Development of a beam flattening

system at J-PARC/JSNS

1

Workshop on Non-linear beam expander systems in high-power

accelerator facilities

ISA, Department of Physics and Astronomy, Aarhus University, Aarhus,

Denmark 26th and 27th March 2012

(J-PARC/JAEA) Shin-ichiro MEIGO

2

Outline

Introduction

Present status of JSNS

Beam flattering system

Effect of miss alignment

Edge peak effect

Requirement of magnet located downstream

Beam peak reduction

3

Proton beam transport line

3-GeV RCS

Ne

utr

ino

lin

e

MLF

Neutron target

Muon target

road to coast

Length of BT: 314m

Ep: 3GeV

Power: 1MW

Rep.: 25Hz

JSNS MUSE

4

NMトンネル M1トンネル M2トンネル

QC10Q2260 LQ30120

QNQ1 QNQ2 QM1Q26100MIC

QM2Q26100MIC

QN1Q26100MIC

QN2Q26100MIC

QN3Q26100MIC

QN4Q26100MIC

Y20S2260AC

M20X20S2260AC

Y21S3060AC

M21X21S3060AC

Y22S2650MIC

M22X22S2650MIC

Y23S2650MIC

M23X23S2650MIC

QC11Q2260

QC12

P6

TMP 7

物質・生命実験施設

LQ30120 LQ30120

261500 314000B.W

ダクト径φ290

Muon Target

FCV3（DN200）

FCS（DN200）

QM1Q26100MIC

QM2Q26100MIC

proton beam window

neutron target

QM3Q26100MIC

QM4Q26100MIC

QM5Q26100MIC

QM6Q26100MIC

Y22S2650MIC

M22X22S2650MIC

Y23S2650MIC

M23X23S2650MIC

muon target collimators

16

00

ベースプレート

水銀ターゲット

ヘリウムベッセル

遮蔽（鉄）

ベッセルサポートシリンダー

反射体

遮蔽（重ｺﾝｸﾘｰﾄ）

アウターライナー

遮蔽（鉄）

遮蔽（コンクリート）

水素輸送管

水素減速材

遮蔽（鉄）

陽子ビーム

陽子ビーム窓

陽子ビーム窓メンテナンス用

ポート

ターゲット台車ベッセル内遮蔽体

P7 P8

垂直偏向部

直線部B

QA4

X05S2260AC

Y05S2260AC

M05

B1UD16150V

QA5

X06S2260AC

QV1Q2260

Y06S2260AC

M06

QV2Q2260

QV3Q2460

QV4Q2260

QV5Q2260

QV6Q2260

QV7Q2260

B1D

B16150V

QB1Q2260

QB2 QB3Q2260

QB4Q2260

QB5Q2260

QB6Q2260

X07S2260AC

Y07S2260AC

M07

X08S2260AC

X08S2260AC

M08

Y09S2260AC

X09S2260AC

M09 X10S2260AC

Y10S2260AC

M10 Y11S2260AC

M11X11S2260AC

Q2260 Q2260

Q2260P4

P3

Foil（DN200）

GV-man（DN200）

GV-man（DN200）

TMP（200 l/s）

水平偏向部 直線部C

QB7Q2260

BH1B16150

QH1Q2260

BH2B16150

QH2Q2260

QH3Q2260

QH4Q2260

QH5Q2260

BH3B16150

BH4B16150

QC2 QC3Q2260

QC4Q2260

QC5Q2260

QC6Q2260

QC7Q2260

QC8Q2260

QC9Q2260

QC1LQ30120

Y12S2260AC

X12S2260AC

M12 X13S2260AC

M13Y13S2260AC

Y14S2260AC

M14X14S2260AC

Y15S3060

M15X15S3060

Y16S2260AC

M16X16S2260AC

Y17S2260AC

M17X17S2260AC

Y18S2260AC

M18X18S2260AC

Y19S2260AC

M19X19S2260AC

LQ30120P5

GV3（DN200）

GV4（DN200）

QC9Q2260

QC10Q2260

← 3GeV RCS beam dump

horizontal bend

neutron

target

muon target

NM tunnel

QX1

SQ3060

X01S3060

QX2 QX3 QX4

Y01S3060

B01

D15150

QX6PB

PB13150

QX7

M01 X02S3060

Y02S3060

M02

QX8 DMP

D15150C11.8deg

QX9QX5 QX10

X03S3060

M03

B02

Y03S3060

QA1 QA2

X04S3060

Y04S3060

M04

QA3

SQ3060 SQ3060 SQ3060 SQ3060 SQ3060 SQ3060 SQ3060 SQ3060 SQ3060 SQ3060 SQ3060 SQ3060D16150

ダンプ入射P1

FCV1 & FCS1(DN200)P2

GV3（φ320）

GV2（φ320） GV1

（φ320）

proton beam

windowcollimators

pulse magnet to

50-GeV MR

B： 9 Q： 54 STR： 45

33m

Whole length 320m 600ns

FWHM ~150ns

Pulse

5

Proton beam at the target 3NBT・MLF beam operational status Beam study with 0.4 MWeq beam (Mar 2012)

1h demo of 0.3 MW beam ( Dec 2009)

User operation 0.2MW beam

Handling of high intensity proton beam Beam profile is important.

Pitting damage (P4 Law)

Profile measurement with high reliability

Condition of JSNS is harder than SNS SNS:60Hz, storage ring without muon target

JSNS:25Hz, RCS with muon target

- Target must be changed when pin holes was detected by the radiation monitor

Control of beam profile at JSNS becomes important

SNS has a record of exchanging new vessel within only 2 weeks after found fail.

In JSNS, failure of the vessel will take much longer time due to difficulties of handle of RI especially for release of tritium.

Peak reduction becomes very important!

Pin holes at target of SNS

Pin holes are permitted at

the inside wall of SNS but

not allowed at JSNS

Target vessel

SNS: 4 walls

JSNS: 3 walls

here

6

Diagnostics of beam profile and halo

Monitors at PBW Imaging Plate MWPM

Hot cell

IP

Target

Placed by RH

MWPM

Halo monitor

・SEC

・ TC

Beam profile monitor and beam halo monitor (online type)

Multi Wire Profile Monitors (MWPMs) (15 sets located) : SiC wires

At the proton beam window (PBW), MWPM located (1.8m from the target)

2D profile： Activation by the Imaging Plate (IP) (Offline type)

After beam operation: IP was attached at the target by the remote handling

TC

7

Beam behavior

0

10

20 Horizontal

0 100 200 3000

10

20

Beam direction position(m)

Beam

rm

s w

idth

(mm

)

Vertical

120 kW 300 kW

Good agreement with the design calculation

300kW:

eh,v 5.7, 4.9 p

120kW:

eh,v 2.7, 2.7 p

Result of RMS emittance Beam emittance and twiss parameters fitted by the observed

beam width

⇒ Good agreement in whole beam line

Consistence from the accelerator to the target.

0

10

20Horizontal

0 100 200 3000

10

20

Beam direction position(m)

Beam

rm

s w

idth

(mm

)

Vertical

300kW twiss parameter

ax -1.84, bx 20.4m

ay 0.57, by 5.26m

unit: mm mrad

e 5.4 p

ax -2.35, bx 24.8m

ay 0.89, by 5.88m

Twiss parameter at exit

of RCS

Calculation of 300 kW

case (by RCS team)

Residual dose at beam line:

Back ground except several points

8

Beam profile at the mercury target

2-D measurement by IP

Result by MWPM

Fitting by Gaussian

• Width of each pulse

obtained

• Beam width transformed to

the width at the target

-100 -50 0 50 1000

100

200

1st IP

Horizontal (mm)

Fit result Peak 1.35mm Sigma 23.8mm

Resultoffset= 01:Int.= 102.7961:Pos = 1.351151:Width=55.99872:Int.= 64.84372:Pos = -16.94012:Width=139.433

Profile result by the IP

• Fitted by two Gaussian curves

Contribution of primary protons and

secondary particles (mainly neutrons)

0.1 MW operation (2009 Dec) 0.2 MW operation (2010 Dec)

Obtained only 6 days of

cooling duration after

irradiation of 0.2 MW

beam

⇒ Possible for 1MW with

certain cooling time

MWPM at the PBW

9

Trend of beam width at the mercury target

2009 Apr

(Run#22)

2009 Nov

(Run#27)

2009 Dec

(Run#28)

2010 Jan

(Run#29)

2010 Dec

(Run#36

200kW)

sh sv sh sv sh sv sh sv sh sv

MWPM 17.3 10.3 22.8 11.0 24.4 11.7 33.8 16.6 54.3 22.6

IP 17.3 12.3 23.8 11.5 27.0 12.7 33.2 15.4 55.7 20.6

Both results by MWPM and IP show good agreement

• Demonstration of reliable method

• Reliable peak density can be obtained by MWPM in real time.

Unit: mm

To obtain low peak density, beam width

gradually expanded Comparison of experimental results

0

10

20

30

40

50

60

Run number

RM

S b

eam

wid

th(m

m) Horizontal

IP MWPM

Vertical IP MWPM

22 27 28 29 36

Found the higher activation at M23 than the estimation calculation, too much wide beam?

10

0 2 4 610-2

10-1

100

101

102

Sigma for V(cm)

Hea

t (W

/cc)

for

1M

W

Reflector Middle section

Reflectory=Σan x

n

a0=-1.37731862e+00

a1=1.57906223e+00

a2=-1.00184180e-01

2.00883063e-01

|r|=9.89000584e-01

Middley=Σan x

n

a0=-2.04023978e-01

a1=-2.42419803e-01

a2=4.15369844e-01

1.69400404e-01

|r|=9.98073455e-01

0 5 1010-2

10-1

100

101

102

Sigma for H(cm)

Hea

t (W

/cc)

for

1M

W

Reflector Middle section

Reflecty=Σan x

n

a0=-1.37747623e+00

a1=7.10670687e-01

a2=-2.02950615e-02

2.00813778e-01

|r|=9.89008212e-01

MiddSecy=Σan x

n

a0=-2.04156726e-01

a1=-1.09012246e-01

a2=8.41059968e-02

1.69345156e-01

|r|=9.98074713e-01

Conserve beam

aspect ratio

H:V=2.2:1

Heat deposition at vicinity＜1W/cc

sh <37mm, sv <17mm

Target blade: no issues

Limit for 1MW operation without beam flattening

system.

Peak at the target: 14J/cc/pulse

-100 -80 -60 -40 -20 0 20 40 60 80 1000

5

10

15

Horizontal position (mm)

Q (

J/c

c/p

uls

e)

Heat loadlimit at bladeB

lad

e

Bla

de

1MW sh 37mmsv 17mm

Beam expand by linear optics

MWPM shows that the distribution is monotonous Gaussian.

Expanding Gaussian beam to decrease the peak density Vicinity of the target < 1W/cc(0.04J/cc/pulse)

Target blade < 90W/cc(3.5J/cc/pulse)

Beam halo was approximated to follow as monotonous Gaussian.

11

Beam halo measurements Heat deposition was measured by the thermo couples (TCs) located at the vicinity of the target and the beam window. Given by the temp rising due to beam irradiation Q(w/cc)=ρCdT/dt ρ：Density(g/cc), C: Thermal capacity (J/g/K), T:Temp (K),t: Time(s)

For beam of 200 and 300kW: Heat at vicinity ～0.3W/cc

Peak density and beam halo ⇒ Giving optimum parameter

Developed expert system for beam control (Beam orbit and profile)

Even a rookie can control the beam with confident as an expert.

Trend of temp at halo monitor at the PBW Result of heat deposition at halo monitor

12

Conceptual design of flattening Beam edge folding by non-linear optics

Linear optics Non-linear optics

w/ octupole

Phase space distribution (horizontal)

Beam flattening using octupoles

Present case is provably the first trial to the high intensity facility such as 1MW class.

Points: For the ideal system, octupole magnets (OCTs) can be placed at an

arbitrary position.

2 set of OCTs for flattening in horizontal and vertical directions

Increase the b function at OCTs within appropriate aperture Expand the beam width at each OCTs

Maximize beam aspect ration At horizontal OCT making large aspect ratio of H/V

Appropriate phase advance to the target

Peak edge appears contrary Beam offset cause the peak edge

S. Meigo, JAERI-Tech 2000-088

Where we should place OCTs? Location of OCTs: Preferable at near the mercury target (M2 tunnel)

-> Pragmatically very difficult High radiation and difficulty of precision alignment at M2 tunnel

Recent beam study shows that beam should be kept smaller in M2 due to the beam loss.

NMトンネル M1トンネル M2トンネル

QC10Q2260 LQ30120

QNQ1 QNQ2 QM1Q26100MIC

QM2Q26100MIC

QN1Q26100MIC

QN2Q26100MIC

QN3Q26100MIC

QN4Q26100MIC

Y20S2260AC

M20X20S2260AC

Y21S3060AC

M21X21S3060AC

Y22S2650MIC

M22X22S2650MIC

Y23S2650MIC

M23X23S2650MIC

QC11Q2260

QC12

P6

TMP 7

物質・生命実験施設

LQ30120 LQ30120

261500 314000B.W

ダクト径φ290

Muon Target

FCV3（DN200）

FCS（DN200）

QM1Q26100MIC

QM2Q26100MIC

proton beam window

neutron target

QM3Q26100MIC

QM4Q26100MIC

QM5Q26100MIC

QM6Q26100MIC

Y22S2650MIC

M22X22S2650MIC

Y23S2650MIC

M23X23S2650MIC

muon target collimators

16

00

ベースプレート

水銀ターゲット

ヘリウムベッセル

遮蔽（鉄）

ベッセルサポートシリンダー

反射体

遮蔽（重ｺﾝｸﾘｰﾄ）

アウターライナー

遮蔽（鉄）

遮蔽（コンクリート）

水素輸送管

水素減速材

遮蔽（鉄）

陽子ビーム

陽子ビーム窓

陽子ビーム窓メンテナンス用

ポート

ターゲット台車ベッセル内遮蔽体

P7 P8

垂直偏向部

直線部B

QA4

X05S2260AC

Y05S2260AC

M05

B1UD16150V

QA5

X06S2260AC

QV1Q2260

Y06S2260AC

M06

QV2Q2260

QV3Q2460

QV4Q2260

QV5Q2260

QV6Q2260

QV7Q2260

B1D

B16150V

QB1Q2260

QB2 QB3Q2260

QB4Q2260

QB5Q2260

QB6Q2260

X07S2260AC

Y07S2260AC

M07

X08S2260AC

X08S2260AC

M08

Y09S2260AC

X09S2260AC

M09 X10S2260AC

Y10S2260AC

M10 Y11S2260AC

M11X11S2260AC

Q2260 Q2260

Q2260P4

P3

Foil（DN200）

GV-man（DN200）

GV-man（DN200）

TMP（200 l/s）

水平偏向部 直線部C

QB7Q2260

BH1B16150

QH1Q2260

BH2B16150

QH2Q2260

QH3Q2260

QH4Q2260

QH5Q2260

BH3B16150

BH4B16150

QC2 QC3Q2260

QC4Q2260

QC5Q2260

QC6Q2260

QC7Q2260

QC8Q2260

QC9Q2260

QC1LQ30120

Y12S2260AC

X12S2260AC

M12 X13S2260AC

M13Y13S2260AC

Y14S2260AC

M14X14S2260AC

Y15S3060

M15X15S3060

Y16S2260AC

M16X16S2260AC

Y17S2260AC

M17X17S2260AC

Y18S2260AC

M18X18S2260AC

Y19S2260AC

M19X19S2260AC

LQ30120P5

GV3（DN200）

GV4（DN200）

QC9Q2260

QC10Q2260

← 3GeV RCS beam dump

horizontal bend

neutron

target

muon target

NM tunnel

QX1

SQ3060

X01S3060

QX2 QX3 QX4

Y01S3060

B01

D15150

QX6PB

PB13150

QX7

M01 X02S3060

Y02S3060

M02

QX8 DMP

D15150C11.8deg

QX9QX5 QX10

X03S3060

M03

B02

Y03S3060

QA1 QA2

X04S3060

Y04S3060

M04

QA3

SQ3060 SQ3060 SQ3060 SQ3060 SQ3060 SQ3060 SQ3060 SQ3060 SQ3060 SQ3060 SQ3060 SQ3060D16150

ダンプ入射P1

FCV1 & FCS1(DN200)P2

GV3（φ320）

GV2（φ320） GV1

（φ320）

proton beam

windowcollimators

pulse magnet to

50-GeV MR

M2

• Before beam commissioning, we hesitated to install octupole magnets. However, due to the following superiority of the J-PARC RCS・3NBT, we understand that the beam flattening by the octupole is possible.

- Very good stability of beam position - Deeply understanding of the beam optics

M1

15

Beam optics using OCT Flattening by the realistic magnet field of OCT(Koct), large beta function (b) at OCTs is necessary for the phase advance of f and RMS emittance e

Koct=1/eb2tanf

Large b makes difficult to have large beam acceptance

RCS collimator: 324pmm mrad (For b =100m -> 0.36m in diam)

From the preliminary result of the beam halo measurements, smaller acceptance than 324pmmmrad may be applicable.

Having large b at OCT1,2 -> Beam acceptance to be100 pmmmrad

Present beam optics Beam optics for flattening by OCTs

Large beta at OCT1,2

16

Octupole magnet

OCT1 and 2 will be installed at upstream of QC12, QNQ1.

at 3NBT tunnel and M1 tunnel

Q-O magnets distance is 1m

length of magnetic pole: 0.6m

Bore diameter 0.3m

Install new steering magnet: At downstream of QC12

OCT1 OCT2

M1 tunnel 3NBT tunnel

1m

OCT1 OCT2

1m

Horizontal view

Vertical view

Octupole magnet (800T/m3)

O3060(Width 1.2m, Length 0.6m, 6t)

Octupole magnets

Excellent of shape of the coil

Installation: Summer in 2013 during long shutdown

Field measurements Mapping by hole probe Field gradient 780 T/m3 @671A

Agreement with calculation

Confirmation agreement with design calculation

立会試験　2号機　励磁特性　0～800[A]　800～0[A],X=70,Y=0,Z=0[mm]

0

100

200

300

400

500

600

0 100 200 300 400 500 600 700 800 900

電流[A]

磁場

　B

y[G

auss]

励磁特性　0～800[A] 励磁特性　800～0[A]

立会試験　2号機　X方向　-160～160[mm]　電流：671[A],Y=0,Z=0[mm]

-6000

-4000

-2000

0

2000

4000

6000

-200 -150 -100 -50 0 50 100 150 200

位置　X[mm]

磁場

By　

[G

auss]

19

Simulation of the profile

DECAY-TURTLE (PSI version)

One of result ignoring the reality Ignore muon production target

Small acceptance of beam: 81p mm mrad

Using this optics let’s discuss the alignment error effect

W/O OCT W/ OCT

Achieved flat shape!

20

Alignment accuracy of octupole magnet

Beam offset at OCT 2mm in horizontal ->Edge peak appears Beam shift at OCT2(for horizontal): 2mm in horizontal

Increase about 8 % at the edge

Beam tuning for flattening system

Beam position monitor (BPM) is placed at each octupole magnets

Additional new steering magnets for horizontal and vertical

Beam center within 1 mm -> 4%

2D Profile at the target

21

Alignment of downstream magnets

Realignment of downstream magnets is quite difficult

Calculation: All magnets at M2 has 2 mm offset in horizontal.

Result: Betatron oscillation is found in the beam orbit, no influence on

the beam shape

⇒ No need realignment at M2

Horizontal Vertical 2D Profile at the target

Effect due to muon target

Phase space distribution distorted as Gaussian

Without scattering effect

Phase space distribution at muon production target

To minimize this effect

Focusing on the muon target as small as possible

Increase divergence ⇒ Smaller effect

With scattering effect

23

Beam profile with flattening system

Muon target: With muon target

RMS Beam emittance: 5p

Peak: 1.3Tp/cm2

9.4J/cc/pulse

Beam Loss: 80.8 kW

(mainly due to interaction at the

muon target)

Acceptance: Horizontal 114p

Vertical 111p

Reduction of peak 42 % of Gaussian

By P4 law: 0.68^4=0.2 (very small)

It is better to have larger beam acceptance at OCTs.

Profile at the mercury target Projection profiles on horizontal and vertical

axis compared with Gaussian

-8 -6 -4 -2102

103

104

Vertical (cm)

Gauss(sh 17mm) OCT (aug24#2)

-5 0 50

10000

20000

30000

40000

50000

60000

Vertical (cm)

Gauss( sh 17mm) OCT (aug24#2)

-20 -15 -10 -5102

103

104

Horizontal (cm)

Gauss(sh 37mm) OCT (aug24#2)

-20 -10 0 10 200

10000

20000

30000

40000

50000

Horizontal (cm)

Gauss(sh 37mm) OCT(aug24#2)

Closing up at vicinity

Beam study for octupoles optics

Revised OCT opt w/o MTG Previous OCT opt w/o MTG

Found slightly beam loss No significant beam loss for 0.18GeV inj

Phase space distribution RCS extraction beam

(0.4GeV injection)

Large aperture >250p

Required aperture ~250 p

Aperture ~100p

0 50 100 150 200 2501

10

100

1000

10000

Beam emittance (p mm mrad)

Be

am

lo

ss (

W)

at 1

MW

150pi painting w/ error

Beam acceptance (p mm mrad)

Beam profiles using flattening system

-20 -10 0 10 200

10000

20000

30000

40000

50000

Horizontal (cm)

Gaussian incld beam loss sh 37mm

H_DEC8_2.DAT

-5 0 50

10000

20000

30000

40000

50000

60000

Vertical (cm)

Gaussian incld beam loss sh 17mm

V_DEC8_2.DAT

Peak density ~11J/cc/pulse

Having larger aperture than 250 p mm mrad

26

0 200 400 600 800 10000

10

20

Beam power for 25Hz (kW)

Peak h

eat

den

sity

(J/c

c/p

uls

e)

JSNS 1MWDesign

Hp <1W/cc

SNS(at 0.3, 0.7 and 1 MW)

JSNS 2009 Dec 2010 Jan 2010 Feb 2010 Jun 2010 Dec

SNS 2MW Design

Peak heat density by flattening

• Even in the beam optics having

large beam acceptance such as 250

p, reduction of the peak enables by

the ~30% of Gaussian case.

• By the beam study, we can carry out

the beam flattening with the

appropriate the beam acceptance

which has beam loss < 1 W/m at the

OCT.

Beam power at 25 Hz (kW)

Beam

flattening

Beam power trend

27

0

1000

2000

3000

4000

5000

6000

2008/5/30

2008/6/22

2008/9/24

2008/12/10

2008/12/24

2009/1/25

2009/2/11

2009/2/19

2009/2/27

2009/6/1

2009/6/9

2009/6/17

2009/10/8

2009/10/18

2009/11/10

2009/11/18

2009/11/26

2009/12/13

2009/12/21

2010/1/13

2010/1/23

2010/1/31

2010/4/16

2010/5/11

2010/5/19

2010/5/27

2010/6/8

2010/6/16

2010/6/24

2010/10/19

2010/11/8

2010/11/16

2010/11/24

2010/12/2

2010/12/10

2010/12/18

2011/1/27

2011/2/4

2011/2/12

Beam

Pow

er[

kWh/D

ay]

0

50

100

150

200

250

300

350

400

450

500

Accum

ula

ted

Beam

Pow

er[

MW

h]

Beam

Pow

er[

kW@

25hz]

Beam Power[kWh/Day]

Accumulated Beam Power【MWh】

Beam Power[kW]

- 0.1-0.2 MW operation begun after earthquake.

- After installation He bubbler in the target, 0.3 MW operation will begin

High power with very high availability (~90%)

(/sr/pulse)

SNS(1MW) 4.2x1012

J-PARC/JSNS(0.3MW) 5.4x1012

Pulse neutron yield

28

Summary

Flattering system 30~40% of peak reduction

Due to scattering at the muon production target, makes difficult for flattening.

Focusing at the muon target is key at J-PARC/MLF

Beam acceptance is key issue of the system.

For 0.18GeV LINAC case, the case of 250 p mm mrad showed acceptable.

Octupole magnet Fabrication has been done

Good agreement with the magnetic field of design

Installation in summer 2013

29

Thank you for attention

- Note: Bulldogs have flat face

30

Y. Yuri et al., Phys rev special topics Accelerators and Beams 10, 104001 (2007)

K2n=(n-2)!(-1)n/2/(n/2-1)!(2eb)n/2-1btanf (n=4,6,8,10…) ⇒ Koct=1/eb2tanf, Kdodeca=-3/e2b3tanf

Settlement

Periodical survey for floor level

-14.0

-12.0

-10.0

-8.0

-6.0

-4.0

-2.0

0.0

2.0

May-10 Jun-10 Aug-10 Sep-10 Nov-10 Jan-11 Feb-11 Apr-11 Jun-11 Jul-11 Sep-11 Nov-11 Dec-11 Feb-12 Apr-12

Date

Dis

plac

em

ent

(mm

)

RCS-03

S1

N1

N7

N81

N82

LM1-11

- Swing of 2mm observed at downstream

- Stable ??? Scheduled periodical beam adjustment is necessary.

RCS-03

EXPJ

EXPJ

S1

N1

N7 N81 N82

LM1-11

MLF

32

To MR To MLF

Kicker and septum

RCS

Beam orbit

3NBT Dump

Bending

B01 Pulse Bending

PB

Bending

B15U

Effect of residual field of pulse bending magnet

In every 80 pulses(3.2s） 4 pulses(K1-4): injected MR 74 pulses(K7-80): transport to MLF

Residual magnetic field distorts the beam orbit to MLF Measurement effect of residual field

Spatial distribution for K9 having kick angle

of 0.06 mrad by residual field (calculation) Beam profile for K9 pulse (0.06mrad)

By the pulse steering magnets

with slow response, possible

to compensate residual field

> 76 pulses can be utilized.

Time dependence of kick angle

by the residual field effect

0 1000 2000 30000

0.1

0.2

0.3

Elapsed Time after K4 pulse(ms)

Kic

k a

ngle

(m

rad)

y=exp(a + b x)a=-2.20276731e+00b=-3.20571963e-036.78503433e-02|r|=9.88279259e-01

Fitted after K7 Estimated by Magnetic Field

K5

2~3 pulses in every pulses

increases ~8 % at the edge

Pulse train K1 K2 K3 K4 K5 K6 K7 K8 K9 ... K80

Not used K7,8 delivered since Run#37

33

Residual field compensator

Orbit by compensation Response of field for compensation

for case1

Pulse steering magnet

Case1: 2 magnets at downstream

Case2: 2 magnets at upstream

Required kick angel

Case1: 0.3 mrad, Case2: 0.6 mrad