Upload
others
View
13
Download
0
Embed Size (px)
Citation preview
Study of Digital Hadron Calorimeter
Tohru TakeshitaShinshu Univ.March 2006
Linear collider physicsHadron Calorimeter at LCdigital Hadron calorimeter
based on Geant4.6.2
Linear Collider Physics clean but narrow jets final state
TPCECAL
HCAL
e+e- " qq@350GeV
Particle Flow
charged energy
neutral hadron energy
photon energy
energy distrubution : one entry for each energy per event
Energy (GeV)
• Tracker : charged 65% in a jet
• ECAL : photon 25 % in a jet
• HCAL : neutral hadron 10% in a jet
e + e- " WW
ECM = 250GeV
Particle Flow cont.
1cm3
HCAL TrackerECAL
clustring,separation,
andmatching
need fine granularity
in 3-Dim.
digital HCAL
• ~1cm3 granularity for PF in LC
• HCAL of ~100M channels
• need to reduce cost of electronics
• digital read out = ON/OFF at 1cm3 = 1bit read out
• Hadron interaction= EM shower +pure hadronic int.
digital HCAL simulation Test detector : 1m x 1m x 1m of 100 channels3
A layer of 8mm Pb + 2mm scintillator (1cm )2
1m 1m
1m
digital HCAL simulation total hit/energy distribution
cell_hits0 20 40 60 80 100 120 140 160 180 200
0
100
200
300
400
500
600
-!Lead/Sci 4GeV no_cut-cell
Entries 9999Mean 102.3RMS 25.19
/ ndf 2" 181.2 / 44Constant 7.8± 640.4
Mean 0.3± 102.9 Sigma 0.17± 24.46
-!Lead/Sci 4GeV
MeV0 20 40 60 80 100 120 140 160 180 200
0
100
200
300
400
500
-!Lead/Sci 4GeV no_cut-etot
Entries 9999Mean 110.6RMS 30.35
/ ndf 2" 420.9 / 43Constant 6.9± 528.3
Mean 0.3± 110 Sigma 0.23± 28.39
-!Lead/Sci 4GeV
cell_hits0 20 40 60 80 100 120 140 160 180 200
0
100
200
300
400
500
600
700
-!Lead/Sci 4GeV cut<10ns-cell
Entries 9999Mean 85.14RMS 23.82
/ ndf 2" 244.6 / 43Constant 8.4± 678
Mean 0.26± 85.66 Sigma 0.17± 22.96
-!Lead/Sci 4GeV
MeV0 20 40 60 80 100 120 140 160 180 200
0
100
200
300
400
500
600
-!Lead/Sci 4GeV cut<10ns-etot
Entries 9999Mean 103.3RMS 29.86
/ ndf 2" 583.5 / 43Constant 7.4± 558.5
Mean 0.3± 102 Sigma 0.22± 26.55
-!Lead/Sci 4GeV
digital analog
mean=103hitsigma=24hit
mean=110MeVsigma=28MeV
0.1Mip threshold4GeV pions
Gaussian
digital HCAL simulation linearity for pions
analog : perfect
digital : not badlinearity for electrons
digital
0
500
1000
1500
2000
2500
3000
0 20 40 60 80 100
Linarity for pionsHCAL-8pb2sc-mth>0.1mip
digital hitsanalog MeV
digi
tal (
hit)
/ana
log(
MeV
)
E(GeV)
digital HCAL simulation 50GeV pi-
Entries 1461Mean 893.8RMS 191.6
/ ndf 2! 24.14 / 37Constant 1.71± 43.29 Mean 4.1± 953.2 Sigma 4.4± 113.9
Nhit0 200 400 600 800 1000 1200 1400
0
10
20
30
40
50
50GeV pi-Entries 1461Mean 893.8RMS 191.6
/ ndf 2! 24.14 / 37Constant 1.71± 43.29 Mean 4.1± 953.2 Sigma 4.4± 113.9
Lead/Sci 100GeV pi-Entries 6175Mean 1504RMS 410.2
/ ndf 2! 64.4 / 23Constant 4.4± 199.2 Mean 5.9± 1705 Sigma 7.8± 240.9
Nhit0 500 1000 1500 2000 2500
0
20
40
60
80
100
120
140
160
180
200
220
100GeV pi-Entries 6175Mean 1504RMS 410.2
/ ndf 2! 64.4 / 23Constant 4.4± 199.2 Mean 5.9± 1705 Sigma 7.8± 240.9
Lead/Sci
50GeV pi-
Entries 1461Mean 1320RMS 159.8
/ ndf 2! 18.07 / 18Constant 4.0± 106.5
Mean 4.0± 1335 Sigma 3.8± 127.3
MeV800 1000 1200 1400 1600 1800 2000
0
20
40
60
80
100
120
50GeV pi-
Entries 1461Mean 1320RMS 159.8
/ ndf 2! 18.07 / 18Constant 4.0± 106.5
Mean 4.0± 1335 Sigma 3.8± 127.3
Lead/Sci 100GeV pi-
Entries 6175Mean 2624RMS 246.3
/ ndf 2! 63.16 / 54Constant 2.6± 133.7
Mean 3.8± 2645 Sigma 4.0± 223.5
MeV2000 2200 2400 2600 2800 3000 3200 34000
20
40
60
80
100
120
140
100GeV pi-
Entries 6175Mean 2624RMS 246.3
/ ndf 2! 63.16 / 54Constant 2.6± 133.7
Mean 3.8± 2645 Sigma 4.0± 223.5
Lead/Sci
analog
digital
50GeV 100GeVenergy distribution
digital HCAL simulation • analog vs digital
analog
digital
100GeV pion
100GeV
50GeV
0.5mip-th
digital HCAL simulation energy resolution for pions
0
0.2
0.4
0.6
0.8
1
0 0.2 0.4 0.6 0.8 1
energy resolution for pions HCAL-8pb2sc-mth>0.1mip
analog-resolutiondigital resolution =sig/digihits
ener
gy resolution
1/sqrt(E)
digitalanalog
digital HCAL simulation segmentation size
1/sqrt(E)
resolutionlinearity
E(GeV)Eth=2mip Eth=0.5mip
0 0.1 0.2 0.3 0.4 0.5 0.6 0.70
5
10
15
20
25
30
35
Analog
Digital
Semi-Digital
Real Data
)-1/2
(GeVE1/
/E(%)
E!
semi(2bits)-digital HCAL • 0.5MIP < EnergyDeposit ≤ 10MIP = 1 hit• 10MIP < EnergyDeposit ≤ 100MIP = 10 hits• 100MIP < EnergyDeposit= 100 hits
σ/E
[%
]
1/√E [GeV-1]
σ/E
[%
]
1/√E [GeV-1]
resolution resolution
Jet energy resolution by DHCAL
HCAL1cm x 1cm segmentation
e+e- " qq@350GeV
500GeV Digital
Z Digi
perfect PF
hardware approach toward DHCAL
• scintillator strips
• homogeneous in strip direction
• to reduce cost
• long strip <> multiple hits in a strip
• photon sensors in strong magnetic field
• MPPC
1cm MPPCWLSF
scintillator strip DHCAL
MPPC R/O
Single pion
0.1
1
10
100
1000
0 20 40 60 80 100z position ( cm )
Hit
tim
e (
ns
)
pi-
4GeV
緑 : ニュートロンのヒット(by G4Track Information)
Z
シンチでトラックから離れたヒットはニュートロン散乱に起因する。
5ns
Sci vs Gas
Pb/Sci
scintillator strip DHCALneutron sesitive
0.1
1
10
100
1000
104
0 20 40 60 80 100
a pion interaction in HCAL layers
time (ns)
Z layer
charged
neutron
0
50
100
150
200
1 10 100 1000
time-cut for pi-4GeV
E(scinti) (MeV)
time cut (ns)
no cut
total energy measured (MeV)
total hits
Pb8+sci2mm
20%
MPPC developmentsemiconductor pixel photon sensor with Geiger or Limited Geiger Mode
2µm
a pixel E (V/cm)
1 10 10 102 4 6
2µm
summary• Digital Hadron Calorimeter investigated
• single particle responses are tested
• pure digital (1bit) HCAL is not sufficient
• semi-digital (2bit) HCAL seems to give similar to analog CAL
• Hardware trial is underway
• test is needed with analog read out and simulate digital case
digital HCAL simulation threshold effect
Log(threshold)
0.1mip
AnalogAnalog vsvs Digital Energy ResolutionDigital Energy Resolution
GEANT 4 Simulation of SD Detector (5 GeV !+)
-> sum of ECAL and HCAL analog signals - Analog-> number of hits with 10 MeV threshold in HCAL - Digital
GaussianLandau Tails+ path length
DigitalAnalog
E (GeV) Number of Hits
"/mean ~22% "/mean ~19%
US-digitalHCAL (US) : RPC Digital-HCAL, PFAAnalog/Digital Response for HadronsAnalog/Digital Response for Hadrons
Showers in a sampling calorimeters are characterized by their spatialdevelopment in terms of “track length” :
Track Length (T) = sum of tracks of all charged particles in a shower – Analogsampling calorimeters sum energy, Digital sampling calorimeters sum hits
T ! E (particle energy) – what about spread in energy?
NhitEmeas
Total Energy SumTotal Energy Sum –– Tracks + Photons + NeutralsTracks + Photons + NeutralsG4v6.1
2 Gaussian fit :mean = 83 GeV! = 7.8 GeV
Poor hadron showerreconstruction with tubes– more sophisticatedapproach
mean = 87 GeV!= 4.7 GeV
Good hadron showerreconstruction
After tuning, very similar resultsv = 3.8GeV (neutrals) 5 1.2 (c)
5GeV r
e + e- " Z " qq