Massimo GiovannozziHB2006 - May 29 - June 2 20061 DESIGN AND TESTS OF A LOW- LOSS MULTI-TURN EJECTION FOR THE CERN PS M. Giovannozzi For PS Multi-Turn

Massimo GiovanMassimo Giovannozzinozzi

HB2006 - May 29 - June 2 HB2006 - May 29 - June 2 2006 2006

11

DESIGN AND TESTS OF A LOW-DESIGN AND TESTS OF A LOW-LOSS MULTI-TURN EJECTION LOSS MULTI-TURN EJECTION

FOR THE CERN PSFOR THE CERN PS

M. GiovannozziM. GiovannozziForFor

PS Multi-Turn Extraction Study GroupPS Multi-Turn Extraction Study Group

Summary:Summary:

Introduction Introduction

Present multi-turn Present multi-turn extractionextraction

New multi-turn extraction New multi-turn extraction (MTE)(MTE)

Measurement resultsMeasurement results

Implementation of Implementation of MTEMTE

Losses estimates Losses estimates



22

IntroductionIntroduction

Multi-turn extractionMulti-turn extractionThe beam has to be “The beam has to be “manipulatedmanipulated” to increase the ” to increase the effective length beyond the machine circumference.effective length beyond the machine circumference.

This extraction mode is used to This extraction mode is used to transfer beam transfer beam between circular machinesbetween circular machines..

AT CERN this mode is used to AT CERN this mode is used to transfertransfer the proton the proton beam beam betweenbetween PSPS and and SPSSPS. In the SPS the beam is . In the SPS the beam is used for used for

Fixed TargetFixed Target physics (broad sense) physics (broad sense)

Neutrino experimentsNeutrino experiments (until 1998) (until 1998)

CERN Neutrino to Gran SassoCERN Neutrino to Gran Sasso (CNGS) (from 2006) (CNGS) (from 2006)

These beams are These beams are high-intensityhigh-intensity (about (about 33××10101313 p p in in the PS). CNGS requested to receive even more beam the PS). CNGS requested to receive even more beam (about (about 4.84.8××10101313 p p in the PS). in the PS).



33

Electrostatic septum Electrostatic septum bladeblade

Present multi-turn extractionPresent multi-turn extraction – – II

Length Length

Kic

ker

Kic

ker

str

en

gth

str

en

gth Four Four

turnsturns

Fifth Fifth turnturn

XX

X’X’

1133 55

22

44

Slow Slow bumpbump

Slow Slow bumpbump

Electrostatic Electrostatic septum septum (beam shaving)(beam shaving)

Extraction Extraction septumseptum

Kicker magnets Kicker magnets used to used to generate a generate a closed orbit closed orbit bump around bump around electrostatic electrostatic septumseptum

Extraction Extraction lineline

EEfieldfield=0=0

EEfieldfield≠0≠0



44

Present multi-turn extractionPresent multi-turn extraction – – IIII

First PS First PS batchbatch

Second PS Second PS batchbatch

Gap for Gap for kickerkicker

CCSPSSPS = 11 C = 11 CPSPS

PSPSPSPS

SPS SPS circumferencecircumference

Beam current Beam current transformer in the transformer in the PS/SPS transfer PS/SPS transfer lineline 1 2 3 4 5 (total spill duration 0.010 1 2 3 4 5 (total spill duration 0.010

ms)ms)



55

Present multi-turn extractionPresent multi-turn extraction – –IIIIII

The main drawbacks of the present The main drawbacks of the present scheme are:scheme are:

Losses (about Losses (about 15%15% of total intensity) are of total intensity) are unavoidable due to the presence of the unavoidable due to the presence of the electrostaticelectrostatic septum used to slice the beam. septum used to slice the beam.

The The electrostaticelectrostatic septum is septum is irradiatedirradiated. This . This poses problems for poses problems for hands-on maintenancehands-on maintenance..

The phase space matching is not optimal (the The phase space matching is not optimal (the various slices have “various slices have “fancy shapesfancy shapes”), thus ”), thus inducing betatronic mismatch in the receiving inducing betatronic mismatch in the receiving machine, i.e. machine, i.e. emittanceemittance blow-upblow-up..

The slices have different emittances and The slices have different emittances and optical parameters.optical parameters.



66

Novel multi-turn extractionNovel multi-turn extraction – – II

The main ingredients of the novel extraction:The main ingredients of the novel extraction:

The beam splitting is not performed using a The beam splitting is not performed using a mechanical device, thus avoiding losses. Indeed, the mechanical device, thus avoiding losses. Indeed, the beam is separated in the transverse phase space beam is separated in the transverse phase space using using

Nonlinear magnetic elementsNonlinear magnetic elements (sextupoles ad (sextupoles ad octupoles) to create octupoles) to create stable islandsstable islands..Slow (Slow (adiabaticadiabatic) tune-variation to cross an ) tune-variation to cross an appropriate resonance.appropriate resonance.

This approach has the following beneficial effects:This approach has the following beneficial effects:Losses are reducedLosses are reduced (virtually to zero). (virtually to zero).The The phase space matching is improvedphase space matching is improved with respect with respect to the present situation.to the present situation.The beamlets have the The beamlets have the same emittancesame emittance and and optical optical parametersparameters..



77

Novel multi-turn extractionNovel multi-turn extraction – – IIII

Right: Right: intermediate intermediate phase space phase space topology. Islands topology. Islands are created near are created near the centre.the centre.

Bottom: final Bottom: final phase space phase space topology. Islands topology. Islands are separated to are separated to allow extraction.allow extraction.

Left: initial phase Left: initial phase space topology. space topology. No islands.No islands.



88

Novel multi-turn extraction - Novel multi-turn extraction - III III

Tune Tune variationvariation

Phase Phase space space portraitportrait

Simulation Simulation parametersparameters::

Hénon-like Hénon-like map (i.e. map (i.e. 2D 2D polynomial polynomial – – degree 3degree 3 - - mapping) mapping) representinrepresenting a FODO g a FODO cell with cell with sextupole sextupole and and octupoleoctupole



99

Novel multi-turn extractionNovel multi-turn extraction – IV – IV

Final stage after 20000 turns (about 42 ms for CERN Final stage after 20000 turns (about 42 ms for CERN PS)PS)

About 6 cm in physical About 6 cm in physical spacespace

Slow (Slow (few thousand few thousand turnsturns) bump first ) bump first (closed distortion (closed distortion of the periodic of the periodic orbit)orbit)Fast (Fast (less than one less than one turnturn) bump ) bump afterwards (closed afterwards (closed distortion of distortion of periodic orbit)periodic orbit)

BBfieldfield ≠ 0≠ 0BBfieldfield = = 0 0At the septum locationAt the septum location



1010

Experimental results - IExperimental results - I

Experimental tests were undertaken since Experimental tests were undertaken since 20022002..

2002 run2002 run: proof-of-principle of the capture : proof-of-principle of the capture process using a low intensity beam.process using a low intensity beam.2003 run2003 run: detailed study of capture process : detailed study of capture process with low-intensity beam and first tests with with low-intensity beam and first tests with high-intensity proton beam.high-intensity proton beam.2004 run2004 run: main focus on high-intensity : main focus on high-intensity beam to solve problems observed in beam to solve problems observed in 20032003..

Overall strategy:Overall strategy:Phase spacePhase space reconstruction using low- reconstruction using low-intensity, pencil beam.intensity, pencil beam.CaptureCapture with with low-intensitylow-intensity, large horizontal , large horizontal emittance beam.emittance beam.CaptureCapture with with high-intensityhigh-intensity beam. beam.



1111

Extraction line

Slow bump

Sextupole magnets

Extraction septum

Octupole magnets

Flying wires

Kicker magnet

Section 71

Section 64

Section 55

Section 16

Section 21

Section 20

Section 54

Extraction line

Slow bump

Sextupole magnets

Extraction septum

Octupole magnets

Flying wires

Kicker magnet

Section 71

Section 64

Section 55

Section 16

Section 21

Section 20

Section 54Experimental results - IIExperimental results - II

Key elements for Key elements for experimental experimental tests.tests.

Phase space Phase space reconstruction is reconstruction is based on fast based on fast digitiser applied digitiser applied to closed orbit to closed orbit pick-ups.pick-ups.

Key elements for Key elements for experimental experimental tests.tests.

Phase space Phase space reconstruction is reconstruction is based on fast based on fast digitiser applied digitiser applied to closed orbit to closed orbit pick-ups.pick-ups.



1212

Experimental results - IIIExperimental results - III

The pencil beam is The pencil beam is kicked intokicked into the islands producing a the islands producing a strong strong coherent signalcoherent signal (filamentation is (filamentation is suppressed).suppressed).Initial Initial wiggleswiggles represent represent beam beam oscillationsoscillations around the around the islands’ islands’ centre.centre.

Normalized Coordinates

-15 -5 5 15-15

15

X

-5

5

H Signal PU1: blue - PU2: red

50 150 250 350 450 550 650 750 850 950 1050 1150 1250 1350 1450 1550 1650 1750 1850 1950-20-10

01020

Turn

y = -0.9211 x + 0.0398

y = -0.7566 x + 0.0408

0.020

0.025

0.030

0.035

0.040

0.045

0.000 0.002 0.004 0.006 0.008 0.010 0.012

Linear invariant

Sec

ond

ary

freq

uen

cy

Measurement results

Numerical simulations

Fit Pesato

Linear (Numericalsimulations)

Measured detuning Measured detuning inside an island inside an island compared to numerical compared to numerical simulations.simulations.

Measured detuning Measured detuning inside an island inside an island compared to numerical compared to numerical simulations.simulations.



1313

Data analysis and beam Data analysis and beam parametersparameters

The wire scanner is the key The wire scanner is the key instrument for these studies.instrument for these studies.

Raw data are stored for off-line Raw data are stored for off-line analysis.analysis.

Five Gaussians are fitted to Five Gaussians are fitted to the measured profiles to the measured profiles to estimate beam parameters of estimate beam parameters of five beamlets.five beamlets.

Beam parametersBeam parametersIntensityIntensity **HH(()/)/**VV(())

Low-intensity Low-intensity pencil beampencil beam 55××10101111 2.3/ 1.32.3/ 1.3

Low-intensity Low-intensity large H emittancelarge H emittance 55××10101111 6.2/ 1.66.2/ 1.6

High intensity High intensity beambeam 66××10101212 9.4/ 6.49.4/ 6.4



1414

Influence of octupole Influence of octupole strengthstrength

Octupole action Octupole action

Island size.Island size.

Detuning with Detuning with amplitude.amplitude.

Problems with the Problems with the fitfit



1515

Crucial part: high-intensity Crucial part: high-intensity beam - Ibeam - I

Reduction of octupole Reduction of octupole strength to move the strength to move the beamlets outwardsbeamlets outwards

- 5.0

- 4.0

- 3.0

- 2.0

- 1.0

0.0

1.0

2.0

3.0

4.0

5.0

0 200 400 600 800 1000 1200 1400 1600Time (ms)

A.

U.

New Octupole function Standard Octupole function

Magnetic field Focusing quadrupoles

Defocusing quadrupoles

14 GeV/c flat-top14 GeV/c flat-top

1.4 GeV flat-1.4 GeV flat-bottombottom

Tune Tune sweep sweep



1616

After optimisation of After optimisation of transverse and longitudinal transverse and longitudinal parametersparameters

Capture losses are Capture losses are reduced to zero…reduced to zero…

Horizontal beam Horizontal beam profileprofile

Horizontal beam Horizontal beam profileprofile

Depleted region: Depleted region: extraction septum extraction septum blade will not blade will not intercept any intercept any particleparticle

Depleted region: Depleted region: extraction septum extraction septum blade will not blade will not intercept any intercept any particleparticle



1717

A movie to show the A movie to show the evolution of beam evolution of beam distributiondistribution

The high-intensity The high-intensity beam is fast beam is fast extracted towards extracted towards the dump D3. the dump D3.

Prior to extraction Prior to extraction beamlets are beamlets are partially merged partially merged back with central back with central core.core.Beamlets projected onto x-Beamlets projected onto x-

axisaxis

Beamlets projected onto x-Beamlets projected onto x-axisaxis



1818

Best result in terms of Best result in terms of capturecapture

Assuming that:Assuming that:

Beamlets are Beamlets are affected by a affected by a different solid angledifferent solid angle

Beamlets are fitted Beamlets are fitted using five gaussians. using five gaussians. Instead of imposing Instead of imposing the same integral for the same integral for the four beamlets the four beamlets (physical (physical arguments), only arguments), only three have such a three have such a constraint (solid constraint (solid angle consideration). angle consideration).

Fit constraint: Fit constraint: same integralsame integral

CaptuCapturere

18%18% three three rightmost rightmost beamletsbeamlets

16%16% single single leftmost beamletleftmost beamlet

Scintillator Scintillator is on this is on this side!side!



1919

Implementation of MTEImplementation of MTE - I - I

Three main items:Three main items:Generation of stable islandsGeneration of stable islands

Extraction proper:Extraction proper:

Slow bump to approach the Slow bump to approach the septumseptum

Fast bump to jump septumFast bump to jump septum

Generation of stable islandsGeneration of stable islandstwo pairs (spaced by 2two pairs (spaced by 2) of ) of

two sextupoles two sextupoles

one octupole one octupole

will be usedwill be used



2020

Implementation of MTEImplementation of MTE - II - II

Extraction proper: slow bumpExtraction proper: slow bumpSix dipoles, independently powered, are foreseen. Six dipoles, independently powered, are foreseen. Large number of magnets -> optimal bump shape.Large number of magnets -> optimal bump shape.Present slow bump: four dipoles powered with a Present slow bump: four dipoles powered with a series/parallel circuit.series/parallel circuit.

-20

-10

0

10

20

30

40

50

60

70

0 100 200 300 400 500 600 700

s coordinate (m)

Bum

p a

mp

litu

de (

mm

)

Slow bump for CTSlow bump for MTE

Origin of s coordinate: SS54



2121

Implementation of MTEImplementation of MTE - III - III

-10

0

10

20

30

40

50

60

0 100 200 300 400 500 600 700

s coordinate (m)

Bum

p a

mplitude (

mm

)

Origin of s coordinate: SS54

Extraction proper: fast bumpExtraction proper: fast bumpFive kicker systems in the PS ring. Five kicker systems in the PS ring. Two kickers to correct the extraction trajectories Two kickers to correct the extraction trajectories in the transfer line.in the transfer line.Maximum kick about 1.8 mrad at 14 GeV/Maximum kick about 1.8 mrad at 14 GeV/c.c.

Fast bump for Fast bump for extracting the extracting the fifth turn (centre fifth turn (centre core)core)



2222

Summary of changes in the PS Summary of changes in the PS ring ring

Legend: Legend: SS -> Straight Section.SS -> Straight Section.

MU -> Magnet Unit.MU -> Magnet Unit.

Red circle -> “heavy” Red circle -> “heavy” intervention: intervention: mechanical design, mechanical design, vacuum intervention.vacuum intervention.

Orange circle -> “light” Orange circle -> “light” intervention: auxiliary intervention: auxiliary magnet exchange.magnet exchange.

SS02SS02

SS13SS13SS12SS12 SS08SS08

SS04SS04SS03SS03

SS21SS21

SS20SS20

SS22SS22

MU14MU14

SS15SS15

MU15MU15

MU16MU16

MU18MU18

MU19MU19

SS60SS60

SS35SS35 SS39SS39SS55SS55

SS18SS18

SS19SS19SS68SS68

SS74SS74

Extraction region

Extraction region

Sextupoles Sextupoles and and octupolesoctupoles



2323

Critical issues: available Critical issues: available mechanical aperture - Imechanical aperture - I

100 200 300 400 500 600

100

50

50

100

150

mm1 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100SLOWBUMP. HOR. nt5



2424

Critical issues: available Critical issues: available mechanical aperture - IImechanical aperture - II

80 100 120 140

100

50

50

100

150

mm1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100FASTBUMP. HOR. nt5



2525

Time scale of Time scale of changeschanges

Install slow extraction sextupoles inInstall slow extraction sextupoles in SS03SS03, keeping , keeping those inthose in SS19SS19..

Replace magnets of slow bumpReplace magnets of slow bump 1616 with typewith type 205205 magnets.magnets.

General clean-up of the machine.General clean-up of the machine.

Remove slow extraction sextupoles inRemove slow extraction sextupoles in SS19 SS19..

Install Install new power converters for bump 16.new power converters for bump 16.

Install kickers.Install kickers.

Install modified vacuum chambers (straight Install modified vacuum chambers (straight sections and magnets).sections and magnets).

Install sextupoles and new octupoles.Install sextupoles and new octupoles.

Move cavity.Move cavity.

Install new wire scanner.Install new wire scanner.

05/0605/06

06/0706/07

07/0807/08



2626

Kic

k a

mp

litu

de

(s)

Kic

k a

mp

litu

de

(s)

Losses estimates: CTLosses estimates: CT - - II

Assumptions for analytical Assumptions for analytical estimatesestimates

Gaussian distribution (transverse).Gaussian distribution (transverse).

Parabolic distribution Parabolic distribution (longitudinal).(longitudinal).

Kickers rise time (5%-95%): 820 nsKickers rise time (5%-95%): 820 ns

Mac

hine c

ircum

fere

nce

Mac

hine c

ircum

fere

nce

Measured values Measured values ofof

Septum Septum thicknessthickness

Septum angleSeptum angle

Beam emittanceBeam emittance

t (t (s)s)



2727

Losses estimates: CTLosses estimates: CT - - IIII

Longitudinal structure

1L p1=emittance

2L p2=septum angle

3L p3=kickers’ rise-time

4L p4=septum width

Continuous 7.51 12.18 0.38 2.88 Bunched h=16 6.52 11.24 2.02 2.63 Bunched h=8 6.27 10.86 2.59 2.05

Normalised Jacobian of Loss Normalised Jacobian of Loss function function

iopii ppLLio

/



2828

Losses estimates: MTELosses estimates: MTE

Losses (%) ->Losses (%) -> ContinuouContinuouss

BuncheBunched d (h=16)(h=16)

BuncheBunched (h=8)d (h=8)

Nominal scheme Nominal scheme (recuperated kickers, (recuperated kickers,

slow rise time)slow rise time)11 0.90.9 0.60.6

Total Total (capture+extraction)(capture+extraction) 3-43-4 2.9-3.92.9-3.9 2.6-3.62.6-3.6

Upgrade (improved Upgrade (improved kickers, faster rise time)kickers, faster rise time) 0.60.6 0.50.5 < 0.1< 0.1

Total Total (capture+extraction)(capture+extraction) 2.6-3.62.6-3.6 2.5-3.52.5-3.5 2.1-3.12.1-3.1

Upgrade (reduced Upgrade (reduced thickness of magnetic thickness of magnetic

septum)septum)0.60.6 0.50.5 0.30.3

Total Total (capture+extraction)(capture+extraction) 2.6-3.62.6-3.6 2.5-3.52.5-3.5 2.3-3.32.3-3.3

No more slicing -> capture lossesNo more slicing -> capture losses

Overall extraction losses -> interplay between Overall extraction losses -> interplay between

kicker rise timekicker rise time

bunch structure bunch structure

septum thicknessseptum thickness

For the nominal MTE scheme the For the nominal MTE scheme the losses are reduced by a factor 3-4 losses are reduced by a factor 3-4 with respect to CT!with respect to CT!

Even higher reduction could be Even higher reduction could be expected (capture losses).expected (capture losses).

For the nominal MTE scheme the For the nominal MTE scheme the losses are reduced by a factor 3-4 losses are reduced by a factor 3-4 with respect to CT!with respect to CT!

Even higher reduction could be Even higher reduction could be expected (capture losses).expected (capture losses).



2929

The members of the PS Multi-The members of the PS Multi-Turn Extraction Study GroupTurn Extraction Study Group

M. J. BarnesM. J. Barnes**, O. E. Berrig, A. Beuret, J. , O. E. Berrig, A. Beuret, J. Borburgh, P. Bourquin, R. Brown, J.-P. Borburgh, P. Bourquin, R. Brown, J.-P. Burnet, F. Caspers, J.-M. Cravero, T. Burnet, F. Caspers, J.-M. Cravero, T. Dobers, T. Fowler, S. Gilardoni, M. Dobers, T. Fowler, S. Gilardoni, M. Giovannozzi (Study Group Leader), M. Giovannozzi (Study Group Leader), M. Hourican, W. Kalbreier, T. Kroyer, F. di Hourican, W. Kalbreier, T. Kroyer, F. di Maio, M. Martini, E. Métral, V. Mertens, K. Maio, M. Martini, E. Métral, V. Mertens, K. D. Metzmacher, C. Rossi, J.-P. Royer, L. D. Metzmacher, C. Rossi, J.-P. Royer, L. Sermeus, R. Steerenberg, G. Villiger, T. Sermeus, R. Steerenberg, G. Villiger, T. Zickler.Zickler.

*On leave from TRIUMF – CA*On leave from TRIUMF – CA



3030

Novel multi-turn extraction Novel multi-turn extraction with other resonanceswith other resonances

The fifth-order resonance is The fifth-order resonance is used, thus giving a six-turn used, thus giving a six-turn extractionextraction

The second-order The second-order resonance is used, thus resonance is used, thus giving a two-turn giving a two-turn extractionextraction



3131

Novel multi-turn Novel multi-turn injectioninjection: : new application!new application!

Simulation Simulation parameters:parameters:

Third-order Third-order polynomial map polynomial map representing a representing a FODO cell with FODO cell with sextupole and sextupole and octupoleoctupole

The fourth-order The fourth-order resonance is used resonance is used for a four-turn for a four-turn injectioninjection

Tune Tune variationvariation

Phase Phase space space portraitportrait

Efficient method to Efficient method to generate hollow generate hollow beams!beams!

Efficient method to Efficient method to generate hollow generate hollow beams!beams!

Study in progress with Study in progress with the contribution by J. the contribution by J. Morel.Morel.

Study in progress with Study in progress with the contribution by J. the contribution by J. Morel.Morel.

Documents

Massimo GiovannozziHB2006 - May 29 - June 2 20061 DESIGN AND TESTS OF A LOW- LOSS MULTI-TURN EJECTION FOR THE CERN PS M. Giovannozzi For PS Multi-Turn