Transcript OptoWGVasey - Indico

Joint Atlas CMS SLHC Optoelectronics Working Group
Francois Vasey on behalf of WG
Working
Group History and Status
Results
Two
(not so common) development
projects
Conclusions
20-Mar-07
[email protected]
0
Joint WG started in 2005
23 June 2006: Proposal for a different operating mode





Move from reporting to collaborating mode
Identify the areas of expertise of collaborators
Collaborate in smaller subgroups, organize topical
meetings
Prepare common R&D proposals
Report at bi-yearly plenary meetings
20-Mar-07
[email protected]
1
27 Sep 2006, LECC06 Valencia:
Working Group Identity Revealed
Joint Optoelectronics Working Group
Resources Distribution
20-Mar-07
5
CERN
11.4
EU
US
6.6
Series1
CERN
Ljubljana
Minnesota
Ohio
Oklahoma
Oxford
Polytechnique
RAL
SMU
Strasbourg
8
7
6
5
FTE 4
3
2
1
0
[email protected]
10.2
ATLAS
12.8
2
CMS
27 Sep 2006 , LECC06 Valencia:
Interpretation of Survey Results



13 institutes, 23 FTE
3 institutes above 3 FTE: SMU, CERN, IRES
2 development projects announced:



10 institutes below 3 FTE (0.8 average)


Optoelectronics developments and testing
Periphery: ex-CMS


Mostly involved in testing
Core activity: ex-ATLAS and CERN


Link on chip SOS (SMU, ongoing)
Versatile bidirectional link (CERN, proposal)
Hybrids, Backends, system modeling
No system level work
20-Mar-07
[email protected]
3
27 Sep 2006 , LECC06 Valencia: Discussion

concentrate on a few common projects only




Work in small subgroups
Work in collaborative mode with convergence dates
Identify sub-projects of general interest before end
2006
Did not agree on common link development
20-Mar-07
[email protected]
4
Dec 06: 3 Themes for joint projects

Group a- Lessons learned and to be learned from LHC.




K.K. Gan (Ohio)
F. Vasey (CERN)
T. Weidberg (Oxford)
Group b- Radiation hardness and reliability of
optoelectronic components for SLHC.



J.M. Brom (Strasbourg)
K. Gill (CERN)
C. Issever (Oxford) plus ATLAS institutes participating to SLHC
radiation test programme
20-Mar-07
[email protected]
5

Group c- Common optical link reference test bench.








J-M Brom (In2P3 Strasbourg)
B. Checcucci (INFN Perugia)
J. Coughlan (RAL)
K.K Gan (UofOhio)
C. Hill (Bristol)
T. Romanteau (In2P3 Polytechnique)
R. Russack (UofMinnesota), D. Baden (UofMaryland)
J. Ye (SouthernMethodistUniversity)
20-Mar-07
[email protected]
6
Joint Atlas CMS SLHC Optoelectronics Working Group
Working Group A: Lessons Learned



K.K Gan
F. Vasey
T. Weinberg

Cost



Quality




[email protected]
Today
Long term monitoring
Technology

20-Mar-07
Material
Personnel
Components
Architectures
7
Cost in CHF, material
ATLAS PIX
Quantity
ATLAS SCT
CMS TK
4144
12264
42800
1’527k
3’486k
12’597k
Development
Cost
455k
250k
2’000k
Production
Cost per link
369
284
294
Development
Cost per link
110
20
47
Production
Cost
20-Mar-07
[email protected]
8
Cost: lessons learned

All optical links ended up being expensive





Overproduction (included in cost)





~ 300 CHF/link, production cost
25%-30% for cables and connectors (more for Atlas Pixels)
70%-75% for actives + hybrid/flex
Plus 10% - 30% development cost
Covers assembly yield, test systems and prototypes, QA testing,
installation yield, unforeseen problems
Up to 130% - 140% * needed quantity
Spares down to % level
Maximize bandwidth usage, limit number of variants,
foresee overproduction
Do not allow cost estimates to drive technology choices.
20-Mar-07
[email protected]
9
Cost in MY, personnel (large uncertainty)
ATLAS PIX
Quantity
ATLAS SCT
CMS TK
4144
12264
42800
R&D
12
17
50
Production
11
23
55
Installation
5
Links/MY
20-Mar-07
148
[email protected]
10
307
372
10
Cost: lessons learned (2)

Over 180 MY of effort




Effort per link seems to stabilize above 10k pcs


R&D:Prod:Install 45%:45%:10%
11 years
Not included in ATLAS SCT and CMS TK project cost
Very few points, large uncertainty
Share resources for systems < 10k links


10k links ~ 30 MY
Institute in charge should have more than 3 FTEs
during 10 years
20-Mar-07
[email protected]
11
Quality (19 March 2007)
ATLAS SCT
Data
Quantity
Dead
ATLAS SCT
TTC
4144
20-Mar-07
42800
0.5%
0.2%
0.04%
ESD mishandling
Broken fibers
Broken fibers
0.4%
?%
0.3%
Slow Turn On
Light coupling
into MM fiber
Excess loss
Broken ribbons
system
packaging
mishandling
packaging
Problematic
CMS TK
[email protected]
mishandling
mishandling
12
Lessons learned: Quality

System






Avoid use of single fibres and pigtails on detector
Do not allow excessive fibre-slack without corresponding
management scheme.
Use ruggedized ribbon/fiber only.
Avoid simplified and/or compact connectors which are
fragile and difficult to dismount and clean.
Always use balanced codes
Separate front-end and opto-modules (i.e. with TP)
20-Mar-07
[email protected]
13
Lessons learned: Quality

QA










Develop and distribute fibre-test tools which allow on-line channel
quality testing, providing immediate feedback during construction.
Ensure QA is for identical conditions to final system
Test system with all final production components before launching
production
Do more longer term testing
Improve traceability of production lots
Respect ESD precautions
Excellent overall quality
Built in redundancy helps mitigate failures
Do not relax QA
Do not compromise ruggedness
20-Mar-07
[email protected]
14
WGa: Lessons learned and to be learned
conclusions

Excellent progress






Quality to be regularly reviewed
System performance to be monitored
Technology lessons still to be written
Full document ready for TWEPP07 Prague?



Technical
Social
Joint ATLAS CMS note
Results are immediately useful
ATLAS and CMS not vastly different, with complementary
experience.
20-Mar-07
[email protected]
15
Working Group B: Radiation Hardness



J.M. Brom (Strasbourg)
K. Gill (CERN)
C. Issever (Oxford) plus
ATLAS institutes
participating to SLHC
radiation test programme




20-Mar-07
[email protected]
Define irradiation test
Evaluate devices thoroughly
 How many?
 Selection criteria?
 Test parameters?
 Annealing?
 Pass/Fail criteria ?
Survey radiation sources
 Which set of beam types
and energies ?
Ageing tests?
16
Radiation hardness

How do we manage all this work with so few people?




Define a minimal-sufficient set of tests required
Define and agree on common protocoles and procedures
select one or a limited number of baseline system options and
then concentrate on testing these components
Attempt to share the load and the results

Next subgroup meeting to be called by C. Issever

In the meantime, 5 independent irradiation tests planned
20-Mar-07
[email protected]
17
Working Group C: Common Reference bench








J-M Brom (In2P3 Strasbourg)
B. Checcucci (INFN Perugia)
J. Coughlan (RAL)
K.K. Gan (UofOhio)
C. Hill (Bristol)
T. Romanteau (In2P3
Polytechnique)
R. Russack (UofMinnesota), D.
Baden (UofMaryland)
J. Ye
(SouthernMethodistUniversity)
20-Mar-07

[email protected]
Define
 Required performance
 Evaluation criteria
 Test procedures
 Test setup
 Support requirements
18
serializer
TP1
data
control
De-serializer
SER
data
TP2
TP3
DES
control
clk


clk
Next subgroup meeting to be called by J. Ye
How many independent setups built in the meantime?
20-Mar-07
[email protected]
19
Two development projects
Joint Optoelectronics Working Group
Resources Distribution
20-Mar-07
5
CERN
11.4
EU
US
6.6
Series1
CERN
Ljubljana
Minnesota
Ohio
Oklahoma
Oxford
Polytechnique
RAL
SMU
Strasbourg
8
7
6
5
FTE 4
3
2
1
0
[email protected]
10.2
ATLAS
12.8
20
CMS
Silicon-On-Sapphire (SOS) 0.25 m Technology Evaluation
J. Ye et al., SMU
 Peregrine 0.25 m SOS has been evaluated and is found to be
suitable for rad-hard ASIC development for LHC upgrade. No
special layout technique (ELT, guard ring) is needed.
 Technology Advantages:
 Low power, low cross talk, good for mixed signal ASIC designs.
 Economical for small to medium scale ASIC development. A multiproject run of 3×3 mm2 for 100+ chips cost ~$35k.
 TID on transistors with gamma (Co-60):
 With a floating substrate and up to 4 Mrad, leakage current
increase is observed but saturates after ~ 100 krad. This increase
anneals back at room temperature in a few weeks.
 With a grounded substrate during irradiation, there is no
measurable leakage current and threshold voltage change in both
NMOS and PMOS.
 The mechanism of the grounded substrate is under study.
 SEE using shift registers with 230 MeV proton:
 No SEE was observed with flux < 1×109 proton/cm2/sec.
 All shift registers function error free after a total fluence of
1.9×1013 proton/cm2 and ionizing dose of 106 Mrad (Si).
March 20 2007
21
Link-On-Chip (LOC) 1st Prototype
J. Ye et al., SMU
 Status:
 A 2.5 Gbps LOC1 prototype chip
has been designed and submitted
for fabrication. Chips will be
back to SMU for tests and
evaluations in mid May.
 If the tests (in-lab and
irradiation) for LOC1 are
successful, LOC2, which is aimed
at a complete serializer chip of
2.5 to 3.125 Gbps data rate, will
be designed and submitted in
Spring to Summer of 2008.
 Fiber attachment schemes will
be studied after the functional
tests of LOC1.
March 20 2007
LOC1
22
GOL Test up to 100 Mrad
J. Ye et al., SMU
 Test with 230 MeV proton beam
 TID: survived 106 Mrad (Si) without current increase. Chip fully
functioning during and after irradiation.
 SEE: no error when flux < 1×109 proton/cm2/sec. When flux = 5×1011
proton/cm2/sec, error cross section is measured to be 1.1×10-13
error·cm2/proton (loss of link) and 1.1×10-14 error·cm2/proton (bit
error).
 Jitter:
Complies with the modified (1.6Gbps vs 1.25 Gbps) IEEE Gigabit
Ethernet standards.
Before irradiation
After irradiation
Jitter Components
Tx clk
Serial Data
Tx clk
Serial Data
Random (RMS)
10.2ps
4.6ps
11.1ps
4.7ps
Deterministic
(Pk-Pk)
67.6ps
55.6ps
67.0ps
57.9ps
Total@BER-14
196.1ps
106.7ps
211.7ps
111.8ps
March 20 2007
23
Versatile link project: P2P architectures
F. Vasey et al. CERN
Module
1
Downstream
Front-End
Back-End
Upstream
TTC
EC
TTC
EC
DAQ
Module
2
Module
3
protocol eTRx
oRx
oTx
oTx
oRx
eTRx protocol
Module
4
Module
n
March 20 2007
24
Versatile link project: PON architectures
F. Vasey et al. CERN
Module
1
Module
2
Module
3
eTRx
protocol eTRx
protocol eTRx
protocol eTRx
March 20 2007
oRx
oTx
oRx
oTx
oRx
splitter
splitter
oTx
oRx
eTRx protocol
oTx
oRx
oTx
25
Versatile link project: the best of both worlds?
F. Vasey et al. CERN
protocol eTRx
oRx
oRx
oTx
splitter
protocol eTRx
protocol eTRx
protocol eTRx
March 20 2007
oTx
TTC
FPGA
oRx
oRx
oRx
oRx
DAQ
FPGA
oRx
oTx
oRx
oTx
oRx
oTx
In detector
26
Building blocks
ld
eTRx
Private
protocol
FPGA
implementation
GBT
tia
Versatile Link
oRx
oTx
splitter
oRx
oRx
oRx
oRx
oRx
oTx
March 20 2007
27
Conclusion


Joint ATLAS CMS SLHC Optoelectronics WG
exists and is well alive
3 subgroups defined




WGa is already producing useful results
WGb and WGc are on the right track, but still need to
get organized to reach cruising speed.
More results at TWEPP07
Independent projects and tests are running in
parallel to common effort.

Keep diversity in common test framework
20-Mar-07
[email protected]
28