CTA report 152:Comparison of photon detection performance
of photomultiplier tube and silicon photomultiplier for Medium Sized Telescopes
Anatolii Zenin, Hiroyasu Tajima, Akira Okumura, Yuki Nakamura
ISEE, Nagoya University
OutlineIntroduction
• CTA and Cherenkov imaging• Motivation• PMT and SiPM• Medium-sized telescope
2
Main Part
• Setup and analysis technique• Measurement results
• Discussion
CTA and atmospheric Cherenkov imaging
3
Cherenkov Photons
Electromagnetic Cascade
Very-high-energyGamma Ray(20 GeV –300 TeV)
LST:20 GeV ~ 150 GeV
SST:5 TeV ~ 300 TeV
MST:150 GeV ~ 5 TeV
K. Bernlöhr
MotivationAdvantages of SiPM:
• Background-tolerant, no degradation over time
• Stable quantum efficiency and photon detection efficiency
• Can be operated with 60V
• Lightweight, compact
4
PMT and SiPMTo ensure that the performance of SiPM cameras will not be inferior to the current PMT cameras, we have to compare:
•QE / PDE
•Sensitivity to signal
•Sensitivity to noise
•High NSB (moonlight) tolerance and performance under high NSB conditions
•Overall telescope performance (the end goal of this work)
5
PMT and SiPM sensitivity comparison (not to scale)
200 300 400 500 600 700wavelength, nm
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
arb.
uni
t
100 GeV γ Cherenkov spectrum
La PalmaNSB spectrum
CTA PMT QECTA SiPM PDE
Medium-sized telescope
6
Cutoff angle 26˚≤αc≤28˚
(technical requirement)
ROBAST simulation of MST optical system
αc
Measurement setup
7
PMT on a rotation stage
Readout module (TARGET) on a rotation stage
8
365 nm
Analysis technique
Crosstalk does not affect 0pe peaks in SiPM
PMT charge spectrum sample
h_spec1Entries 39900Mean 15.31Std Dev 18.79
40− 20− 0 20 40 60 80 mV)×Integral Value (ns
0
100
200
300
400
500
600
700
800
Entri
es
h_spec1Entries 39900Mean 15.31Std Dev 18.79
200− 0 200 400 600 800 1000 1200 1400 1600 ns)×Charge (mV
1
10
210
310
410 Signal + dark count
Only dark count
noise
SiPM charge spectrum sample
P (k) =�k
k!e��
<latexit sha1_base64="oeIIB7kHc47mJxUDmVkUPZ7sMuE=">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</latexit><latexit sha1_base64="oeIIB7kHc47mJxUDmVkUPZ7sMuE=">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</latexit><latexit sha1_base64="oeIIB7kHc47mJxUDmVkUPZ7sMuE=">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</latexit><latexit sha1_base64="oeIIB7kHc47mJxUDmVkUPZ7sMuE=">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</latexit>
� = � lnP (0)<latexit sha1_base64="zWnPk8B8mEXbQb2nXx/pjxwn5fk=">AAAB/nicdVDLSsNAFJ3UV62vqLhyM1iEujAkpdB2IRTduKxgH9CEMplM2qGTSZiZCCUU/BU3LhRx63e482+cphVU9MDA4ZxzuXeOnzAqlW1/GIWV1bX1jeJmaWt7Z3fP3D/oyjgVmHRwzGLR95EkjHLSUVQx0k8EQZHPSM+fXM393h0Rksb8Vk0T4kVoxGlIMVJaGppHLtPhAMELeA5dxrN2xT6bDc2ybTWbjWq1BnNSc5wFceoN6Fh2jjJYoj00390gxmlEuMIMSTlw7ER5GRKKYkZmJTeVJEF4gkZkoClHEZFelp8/g6daCWAYC/24grn6fSJDkZTTyNfJCKmx/O3Nxb+8QarChpdRnqSKcLxYFKYMqhjOu4ABFQQrNtUEYUH1rRCPkUBY6cZKuoSvn8L/SbdqObbl3NTKrctlHUVwDE5ABTigDlrgGrRBB2CQgQfwBJ6Ne+PReDFeF9GCsZw5BD9gvH0CUKSUbw==</latexit><latexit sha1_base64="zWnPk8B8mEXbQb2nXx/pjxwn5fk=">AAAB/nicdVDLSsNAFJ3UV62vqLhyM1iEujAkpdB2IRTduKxgH9CEMplM2qGTSZiZCCUU/BU3LhRx63e482+cphVU9MDA4ZxzuXeOnzAqlW1/GIWV1bX1jeJmaWt7Z3fP3D/oyjgVmHRwzGLR95EkjHLSUVQx0k8EQZHPSM+fXM393h0Rksb8Vk0T4kVoxGlIMVJaGppHLtPhAMELeA5dxrN2xT6bDc2ybTWbjWq1BnNSc5wFceoN6Fh2jjJYoj00390gxmlEuMIMSTlw7ER5GRKKYkZmJTeVJEF4gkZkoClHEZFelp8/g6daCWAYC/24grn6fSJDkZTTyNfJCKmx/O3Nxb+8QarChpdRnqSKcLxYFKYMqhjOu4ABFQQrNtUEYUH1rRCPkUBY6cZKuoSvn8L/SbdqObbl3NTKrctlHUVwDE5ABTigDlrgGrRBB2CQgQfwBJ6Ne+PReDFeF9GCsZw5BD9gvH0CUKSUbw==</latexit><latexit sha1_base64="zWnPk8B8mEXbQb2nXx/pjxwn5fk=">AAAB/nicdVDLSsNAFJ3UV62vqLhyM1iEujAkpdB2IRTduKxgH9CEMplM2qGTSZiZCCUU/BU3LhRx63e482+cphVU9MDA4ZxzuXeOnzAqlW1/GIWV1bX1jeJmaWt7Z3fP3D/oyjgVmHRwzGLR95EkjHLSUVQx0k8EQZHPSM+fXM393h0Rksb8Vk0T4kVoxGlIMVJaGppHLtPhAMELeA5dxrN2xT6bDc2ybTWbjWq1BnNSc5wFceoN6Fh2jjJYoj00390gxmlEuMIMSTlw7ER5GRKKYkZmJTeVJEF4gkZkoClHEZFelp8/g6daCWAYC/24grn6fSJDkZTTyNfJCKmx/O3Nxb+8QarChpdRnqSKcLxYFKYMqhjOu4ABFQQrNtUEYUH1rRCPkUBY6cZKuoSvn8L/SbdqObbl3NTKrctlHUVwDE5ABTigDlrgGrRBB2CQgQfwBJ6Ne+PReDFeF9GCsZw5BD9gvH0CUKSUbw==</latexit><latexit sha1_base64="zWnPk8B8mEXbQb2nXx/pjxwn5fk=">AAAB/nicdVDLSsNAFJ3UV62vqLhyM1iEujAkpdB2IRTduKxgH9CEMplM2qGTSZiZCCUU/BU3LhRx63e482+cphVU9MDA4ZxzuXeOnzAqlW1/GIWV1bX1jeJmaWt7Z3fP3D/oyjgVmHRwzGLR95EkjHLSUVQx0k8EQZHPSM+fXM393h0Rksb8Vk0T4kVoxGlIMVJaGppHLtPhAMELeA5dxrN2xT6bDc2ybTWbjWq1BnNSc5wFceoN6Fh2jjJYoj00390gxmlEuMIMSTlw7ER5GRKKYkZmJTeVJEF4gkZkoClHEZFelp8/g6daCWAYC/24grn6fSJDkZTTyNfJCKmx/O3Nxb+8QarChpdRnqSKcLxYFKYMqhjOu4ABFQQrNtUEYUH1rRCPkUBY6cZKuoSvn8L/SbdqObbl3NTKrctlHUVwDE5ABTigDlrgGrRBB2CQgQfwBJ6Ne+PReDFeF9GCsZw5BD9gvH0CUKSUbw==</latexit>
Poisson distribution:
Can extract mean value from pedestal:
9
365 nm
Photoelectron distribution in SiPM
0 1 2 3 4 5 6 7 8x
0
1
2
3
4
5
6
7
8y
0
200
400
600
800
1000
1200
1400
phot
on c
ount
0.0027222 0.00747551 0.0469161 0.09266 0.0896353 0.0360592 0.00623342 0.000213355
0.0471951 0.0805923 0.0508806 0.0415147 0.0400246 0.0471579 0.0808369 0.0388827
0.0817675 0.0739368 0.0170783 0.0146918 0.0229714 0.02124 0.0693322 0.0787186
0.0509001 0.0145181 0.0176383 0.0133915 0.01318 0.0485025 0.0478063
0.0667145 0.0487292 0.0168159 0.011876 0.0162426 0.0156332 0.0485464 0.0633529
0.080206 0.0730602 0.0395963 0.0324879 0.0133267 0.034744 0.077408 0.0757741
0.0181476 0.0732873 0.0366851 0.0603641 0.0581927 0.0318337 0.0642246 0.013782
0.00291888 -0.00117109 0.0395916 0.0910576 0.0869906 0.0381313 -0.000498826 0.00173501
0 1 2 3 4 5 6 7 8x
0
1
2
3
4
5
6
7
8y
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
mea
n PE
SiPM photo
SiPM geometry
PE distribution in data
Simulated photon distribution
0 5 10 15 20 25 30 35 40 angle, deg.θ
0
0.5
1
1.5
2
2.5
3λ PMT simulationPMT experimentSiPM simulationSiPM expermiment
465 nm
10
Measurement resultsResults for azimuthal angle φ = 0
(see the figure on the left)
cutoffangle
PreliminaryOur simulation can not reproduce absolute values, so it is normalized to fit on-axis SiPM
data
There are still some issues with the measurement setup and analysis
It is likely that consideration in terms of only geometrical optics is not good enough to
simulate SiPM behavior properly
0 5 10 15 20 25 30 35 40 angle, deg.θ
0
0.5
1
1.5
2
2.5
3
3.5λ PMT simulationPMT experimentSiPM simulationSiPM expermiment
375 nm
Preliminary
cutoffangle
Discussion
11
• Although the results are not final, our measurement technique is usable and only needs a little polishing
• There are still some difficulties with the setup, but that will be solved in a short time
• Relative performance is not consistent with the simulation, likely due to interference effects in Si3N4 and SiO2 layers and on silicon surface
• Cutoff angle for SiPM configuration has to be checked
To do
12
Several things have to be done before making any solid conclusions:
• Design a new light concentrator to remove the gaps and optimize the cutoff angle (in progress)
• Compare the performance of PMT and SiPM by a simple ray-tracing simulation(after designing a new light concentrator)
• Perform a full telescope simulation with CORSIKA and sim_telarray and study how NSB affects new cameras(the final step)