BROOKHAVEN SCIENCE ASSOCIATES SRF cavities for high current

BROOKHAVEN SCIENCE ASSOCIATES SRF cavities for high current ERLs Wencan Xu Brookhaven National Laboratory Wencan Xu ERL workshop June 7 -12 2015 1

Outline Ø Considerations of SRF cavities for high current ERLs Ø High current SRF cavities development at BNL Ø Summary Wencan Xu ERL workshop June 7 -12 2015 2

Cavity design consideration Ø Transversal Beam-Break-Up induced by one dipole mode (1 -Pass ERL)*: Where, Ø Longitudinal Beam-Break-Up induced by one monopole mode (1 -Pass ERL)*: *C. Song, G. Hoffstaetter , Beam breakup simulation for the Cornell X-Ray ERL, PAC 07. ØHOM power built up in a CW machine (one monopole mode): Resonant exciting Non-resonant exciting Wencan Xu ERL workshop June 7 -12 2015 3

704 MHz BNL 1 cavity for R&D ERL Ø Ø Ø Wencan Xu Enlarged beam tube: to attenuate the fundamental mode but propagate all the HOMs. The cutoff frequency is 790 MHz, the first dipole mode is 810 MHz; RT ferrite as HOM damper: HOM measurement at both room temperature and low temperature showed extremely good damping. 50 k. W CW fundamental power coupler: SNS type, water cooling on the RF window, no cooling for the inner conductor. Mechanical tuner with 100 k. Hz tuning range, piezo provides 9 k. Hz fast tuning First horizontal test in 2009 and several tests were carried out after that(Q 0 ~ 1010 at 18 MV/m). ERL workshop June 7 -12 2015 4

704 MHz BNL 1 cavity for R&D ERL Ø For ERL prototype, it is necessary to have 15 MV/m in the 5 -cell cavity , due to beam-dynamics constraints in the beam combiner. Ø However, 12 MV/m was observed to be threshold in CW mode, due to the Al. Mg 3 seal located between the Nb. Ti and stainless steel flanges on the beam pipe. Ø Quasi-CW operation mode: Test showed that the cavity can safely (thermal stability) run at 18 MV/m with a 6. 25% duty factor. Wencan Xu ERL workshop June 7 -12 2015 5

704 MHz BNL 3 cavity for conventional lattice ERL e. RHIC Ø High current ERL based (3. 5 n. C, 50 m. A 6 -pass ) => Heavy damped HOM Ø All in the RHIC existing tunnel => limit the linac length (required efficient HOM damper) Ø The luminosity can exceed 1034 cm-2 s-1 Wencan Xu Parameters BNL III Frequency [MHZ] 704 703. 79 Cells No. 5 5 Geometry Factor 225 283 (R/Q)/cell [Ω/cell] 80. 8 101. 26 Epeak/Eacc 1. 97 2. 46 Bpeak/Eacc [m. T/MV/m] Coupling factor [%] 5. 78 4. 26 3. 00 3. 02 Length (flange to flange) (cm) Beam pipe radius (mm) 300 160 (e. RHIC) 120 110 ERL workshop June 7 -12 2015 6

HOM damping for conventional lattice ERL at BNL ØThe average monopole power: Loss factor k// =3. 6 V/p. C, Qbunch: Bunch charge, Ibeam : Beam Current For conventional e. RHIC: 50 m. A x 3. 5 n. C x 3. 5 V/p. C x 12 passes => 7. 3 k. W per cavity ØDamping requirement for dipole modes: Wencan Xu ERL workshop June 7 -12 2015 7

HOM measurements on BNL 3 cavity Black: closed Red: Open Dipole Quadrupole Ø A 3 -D bead-pulling setup with capability of motion in longitudinal, radial and azimuthal directions was developed to map the field in the cavity; This is very useful for identify HOMs. Ø The detachable design of beam pipes allows us to measure/compare HOMs’ information due to different layouts. Ø The measurement results verifies the excellent propagating properties of the BNL 3 cavity through the dramatic reduction of the loaded Q values in the open beam pipe case. Wencan Xu ERL workshop June 7 -12 2015 8

Vertical test of BNL 3 Nb cavity ØIt reached 17. 1 MV/m with CW RF input, and Q 0 is 2. 1 E 10, which results in a total Rs only 9 nΩ ( RBCS is estimated 4 nΩ and residual R is 5 nΩ). Ø The gradient was only tested up to 19. 7 MV/m, due to safety administration limitation (20 MV/m). Wencan Xu ERL workshop June 7 -12 2015 9

BNL 3 Nb cavity for Coherent Electron Cooling POP RHIC beam electron beam ØThe cryomodule will be delivered to BNL in a few weeks. Wencan Xu ØCommission the cryomodule before and during next RHIC run. ERL workshop June 7 -12 2015 10

422 MHz BNL 4 cavity for FFAG lattice ERL e. RHIC Ø The frequency of the main linac cavity is 422 MHz cavity (comparing with 704 MHz cavity) for higher BBU threshold current, longer bunch length, smaller energy spread, higher polarization, smaller transient effect, lower HOM power, easier path control, higher RF power efficiency… Ø The linac consists of 42 5 -cell cavities with gradient of 18. 5 MV/m, which provide 1. 322 Ge. V energy gain in one pass. Ø Compact and efficient HOM damping is required due to space limit in the RHIC tunnel Wencan Xu ERL workshop June 7 -12 2015 11

cavity for FFAG lattice ERL Parameters BNL 1 BNL 3_scale BNL 4 Frequency [MHZ] 704 703. 87 422. 2 beta 1 1 Cells No. 5 5 Geometry Factor 225 283 273 (R/Q)/cell [Ω/cell] 80. 8 101 100. 6 Epeak/Eacc 1. 97 2. 46 2. 27 Bpeak/Eacc [m. T/MV/m] 5. 78 4. 26 4. 42 Coupling factor [%] 3. 00 3. 02 2. 83 Length with HOM damper (cm) 272 160 (e. RHIC) 267 256 Beam pipe radius (mm) 120 110 183 180 Radius after tapered (mm) N/A 50 83 80 483 498 First HOM frequency (MHz) Wencan Xu ERL workshop June 7 -12 2015 12

Transverse BBU estimation Note: 1) Beam parameters: 18 m. A 16 -pass ERL, T 21= 150 m; 2) The threshold impedance has divide by N^2 = 256 for 16 passes; 3) So, very conservative beam parameters are used in threshold estimation. Wencan Xu ERL workshop June 7 -12 2015 13

Beam-break-up simulation in FFAG e. RHIC (21 Ge. V) ØInjection energy 12 Me. V; ØTotal length of linac 115 meters, and energy gain is 1. 322 Ge. V; ØTwo cases: 12 passes to reach 15. 8 Ge. V, 16 passes to reach 21. 1 Ge. V. Δf/f Threshold (m. A) STD (m. A) 0 53 N/A 5 e-4 95 7 1 e-3 137 14 3 e-3 225 22 1 e-2 329 37 ØThe BBU simulation for the 21 Ge. V case shows that the threshold is way above the operation current 18 m. A; ©Y. , Hao Wencan Xu ERL workshop June 7 -12 2015 14

Longitudinal BBU Note: 1) Beam parameters: 18 m. A 16 -pass ERL, R 56 = 10 cm; 2) The threshold impedance has divide by N^2 = 256 for 16 passes; 3) So, very conservative beam parameters are used in threshold estimation. Wencan Xu ERL workshop June 7 -12 2015 15

HOM damping requirement for 422 MHz cavity ØDamping 7. 8 k. W (per cavity) of HOM power is required for full luminosity e. RHIC. ØA compact, high efficient HOM damping system should be developed. Wencan Xu ERL workshop June 7 -12 2015 16

HOM spectrum of e. RHIC electron beams Ø Frequency range: 0. 5 to 30 GHz (4 mm rms bunch length). Ø One of the optimization goals of the cavity design was to reduce to loss factor, which resulted in 1. 9 V/p. C. Ø Plan to use two types of HOM couplers to damp the full spectrum HOM. Wencan Xu ERL workshop June 7 -12 2015 17

Frequency jump for e. RHIC modes ØLorentz detunig factor is 0. 6 Hz/ (MV/m)^2 ØTuning sensitivity: 55 k. Hz/mm; ØSo the total tuning will be about 2. 05 mm or 113. 2 k. Hz, which stress is still smaller than 7000 PSI. Wencan Xu ERL workshop June 7 -12 2015 18

HOM damping concept tests with prototype 3 -cell cavity Ø 5 -cell 422 MHz cavity is largest multi-cell cavity over the world, there are rare fabricating/ processing facilities large enough in industry and labs and there is no experience in handle such a big cavities. ØA 3 -cell prototype cavity (end cells + one middle cell) will be built to resolve potential issues (because of its size and weight) for making e. RHIC cavities, and its performance will be studied to reach: 18. 5 MV/m, Q 0 > 5 E 10. ØHOM damping scheme will be tested with the prototype cavity at 2 K. ü 6 coaxial-line HOM couplers on the cavity for low frequency HOMs: ~ 5. 6 k. W ü 3 rectangular waveguide HOM damper on the beamline taper for high frequency HOMs: ~2. 2 k. W Wencan Xu ERL workshop June 7 -12 2015 19

Preliminary results of HOM damping S 21/d. B Coaxial line HOM coupler can transform to high frequency Power/W HOM ports takes more power than beampipe ports ØCombination of these two types HOM couplers will damp full spectrum up to 30 GHz. Wencan Xu ERL workshop June 7 -12 2015 20

Summary Ø We gained a lot of experience during the operation of the BNL 1 cavity in ERL prototype. Ø BNL 3 cavity has been designed for the conventional lattice e. RHIC. Ø HOM measurements and studies were carried out on the detachable copper prototype. And it shows that the HOMs were well-damped. Ø The BNL 3_AES cavity test results shows good performance. A cryomodule with this cavity was built for CEC POP project and will be commissioned this year. Ø A 5 -cell 422 MHz cavity was optimized for FFAG lattice e. RHIC. A 3 -cell prototype cavity is going to be built to address all the potential issues in cavity fabrication and processing. The cavity will deliver to BNL next summer and performance study will be carried out after that. Ø HOM damping for the whole HOM spectrum is a challenge for high current ERLs, we are making effort on design of a full spectrum HOM damping scheme and the prototype will be tested in next 2 to 3 years. Wencan Xu ERL workshop June 7 -12 2015 21