Transcript 20030626_Muheim_mapmt

MaPMT L0 Electronics &
Photo detector mounts

Report from meeting
– at CERN 22 May 2003 and follow-ups

Conceptual Design
L0 motherboard
– Design completion scheduled
for end 2003
– on critical path for MaPMTs

Photo detector mounts
– RICH 2 end of 2003
solution required which satisfies
boundary conditions
– RICH 1
more severe space constraints
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
1
3 x 3 MaPMT Cluster Set-up
40 MHz Read-out
APVm chip
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
2
Bleederboards
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
3
MaPMT Module

Modules of 4x4 MaPMTs
–
–
–
–
–

Proposal for photodetector
choice / TDR
16 tubes, 1024 channels
Size: 113.8 x 113.8 mm2
Filling factor: 0.72
Frame: carbon fibre, G10
Cooling: 16 W / module
-metal shield
– for individual tubes 0.9 mm thick

Lenses
– incorporated in module
– no optical contact

RICH system
– # of modules: 232
– 3504 MaPMTs
– # of channels: 237 kc
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
4
LHCb RICH2 MaPMTs
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
5
Cooling

Power Dissipation
– Bleeder board
0.55 W / resistor chain at HV = 1000 V
– Power dissipation 8.8 W / module (MaPMTs)
2.2 W / module if 4 tubes share resistor chain
– Beetle chip 750 to 900 mW or 6-7 mW / channel
– Power consumption 7.2 W / module (8 Beetle chips)
– Total power dissipation
16 W / module
~ 1000 W/ half plane
– compare EUSO
200 W / plane
need to understand this difference
– Cockroft-Walton

Cooling design
– needs input from L0 motherboard design
Connections, Beetle pad layout
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
6
RICH2 Photo Detector Plane

Proposal by Olav

Assumptions
– basic unit for electronics
1024 channels, 8 Beetles, 16 tubes
– From cooling simulations
heat sink not more than
2 tubes wide


Modules of 8 x 2 MaPMTs
– MaPMT pitch
28 mm
– module width
56 mm
– module height
224 mm
RICH 2 implementation
– # of modules
72 / half plane
– # of MaPMTs
1152 / half plane
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
7
Photo Detector Footprint

Assumptions
– MaPMT pitch

28.45 mm (Olav used 28 mm)
RICH 2 half plane
– photo detector area 1300 mm x 682 mm
(RICH2 EDR)
– 6 columns with 11 modules of 4x4 tubes, total 1056 MaPMTs
12 columns with 6 modules of 2x8 tubes, total 1152 MaPMTs

RICH 1 half plane
– photo detector area 1200 mm x 540 mm
(email Dave Websdale)
– 11 rows with 5 modules of 4x4 tubes, total 880 MaPMTs6 rows
with 10 modules of 2x8 tubes, total 960 MaPMTs

Total RICH 1 and RICH 2
– # of MaPMTs with 4x4 modules
3872
– # of MaPMTs with 8x2 modules
4224
– less flexibility with 8x2 modules
up to 9% more MaPMTs and electronics needed
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
8
Magnetic Shield/Lenses

-metal shield
– for individual tubes
– thickness: 0.9 mm
increased B field in RICH 1
– not necessary in RICH 2

Lenses
– could be incorporated in
magnetic shielding
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
9
Magnetic Shields

Magnetic Field Measurements
– documented in note LHCb 2003-042
– requirement
10% maximum loss of photons
– MaPMT needs shielding above 20 G longitudinal field
– with -metal shielding can operate up to 80 G

RICH 2
– maximum B-field in box 4.1 … 9.1 G
– about 5% loss of photo electrons at 10 G
– safety factor >2 for <10% loss requirement
(RICH2 EDR)
– -metal shielding not required for RICH 2

RICH 1
– see talk by Bill Cameron
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
10
Longitudinal B-Field
-metal shielding:
0.9 mm thick
13 or 20 mm extension
0.9
90%
MAPMT
13
shielded MaPMT with
13 mm extension:
MaPMT functions in
longitudinal fields
up to 8 mT (80 G)
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
11
Lenses & -metal Design

Different mounting designs possible

Lenses / -metal mounted onto modules
–
–
–
–

lenses glued to MaPMT or spring loaded
easy to exchange, little clearance needed
costs if lenses are not glued to MaPMT
lenses glued to MaPMT, transparency of glue
Lenses / -metal separate structure from modules
–
–
–
–
–

(Geoff Barber et al)
lenses or lens-arrays glued together
 can be used as gas seal, cost savings, no more quartz window
 cost savings, structure is glued
larger clearance wrt to MaPMTs, seal must be gas tight
cannot replace lens if it breaks or turns yellow
Lenses are separate structure from -metal
– -metal can be part of modules
–  can combine advantages of above 2 designs
– cost, complexity likely increases
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
12
Transparent Glue

EPO-TEK 301-2
– 2 component epoxy
– refractive index 1.564
– increased B field in
RICH 1
– not necessary in RICH 2

Transmission
– > 97% from
300 nm .. 2.5 m

Needs study
– Is EUSO using it?
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
13
Issues

Footprints of RICH2 and RICH2
– agree on dimensions for basic components
– 4x4 vs 8x2 decision

Magnetic Shields
– RICH 2 no -metal shields
– RICH 1 Is -metal shield really needed
urgent
when?
ok
when?
– Lenses
– study moulded lenses
– agree on dimensions could be different for RICH 1 and RICH 2
when?
– Decide if lenses make gas seal
soon

Photo detector mounts
– decide on design of lenses / -metal

when?
Next steps
– Produce plan with decision points
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
14
Backup Slides
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
15
MaPMT
L0 motherboard
MaPMT meeting
CERN
22 May 2004
Franz Muheim
University of Edinburgh
on behalf of MaPMT group
L0 Interface Board
1st prototype
GOLs
VCSEL
JTAG
connector
Power
supply
regulators
MaPMT
College,
TTCrx
onWorkshop,
reverseImperial
of board…
26/27 June 2003
PINT
DACs
F. Muheim
17
connectors to PIXEL chip
Reverse of L0 Interface Board
1st prototype
TTC receiver chip
MaPMT Workshop, Imperial College,
26/27 June 2003
TTCrx
F. Muheim
18
Transverse B-Field
Transverse field in x-direction
90%
MaPMT insensitive
to transverse fields
up to 23mT (230 G)
without shielding
normalised light yield in whole MaPMT
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
19
Longitudinal B-Field
No shielding
Colour
Legend
light yield: rows
6mT
light yield: columns
R1
 1 mT
 2 mT
 3 mT
 4 mT
 5 mT
R5
 6 mT
 7 mT
 8 mT
R6
 9 mT
C1
 10 mT
 15 mT
CoG: rows
CoG: columns
 20 mT
C6
Conclusions:
Agreement with previous test
 25 mT
 30 mT
Top and bottom row drop first
 35 mT
Sizable loss for edge rows at 3mT
Decrease due to loss of p.e.
Row 5 anomaly understood from
x-talk in APVm readout
MaPMT Workshop, Imperial College,
26/27 June 2003
F. Muheim
20