Do we need

Wolfgang Weingarten CERN2nd revised version including additional information as collected during the workshop15 -17 December 2010Rev. 7 Feb. 2011LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. Weingarten1 Coupler (FPC & HOM) concepts for compact crab cavities OutlineSpectrum of LHC beamSome basic relations 1)Compact crab cavity characteristicsCoupler/damping, cryogenic and clean work requirementsFundamental Power Couplers for Crab Cavities, possible designs 2)Conclusion

15 -17 December 20102LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. WeingartenI owe much of clarification to Joachim Tckmantel: Crab cavities: speed of voltage change, CCinS WG, 27 Nov.2009 (cf. also J. Tckmantel, BE-RF, Cavity-Beam-Transmitter Interaction Formula Collection with Derivation, CERN-ATS-Note-2011-002 TECH, January 24, 2011).The respective slides were provided by Eric Montesinos, whom I gratefully thank.

Spectrum of LHC beam 1/215 -17 December 2010LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. Weingarten3

from LHC design report CERN-2004-003 15 December 20041.8 ms0.3 ms

Spectrum of LHC beam 2/215 -17 December 2010LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. Weingarten4476 kHz 84 bunches40 MHz 1 bunchLHC beam approximated by pulsed beam without gaps

replaced by bunch string without gaps (hence preliminary)

Some basic relations15 -17 December 2010LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. Weingarten5

Compact crab cavity characteristics 1/215 -17 December 2010LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. Weingarten6

Compact crab cavity characteristics 2/215 -17 December 2010LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. Weingarten7

Coupler/damping, cryogenic and clean work requirements 1/215 -17 December 2010LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. Weingarten8The competition among lower-order modes, same-order modes, and higher-order modes is to a large measure the main technical and engineering issue encountered in crab cavities.Fundamental power coupler (FPC)must provide about 15 kW RF power with Qext = (2 - 6)106 (depending on beam offset)HOM/LOM/SOM couplerMust provide Qext = 100 - 300 for specific modes without affecting the crab mode.Consequence: cavity with large beam tube to guarantee leaking out of all modes seems to be most adequate but at the cost of an effective filter to block the crab mode (an exception is the JLAB cavity)Cryogenic needsThe dynamic heat load at 2/4.5 K is in the range of 10/100 and even more Watts.Clean workThe required electromagnetic fields correspond to an equivalent of 25 - 31 MV/m in conventional acc. cavities; hence a crab cavity study can largely profit from the planned upgrade in CERNs SM18 treatment and assembling facilities 15 -17 December 2010LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. Weingarten9

15 -17 December 2010LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. Weingarten10

15 -17 December 2010LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. Weingarten11

15 -17 December 2010LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. Weingarten12

Conclusion15 -17 December 2010LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. Weingarten13The compact crab cavity for the LHC is challenging in obtaining the large electromagnetic fields (31 MV/m equivalent to acc. cavity) - an additional argument in favor of the planned refurbishment of the SRF infrastructure in SM18.The competition among lower-order modes, same-order modes, and higher-order modes is to a large measure the main technical and engineering issue encountered in crab cavities;The damping studies for LOM,SOM and HOM must go on. The fundamental power is low (tens of kW depending on tolerated beam offset).There exist several well advanced and robust designs.Fundamental Power Couplersfor Crab CavitiesDepending on the cavity design and available space around the cavity, several solutions can be considered

We recently launched at CERN several Fundamental Power Couplers developments which cover various window types:

SPL coaxial disk window coupler

SPL/ESRF/SOLEIL/APS coaxial cylindrical window coupler

Linac 4 waveguide disk window couplerCERN, BE-RF-SR, Eric MontesinosLHC-CC10, 4th LHC Crab Cavities Workshop, 15-17 December 2010, CERN14f0400 MHzAverage Power15 kWQext of input couplerFixedCoolantAir cooledEven if designing a new Fundamental Power Coupler will never be an easy job, we can try to design a new coupler to meet the proposed parameters and fit any cavity designThank you for your attention15 -17 December 2010LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. Weingarten15

Men do not stumble over mountains, but over mole hills - Confucius -Spare slides15 -17 December 201016LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. WeingartenCoaxial Disk windows

CERN, BE-RF-SR, Eric MontesinosLHC-CC10, 4th LHC Crab Cavities Workshop, 15-17 December 2010, CERN17CouplerFrequency[MHz]Average Power [kW]Peak power[kW]# in operation or constructedSPS20055080016KEKB50930014208CEA-HIPPI70412012002IHEP5001502702JPARK97230220023SNS80578200093SPL7041001000Prototyping ongoing

Waveguide windows

CERN, BE-RF-SR, Eric MontesinosLHC-CC10, 4th LHC Crab Cavities Workshop, 15-17 December 2010, CERN18CouplerFrequency[MHz]Average Power [kW]Peak power[kW]# in operation or constructedSPS801225225 (more ?)8Cornell5003503504FNAL / TTF II13004.5100032LBNL7008008004Linac 43521401400Prototyping ongoing

One cylindrical windowCERN, BE-RF-SR, Eric MontesinosLHC-CC10, 4th LHC Crab Cavities Workshop, 15-17 December 2010, CERN19CouplerFrequency[MHz]Average Power [kW]Peak power[kW]# in operation or constructedLHC400550 sw cw,(i.e 2200 tw cw)i.e. 2200 tw cw16LEP352550 tw cw565 tw cw252SPS (1976-2000)20037550016SPL7041001000Prototyping ongoingESRF / Soleil / APS3525001000Prototyping ongoing

Two cylindrical windowsCERN, BE-RF-SR, Eric MontesinosLHC-CC10, 4th LHC Crab Cavities Workshop, 15-17 December 2010, CERN20CouplerFrequency[MHz]Average Power [kW]Peak power[kW]# in operation or constructedTTF family - XFEL13004.5110016 (+ 1064)Cornell ERL130075752

Cold window (beam vacuum / coupler vacuum) is difficult to cool down enough for high average power couplerEngineering points forFundamental Power CouplersRF power capabilityMinimize heat loadGas coolant (air preferably) is preferred to liquid coolant to ease beam vacuum leak detection in case of failure

Ceramic design:The critical item of a power coupler is the main ceramic.The coupler robustness mainly depends on the ceramic designSingle ceramic room temperature window couplers have proven to be reliableBrazing and EB welding neededCoupler design:ModelingCeramic location in E-field minimumPrototypingMatching and VSWRBandwidthQext, tuningMultipacting simulationMonitoring (vacuum, e- probe, arc detector)

Coatings:Plating or/and sputteringMultipacting suppression (Ti, TiN)CERN, BE-RF-SR, Eric MontesinosLHC-CC10, 4th LHC Crab Cavities Workshop, 15-17 December 2010, CERN21Engineering points forFundamental Power CouplersMechanical:Clean room cleaning and assembly to guarantee high fieldHigh pressure rinsingContamination during beam vacuum part assemblyEasy installationIntegration with the cryomodule

Easy operationMaintenance (air blowers, water pumps)Failure consequencesRepairSparesRF ConditioningVacuum bakeoutHigh vacuum pumping for shorter processesShort pulses up to full power, with pulse length increased to cw with vacuum feedback conditioning methodDC biasingTemperature sensorsRoom temperature testingHigh power tests benches available from the beginning

CostCERN, BE-RF-SR, Eric MontesinosLHC-CC10, 4th LHC Crab Cavities Workshop, 15-17 December 2010, CERN22Long tradition at CERN15 -17 December 2010LHC-CC10, 4th LHC Crab Cavity Workshop CERN Geneva - W. Weingarten23

~3 mLongitudinalTransversal

Energy-momentum 4-vector

Panofsky-Wenzel theorem

Wave propagation

Legend to symbolsPr: reflected powerPg: generator powerfb: bunch form factor (close to 1)x : maximum beam offsetIb,DC: DC beam currentV: kick-voltageE: energy of beamp: momentum of beam: Phase angle: RF frequency of crab mode(R/Q): long. or transversal R/Q valueQext, opt: ext. Q-value of crab mode that minimizes the reflected power (this is not necessarily the most favourable option!)

Compact crab cavity candidates under study

Cavity characteristics

Deflecting mode at 400 MHz

R/Q = V2/(2U)Parallel bar elliptical TEM cavity (JLAB) Half wave spoke resonator (SLAC)

Four Rod Compact CrabCavity (Cockcroft) Compact crab cavity (KEK)

Total kick voltage [MV] per beam10

Kick voltage [MV] per cavity5

Total number of cavities16 (4 cavities per beam and interaction point)

Frequency [MHz]400

Cavity width/diameter [mm]295290286150

Cavity height [mm]406391.5236668

Cavity length [mm]445580408 > 668

Beam pipe [mm]84848475

Peak el. field [MV/m]365262145 (n/a)

Peak mag. field [mT]8097.599250 (n/a, > Bc of Nb)

R/QT [] per cavity131.3107.54758

Surface resistance @2 / 4.5 K [n]1)6/1808/2608/260n/a

Geometry factor []108.960 (best guess)7078

Respective Q-values [109]12/0.47/0.28/0.3n/a

Power dissipation [W]5/16017/5003.4/98n/a

1) Derived from a least square fit of a multitude of cavity data: W. Weingarten, On the Field Dependent Surface Resistance Observed in Superconducting Niobium Cavities, Proceedings of SRF2009, Berlin, Germany, TUPPO052.

Compact crab cavity candidates under study

= 90

|x|< 0.5 mm

Crab mode at 400 MHzR/Q = V2/(2U) (circuit definition)Parallel bar elliptical TEM cavity (JLAB) Half wave spoke resonator (SLAC)

Four Rod Compact CrabCavity (Cockcroft) Compact crab cavity (KEK)

Crab modeR/QT [] per cavity @ f [MHz]131.3 @ 400107.5 @ 400475 @ 400 MHz8 @ 400 MHz

Qex, opt6106810621061108

Generator power [kW]15151515

Longitudinal mode with largest R/QMaximum R/Q [] per cavity @ f [MHz]~75 @ 680~90 @ 33060 @ 375?

Max. required Qex for 10 kW deposited power200100200?

Max. required Qex for 1 MV max voltage build-up104104104?

Max. required worst case Qex for 40 k per cavity2)270220330?

Transverse non-crab mode with largest R/QTMaximum R/QT [] per cavity @ f [MHz]~45 @ 610~70 @ 880~40 @ 1050?

Max. required Qex for 0.4 M/m per cavity for 2 identical cavities 2)700310450?

1)This is not the most prominent but the mode closest to the crab mode, 2)This number is based on E. Shaposhnikovas talk at LHC-CC10, slide # 19