1

JHF synchrotrons(design philosophy and resent status)

Shinji Machida

ＫＥＫApril 17, 2002

2

contents

• Introduction

– Purpose of JHF

• Design philosophy

– Minimizing beam loss

• Hardware R&D

• Prospective

3

Physics motivation

• １ｘ１０１４ｐｐｐ source of secondary particle （Ｋ，ｐｐ、ｍｕ，ｎｕ、ｅｔｃ．）– ５０ ＧｅＶ Ｐｒｏｔｏｎ Ｓｙｎｃｈｒｏｔｒｏｎ（Ｐ

Ｓ）

• １ＭＷ spallation neutron source– ３ ＧｅＶＰｒｏｔｏｎ Ｓｙｎｃｈｒｏｔｒｏｎ（Ｐ

Ｓ）

• Transmutation, both for physics and enginearing.– ６００ ＭｅＶ Ｌｉｎａｃ

4

Secondary particle source （Ｋ，ｐｐ、ｍｕ，ｎｕ、ｅｔｃ．）

• At least 30 GeV

– Yield does not change much above 30 GeV.

• At least １ｘ１０１４ｐｐｐ– Above AGS ＠ BNL

– Neutrino physics

5

Spallation neutron source

• 1 or a few GeV

– Neutron yield is almost proportional to the beam power.

• Beam power must be the order of 1 MW.

– ＳＮＳ（ US ）１ - ２ＭＷ .

– ＥＳＳ（ Europe ）５ＭＷ .

6

３ＧｅＶＰＳ：　ＬＡＲ（ｌｉｎａｃ and accumulator ）

ｖｓ．ＲＣＳ（ rapid cycling synchrotron ）

• 3GeV RCS is an injector of 50GeV MR.

– Higher injection energy of 50 GeV MR is preferable.

• The higher the extraction energy is, the lower the current.

– Beam power ＝ energy 　ｘ　 current

• More beam loss is tolerable at injection.

– The same reason above.

• Beam stays longer in the machine.

– ２０ｍｓ（ RCS ） ｖｓ．　１ｍｓ（ LAR ）• Hihger RF total voltage is necessary.

• All the magnet should be synchronized, especially the tracking of quads.

• Need some cure against eddy current.

7

核変換実験：工学実験施設と物理実験施設

• 本当に必要なのは、エネルギー１ＧｅＶ以上、ビームパワー１０ＭＷ級のＣＷビーム。

• 統合計画で実現されるのはあくまでその前段階を６００ＭｅＶ超伝導リニアックで実現。– 工学実験施設

• ６００ＭｅＶ超伝導リニアック、２００ｋＷでＡＤＳ構造材料の照射試験

– 物理実験施設• １０ｋＷ程度のビームで、ＡＤＳにおける中性子挙動の物理

的特性

8

Layout

9

Specifications

• ５０ＧｅＶＰＳ– Protons per pulse ３．３Ｅ１４　ｐｐｐ– Repetitions ０．３ Ｈｚ　（３．４２　ｓ）– Average current １５ ｍｉｃｒｏＡ– Beam power ０．７５ ＭＷ

• ３ＧｅＶＰＳ– Injection ４００ＭｅＶ– Protons per pulse ０．８３Ｅ１４　ｐｐｐ– Repetition ２５ Ｈｚ　（４０　ｍｓ）– Average current ３３３ ｍｉｃｒｏＡ– Beam power １ ＭＷ

10

Comparison with other running or planned project

• ５０ＧｅＶＰＳ like– running

• ＡＧＳ＠ＢＮＬ ３０ＧｅＶ　　０．６－０．７ｘ１０１４ｐｐｐ

• ＭａｉｎＩｎｊｅｃｔｏｒ＠ＦＮＡＬ １２０ＧｅＶ　０．２－０．３ｘ１０１４ｐｐｐ

– planned• Nothing else

• ３ＧｅＶＰＳ like– running

• ＩＳＩＳ＠ＲＡＬ ８００ＭｅＶ　０．２ＭＷ• ＰＳＲ＠ＬＡＮＬ ８００ＭｅＶ　０．０６ＭＷ

– planned• ＳＮＳ １．３ＧｅＶ　１－２ＭＷ• ＥＳＳ（ not started yet ） １．３３４ＧｅＶ　５ＭＷ

11

Design philosophy

Minimizing beam loss

12

Definition of beam loss and its limit

• Controlled loss– Localized and shielding takes care– ４ｋＷ　＠　３ＧｅＶＰＳ collimator– ７．５ｋＷ　＠　５０ＧｅＶＰＳ ESS septum

• Uncontrolled loss– Not localized and no special shielding. If this

happens to be large, total beam intensity is decreased.

– About １Ｗ／ m

13

Expected at each area (or element)• ３ＧｅＶＰＳ

– Injection and collimator ４ｋＷ ２％ ｃｏｎｔｒｏｌｌｅｄ– Around the ring １Ｗ／ｍ ０．１％ ｕｎｃｏｎｔｒｏｌｌｅｄ– extraction １ｋＷ ０．１％ ｕｎｃｏｎｔｒｏｌｌｅｄ　　 total ５．３３ｋＷ ２．２％

• ３ＧｅＶＢＴ– collimator ４５０Ｗ １％ ｃｏｎｔｒｏｌｌｅｄ– Other area except collimator １Ｗ／ｍ ０．４４％ ｕｎｃｏｎｔｒｏｌｌ

ｅｄ• ５０ＧｅＶＰＳ

– injection １３５Ｗ ０．３％ ｕｎｃｏｎｔｒｏｌｌｅｄ– collimator ４５０Ｗ １％ ｃｏｎｔｒｏｌｌｅｄ– Around the ring ０．５Ｗ／ｍ　０．３６％　ｕｎｃｏｎｔｒｏｌｌ

ｅｄ– Slow extraction ７．５ｋＷ １％ ｃｏｎｔｒｏｌｌｅｄ– Fast extraction １．１２５ｋＷ　０．１５％　ｕｎｃｏｎｔｒｏｌ

ｌｅｄ　　 total ８．９ｋＷ ２．７％ （ slow extraction ）

２．５ｋＷ １．８％ （ fast extraction ）

14

Example of radiation calculation（３ＧｅＶＰＳ collimator region ）

•Inside collimator （ near ＪＡＷ）

> １Ｓｖ／ｈ　（１．２ｋＷ / １台）

•Outside the shield （３００ｍｍFe and ４００ｍｍ concreate ）

７ｍＳｖ／ｈ

＊ radiation after 30days operation and 1day cooling

15

What makes beam loss?

• We do not know the behavior of a high intensity beam

– Space charge effects

– E-p instability

– Some other reasons ？

• Intensity independent effects

– Fringe fields due to large core magnets

16

Space charge tune spread

• ３ＧｅＶＰＳ– ０．１５　 assuming the following value

• emittance ２１６ｐｉ　ｍｍ－ｍｒａｄ• intensity ８．３３ｘ１０１３ｐｐｐ• Bunching factor ０．４２

• ５０ＧｅＶＰＳ– ０ . １４　 assuming the following values

• emittance ５４ｐｉ　ｍｍ－ｍｒａｄ• intensity ３．３３ｘ１０１４ｐｐｐ• Bunching factor ０．２７• Form factor １．７

17

Some efforts to minimize space charge effects

• Enlarge physical aperture to minimize vertical beta function at the bending magnets.

• Enlarge transverse emittance and control it with injection painting ＠３ＧｅＶＰＳ

• Control longitudinal emittance with RF manipulation.– Shaking of rf phase or voltage in 3 GeV RCS.

• Search best bare tune points away from any resonances.

18

Lattice function of ３ＧｅＶＰＳ

• Vertical beta function at a bend– Less than １６ｍ

• Vertical beam size at the extraction kicker– Less than １８０

mm

• No dispersion in long straight.

19

Emittance and acceptance （ transverse- １）

emittance（ｐｉ　ｍｍ－ｍｒａｄ）

Collimator acceptance

Physical acceptance

Ｌ３ＢＴ

Exit of linac １０

Right after Ｌ３ＢＴ collimator

４

３ＧｅＶＰＳ

After painting ２１６ ３２４ ４８６

extraction ８１（ｃｏｒｅ）

３２４（ｔａｉｌ）

－ core means the number simply determined by adiabatic damping of 1.5 times the painted beam.－ tail means the maximum possible amplitude within collimator.

20

Emittance and acceptance （ transverse- ２）emittance（ｐｉ　ｍｍ－ｍｒａｄ）

Collimator acceptance

Physical acceptance

３ＧｅＶＰＳ

Paining injection １４４ ３２４ ４８６

extraction ５４（ｃｏｒｅ）

３２４ ４８６

３２４（ｔａｉｌ）

３２４ ４８６

３ＧｅＶＢＴ

Right after ３－５０ＧｅＶＢＴ collimator

５４ ５４ １２０

５０ＧｅＶＰＳ

injection ５４ ５４－８１ ８１

extraction （３０ＧｅＶ）

１０

extraction （５０ＧｅＶ）

６．１

21

Longitudinal emittance and dp/p

emittance（ｅＶ　ｓｅｃｏｎｄｓ）

dp/p （％）

Ｌ３ＢＴ

Right after Ｌ３ＢＴ collimator ＋－０．１

３ＧｅＶＰＳ

injection ５ ＋－０．７５

extraction （ for ３ＧｅＶ users ） ５ ＋－０．４２

Extraction （ for ５０ＧｅＶＰＳ）

５ ＋－０．２４

22

Injection painting• Ａｎｔｉ－ｃｏｒｒｅｌａｔｅｄ　ｐａｉｎｔｉｎｇ

– Start from small emittance in horizontal and large emittance in vertical.• Ｃｏｒｒｅｌａｔｅｄ　ｐａｉｎｔｉｎｇ

– Start from small emittance in both horizontal and vertical.

• Which is a better way?• How we match the linac beam to the acceptance

– What is the optimizedtwiss parameters of linac beams.

ＣｏｒｒｅｌａｔｅｄＡｎｔｉ－ｃｏｒｒｅｌａｔｅｄ

23

Maximize bunching factor

• Ｏｆｆ－ｍｏｍｅｎｔｕｍ（＋０．３％ ) injection０．２５　　　　　　　　　　　　　　　　　　　

　０．４０

24

Beam loss at different bare tune

(7.30, 6.10) (7.30, 6.25)

(7.30, 6.40) (7.30, 6.45)

Ｎｙ＝６．１０：　 near integerＮｙ＝６．４０：　 near sum resonace

25

５０ＧｅＶＰＳ lattice

• Negative dispersion at bend magnets with missing magnets.

• Transition gamma is ３２ｉ• 3DOFO module with total 11 quad families.

26

E-p instability

• Serious problem at ＰＳＲ＠ＬＡＮＬ and e+-e- colliders

• We just start looking at the effects based on the JHF parameters.

27

Localized beam loss

• Makes beam loss controlled.• Collimator at

– Ｌ３ＢＴ– ３ＧｅＶＰＳ– ３ＧｅＶＢＴ– ５０ＧｅＶＰＳ　　　（ only transverse ）

• What is the necessary Physical aperture when theCollimator aperture is fixed.?

28

R&D of hardware

• ３ＧｅＶＰＳ– Injection and extraction elements

– Tracking 11 families of quadrupoles

– Ceramic chamber and RF shielding

– RF acceleration cavity with 450kV total voltage

– Charge exchange foil

• ５０ＧｅＶＰＳ– Bending magnet with 1.9 T

– Patterned power supply with IGBT.

– RF acceleration cavity with 280 kV total voltage

– Slow extraction with 1% beam loss.

29

Characteristics of materials (Ferrite vs. FINEMET)

1.00E+09

1.00E+10

1.00E+11

1 10 100 1000 10000

Brf[Gauss]

up'Q

f

SY2

N5C

4M2-302

FT-small

FT-large

200V/div, 5ms/div

High Loss Effect

Magnetic Alloys

Ferrites

2000 Gauss

横軸：高周波磁束密度（電圧に比例）

縦軸

　

磁性体特性

（

インピー

ダンスに比

例

）

30

High gradient MA Cavity

Protron Synchrotron RF System

0

10

20

30

40

50

60

0 2 4 6 8 10 12

Frequency (MHz)

Fiel

d G

radi

ent (k

V/m

)SATUNEMIMASCERN PSBCERN PSAGS BSTRAGSISISKEK BSTRKEK PS J KJ 50GeV PSJ KJ 3GeV PS 50GeV PS UpgradeMITSUBISHIKEK- HGC

50 GeV Synchrotron

3GeV Synchotron

AGS

KEK PS

50 GeV PS UpgradeKEK- PS HGC

陽子加速器における高周波空洞の勾配

31

Cut Core

0.E+00

1.E+09

2.E+09

3.E+09

4.E+09

5.E+09

6.E+09

7.E+09

0 5 10

f

uQf w/ o Cut(#2- 9,

97/ 10)

Cut Core #2with Epoxy andSiO2, 2mmspacing (98/ 10)

Cut Core,O.D. 95cm

カットによるインピーダンスの変化

Increase Q-value by cutting the Core. Impedance is not changed.

32

Ｒ＆Ｄ of cavitiy

• Indirect cooling– Cu plate in between.

• Good thermal conductivity ～ 4 W/mK

• 5mm thickness is OK.

• Can use at 300 degree C.

• High Resistance, Strong for High Voltage Cutting core with water jet.

33Cavity with indirect cooling

34

Ceramic chamber

●TiN Coating to decrease secondary emission from surface Thickness : < 2 nm Requirement form Impedance Possible to use the Hollow Cathode Discharge Ion Plating Method

●RF Shield to decrease the impedance generated the beam Thickness,Width,Pitch : Performing to Estimate Possible to make Stripe with Metal Coat and Plating

●Duct Joint to make long duct ( > 3 m ) Two kinds of joint method (1) Glazing (2) Metalizing & Grazing

Ceramic Duct

35

prototype

•メタライズとロウ付けによりチャンバーを　接合。

- 接合強度がアップ。

•同様に金属フランジの接合に成功。- 金属フランジの使用が可能となった。

•表面ストライプによるＲＦシールドを構成。

•２次電子抑制のために内面ＴｉＮコーティ　ング

- 電子 陽子不安定性の抑制にも効ｰ果が期待される。

36

５０ＧｅＶＰＳ benging magnet （Ｒ＆Ｄ model ）

Ｇａｐ　Ｈｅｉｇｈｔ １０６ｍｍ

Ｕｓｅｆｕｌ　Ａｐｅｒｔｕｒｅ １２０ｍｍ

Ｆｉｅｌｄ ０．１４３－１．９Ｔ

Ｌｅｎｇｔｈ ５．８５ｍ

37

Field measurement of Ｒ＆Ｄ model

Good agreement with calculation so far.

38

５０ＧｅＶＰＳ quadrupole （Ｒ＆Ｄ model ）

Ｂｏｒｅ　Ｒａｄｉｕｓ６３ｍｍ

Ｕｓｅｆｕｌ　Ａｐｅｒｔｕｒｅ１３２ｍｍ

Ｍａｘ．Ｆｉｅｌｄ １８Ｔ／ｍ

Ｌｅｎｇｔｈ （ｍａｘ．） １．８６ｍ

39

Power supply （ＩＧＢＴ、 IEGT ）

• Rapid improvement of high power switching device makes the big change in magnet power supply for last 2-3 years.

40

Magnet power supply• Current pattern is controlled by the chopper downstream of

capacitance after the converter.

• No reactive power.

• １００ｋＨｚ switching 。• Ripple １０－６。• Tracking error １０－４。

41

ESS septum for slow extractionＲ & Ｄ

• We have achieved ２３７ kV– 1.4 times of the design value.

• Ceramic support is broken.

• Redesigning the shape and quality control of ceramic.

42

Future prospective

• Neutrino factory （ Japanese scenario ）– ＦＦＡＧ　ｂａｓｅｄ

• ５０ GeVPS 　ａｓ　Ｐｒｏｔｏｎ　Ｄｒｉｖｅｒ　ｏｆ　１ - ４ＭＷ。