RF design studies of 1300 MHz CW buncher for European X-FEL. Shankar Lal PITZ DESY-Zeuthen

Transcription

1 RF design studies of 1300 MHz CW buncher for European X-FEL Shankar Lal PITZ DESY-Zeuthen

2 Outline Introduction Buncher design: Literature survey RF design of two-cell buncher: First design Two- cell buncher: alternate design options Three-cell buncher design Summery Outlook

3 Introduction XFEL needs CW injector for future upgrade Three operation modes envisioned for FLASH and XFEL Possible injector systems 1. CW SC DESY gun: under experimental demonstrations 2. CW NC gun based injector system: Under design study Up to 27 μa Up to 10 μa 187/217 MHz 1300 MHz 1300 MHz Up to 20 μa Borrowed from Dr. Houjun Qian, PITZ DESY Page 3

4 1300 MHz buncher : Literature survey Cornell/Jlab: ERL KEK: ERL LBNL:APEX Ref: (1). V. Veshcherevich and S. Belomestnykh, Buncher cavity for ERL, PAC 2003; (2) T. Takahashi et al., Development of a 1.3 GHz buncher cavity for the compact ERL, IPAC 2014; (3). H. Qian et al., Design of a 1.3 GHz two-cell buncher for APEX, IPAC 2014 Page 4

5 1300 MHz buncher designs: comparison Parameter Cornell/Jlab KEK LBNL DESY (proposed) No. of cells or 3 Frequency (MHz) R sh = V2 P c Nominal Acc. Voltage (kv) * Power dissipation (kw) (study underway ) Proposed PITZ/DESY design: KEK design (highest shunt impedance/cell ) with multiple cells Page 5

6 Two - cell 1300 MHz pre-buncher: First design Parametric view of cavity RF parameters mode Parameter value f (MHz) 1300 Q R (M ) P (for 400 kv) kw f -f 0 (MHz) 0.2 Electric field array plot for mode 5.1 mm Practical issues Small mode separation: Mode mixing /2=105.1 mm R 1 = 50mm Minimum wall thickness between to cell is only ~5 mm Page 6

7 Two-cell 1300 MHz pre-buncher: First design improvements Minimum wall thickness increased ~10 mm Diameter of inter-cell coupling iris increased to 49 mm RF parameters of mode Parameter Value f (MHz) 1300 Q R ( M ) 9.35 f - f 0 ( MHz) 0.96 P (for 400 kv) kw E peak /E z Electric field array plot for mode LBNL PITZ R sh is ~ 20 % higher compare to LBNL design (7.8 M ) Mode separation ~ 0.96 MHz RF power required to generate nominal cavity voltage (total) of 400 kv is ~ 17 kw Pc with accelerating voltage Page 7

8 Alternate design considerations Mode separation 3 MHz Higher shunt impedance 1. Modify geometry 2. Increase No. of cells Page 8

9 Alternate design: Option 1 Central geometry similar to TESLA (SCRF) cavities with re-entrant at end Elliptical inter-cell coupling iris: Mode separation increased to ~3 MHz Diameter of beam pipes smaller (36 mm) compare to inter-cell coupling iris (44 mm) Shunt impedance (~9.2 MHz) Peak magnetic field shifted near end pipes: simplify the cooling design E peak /E z ~ 3, higher but not problematic at operating gradient (~2 MV/m) Electric field array plot for mode Field profile is non- sinusoidal but qualified by beam dynamics simulations Need investigation of multipacting issues Surface current distribution Electric energy density distribution On- axis Electric field profile Page 9

10 Alternate design: Option 2 Geometry similar to TESLA (SCRF) cavities Cell length and Inter-cell coupling iris modified Inter-cell coupling iris reduced to 44 mm (TESLA cavity 70 mm) Mode separation ~ 3MHz Diameter of beam pipes smaller (36 mm) compare to inter-cell coupling iris (44 mm) Shunt impedance ~ 7.7 M E peak /E z ~ 1 Electric field array plot for mode MP remedies are available Magnetic field distribution Electric energy density distribution On- axis Electric field profile Page 10

11 Comparison of RF properties of different designs Geometry Parameter Re-entrant Center-elliptical with end nose cones Elliptical Q R ( =1) R ( =0.91) W =750 kev Δf (MHz) E peak /E z Peak surface power density (W/cm 2 ) P c (kw) for 400 kv Best option : but MP need to be investigate Page 11

12 Design of three-cell buncher Page 12

13 Option1:elliptical inter-cell coupling iris and nose cone at end Central geometry similar to TESLA (SCRF) cavities with re-entrant at end Elliptical inter-cell coupling: Higher mode separation Diameter of iris inter-cell coupling iris optimized for f - /2 ~ 3MHz Diameter of beam pipes smaller (36 mm) compare to inter-cell coupling iris (52 mm) Shunt impedance ~12 M E peak /E z ~ 3.5, higher but not problematic at low gradients RF power required ~13 kw for total gap voltage of 400 kv Need investigation of multipacting Electric field array plot for mode Magnetic field distribution On- axis Electric field profile Page 13

14 Geometry similar to SCRF cavities Option2: Elliptical cavity Cell length and inter-cell coupling iris modified Dimeter of inter-cell coupling iris is 52 mm: Mode separation ~ 3MHz Diameter of beam pipes smaller (36 mm) compare to inter-cell coupling iris (52 mm) Shunt impedance ~10.7 M E peak /E z ~1 RF power required ~15 kw Electric field array plot for mode MP remedies are available Magnetic field distribution Electric field distribution On- axis Electric field profile Page 14

15 Comparison of 3- cell pre-buncher structures Geometry Center-elliptical with end nose cone Elliptical Parameter Q R ( =1) R ( kev Δf (MHz) E peak /E z Peak surface power desity (W/cm 2 ) P c (kw) for 400 kv Page 15

16 Summary First RF design of re-entrant type two-cell 1300 MHz buncher is carried out. Proposed design has ~ 20 % higher shunt impedance compared to LBNL design but mode separation ~1 MHz. Alternate designs are also investigated with higher mode separation and higher shunt impedance. Two-cell design with elliptical inter-cell coupling iris and re-entrant at ends is most efficient, but MP issues need to be investigated. Three cell design also studied. Elliptical shape cavity require ~20 % higher RF power compare to re-entrant geometry with elliptical inter-cell coupling iris but thoroughly tested and MP remedies available in literature. Page 16

17 Out look Study of variation in RF parameters with geometrical dimensions Design of RF power coupling and pick ups Study of multipacting and thermal issues Design of RF tuners Page 17

18 Acknowledgments I would like to thanks Dr. Houjun Qian, Dr. Guan Shu, Dr. Hamed Shaker, Dr. Frank Stephan and all PITZ team members for useful discussions and critical feedback. I would also like to thanks Dr. Valentin Paramonov from Institute for Nuclear Research of Russian Academy of Sciences, Moscow, Russia for his feedback on design optimization. Page 18

19 Thank you for your attention If you are still alive Give your feedback and comments Page 19