The development and performance evaluation of a hybrid photo- - - PowerPoint PPT Presentation

▶
the development and performance
SMART_READER_LITE
LIVE PREVIEW

The development and performance evaluation of a hybrid photo- - - PowerPoint PPT Presentation

Jun 01, 2023 440 likes •889 views

The development and performance evaluation of a hybrid photo- detector for Hyper-Kamiokande M. Jiang, Y. Nishimura A , Y. Suda B , Y. Okajima C , R. Akutsu A , D. Fukuda D , S. Hirota, M. Yokoyama B , M. Nakahata A , M. Shiozawa A , Y. Hayato A ,

slide-1
SLIDE 1

The development and performance evaluation of a hybrid photo- detector for Hyper-Kamiokande

  • M. Jiang, Y. NishimuraA, Y. SudaB, Y. OkajimaC, R. AkutsuA, D. FukudaD,
  • S. Hirota, M. YokoyamaB, M. NakahataA, M. ShiozawaA, Y. HayatoA,
  • S. NakayamaA, H. TanakaA, T. Nakaya, A. Minamino, A.K. Ichikawa,
  • M. KuzeC, A. TaketaE, Y. KawaiF, T. OhmuraF, M. SuzukiF

Kyoto.U, ICRRA, U.TokyoB, Tokyo TechC, Okayama.UD, ERIE, HPKF

2015/7/7 Photo Det 2015 @ Moscow 1

slide-2
SLIDE 2

Hyper Kamiokande experiment & PD candidate

  • A proposed next generation

underground water Cherenkov detector

  • 1Mt Water tank (Super K×25)
  • 100, 000 PDs (Super K ×9)
  • Three candidates of 50cm PD

2015/7/7 Photo Det 2015 @ Moscow

  • Performance is

known well

  • Stable
  • Better energy

resolution

  • Better time

resolution

  • Simpler

structure

  • Much better

resolution

High performance

2

Venetian Blind PMT (Super Kamiokande (Super-K) PMT) (Hamamatsu R3600 NQE) Box & Line PMT (B&L PMT) (Hamamatsu R12860 HQE) Hybrid Photo-detector (HPD) (Hamamatsu R12850 HQE)

slide-3
SLIDE 3

Principle of HPD

  • Using an Avalanche Diode (AD)

instead of dynode to multiply photoelectron

  • Simpler structure

→ Cost reduction

  • Fast and low fluctuation process of

electron convergence and multiplication → Good timing resolution

  • High first stage gain

→Good 1 p.e. detection capability

2015/7/7

ｅ

~8kV

Avalanche Diode (AD)

× ~1600 ~300V × ~200 AD gain Bombardment gain ΔV ~ -300V

Amp

Typical values in an HPD prototype

Total gain ~105 ×Gain of Amp

Vacuum Tube

Photo Det 2015 @ Moscow 3

slide-4
SLIDE 4

Outline

  • Introduction
  • Development status
  • HPD system
  • Performance Evaluation of 50cm HPD with 5mm AD
  • Development of 50cm HPD with 20mm AD
  • Development status and issues due to large area AD
  • Some solutions proposed
  • Multi-ch AD
  • New preamplifier with decoupling transformer
  • Summary

2015/7/7 Photo Det 2015 @ Moscow 4

slide-5
SLIDE 5

Introduction

2015/7/7 Photo Det 2015 @ Moscow 5

slide-6
SLIDE 6

Development of 50cm HPD

20cm HPD (w/ 5mm AD) 50cm High QE HPD (w/ 20mm AD)

  • Operate properly
  • Performance has been

evaluated fully

  • Testing in a 200t tank
  • QE: ~22% @ Maximum (~400nm)
  • Target design.
  • High noise level due to large AD stray

Capacitance

  • Some solution has been

proposed

2015/7/7 Photo Det 2015 @ Moscow

  • Operate properly
  • Performance has been evaluated

50cm High QE HPD (w/ 5mmAD)

Theme of today’s talk

Collection Efficiency ~ 40% CAD ~ 60pF QE: ~30% @ Maximum(~400nm) Collection Efficiency ~ 90% CAD ~ 800pF QE: ~30% @ Maximum (~400nm)

6

slide-7
SLIDE 7

HPD System

2015/7/7 Photo Det 2015 @ Moscow

  • Equipped with
  • a preamplifier
  • a HV power supply.

Low pass filter to reduce the noise from HV

surge protection

7

slide-8
SLIDE 8

Evaluation of 50cm HPD with 5mm ΦAD

2015/7/7 Photo Det 2015 @ Moscow 8

slide-9
SLIDE 9

Pulse shape and 1p.e charge response

2015/7/7 Photo Det 2015 @ Moscow

50 cm HPD w/5mm AD 50 cm Box&Line PMT (B&L PMT) 50 cm Super-K PMT

The performance of HPD is mainly determined by preamplifier

Valley Peak

9

Pulse Shape 1PE Charge Distribution In the 3 candidates of PDs, HPD (w/ 5mm AD) has

  • fast response
  • the best 1p.e. resolution in charge distribution

Condition

HV:8kV AD Bias V:293V AD:5mm

50 HPD w/ 5mm AD (w/ preamp) Super-K PMT Rise Time 7.4 10.6 Fall Time 11.5 13.1 Pulse width (FWHM) 17.1 18.5 1pe resolution (σ/μ) 16% 53% Peak / Valley Ratio 4.0 2.2

HPD B&L PMT Super-K PMT

slide-10
SLIDE 10

Time resolution

2015/7/7 Photo Det 2015 @ Moscow

Better time resolution for higher luminosity

  • 1p.e. ：1.0ns (σ) 3.6ns(FWHM)
  • High luminosity：～120ps(σ) ~400ps(FWHM)

Time resolution of 50cm HPD w/5mm AD is better than that of Super-K PMT (Hamamatsu R3600) and Box & Line PMT After the time walk correction

10

σ FWHM

Super-K PMT Super-K PMT

Condition

HV:8kV AD Bias V:293V AD:5mm

slide-11
SLIDE 11

Linearity

2015/7/7 Photo Det 2015 @ Moscow

Charge (pC) p.e. We confirmed the linearity is kept in 6% from 1p.e. to 140 p.e. It is limited by the maximum

  • utput of preamplifier

(～800mV)

The expected output according to gain calculated from 1 p.e

The intrinsic linearity of HPD is better than that

  • f PMT because it is harder to saturate for AD

than for dynode.

11

  • We use two laser diodes to

evaluate the linearity of the

  • utput
  • We fixed the output luminosity
  • f A and change the one of B

from 1 p.e to several hundred p.e. by to check the linearity

  • f wide region

Number of Photon

Laser Diode A (Fixed) Laser Diode B (Variable) Laser Diode A+B lighting at the same time

The expected output is O(A+B)=O(A)+O(B)

A Bi A+Bi O(A) O(B) O(A+B) Output Condition

HV:8kV AD Bias V:293V AD:5mm

slide-12
SLIDE 12

Temperature dependency

2015/7/7 Photo Det 2015 @ Moscow

Gain

50 cm HPD (w/preamp) 20 cm HPD (w/preamp)

Temperature coefficient @20℃

  • 2.1%/℃

AD Bias 253V

HK Temperature :13℃

  • Both of them use an AD with a diameter of 5mm,

their temperature coefficient also has a similar value.

  • The temperature dependency of HPD is stronger for higher Bias voltage.
  • In Hyper-K, surrounding water temperature is expected be 13℃ and stable.

12

Condition

HV:8kV AD:5mm

Temperature coefficient @20℃

16 14 12 10 6 8

Temperature℃

AD Bias V 339V 300V 280V

HK Temperature :13℃

  • 3.1%/℃
  • 2.2%/℃
  • 2.4%/℃

×106

Gain

Temperature℃

slide-13
SLIDE 13

Development of 50cm HPD with 20mmΦ AD

50cm HPD with 5-ch 20mm AD New preamplifier with decoupling transformer (tested in a 15mm AD)

2015/7/7 Photo Det 2015 @ Moscow 13

slide-14
SLIDE 14

Development status of 50cm HPD with large AD

2015/7/7

HV module:

  • In: 5V~10V
  • Out: 8kV

Preamplifier Avalanche Diode(AD) Photocathode

  • Large photocathode

→Large AD →Large Stray Capacitance →Low S/N

  • Some solution
  • Multi-ch AD (Segmented AD)
  • Improvement on Preamp
  • Decoupling Transformer
  • Bootstrap type
  • Low capacitance AD

Photo Det 2015 @ Moscow

(50cm HPD)

14

slide-15
SLIDE 15

50cm HPD with 5ch 20mmΦ AD

2015/7/7 Photo Det 2015 @ Moscow 15

slide-16
SLIDE 16

Multi-ch 20mmΦ AD

  • A smaller area leads a smaller noise
  • The area of each channel is same.
  • For the 5ch AD, the noise level is

expected to be the same as √5 × 8.9mm AD ~ 44% of 20mm AD 5ch AD Ch1 Ch5 Ch4 Ch3 Ch2

2015/7/7 Photo Det 2015 @ Moscow 16

slide-17
SLIDE 17

5-ch 20mm AD of HPD

  • The performance of 5-ch 20mm AD is tested

in a 50cm HPD

  • For the readout electronics, a sum amp is

used to combine the output signal of each channel

2015/7/7

50ns 5mV

Photo Det 2015 @ Moscow

1pe Signal Output of 5ch AD

filter filter

I-V inverted amplifier

Sum amp （+8kV） （+8kV - VAD）

5-ch AD

HV Power

50V/mA

protection I-V I-V

Signal out

VAD

filter

… …

Ch1 Ch2 Ch3

…

17

Condition

HV:8.4kV AD Bias V: 415V AD: 5ch 20mm BW: 20MHz

slide-18
SLIDE 18

Pulse shape and 1p.e charge response

2015/7/7 Photo Det 2015 @ Moscow

  • 4 0

4 8 12 16 20 24 28

pC Pedestal 1pe 2pe 3pe

Time (ns)

Voltage (mV) Entries

18

5.57mV

Condition

HV:8.4kV AD Bias V: 415V AD:5-ch 20mm

HPD (w/ 5ch 20mm AD) has

  • a slow response (even comparing with Super-K PMT)
  • better 1 p.e. resolution than the Super-K PMT, but lower than that of 50cm HPD w/ 5mm AD

50cm HPD w/ 5ch 20mm AD 50 cm HPD w/ 5mm AD Super-K PMT Rise Time (ns) 11.7 7.4 10.6 Fall Time (ns) 18.9 11.5 13.1 Pulse width (FWHM)(ns) 32.3 17.1 18.5 1pe resolution (σ/μ) 28% 16% 53% Peak / Valley Ratio 3.08 4.0 2.2

slide-19
SLIDE 19

New amplifier with decoupling transformer

Tested on a 50cm HPD w/ 15mmΦ AD

2015/7/7 Photo Det 2015 @ Moscow 19

slide-20
SLIDE 20

New amplifier with decoupling transformer

  • A usual capacitor coupling for deriving a signal from AD at 8 kV brings

a large noise due to a large AD stray capacitance. As a result, another coupling transformer is added to suppress the noise by reducing the capacitance seen by the amplifier.

  • The prototype circuit of the

amplifier with decoupling transformer has been made and tested on HPD with 15mm AD.

2015/7/7 Photo Det 2015 @ Moscow

Circuit w/o trans Circuit w/ trans

In this study, the impedance ratio

  • f the transformer is 1:9

20

NEW!

slide-21
SLIDE 21

Performance in a 50cm HPD

2015/7/7

Distribution noise signal (μ) P/V 1p.e. resolution w/ trans Charge 0.31pC (σ) 1.63pC 4.90 26.9% Pulse Height 0.92mV (μ) 3.54mV 11.31 29.5% w/o trans Charge 0.55pC (σ) 2.62pC 1.38 49.1% Height 0.69mV (μ) 4.47mV 2.59 64.1%

Photo Det 2015 @ Moscow

Charge Distribution Pulse Height Distribution w/o trans w/ trans

20ns

5mV

Decoupling transformer can improve the performance

  • Faster response and better 1p.e. resolution

21

w/o trans w/ trans w/o trans w/ trans

  • 2 -1 0 1 2

3 4 5 6

pC

  • 2 0 2 4 6 8 10

mV

Condition

HV:11kV AD Bias V:415V AD:15mm BW: 20MHz

slide-22
SLIDE 22

Pulse shape and response in different HV

2015/7/7 Photo Det 2015 @ Moscow

50cm HPD w/ 15mm AD and transformer 50 cm HPD w/ 5mm AD Super-K PMT Rise Time 14.3 (8kV) 13.5(11kV) 7.4 10.6 Fall Time 16.7(8kV) 15.7(11kV) 11.5 13.1 Pulse width (FWHM) 29.5(8kV) 28.8(11kV) 17.1 18.5 1pe resolution (σ/μ) 51.7%(8kV) 26.9%(11kV) 16% 53% Peak / Valley Ratio 1.08 (8kV) 4.90(11kV) 4.0 2.2

HV: 11kV HV:8kV HV: 11kV HV:8kV

The performance of HPD @ 8kV with the new amplifier is not satisfactory. Combination use of the 5-ch AD and the transformer coupling is expected to give a sufficient performance.

Normalized Pulse Height

22

Charge Pulse shape

  • 4 -2 0 2 4 6 8

0 2 4 6 8 10 12 14 16 18 10ns p.e

Condition

HV:8kV & 11kV AD Bias V:415V AD:15mm

slide-23
SLIDE 23

Transit time spread of the 1p.e. signal

  • The transit time spread (TTS) of the 1p.e signal has been evaluated in

different HV.

  • Distribution is fitted with Exponential Modified Gaussian
  • The TTS decreased with the HV increased until 10kV.
  • It is limited by noise and the performance of preamplifier.
  • It might be improved the redesign of preamplifier

2015/7/7

Transit time distribution @8kV ns(a.u.) 130 135 140 145 150 155 Transit time spread (σ) in different HV σ(ns) HV (V) 1.9 1.8 1.7 1.6 1.5 1.4 1.3 Thres @ 0.5p.e.

Photo Det 2015 @ Moscow

Super-K PMT: 1.4ns

23

Condition

AD Bias V:415V AD:15mm

slide-24
SLIDE 24

Dark count rate

2015/7/7 Photo Det 2015 @ Moscow

pC 2 4 6 8

  • 2

V Dark Count Dark Count Integrated charge of 1p.e. @11kV Pulse height of 1p.e. @11kV

HV 8kV 9kV 10kV 11kV Dark Rate 8.33kHz 13.1kHz 13.8kHz 47.3kHz

  • We recorded the waveform in a long gate (50μs × 3000) and count the

number of pulses.

  • Cut on pulse height and integrated charge is used to select the dark count

from dark noise (including dark count and electric noise) Plan to use HPD with HV of 8kV

24

Condition

AD Bias V: 415V AD:15mm

slide-25
SLIDE 25

Future plan

  • Preamplifier
  • Finalize the preamplifier design by autumn, 2015
  • Water proof housing
  • Will be designed and tested in autumn, 2015
  • External HV supply (in 2016)
  • The HV is powered and controlled by voltages under 10V now.
  • Test of HV cable
  • Test of HV connecter
  • Our goal is to finish the development of 50cm HPD in 2015
  • A test in a 200-ton water tank would start within a year.

2015/7/7 Photo Det 2015 @ Moscow 25

slide-26
SLIDE 26

Summary

  • The 50cm HPD has been being developed for Hyper Kamiokande
  • The prototype of 50cm HPD with 5 mm AD can operate properly.
  • It shows better performance than Super-K PMT and B&L PMT
  • Better time resolution, better charge resolution
  • But 20 mm AD is desired in order to get a full efficiency
  • The prototype of 50cm HPD with 20mm AD has been produced
  • S/N ratio is low due to large stray capacitance
  • Some solution has been proposed to deal with the noise
  • Multi-ch AD
  • New preamplifier with decoupling transformer
  • After applying these solution, the S/N has been improved,

although high frequency noise still remains.

  • HPD will be ready in 2016

2015/7/7 Photo Det 2015 @ Moscow 26

slide-27
SLIDE 27

Back up

2015/7/7 Photo Det 2015 @ Moscow 27

slide-28
SLIDE 28

High frequency noise

HV:11kV LV:415V Bandwidth: full

2015/7/7 Photo Det 2015 @ Moscow

The high frequency noise still remains after the decoupling transformer. 50.0 ns/div 200 μs 500 ns 20mV 20mV 50.0 μs/div

28

slide-29
SLIDE 29

The effect of the signal cable with a length of 70m

Pulse height Noise 1p.e. signal P/V 4m 2.20mV 4.37mV 1.70 4m+70m 1.13mV 2.69mV 1.72

4m cable 4m+70m cable

HV:11kV LV:415V Bandwidth: full

Charge Noise 1p.e. signal P/V 4m 0.28pC 1.08pC 1.60 4m+70m 0.20pC 0.89pC 2.49 Noise=

n pedestal

n

) (

min

Ｖ The minimum voltage in a gate

Noise =

)) ( ( pedestal Q  Charge distribution of pedestal

The S/N ratio improved after a 70m signal cable: the noise level decreased.

2015/7/7 Photo Det 2015 @ Moscow

A co-axis signal cable with a length of can be used as a low pass filter

Pulse height Integrated charge in 60ns gate Integrated charge in 100ns gate

20ns 2mV

29

slide-30
SLIDE 30

Response of 1p.e. signal in different HV

HV 8kV 9kV 10kV 11kV SK PMT HK requirement Pulse Height 2.05mV 2.52mV 2.83mV 3.54mV 4mV

  • Noise

(Pulse Height) 0.63mV 0.71mV 0.91mV 0.92mV

  • 1p.e. resolution

(pulse height) 68.3% 41.2% 31.7% 29.5%

  • Charge

0.83pC 1.15pC 1.36pC 1.62pC 2.2pC

  • Noise

(Charge) 0.28pC 0.31pC 0.34pC 0.31pC

  • 1p.e. resolution

(charge) 51.7% 35.7% 29.8% 26.9% 53% <50% Dark Rate (@0.5p.e.) 8.33kHz 13.1kHz 13.8kHz 47.3kHz 4kHz

  • TTS (σ)

1.91ns 1.70ns 1.36ns 1.34ns 2.2ns 2.2ns P/V (Charge) 1.08 1.98 3.92 4.9 2.2 1.9

2015/7/7 Photo Det 2015 @ Moscow

The performance, except dark count rate, is improved with the increase of HV. Some issue still remains: low gain, high overshoot and high frequency noise.

LV: 415V BW: 20MHz AD:15mmΦ

30

slide-31
SLIDE 31

Principle of HPD

  • The photon-electrons produced by the

photon cathode is accelerated by the high voltage in the vacuum tube then avalanche multiplied in the avalanche diode (AD).

  • Advantage
  • The structure of AD is simpler than that of

dynode therefore cost reduced

  • Good timing resolution because of the fast

and low fluctuation process of electron convergence and multiplication

  • Good single p.e. detection capability because
  • f the high first stage gain.
  • Disadvantage
  • The HV between the photon cathode and the

AD is high (~8kV), therefore a new high voltage system are needed to be developed

  • A preamplifier is needed due to a low total

gain

2015/7/7

ｅ

~8kV

Avalanche Diode (AD)

× ~1600 ~300V × ~200 AD gain Bombardment gain ΔV ~ -300V

Amp

Typical values in 20 cm HPD prototype

Total gain ~105 ×Gain of Amp

Vacuum Tube

６０ns

Signal of HPD

日本物理学会 2015年春季大会 早稲田大学 Photo Det 2015 @ Moscow 31

slide-32
SLIDE 32
  • We measured the output of 20mm

AD with same incident light and Bias Voltage.

  • The amplitude of noise is reduced
  • For the pulse height
  • Although the height of signal

becomes about 82%, S/N is larger than before by 4 times.

  • For the integrated charge
  • The charge of noise is reduced

to 73.4%

  • The output charge of signal

increased to 1.5 times.

Evaluation on AD

2015/7/7

Pedestal Signal (High luminosity) w/o trans w/trans

AD Bias:400V BW: full

V V Charge distribution

Impedance of trans Primary (Input): 9μH Secondary (Output): 81μH BW :0.3MHz~500MHz

w/o trans w/ trans

Pulse height distribution

pC 0 2 4 6

  • 2

noise signal noise signal (100ns) w/ trans 8.1mV 346mV 0.69pC 11.1pC w/o trans 38.4mV 419mV 0.94pC 7.0pC Pulse height distribution

Strong Overshoot for AD!

Photo Det 2015 @ Moscow 32

slide-33
SLIDE 33

Charge distribution

2015/7/7 Photo Det 2015 @ Moscow

Pedestal Signal (High luminosity)

noise (pedestal RMS) signal w/ transformer 0.69pC 11.1pC±4.4pC w/o transformer 0.94pC 7.0pC±4.5pC

  • The charge of noise is reduced to

73.4%

  • The output charge of signal

increased to 1.5 times.

  • Although height is lower,

the influence to the output charge from overshoot is smaller because faster oscillation pC

  • 6 -4 -2 0 2 4 6 8 10 12
  • 20 -10 0 10 20 30

pC

w/o transformer w/ transformer

50 100 150 200 t/ns

200mV/Div

Wave form

w/o transformer w/ transformer w/o transformer w/ transformer

33

slide-34
SLIDE 34

The waveform of 1p.e. signal in different HV

2015/7/7 Photo Det 2015 @ Moscow

5mV 20ns

Bandwidth:20MHz LV: 415V

HV:8kV HV:9kV HV:10kV HV:11kV 5mV 20ns 2mV 20ns

0.5 1 1.5 2 7000 8000 9000 10000 11000 12000

1p.e. charge(pC) HV(V)

1p.e. integrated charge in 60ns Pulse Height

1.7 2.4 3.1 3.8 7000 8000 9000 10000 11000 12000

1p.e. pulse height (mV) HV(V)

1p.e. pulse height

Integrated charge in 60ns Integrated charge in 100ns

34

slide-35
SLIDE 35

Peak time distribution of 1 p.e. signal

2015/7/7

3.25ns 2.75ns 1.86ns 1.85ns

Photo Det 2015 @ Moscow 35

slide-36
SLIDE 36

Transit time distribution

2015/7/7

http://arxiv.org/abs/1307.0162 Transit time distribution @8kV ns(a.u.) 130 135 140 145 150 155 Fitted to Exponential Modified Gaussian, a convoluted function of an exp func. and a Gaussian func. For the fast part (than the peak), this result can be used to compare with the result of SK PMT. σ σ(SK PMT TTS)

Photo Det 2015 @ Moscow 36

slide-37
SLIDE 37

Multi-ch AD of HPD

2015/7/7

mV

0 5 10 15 20

  • 4 0 4 8 12 16 20 24 28

pC Pedestal 1pe 2pe 3pe Pedestal 1pe 2pe 3pe

noise 1pe 1pe resolution S/N P/V Pulse height 1.74mV 5.57mV 44%

  • 1.98

Charge 0.72pC 3.24pC 28% 4.50 3.08

5ch HPD

HV: 8.4kV LV: 415V BW: 20MHz

Photo Det 2015 @ Moscow

The 50cm HPD with 5-ch AD shows good performance

37

slide-38
SLIDE 38

Time walk Effect

2015/7/7 Photo Det 2015 @ Moscow

Threshold Charge(0.05pC)

t1 tA tB

ｔ1: Lighting time ｔi:I Detection time (i=A,B)

信号A 信号B Time(a.u. ns)

The difference between signal A and signal B is due to time walk effect.

  • The time when it reached the threshold

is different due to the different pulse shape TQ correction baseline

Signal A Signal B

Time walk effect

時間 Voltage

38

slide-39
SLIDE 39

Time walk Correction

2015/7/7 Photo Det 2015 @ Moscow

Origin After correction Total time distribution Time (ns)

39

slide-40
SLIDE 40

Measurement of linearity（low luminosity）

  • The relationship between the HPD output and incident light is determined

by the charge distribution of multi p.e.

  • The linearity is kept in 2%

2015/7/7 Photo Det 2015 @ Moscow

Multi p.e. Charge distribution

Output(0.05pC)

ped 1 p.e. 2 p.e. 3 p.e. ped 1 p.e. 2 p.e. 3 p.e. Linearity

Entries Output(0.05pC)

一次関数に フィットした結果

40

slide-41
SLIDE 41

Measurement of linearity（high luminosity）

  • We use two laser diodes to

evaluate the linearity of the

  • utput
  • We fixed the output luminosity of A

and change the one of B from 1 p.e to several hundred p.e. by to check the linearity of wide region

2015/7/7

Number of Photon

Laser Diode A (Fixed) Laser Diode B (Variable) Laser Diode A+B lighting at the same time

The expected output is O(A+B)=O(A)+O(B)

A Bi A+Bi O(A) O(B) O(A+B) Output

Photo Det 2015 @ Moscow 41

slide-42
SLIDE 42

Some remained issue & To do

  • Low gain
  • Some loss due to trans
  • Both of the gain of AD and preamplifier are not enough
  • To do: We will try to design the preamplifier with higher gain
  • High frequency noise
  • The high frequency noise, including from HV module and outside of the HPD,

reduced the S/N ratio of HPD severely

  • To do: Optimize the HV module and try to use low pass filter for HPD
  • Strong Overshoot (30% ~ 40%)
  • The wave form of HPD has strong overshoot, which can influence the

integrate charge measurement.

  • In integrate charge distribution, 1p.e peak can be seen clearly when the integrate gate is

60ns, but cannot be seen when gate is 100ns.

  • To do：Optimize the parameter in the preamplifier

2015/7/7 Photo Det 2015 @ Moscow

HV:8kV Pulse height Integrated Charge Measured 2.1mV 0.836pC Target ~8mV ~3pC

42

slide-43
SLIDE 43

Dark count rate

2015/7/7 Photo Det 2015 @ Moscow

1 p.e. 0.5 p.e. 0.8 kHz

Seems lower than the SK PMT(4 kHz)

  • Lower collection efficiency
  • Large individual difference

2 kHz

20 cm HPD The dark count rate is at the same level of 20cm HPD HV:8kV LV:293V AD:5mm

43