The RFQ for IFMIF-EVEDA project: status and high-power tests.static.sif.it/SIF/resources/public/files/congr14/ip/Fagotti.pdf · •The focusing in the Gentle Buncher is strong (B=7)

Istituto Nazionale di Fisica Nucleare (Italy)

The RFQ for IFMIF-EVEDA project:

status and high-power tests.

Enrico Fagotti

INFN-LNL

Michele Comunian

INFN-LNL


International Road Map

Advanced Materials are at a critical path

ITER

1-3 dpa/lifetime

20-40 dpa/year

DEMO < 150 dpa

IFMIF

International Fusion Materials Irradiation Facility Dpa: displacement per atom: measure the integral of received radiation


Typical reactions: 7Li(d,2n)7Be, 6Li(d,n)7Be, 6Li(n,T)4He

Beam footprint on Li target 20cm wide x 5cm high (1 GW/m2)

IFMIF Principles

Accelerator based neutron source

using the D-Li stripping reaction

intense neutron flux with the

appropriate energy spectrum

2 x 125 mA D+ CW at 40 MeV

Test Modules


A specific source is needed to simulate DEMO

Spectrum fission

n

Neutron flux compared with DEMO’s

in available and planned neutron sources

Steven J. Zinkle, A. Moeslang, Evaluation of irradiation facility

options for fusion materials research and development,

Nuclear Engineering & Design, pre-printed (2013)

• The accumulation of gas in the materials

lattice is intimately related with the

neutron energy

• 54Fe(n,α)51Cr

• (incident n threshold at 2.9 MeV)

• and

• 54Fe(n, p)54Mn

• (incident n threshold at 0.9 MeV)

• Swelling and embrittlement of materials

takes place


Example of after irradiation test of small speciments


Facility design

7 Juan Knaster IPAC 2013 - Shanghai

Accelerator facility of IFMIF

10

the highest intensity the highest beam power the highest space charge the longest RFQ

2 x 5 MW beams

Individual availability of >90%

2 accelerators, 2 x 125 mA, 2 x 5MW

D+ source CW 140 mA

100 keV

SRF Linac HWR 175 MHz

40 MeV

Beam shape 200 x 50 mm2

RFQ 175 MHz

5 MeV

A Pisent SIF 2013

A Pisent Follow up Jannuary 2012

Linear IFMIF Prototype Accelerator

A Pisent Follow up Jannuary 2012

Linear IFMIF Prototype Accelerator


RFQs general parameters

Name Lab ion energy vane beam RF Cu Freq. length Emax Power density

voltage current power power ave max

MeV/u kV mA kW kW MHz m lambda kilpat W/cm 2 W/cm 2

IFMIF EVEDA LNL d 2.5 79-132 130 650 585 175 9.8 5.7 1.8 3.5 60

pulsed CERN linac 2 CERN p 0.75 178 200 150 440 202 1.8 1.2 2.5

SNS LBNL H- 2.5 83 70 175 664 402.5 3.7 5.0 1.85 1.1 10

CERN linac 3 LNL A/q=8.3 0.25 70 0.08 0.04 300 101 2.5 0.8 1.9

CW LEDA LANL p 6.7 67-117 100 670 1450 350 8 9.3 1.8 11.4 65

FMIT LANL d 2 185 100 193 407 80 4 1.0 1 0.4

high p IPHI CEA p 3 87-123 100 300 750 352 6 7.0 1.7 15 120

TRASCO LNL p 5 68 30 150 847 352 7.3 8.6 1.8 6.6 90

CW SARAF NTG d 1.5 65 4 12 250 176 3.8 2.2 1.6

mid p SPIRAL2 CEA A/q=3 0.75 100-113 5 7.5 170 88 5 1.5 1.65 0.6 19

CW ISAC TRIUMF A/q=30 0.15 74 0 0 150 35 8 0.9 1.15 - -

lp PIAVE LNL A/q=6 0.58 280 0 0 8e-3 (SC) 80 2.1 0.5 - - -

Cryo pump

RF coupler

Tuner

0.55 m m

odule

IFMIF-EVEDA RFQ

• 18 modules 9.8 m

• Powered by eight 220 kW rf chains and 8 couplers

• High availability 30 years operation.

• Hands on maintenance

• First complete installation in Japan

The RFQ is

INFN Italy responsibility

LNL

Padova

Torino

..Bologna


INFN organization

11

RFQ system organization

Torino

Bologna

Legnaro

Padova • Responsible A. Pisent

– Responsible for Padova: A. Pepato

– Responsible for Torino: P. Mereu

– Responsible for Bologna: A. Margotti

About 30 persons involved, 20 FTE, 10 dedicated

contracts

The participation of INFN to IFMIF-EVEDA includes

– RFQ construction

– Participation to final IFMIF design activity

– Participation to the man power of the project team in

Japan

– Participation to beam commissioning in Japan


IFMIF/EVEDA Accelerator building by JAEA (Rokkasho)


IFMIF RFQ

modulation design

• The voltage is increased (79-132 kV)

following an analytic law

• The focusing in the Gentle Buncher is

strong (B=7) so to keep the tune

depression above 0.4 for the best

control of space charge.

• Main resonances are avoided in the

accelerator section

• The focusing in the shaper raises from

4 to 7 to allow an input with smaller

divergence.

Ions d

Energy range 0.1-5 MeV

input-output nom emitt 0.25 mmmrad (rms)

Ouput long emitt. 0.2 MeV deg (rms)

Output current 0.2

Tansmission 98 % WB distr.

95 % Gsussian distr.


Current Loss [uA/m]

0

1

2

3

4

5

6

0 200 400 600 800 1000

RFQ Length [cm]

Cu

rre

nt

Loss

[u

A/m

]

Serie1

Power Loss [W/m]

0

50

100

150

200

250

0 200 400 600 800 1000

RFQ Length [cm]

Po

we

r Lo

ss [

W/m

]

Serie1

Neutron production [n/(s*m)]

0.00E+00

2.00E+08

4.00E+08

6.00E+08

8.00E+08

1.00E+09

1.20E+09

1.40E+09

1.60E+09

0 200 400 600 800 1000

RFQ Length [cm]

ne

utr

on

s [n

/(s*

m)]

Serie1

Integral 0.96 mA

Integral 2.4 10^9 n/s

Integral 363 W

WB distribution 0.25 mm mrad rms norm

Beam losses

• To achieve Beam

losses concentrated

in the low energy

part is very

important since

neutron production

is proportional to w2

1.2710*15.5 Nwn


IFMIF EVEDA RFQ challenges

• 650 kW beam should be accelerated with low beam losses and activation of the structure so as to allow hands-on maintenance of the structure itself ( Beam losses<10 mA and <0.1 mA between 4 MeV and 5 MeV). (Tolerances of the order of 10-50 um)

• 600 kW RF dissipated on copper surface: necessity to keep geometrical tolerances, to manage hot spots and counteract potential instability.

• The RFQ will be the largest ever built, so not only the accelerator must be reliable, but also the production, checking and assembling procedure must be reliable

– Fully exploit INFN internal production capability (design machining, measurement and brazing)

– Make production accessible for different industrial partners

• At present and we are in the production of the modules phase.

• 9/18 have been accepted today


Modules construction


RF cavities built in three supermodules

(6 modules each)

• High energy SM in construction at Cinel, Padua (Italy), Intemediate energy in INFN

Padova, Low energy by RI Koln (Germany)


Main news

• Module production:

– Cinel production completed (6 modules),

– INFN production, 6/6 final machining, 2/6

brazed (one need repairing).

– RI electrodes machined, brazing late

• High power tests

– Segment (four modules) tuned to 175 MHz

and +1%

– The high power test stand (approx 500 kW

installation) is fully operational.

– The couplers by JAEA have shown

unacceptable power losses at the window.

As a risk reduction procedure a new

couple couplers have been designed and

built by INFN (machining and detailed

mechanical design at CINEL)

– INFN couplers have been tested to the

nominal power of 200 kW cw

.

RFQ inside view

RFQ couplers by INFN


• Based on vacuum brazing, LNL mechanical experience with TRASCO, CERN experience for RFQ brazing, design compatible with oven at CERN, LNL and in industry;

• Due to the relatively large transverse dimensions of the RFQ, the procurement of the CUC2 raw material blocks is limited by the total mass amount (length 550 mm).

• To minimize the use of Ultra-pure CUC2 and to limit the induced stresses on the raw material, a rough-cut of the shape of the module components from a starting block of about 500x280x570 mm will be performed, by using a EDM (wire electroerosion).

• The accelerator is composed by 18 of these modules.

Mechanical design

E

T

Brazing joint

Prototype before brazing at CERN

Brazing joint


Mechanics details

Head flange

Vacuum grids machined from bulk

650


Cooling circuit

• About 600 kW RF power are removed by

means of 28 channels longitudinally drilled

along the RFQ modules; the water velocity

is approximately 3 m/s,

• 12 channels at fixed low temperature on

the vanes

• 16 channels on the cavity wall with

variable temperature for frequency tuning

• the temperature of the channels on the

vane and on the cavity wall can be

separately tuned so to achieve a tuning

range of +100kHz.

• 3 modules are in series (supermodule

water-in water-out

module (~0.55 m)

super-module (~3.3 m)

9.78 m

water-out water-in


3D details

• Dummy tuners, vacuum grids and end cells

• In the end cell the 45° angle of the undercut

guarantees the access of the cooling channel

as close as possible to the hot spot at the

electrode base (~80 W/cm2*), which is the

most severe of the entire RFQ • Deformations of 70 um and field perturbation

less than 1%

70 deg C

Slug tuners (CF100)

*with margin of 10% higher field

Tuning range +1 MHz



First construction step module n. 16

• Rough machining of block 550 mm long via EDM for minimal stresses

and deformations during annealing and brazing

EDM at Padova

Remaining copper 550 mm electrode before cleaning


Finishing

• 0.7 µm roughness

• 3D modulation

• 20 µm tolerances on vane

tip geometry


Calypso

4.10.06

0016049000Meas. Plan Name

Date September 29, 2

OrderCarl Zeiss

Part Number

8

CMM Type

ACCURA_MASS

Drawing No. Department:

Operator IFMIF

Signature:

1: Curve Form Full Profile Z 1 mm

X

Y

-200.0000 -150.0000 -100.0000 -50.0000 0.0000 50.0000 100.0000 150.0000 200.0000

-200.0000

-150.0000

-100.0000

-50.0000

0.2 mm

50 : 1

X

Y

No Identifier Sigma [mm] Form [mm]Number

of Points

Lower Tol.

[mm]

Upper Tol.

[mm]

Min Dev.

[mm]MaxInd

Max Dev.

[mm]MinInd Best Fit X [mm] Y [mm] Z [mm] X Y Z

Curve Form Full Profile Z 1 mm 0.0362 0.1255 2396 -0.0200 0.0200 -0.0169 13 0.1086730 Translation Rotation0.0000 0.0000 0.0000 0.0000 0.0000 0.00001

Point 39

Characteristic: Curve Form Full Profile Z 1 mmSegment: Segment 001

Lower Tolerance: -0.0200 mmUpper Tolerance: 0.0200 mmDeviation: 0.1073 mm

Position: X: -201.1000 mmY: -67.5000 mmZ: 1.0000 mm

Direction: X: 0.0000 mmY: -1.0000 mmZ: 0.0000 mm

Status: Valid

Point 1336



Position: X: 7.7707 mmY: -176.8850 mmZ: 1.0000 mm


Status: Valid

Point 1461




Direction: X: 1.0000 mmY: 0.0000 mmZ: 0.0000 mm

Status: Valid

Point 2353





Status: Valid

Four electrodes of module #16 electrodes

(machined by Cinel) in specs

Calypso

4.10.06

0016043000Meas. Plan Name

Date September 15, 2

OrderCarl Zeiss

Part Number

7

CMM Type

ACCURA_MASS

Drawing No. Department:

Operator IFMIF

Signature:

Z

Y

-250.0000 -200.0000 -150.0000 -100.0000 -50.0000 0.0000 50.0000 100.0000 150.0000 200.0000 250.0000

0.1 mm

150 : 1

Y

Z

No Identifier Sigma [mm] Form [mm]Number

of Points

Lower Tol.

[mm]

Upper Tol.

[mm]

Min Dev.

[mm]MaxInd

Max Dev.

[mm]MinInd Best Fit X [mm] Y [mm] Z [mm] X Y Z

Curve Form Middle Curve 0.0026 0.0099 3491 -0.0100 0.0100 -0.0039 568 0.00611166 Translation Rotation0.0038 0.0090 0.0000 0.0000 0.0000 0.00051

Max Deviation: 6.1 mm

Max

Deviation:

10.5 mm

Max

Deviation:

18.6 mm

material stock


INFN development for Brazing

Necessary for the large production (18 modules could not

be done at CERN).

Vacuum oven in INFN LNL


Brazing at LNL (1/2)

• Upgrade of the vacuum system

• Construction of the assembly lab

• Test of brazing geometry with test

pieces.

• Ultrasonic check of brazing


• Chemical preparation

• Brazing

Brazing at LNL (2/2)


• Construction of the assembly lab

• One step vertical

• Ultrasonic check of brazing

• Chemical preparation

• Brazing

• Vacuum ok dimensions not significative

Succesfull Brazing at LNL and Cinel

of two small prototypes (half scale transversally)


Module 16 brazing

After first brazing (Vertical, Cinel)

vacuum and dimensions were ok

After second brazing (Vertical)

Vacuum leak reparing on vacuum port and severe problems with brazing of head

flange plates. Deformation at the limit of PA acceptance (100 um, 1 MHz)


INFN production of modules

• Machining and QA at

INFN Padova

• Chemical processing at

LNL

• Assembly at LNL (sez pd

responsibility)

• Brazing by LNL


High power test stand at LNL

• A 500 kW test stand able to test 4

RFQ modules, to test at full power

density the structure (200 kW RF

power)

• The test is necessary to validate the

design during the module

construction (it was ment to be used

at the beginning, now we have built

half of the modules)



JAEA couplers history

• All the components for coupler test-stand arrived to LNL at the

beginning of July 2013.

• Before connecting the system to the RF line, a Network

Analyzer was used to measure RF properties.

• Couplers resulted to be underperforming: 25% power losses


Backup Solution

A new RF Power Coupler was designed Nov 2013

A new coupling cavity was designed Nov 2013

Two new RF windows were ordered Nov 2013

and delivered to LNL Feb 2014

The two couplers were machined (CINEL) Feb 2014

and brazed (LNL) Apr 2014

The new coupling cavity was produced (CINEL) Apr 2014

A new coupler test-stand was assembled May 2014

Coupler conditioning started 25-May-2014

Status Input power (kW) Duty Cycle (%)

27/05/2014 200 0.02

28/05/2014 200 2.50

29/05/2014 150 100

03/06/2014 200 100


Power Coupler - Concept

• Coupler and RF window are two separate devices coupled with a standard 6-

1/8 IEC coaxial interface

• Coupler material is copper OFHC

• Water heat exchange coefficient is maintained near 10000 W/m2K in the whole

heat exchange surface: water velocity is above 2.5 m/s both in the external and

internal spiral and in the loop

• Coupler inner connection is used to remove power from the RF window


HPC – Machining and brazing


High Power Couplers


HPC test-stand assembly and measurements

S12 = -0.34 dB (7.6% insertion loss) (with dummy couplers) S12 = -0.44 dB (9.7% insertion loss) (with final couplers without joints)


HPC test-stand control system


RFQ alignment (outside view)


RFQ alignment (inside view)


RFQ Tuning for the High Power Test

1. Tuner flush and end plates @ nominal settings (g=12 mm insertion each): Mode Spectra

Q modes

fq0=174.77 MHz fq1=190.79 MHz

fq2=226.52 MHz

Q0 (fq0) = 7040

D modes

fd0=171.38 MHz

(degeneracy= 10 kHz)

fd2=190.86 MHz


fq2=228.68 MHz


Beadpull measurements (metallic bead)

Dipole components are very low:

Strong Q component tilt at LE side

Due to the P2 module frequency mismatch

03/06/2014 44


RFQ Tuning for the High Power Test (2)

3. Measurement after 2 tuning steps

4. Measurement after the insertion of the dummy coupler

fq0=175.001MHz

fq0=175.014 MHz The attainment of the maximum

value of ± 2% voltage

perturbation is verified almost

everywhere in the RFQ. The

zone exceeding this value is

related only to the P2 RF plug.

The dummy coupler had only a

very slight effect on both fq0 and

voltages.


Documents

The RFQ for IFMIF-EVEDA project: status and high-power tests.static.sif.it/SIF/resources/public/files/congr14/ip/Fagotti.pdf · •The focusing in the Gentle Buncher is strong (B=7)