The%20CMS%20Silicon%20Strip%20Tracker:%20Design%20and%20Current%20Status

1
The CMS Silicon Strip TrackerDesign and Current
Status

• Anthony Affolder
• University of California, Santa Barbara
• On behalf of the CMS SST collaboration

2
Outline

• Detector Layout
• Components and Sub-structures
• Organization
• Status of Construction
• Schedule
• Module Assembly and Quality Assurance
• Substructure Assembly and Quality Assurance
• System and Beam Test Results

3
CMS compared to CDF
4
CMS Tracker
Outer Barrel (TOB)
Pixels
End Caps (TEC 12)
Inner Barrel Disks (TIB TID)
2,4 m
5.4 m
volume 24.4 m3 running temperature 10 0C
5
Silicon Strips
6 layers of 500 mm sensors high resistivity,
p-on-n
93 disks per end
Blue double sided
Red single sided
4 layers of 320 mm sensors low resistivity, p-on-n
Strip lengths range from 10 cm in the inner
layers to 20 cm in the outer layers. Strip
pitches range from 80mm in the inner layers to
near 200mm in the outer layers
6
Some Tracker Numbers

• 6,729 Thin sensors (320 µm)
• 19,280 Thick sensors (500 µm)
• 6,472 Thin modules (1 sensor)
• 9,640 Thick modules (2 sensors)
• 3300 3172 Thin modules (ss ds)
• 5840 3800 Thick modules (ss ds)
• 10,016,768 individual strips and readout
  electronics channels
• 78,256 APV chips
• 26,000,000 Bonds
• 223 m2 of silicon sensors (175 m2 48 m2)

7
Tracker Inner Barrel
Support mechanics CF space frames and/or
Honeycomb structures
8
Tracker End Caps
Digital Optical Hybrid
R6
Interconnect Board
Analogue Optical Hybrid
R4
Frontend Hybrid
R2
9
Tracker Outer Barrel
0.9 m
1.2 m
10
CMS SST Organization
Sensors
Pitch adapter
Hybrid
Frames
Hybrids
factories
Brussels
Brussels
CF carrier
Strasbourg
CERN
Perugia
Wien
RU
Louvain
Sensor QAC
Karlsruhe
Strasbourg
Module assembly
Perugia
UCSB
FNAL
Wien
Lyon
Bari
Bonding testing
Aachen
Karlsruhe
Strasbourg
Zurich
Wien
UCSB
Hamburg
Integration into mechanics
ROD INTEGRATION
PETALS INTEGRATION
TIB-TID INTEGRATION
Aachen
Louvain
FNAL
Brussels
Karlsruhe
Lyon
Strasbourg
UCSB
TOB Assembly
TIB-TID Assembly
TEC Assembly
TEC Assembly
Sub-assemblies
CERN
Pisa
Aachen
Karlsruhe
.
--
gt Lyon
TK ASSEMBLY
CERN
11
Module Components
Front-End Hybrid
Pins
(Now flex hybrid)
Kapton cable
Pitch Adapter
Kapton-bias circuit
Carbon Fiber Frame
Silicon Sensors
12
Silicon Strip Readout Hybrid

• Holds 4 or 6 APV25 analog readout chips
• Distributes/filters power, data, and control
  signals
• Holds 3 additional support chips
• DCU
• Measures LV and HV currents.
• Measures hybrid and sensor temperatures
• PLL
• Fine tunes clocks/triggers
• APVMUX
• Multiplexes 2 chips data output
• Produced at Cicorel/Hybrid SA
• Flex kapton circuit laminated onto ceramic

13
Hybrid Problems

• A few potentially serious problems have been
  uncovered during production
• Flex cable fragility
• Wire bonding weakness
• Power via opens
• Problems quickly solved because of the great
  relationship with the vendors
• Problems typically diagnosed, understood, and
  removed in 1-4 weeks

14
CMS Silicon Sensors

• Radiation Hardness Lessons
• p-on-n sensors work after bulk type inversion,
  Provided they are biased well above depletion
• High Breakdown Voltages
• Follow simple design rules for guard strip
  geometries
• Use Al layer as field plate to remove high field
  Region from Si bulk to Oxide
• Careful processing especially implants
• Surface radiation damage can
• Increases strip capacitance noise
• Use lt100gt crystal instead of lt111gt

Single-Sided Lithographic Processing ( AC,
Poly-Si biasing )
Al Strips
p implants
n Bulk
n Implants
15
Sensor Production
HPK-Thin and Thick Sensors
STM -Thick Sensors

• Thin Sensors (320 mm)
• Large fraction of sensor delivered
• 6273 of 6877 (91)
• High rate of acceptance after QCA sensor testing
  (gt99)
• Issue of high resistivity solved
• Thick Sensors (500 mm)
• Recently a fraction of thick sensors moved to HPK
  to meet schedule
• First prototypes of each sensor type to be
  delivered in June
• First large (600 sensors) batch to be delivered
  in July
• HPK can increase rate of production to level
  needed

• Thick Sensors (500 mm)
• Newer production line with historically higher
  rejection rate in internal testing (10)
• Would require full testing of all sensors. NOT
  POSSIBLE.
• Various potentially serious problems found
• Low inter-strip resistance
• Micro-discharge
• Due to hard work of the CMS sensor group and STM,
  internal yield has increased to about 98 and
  uniformity has improved
• First delivery with final processing is currently
  being tested by the sensor group

16
Production Schedule

• Component availability has delayed the beginning
  of large-scale production worldwide
• To meet scheduled completion date of June 2005,
  module construction schedule compressed from 2
  years? 1-1½ year
• Each sub-systems effected differently
• Hybrid and sensor problems has effected
  production of thick sensor modules (TOB and outer
  TEC) more
• TIB and thin sensor TEC module production
  proceeding at high rate
• How will we be able to meet such to new,
  compressed schedule?
• Will the quality of the finished detector suffer?

17
Module Quality Control and Assurance

• Uniform equipment and procedures ensures the high
  quality of the modules in the highly distributed
  system
• 7 module assembly centers
• 14 module bonding and testing centers
• 9 sub-detector assembly and testing centers
• Cross-calibrations are performed between sites in
  which standard candles (glass plates, hybrids,
  modules, etc.) and results compared
• Module defects appear in the same manner at all
  sites
• Require a high level of traceability of
  component/module flow and of testing results at
  all stages (Database)
• Global quality of production can be monitored
• Quick feed-back to production center, improving
  process quality and uniformity

18
Module Assembly

• Robotic gantry systems assemble modules from
  components
• In order to meet new schedule, module production
  capacity has increased significantly in the last
  year
• Improvements in technique and efficiency
• Two person operation where the next plate
  prepared while current plate is being assembled
• Surveying modules on independent machine (OGP) in
  order to free gantry for production
• Further increases in capacity already possible
  with a little fine tuning of procedures/experience
  
• If necessary capacity can be increased further
  with split-shifts

19
Module Assembly Status

• Bari and Perugia gt TIB/TID
• Demonstrated capacity of 24 modules a day
• gt800 modules assembled
• Brussels, Lyon, Wien, and UCSB gt TEC
• Demonstrated capacity of 30 module a day
• gt450 modules assembled
• FNAL and UCSB gtTOB Modules
• Demonstrated capacity of 30 modules a day
• gt400 modules assembled
• Demonstrated total capacity of more than 90
  module/day

20
Module Mechanical Precision

• Fiducial markers on the sensors, the hybrid, and
  the carbon fiber frame are measured
• Requirements on relative positions and angles
  between the sensors, the hybrid and the frame are
  made
• Dx and angle between sensors and between the
  frame and the sensors are most critical

Frame
Silicon
21
Module Mechanical Precision

• 97 modules meet the current strindent selection
  requirements
• Most failures are just outside the relative
  angular requirement
• After building 10 modules per assembly plate,
  individual placement corrections can be applied
• Further improved mechanical quality of modules
• No new failures after corrections

Dx(Frame-Sensor) (mm)
Dx(Sensor-Sensor) (mm)
Dq(Frame-Sensor) (mdeg)
Dq(Sensor-Sensor) (mdeg)
22
Bonding Centers

• All centers are equipped and fully operational
  (mostly Delvotec and KS bonders)
• Bari,Catania, Firenze, Padova, Pisa, Torino ?
  TIB/TID Modules
• Aachen, Hamburg, Karlsruhe, Strasbourg, Wien,
  Zurich, UCSB ? TEC Modules
• FNAL, UCSB ? TOB Modules
• Bonding capability easily matches production
  capability of the gantry centers
• The group has the ability to place over 130,000
  bonds/day
• Equivalent number of bonds as entire CDF Run II
  upgrade in two weeks

23
Hybrid Module Electrical Testing

• Hybrids and modules are tested at all stages of
  module production
• Ensure that the performed operations do not
  introduce any or very few defects
• Require lt2 Faulty Channels Per Module
• Noise performance and shielding standardization
  has allowed for the same fault finding algorithms
  to work on the TIB, TEC TOB
• Eliminates external noise sources
• Results can be combined for the same module type
  measured at different sites in order to further
  refine testing

24
Hybrid Module Electrical Testing

• Module testing has matured greatly with the
  beginning of the large scale production
• Minimum set of tests defined significantly
  reducing testing time
• Modules are now fully characterized in under 20
  minutes
• Fault finding algorithms are now tuned to
  maximize fault finding and fault type
  identification while minimizing false bad channel
  flagging
• Test failures are correlated in order to diagnose
  fault type
• Faults are found gt99 with correct fault type
  identified 90 of the time
• Less than .1 of good channels flagged as faulty

25
Electronic Testing Cycle
Gantry makes modules
Quick test unbonded module
Thermal/quick test hybrid
Wirebond
Bonded module test
Thermal cycle module
TEC/TOB
Assemble/test petals/rods
Petal/rod burn-in
Final pinhole test
26
Module Fault Finding
Noise Measurement
Pulse Height Measurement (Using Calibration Pulse)
Opens
Noisy
Shorts
1 sensor open
Pinhole
2 sensor open
Pinholes
Bad Channel Flags
Bad Channel Flags
27
Module Quality

• Goal of less than 1 faulty channels per module
• Single Sensor Modules
• 0.20 Faulty Channels Per Module
• Production introduced faults at less than 0.1
  rate
• Two Sensor Modules
• 0.55 Faulty Channels Per Module
• Production introduced faults at less than 0.1
  rate
• Over 1500 modules produced with industrial
  methods with historically low rate of faulty
  channels
• Made possible by the design of the modules which
  emphasizes robustness and simplicity

28
Substructure Production

• Shell, petal and rod production is underway
• Readout and control electronics, optoelectronics,
  and cooling elements are being assembled on
  carbon fiber frames
• 10 rods and 2 petals have had modules mounted
  and tested at final assembly sites
• Systems tests aid in the finalization of
  components and procedures
• Petals modified to improve cooling efficiency
• Rod/petal grounding and shielding modified to
  reduce noise

29
Substructures Beam Tests

• May 2003 Beam Test
• Bunched 25 ns beams of muons and pions
• Tested subsystem integration
• 6 TIB modules on layer 3 shell, 10 TEC modules on
  petal, and 6 TOB modules
• Studied signal-to-noise distributions, power
  supplies, electronics cross-talk, cooling
  performance, etc.
• Detector performance as expected and needed
• May 2004 Beam Test
• Multiple rods, petals, and shells
• Larger system integration tests
• Tracking tests
• Position resolution, hit efficiency

Beam Direction
30
Conclusion

• This has been a very eventful year
• Many potentially serious problems found and
  addressed
• These setbacks have delayed the beginning of
  full-scale production significantly
• In order to meet the delivery date, the entire
  production/testing procedures have been
  optimized, doubling our module production
  capacity
• Full scale production of all subsystems will
  begin in the summer
• Module production should finish summer of 2005

31
Why a Silicon Tracker?

• Efficient robust tracking
• Fine granularity to resolve nearby tracks
• Fast response time to resolve bunch crossings
• Radiation resistant devices
• Ability to reconstruct high pT tracks and jets
• 1-2 pT resolution at 100GeV
• Ability to tag bottom/top quarks through
  secondary vertex
• Asymptotic impact parameter sd 20mm

Golden Channel
32
Inner Detector

• Inside the 4T Solenoid Field
• Pixels 3 Layers
• Si Strips 10 Layers and 9 3 disks per end
• EM Calorimeter PbWO4 crystals w/Si APDs
• Had Calorimeter CuScintillator Tiles