Data%20Quality%20Checks%20of%20CERN%20Testbeam%20Data%20--Run%20Classification%20and%20Trigger%20Analysis - PowerPoint PPT Presentation

View by Category
About This Presentation
Title:

Data%20Quality%20Checks%20of%20CERN%20Testbeam%20Data%20--Run%20Classification%20and%20Trigger%20Analysis

Description:

Most of the run have a majority of events with Cherenkov Bit On; ... Mu Scinitllator could used to clean off the Minimal Ionization peak as well as ... – PowerPoint PPT presentation

Number of Views:13
Avg rating:3.0/5.0
Slides: 42
Provided by: Admini331
Learn more at: http://nicadd.niu.edu
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Data%20Quality%20Checks%20of%20CERN%20Testbeam%20Data%20--Run%20Classification%20and%20Trigger%20Analysis


1
Data Quality Checks of CERN Testbeam
Data --Run Classification and Trigger Analysis
  • Manqi

2
Outline
  • Reconstructed data file classification Runs with
    different type of Noise and list of clean runs
  • Trigger Analysis Cherenkov counter behavior
    trigger bit based events classification and bump
    events analysis

3
CERN 2006 test beam set up
Beam direction
4
Data Classification Introduction
  • By using Grid, I scan over totally 91 Runs in
    /grid/calice/tb-cern/rec/rec_v0402
  • Run without beam 300090, 300095, 300106, 300107,
    300351, 300355, 300356, 300362, 300367-300370,
    300372, 300376 310069 (garbage run with
    problem in Ecal slow control), 310070
  • Run with less than 1500 beam events 300093,
    300094, 300097, 300098, 300100, 310049, 310057,
    310058, 310061, 310066-310068

5
Run 300085, 50 GeV electron. Noisy layer 1, 2, 14
Noise 1 Mip
6
Example Longitudinal Profile in hit for event
10000
Events with many hits but low energy ?
corresponding to Noisy layer
7
Noisy layers observed in hadron Run 300363, 80
GeV pion
8
Run 310065 with large Noise in Layer 19, 30GeV
electron
Noise 10 Mip
Noise 1 Mip
Taken in 07/09, No special information about this
run mentioned in e-Log
9
Run 300381, 15 GeV electron. Bump emerge at
2500Mip
Noise 1 Mip while layer actived sometime
10
Bump in energy spectrum observed in many Runs
Run 300384, 45 GeV electron
Also seen in 300094, 300105, 300350, 300371,
300380, 310050, 310051, 310059,
310062 Generally, the Bump is located at 2/3 ?
4/5 of the Peak Position
11
Summary of Noise behavior
  • Electron run with Noise 1 Mip 300085, 300094,
    300105, 300354, 300371, 300380, 310050, 310051,
    310059, 310062, 300381 (most of those run have
    noise at layer 1, 2, 14)
  • Hadron run with Noise 1 Mip 300363
  • Electron run with Noise 10 Mips 310065
  • Runs with Bump Events 300350, 300352, 300353,
    300364, 300365, 310064, 300377-300379, 300384

12
Brief Classification of current Reconstructed Runs
Beam Type Electron (10-80 GeV) Hadron Mixed
Clean Runs (39 Runs) 300087-300089, 300091, 300096, 300099, 300101, 300104, 300366, 300383, 300373-300375, 310060 310046-310048, 310054-310056, 310059, 310060, 310063, 310064 300358, 300360, 300361, 300382 300084, 300086, 300676, 300092, 300670 -300676
Runs with large Low energy tail (3 Runs) 300102, 300103, 310052 (might caused by hardon component in the beam)
Runs with Noisy Layer (12 Runs) 300085, 300105, 310065 300354 (with many bi-particle events), 300371, 300380, 310050, 310051, 310059, 310062, 300381, 300363
Runs with bump (9 Runs) 300350, 300352, 300353, 300364, 300365, 300377-300379, 300384
13
Trigger System
  • Activity information on the trigger inputs is
    recorded in a FIFO. One word in the FIFO, called
    Main Word, does contain the trigger information
    at the moment when the trigger was fired.
  • The two 1010 scintillators act as the production
    Trigger

14
Cherenkov Counter Behavior
Cherenkov Counter can select out the electron
events with low beam energy ( but failed in high
energy ) Most of the run have a majority of
events with Cherenkov Bit On Cherenkov Counter
was not set up properly for runs in early running
period (Off for 300084-300087, 300097, 300098
Sometimes On for 300088-300092)
15
Trigger information based events classification
Mu Scinitllator could used to clean off the
Minimal Ionization peak as well as some pion
background EventNum(Scint_3by3_On)/
EventNum(Scint_3by3_Off) 6 ? the 33
Scintallators cover over 80 of the beam
particles
16
Back to bump events analysis using trigger
information
Also seen in 300094, 300105, 300350, 300371,
300380, 310050, 310051, 310059,
310062 Generally, the bump is located at 2/3 ?
4/5 of the Peak Position
17
No obvious hadron component in the beam Some
events smear out the electron Peak, with larger
Num of Hits Majority of the bump events active
the 1010 and 33 Scintillators, also the
Cherenkov Seems to have much more narrow peak
for events having this trigger pattern
18
Num of Hits Vs Ecal Energy Comparison
Events with Cherenkov, 33, 1010 Scintillator On
All Other Events
Pump Events 3423 Events in Region energy
(3200,4500), Hits (400,500) Peak Events 70343
Events in Region energy (4500,6200), Hits
(300,500)
19
Longitudinal Profile Comparison for Peak Events
and Bump Events
Longitudinal profile comparison in Hits
Longitudinal profile comparison in Energy
More Hits in Layer 12, 13 for Bump Events like
Noise Actually, Layer 4-18 seems Not too good ?
coherent electronic effect ?
20
Partially square pattern observed on layer 12
21
Summary
  • For Data Classification
  • -- Majority of the runs have good quality.
    (typical plots for clean runs are attached)
  • -- Noise layer with noise vary from 1 Mip ? 10
    Mips observed
  • -- The corresponding reconstructed root file
    had been placed on grid /grid/calice/tst1441/re
    sult
  • For Trigger Analysis
  • -- Cherenkov Counter works well at low energy
    region
  • -- Trigger information can be used for cleaning
  • -- Bump events associate both with trigger and
    ECAL Needs further understanding

22
Thank you
23
Typical Runs with different beam components
  • Typical electron Run 300089
  • Typical hadron Run 300358
  • Mixed Run with very few electron component
    (300386), comparable electron and hadron
    component (300670), and few hadron component
    (300102)

24
Define a parameter, ScMainWord to record the
trigger information
ScMainWord 100Cherenkov2Sc1_3Sc2_38Sc1_104
Sc2_1032Sc1_10016Sc2_100
Example a event which Active the Cherenkov
Counter, and the 2 1010 Scint, with all other
Trigger not actived, will have ScMainWord1001
32016081412010112
25
Typical electron Run 300089, 30 GeV beam
26
A few bi-particle events
mu events will hit the 100100 Scintillator
27
Typical electron Run 300358, 80 GeV beam
28
Concentrated mu peak and a largely smeared plane
Most events active the Cherenkov Counter
29
Mixed Run 300086, hadron with few electron
component
30
Electron stopped in ecal
Electron component of the beam
Minimal Ionization Component
31
Mixed Run with large electron component and many
bi-particle events Run 300670 same pattern
observed also in 300671?300676
32
Bi particle events 2 electrons
Majority of the beam electrons, will stop in
ecal and active the Cherenkov
33
Clean Run with relatively Large Low energy tail
(corresponding to hadron) 300102, 300103
34
Some hadron component in the beam contribute to
the low energy tail Its the same pattern for
run 300103, 310052
35
Bump in energy Spectrum observed for many
electron Run
Run300350 45 GeV electron
36
Same pattern as 300384 observed in data file
300364
ScMainWord115
37
Same as in Run 300384 Obvious Noise at Layer 12
Layer 4 ? 18 seems Not too good
38
Something more in Run300364 beam component
change with time?
39
Bump Event is also more likely to be declared
Dirty by ErrorBit With ErrorBit Value 1024 or
1536
40
  • The Bump pattern also exist in some
  • files that are declared clean here if
  • we looked detailly into the events that
  • active the 33 Scintllators
  • (i,e, with ScMainWord15 or 115)

41
A list of cells that might be dead only checked
on data file 300384
  • Show in the format of (CellID_X,Y,Z) count from
    0
  • (1,8,0) (16,2,0) (10,9,1) (6,3,2) (8,4,2)
    (15,4,2) (16,4,2)
  • (0,2,5) (1,2,5) (13,7,5) (13,8,5) (13,9,5)
    (15,7,6) (17,6,6)
  • (7,0,20) (11,0,20) (0,11,22) (1,6,25) (2,6,25)
    (1,11,25)
  • (2,11,25) (10,5,29)
About PowerShow.com