Progress report of the GLAST ACD Beam Test at CERN (Backsplash study) simulation and analysis

1
Progress report of the GLAST ACD Beam Test at
CERN (Backsplash study) simulation and analysis

• Tsunefumi Mizuno, Hirofumi Mizushima (Hiroshima
  Univ.) and Tuneyoshi Kamae (SLAC)
• October 1st, 2002

• Detector Geometry/ACD tile numbering pp.2-3
• 350GeVProton Calibration and simulation pp.4-6
• Real Data for 200GeV/50GeV e-pp.7-11
• Comparison between data and simulation pp.12-15
• Effect of the gap pp.16-18
• Summary p.19

2
Detector Geometry
ACD 54cm x 24cm x 1cm
Veto Scint 20cm x 20cm x 1 cm with a hole (1cm
diameter)
Calorimeter Absorber 20cm x 20cm thickness and
material is adjustable
Trigger Scint 1cm x 1cm x 1cm
1cm
Beam
45cm
47cm
53cm
Absorber (Pb) 20cm x 20cm x 1 cm with a hole (1cm
diameter)
Calorimeter Gap 20cm x 20cm thickness and
material is adjustable

• Note So far, we have been using G4 simulation
  program developed before the Beam Test. Detector
  Geometry (shown above) is not the same as, but
  similar to, that of experiment.

3
Nomenclature of ACD tile numbering

• There were 8 ACD tiles in the Beam Test. Here,
  the leftmost tile seen from Calorimeter is called
  1st and the rightmost one is called 8th.

Calorimeter
1st tile
8th tile
ACD tiles
Beam
4
Proton (350GeV) Calibration (1)
1st tile (run09) Peak1220 Pedestal149
3rd tile (run06) Peak1180 Pedestal146
2nd tile (run07) Peak690 Pedestal147
4th tile (run05) Peak1150 Pedestal144
5
Proton (350GeV) Calibration (2)
7th tile (run02) Peak730 Pedestal158
5th tile (run04) Peak1090 Pedestal164
6th tile (run03) Peak1630 Pedestal153
8th tile (run01) Peak1250 Pedestal141
6
Energy deposition of 350 GeV Proton
G4 Simulation with 350 GeV Proton

• We assume that the peak position for 350 GeV
  proton corresponds to 1.7 MeV.

7
Real data for 200 GeV e- (1)

• data run45 (2inch Pb without gap)
• background run11

1st tile
3rd tile
background
data
4th tile
2nd tile
8
Real data for 200 GeV e- (2)

• data run45 (2inch Pb without gap)
• background run11

5th tile
7th tile
background
data

• Note In this report, we just subtract background
  run histogram (blue) from backsplash run one
  (black).

6th tile
9
Real data for 200 GeV e- (3)

• data run45 (2inch Pb without gap)
• background run11

Calorimeter
1st tile
8th tile
ACD tiles

• backsplash level differs by a factor of 2
  between tiles

Beam
10
Real data for 50 GeV e- (1)

• data run47 (2inch Pb without gap)
• background run11

1st tile
3rd tile
background
data
4th tile
2nd tile
11
Real data for 50 GeV e- (2)

• data run47 (2inch Pb without gap)
• background run11

5th tile
7th tile
background
data

• Note In this report, we just subtract background
  run (blue histogram) from backsplash run (black
  histogram).

6th tile
12
Comparison between data and simulation 200 GeV
e- (1)
1st tile
3rd tile
data
simulation
2nd tile
4th tile

• Note1 Statistical errors for simulation are
  similar to those of real data, but we do not show
  them for clarity.
• Note2 In blue histograms (simulation), we do not
  take scintillator response into account.

13
Comparison between data and simulaiton 200 GeV
e- (2)
5th tile
7th tile
data
simulation
6th tile

• Simulation well predicts backsplash data for 200
  GeV e-.

14
Comparison between data and simulation 50 GeV e-
(1)
1st tile
3rd tile
data
simulation
2nd tile
4th tile

• Note1 Statistical errors for simulation are
  similar to those of real data, but we do not show
  them for clarity.
• Note2 In blue histograms (simulation), we do not
  take scintillator response into account.

15
Comparison between data and simulaiton 200 GeV
e- (2)
5th tile
7th tile
data
simulation
6th tile

• Simulation well predicts backsplash data for 50
  GeV e-. To solve remaining difference, we may
  need to improve background subtraction and take
  response of scintillators into account.

16
Effect of the gap (1)

• Calorimeter with gap was used for some data and
  this might affect the backsplash as illustrated
  below. To estimate this effect, we run simulation.

• Calorimeter with gap
• (1inch Pb and air gap)x4

• Calorimeter without gap
• (1inch Pb)x4

Excess in backsplash
ACD tiles
ACD tiles
Beam(50GeV/200GeV e-)
Beam (50GeV/200GeV e-)
17
Effect of the gap (2) 200GeV e-
1st tile
7th tile
with gap
without gap

• Gap might affect the backsplash of the 1st
  (outermost) tile by a factor of 2 for 200GeV e-.
  The effect could be negligible for the inner
  tiles.

4th tile
18
Effect of the gap (3) 50GeV e-
1st tile
7th tile
with gap
without gap

• Gap might affect the backsplash of the 1st
  (outermost) tile by a factor of 1-2 for 50GeV e-.
  The effect could be negligible for the inner
  tiles.

4th tile
19
Summary

• We analyzed 50GeV and 200GeV e- data (2inch Pb
  without gap) and compared them with simulation
  predictions.
• The simulation reproduces the data well.
• The effect of the gap of calorimeter might be up
  to by a factor of 2 for the outermost tile, but
  could be negligible for inner ones.
• We will update simulator geometry.
• We may need to improve background subtraction and
  take scintillator response into account.
• We will analyze/simulate data with calorimeter
  gap.