SC Spoke Resonators for Project X

Development of SC Spoke Resonators at FNALG. Apollinari Fermilab4th SPL Collaboration Meeting with ESSLund, Sweden July 1st 2010In the last year Project-X has evolved within the financial constraints of DOE to better meet the physics mission of US HEP community along 3 major lines of research:Long baseline neutrino beam High intensity, low energy protons for kaon and muon based precision experimentsA path toward a future muon facility neutrino factory or muon colliderInitial Configuration 1 (IC-1)8 GeV Linac in MI, ILC paramtersInitial Configuration 2 v12 GeV CW Linac + 2-8 GeV RCSInitial Configuration 2 v2 (IC-2v2)3 MW @ 3 GeV CW LinacProject-X EvolutionCW nature requires SCRF acceleration from very low energies (2.5 MeV)The zero-current phase advances of transverse and longitudinal oscillations should be kept below 90 per focusing period to avoid instabilities at high current.The wavenumbers of transverse and longitudinal particle oscillations must change adiabatically along the linac. This feature minimizes the potential for mismatches and helps to assure a current-independent lattice.Minimize derivative of zero-current longitudinal phase advance along lattice, to reduce halo excitation.Avoid the n=1 parametric resonance (zero current) between the transverse and longitudinal motion.Avoid energy exchange between the transverse and longitudinal planes via space-charge resonances either by providing beam equi-partitioning or by avoiding instable areas in Hofmanns stability charts . Provide proper matching in the lattice transitions to avoid appreciable halo formation.The length of the focusing period must be short, especially in the front end.Beam matching between the cryostats: adjust parameters of outermost elements (solenoid fields, rf phase)

See: P. Ostroumov talk, Feb.2, 2010, FNALPrX Beam Physics Gospel

To be efficient at low-bLow cryogenic lossesHigh (R/Q)GHigh GradientLow Ep/Eacc and low Bp/EaccLarge Velocity acceptanceFew accelerating gapsFrequency ControlLow sensitivity to microphonics & low energy contentChoice of Cavities

http://www.lns.cornell.edu/public/SRF2005/talks/sunday/SuA04_talk_srf2005.pdf

Spoke Resonators

AdvantagesNo dipole SteeringHigh performanceLower Rsh than HWRsWide b rangePotential ProblemsNot easy accessDifficult to tuneLarger size the HWRsMore expensive than HWRsQuadrupole Steering325 < f < 805MHz, 0.15 < b < 0.6PrX - Initial Configuration 2

For historical reasons(2005 Proton Driver) SSR(1) was the first SC low-b cavity developed at FNAL within the context of a pulsed 8 GeV Linac with SC cavities from 10 MeV.CM segmentation, number of cavities/CM and the gap between CMs

88 SSRs (325 MHz)138 Ellipt. (650 MHz)64 Ellipt. (1.3 GHz)

(Initial) Performance GoalsFreq (MHz) Bpk(mT)G (MV/m)Q@T (K)325 60151.4E102 650 72161.7E1021300 72151.5E102

PrX Initial Beam Elements SpecsBeam dynamics design completed and optimized for regular lattice, with break points and cavity types determined

SSR0SSR0SSR1SSR2LB 650SSR1SSR0LB 650

Spoke Resonator ChallengesDesign Optimization (SSR1)Construction(Feasibility of) AssemblyEngineering Safety OperationsGradient and Q0 performanceTunabilityLorentz Force Detuning (Pulsed) and Microphonics (CW)High Power OperationsCryomodule Integration and Test Facility(Beamline) IntegrationFocusing ElementsInstrumentation Cryomodule Assembly & CommissioningMeson Detector Building Test Facility

RF Design optimizationMechanical analysis and optimization

L

SSR1 Design Optimization

Total Deformation at 2 atm.Reinforcement TypeTotal DeformationVon Mises[MPa]none329352.8flat283256.0tubular266254.4Flat + gussets13765.08Tubular + gussets10965.85VBuilt 2 prototypes + 2 additional from IUAC (India).

ShellDonut ribEnd wallCoupler portVacuum portBeam portBridge ribsspokeDaisy ribsAssembly Model

Spoke AssemblySpoke FormingCollar FormingSpoke Welding 11

Shell Assembly

End-walls Forming

Stiffening RibsCaveat: Ribs designed for Pulsed Operation Stiffening

ShellDonut ribEnd wallCoupler portVacuum portBeam portBridge ribsspokeDaisy ribsBrazed Ports for He VesselBraze spikeCool downRamp upsoakventTtBrazing Process Initially developed at CERN (1987) later modified at ANL (2003) Filler metal: CDA-101 high purity copper wire SST flanges pre-machined, stress relieved at 1100 C and finishedYield limit 6700 lbs Allows assembly of SS He-Vessel

Cavity Tuning

At a glanceWeight~ 40 kgLength342 mmHeight615 mmNb thickness2.8 - 3.2 mmRRR Nb~ 18 ft2RG Nb~ 5 ft2TransitionsCu BrazeMAWP34 psiSpring K~ 20 N/m

BCP at ANLHelium Vessel Assembly17

3-D ModelBellow and End-plate welding process SSR Tuner

Safety Rod attached cavity flange to slow tuner arm Stepper Motor & Harmonics Drive Slow Tuner Arm ASME Boiler and Pressure Vessel code (US) introduced in 1905 to address exploding boilers (before then, individual state regulations).Dp > 15 psi (~1 atm.) & Dimension > 6 in (15 cm)EU has Pressure Equipment Directive (PED) since ~2005, individual countries regulations before thenDp > 0.5 atm & Volume > 1 LSCRF assemblies cannot meet fully the requirements of the US-ASME code (ex: Nb is not a code-allowed material)(Possible) strategies: Directors magic wand for exceptional vessel approvalSNS/JLAB approach: coded cryomodule vacuum vessels as pressure containmentDevelop standards such that necessary deviation from code (Nb) are handled by special procedures (measuring the mechanical properties of samples of the niobium from the lot of material from which the cavities are made)Engineering Note & SafetyEngineering AnalysisPresent Strategy: Minimize number of exceptions to codeUse of Nb (not explicitely allowed by the code)No ultrasonic examination of EB welding along entire lengthLack of WPS (Weld Procedure Specs) PQR (Procedure Qualification Records) and WPQ (Welder Performance Qualification) for Nb and SS assembliesNb-SS brazing did not have a Brazing Procedure SpecificationDemonstrate safety by engineering analysis and pressure testing

Spokes Cavity Vertical Test21SSR1-1Four VTS tests between March 2008 and March 2009Vacuum problems in first two testsActive pumping added to VTS before 4th test4th test included cool-down dwell at 100 K in attempt to induce Q-diseaseWill next be tested in new test cryostat in coming monthsSSR1-2One VTS test in 2009Reached gradient 33MV/mEacc=Acc. Voltage/Liris =Acc. Voltage/2/3 bl

21SSR1-2 First VTS Test22

60 mTPC High Power Test Fixture23

Three Fermilab-designed couplers produced and in houseAverage power of 4.2 kW (2 Hz, 3 ms, 700 kW) was sustained for 2 hours, and an additional 3 hour test at 3.3 kW (2 Hz, 3 ms, 550 kW) was performed.

23Horizontal Test Cryostat for High Power (~250 kW) SSR testingHigh Power Test

Design Optimization (SSR1)Construction(Feasibility of) AssemblyEngineering Safety OperationsGradient and Q0 performanceTunabilityLorentz Force Detuning (Pulsed) and Microphonics (CW)~ High Power OperationsCryomodule Integration and Test Facility(Beamline) IntegrationFocusing ElementsInstrumentation Cryomodule Assembly & CommissioningMeson Detector Building Test Facility

Spoke Resonator Challenges

SSR1 Cryomodule ModelPresent concept of SSR1 CryomodulesContain 9 SSR1 cavities and 9 solenoidsProject X expects that these designs could be extended to SSR0 and SSR2 requirements

26Cryomodule Assembly27

Assembly performed using the same or similar tooling to that used for 1.3 GHz (XFEL/ILC) final assembly Also studying the advantages/disadvantages of bath-tub assembly273-Cavity Cryomodule Concept28

28

E FieldB FieldF(MHz)325325325325optimal0.1350.1170.110.22Rcavity,mm210191.5180245.6R/Q, 150130120240TTF, Average0.8910.9440.9530.952Emax/Eacc/Emax/Eacc*5.4/6.16.8/7.17.0/7.33.9/4.1Hmax/Eacc/Hmax/Eacc*(mT/MV/m)10.9/12.310.3/1110.8/11.35.8/6.1Deff =(2*/2),mm124.6108101.5203

greenredblueEacc*=Eacc(optimal)* TTF, AverageSSR0 RF Design OptimizationSSR0 Mechanical Optimization

Displacement 0.1 mm

Force reaction 315 x 4

K = 315*4/0.1 = 12600 N/mm ~ 13 N/mMeson Detector Building SetupMDB/HINS initially (2005) conceived as development ground for 325 MHz Front End for the initial Project X configuration (8 GeV Pulsed Linac, SC from 10 MeV)Initial Goals:Testing ground/conditioning for all 325 MHz equipmentAccelerate beam from 0 to ~60 MeV through RFQ, MEBT, RT section (2.5-10 MeV) and SC SSR1 section (10-60 MeV)Control RF power to cavitiesBeam Chopper DevelopmentBeam Instrumentation DevelopmentRevised Goals:In CW design, RT section eliminated. Probably replaced by short SSR0 cryomodule (~3 cavities) for proof of principle acceleration with SSRsTesting ground for MEBT section, chopper and beam diagnostic

13.4 m16.9 m10.5 foot ceilingLayout for 3-Cavity SSR0 Cryostat2.4 m cryostat0.5 m end0.5 m end0 m10.5 m14.2 m18 spectrometer ~2.7 m lengthExisting ion source and RFQ10 m MEBT/CHOPPERactual HINS absorber/shieldingRFQ and 2.5 MeV Beamline

First 2.5 MeV Beam through RFQPage 34

Signals from toroid and two BPM buttons, all downstream of the RFQ

Upper display: 2 sec/div Lower display: 20 nsec/div

Lower display shows the 44nsec delay expected for transit of 2.5 MeV beam between the BPM two buttons separated by 0.96 meters

Beam current is about 3 mA

ConclusionsProject X has adopted Superconducting Spoke Resonators for the Front EndDesign and development of bare cavities is by now a routine operationValuable experience is being gained in the assembly and operation of dressed cavities.Design of the SSR cryomodule is still in its early stages.Supporting SlidesAdvantagesCompactModularHigh performanceLow CostLow BetaPotential ProblemsField asymmetry (to be compensated by dipole steering or gap shaping)Mechanical Stability Quarter Wave Resonator

40 < f < 160 MHz, 0.001 < b < 0.2OPERATINGHalf-Wave ResonatorAdvantagesNo dipole SteeringHigh performanceLower Ep than QWRWide b rangeVery compactPotential ProblemsNot easy accessDifficult to tuneLess efficient than QWRs

160 < f < 350MHz, 0.09 < b < 0.3Provide proper matching in the lattice transitions to avoid appreciable halo formation. In the perfect current-independent design, matching in the transitions is provided automatically if the beam emittance does not grow for higher currents.

Stability for zero current beam, defocusing factor should be < 0.7. (Defocusing factor is less for lower frequencies for given Em)

The length of the focusing period must be short, especially in the front end.

Beam matching between the cryostats: adjust parameters of outermost elements (solenoid fields, rf phase)

In the frequency transition, the longitudinal matching is provided by 90 bunch rotation, or bunch compression

PrX Beam Physics Gospel (cont.)

SSR1-1 Early VTS Results

O

SSR1-1 Q vs. EAccelerating Gradient MV/mDressed Cavity Operating Goal @ 4 K (Pulsed)This test ended when multipacting due to poor cavity vacuum became unacceptable.

SSR1-1 Final Test ResultsSpoke Cavity Horizontal Test Cryostat in MDB

42

Complete LayoutCaveats: Physical Sizes of RF4-1300P1000, RF7-650P100 and RF10-325CW30are not known

Layout does not show RF5-1300P100. It is assumed that RF2-1300P250 or RF4-1300P1000 serves that purpose.HTS-2325CAGEExisting RF1-3900P80 and RF2-1300P300NEW RF3-1300CW30 and RF6-650CW30Existing RF8-325P2500Existing RF9-325CW0.4NEW RF4-1300P1000and RF7-650P100NEW RF10-325CW30Relocated 325 MHz RF Component Test FacilityNEW RF11-325CW1.543GoalsComplete Six-Cavity Test June 2011Demonstrate individual Phase/Amplitude control with DQM shifters.Demonstrate that solenoid beam axis can be aligned to 0.5 mm rms by Oct 2011Select bunch frequency (162.5 or 325) by demonstrating a broad-band chopper by July 2013 (CD2)Complete test of SSR0 short cryomodule (3/4 cavities, 4/5 solenoids + correctors, BPMs) (prototype for a long cryomodule) with beam and broad-band chopper by Sept 2014Ongoing development of instrumentation, optics, couplers, LLRFIon Source Emittance Scan Data

Ib = 4 mA

HorizontalVerticalIb = 12 mA50 keV beam from HINS proton ion sourceHINS 2.5 MeV Beam Profiles

Linac Enclosure (Under Construction)

RFQ Problem & its solution

Note RF joint seal

RF joint seal buckledChart10.45670134620.6587030660.74546031620.65483993150.81651514990.87781361010.7999692610.91581396140.95378214730.89700277360.97117513290.99008168560.95631611470.99587275550.99999995050.98795075640.99999944210.99297559850.99999894320.99070950310.97540815290.99851677680.97291643930.95154445410.98788085260.94994301750.92417139740.9711987190.92401697840.89510340240.95065614570.89662206180.8655113687

Normalized Transit Time Factor vs beta

ttf27.00E-020.36082603260.45670134620.51735197510.6587030660.5830871320.74546031628.00E-020.51736938470.65483993150.64129916390.81651514990.68661176080.87781361019.00E-020.6320317140.7999692610.71928944340.91581396140.74603313510.95378214731.00E-010.70869498130.89700277360.76277066110.97117513290.77442605320.99008168560.10999999940.75555667280.95631611470.78216842090.99587275550.78218396130.99999995050.11999999920.78055025410.98795075640.78540956180.99999944210.77668962550.99297559850.1299999990.7900691650.99999894320.77811315090.99070950310.76294865070.97540815290.13999999870.78889814990.99851677680.76413830060.97291643930.74428284730.95154445410.14999999850.78049502520.98788085260.74609474540.94994301750.72287208030.92417139740.15999999830.76731497190.9711987190.7257321750.92401697840.70013555970.89510340240.16999999810.7510849010.95065614570.70421593350.89662206180.67698914440.8655113687

ttf2

Normalized Transit Time Factor vs beta