1. Overview of NLC X-Band Technology and Applications Including a Compact XFEL Chris Adolphsen SLAC

2. Chose X-Band technology as an evolutionary next step for the Next Linear Collider (NLC) from the SLAC Linac S-Band (2.86 GHz) technology. In general want higher rf frequency because: -Less rf energy per pulse is required, so fewer/smaller rf components. -Higher gradients achievable, so shorter linacs (with reasonable efficiencies) Offsetting these advantages are the requirements of: -High power (100’s MW) HV pulses with fast rise times (100’s ns). -High surface gradients in the klystrons, waveguide transport system and accelerator structures. -Tight alignment tolerances due to stronger wakefields (much less an issue for light sources where the bunch charge is low, and bunch length short). X-Band (11.4 GHz) RF Technology

3. SLAC Linac RF Unit (One of 240, 50 GeV Beam)

4. NLC/GLC Linac RF Unit (One of ~ 2000 at 500 GeV cms, One of ~ 4000 at 1 TeV cm)

5. Next Linear Collider Test Accelerator (NLCTA) In 1993, construction began using first generation X-Band components. In 1997, demonstrated 17% beam loading compensation in four, 1.8 m structures at ~ 40 MV/m. In 1998-99, added second klystron to each linac rf station. In 2000-10, used for high gradient studies and other programs.

6. Klystrons Dual-Moded SLED II Pulse Compression Load ‘Tree’ Eight-Pack Project: Second Generation RF Unit Test Phase I: Generate RF Power and Transport to Loads Eight-Pack Modulator

7. RF Component Performance

8. Solid State Induction Modulator

9. Waveforms When Driving Four 50 MW Klystrons at 400 kV, 300 A Each Eight-Pack Modulator 76 Cores Three-Turn Secondary > 1500 Hours of Operation

10. Next Generation Induction Modulator: The ‘Two-Pack’ Features - 6.5 kV IGBTs with in-line multi-turn 1:10 transformer. - Industrialized cast casings. - Improved oil cooling. - Improved HV feed through. Bechtel-LLNL-SLAC 20 kV Test Stack A Hybrid 2-Pack Modulator (15 core stack driving a conventional 1:10 transformer) was built and is still being used – also built a version to drive a single ‘5045’ S-Band Tube 2-Pack Layout (never built)

11. RF Component Performance

12. X-Band Klystrons ‘XL4’ – Have built at least 15, now producing 12 GHz ‘XL5’ versions

13. PPM Klystron Performance (75 MW, 1.6  s, 120/150 Hz, 55% Efficiency Required) SLAC Two tubes tested at 75 MW with 1.6  s pulses at 120 Hz. Efficiency = 53-54%. KEK/Toshiba Four tubes tested at 75 MW with 1.6  s pulses at 50 Hz (modulator limited). Efficiency = 53-56%.

14. Character of events suggest they originate in output cavity – visual inspection inconclusive so far. At 75 MW, iris surface field ~ 70 MV/m, lower than in 3% v g structures, but higher than sustainable (~ 50 MV/m) in waveguide with comparable v g (~ 20%) as the klystron TW output structures. May require multi-beam klystron approach for stable > 50 MW, 1.6  s operation. KEK PPM2 Output Structure Time (100 ns / div) Reflected Transmitted Power – Normal Pulse Power – Tear Event Klystron Tear Events 14 mm

15. For RF Unit Test, Four 50 MW Solenoid-Focused Klystrons Installed in the Eight-Pack Modulator (In Place of Two 75 MW PPM Klystrons)

16. RF Component Performance

17. First Generation RF Pulse Compression (SLED II) at NLCTA

18. TE 01 TE 02 TE 01 For NLC/GLC, Use Dual Moded Delay Line to Reduce Delay Line Length in Half

19. Also Use Over-Height Planar Waveguide to Lower Surface Fields Design for < 50 MV/m for 400 ns pulses Example: Power Splitter

20. Dual-Moded SLED-II Performance (475 MW, 400 ns Pulses Required) Combined Klystron Power Output Power (Gain = 3.1, Goal = 3.25)

21. RF Component Performance

22. NLC Accelerator Structure Requirements Convert rf energy to beam energy efficiently. Short-range transverse wakefields small to limit linac emittance growth: iris radius limited to 17% of rf wavelength (i.e. a/ = 17%). Long-range wakefields suppressed so bunch train effectively acts as a single bunch. Dipole mode power coupled out for use as guide for centering the beam in the structure. Operate reliably at the design gradient and pulse length. NLC/GLC Structures were developed by a FNAL/KEK/SLAC collaboration – CERN/KEK/SLAC now developing 11%-13% a/ structures that run at ~ 100 MV/m

23. High Gradient Structure Development 50 cm ‘T53’ Structure Since 1999: -Tested about 40 structures with over 30,000 hours of high power operation at NLCTA. -Improved structure preparation procedures - includes various heat treatments and avoidance of high rf surface currents. -Found lower input power structures to be more robust against rf breakdown induced damage. -Developed NLC-Ready ‘H60’ design with required wakefield suppression features.

24. H60 Structure Cells and Coupler Assembly Cells with Slots for Dipole Mode Damping Port for Extracting Dipole Mode Power High Power Output Coupler

25. Wakefield Damping and Detuning Time of Next Bunch Time After Bunch (ns) Wakefield Amplitude (V/pC/m/mm) Measurements Ohmic Loss Only Damping and Detuning Detuning Only Dipole Mode Density Frequency (GHz)

26. Hours of Operation Breakdown Rate History (Goal < 0.1/hr) Four H60 Structures at 65 MV/m  = 400 hr  = 600 hr  = 400 hr Breakdown Rate at 60 Hz (#/hr) with 400 ns Pulses Hours of Operation

27. RF Unit Test in 2003-2004 From Eight-Pack 3 dB Beam From Station 2 From Station 1 Powered Eight H60 accelerator structures in NLCTA for 1500 hours at 65 MV/m with 400 ns long pulses at 60 Hz and accelerated beam

28. Efficiency (%)DesignAchievedComment Two-Pack Modulator 7060 Use Integrated Transformer (> 70% Expected) PPM Klystron5553-56~ 60% Possible SLED II8178Improve Flanges/Fab/Assembly Wave Guide Transport 9277 Use Design Layout / Reduce High-Loss WG Structures3129Use Round Cell Shape RF Component Efficiencies

29. High Power (Multi-MW) X-Band Applications Energy Linearizer –Single Structure: in use at LCLS, planned for BNL, PSI, Fermi/Trieste and SPARX/Fascati Deflecting Cavity for Bunch Length Measurements CERN Linear Collider Structure Development 100’s of MeV to Many GeV Linacs –LLNL 250 MeV linac for gamma-ray production –SLAC 600 MeV energy ‘dither’ linacs for LCLS II –LANL 6-20 GeV linac for an XFEL source to probe proton-matter interactions –SPARX 1-2 GeV X-Band linac for their FEL –SLAC study of a 6 GeV Linac for a Compact XFEL (CXFEL) source See my Thursday talk in WG4 for more details

30. CERN/CLIC X-band Test-Stand (Under Construction) CERN - CEA – PSI – SLAC High voltage modulator Klystron XL5 Directional coupler Circular waveguide  =50 mm SLED Pulse compressor Mode convertors RF Valve Circular pumping port

31. CLIC WS 10/2009KM Schirm BE-RF 31

32. ParametersymbolLCLSCXFELunit Bunch Charge Q 250 pC Electron Energy E 146GeV Emittance  x,y 0.4-0.60.4-0.5 µmµm Peak Current I pk 3.0 kA Energy Spread  E /E 0.010.02% Undulator Period u 31.5cm Und. Parameter K 3.51.9 Mean Und. Beta ββ 308m Sat. Length L sat 6030m Sat. Power P sat 3010GW FWHM Pulse Length  80 fs Photons/Pulse NN 20.710 12 Compact X-Ray (1.5 Å) FEL

33. Linac-1 250 MeV Linac-2 2.5 GeV Linac-3 6 GeV BC1BC2 Undulator L = 40 m S rf gun X undulator LCLS-like injector L ~ 50 m 250 pC,  x,y  0.4  m X-band Linac Driven Compact X-ray FEL X X X-band main linac+BC2 G ~ 70 MV/m, L ~ 150 m Use LCLS injector beam distribution and H60 structure (a/ =0.18) after BC1 LiTrack simulates longitudinal dynamics with wake and obtains 3 kA “uniform” distribution Similar results for T53 structure (a/ =0.13) with 200 pC charge

34. UnitsCXFELNLC Final Beam EnergyGeV6250 Bunch ChargenC0.251.2 RF Pulse Width * ns150400 Linac Pulse RateHz120 Beam Bunch Lengthμm7110 Operation Parameters * Allows ~ 50-70 ns multibunch operation

35. Layout of Linac RF Unit 50 MW XL4 100 MW 1.5 us 400 kV 480 MW 150 ns 12 m Nine T53 Structures (a/ = 13%) or Six H60 Structures (a/ = 18%)

36. NLC RF Component Costs (2232 RF Units) Cost (k$) per Item LLRF26.1 Modulator83.7 Klystron56.6 TWT13.3 SLED-II242.3 Structures21.5 For the 6 GeV CXFEL, assume the cost per item will be 4 times higher. For 70 MV/m operation: UnitsT53H60 Total RF units1824 X-Band Linac Lengthm122108 Total Accelerator Lengthm192178 X-Band Linac CostM$5662

37. Assuming 1) Tunnel cost 25 k$/m, AC power + cooling power 2.5 $/Watt 2) Modulator efficiency 70%, Klystron efficiency 55%. Gradient Optimization

38. Structure Breakdown Rates with 150 ns Pulses 1)H60VG3R scaled at 0.2/hr for 65 MV/m,400 ns, 60Hz 2)T53VG3R scaled at 1/hr for 70 MV/m, 480 ns, 60 Hz 3)Assuming BDR ~ G 26, ~ PW 6 At 70 MV/m, Expect Rate Less Than 0.01/hr at 120 Hz

39. Single Bunch Wake Tolerances In both Linac-2 and Linac-3, short-range transverse wakefields in H60 are not a major issue in that: - An injection jitter equal to the beam size yields a 1% emittance growth in Linac-2 and.003% growth in Linac-3 - Random misalignments of 1 mm rms, assuming 50 structures in each linac, yields an emittance growth of 1% in Linac-2, 0.1% in Linac-3. With the T53 structure, the jitter and misalignment tolerances are about three times smaller for the same emittance growth. The wake effect is weak mainly because the bunches are very short.

40. X-Band Revival The 15 year, ~ 100 M$ development of X-band technology for a linear collider produced a suite of robust, high power components. With the low bunch charge being considered for future XFELs, X-band technology affords a low cost, compact means of generating multi-GeV, low emittance bunches. –Gradients of 70-100 MV/m possible vs ~ 25 MV/m at S-Band and ~ 35 MV/m at C-Band The number of XFELs is likely to continue to grow (e.g., normal conducting linacs being considered in Korea and China). To expand X-band use, need to have components industrialized and a small demonstration accelerator built, such as the 150 MeV C-band linac at Spring-8 in Japan where they have done light source studies.