15 th ESLS RF Meeting 2011 ESRF 5

15 th ESLS RF Meeting 2011 ESRF, 5 th – 6 th October Status of ESRF RF Upgrade Jörn Jacob, ESRF 15 th ESLS RF, Grenoble, 5 -6 October 2011 J. Jacob: Status of ESRF RF upgrade 1

RF upgrade overview Cell 5 Cav 1 & 2 Cell 7 Cav 3 & 4 Recall [14 th ESLS RF’ 2010] Klys 1 6 Ge. V e- Storage Ring Klys 2 Teststand Existing 352. 2 MHz ESRF RF system SY Cav 1 & 2 Klys 0 Booster 352 MHz 1. 3 MW Klystron Klys 3 5 -cell cavities: strong HOM ! Cell 25 Cav 5 & 6 15 th ESLS RF, Grenoble, 5 -6 October 2011 Thales TH 2089 J. Jacob: Status of ESRF RF upgrade 2

RF upgrade concept Cell 5 Cell 7, length 5 m 7 m 1. 18 new single cell HOM damped cavities for 300 m. A operation ID ID 150 k. W End of 2004 start R&D Today: 2 prototypes delivered 2. Solid State Amplifiers 150 k. W 150 k. W Extremely Modular, Klys 2 Intrinsic redundancy, Teststand High global operation MTBF, … à 4 x 150 k. W SSA for the Booster 150 k. W à 18 x 150 k. W SSA for the Storage Ring à 1 st SY Cav 1 & 2 batch: 7 SSA in fabrication by ELTA 150 k. W Recent review of ESRF upgrade strategy pulsed Storage Ring 150 k. W Budget reductions: 3. In house development of SSA: 150 k. W no additional long beam line in cell 7 Cavity combiner for 132 modules Booster Planar balun transformer 300 m. A option not retained in phase 1 of ESRF upgrade cavity series fabrication frozen 150 k. W 150 k. W Cell 25 150 k. W ID SSA still a priority as alternative to klystrons, however, also SSA program will be adapted ID Cell 23, length 5 m 7 m 15 th ESLS RF, Grenoble, 5 -6 October 2011 J. Jacob: Status of ESRF RF upgrade. . . 3

Cell 5 Cav 1 & 2 RF upgrade phase 1 well in progress ! Cell 7 Cav 3 & 4 Klys 2 Klys 1 Teststand SY Cav 1 & 2 4 SSA from ELTA for the booster: 150 k. W Commissioning of first 150 k. W Amplifier unit on variable dummy load (EH tuner) underway pulsed 150 k. W 3 SSA from ELTA for SR: Storage Ring Commissioning of all 4 150 k. W Powering 3 new HOM damped cavities on the storage ring amplifiers on the ring in March 2012 Booster 3 prototype HOM damped cavities … Test with beam one by one on cell 25 with klystron transmitter TRA 3: installation of RI cavity started yesterday 5 th October 2011 ! 150 k. W Klys 3 150 k. W … 3 prototype HOM damped cavities 2013: all 3 cavities in new 7 m section/cell 23 with 3 SSA Cell 25 Cav 6 & new cavity for test ID ID Cell 23, length 5 m 7 m 15 th ESLS RF, Grenoble, 5 -6 October 2011 J. Jacob: Status of ESRF RF upgrade 4

* This work, carried out within the framework of the ESRFUP project, has received research funding from the EU Seventh Framework Programme, FP 7. Summary of Cavity Status [for details, see presentation by V. Serrière, project leader] 3 cavity prototypes: • RI – Research instruments: Ø Fabricated according to initial specification by e-beam welding of HOM coupling sections and vacuum brazing for all remaining assemblies Ø Delivered 15 June 2011 Ø Excellent fundamental mode impedance: Rs = 4. 9 M , Q 0 = 33800 Ø HOM spectrum a factor two lower than design goal Ø 600 k. V obtained in CW on RF power test stand RF fingers between HOM coupling section and damper: no sign of degradation 1 st successful step in validating our approach Problem with fieldprobe housings: leak after exposure to high field Research Instruments Cavity in ESRF power teststand: 0. 65 MV 20 nm Ti coating (solved similar problems on the booster in 2008) Ø Cavity being installed on the storage ring / cell 25 for beam tests • SDMS cavity: Ø Fabricated according to alternative design using e-beam welding for most of the assembly steps Ø Delivered on 20 July 2011 Ø Installation on cell 25 in summer 2012 • CINEL Ø Still in fabrication Ø Beam tests foreseen in first half of 2013 • Installation of all 3 cavities on cell 23 in Summer 2013 15 th ESLS RF, Grenoble, 5 -6 October 2011 SDMS cavity J. Jacob: Status of ESRF RF upgrade 5

Status of SSA from ELTA [ESRF project leader: J. -M. Mercier] Order 7 x 150 k. W SSA from ELTA / AREVA: Qualifying SSA as alternative to high power klystrons to guarantee long term operation of ESRF preparing possible replacement of other klystrons in further upgrade phases. Saving electrical power on the booster thanks to the anti flicker system needed for the cycling at 10 Hz. Qualifying with beam a new storage ring RF unit comprising 3 HOM damped cavities powered by 3 SSA, which could replace one klystron powered five-cell copper cavity. ELTA design based on SOLEIL SSA: (Transfer of technology from SOLEIL) 650 W – 352. 2 MHz RF module Ø developed by SOLEIL Ø 6 th generation LDMOSFET = BLF 578 / NXP Ø Vds = 50 V Ø Gain: 20. 3 ± 0. 2 d. B Ø Efficiency: 68 to 70 % Ø Low transistor & circuit temperature: < 60 °C Ø Circulator: Insertion loss < 0. 2 d. B, Isolation > 26 d. B Summer 2010: • 500 hours fatigue test with 7500 ON/OFF cycles • 16 RF modules at 10 k. W: without any detected degradation 15 th ESLS RF, Grenoble, 5 -6 October 2011 J. Jacob: Status of ESRF RF upgrade 6

Status of SSA from ELTA, followed 75 k. W tower Ø 128 RF modules Ø Coaxial combiner tree with l/4 transformers 150 k. W amplifier = 2 towers Spec: h > 55 % at 150 k. W h > 45 % at 100 k. W 150 k. W / full reflection for 20 ms 150 k. W / 50 k. W reflection permanently 80 k. W / full reflection permanently April 2011: First 75 k. W tower at the ESRF, tested at full power on load Mismatch checked for all phases with E/H tuner Initially h 59 % degradation after 1000 h run test h 57 % (still above spec) 15 th ESLS RF, Grenoble, 5 -6 October 2011 J. Jacob: Status of ESRF RF upgrade 7

Status of SSA from ELTA, followed Degradation of one module 80 Efficiency [%] 60 IDS BBEF meas. ESRF 1000 h ESRF new polyn. fit 40 20 Saturation at maximum power 0 0 200 400 600 800 Output power [W] VGS 1/Rload VDS 50 V t t Tentative explanation by ESRF / simplified transistor model showing fundamental RF signal / needs to be confirmed ü NXP: checked transistors are still in spec ü ESRF experiment: Partial cure by UV radiation indicates electron trapping in crystal defects of LDMOSFET channel, but not only explanation ü ELTA is currently investigating this phenomenon ü Only little degradation on SSA efficiency, since nominal module power around 610 W, i. e. at the limit of the observed effect 15 th ESLS RF, Grenoble, 5 -6 October 2011 J. Jacob: Status of ESRF RF upgrade 8

Status of SSA from ELTA, followed EH-tuner: to adjust any mismatch at any phase for the SAT • • • Delivery of first 150 k. W SSA for the booster on 2 September 2011 (late by 2 months) Currently in 200 hours run test since last week (interrupted at 160 hours for 2 days due to shut down of services) Efficiency at 58. 5 ± 1%, no significant degradation so far Delivery of remaining amplifiers for the booster still foreseen this year: 1 tower per week This run test with local control from ELTA, however interlocked by ESRF control system 15 th ESLS RF, Grenoble, 5 -6 October 2011 J. Jacob: Status of ESRF RF upgrade 9

400 V ac / 280 V dc power supply for the booster SSA Recall [14 th ESLS RF’ 2010] Constant power 300… 400 k. W Reduced power consumption 10 Hz Mains 400 V ac Peak DC power 1000… 1200 k. W 600 k. W RF 280 V dc Power Supply 400 k. W average Anti-flicker C 3. 2 F In house built capacitor cabinets • One 280 V dc / 400 k. W PS for the booster SSA 2 booster cavities • Four 280 V dc / 360 k. W PS: 3 for the 3 SR x “ 2 for couplers SSA and 1 as “hot spare the booster SSA (slightly reduced power for normal injection use of the booster) Developed by ESRF Power Supply Group 15 th ESLS RF, Grenoble, 5 -6 October 2011 J. Jacob: Status of ESRF RF upgrade 10

Booster RF : 4 150 k. W amplifiers 2 five-cell cavities x 2 couplers Recall [14 th ESLS RF’ 2010] 4 Waveguide switches to 4 water loads Directional couplers SY: Syn Booste chro r tron 75 k. W tower Existing transmitter room SYRF 15 th ESLS RF, Grenoble, 5 -6 October 2011 J. Jacob: Status of ESRF RF upgrade 11

SSA control SSA - monitoring of: • For each RF module: Ø DC Current Ø Transistor socket and circulator load temperatures • RF forward and reflected power at each 5 k. W combination and each 75 k. W tower • Cooling water flow and temperature interlocks • “Muxboxes” interconnected via RS 485 4 control racks: (1 per SSA) • LLRF & Driver crate with RS 232 communication • Hardwired Fast Interlock crate with RS 232 communication • PLC crate / slow interlocks with TCP/IP communication • RF amplitude & phase measurement crate with RS 232 communication FPGA rack for booster waveform control 1 common rack: • Transmitter control computer (PCI Express) • PCI Rocket ports (RS 232/422/485) communication with local control racks and ELTA “muxboxes” • LLRF crate with FPGA based IQ regulation loop for the 10 Hz / 50 ms booster pulse modulated RF signal distributed to 4 individual SSA 15 th ESLS RF, Grenoble, 5 -6 October 2011 IQ modulator / demodulator LLRF rack [project leader: G. Gautier] J. Jacob: Status of ESRF RF upgrade 12

The R&D of cavity combiner will receive funding from the EU as working package WP 7 in the framework of the FP 7/ESFRI/CRISP program, involving CERN-SLHC, ESS, GSI-FAIR as partners. In house R&D of SSA [for details, see presentation by M. Langlois, project leader] Objectives develop the required internal expertise for the follow up of more than 1 MW of SSA further develop this technology to reduce size & costs and to increase the efficiency 1. RF module with planar balun easier mass production, reduced fabrication costs ü 1 st version limited to 400 W by component temperature ü 2 nd improved version under test 2. Cavity Combiner • 6 rows x 22 Columns x 600 … 800 W per transistor module 75 … 100 k. W • More compact than coaxial combiners ßwaveguide nmodule x ßmodule >> 1 • Easy to tune if nmodule is varied • Substantial reduction of losses higher h ü Lab prototype with 18 inputs under test ü Goal within FP 7/ESFRI/CRISP: 75 k. W prototype for ESRF & feasibility studies for CERN, ESS and FAIR at other frequencies 15 th ESLS RF, Grenoble, 5 -6 October 2011 J. Jacob: Status of ESRF RF upgrade 13

New cavity high power coupler using LHC window • Already reported at previous ESLS RF meetings • Collaboration contract CERN / ESRF / SOLEIL • First prototype coupler: Ø Delivered at ESRF last week Ø “Plug & Play” coax to waveguide transformer expected this month, allowing to replace existing coupler without modifying waveguide layout Ø Tests on spare 5 -cell cavity foreseen at ESRF in October- December 2011 • Prototype coupler for SOLEIL cavity will be tested later at ESRF using a dedicated warm cavity from CERN • Furthermore, fabrication of Ø 2 series couplers for ESRF Ø 2 series couplers for SOLEIL Ø 1 prototype to be tested this autumn + possibly several series couplers for APS (extra contract) Ø 4 prototypes for SPL Many thanks to Eric Montesinos / CERN for his great involvement and for leading this project ! 15 th ESLS RF, Grenoble, 5 -6 October 2011 J. Jacob: Status of ESRF RF upgrade 14

Thank you for your attention ! Marc Dubrulle Hervé Delamare Jean-Maurice Mercier Jörn Jacob Eric Rabeuf ESRF Linac / Injection. Extraction / RF Group Alain Panzarella Baroudi Boucif Vincent Serrière Paul De Schynkel Philippe Chatain Michel Langlois Georges Gautier Nicolas Michel Massimiliano De Donno Didier Boilot 15 th ESLS RF, Grenoble, 5 -6 October 2011 Denis Vial Pierre Barbier J. Jacob: Status of ESRF RF upgrade 15