will dynamically monitor the efficiency of the detector to identify cosmic muons.

16. Using tower with maximum energy as evidence for cosmic muon   
  16a. E_max_tower spectrum
  16b. Number of towers (per event) with E_tower > E_threshold (pick 
         a sensitive E_threshold)  
  16c. Eta and/or phi distribution of the towers with maximum energy 
  16d. Eta-phi plot of the towers with maximum energy 
17. Using muon identification algorithm  
  17a. E_muon spectrum
  17b. Number of identified muons per event
  17c. Eta and/or phi distribution of identified muons 
  17d. Eta-phi plot of identified muons 

During the cosmic runs, if the detector systems are in a "healthy" state, essentially the content of the histograms 16(a-d) and 17(a-d) must show similar features, and the histograms 17(a-d) must show a distinct evidence of efficient muon identification.