R

1
RD Program for any detectors using large
number of photodetectors
2
Large optical surface detector
Huge amount of very large photodetectors (PMTs
of 20 size)
Proposition PMm2
Replace large PMTs (20) by groups of
smaller ones (12)
Integrated electronics (Multichannel, close to
the PMTs)
3

• PMm2 philosophy for large detectors
• Replace large PMTs (20) by groups of smaller
  ones (eg. 12) originally proposed by Photonis
  Co. at NNN05

No possible local coincidence (low energy
event 10PMs/MeV over 81000 PMs) ? TRIGGER LESS
x16
Twisted pairs cables for Clock sync. - Digitized
Data - Power
MEMPHYS 3 x 81,000 PMTs LENA GLACIER
20,000 30,000 PMTs
4
50
115
53
k
Post-Doc (LAL)
119
167
500k/3yrs funded by new French Agency (ANR)
Starts officially 25 Jan. 07
5
The team 5 FTE (Ing.), 1 post-doc

• LAL
• P. Barillion, S. Blin, Th. Cacérès, J.E
  Campagne, Ch. de La Taille, G. Martin-Chassard,
  N. Seguin-Moreau
• Wei-Wei (Chinese post-doc since Sept. 07 for 6
  months)
• IPNO
• B. Genolini, Th. Nguyen Trung, J. Peyré, J.
  Pouthas,E. Rindel, Ph. Rosier
• LAPP
• N. Dumont-Dayot, D. Duchesneau, J. Favier, R.
  Hermel, J. Tassan-Viol
• Photonis
• P. Lavoute, C. Moussan

coordinator
http//pmm2.in2p3.fr
6
Front-End Electronics Requirements

• Auto trigger
• 100 trigger efficiency _at_ 1/3 p.e (50fC)
• Excellent time resolution lt 1ns
• Dynamic range up to 300 p.e
• Provide digitized signals
• Scalability
• Low cost
• Profit from progress in micro-electronics and DAQ
  !
• Many issues in common with HPD or large PMTs
  developments by KEK (see M. Tanaka)

7
MAROC 64 ch MaPMT chip for ATLAS lumi
5x5 array of 64 anodes PMT
Hamamatsu H7546B
BOTTOM side
LAL has a solid expertise in micro-electronic
developments and we will reuse efficiently the
existing and validated blocs.
MAROC1
Chip On Board
8
MAROC performance

• Gain adjustment
• Trigger efficiency 100 _at_ 50fC (1/3 p.e _at_ 106)
• Can be lower down to 10fC

P. Barillon et al.,  MAROC Multi-Anode ReadOut
Chip for MaPMTs , IEEE 2006, October 29 Nov.
4, 2006 San Diego, California P. Barrillons
talk at TWEPP07
(50 3) fC over 64 chan.
9
New ASIC Architecture
10
Complementarily with PMm2 a collaboration
with KEK starts for HPD read out

• Since NNN06 a new thread of collaboration has
  emerged.
• M. Tanaka-san came March 07 to test MAROC2 at LAL
  and discuss informally of possible common RD.
  Right after, a MAROC2 chip with Test Board had
  been send to KEK for evaluation with Hamamatsu
  13 HPD.
• Ch. de La Taille N. Seguin-Moreau had attended
  the FJ-AIL meeting and a new project between
  Japan France had been accepted in the FJ-AIL
  context for HPD Read-Out join RD.
• M. Tanaka came at LAL end Sept. 07.

11
 All ideas on photodetection designs are
certainly interesting But
if a mass production is foreseen
Constraints from industry must be considered
from the beginning 
J. Pouthas at NNN06
12
Improved photocathode
Quantum efficiency (400 nm) Standard
26 Improved 32
Control by Pulse measurements in Single Electron
Response (Relative detection efficiency)
Drawbacks ?
Dark count rate Same at low temperature Increase
with temperature
D. Dornic et al, Beaune Conference, France, June
2005. Nucl. Instr. and Meth. A567 (2006) 27.
13
New Ph. D Cl. Périnet (? Oct. 09)

• Specific RD for PMm2
• Parameter correlation studies
• Signal/Noise _at_ 1p.e vs Dark current vs Quantum
  and Collection Eff.
• ? What is the optimum ?
• Potting preliminary studies done
• Water Pressure test facility (_at_ 10bars) under
  study
• Water tight Box for the Read Out electronics
  under design
• Glass shape optimisation for 10bars mechanical
  studies undertaken
• 12 PMTs will produced by Photonis and qualified
  by IPNO
• Mechanics of the demonstrator

14
Digital part details

• SPI bus Dialog with FE ASIC, Read IRIGB time
  stamping (UTC pps)
• I2C bus Monitoring and control
• Ethernet (data) Power Over Ethernet ( power
  supply)
• 50V for HV generation locally
• Ok at least for 40 Mb/s _at_ 5kHz/PM Dark Current
  

15
Summary Outlook

• PMm2 is funded and the RD began on Jan. 07.
• A lot to do
• DAQ-Clock system is for the moment at a stage of
  selection of the appropriate technology.
• New generation of FE ASIC will be submitted soon.
• Photodetector specific test bench pressure
  under water is under study.
• PMT detailed specifications still to be
  finalized. Eg. Dark Current is certainly an
  issue.
• A complementary RD has also started for HPD
  readout between LAL and KEK in 07.

See you in Paris for NNN08 (Sept.)
16
BACKUP
17
Some detectors presented at NNN Workshops
Start 99, Aussois 05, Seattle 06, Hamamatsu 07,
Paris 08
Water Cerenkov 500kT?1Mt
MEMPHYS
HyperK
UNO
65m
80m
European initiative
Liq. Argon ?100kT
Liq. Scintillator ?50kT
GLACIER
LENA
Large Apparatus for Grand Unification and
Neutrino Astrophysics LAGUNA
18
Charge output linearity
40p.e

• Above measurements performed with the external
  ADC of the test board
• The pedestal (measured with the first TH) was
  suppressed
• Linearity of 2 approximately
• The 64 Wilkinson ADCs are also working 12bits,
  80 µs conversion time

19
Noise (dark pulses)
All the main results in PhD of D. Dornic Sept.
06 And are inputs for PMm2 photodetector
requirements.
20
MAROC2 architecture

• MAROC2 MAROC1 additional features (ADC
  Wilkinson, 3 discriminators, Encoder)

• Similar to OPERA ROC
• Low input impedance (50-100 O)
• 6 bits gain adjustment (G0-4) per channel
• 64 discriminator outputs
• 100 sensitivity to 1/3 photoelectron (50fC).
  Counting rate up to 2 MHz
• Common threshold loaded by internal 10bit DAC
  (step 3mV)
• 1 multiplexed charge output with variable shaping
  20-200ns and Track Hold.
• Dynamic range 11 bits (2fC - 5 pC)
• Crosstalk lt 1

• Technology AMS SiGe 0.35µm
• Submitted March 06
• Area 16 mm2
• Received in june 06
• 240 pins

21
MAROC 64 ch MAPMT chip for ATLAS lumi
Hold signal

• Similar to OPERA ROC
• Low input impedance (50-100 O)
• 6 bits gain adjustment (G0-4) per channel
• 64 discriminator outputs
• 100 sensitivity to 1/3 photoelectron (50fC).
  Counting rate up to 2 MHz
• Common threshold loaded by internal 10bit DAC
  (step 3mV)
• 1 multiplexed charge output with variable shaping
  20-200ns and Track Hold.
• Dynamic range 11 bits (2fC - 5 pC)
• Crosstalk lt 1

Multiplexed Charge output
Variable Slow Shaper

SH




Photons
64 inputs
Variable Gain Preamp.
Bipolar Fast Shaper

Photomultiplicator









64 trigger outputs


Synoptic diagram of MAROC1
Gain correction 6 bits/channel

discriminator threshold 10 bits DAC
MAROC1

• Technology AMS SiGe 0.35µm
• Submitted 13 june 05
• Area 12 mm2
• Received in november 05
• Dissipation 130 mW _at_VDD3.5V

OPERA_ROC successor
22
IPN Orsay / Photonis Overview on results
Improved photocathode
D. Dornic et al, Beaune Conference, France, June
2005 In press in Nucl. Instr. and Meth.
XP1805 (9, AUGER PMT)

• Standard (800 PMTs)
• Sk CB 9.32 µA/lmF
• Sk White 68. 37 µA/lm

Blue measurement (Corning Blue filter)

• Improved (25 PMTs)
• Sk CB 11.35 µA/lmF
• Sk White 118.00 µA/lm

Increase of Sk CB 19 Increase of Sk
White 42
White measurement
Joël Pouthas IPN Orsay
23
ASIC Chip TEST BOARD designed at LAL
GPIB port
USB port
64ch PM socket
MAROC (COB)
Control Altera
24
ADC performance

• Wilkinson type
• 64 channels
• 12 bits
• 80 µs conversion time

25
MAROC 2 ASIC

• 64 channels
• Preamps
• Fast shaper 15ns
• Discriminators
• Slow shaper
• TrackHold
• 12bit ADC
• 10bit DAC
• Bangap reference
• Digital formatting
• Silicon Germanium
• 0.35µm BiCMOS
• 16mm2 area

26
MAROC 64 ch MAPMT chip for ATLAS lumi

• Similar to OPERA ROC
• Low input impedance (50-100 O)
• 6 bits gain adjustment (G0-4) per channel
• 64 discriminator outputs
• 100 sensitivity to 1/3 photoelectron (50fC).
  Counting rate up to 2 MHz
• Common threshold loaded by internal 10bit DAC
  (step 3mV)
• 1 multiplexed charge output with variable shaping
  20-200ns and Track Hold.
• Dynamic range 11 bits (2fC - 5 pC)
• Crosstalk lt 1

27
MAROC Main Features

• Technology AMS SiGe 0.35 mm
• Package CQFP240
• Power consumption 350mW
• 5 mW/ch
• Area 16 mm2
• Submitted March 2006
• Received July 2006

Second version