Centre de rencontre de Cartigny D. Ferrère, 11 June 2008 On behalf of the University of Geneva.

ATLAS UpgradeATLAS Upgrade

Centre de rencontre de CartignyD. Ferrère, 11 June 2008

On behalf of the University of Geneva

Possible Machine ScenariosPossible Machine Scenarios

3 phases considered:

Phase 0: Push the machine to its maximum performance without hardware changeLuminosity of 3x1034 cm-2s-1. and ultimate energy 7.54 TeV (dipole field @ 9T)

Phase 1 (SLHC): Interaction Region (IR) quadrupoles life expectancy < 10 years Modify the insertion quadrupoles and their layout *

Phase 2: Rebuild the SPS with superconducting magnets, the transfer line to inject into LHC at 1 TeV and new 15T dipoles proton energy of 12.5 TeV

ParametersLHC SLHC - Phase1

Nominal Ultimate Scenario 1 Scenario 2

Bunch spacing [ns] 25 25 50 25

Proton/bunch Nb[1011] 1.15 1.7 4.9 1.7

* at IP1&5 [m] 0.55 0.5 0.25 0.08

Longitudinal profile Gaussian Gaussian Flat Gaussian

Rms bunch length z[cm] 7.55 7.55 11.8 7.55

Peak luminosity [1034 cm-2s-1] 1 2-3 10.7 15.5

Effective luminosity (5h) [1034 cm-2s-1] 0.56 1.15 3.5 3.6

Peak events per crossing 19 44 403 294

CommentsWire comp. Crab + D0

(+ Q0)

NB: Original scenario of 12.5 ns excluded since exceed the max local cooling capacity

Ba

se

lin

e

Alt

ern

ati

ve

Far future

Cartigny, June 11th 2008 ATLAS Upgrade, Didier 2

20

14

-15

20

11

-12

Possible sLHC SchedulePossible sLHC Schedule


Fro

m N

igel

Hes

sey

– U

pgr

ade

Ste

erin

g G

roup

Rough ATLAS Upgrade ScheduleRough ATLAS Upgrade Schedule


Other factors:• CMS do not believe they can upgrade their inner tracker to run before 2018 • LHC expects new injectors (new PS replacement) ready for full sLHC in 2017 (so no data in 2016 would get us ready in time)• ATLAS could need to install in 2015 if LHC starts up fast and reaches “Ultimate” 2-3.1034

• Clearly we should only have one major shutdown• Propose discussion within LHCC to agree which year is the major shutdown

Very rough outline for ID Upgrade:• Start 2007: 3 years R&D 2007-2009• TDR April 2010• Other documents 2010; design work; final prototypes• PRR's, start procurement 2011• 2 years building modules (and continuous parts procurement) 2012-2013• Surface assembly and test 2013-2014• Stop LHC end 2014• Install 2015• Take data April 2016

ATLAS Upgrade - DetectorATLAS Upgrade - Detector


Due to increase of radiation level, pile-up, background Due to increase of radiation level, pile-up, background each sub-detectors has to think of the consequences in each sub-detectors has to think of the consequences in term of detector performance, aging, radiation hardness!term of detector performance, aging, radiation hardness!

Inner Detector (ID):Completely new design and detectors – No TRT, more pixel and strips: New detector and ASICs technologies to withstand the radiation level Simulations drive optimal geometry (Strawman layers) and occupancies

LAr Calorimeter:FCAL: To be replaced with smaller gap size and special cooling close to the beam pipe.HEC: Cold electronics may have to be replaced (no need to get HEC wheels apart)Others: More R&D require about the Ion buid-up

Tile Calorimeter:Front-end electronics and Low Voltage Power Supplies

Muon System:MDT: Chambers and readout electronics are radhard at LHC or can stand 10 times the max nominal flux but further studies need to be performed in SLHC environment

2008: detail upgrade plan based on some tests and R&DTGC: Thick GEMs or MM can possibly replace forward chambers due to high occupancy.

Focus


SLHC - ChallengesSLHC - Challenges

ATLAS Upgrade – OrganizationATLAS Upgrade – Organization


ATLAS Upgrade – POATLAS Upgrade – PO


Module Integration – SLHC FP7-PPModule Integration – SLHC FP7-PP



Inner Detector Upgrade - LayoutInner Detector Upgrade - Layout

The goal for b-layer replacement is fall 2012Actual b-layer is expected to survive 3 years at LHC design luminosity

(~300fb-1)The Upgrade of the entire tracker should take place in 2016

Strawman 06Including disks this leads to:

Pixels: 5 m2, ~300,000,000 channels

Short strips: 60 m2, ~30,000,000 channels

Long strips: 100 m2, ~15,000,000 channels

1.6%

1.2%

Short and Long Strip Occupancy

0.8%


Inner Detector Upgrade – Layout cont’dInner Detector Upgrade – Layout cont’d

Strawman 07

Nb of hits into the ID volumeStrawman06 vs Strawman07

A new version is about to come with single length barrel for Short- and Long-Strips…


Effect of Tilt AngleEffect of Tilt Angle

Timothy Greenshaw (Liverpool) – Presented a MIWG 05.06.2008

Some simulations have started and comparisons with ATLAS Pixel data were done

ATLAS – ID DetectorATLAS – ID Detector


https://twiki.cern.ch/twiki/bin/view/Atlas/AtlasTechnicalPaper

Next time, lets not just get hung up over material at η=0. This is supposed to be a tracker up to |η|≤2.5


ATLAS - SCTATLAS - SCT

A piece of art in term of services and engineering!

Barrel modules were mounted with a robot not the endcap!


Radiation Background in ATLAS at SLHCRadiation Background in ATLAS at SLHC

1 MeV equivalent neutron fluences assuming an integrated luminosity of 3000fb-1 and 5cm of moderator lining the calorimeters (reduces fluences by ~25%)

•With safety factor of two, design short microstrip layers to withstand 1015neq/cm2 (50% neutrons)

•Outer layers up to 4×1014neq/cm2 (and mostly neutrons)

Thermal management and shot noise. Silicon looks to need to be at ~ -25oC (Thermal runaway).Si power: 1W @ -20°C

4W @ -10°C10W @ 0°C

High levels of activation will require careful consideration for access and maintenance.

IssuesIssues

Simulation using FLUKA2006


Silicon Sensors Technology versus RadiationSilicon Sensors Technology versus Radiation

Choice of adequate material thanks to RD50 collaboration

Pixel and Strips: n-in-p (planar technology)

No type inversion, full depletion is on structured side

Collection of electrons (faster than p-in-n)

0 20 40 60 80 100 120eq [1014 cm-2]

0

5000

10000

15000

20000

25000

sign

al [e

lect

rons

]

3D FZ Si, 235 m, (laser injection, scaled!), pad [Da Via 2006]n-FZ Si, 280 m, (-10oC, 40ns), n-in-n pixel [Rohe et al. 2005]p-FZ Si, 280 m, (-30oC, 25ns), strip [Casse 2004]p-MCZ Si, 300 m, (-30oC, s), pad [Bruzzi 2006]n-epi Si, 150 m, (-30oC, 25ns), pad [Kramberger 2006]n-epi Si, 75 m, (-30oC, 25ns), pad [Kramberger 2006]sCVD-Diamond, 770m, (RT, s), [RD42 2006] (preliminary data, scaled)pCVD-Diamond, 500m, (RT, s), strip, [RD42 2002-2006] (scaled)SiC, n-type, 55 m, (RT, 2.5s), pad [Moscatelli et al. 2006]

M.Moll 2007

Pixel b-layer: 3D technology is an option(should be ready for b-layer replacement in 20012)

electrons swept away by transversal field holes drift in

central region and diffuse towards p+

contact

Exist in single and double column type

n+-columns

ionizing particle

p-Si

3D FZ 235m

P-FZ 280m

Still ~15000e- at 1.1015 cm-2s-1 (FZ or MCZ)


Short Strip Sensors - HPKShort Strip Sensors - HPK

9.75cm

9.75

cm

• Dimension– Full square

• Wafer– 150 mm p-type

FZ(100)– 138 mm dia. usable– 320 µm thick

• Axial strips– 74.5 µm pitch

• Stereo strips– 40 mrad– 71.5 µm pitch

• Bond pads location– accommodating 24-

40 mm distances• n-strip isolation

– P-stop– Spray on

miniatures Y. Unno

•Strip segments– 4 rows of 2.38 cm strips (each row

1280 channels)

130 detectors ordered end of last year (15 for UniGe)!


First Engineering LayoutFirst Engineering Layout

Knowing the detector size and module and the layout…


FE Readout for Strip UpgradeFE Readout for Strip Upgrade

ABC-N Readout architecture

Minimum requirements:

Implementation of an on-chip shunt regulation to allow any powering scheme (Serial or DC-DC)

Increased data bandwidth of 160 Mb/s

Compatibility with existing ATLAS SCT DAQ ROD hardware

Front-end for two signal polarities, 3 to 12 cm long silicon strips

Minimum power circuit techniques

Delay adjustment for control and data signals

Similar “core” functionality as for the present silicon strip tracker: signal amplification, discriminator, binary data storage for L1 latency, readout buffer with compression logic, and data serializer.

Goal: Prepare a design of the FE ASIC in deep submicron radiation tolerant.Design and evaluation with 250 nm towards 130 nm technology


FE Readout for Strip UpgradeFE Readout for Strip Upgrade

• Daniel La Marra, Sebastien Pernecker, Geneva University• Wladek Dabrowski, Krzysztof Swientek, AGH-Cracow• Jan Kaplon, Karolina Poltorak, Francis Anghinolfi, CERN• Mitch Newcomer, Pennsylvania University

Design team:

Front-End Opt. short strip doneLayout started

33mA/chip 750enc (2.5cm strips)Final S/N > 10

Back-End Main change in DCL block to handle 160MHz

92mA/chip at 2.5V estimated

Powering Integrated shunt regulators possible

Current limits to impose uniformity

Floor Plan First Checks now 7.5mm by 6±1mm

P&R Examples with pipeline and derandomizer OK

Submission Summer 2008 Deliver Autumn 2008


A Break…A Break…

What do we have clear today?- Silicon sensors n-in-p SS and LS OK- FE chips Proceed with 0.25 m technology then move this year to 0.13 m

What are still unclear?- Layout: Hopefully a baseline will satisfy soon most of us!- Cooling: C3F8 or CO2… Consequences are huge for the design- Powering with 4 options: Serial Powering + 3 flavors for the DC-DC- Modules integrations with 4 options: Stave + 3 module flavours (Just reviewed)- Endcap is not clearly defined in a modular or stave concept- Services: Keep or replace the services between PP1 and PP2

The engineering is requiring to work soon on one viable scenario!The engineering is requiring to work soon on one viable scenario!


CoolingCooling

Cooling System is one of the key points for ID operationHeat density is 4-7 times higher than the current SCT module

Detector thermal runaway impose to operate the silicon wafer below -20ºCTsensor = Tcoolant + THeat_Path

Coolant: Less than -30 C is considered for the pipe temperature Module design and choice of material is critical – Thermal FEA

2 candidates for the cooling:• CO2 (~100 bar)• C3F8 (Medium-Low pressure)

1g of evaporation @ -30ºC ~280 J(Where for C3F8 it is 100 J/g)

CO2Enthalpy [kJ/kg]

Pre

ss

ure

[b

ar]

Existing pipes are considered:-SCT & Pixel (ok for C2F6, C3F8)-TRT (ok for CO2 but limited number)


Cooling – cont’dCooling – cont’d

ID

Calorimeter

Inner muon

Middle muon

Access

Conceptual, not to scale

• HEX is located at PP2– Ex-TRT space available (+more?

Other φ?),– Needs to be thermally neutral.

• Plant stage – uses existing transfer network

underneath inner muon system.– Throttling, back pressure regulation

and heater close to HEX → accessible and “moderate” radiation.

• Detector stage – uses new, insulated pipes from PP2

along calorimeter endplate to ID.– Pump close to PP2 → accessible

and “moderate” radiation, but in magnetic field

– No heaters or further HEXs needed– If we need throttle, it should also be

located close to HEX for access, if the control allows.

A very preliminar proposal for CO2

Georg Vihausser – Presented at Thermal Management WG 10.06.2008


PoweringPowering

Serial powering

BNL, Bonn, Cracow, Penn, Prague, RAL, Wuppertal, (FNAL)

DC-DC buck converter

BNL, CERN, Yale, (PSI)

DC-DC charge pump

LBNL

Minimize current through cables by a) “recycling” current (SP) or b) “high-voltage” power lines (DC-DC)

Piezo transformer

Masatoshi Imori, KEK and industry

From Marc Weber (RAL)


Element of Power SystemsElement of Power Systems

IP: independent powering; SP: Serial powering; PP: parallel powering

Off-detector power supplyConstant-current source;

HV supply

Cables and power distribution scheme:

IP; SP; PP

On-detector power supplies:

regulators; converters;transformers

Monitor + control system

DCS; protection; by-pass

On-detector supplies and understanding cables is most urgent, but we need all element!

It is also time to see how elements would come together in a system.


Proposed DCS ArchitectureProposed DCS Architecture

SCTDAQ

PS Crate

Cooling Temp

Environmental (Super-Module)

Hybrid Temp

Hybrid Power

SMC Hybrid

BBIM

Ibox

Matrix

Elmb

CAN Bus

CAN Bus

Global InterlockROD

BOC

5 GB/s optical linkData/DCS

Channel Interlock

LAN

DCS Type 2 cables Env.

Structure

DetectorPS Type 2 Type 4 cables

SCTDCS

Elmb

CAN Bus

SPI bus daisy chain

Option 2 Option 1

DCS link

D. Ferrere – PO April 08


Module Integration - StaveModule Integration - Stave

• ATLAS specific design: 10, 10x10 cm sensors/side, doubled sided measurement• Embedded cooling and bus cable• Thickness and mass (Xo) depend upon choice of coolant, hybrid technology• Complete and inclusive except for barrel support cylinder and fixation points• Straightforward generalization to outer barrel modules• Baseline design has hybrids glued directly on sensors and small gaps in Z• Alternatives have been studied which include bridges and closed gaps

Bus cable

Hybrids Coolant tube structure

Carbon honeycomb or foam

Carbon fiberfacing

Readout IC’s

Silicon sensors

Material budget announced to be relatively light: ~ 2.2 % for CO2~ 2.7 % for C3F8

(Only for th

e Stave)


30 modules with ABCD & serial power(ABCnext will be 30 V)

2 adjacent modules glued on a stave

Output noise

Input Noise vs Strip Length

0

200

400

600

800

1000

1200

1400

1600

0 2 4 6 8 10 12 14

Strip Length [cm]

Electrons

Unbonded

Stave 2007 Stave 2006

SCT

Comparison

Module Integration – Stave07Module Integration – Stave07

Module Integration – Module ConceptsModule Integration – Module Concepts


• Fundamental principle– Maximum Repairability and Replaceabiltiy– Separated function

• Modularity– Double-sided module

• 2D information in-situ - effective for track/hit finding – Minimum separation of axial and stereo sensors

• Identical objects- effective for alignment– Above two are best realized with precision alignment of front-

back sensors• Repairable and replaceable unit

– Separated function of modules and service structures• Fabrications, QA's, Repairs, ..., in parallel• thus, minimizing risks and costs

Module strongly inspired from SCT barrel.

Module Integration – Super-Module JapanModule Integration – Super-Module Japan


2m long superframe for 20 modules

Module Integration – Super-Module JapanModule Integration – Super-Module Japan

Circumferential mounting with support jigs for 2 m long SM


Module Integration – Local Support UniGeModule Integration – Local Support UniGe


Individual module mountingFeb 07

10 Module structure

Apr 07

With longitudinal insertion mechanism

Jul 07Local Support Row (LSR)

Dec 07 Local Support with anchorage points (Valencia)

Jan 08 Simplified Local Support

Fabrication aspect taken into account

Prototype program en route

DemonstratorMay 08

June 08 Review



Stiff and self-handling structure.Module and barrel fixation precision as assembled!

Local support on the jig for the assembly



Identified advantages of the end-insertion:Identified advantages of the end-insertion:• Flexibility: Barrel structures can be assembled before the SMs are integrated Flexibility of the assembly procedure• Time saving: No need to test a fully populated barrel after another and then integrate the 5 barrel layers Time saving about ~1 year (2-3 months per layer)• Parallel work: Assembly of the structure can be done in parallel with the SM/Stave production. Time saving can be considered!• Rework: The rework or replacement of a SM could occur even late (even after the commissioning). But when final services installed access is difficult!• SM/Stave replacement could happen in any working area (few light jigs).• It allows the use of either a single skin barrel cylinder with reinforcing rings or a CF sandwich • It allows double layer structure BUT … (see next slides)• Geometrical survey of the pre-assembled structures with and without SM/Stave.



Si-waferTmax = -22.3 oC

ASIC’sTmax = -14.5 oC

-30

-25

-20

-15

-10

-5

0

0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50

Q (Si) T = 0 oC ( mW/mm2 )

Tem

per

atu

re (

oC

)

with HTC -27 C/ -30 C -27 C / 30 C MCC

Excellent performance with respect to the thermal runaway

sLHC

K.P. Streit Preview 05.06.2008

Module Integration – Wing Support UKModule Integration – Wing Support UK


• Space available to services is high making design insensitive to exact powering and data scheme

– In particular space for stubs and transition boards required for fan-out and cross-over of traces

• No additional service supports and/or protection needed.

• Services not cantilevered avoiding possibility of stresses being applied to modules

• Services do not dictate assembly order

• Notes on rendering– Extreme number of connectors

shown on module – Exact shape of tapes to be

driven by electrical layout (in progress)

– Both top and bottom of module plug in to backbone

SMC envelope

Tape backbone

Module connectors

Module connectors

Pipe not sliding and constrained into cooling blocksAlso end-insertion proposed…

I. Wilmut Review 06.06.2008

Module Integration – EndcapModule Integration – Endcap


C. Lacasta Review 06.06.2008

Module Integration – Endcap Cont’dModule Integration – Endcap Cont’d


C. Lacasta Review 06.06.2008

Module Integration – Review CloseoutModule Integration – Review Closeout


• The stave appears to have a number of advantages and is the most advanced development. It should be the collaborations priority to complete this understanding by setting up collaboration-wide subgroups to tackle critical issues (see p6 but start to complete this list and set up groups today).

• The second priority should be to accelerate the development of an EC solution. The hope is that the stave concept can be morphed into a petal or ring and benefit from the barrel development.

• There are open issues with the stave which might result in a fall-back. Concentrating all effort on the stave is not risk-free. Revisit progress at Nikhef.

• Review committee will circulate a detailed list of concerns in a few days

Recommendation summaryRecommendation summary

M. Tyndel Review 07.06.2008

Module Integration – EngineeringModule Integration – Engineering


1 single ID element Integrated and commissioned at the surface building

A. Catinaccio Review 07.06.2008

Module Integration – Engineering & ServicesModule Integration – Engineering & Services


Possible connector implementation on the SMC!

Only 60 mm space between interlinks!

D. Ferrere Preview 05.06.2008

ATLAS Strip Upgrade - UniGe InvolvementsATLAS Strip Upgrade - UniGe Involvements


Asics: Digital Architecture & Simulation: pipeline, Derandomizer, Data Compression Logic, Readout Logic and Control; Check after place and route!

250nm submission January 2008 130nm design from January 2008

Detectors: 135 sensors of size10X10cm2 ordered to HamamatsuUniGe ordered 15 sensors: 12 p-stop and 3 p-spray Sensor characterization will be made and compared with other institutes

Module Concept: UniGe investigated 2 concepts module directly mounted on barrel or mounted on an intermediate local support!

UniGe operates 3D Thermal FEA to evaluate and optimize the designStrategy is to make prototype modules in 2008 with above detectors and Asics

DCS: Involvement in the strategy and decision for the detector safety. DCS will be part of the system design to optimize the service and resource issues.

Mechanical Engineering: Involvement in the structure and service design.

Few ConclusionsFew Conclusions


SLHC is clearly in the continuity of the machine and detector operation Phase 1 will be to increase the luminosity up to a factor 10 (1035 cm-2s-1 ) ATLAS and CMS are considering options to Upgrade the detectors in 2016 The ATLAS Inner Tracker will be completely renewed (no TRT) and is in a design and specification phase The B-layer replacement is foreseen already in 2012 using 3-D sensors The major challenges are: Service and material reductions, Cooling, Powering scheme The time scale is short and there is almost no time for R&D Working groups start smoothly to grow and to be active UniGe involved in various topics and keep an eye on the B-layer replacement activities Next event:

•ID Upgrade Workshop Nov 2008 at Nikhef

Centre de rencontre de Cartigny D. Ferrère, 11 June 2008 On behalf of the University of Geneva.

Documents

Transcript of Centre de rencontre de Cartigny D. Ferrère, 11 June 2008 On behalf of the University of Geneva.