1 / 23

New PS one-turn delay feedbacks and further developments

New PS one-turn delay feedbacks and further developments. Thanks to Alfred Blas, Heiko Damerau , Valentin Desquiens , Simone Gilardoni , Guido Sterbini , Letizia Ventura, and all involved people from sections BE-RF-FB, IS and CS. Damien Perrelet BE-RF-FB CERN

molly-cooke
Download Presentation

New PS one-turn delay feedbacks and further developments

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. New PS one-turn delay feedbacksand further developments Thanks to Alfred Blas, HeikoDamerau, Valentin Desquiens, Simone Gilardoni, Guido Sterbini, Letizia Ventura, and all involved people from sections BE-RF-FB, IS and CS Damien Perrelet BE-RF-FB CERN LIU-DAY 2014, 11 April 2014

  2. Content Overview of the new feedback board Current implementations: PS 10 MHz 1-turn feedback and AVC digital loops PS transverse damper Future developments: PS 1-turn feedback cavities 13/20/40/80 MHz PS longitudinal coupled-bunch feedback PSB/AD/LEIR transverse damper upgrade Summary and outlook Damien Perrelet BE-RF-FB CERN

  3. The new hardware Damien Perrelet BE-RF-FB CERN

  4. DAC1 LP LP LP LP LP LP LP LP ADC1 4 SERIAL LINKS VME + TRIGGERS ADC2 DAC2 4 ADC RF INPUT CHANNELS → FPGA → 4 DAC RF OUTPUT CHANNELS ClockDistri NODLY dUPSTR dnUPSTR dDWSTR1 dDWSTR2 FPGA Stratix II EPS2S90F1020C4 LMH6702 ADCs AD6645 DACs AD9754 AD8009 CLK DWSTR1 Gap Return 1TFB TP New PS feedback hardware ADC3 DAC3 DRIVE IN Power Combiner G Ctrl Vprog Test VPROG TP 50 MHz DAC4 ADC4 G Ctrl h200 CLK UPSTR Digital BP1 G Ctrl AVC TP RF from BC Power Combiner G Ctrl Spare VGAP DET TP 50 MHz CLK DWSTR2 SRAM Damien Perrelet BE-RF-FB CERN

  5. Electronic board • Generic VME board • STRATIX IIFPGAand external SRAM • 4 RF channels ADC and DAC, low latency, 14 bits and running at 105/125 MSPS • 2 multiplexed fine delay lines for glitch-free delay changes (DAC and FPGA clocking) • 4 serial links CVORB or fast 40 Mbit/s Damien Perrelet BE-RF-FB CERN

  6. PS 10 MHz loops Damien Perrelet BE-RF-FB CERN

  7. Final amplifier, 2.8-10 MHz cavity, fastwidebandFB PS 10 MHz cavity feedback overview Gap return - Slow voltage control loop (AVC)  Gain control at fRF Reduction of cavity impedance seen by the beam: - Fast wide-band feedback around amplifier (internal)  Gain limited by delay - 1-turn delay feedback  High gain at n  frev Drive 1TFB DAC ADC AVC DAC ADC hn h200 Vprog RF from BC Damien Perrelet BE-RF-FB CERN H

  8. PS 1-turn delay feedback Damien Perrelet BE-RF-FB CERN

  9. FPGA VGAP Detected ADC NOTCH FILTER COMB FILTER AUTO DELAY DAC 1TFB OUT to Drive hn VME INTERFACE VPROG NOTCH BYPASS RF CLK h200 d UPSTR d DWSTR Cascade of notch + comb IIR fractional notch hn/hs Periodic comb at frev 1-turn delay feedback Suppress the RF component to not work against AVC High gain at frev multiples DPRAM hn Fine delayed clock Unique sampling clock h200integer multiple of the revolution frequency Damien Perrelet BE-RF-FB CERN

  10. Spectrum at cavity gap voltage of a single high intensity bunch 8∙1012 ppb Filters only and closed loop transfer functions Beam induced spectrum of four bunches during batch compression 10 db/div h = 8 Measurements with beam before LS1 time FB off h = 8  13  20 10 db/div Feedback succesfully follows harmonic changes FB on time amplitude frequency frequency Damien Perrelet BE-RF-FB CERN

  11. PS digital AVC loop Damien Perrelet BE-RF-FB CERN

  12. Collaboration RF from beam control ADC RF-FB RF-IS RF veto Voltageprogram Drive + Error[k] PID controller AVC loop and AVC surveillance DAC - Modulator Detected gap voltage Veto Power amplifiers hn I CORDIC Gap return Cavity 10MHz non-IQ detection ADC Q • Amplitude looponly, future phase loop or move to IQ loop • Voltage detectionusingnon-IQ algorithm at givenharmonichn • High power hardware safetyensured by separate surveillance module H. Damerau, V. Desquiens, M. Haase, D. Perrelet Damien Perrelet BE-RF-FB CERN

  13. Stepresponse of the controlleraround a cable 500 ms test pulse voltage program Response with fast voltage rise Vprog Digital AVC loop measurement Vdet Vprog Vdet Error RF drive 15 turns Error Error RF drive 100 ms/div 10 ms/div • Digital PID withadjustableparameters • Regulation to full voltage withoutovershoot in about 15 turns Damien Perrelet BE-RF-FB CERN

  14. Four new VME crates installed in building 359 (one for each tuning group) • Example for tuning group B(cavities C56, C66, C76 and C81) Installation in the machine • Full commissioning of the new system during start-up 2014 Damien Perrelet BE-RF-FB CERN

  15. PS transverse damper Damien Perrelet BE-RF-FB CERN

  16. PS transverse feedback Firmware: A. Blas • PU flight time compensation • Periodic notch filter for orbit offset suppression • Hilbert filter for phase compensation at betatron frequencies • Automatic delay to complete 1 turn Damien Perrelet BE-RF-FB CERN

  17. See talk by Raymond Wasef Damper to cure horizontal and vertical transverse instabilities Damping of head-tailinstability PS transverse feedback Damping of injection errors PS TFB ON PS TFB OFF 10 ns/div 10 ns/div → Present PS TFB designed to damp 44 μm/turn at 1.4 GeV (BW 23 MHz). A.Blas, S.Gilardoni, G.Sterbini Damien Perrelet BE-RF-FB CERN

  18. New applications Damien Perrelet BE-RF-FB CERN

  19. Frequency conversion board • 1) Analog conversion to IF and IQ digital processing • → Using existingand analog frequency conversion boards • or • 2) Direct sampling • → Modification of existing board • to achieve higher sampling capabilities • or • → Build a complete similar new and fast board PS 1-turn feedbacks cavities 13/20/40/80 MHz Thomas Truszczyński Mini-PCB to fast ADC adaptation Damien Perrelet BE-RF-FB CERN

  20. PS longitudinal coupled-bunch feedback See talk by Letizia Ventura Cross-damping: hdet+ hkick = hRF = 21 → Detect at h = 13 and Damp at h = 8 → Make a new digital coupled-bunch feedback for the Finemet cavity ADC DAC Beam (WCM) Longitudinal kicker (Finemet) H. Damerau, L. Ventura: • Demonstrated with measurements using existing feedback and the spare cavity Damien Perrelet BE-RF-FB CERN

  21. Future developments • Target to other machines: • Upgrade of PSB, AD and LEIR transverse dampers • PS 10 MHz possible improvements: • Cavity phase loop or IQ loop (Reliability of RF manipulations) • Cavity phase compensation (More gain for 1-turn feedback) • Multi-harmonic RF generation on-board and use of hs=256 Damien Perrelet BE-RF-FB CERN

  22. Summary and outlook - New generic feedback board meets requirements → Complex design buteasy to use and allows lots of features Two operational loops: PS 10 MHz 1-turn feedback and AVC → Fully digital and remote controlled → Commissioning during setting-up 2014 - New operational PS transverse damper → Good results and required for LIU beams - Many future developments and projects using this hardware → FPGA gives reliability and flexibility for new ideas Damien Perrelet BE-RF-FB CERN

  23. Thanks for your attention Many thanks to people from sections RF/FB-CS and RF/FB/IS for collaboration and support : Alfred Blas, Philippe Baudrenghien, Andy Butterworth,HeikoDamerau, Valentin Desquiens, Simone Gilardoni, GregoireHagmann, Matthias Haase, Wolfgang Hofle,Michael Jaussi, GerdKotzian, Tom Levens, Gérard Lobeau, John Molendijk, José Noirjean, Anthony Rey, Maarten Schokker, Guido Sterbiniand Letizia Ventura. Damien Perrelet BE-RF-FB CERN

More Related