EMMA MA RF System C. Ohmori and J. S. Berg EMMA MA System * Many - PowerPoint PPT Presentation

EMMA MA RF System C. Ohmori and J. S. Berg

EMMA MA System * Many FFAG applications require slow acceleration  * Linear non-scaling FFAGs cross many resonances  - Nonlinear resonances - Imperfection resonances * Resonances damage beam more when you cross them slowly  * There is thus a minimum rate at which you can cross resonances  - May depend on magnitude of errors (Machida) 10 -2 6 4 normalized DA [ ฀ m rad] * Low-frequency RF to allow slow acceleration  2 - EMMA as-is only allows very rapid acceleration 10 -3 - Primarily due to high-frequency RF system 6 4 2 * Accelerate rapidly then reduce rate 10 -4  - Start with 100 turns to insure success 6 4 - Reduce acceleration rate and study effects 100 Turns 2 1,000 Turns 10 -5 FFAG09 0 20 40 60 80 100 1 ฀ of alignment error [ ฀ m]

Parameters of MA system Frequency 18 MHz comments Frequency sweep 3 % Possible up to ~8% Total voltage 100 kV 100 turns / cycle Number of cavities 3 Voltage 33.3 kV / cavity 50 kV to run further off-crest Length of cavity 10 cm Number of cores 2 / cavity Size of core 27 O.D, 10 cm ID, 2.5 cm thickness Cut/un-cut core Cut core Q-value About 11 1000 Ω Cavity impedance FT3L will be used. Core material FT3M FFAG09

EMMA MA CAVITY FFAG09

Technical Issues to solve Need high impedance to get > 33 kV/cavity • High impedance MA cores  Resonant frequency at 18 MHz • Cut core cavity  High voltage in a small space • Space around beam pipe to avoid discharge  Ceramic gap to stand high voltage of > 1.6 kV/mm  High power amplifier and power supply • Compact AMP and PS for low duty  FFAG09

Cavity Impedance  Assuming ordinary MA material is used  The cavity will have about 600 Ω at 18 MHz  Large O.D/I.D ratio is necessary.  If I.D. is 10 cm, O.D. will be about 30 cm.  Assuming to use PRISM amplifier to drive − Rf current of 60 A is available.  About 30 kVp per cavity might be achievable ?  If cavity impedance is high, higher voltage can be obtained and/or system will be compact . FFAG09

Improvements of cavity impedance uQf (磁性体コアの特性） We are interested in higher • impedance core for J-PARC 14 beam-power upgrade. 12 ∝ shunt impedance 10 A new material (FT3L, 13 µ m • uQf (GHz) 8 thickness) has two times higher impedance than 6 ordinary Magnetic Alloy 4 uQf(FT3L,13um) (FT3M, 18 µ m) . uQf(FT3L,18um) 2 uQf(FT3M,18um) 0 So far, up to 10 cm core was • 0 1 2 3 4 5 6 available. Now, up to 27 cm frequency （MHｚ） core becomes available. More than two times higher impedance is expected at higher  frequency. Gap voltage will be twice higher if amplifier has enough power. FFAG09

Improvement of cavity impedance 4 MA cores using a new material is  produced (27 cm O.D) for J-PARC R&D. Impedance measurement, next  week. Preliminary results are good as we planned : 500 Ω /core. Test of immersion and cut core are  next step for J-PARC and EMMA. These cores will be references (or  used) for EMMA MA system design. Using high impedance core, FT3L core with 27 cm O.D. system design will be easy. FFAG09

Cut core configuration Un-cut core has a larger inductance and the resonant frequency will  be low because of floating capacitance in cavity and amplifier. Cut core configuration is a possible way to reduce the inductance.  High impedance will increase Q-value and it is preferable as it can  avoid waveform distortion. Obtained bandwidth is wide much enough to sweep frequency by 3%.  Floating capacitance About 100 pF 0.8 µ H Core inductance 1000 Ω Shunt impedance Q-value About 11 Bandwidth 1.5 MHz FFAG09

Ceramic Gap  Length of ceramic gap will be about 2 cm because cavity length is 10 cm.  30 (-60) kV will applied on it.  Ordinary ceramic pipe can not stand for 1 kV/mm on surface.  Need to stand at 1.5 (-3) time higher voltage.  A new ceramic made by Kyocera can stand 4 times higher field (4kV/mm). A beam pipe for J-PARC RCS cavity #12 will use it. FFAG09

Compact amplifier and anode power supply  Peak RF power is 400 kW  33.3 kV X 24 A  To save the system cost,pulse operation of small vacuum tubes is preferable. Duration is 5.6 µ s (100 turns/18 MHz) and repetition rate is small (~1 Hz).  PRISM amplifier and anode power supply can be used.  50 kW tetrodes, 4CW50,000, can be used. (150 kW tubes, 4CW150,000, are used for PRISM)  Specification of APS is much smaller than PRISM. Combination of small DC PS and capacitor bank will reduce cost. o  PRISM amplifier still sufficient for 50 kV FFAG09

summary  RF system using MA cavity is designed.  Using a new MA material, FT3L, the system will be compact and fit all requirements.  If you need 2 times faster acceleration, PRISM amplifier and APS will be a suitable system to drive.  R&D on FT3L and new ceramic are carried on at J-PARC for upgrade. FFAG09