linac4 to psb transfer line (slides by m. garcia-tudela & a. lombardi slightly adapted)

Status of the LINAC4 to PSB beam dynamics studies and open questions

Christian Carli 25th May 2011On behalf of many people contributing

Linac4 to PSB transfer line (slides by M. Garcia-Tudela & A. Lombardi slightly adapted)Strong space charge regime and simulationsRecap on plans for the PSB with Linac4Some (not that) recent ResultsStatus and OutlookLosses, activation, collimation A few words on 2 GeV upgrade and RCS studyStatus and Plans and open Questions


L4 to PSB line: Motivation(slides by M.Garcia & A.Lombardi slightly adapted)L4T new layout in order to make power converters suitable for quadrupoles to be constructed. Possible solutions Quads:Longer quadrupolesAperture reduction Keep 50 mm radiusBeam dynamics:Reduce the integrated gradientHighest gradients in the vertical offset Increased distance between the magnets25th May 2011Status of the LINAC4 to PSB beam dynamics studies and open questions


L4 to PSB line: Modified Vertical offset

Before : Steep slope required, 28 deg, 5.35 mIncreased distance between the magnets (equal spacing between vertical bends and quads)Factor of 2.5 gained in integrated gradient.

25th May 2011Status of the LINAC4 to PSB beam dynamics studies and open questions


L4 to PSB line: Proposal to modify first horizontal deflectionIdea: to Reduce the distance between MBH.150 and MBH.350 to 8m.Not required to reduce maximum gradient, but motivated by beam dynamics, for:First horizontal bending further from Linac output where the beam is still bunched. ~3.6mReduce the space charge effects.Make easier the matching downstream.



L4 to PSB line: Comparison between layouts~ 8m10 m5.35 m13.82mPrevious LayoutProposed LayoutBHZ.20BHZ.2025th May 2011Status of the LINAC4 to PSB beam dynamics studies and open questions


L4 to PSB line: Top View proposed layoutNb: Final positions to be discuss with the design office. Not corresponding to the MADX file.25th May 2011Status of the LINAC4 to PSB beam dynamics studies and open questions


L4 to PSB line: Quad settingsPrevious : 17 Quads (1 PIMS type)Proposed : 16 quads (1 or 2 PIMS type)L = 250 mm ALL QUADS BELOW 1.8 T25th May 2011Status of the LINAC4 to PSB beam dynamics studies and open questions


L4 to PSB line: Beam dynamics (New Part)Previous LayoutProposed Layout25th May 2011Status of the LINAC4 to PSB beam dynamics studies and open questions


L4 to PSB line: Up to PSB Injection Dx=025th May 2011Status of the LINAC4 to PSB beam dynamics studies and open questions


L4 to PSB line: Status and next stepsFeasible? Bend and quad positions o.k. for integration (discussion last Wednesday) Ordered converters now suitable with weaker magnetsNext steps: Review diagnostics and correctors (new positions what about numbers) 14 correctors foreseen for steering, magnet experts plan for 5 correctorsRestrictions for SE monitors due to losses? Sensitivity studies (expected to be o.k. similar beam sizes & weaker quads) (Rematch of the line with new quads (if necessary))Re-discuss integration and make sure that plans are consistent



Status of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011Strong space charge regime and simulations -Recap on plans for the PSB with Linac4Rationale for Linac4:PS Booster limited by direct space charge effects, quantified by direct space charge tune shift Q, at injection energyIncrease injection energy such that the brightness can doubled (double intensity within unchanged normalized emittances) without changing QRaise injection energy from 50 MeV with Linac2 to 160 MeV with Linac4Convert to H- charge exchange injectionReach the Phase Space Density expected in the PSBAllows - with a Linac4 Beam chopper - for longitudinal PaintingNew injection hardware for increased injection energy and charge exchange injection

Next bottleneck around the corner: PS and SPS at low energy


Status of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011Strong space charge regime and simulations -Recap on plans for the PSB with Linac4Longitudinal PaintingPrinciple:Triangular energy modulation (slow, ~20 turns for LHC)Beam on/off if mean energy inside a contur ~80% of acceptanceNominal LHC: intensity with 41mA (!!!) after 20 turns High intensity: several and/or longer modulation periodsPotential limitations: Linac4 jitter, debunching of Linac4 structure in BoosterNote that D ~ 1.4 m at the PSB injection


Strong space charge regime and simulations -Recap on plans for the PSB with Linac4H- charge exchange injection and transverse paintingInjection chicane required to merge incoming H- beam with circulating protonsNo use of main bend field (as done e.g. at BNL)Little space available in injection section (existing machine) to install chicane BS magnets

Perturbation of lattice due to BS magnets adding focusingStrong beta-beating in vertical plane due to tune close to (and even above) half-integer (mandatory for good performance with high intensity/brightness beams)Make BS magnets as long as possible to reduce additional focusingCompensate lattice perturbations (as much as possible) by additional quadrupolar fields

Status of the LINAC4 to PSB beam dynamics studies and open questionsIncoming H-Reference trajectory without bumpsClosed orbitwith chicane andpainting bumpBS1BS2BS3BS4Injection straight at the beginning of the injection (trajectory of incoming H- coincides with closed orbit)Stripping foilBending magnet25th May 2011


Strong space charge regime and simulations -Recap on plans for the PSB with Linac4Active compensation of injection chicaneLook for appropriate locations to add (time-varying) quadrupolar fieldsTrims on QDE magnets (on additional windings ) in period 03 and 14 with appropriate phasesEffective compensation with large vertical b-functions (small horizontal perturbation small hor. bs)

Slow chicane fall (say 5 ms) to make sure that quad trim PSs can follow programmed currentsPassive compensationDo not use R-Bends, but add pole-face rotations to bring part of perturbation in (less sensitive) horizontal planeNo perfect compensation during chicane fallFast chicane fall possible without compensation currents (if Eddy currents are nor a problem)Status of the LINAC4 to PSB beam dynamics studies and open questionsInjection straight sectionQDE03 and QDE14appropriate for Compensating 2QV = 925th May 2011


Strong space charge regime and simulations -Recap on plans for the PSB with Linac4Active compensation with trims on appropriate QDE quads:Proposal (M. Aiba, B. Goddard ..) for additional quads ruled out initially: complicated scheme with similar b-function in both planes for quads in straightsTrims on QDE quadrupoles are appropriate (chicane made of rectangular bends):Large (b-function and) compensation in vertical plane with little (b-function and) perturbation in the horizontal planeIn practice: trims on appropriate Q-strips (additional windings of PSB quads)Appropriate quads for lower vertical tune: QDE03 and QDE14

Compensation during (no compromise) whole chicane fallPreliminary discussion on Power Converters: feasible provided chicane fall time is not too short (at least a few ms)

Status of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011-beating with (dashed) and without (solid) compensationBlue: vertical; red: horizontal


Strong space charge regime and simulations Some (not that) recent ResultsTime dependent lattice implemented following a procedure proposed by M. Aiba (found by looking into the code!!) Chicane magnets modeled as bends with time-varying properties during chicane fall For active compensation, variation of quadrupole components as well. Lattice regularly (every turn during chicane fall) read from file, properties of some magnets redefined

Emittance evaluations: Dispersion (from file) used for emittance evaluations not any more the one of the lattice in the simulation (magnet properties redefined)Emittances and lattice functions from analysis of dumped macro-particle distributions

Simulations: both active and passive compensation work (formar with slighty less losses)active compensation will be used (simpler for hardware)

Status of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011


Strong space charge regime and simulations Some (not that) recent ResultsComparison of compensation schemes (M. Scholz, M. Martini )

Status of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011eV (mm)Plots from Michels presentation on 2nd december 2010


Strong space charge regime and simulations Some (not that) recent ResultsStudy on painting with slower energy modulation (M.Martini, J.-B. Lallement)Injection painting bump optimization (C.Bracco et al.)

Status of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011Plots from Chiaras presentation on 2nd december 2010


Strong space charge regime and simulations Status and OutlookSimulations with original ORBIT version concentrated on Injection Studies Shorter duration, blow-up due to injection hardware easier to assess Choice of active compensation of chicaneMore recently (still with original ORBIT version) First studies on longer energy modulation period (J.-B. Lallement and M. Martini) Studies on transverse painting (C. Bracco)

Imperfections (multipoles) of chicane magnets to be modeled (urgent!!)Lattice imperfections, slow blow-up, working point studies Feasibility, resources can only be a long-term aim (dream) in present situation

PTC ORBIT version installed at present at CERNVery recently still discussions on basics for time-dependent latticeAdvantage: easier for ORBIT simulations with lattice imperfections and time-dependenceStatus of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011


Losses, activation, collimation Impact of less losses at higher energy on activation and operationStudy on hold due to lack of resources potential showstopper for connection in LS1??Very first study on collimation by a doctoral student (P.Jackson)First approach: multistage system as in LHC however without possibility to place many secondary collimatorsVery very thin primary collimator needed at these low energiesExcessive heating for full beam loss

May-be a single stage system feasible?? .. Better understanding of what happens with lost particleStatus of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011


A few words on 2 GeV upg. and RCS studyStudy in 2010 on 2 GeV upgrade with little involvement from our side Our initiative to start some studies Study on transfer of 2 GeV LHC beams with present kickers Is it really necessary to exchange the MPS (upgrade required in any case)? Discussions (recent) on transfer of high intensity hPSB=2 beams to PSABP contributions to present RCS studies mainly by M.Fitterer Very quickly decision for longer version (4/21) times the PS and construction outside the PS (justifications??) Several interesting lattices proposed (triplet versus doublet, triangular, square, racetrack, different number of cells kind of dispersion suppression scheme with contrib. by B.Holzer)Decision to stick to lattice used for CE estimatesUnclear situation on contributions expected from usE.g.: design of the transfer line?Status of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011


Status and Plans and open QuestionsLinac4 to PSB transfer line (new task) Geometry and optics (Alessandra & Margarita) hopefully frozenNext steps: instrumentation, imperfections/steering, discussions to make sure everything is consistent measurement line to be looked into (LBS) urgentlyLongitudinal painting Basic principles for implementation discussed and documentedAdaptation of PSB injection timing and application to generate data for choppingInstrumentation: Agreement on basic description of measurements and knobs required to commission and operate H- injectionHigh intensity, instabilities Studies by Giovanni et al. Impedance of new injection hardware and impact (instabilities) .. urgent Losses, activation On hold due to lack of resources

Clarification on expectations and available resources Expectations MUST fit to available resources!!Status of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011


Status of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011Introduction CERN PS Booster (PSB) OverviewRecapitulation of the present Booster with Linac2:Constructed in the 70ies to improve the performance (intensity) of the PSFour superimposed rings (chosen instead of fast cycling synchrotron or 200 MeV Linac)16 cells with triplet focusing, phase advance >90o per periodLarge acceptances: 180 m and 120 m in the horizontal and vertical plane, respectivelyMultiturn injection with betatron stacking and septum at 50 MeVAcceleration to 1400 MeV (was 800 MeV at the beginning)Sketch of the PS Booster with: - Distribution of Linac beam into 4 rings - Recombination prior to transfer


Status of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011Introduction CERN PS Booster (PSB) OverviewMandatory for high Intensity and BrightnessDouble harmonic RF (now: h=1 and h=2), Resonance compensation, Dynamic working point Maximum direct space charge tune shifts larger than 0.5 in the vertical planeIntensities above 1013 protons per ring (more then four times design) obtainedVersatile machine - many different beams with different characteristics

PSB to PS Transfer Energy:Increased (in two steps) from 800 MeV to 1400 MeV to reduce direct space charge effects in the PSDirect Space Charge at PSB Injection a BottleneckDynamic Working point with lower vertical tune no syst. 3rd order resonance (taken from a presentation by M. Chanel)


Status and Plans and open QuestionsPlans and open questions (optimistic wish-list? resources needed)Feasibility to study long term (~100 ms) blow-up? Comparison (Benchmark) with expsMachine imperfections (misalignements, multipoles from mech. tolerances ..)Working point optimizations ??Injection of 50 MeV protons from Linac2 and Linac4 with conventional multiturn injectionRecent question: to which extent Linac4 can replace Linac2 in case of failure (>2014 when Linac4 may be ready but not yet connected to the Booster)Injection with a magnetic septum (PTC ORBIT suitable?) to be studiedImperfections of the chicane magnets (multipoles . with time dependence)Completion studies on injection with constraints on (slower) energy modulationCompletion of injection painting optimizationsRCS studies (proposed/studied as new PS injector):Asses performance to be expected (Laslett tune shifts > 0.5 to be expected)Space charge limit with low periodicity and, possibly, large variations of beta functionsLosses and activationAny Emittance and lattice function estimates only based on analysis of macro-particle distributions



Issues Related to the new H- Injection Conceptual Aspects for the InjectionDynamic working point and vertical tune at injection:At present: highest intensity with vertical tune at injection above half-integer resonance (but below for LHC type beams)With Linac4 always below half-integer resonance:Double brightness for LHC, Less than twice the intensity for high intensity (less losses, limited by machines further downstream as well)Dispersion matching:Large energy modulation for longitudinal painting and dispersion of PSBIf beam arrives from Linac4 with D=0 m (dispersion mismatch): correlation of transverse emittance and position in longitudinal phase spaceShould beam arrive with D matched to the PSB lattice (see chapter on simulations)Acceptance with Linac4At present: Large beams close to limit of ejection channel acceptance (losses)With Linac4: aim for same maximum emittances at ejectionSmaller physical emittances at injection -> reduce machine acceptanceFor acceptances: AH=130 mm, AV=60 mm and ADp/p=2.2 10-3:Half-apertures in long straights: horizontal 33 mm and vertical 18 mmCan one take profit of a localized aperture to concentrate losses?Status of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011


Issues Related to the new H- Injection Conceptual Aspects for the InjectionLayout of the injection region (mainly by TE/BT team working on hardware)Superposition of two closed orbit bumps:Chicane to merge incoming H- beam with circulating protonsInjection painting bump: horizontal painting and removes beam from foilOptions (with different conceptual approaches)Small injection bump and high chicane (baseline):No problems with aperturesChicane is used as well to remove circulating beam from foilLarge injection bump and small height of chicane (ABP proposal):Foil outside of machine acceptance after fall of chicaneAperture in bends within injection painting bump criticalPlans for the near future:Evaluation of the two options by simulation with failures (e.g. painting triggered bump too early and injection with large betatron amplitudes)Fix layout (and specify maximum chicane fall time)Status of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011


Issues Related to the new H- Injection Scenario proposed (foil outside acceptance without painting bump)Limiting aperture (BeamScope like) in straight ~2p from foil (5L1)Aperture displaced towards inside of PSB -23 mm to +43 mm to save aperture for the injected beam -> average momentum (for large beams) Dpav/p = -10mm/DHeight of painting bump:Distance chicane to inner edge of foil: 33 mm + 7mm = 40 mmDistance edge of foil to incoming beam (Dpav/p) : 3*(2b erms)1/2 + D Dppaint/p ~ 15 mmHeight of the injection painting bump: 55 mm (instead of 35 mm proposed by BT)Problem: tight for aperture (taken by painting bump)

Make use of acceptance reduction to implement rough collimationThin aperture defining acceptance (at beginning) in straight sectionThicker absorber behind in same straight section and/or in straights further downstreamFirst step: look into feasibilityStatus of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011


Simulations of Beam Dynamics with strong direct Space Charge ForcesAssessment of slow blow-up by Simulations feasible?:Benchmark comparing measurements (by M. Chanel) with simulations (by M. Aiba and M. Martini) and between ACCSIM and ORBIT (M. Aiba and M. Martini)For the moment, no satisfactory agreement (yet?)Example:2nd harmonic RF to shorten bunches to increase tune shiftACCSIM largely overestimates blow-upORBIT overestimates blow-up as wellDependance on number of macro-partsNumerical artefacts?

Successful benchmark is prerequisite to apply simulations for blow-up and loss estimatesStatus of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011nH=19.2mnV=7.1mnH=20.4mnV=7.3mACCSIMORBIT


Simulations of Beam Dynamics with strong direct Space Charge ForcesImpact of injection matching and painting on PSB performanceFilamentation of structure due to injection dispersion mismatch?Investigated by M. Aiba with input from B. Goddard longitudinal and transverse painting of an LHC type injected over 20 turns

No significant difference with and without dispersion mismatchTo do: Impact (minor ?) of transverse matching and painting on performance?Simulation on impact on and of the injection foil by BT team and M. AibaStatus of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011Evolution of rms emittances after injection - blow-up (extrapolation ?) may be too large (slightly larger relative blow-up for 99% and 100% emittances)


Simulations of Beam Dynamics with strong direct Space Charge ForcesEvolution of beam parameters during fall of chicane with lattice perturbations and passive compensations and different fall times (simulations by M. Aiba)Passive compensation (see above) by pole-face rotation of chicane magnets yieldSimulations with ORBIT and additional routine for time dependant focusing of chicane, but no synchrotron oscillationsSimulations indicate significant reduction of blow-up

Chicane: LBS = 0.25 m and sector bends Chicane: LBS = 0.25 m and tapered bends

Lattice perturbation and active (by trims on QDEs) compensationInjection painting, static chicane and compensation and with synchrotron motionSimilar results (compensation reduces blow-up)Status of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011


Status of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011Activities and Plans for the working Group on PSB Beam Dynamics with Linac4Proposed longitudinal painting proposed and follow-up (implementation, see talk on RF aspects and possible limitations)Beam dynamics with strong direct space charge forcesBenchmark to which extent simulation are appropriate toolsApply simulations: filamentation of structure from injection, losses slow blow-up ?)Beam dynamics issues related to the new H- charge exchange injection (with BT team)Integration into the CERN Complex (see talk on Operational Scenarios ):Generation of all (also low intensity without painting) beams to be provided, minimization of losses and activationTransfer of LHC type beams between PSB and PSCheck limitations of present Booster hardware:Instabilities (existing damper with higher intensities) (Beam loading problems of h=2 cavities for h=1 beams limitations ISOLDE beams ?)Beam Losses, Activation (normal losses, failure scenarios ):Losses at Injection (Line and Ring) in collaboration with or by injection hardware team ?Feasibility of rough Collimation System (if feasible, resources need to be requested and allocated)


Summary, Status and QuestionsLinac4 aims at increasing the brilliance out of the PSB by about a factor 2.Beam dynamics studies aim:Thorough understanding of implications of Linac4 on PSBAnticipating problems to implement measures ensuring fast and efficient (i) restart of the PSB with Linac4 (ii) performance increase to expected levelsStatusProgress on many topics during the last year, but work on some topics still in starting phasePresent priority on studies with possible impact on hardware (e.g. finalization of injection)Questions to the review board:Validation of proposed longitudinal paintingAdvice on finalization of injection layoutValidation of proposed reduction of the machine acceptanceAdvice on simulations with strong direct space charge effectsStatus of the LINAC4 to PSB beam dynamics studies and open questions25th May 2011


**

linac4 to psb transfer line (slides by m. garcia-tudela & a. lombardi slightly adapted)

Documents