linac4 status

20
1 Linac4 Status M. Vretenar, 30.10.2008

Upload: karyn-bowman

Post on 30-Dec-2015

49 views

Category:

Documents


0 download

DESCRIPTION

Linac4 Status. M. Vretenar, 30.10.2008. Linac4 on the CERN site. Linac4 will be built at the place of the “Mount Citron”, a small hill made in the 50’s from the excavation materials of the PS. Position and orientation allow a future extension to the SPL. PS. PSB. PS2 - SPS. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Linac4 Status

1

Linac4 Status

M. Vretenar, 30.10.2008

Page 2: Linac4 Status

2

PS2 - SPS

PS

PSB

Future SPL

Transfer line to Linac2

Linac4

Linac4 on the CERN site

Linac4 will be built at the place of the “Mount Citron”, a small hill made in the 50’s from the excavation materials of the PS.

Position and orientation allow a future extension to the SPL.

Page 3: Linac4 Status

3

Linac4 Groundbreaking

16 October 2008, Linac4 Groundbreaking by CERN Director, R. Aymar

Page 4: Linac4 Status

4

Linac4 Groundbreaking

Page 5: Linac4 Status

5

Linac4 Groundbreaking – The team

Page 6: Linac4 Status

6

Linac4 – Milestones

June 2007: approval of the project by the CERN Council. January 2008: official project start (budget available).

May 2007: Defined linac position and type of tunnel, start pre-integration. November 2007 – April 2008: Preparation of CE tendering drawings. May 2008: CE Tendering September 2008: Contract attribution. 22 October 2008: Start Civil Engineering works.

December 2010: Tunnel and Hall delivery.

December 2011: Move Front-End to the linac tunnel.January 2012: Start installation of DTL, CCDTL, PIMS.February 2012: Front-end re-commissioning.May-October 2012: DTL, CCDTL, PIMS commissioning.

November 2012: Transfer line commissioning, start PS Booster modifications.March 2013: Start of PS Booster commissioning with Linac4 beam.June 15th, 2013: Start physics run with Linac4 beam.

Page 7: Linac4 Status

7

Linac4 Parameters

Structures and klystrons dimensioned for 50 Hz Power supplies and electronics dimensioned for 2 Hz, 1.2 ms pulse.

Re-use 352 MHz LEP RF components: klystrons, waveguides, circulators.

Ion species H−Output Energy 160 MeVBunch Frequency 352.2 MHzMax. Rep. Rate 2 HzMax. Beam Pulse Length 1.2 msMax. Beam Duty Cycle 0.24 %Chopper Beam-on Factor 65 %Chopping scheme:

222 transmitted /133 empty bucketsSource current 80 mARFQ output current 70 mALinac pulse current 40 mAN. particles per pulse 1.0 × 1014

Transverse emittance 0.4 mm mrad

Max. rep. rate for accelerating structures 50 Hz

H− particles + higher injection energy (160/50 MeV, factor 2 in 2) same tune shift in PSB with twice the intensity.

Chopping at low energy to reduce beam loss at PSB.

Page 8: Linac4 Status

8

Linac4 Layout

CCDTL PIMS

3MeV

50MeV 94MeV 160MeV

Drift TubeLinac

352 MHz18.7 m3 tanks3 klystrons4 MW111 PMQs

Pi-Mode Structure

352 MHz22 m12 tanks8 klystrons~12 MW12 EMQuads

Cell-Coupled Drift TubeLinac352 MHz25 m21 tanks7 klystrons6.5 MW21 EMQuads

Beam Duty cycle:0.24% phase 1 (LP-SPL)<6% phase 2 (HP-SPL)

4 different structures, (RFQ, DTL, CCDTL, PIMS)

Total Linac4: 80 m, 18 klystrons

Ion current: 40 mA (avg. in pulse), 65 mA (bunch)

CHOPPERRFQ

Chopper

352 MHz3.6 m11 EMquad3 cavities

Radio FrequencyQuadrupole(IPHI)352 MHz6 m1 Klystron1 MW

H-

3MeV45keV

RF volumesource

DTL

Page 9: Linac4 Status

9

The 3 MeV Test Stand

In construction in the South Hall extension.

- H- source (2008) - LEBT (2008-09)- RFQ (February 2010)- Chopper line (2008)- Diagnostics line (2010)- Infrastructure (1 LEP Klystron, pulsed modulator, etc.) - ready

In the front end are concentrated some of the most challenging technologies in linacs, and this is where the beam quality is generated.

Early understanding and optimisation of front-end is fundamental for a linac project.

klystronmodulator

sourceRFQchopper line

diagnostics line

Page 10: Linac4 Status

10

The RFQ

Original idea:Use the RFQ being built in France for the

IPHI project, after the tests at Saclay.

However,1. This RFQ is now late by >2 years.2. Problems (microleaks) on 1st

segment.3. The design parameters (95 kV

injection energy, length 6 m) are not optimum for Linac4.

Conclusion:Decision (summer 07) to build a new

CERN RFQ optimised for Linac4:45 kV injection, 3 m length.

a. CERN+CEA beam dynamics and RF design,

b. CERN mechanical design, based on INFN and CEA designs (TRASCO and IPHI).

c. Construction and brazing in the CERN Workshops.

Page 11: Linac4 Status

11

Linac4 accelerating structures

Linac4 accelerates H- ions up to 160 MeV energy:

in about 80 m length

using 4 different accelerating structures, all at 352 MHz

the Radio-Frequency power is produced by 19 klystrons

focusing of the beam is provided by 111 Permanent Magnet Quadrupoles and 33 Electromagnetic Quadrupoles

PIMS

A 70 m long transfer line connects to the existing line Linac2 - PS Booster

RFQ DTL CCDTL PIMS Output energy 3 50 102 160 MeV Frequency 352 352 352 352 MHz No. of resonators 1 3 7 12 Gradient E0 - 3.2 2.8-3.9 4.0 MV/m Max. field 1.95 1.6 1.7 1.8 Kilp. RF power 0.5 4.7 6.4 11.9 MW No. of klystrons 1 1+2 7 4+4 Length 6 18.7 25.2 21.5 m

Page 12: Linac4 Status

12

Linac4 accelerating structures

DTL CCDTL PIMS

7-cell cavities in -mode. With respect to the SCL, the PIMS : has 5 times less cells (less machining time and cost). needs about the same quantity of copper. allows a simpler tuning (7 cells instead of >100, dummy tuners). allows standardisation of the Linac4 frequency to 352 MHz. has only about 12% less shunt impedance.

Conventional DTL structure with: PMQs in vacuum inside drift tube. no drift tube adjustment after installation.

Modules of 3 SDTL-type cavities with 2 drift tubes, coupled by 2 coupling cells.

Easy access and alignment of electromagnetic quadrupoles. Relaxed tolerances on drift tube alignment.

Page 13: Linac4 Status

13

Accelerating structures: gradients, peak fields, margins on RF power

Optimum gradient in the different structures determined by cost considerations.

For calculations, structure cost 180-250 kCHF/m, RF cost 450-650 kCHF/W

Maximum surface field limited to 1.8 Kilpatrick (considered as safe enough for a pulsed machine)

Maximum cavity power (copper, beam) 1 MW for klystrons (LEP) giving 1.2-1.3 MW

(~20% margin for regulation and losses in waveguides)

20% margin from theoretical Q-value for additional losses in cavities

Optimum E0 (LEP klystron)

Optimum E0 (pulsed klystron)

Optimum E0

(10 yrs. op.) Design E0 Max. field

(Kilpatr.) DTL 5.9 MV/m 5.1 MV/m 4.3 MV/m 3.2 MV/m 1.6 CCDTL 4.2 MV/m 3.7 MV/m 3.1 MV/m 3.5 MV/m 1.7 SCL - 3.8 MV/m 3.3 MV/m 4 MV/m 1.2 PIMS - 3.1 MV/m 2.7 MV/m 3.9 MV/m 1.8

Page 14: Linac4 Status

14

The Linac4 RF system

RFQ DTL CCDTL PIMS3 MeV 50 MeV 100 MeV 160 MeV

1.3 MW klystron (LEP, CW)

2.5 MW klystron (pulsed)

Modulator for 1.3 MW RF

Modulator for 2.5 MW RF

Initial configuration: 13 klystrons 1.3 MW, 6 klystrons 2.5 MW, 3 modulators 1.3 MW, 11 modulators 2.5 MW

RFQ DTL CCDTL PIMS3 MeV 50 MeV 100 MeV 160 MeV

Final configuration (at the end of the stock of LEP klystrons):

3 klystrons 1.3 MW, 11 klystrons 2.5 MW, 3 modulators 1.3 MW, 11 modulators 2.5 MW

Linac4 will reuse the stock of high-power klystrons coming from the old LEP accelerator:

Page 15: Linac4 Status

15

Linac4 civil engineering

Linac4 tunnel

Linac4-Linac2 transfer line

Equipment building

Access building

Low-energy injector

ground level

Pre-integration May – October 2007Tendering drawings November 2007 – April 2008Tendering May 2008, Contract to FC September 2008.

Page 16: Linac4 Status

16

Linac4 Work Breakdown Structure

2. LinacSystems

3. BoosterSystems

4. Installationcommiss.

5. Building,Infrastruct.

2.1 Ion Source and LEBT, R. Scrivens, AB/ABP

2.2 Radio Frequency Quadrupole

2.3 Chopper Line, A. Lombardi, AB/ABP

2.4 Accelerating Structures, F. Gerigk, AB/RF

2.5 Linac Beam Dynamics, A. Lombardi, AB/ABP

2.7 Beam Instrumentation, U. Raich, AB/BI

2.6 Radio Frequency Systems

2.8 Transfer Line, S. Maury, AB/ABP

2.9 Magnets

2.10 Power Converters, C. De Almeida, AB/PO

2.11 Vacuum Systems, G. Vandoni, AT/VAC

2.12 Control Systems, J. Serrano, AB/CO

3.1 Booster Injection Mod., W. Weterings, AB/BT

3.2 PSB Beam Dynamics, C. Carli, AB/ABP

4.1 Test Stand Operation, C. Rossi, AB/RF

5.1 Building Design, Construction, N. Lopez, TS/CE

5.2 Cooling and Ventilation, Y. Body, TS/CV

5.3 Electrical Systems, J. Pierlot, TS/EL

5.4 Access Systems, D. Chapuis, TS/CSE

4.2 Transport and Installation, S. Prodon, TS/IC

4.3 Survey, M. Jones, TS/SU

4.4 Linac Commissioning, A. Lombardi, AB/ABP

4.5 PS Booster Commissioning

LINAC4

2.6.2 High Power RF Systems, O. Brunner, AB/RF

2.6.1 Low Level RF Systems, P. Baudrenghien, AB/RF

2.6.3 Slow Tuning and Interlocks, L..Arnaudon, AB/RF

2.6.4 Tube Amplifiers, J. Broere, AB/RF

2.9.2 Magnetic Measurements, M. Buzio, AT/MEI

2.9.1 Magnet Design and Procurement, T. Zickler, AT/MCS

2.9.2 RFQ Mech. design, Fabrication, S. Mathot,TS/MME

2.2.1 RFQ Funct. design, Commissioning, C. Rossi, AB/RF

4.5.2 PSB Commissioning to ultimate, C. Carli, AB/ABP

4.5.1 PSB Commissioning to nominal, K..Hanke, AB/OP

1. Management, M. Vretenar + core team

2.13 Beam Intercepting devices, Y. Kadi, AB/ATB

LINAC4 WORK BREAKDOWN STRUCTURE

15.04.2008

Total of 30 Workpackages

Page 17: Linac4 Status

17

Linac4 Master Plan

End CE works: December 2010

Installation: 2011

Linac commissioning:

2012

Modifications PSB: shut-down 2012/13 (7.5 months)

Beam from PSB: 15.6.2013

White Paper

Task Name

Linac4 project start

Linac systems

Source and LEBT construction, test

Drawings, material procurement

RFQ construction, test

Accelerating structures construction

RF equipment procurement

Transfer l ine construction

Magnets design and procurement

Power converters procurement

Building and infrastructure

Building design and construction

Infrastructure instal lation

PS Booster systems

PSB injection elements construction

Installation and commissioning

Test stand operation (3 MeV)

Acc. structures testing, conditioning

Cabling, waveguides installation

Accelerator instal lation

Equipment installation

Hardware tests

Front-end commissioning

Linac accelerator commissioning

Transfer l ine commissioning

PSB modifications

PSB commissioning with Linac4

Start physics run w ith Linac4

01/01

15/06

Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q12007 2008 2009 2010 2011 2012 2013 2014

Page 18: Linac4 Status

18

Installation and commissioning

Page 19: Linac4 Status

19

Linac4 Budget

Linac Systems 47.3 MCHF 50.9%Booster Injection Upgrade 10.1 MCHF 10.9%Installation, Commissioning 2.3 MCHF 2.5%Building, Infrastructure 33.3 MCHF 35.8%

TOTAL LINAC4 93 MCHF

Project Management

Front End

Accelerating Structures

Radio Frequency

Beam Instrumentation

Magnets

Power Converters

Controls

Transfer Line

Vacuum

Booster Injection Upgrade

Installation, Commissioning

Building, Infrastructure

Does not include: chopper line (built in HIPPI) 13 klystrons + waveguides from LEP

Budget is covered by White Paper funding (55 MCHF)CERN budget for 2011-12 (38 MCHF)

Page 20: Linac4 Status

20

Linac4 Status (10/2008)

o Civil Engineering works started 22.10.2008, delivery of building end 2010.o Safety File submitted to CERN Safety Commission, June 2008. Building

approved.

o Ion source almost completed, first beam tests end 2008.o 3 MeV Test Stand infrastructure completed.o Prototype modulator tested with LEP klystron in pulsed mode.o Prototypes of accelerating structures tested (CCDTL), in construction (DTL),

in final design stage (PIMS). Material being ordered, construction of DTL and CCDTL will start in 2009.

o Started preparation for large contracts (klystrons, modulators, magnets,…).

o Detailed descriptions of Workpackages in preparation, project baseline will be frozen at end 2008.

o Advanced negotiations for in-kind contributions with France, Russia (via the ISTC Organisation), Poland, Spain (via ESS-Bilbao), India, Pakistan, Saudi Arabia.