design and simulation and implementation

ahmedabed37 10 views 63 slides Oct 12, 2024
Slide 1
Slide 1 of 63
Slide 1
1
Slide 2
2
Slide 3
3
Slide 4
4
Slide 5
5
Slide 6
6
Slide 7
7
Slide 8
8
Slide 9
9
Slide 10
10
Slide 11
11
Slide 12
12
Slide 13
13
Slide 14
14
Slide 15
15
Slide 16
16
Slide 17
17
Slide 18
18
Slide 19
19
Slide 20
20
Slide 21
21
Slide 22
22
Slide 23
23
Slide 24
24
Slide 25
25
Slide 26
26
Slide 27
27
Slide 28
28
Slide 29
29
Slide 30
30
Slide 31
31
Slide 32
32
Slide 33
33
Slide 34
34
Slide 35
35
Slide 36
36
Slide 37
37
Slide 38
38
Slide 39
39
Slide 40
40
Slide 41
41
Slide 42
42
Slide 43
43
Slide 44
44
Slide 45
45
Slide 46
46
Slide 47
47
Slide 48
48
Slide 49
49
Slide 50
50
Slide 51
51
Slide 52
52
Slide 53
53
Slide 54
54
Slide 55
55
Slide 56
56
Slide 57
57
Slide 58
58
Slide 59
59
Slide 60
60
Slide 61
61
Slide 62
62
Slide 63
63

About This Presentation

design and simulation and implementation


Slide Content

Strips and Pixels Detectors
Associated Electronics
Jean-François Genat
LPNHE Paris
Louvain la Neuve,
Jan 16th 2008
J-F Genat, LLN Jan 2008

Outline
J-F Genat, LLN, Jan 16th 2008

Solid-state vs Gaseous, basics
Strips
Timing
Pixels
Monolithic
DEPFETs
SOI
Hybrids
3D detectors
Readout electronics
Perspectives

Solid state detectors vs Gaseous
Gaseous Silicon
Ionization energy 20 eV 3.6 eV
Primary ionization 3e-/mm 60k e-/mm
Amplification 100k No
Exceptions: DEPFETs
(APDs, Silicon PMTs)
Carriers velocity 50 m/ns 500 m/ns (electrons)
Signal processing No Yes
integration
Radiation hardnessYes No, if not designed for
J-F Genat, LLN, Jan 16th 2008

Basic mechanisms
Primary ionization
Low electron-hole pair generation energy: 3.6 eV (Silicon)
High loss/length: 1 MIP = 60k e-/mm
Small range  rays, high position resolution using thin detectors
Collection
- Drift under depletion field: reverse biased PN junction
Voltage needed depends upon resistivity : high resistivity Silicon
(1e4 /cm) to deplete at low voltages (100V)
Drift velocity saturates: v =  E, 1/ = Nq 
- Diffusion
Total collected charge = primary ionization
Unless reduced by :
- recombination with impurities
- radiation damage
Current signal as large as:
- number of moving carriers,
- electric field (as high as closer to small electrodes)
J-F Genat, LLN, Jan 16th 2008

Noises
Sources of noise
Parallel
Detector leakage current
Biasing resistor
Series
Electrode resistance
input transistor (thermal, 1/f)
« System » noise
All other noises !
Reduce detector capacitance (segmentation helps)
Use high good quality material (leakage current)
Use appropriate shaping times
Parallel 1/f Series
f
Input noise
J-F Genat, LLN, Jan 16th 2008
H. Spieler
http://www-physics.lbl.gov/~spieler/Heidelberg_Notes/pdf
Charge amplifier

Outline
J-F Genat, LLN, Jan 16th 2008

Solid-state vs Gaseous, basics
Strips
Timing
Pixels
Monolithic
DEPFETs
SOI
Hybrids
Semi-3D, 3D detectors
Readout electronics
Perspective

Silicon strips DC coupled
Single sided
Double sided
DC or AC coupled
n+
al
al
P+
n
+V
Amplifier
bias voltage
Single sided
DC coupled
DC leakage current flowing through amplifier may saturate
particularly under irradiation
J-F Genat, LLN, Jan 16th 2008
Fully depleted
PN junction
G. Lutz Semiconductor Radiation Detectors.

Springer Verlag 2001

AC coupled strips
AC coupled
No DC flowing through insulated amplifiers
Need for biasing resistors
n+
al
al
P+
n
+V
0V
SiO
2
20% more expensive
’’pinholes’’ through oxide
J-F Genat, LLN, Jan 16th 2008

Silicon strips parameters
Size 100 x 100 mm
2
(6-8’’ wafers)
DC or AC coupled
Pitch 20-200m
Substrate resistivity 2-10k/cm
Capacitance to substrate 100fF/cm
Interstrip capacitance 1pf/cm
Full depletion voltage60-100V
Thickness 50-300 m
Leakage current 50nA/cm
2
at 100V bias
Junction breakdown >200V
Interstrip resistance> 2 G
Edgeless < 10 m
Defective strips <1%
J-F Genat, LLN, Jan 16th 2008

Silicon strips space resolution
Digital readout (ATLAS, CERN ABCD chips):

m
m
Analog readout (CMS, RAL APV25 chips):
Several adjacent strips collect the signal:
Centroids
Improved space resolution:
 = <10 m
J-F Genat, LLN, Jan 16th 2008
12
ATLAS Unno et al. NIM A453 pp109-120
CMS K. Klein. IECHEP 2005, Lisboa

Magnetic field effect
J-F Genat, LLN, Jan 16th 2008
(S. Cucciarelli)
Degrades space resolution

Silicon strips modules
J-F Genat, LLN, Jan 16th 2008
CMS
(K. Klein)
Radiation hardness
is a big issue, as close to the beam

Outline
J-F Genat, LLN, Jan 16th 2008
Solid-state vs Gaseous, basics
Strips
Timing
Pixels
Monolithic
DEPFETs
SOI
Hybrids
3D detectors
Readout electronics
Perspective

Timing
J-F Genat, LLN, Jan 16th 2008
The faster detector and front-end electronics,
The better the timing
Obvious, but still to keep in mind…

Rise time and noise
A
Threshold
1

Slope 1/

A

t
 .
AtA
 .
At

noise
Effects of noise
t
Time spread proportional to rise-time and noise
J-F Genat, LLN, Jan 16th 2008

Amplitude, rise-time
Amplitude and/or Rise-time spectra translate into time spread
J-F Genat, LLN, Jan 16th 2008

Fast detectors
Moving charges:
i = n q v / d
Rise-time di/dt= q [ n dv/dt + dn/dt v]
collection multiplication
Maximize
n primary ionization, detector thickness
dv/dt qE/m electric field
- Electron multiplication
- High electric fields, segmented electrodes
- Reduce collection distance d
-
Detector capacitance (signal/noise)
- Landau amplitude distribution (rise time spreads)
Bias
J-F Genat, LLN, Jan 16th 2008
d
transimpedance

Gaseous vs Silicon
Signals (e-) Rise-time Time resolution
Gaseous
• MWPCs 10
6
3ns 500ps
• RPCs 10
7
1ns 100ps
Silicon
•Strips 10
4
5 ns 2 ns
•3D Silicon 10
4
1ns ?
•Silicon PMs 10
7
1 ns 100 ps
J-F Genat, LLN, Jan 16th 2008

Time picking techniques
J-F Genat, LLN, Jan 16th 2008
Time picking:
Leading edge
Zero crossing
Constant fraction
Multiple thresholds
Sampling
Leading edge vs CFD

Sampling vs CFD
MATLAB Simulation with Silicon signals
Better compared to CFD by a factor of two depending
on noise properties and signal waveform statistics
- MATLAB simulation package
J-F Genat, LLN, Jan 16th 2008

Expected time resolution with Silicon
Simulated time resolution using sampling
- S/N =25
- 16 samples
- 40 ns shaping
1 ns time
resolution
J-F Genat, LLN, Jan 16th 2008

Timing using Silicon strips
M. Friedl et al NIM 572 pp 385-387
Using sampling at 25ns rate, digitization and signal processing,
a time resolution of 2ns id obtained (CMS detectors + APV25 chip)
using a deconvolution algorithm
J-F Genat, LLN, Jan 16th 2008

Coordinate along the strip

15 ns
LCV/1
V = c/4.7
L =50nH R =5 
C
i=500 fF
C
s= 100 fF
120cm
V = c/3.7
8 cm /ns
SPICE
J-F Genat, LLN, Jan 16th 2008

Measured Pulse Velocity
4.5 ns/cm Measured moving a laser diode along 24 cm
Moving a laser diode along the Silicon strip detector
Threshold
V = 4.5cm/ns
J-F Genat, LLN, Jan 16th 2008

Outline
J-F Genat, LLN, Jan 16th 2008
Solid-state vs Gaseous, basics
Strips
Timing
Pixels
Monolithic
DEPFETs
SOI
Hybrids
3D detectors
Readout electronics
Perspective

Monolithic Active Pixels Sensors (MAPS)
Integrate pixel detector and readout electronics
using a single standard opto CMOS process (cheap)
Matrices of 25 x 25 microns pixels
Built-in 2D device
Detector segmentation highly improves Signal/noise
Opto-CMOS allows pixel signal processing
Slow collection (diffusion)
Slow readout (unless on-pixel sparsifier implemented)
Power needed by amplifiers (can be cycled)
n+
n well
p epi
P sub
J-F Genat, LLN, Jan 16th 2008
Dulinski et al. Trans Nucl Sci V49 N2 p601

DEPFETS
One MOSFET on top of a fully depleted PN junction:
MOSFET current is modulated by the deposited charges stored in the bulk
under the channel: amplification . Multiple readout possible (reduced noise)
using two ping-pong DEPFETs.
Thickness down to 50 m 4000 e-/MIP
Noise down to ¼ electrons ! Wolfel et al. NIMA253, 356-377
n+ p+
p
p+ n+ n+
n+
p
Source Gate Drain Clear Bulk
Internal gate
p+
50
MIP
---
+++
n-
1 
G. Lutz Semiconductor Radiation Detectors. Springer Verlag 2001
J-F Genat, LLN, Jan 16th 2008
Gain: 500pA/e-

DEPFETS Readout
J-F Genat, LLN, Jan 16th 2008
Readout chip
• Current based
• 3-fold sampling
• Noise 45 nA (<100 e- equivalent)
• Signal/noise = 40
• Mixed signal FIFO
• Analog pedestal subtraction
• Zero-suppression up to 110 MHz
• 4.5 x 4.5 mm
2
250nm CMOS
M. Trimpl et al. NIMA 511 p257

Silicon on Insulator (SOI)
Superimpose:
- CMOS process on low resistivity Silicon
- Oxide (100-200nm)
- High resistivity Silicon (detector)
- Allow connections through oxide
PMOS
n well
p
NMOS
oxide
n detector
al
p+
+HV
gnd
p+
gnd
J-F Genat, LLN, Jan 16th 2008

SOI features
Full thickness of fast collection detector
High performance CMOS electronics available
Availability of SOI wafers from industry (SOITEC)
Processes available: Japan OKI 150nm, US ASI 180nm.
100% fill factor
Not so radiation hard (oxide)
Not so cheap
J-F Genat, LLN, Jan 16th 2008

SOI pixels
Fermilab pixel design:
64 x 64 matrix of
26 x 26 m counting pixels (12bit, max 1MHz)
350 m detector thickness
Pixel includes:
Amplifier 150 mV/1k e-
CR-RC shaper150ns peaking time
Discriminator
12bit counter
ASI US
OKI Japan
(Numbers from Ray Yarema, FNAL)
J-F Genat, LLN, Jan 16th 2008
Y.Arai et al. NSS 2007 N20-2 pp 1040

Hybrid pixels
Flexibility for detector/electronics process choice
Readout chips available (CERN), but fixed pixel size
Radiation hard (as far as detector and electronics are,
but the choice is more open)
Pixel level bump bonding connexions
J-F Genat, LLN, Jan 16th 2008
Mature technology

Hybrid pixels at LHC
Detector and electronics as two independent Silicon structures
ATLAS Vertex detector:
- Detector:
ATLAS rad-tol: Partially depleted n+ on_inverted_n on
Fz-oxygenated Silicon after irradiation
- Electronics:
FE-I3 chip IBM 250nm (LBL Bonn CPPM)
- Bump-bonding (Bonn – IZM)
CMS Detector: Cz-oxygenated
Electronics: IBM 250nm (PSI)
J-F Genat, LLN, Jan 16th 2008
L. Cremaldi et al. NIM 511-1, pp64-67

FE-I3 chip (LBNL, Marseille, Bonn)
•18 x 160 pixels columns, active area 8 x 7.2mm
2
•Time over threshold (7 bits)
•Level 1 buffering: 3.2s. Max rate: 0.15 hit/BC/column pair=1.35
hit/BCO/chip.
•Trigger output as a single fast OR bit of all (selectable) pixels.
50 m
7.4 mm
8 mm
10 m
I/O pads
1100 mm
400 m
7.2mm
100 m
Output data format:
| Header 1 | BC Id 4 | row 8 |column 5 | Parity 1 | ToT 7 |
J-F Genat, LLN, Jan 16th 2008

ATLAS pixel module
•Pixel module (50,000 pixels)
J-F Genat, LLN, Jan 16th 2008

Outline
J-F Genat, LLN, Jan 16th 2008
Silicon detectors
Solid-state vs Gaseous, basics
Strips
Timing
Pixels
Monolithic
DEPFETs
SOI
Hybrids
3D detectors
Readout electronics
Perspective

3D Silicon
J-F Genat, LLN, Jan 16th 2008

3D Silicon detectors
Full 3D: Charges move parallel to the detector plane instead
transverse, the field created using vertical electrodes
through the detector (MEMS technology). Built-in pixels.
Semi-3D:
The backplane is still an electrode.
Electrons are collected by vertical electrodes, holes by backplane.
J-F Genat, LLN, Jan 16th 2008

3D Active edges Silicon
Full 3D:
Both type of carriers collected by electrodes of two types
Active edges allow sensitivity to 5 microns from the cut,
Abutt detectors without dead zones.
J-F Genat, LLN, Jan 16th 2008

3D Silicon
 Collection distance reduced by 10:
- Faster (1-2ns rise-time)
      - Position resolution down to 10-15 mm in the two dimensions.
- Same collected charge (24k electrons/300 mm)
- More radiation tolerant detectors (10
15
n/cm
2
).
- Depletion voltage reduced to a few Volts.
Any connexions between electrodes are possible (as long as the
total capacitance remains small, if not noise increases)
        
         - Active edges (conducting): edgeless capability down to 10 mm.
         - Readout using CMOS chips available at CERN (ATLAS pixels)
        - Moderate passive cooling.
J-F Genat, LLN, Jan 16th 2008

Available 3D detectors
Active area
50 m
7.4 mm
8 mm
10 m
I/O pads
1100 mm
400 m
7.2mm
100 m
7.4 x 8mm
2
detectors bump-bonded to FE13 chips (CERN)
400 x 50 m
2
pixels
Plans to reduce pixel size to 50 x 50 m
2
J-F Genat, LLN, Jan 16th 2008

220m Roman Pots at ATLAS
IP 220m
8m
One horizontal pot and two vertical pots at 216 and 224 m on each arm
with detectors as close as possible to the beam: 10  = 1mm
3cm
Silicon detectors
One Horizontal pot
Two Vertical pots
J-F Genat, LLN, Jan 16th 2008

3D Silicon detectors for RP 220
J-F Genat, LLN, Jan 16th 2008

Roman Pot Layout
MCP-PMT
Light
Guide
Radiator
8 x 8
Pixels
Read Out
ASIC
64 Ch.
P. Le Dû
XX
2,54 x 2,54 cm
Beam
4,5 cm
FE ASIC
(PA,SH,Pipeline)
Plus FAST OR
Read out and Trigger LOGIC (FPGA)
Power and Clock distribution
Slow control
Cooling
To
Alco
ve
Edgeless
Silicon
J-F Genat, LLN, Jan 16th 2008

Acceptance for diffracted protons
Acceptance for diffracted p[rotons at 220m
Simulation
J-F Genat, LLN, Jan 16th 2008

3D Silicon layout
J-F Genat, LLN, Jan 16th 2008

Silicon Manufacturers
Hamamatsu Japan
Canberra Belgium
SINTEF Norway 3D
Micron UK
CSEM Switzerland
VTT Finland 3D
CiS Germany
IST Italy
STM France
J-F Genat, LLN, Jan 16th 2008

Outline
J-F Genat, LLN, Jan 16th 2008

Solid-state vs Gaseous, basics
Strips
Timing
Pixels
Monolithic
DEPFETs
SOI
Hybrids
3D detectors
Readout electronics
Perspective

Digital sampler chip for Silicon strips
J-F Genat, LLN, Jan 16th 2008
Technology CMOS 130nm

Digital sampler architecture
Storage
Waveforms
Technologies: Deep Sub-Micron CMOS 180-130nm

Deep Sub-Micron CMOS
LPNHE Paris
Counter
Wilkinson
ADC
‘trigger’
Ch #
Charge 1-40 MIP, S/N~ 15-20, Time resolution: BC tagging, fine: ~ 2ns
Calibration
Control
Analog samplers, slow, fast

iV
i > th
Sparsifier
Channel n+1
Channel n-1
Time tag
Preamp +
Shapers
Strip
reset
reset
J-F Genat, LLN, Jan 16th 2008

Digital sampler prototype chip layout
J-F Genat, LLN, Jan 16th 2008
Technology CMOS 130nm

Prototype Sampler chip beam tests
J-F Genat, LLN, Jan 16th 2008
120 GeV pions beam tests
S/N= 18
Detectors+ 130nm chip

Issues
Inter-layers vias through oxide (etching)
Wafer thinning to facilitate through wafer vias
Alignment (< 1m)
Present:
6’’ wafer thinned to 6 m, mounted on 75 m Kapton !
J-F Genat, LLN, Jan 16th 2008

Outline
J-F Genat, LLN, Jan 16th 2008

Solid-state vs Gaseous, basics
Strips
Timing
Pixels
Monolithic
DEPFETs
SOI
Hybrids
3D detectors
Readout electronics
Perspectives

MAPS
Faster collection time by drift in a depleted PN junction
Integrate pixel discriminators:
Sparse readout to fasten the readout
Improve fill factor
Integrate pixel DAC for threshold match
Storage
350nm standard HighVoltage CMOS process
(used for I/Os)
Mannheim Germany
I. Peric IEEE NSS 2007 N20-1 pp1033-1039
J-F Genat, LLN, Jan 16th 2008

3D Silicon
Merge pixels and strips
Pixels top
Strips underneath
Readout with bump-bonded pixel chip, wire bonded
to strip readout chip
Improved space resolution, trigger capabilities with the
strips.
C. Da Via’ 2005
J-F Genat, LLN, Jan 16th 2008

Silicon at low T
J-F Genat, LLN, Jan 16th 2008
Electron and hole mobilities in Silicon vs dopants densities
at various temperatures

Front-ends
NA60 AFP chip (CERN)
Cryogenic Silicon beam tracker for high intensity proton
beam and heavy ions hodoscopes
• 32 channels 0.25m CMOST=80-300K
• Transimpedance amplifier optimized for 4pF input
• CMOS 250nm
• Gain: 10mV/MIP
• Noise: 350 e- (@ 300K)
• Fall time: 3ns @ 300K, 1.5ns @ 130K
• Radiation hard to 10
15
p/cm
2
Designed for 20 m pitch Silicon strips detectors
Rencently tested with 3D detectors
G. Anelli et al
A high speed low-noise transimpedance… NIMA 512 1-2 pp117-120
J-F Genat, LLN, Jan 16th 2008

Silicon readout
J-F Genat, LLN, Jan 16th 2008

Merge low-noise amplification and sampling
Increase number of channels up to 1k
Implement low level Digital Signal Processing
Calibrations, centroids, fits
Flip-chip 3D interconnect

Equip on-detector Silicon strips readout
3D interconnect to pixels (3D or others)

Signal processing
• Fast samplers in CMOS technology
• Saclay (France) 2 GHz 12 bits
• Germany 5 GHz 12 bits
• Hawaii 6 GHz 10 bits
Push sampling speed up to higher frequencies:
Get Charge and Time with ultimate precision
J-F Genat, LLN, Jan 16th 2008

Vertical integration
As device feature size cost increases, vertical integration
allows stacking any processes:
DEPFET + CMOS/SOI
CCD + CMOS/SOI
MAPS + CMOS/SOI
Solves the 2D interconnection bottleneck:
- Reduces I/O pads, crosstalk, power, increases speed for
same functionnality
Ray Yarema, FNAL
Lincoln Labs
IZM Germany
Digital
Analogue
Sensor
J-F Genat, LLN, Jan 16th 2008

Vertical integration references
Ray Yarema Fermilab
Marcel Demarteau ’’
R. Yarema, 3D Integrated Circuits for HEP,
6
th
Front-End Electronics Workshop Perugia 2006
C.Bower et al, High Density Vertical Interconnects
56
th
Electronics Components TC Conference May 30
th
-June 2 2006 San Diego
A.Klump, ‘’3D System integration’’, FEE Perugia 2006
B. Aull et al. Laser Radar Image based on 3D APDs IEEE SSCC 2006, p26
J-F Genat, LLN, Jan 16th 2008

The end…
Lots of ideas
Technology exploding
So much to do…
Thanks !
Tags