63
11-Nov-10 Pac Mtg - Unv of Oregon Global Design Effort 1 Barry Barish PAC Meeting University of Oregon 11-Nov-10 GDE Director’s Report Grooved Insert for CesrTA Wiggler

Barry Barish PAC Meeting University of Oregon 11-Nov-10

  • Upload
    rea

  • View
    23

  • Download
    0

Embed Size (px)

DESCRIPTION

GDE Director’s Report. Grooved Insert for CesrTA Wiggler. Barry Barish PAC Meeting University of Oregon 11-Nov-10. Outline. Status of R&D and design Plans through 2012 Technical Design Report Implementation Plan Future beyond 2012. Our Plan --- Updated “Living” Document. - PowerPoint PPT Presentation

Citation preview

Page 1: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 1

Barry BarishPAC Meeting

University of Oregon 11-Nov-10

GDE Director’s Report

Grooved Insert forCesrTA Wiggler

Page 2: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 2

Outline

• Status of R&D and design

• Plans through 2012– Technical Design Report– Implementation Plan

• Future beyond 2012

Page 3: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 3

Our Plan --- Updated “Living” Document

Page 4: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 4

R & D Plan Resource Table

• Resource total: 2009-2012

• Not directly included:– Other Project-specific, general infrastructure or

generic R&D resources overlap with ILC R&D

FTE SCRF CFS & Global AS TotalAmericas 243 28 121 392Asia 82 9 51 142Europe 108 17 64 189

433 55 236 724

MS (K$) SCRF CFS & Global AS TotalAmericas 18080 2993 6053 27126Asia 23260 171 5260 28691Europe 9890 921 530 11341Total 51231 4085 11843 67158

Page 5: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 5

R&D Progress

Page 6: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 6Global Design Effort

Major R&D Goals for Technical Design

SCRF• High Gradient R&D - globally coordinated program to

demonstrate gradient by 2010 with 50%yield; improve yield to 90% by TDR (end 2012)

• Manufacturing: plug compatible design; industrialization, etc• Future systems tests: NML (FNAL), STF2 (KEK)

Test Facilities• ATF2 - Fast Kicker tests and Final Focus design/performance• CesrTA - Electron Cloud tests to establish damping ring

parameters/design and electron cloud mitigation strategy• FLASH – Study performance using ILC-like beam and

cryomodule

M Ross for details

A Yamamoto for details

Page 7: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 7

SCRF Status/Progress

A Yamamoto for details

Page 8: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 8

The ILC SCRF Cavity

- Achieve high gradient (35MV/m); develop multiple vendors; make cost effective, etc

- Focus is on high gradient; production yields; cryogenic losses; radiation; system performance

Page 9: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 9

Creation of a Global Database

• Global Data Base Team formed:– Camille Ginsburg (Fermilab) – Rongli Geng (JLab) – Zack Conway (Cornell University)– Sebastian Aderhold (DESY)– Yasuchika Yamamoto (KEK)

• Activities – July 2009:

- Determine DESY-DB to be viable option, – Sept., 2009: (ALCPG/GDE)

- Dataset, web-based, support by FNAL/DESY,– Dec., 2009:

- 1st update of the yield statistics– March, 2010

- 2nd update- July, 2010

Page 10: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 10

Global Plan for SRF R&D

Year 07

2008 2009 2010 2011 2012

Phase TDP-1 TDP-2

Cavity Gradient in v. test

to reach 35 MV/m Yield 50%

Yield 90%

Cavity-string to reach 31.5 MV/m, with one-cryomodule

Global effort for string assembly and test

(DESY, FNAL, INFN, KEK)

System Test with beam

acceleration

FLASH (DESY) , NML (FNAL)

STF2 (KEK, test start in 2013)

Preparation for Industrialization

Production Technology R&D

Page 11: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 11

Cavity Gradient Milestone Achieved

2010Milestone

TDRGoal

Page 12: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 12

S1-Global Cryomodule Test in ProgressDESY, FNAL, IHEP, INFN, KEK, SLAC Cooperation

Vertical cavity test • CW low power test reached: < 30 MV/m >

S1-Global cryomodule • 1ms, 5 Hz pulse Individual test reaching: < 28 MV/m >

{as of Oct. 22, 2010}

Page 13: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 13

NML Cryomodule

NML CM1 cryomodule (Fermilab, DESY, INFN).

Closed and cool down is imminent.

Page 14: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 14

Test Facilities: FLASH

M Ross for details

Page 15: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 15

Page 16: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 16

Energy stability over 8hrs(3mA, 800us bunch trains)

Beam Energy

RF Vector Sums (Normalized)

Time (hrs)

2MeV(0.25%)

0.2%

844 MeV

Nominal

8 hrs

Time (hrs)

Tuning change (Spec: +/-0.1%)

16

J. Carwardine

Page 17: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 17

Test Facilities: ATF-2

M Ross for details

Page 18: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 18

ATF2 – Beam size/stability and kicker tests

IP Shintake Monitor

Final Doublet

Page 19: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 19

ATF2 (KEK) Status/PlansT. Tauchi

Page 20: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 20

Test Facilities: Cesr-TA eCloud

M Ross for details

Page 21: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 21

• Mitigating Electron Cloud

• Simulations – electrodes; coating and/or grooving vacuum pipe• Demonstration at CESR critical tests

eCloud R&D

Page 22: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 22

Mitigation - Simulation Studies

Page 23: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 23

Page 24: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 24

CesrTA - Wiggler Observations

IWLC2010 - CERN, Geneva, Switzerland

0.002”radiusElectrode best performance

Page 25: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 25

CESRTA - eCloud

• Mitigation performance:– Grooves are effective in dipole/wiggler fields, but challenging to make when size is

small– Amorphous C, TiN and NEG show similar levels of EC suppression so each is a

potential candidate for DR use• TiN and a-C have worse dP/dI than Al chambers at our present level of processing

• In regions where TiN-coated chambers are struck by wiggler radiation (high intensity and high Ec), we observe significant concentrations of N in the vacuum system

– EC suppression with the clearing electrode in the wiggler is significantly better than other options

• No heating issues have been observed with the wiggler design in either CESRTA or CHESS operating conditions

– Work is in progress to take RFA measurements in chambers with mitigations and convert these to the effective SEY of the chamber surfaces

• Agreement between data and simulation looks very promising

• Magnetic field region model requires full inclusion of RFA in simulation

– Trapping and build-up of the EC over multiple turns in quadrupole and wiggler chambers

• Simulation and experimental evidence

• Further evaluation of impact on the beam is required

M Palmer

Page 26: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 26

Design Update

Global Design and Decision Making

(SB2009)

N Walker for details

Page 27: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 27

Why change from RDR design?

• Timescale of ILC demands we continually update the technologies and evolve the design to be prepared to build the most forward looking machine at the time of construction.

• Our next big milestone – the technical design (TDR) at end of 2012 should be as much as possible a “construction project ready” design with crucial R&D demonstrations complete and design optimised for performance to cost to risk.

• Cost containment vs RDR costs is a crucial element. (Must identify costs savings that will compensate cost growth)

Page 28: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 28

Proposed Baseline Changes for TDR

Page 29: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 29

Proposed Design changes for TDR

RDR SB2009 • Single Tunnel for main linac

•Move positron source to end of linac ***

• Reduce number of bunches factor of two (lower power) **

• Reduce size of damping rings (3.2km)

• Integrate central region

•Single stage bunch compressor

Page 30: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 30

Top Level Change Control Themes

Page 31: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 31Global Design Effort 31

Top Level Change Control Process

keywords: open, transparent

31

Page 32: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 32Global Design Effort 32

TLCC Process

When Where What

BAW 1 Sept. 7-10, 2010

KEK 1. Accelerating Gradient2. Single Tunnel (HLRF)

BAW 2 Jan 18-21, 2011

SLAC 3. Reduced RF 4. e+ source location

Baseline Assessment Workshops• Face to face meetings• Open to all

stakeholders• Plenary

Physics and detector input / representation mandatory

Physics and detector input / representation mandatory

Walker

Page 33: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 33Global Design Effort 33

Formal Director Approval• Change evaluation

panel• Chaired by Director

TLCC Process• Final formal step (recommended by

AAP)

• Change Evaluation Panel– Chaired by director– Experts to evaluate impact on

performance, cost, schedule, risk

– F. Asiri, K. Buesser, J. Gao, P. Garbincius, T. Himel, K. Yokoya

• Decision by Director– Accepts – becomes baseline;

guidance in decision memo– Rejects – sent back for further

work with comments33

Page 34: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 34

Proposals Received

Page 35: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 35

• Critical technical challenge for one-tunnel option is the high level RF distribution.

• Two proposed solutions :

– Distributed RF Source (DRFS)• Small 750kW klystrons/modulators in tunnel• One klystron per four cavities• ~1880 klystrons per linac• Challenge is cost and reliability

– Klystron Cluster Scheme (KCS)• RDR-like 10 MW Klystrons/modulators on surface• Surface building & shafts every ~2 km• Challenge is novel high-powered RF components

(needs R&D)

– Backup: RDR-like single tunnel HLRF

35

Single TunnelHigh-Level RF Solution

Page 36: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 36

Distributed RF – Single Tunnel

Page 37: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 37

Klystron Cluster – Single Tunnel

Page 38: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 38

RDR-type RF – Single Tunnel

Page 39: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 39Global Design Effort 39

TLCC Process

Page 40: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 40Global Design Effort 40

TLCC Process

Page 41: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 41

Plans through 2012--------------

Technical Design Report (TDR)

Page 42: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 42Global Design Effort 42

Five Themes to Develop

>2012

Remains

special

case

Project Implementation

Plan

Industrialisation

in-kind contribution

models

Site requirements

Project Schedule

Remaining Technical activities

N Walker

Page 43: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 43

Technical Design Phase and Beyond

AD&I studies

2009 2010

RDR ACD concepts

R&D Demonstrations

TDP Baseline Technical Design

2011 2012 2013

RDR Baseline

Be

ijing

Wo

rks

ho

p

TDR

TDP-1 TDP-2ChangeRequest

SB2009 evolve

change control processAAPPACPhysics

CE

RN

Wo

rks

ho

p

Page 44: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 44

• Timescale: Produce final reports end of 2012– Technical Design (TDR)– Project Implementation (PIP)

• First goal: Technical Design (TDR)– SCRF – S0 gradient; S1 Global Tests continue past 2012– Detailed technical design studies from new baseline – Updated VALUE estimate and schedule. – Remaining critical R&D and technology demonstration

(CesrTA complete; ATF-2; FLASH; etc)

• Second Goal: Project Implementation Plan (PIP)– Studies of governance; siting solicitation and site

preparations; manufacturing; etc

Technical Design Phase 2

Page 45: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 45

Essential Elements of TDP

• Optimize the design for cost / performance / risk– Top down approach got SB2009; value engineering; risk

mitigation

• Key Supporting R&D Program (priorities)– High Gradient R&D - globally coordinated program to

demonstrate gradient for TDR by 2010 with 90%yield– Electron Cloud Mitigation – Electron Cloud tests at Cornell

to establish mitigation; determine mitigtion plan; verify one damping ring is sufficient and size.

– Final Beam Optics – Tests at ATF-2 at KEK

• GOAL – Bring us ready to propose a solid and defendable “construction project” to world’s governments any time after 2012

Page 46: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 46

Project Implementation Plan

Page 47: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 47

Governance – Interim Reportpresented to FALC& ILCSC and at IWLC10

Page 48: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 48

Governance – Interim Reportexample – approach to “in-kind” contributions

Page 49: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 49

Page 50: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 50

Presently – Outline Stage

Recent GDE Executive CommitteeMeeting during IWLC10

Page 51: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 51

(Global) Mass Production (SCRF)

• Critically important TDP2 activity

• Learn from XFEL experience• 5% ILC

• Learn from LHC experience

• Develop realistic models on which to base cost estimate

• With industry

TLCC-1

Global S1TDP-2

PIPGovernance

Akira

Page 52: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 52

19m x 14m ISO class-5 clean room

Chemical Polish room

KEK Industrial R&D Pilot Plant

Press machine

Triming machine

Electron Beam Welder

work together with industry to develop cost-effective cavity production techniques

Page 53: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 53

Page 54: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 54

ILC: Future after 2012

Page 55: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 55

Why TDR in 2012 ?

• The R&D program is effectively ballistic at this point and slowing it down is not cost effective.

• Synergy with the CLIC CDR (in 2011).

• Available for the European Strategy for Particle Physics (2012) consideration.

• Ready on a time scale consistent with the first LHC physics results.

• TDR can serves as a basis for value engineering and industrial tech transfer programs.

• Project Implementation Plan available for potential hosts

Page 56: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 56

CY

Technical Design Report completeBaseline established

2011 2012 2013 2014 20162010

ILC

2015

Technical design & R&D program

2017 2018

ILC possible timeline

SRF system tests

TDR reviews

Site EOI’s

Cost Estimating

Decision to proceed

Site/host established

Project Implementation Plan complete XFEL operation

LHC

Physics Run 1 Physics Run 2Interconnect repair

Existence of low-lying SUSY known

Higgs energy scale known

timegap

Page 57: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 57

CY

Technical Design Report completeBaseline established

2011 2012 2013 2014 20162010

ILC

2015

Technical design & R&D program

2017 2018

ILC possible timeline

SRF system tests

TDR reviews

Site EOI’s

Cost Estimating

Decision to proceed

Site/host established

Project Implementation Plan complete XFEL operation

LHC

Physics Run 1 Physics Run 2Interconnect repair

Existence of low-lying SUSY known

Higgs energy scale known

timegap

Page 58: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 58

Pre-project ILC Organization

• GDE will have successfully completed its mandate after TDR + reviews (mid-2013 at latest)

• ILCSC / ICFA considering transition organization

http://cpdg.kek.jp (cpdg username and password).

Page 59: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 59

GDE - Technically-driven post 2012 program

• SCRF Systems tests; Mass production; Value Engineering, etc.

• Design evolution: 1 TeV; Positrons; R&D toward major technical advances

• Must preserve GDE-like global decision making and coordination in new pre-project organization

Page 60: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 60

From Technical Design Report to ILC(or beyond 2012)

• Steps to a Project – Technical (2-3 years)– R&D for Risk Reduction and Technology Improvement– Engineering Design– Industrialization

• Project Implementation– Government Agreements for International Partnership– Siting and site dependent design– Governance

• Time to Construct– 5-6 years construction– 2 years commissioning

• Project Proposal / Decision keyed to LHC results

• ILC Could be doing physics by early to mid- 2020s

Page 61: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 61

Possible GDE program - post 2012Possible GDE program - post 2012

• As the TDR phase and the associated R&D program concludes then the technical elements of the program will be de-emphasized (CESR TA, electron source, HLRF, LLRF, cryomodule design, BDS design etc…….).

• We will switch to operating the systems test facilities that were fabricated as part of the R&D program e.g. NML, STF. The Fermilab SRF string test will be commissioned in 2012 but the regular facility operations will not start until FY13. XFEL commissioning can be considered a large ILC system test of sorts.

• We will continue to support beam delivery system development at the KEK test beam facility (ATF2). This of course is contingent on KEK deciding to continue to support ATF program past the currently approved JFY12.

Page 62: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 62

Possible GDE program - post 2012

• We will support a core team to maintain corporate knowledge and be available for TDR reviews

• We plan to keep the SRF industrial base active at a minimally useful level and engaged in pre-production (value) engineering.

• It’s likely that positron production will benefit from R&D past 2012.

• It is likely that machine-detector interface activities will need to continue. This will help to facilitate the detector program.

Page 63: Barry Barish PAC Meeting  University of Oregon    11-Nov-10

11-Nov-10 Pac Mtg - Unv of Oregon

Global Design Effort 63

Final Remarks

• ILC accelerator R&D and design evolution is on track for Technical Design Report at end of 2012. This will be accompanied a Project Implementation Plan

• We are broadly collaborating with CLIC, which should enable an informed decision in the future

• Planning for ILC development beyond 2012 is very important. It will be very difficult to maintain viable support until a decision will be made.

• LHC will open the TeV energy frontier and the resulting physics will point our way to the future of HEP.