Status of ATF Damping Ring low vertical emittance studies

2009.10.K. Kubo

Required or target of low vertical emittanceOriginal design: 12 pmAssumption in ATF2 design: 12 pmILC damping ring design: 2 pm

Extraction and Final Focus Test Beam Line

Injector Linac

Circumference: 139 mEnergy: 1.3 GeV

Test Facility for LC•Test of Low emittance beam tuning•Deliver low emittance beam, e.g. for final focus test (ATF2)•R & D of instrumentations, etc.

ATF Damping Ring (KEK)

Low Emittance Tuning

Three consecutive corrections• COD correction• Vertical COD-dispersion correction• Coupling correctionMonitor:

BPM (Beam Position Monitors) (total 96)

Corrector:

Steering magnets (47 horizontal and 51 vertical)

Skew Qauds (trim coils of sextupole magnets, total 68)

Skew correctors - trim coils of sextupole magnets

+ I

+ I

- I /2

- I /2- I /2

- I /2

The trim windings of all 68 sextupole magnets have been arranged to produce skew quadrupole fields, used as correctors.

Currents of the top and the bottom poles are the same. Currents of other poles are one half.

Skew Quad Field by Sextupole Magnet

(Suggested by T. Raubenheimer)

Low Emittance Tuning(a) COD correction: using steering magnets,

minimize BPM reading

and , :x(y): horizontal (vertical) BPM reading.

(b) V-COD-dispersion correction: using steering magnets,

minimize dispersion and orbit y: measured vertical dispersion.

r : weight factor = 0.05

(c) Coupling correction: using skew quads,

minimize vertical response to horizontal steering

x(y): horizontal (vertical) position change at BPM due to excitation of a horizontal steering magnet.

Two horizontal steering magnets were used (Nsteer=2). About (n+1/2)p phase advance between the two.

x2

BPM

BPM

2y

y2

BPM

r2 y2

BPM

Cxy y2

BPM

x2

BPM

H steers

Nsteer

Old History of low emittance in ATF DRBy 2004, we confirmed very low emittance beam, around 4 pm.Since then, pursuit of low emittance had not been a major study item. The emittance deteriorated. Over the past year, renewed efforts have been made for low emittance.

10-12

10-11

10-10

10-9

8/1/97 8/1/98 8/1/99 8/1/00 8/1/01 8/1/02 8/1/03 8/1/04

Single bunch emittance history

Y emittance (EXT wire, SR)Y emittance (LW)

Y e

mit

tanc

e [r

ad.m

]

Date

RF

gun

2µm

BP

M

210m

A l

imit

DR

Ion

Pu

mp

20µ

m B

PMEX

T l

ine

MB

op

erat

ion

star

t M

Bra

d.

shie

ld

rad

. sh

ield

X e

mi

con

firm

ed

GL

CT

A

rad

. sh

ield

refine beam tuning by H. Hayano

Recent efforts for low emittance

Re-Alignment of magnets

Related to beam measurement• BBA (Beam Based Alignment) measurement• Optics matching (Beta-beat correction)

• ORM (Orbit Response Matrix) analysis

Effectiveness of each still needs to be understood.

DR Magnet Level Measurement

M. Takano

3

0/2cos

nnn LsnC

Smooth curves fromfitting

No data for SF, SD and QF1R in the first three measurements.

Deviation from smooth curve

RMS(DY) [mm] MAR08: 0.086 APR08: 0.084 OCT08: 0.076 NOV08: 0.057 FEB09: 0.056

Improved?

M. Takano

Beam Based AlignmentSimulation of the tuning showed importance of BPM - magnet center offset error

Emittance vs. BPM offset error with respect to the nearest magnet

Beam Based Alignment - Method

1. Make vertical local bump at the Magnet-BPM.2. Change Magnet strength.3. Measure the orbit difference for all the BPMs.

Normal Quads: Vertical, Skew Quads: Horizontal

4. Estimate the minimum orbit difference point.

BPMBPMs BPMs

Magnet (quad or skew quad)

Trim winding of sextupole magnets

T. Okugi

Beam Based Alignment - results

0

5

10

15

20

25

30

35

40

-1500 -1000 -500 0 500 1000 1500

Ver

tica

l Orb

it D

iffe

renc

e (

m)

Veritical Position (m)

0

10

20

30

40

50

60

70

80

-1500 -1000 -500 0 500 1000 1500Hor

izon

tal O

rbit

Dif

fere

nce

(m

)

Veritical Position at BPM2 (m)

RMS of beam position change at all (except for noisy) BPMs vs. local bump amplitude.Fitted as 22 )( aycbRMS

ExamplesBump at Quad Bump at Skew Quad (Sextupole)

offsetMagnet -BPM :a

Typical error of offset: ~ 30 m for Quad ~ 80 m for Skew Quad

T. Okugi

Optics matching (Beta-beat Correction)

0

5

10

15

20

25

0 20 40 60 80 100 120 140 y (

m)

arc arcarc

0

5

10

15

20

0 20 40 60 80 100 120 140

y (m

)

s (m)

Calculated vertical beta functions of two different optics matching conditions.

December 1999, when we observed small vertical emittance (about 5pm).

May 2008, when we could not achieve low emittance.

Beta-beat can increase emittance sensitivity to errors.

Optics matching (Beta-beat Correction)• Start from setting a “design” optics. (2008 fall)• Measure beta-function at every quadrupole magnet

– Beta-beat was observed.

Correction based fully on the model– We tried three (?) times. Results were different from the

calculations.– Our model was not good enough for this methods. (?)

Partially rely on model, partially empirical method Concentrate on vertical beta-function at QF1Rs (vertical beta-beat in

arc sections) – Look for quadrupole magnets whose change would partly

correct the beta-beat at QF1Rs from optics model. – Some improvement

Optics matching (Beta-beat Correction)

0

5

10

15

20

0 5 10 15 20 25 30

Before correctionCorrected

y (m

)

Quad QF1R

West arc

East arc

Vertical beta-function at all quadrupole magnet of one family in the arc sections. (Should be flat for matched optics.)

Before and after a beat correction. (Example)

Not completely satisfactory. Need more study for better modeling.But condition was improved.

ORM Analysis• Measure changes in the closed orbit with

respect to changes in strength of a number of orbit correctors

• Fits a machine model to the data, by adjusting:– Quadrupole strengths, – BPM gains and couplings,– Corrector magnet strengths and tilts.

Need more studies for understanding how to use information from ORM.

Coupling correction using ORM Analysis

ORM analysis effectively projects the betatron coupling sources onto the skew quadrupolesDetermination of skew quadrupole strengths required to cancel the coupling sources

Possible limitations• Present analysis do not include orbit distortion

– which can affect predictions of effects of correctors– will be tried after BBA and more accurate orbit corrections

• Degeneracy between errors causing apparent coupling

ORM AnalysisCorrelation between • Changes in skew quadrupole strengths determined from ORM analysis, and • Known changes in currents applied to correctors

Deviations from the straight line indicate error of present model.

Recent history of emittance in ATF DR

Vertical emittance < 10 pm (from Laser Wire measurement)Smaller than limits of other monitors?

0

10

20

30

40

50

60

Nov/1/2007 Mar/1/2008 Jul/1/2008 Nov/1/2008 Mar/1/2009 Jul/1/2009

y (pm) X-SR

SR-ILW

y (p

m)

S. Kuroda and N. Terunuma

Example of DR Laser Wire measurement

s = 8.6 m(convolution of beam size andlaser size)

H. Shimizu

Summary and Future Plans • Low emittance tuning and efforts for improving DR emittance

– Re-alignment– BBA (BPM - Magnet offset measurement)– Optics matching (Beta-beat correction)– ORM (Orbit Response Matrix) analysis

• The emittance performance has been recovered.– y < 10 pm in April and May 2009. Good enough for FF test.– Effectiveness of each item for this recovery is not clear yet.

• Plans for smaller emittance (2 pm is ILC DR design.), – More simulation studies on the tuning procedure– Analysis of beam measurement, e.g. ORM.– Upgrade of all BPM electronics (20 out of 96 BPMs were already

upgraded) – Re-alignment of magnets.

Backup Slides

DR Magnet Level Measurement

Average of 100 seeds 90% CL (90th among 100 seeds)

0

2 10-11

4 10-11

6 10-11

8 10-11

1 10-10

1.2 10-10

1.4 10-10

0 0.1 0.2 0.3 0.4 0.5 0.6

Alignment 2008 Feb.Alignment 1999

90%

y (

m)

BPM offset w.r.t. magnet (mm)

0

2 10-11

4 10-11

6 10-11

8 10-11

1 10-10

0 0.1 0.2 0.3 0.4 0.5 0.6

Alignment 2008 Feb.Alignment 1999

Mea

n

y (m

)

BPM offset w.r.t. magnet (mm)

Emittance vs. BPM offset w.r.t. nearest magnet

with two different magnet misalignment

1999 data and 2008 data used different measurement methods.Measurement error may be dominant in2008 data. Can not tell change of alignment conditions from these data.

Result of tuning simulation, 3 optics Number of random seeds giving resultsEmittance vs. BPM-Magnet offset error

0

5 10-12

1 10-11

1.5 10-11

2 10-11

2.5 10-11

3 10-11

0 0.1 0.2 0.3 0.4 0.5

optics1999Dec

optics2008May

opticsBad

< y>

(m

)

BPM-Quad misalignment (mm)

For some random seeds, SAD cannot find closed orbit, betafunctions, etc.The left figure shows number of seeds out of 100, which give results.The reason was not well studied and how to treat these results is not clear.

0

20

40

60

80

100

0 0.1 0.2 0.3 0.4 0.5

optics1999Decoptics2008MayopticsBad

Solv

ed n

umbe

r of

see

ds

BPM-Quad misalignment (mm)

Beam Based Alignment

-30

-20

-10

0

10

20

30

-500 0 500 1 103 1.5 103 2 103 2.5 103

BBA for Skew Quad

Maget Trim I = -1 AMaget Trim I = +1 A

BP

M60

X [m

]

BPM41 Y [m]

-30

-20

-10

0

10

20

30

40

-1.5 103 -1 103 -500 0 500 1 103 1.5 103

BBA for Quad

Magnet Trim = -1 AMagnet Trim = +1 A

BP

M4

Y [m

]

BPM1 Y [m]

Position at BPM outside local bump vs. amplitude of bump

Bump at Quad Bump at Skew Quad

Examples

Example of data from coupling correction

Changes in vertical orbit response to a horizontal steering.Measurement and prediction from present model.

-0.2

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0.2

0 20 40 60 80 100 120 140

ZH1RMeasuredModel

s (m)

Change of Response at all BPM to one horizontal steering magnet.

Measured April 10, 2009

ORM AnalysisResponse ｏｆ all BPMs to all horizontal steering magnets (But omitted if error<0.03)

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

-0.15 -0.1 -0.05 0 0.05 0.1 0.15

Mea

sure

d r

espo

nse

chan

ge (

m/r

ad)

Model response change (m/rad)

Measured April 10, 2009