RF Commissioning in Point 4 Hardware Commissioning: • ACS RF System - Power Systems • ACS Cavities : Sector 4-5 • ACS Cavities : Sector 3-4 • ADT Power and Feedback Systems Commissioning Preparation for beam: • RF Synchronization • Beam Controls - ACS and ADT • Diagnostics and other Facilities Summary/Conclusions E. Ciapala On behalf of the AB-RF Group LHC MAC June 08 – RF commissioning in Point 4 1
28
Embed
RF Commissioning in Point 4 Hardware Commissioning: ACS RF System - Power Systems ACS Cavities : Sector 4-5 ACS Cavities : Sector 3-4 ADT Power and Feedback.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
RF Commissioning in Point 4
Hardware Commissioning:
• ACS RF System - Power Systems• ACS Cavities : Sector 4-5 • ACS Cavities : Sector 3-4• ADT Power and Feedback Systems Commissioning
Preparation for beam:
• RF Synchronization• Beam Controls - ACS and ADT• Diagnostics and other Facilities
ACS Cavity Commissioning – Cryogenics for RF cavities
LHC MAC June 08 - RF Commissioning in Point 4
Cavityies He tanks & circuits are low pressure systems, <2 bar but fed from magnet QRL
Concern: Overpressure in D-Line on multiple magnet quench (<20 bar estimated)Rely on: Process control, auto blocking outlet valve, non-return valve (NRV) in D-Line.Safety study: (SC, AT-CR & AB-RF ) Analysis of all risk situations (~10): EDMS 880723 => Recommendations => 1.5/1.8/2.1 release valves/rupture discs, procedures.Warm Recovery Line (WRL) installed to recuperate static losses when D-Line closed, to avoid release valves opening. Late decision. Extra outlet dome on module, cryo lines, heaters etc.
ACS Cavity Commissioning – Cool-down and first tests Sector 4-5
C8.B1
C7.B1
C6.B1
C5.B1
C4.B2
C3.B2
C2.B2
C1.B2
400.55
400.60
400.65
400.70
400.75
400.80
400.85
400.90
400.95
401.00
0%
100%
•Pressure test 2.1 bar done all modules (incl. Sector 3-4) Mid 2007
•Start cooling 2 modules (sector 4-5) on Nov 20th 2007•Careful setting up of cryogenic processes and safety systems•2 modules cold on Nov 22nd - 24 hours to stabilize
• Low-Power tests, Generator connected to w/g transition piece, measure resonant frequency and Qext
•Check of Tuning Range on 8 cavities Sector 4-5 (frf = 400.790 MHz)
ACS – Low Level RF systems – Cavity Controllers in UX45 Faraday Cages
Cavity Controller
•One system per cavity (in 2 VME crates)•Located in two Faraday cages in the UX45 cavern•Control phase and amplitude of cavity voltage.
Take reference from AB-PO standard function generators interfaced via serial link into cavity contoller.
• Minimize disturbances from•HT ripples from Power Converters: 1 % HT ripple -> 8.4
degrees @ 400.8 MHz•Transient Beam loading•Keep demanded klystron power reasonable and avoid
saturation (300 kW max). •Largely digital implementation•Operates at the bunch rate (40 Msps)
Modules/Functions:
Tuner Loop: Keeps cavity at optimum tuning to minimize klystron power (Also ‘half detuning ‘ to keep the power flat across beam segments and gaps)RF Feedback Loop: Reduces the cavity impedance at the fundamental (by 20 linear for Q=20000, by 180 at Q = 180000). Precision of RF voltage, transient beam loading and longitudinal stability1-T Feedback: Adds factor 10 reduction on the revolution frequency side-bands. (Transient beam loading + longitudinal stability)Klystron Polar Loop: Compensates for the klystron gain/phase changes. (HT drifts and ripples, 50 Hz components and multiple).Set Point: Voltage & Phase control, Interface to the function generator, can also customize the voltage reference for each bunch, phase and amplitude. Conditioning: Automatic conditioning system integrated. with local synthesizer. Also used for open loop set-upLongitudinal damper: Damps the injection phase and momentum errors, batch by batch Acts on 400 MHz cavities. (Postponed)
Cavity Controllers installed in FC B for Sector 3-4 cavities
•Processing simultaneously cavity and power coupler, fast gain loop acting on vacuum activity•Use same strategy as for initial conditioning in SM18 test stand:
Initial pulsing with slow rise-fall envelope, increasing voltage, increasing pulse width then DC, at all coupler positions. Keep low vacuum set limits throughout.
Conditioning time as expected. Quickly reached 2 MV/cavity (Nominal 5.5 MV/m) in pulsed mode
Then ~ 2weeks overall net time to get to CW operation at 300 kW
•Process handled by local DDS in cavity controller,
•Full remote control via network,
•Run all cavities simultaneously => time saving
Radiation measurements:
•Peak value: few mSv/h close to the cavities (8 cavities running)
Might get higher radiation above 2MV…
•No radiation measured nor in UX45 nor outside the RF zone
Access system de-bugged. Two systems: main tunnel system and upper UX45 system.
Some delays due to co-activities on equipment in RF zone. Handling of 18kV reset on door forcing, etc..
RAMSES RP monitoring systems tested & validatedPeak value: few mSv/h close to the cavities (8 cavities
conditioning simultaneously)Can expect higher radiation above 2MV, conditioning before
installation (in SM18) was done to 3 MV (8 MV/m)No radiation measured in UX45 nor outside the RF zone
Cryogenics systemClose collaboration with AT-ACR in setting up (important)Availability and reliabilty good..Some initial cryo control issues, (He lost through safety valves
instead of through WRL)Excellent cryogenic regulations (level & pressure) in the modulesSo far no observation of magnet quenches on RF cavities
RF Summary Sector 4-5
5 weeks cold in 2007, then 7 weeks in 2008Low power tests on all 8 cavitiesConditioning to nominal field and full powerSet up of loops on three out of 8 cavities. Need ~ 2weeks to finish the other 5….
Power Tests since Sector 4-5 commissioning Test with all 4 power converters simultaneously
at full power Optimize HV ripple on power converter (600Hz) Test spare power converter Test spare HV cables (surface to tunnel)
RF Commissioning in Sector 3-4….Cryo instrumentation etc. starting nowLow power tests on all 8 cavities NEXT
Front-end hardware:• PLCs for equipment control• VME Crates for LLRF systems and feedbacks
All front-end software based on FESA
Applications software:• LSA (Control room)• Applications for RF experts and system commissioning
Signal diagnostics accessible via OASIS:• Fast digitizers in Compact PCI crates• Embedded acquisition buffers in digital LLRF hardware
Slow Controls• PLC, FESA classes, specialist applications for power systems, Developed & tested• ACS & ADT slow control systems installed in UX45, tested and fully commissioned locally.• Remote control GUIs, based on Labview and AB/CO ‘Knobs’, operational and extensively used during commissioning.
LLRF controls (Cavity Controller, Beam Control, RF synchro)• Drivers for 30 different VME modules, based on rigorously defined memory map and functionality (streamlined design
environment built up)• Cavity Controller and Synchro front-end software operational; ‘Expert’ GUIs available• Beam Control front-end software implementation in progress
Interface with operational software• Parameter model and settings management defined in LSA; implementation in progress by AB/OP• Full remote control of cavity systems via LSA to be tested after Sector 3-4 commissioning (end June)
Synchronization of the SPS-LHC RF bunch into bucket transferGeneration of beam synchronous signals: 40 MHz bunch clocks, revolution frequencies, 40 MHz 7TeV reference
injection pulsesTransmission of timing & clocks to the users: BI, BT and Experiments, from SR4 via fibre linksFine-rephasing of the two rings before physics
Locking each ring on a low-noise fixed-frequency synthesizer.
One system per ring, located on the surface (SR4)Generates Master RF @ 400.8 MHz (VCXO output), phase is adjusted continuously. Update at 11 kHz revolution frequency
Three loops for each beam:Phase Loop locks the Cavity-Sum voltage (8 cav/beam) onto the Beam PU signal in order to minimize the RF phase noise. The loop is switched ON at first injection.Synchro Loop locks the VCXO output on a Frequency-Program DDS. The DDS output frequency is the injection frequency during filling and follows the B field during acceleration. The loop is ON before injection and remains ON all the time in normal operation (until re-phasing before physics when the DDS is replaced by a Synthesizer)Radial Loop to keep the beam centered during acceleration ramp. Can use for commissioning in place of Synchro loop.
R1
Low-levelLoops
Processor
Radial PUFront-end
VCXO
Phase Discri
F RF Prog 1
Radial loop
Phase loop
Synchro loop
DDS1 DDS2
Sync
F1
,P1
F2
,P2
F1 F2
1/h divider
Master F rev
To Ring 1 Cavity Controllers (fibers)
Dual Frequency Program and
Rephasor FPGA
Function Gen. Function Gen.
F RF Prog 2
DUAL FREQUENCY
PRGM
Master F RF
Beam 1
Rad. PU
Beam/Vt phase
RF/Fprog phase
F out
Ra
d P
os.
Fiber Optic TX
BEAM CONTROL
LOOPS MODULE
Radial steering with radial loop
Coarse F1
LF switch
Phase Discri
7 TeV synthesizer
Phase shifter
180 deg hybrid
a b
BEAM POS
MODULE
FPGA
AD
C
SYNCHROMODULE
Ib RF Summing Network
Vt
Phase PU
AD
C
CORDIC (+ AGC?)
Bunch/RF phase
CORDIC (+ AGC?)
Vt/RF phase
Phase Difference
and Averaging
BEAM PHASE
MODULE
AD
C
Cavities
Analog I/Q demod
Analog I/Q demod
Master FrfMaster Frf
AD
C
AD
C
Delay adjust
Function Gen.
fsync
fphase
Injection frequency, injection phase and stable phase will be adjusted by observing these two signals
A function sets the RF frequency on the injection plateau and through the ramp
The VCXO generates the RF sent to the Cavity Controllers
This synthesizer replaces the frequency program during physics
Critical Modules: Status -•Beam Phase - in final
development•Dual frequency program
– firmware testing•Beam Position –
firmware testing
System Test - with VCXO, Beam Control
loops, Synchro modules undergoing measurements in Lab.
Beam position VME moduleProduces normalized beam position digitally at 40 MHzSeries hardware completedFirmware developmentf
Signal processing VME moduleDSPU (“Damper Loop”)Produces vector sum of two PU signalsProvides notch filter at n * Frev, individual tof compensation for each module, loop gain adjust (via op DAC)Based on ACS 1T-FB moduleProto tested, series hardware being assembles, then firmwareSEC
Two Wideband Longitudinal Monitors per beam• One per beam to surface SR4, beam control system• One to local racks on cryo side of RUX45 tunnel - observation• Multiplexers and CPCi acquisition crates in place in UX45.• ‘Mountain range’ display & bunch length/profile measurements
Conclusions – Remaining Commissioning and Beam Preparation
Hardware Commissioning - Status and Remaining work
• ADT and ACS power systems fully set up, with all required facilities in place. Heat run of damper power systems to be done.• Sector 4-5 cavities conditioned to nominal voltage and power. Cavity controller loops set up on 3
cavities. 5 cavities still to complete, as soon as sector is cold• Sector 3-4 cavites; He tests (starting now) Low Power measurements, Conditioning and cavity
controller set-up as soon as possible• Check of function generators and software to complete (“vertical slice”)
Preparation for Beam
•RF synchro in place – clocks and timing now going to all users•ACS Beam control systems in advanced state but some items critical.•ADT electronics in test.•Software for beam control also critical, but basic functionality will be available for this run •Procedures for beam commissioning well defined.•Longitudinal Diagnostics in good shape to commission and study first beams….
A1, First Turn Pilot: Inject pilot and center first turn with RF OFFAdjust front end gains to see PU signals, Label bucketsPrepare injection frequency and bunch timing, injection timing, beam dump,
experiment clocks, cavity initial phasing
Beam Commissioning – ACS 400 MHz RF system
A2 Capture and Circulating Pilot at Injection energy.Commission phase loop and synchro loop. RF ON/OFfCaptureCheck cavity phasingAdjust relative positions of the 2 rings for collisions in IPs (cogging)
A3 to A11, Increasing Intensity to Collision on Flat TopPrecise measurements of lifetime, longitudinal profile Set up radial loopSet up multi-bunch injection, Commission the Filling Pattern maskSet up multi-batch injection, Commission the changing filling pattern mask, & update
in phase loopRamping, with function generators and software facilities, fine tuning of rampOptimization of the RF voltage on the flat topRephasing each ring to the 7 TeV Synthesizer (see diagram slide 15)Fine adjustment of collision point (OP and EXP)
A4, [450 GeV] Commissioning Damper Loop Measure de-coherence time with damper offMeasure open loop transfer function (mainly at ~low frequency)Make necessary adjustments (gain, phase, delay), Close damper loopScan gain, phase, delay and measure damping time and stability limitsMeasure beam lifetime as function of damper gainScan injection kicker pulse by moving bunch.
Beam Commissioning – ADT Transverse systems
Phase A1 and A2 First turn and circulating beam (1 to 156 bunches Single batch)Observation of beam at damper pick-ups Q7, Q9 and delay equalization:Verification of signal levels (sum signals, calibrate using orbit system)Delay equalization of damper pick-up signals from Q7 and Q9 (in SR4)Kick calibration:-Excite transverse oscillations (phase A2) in order to check available damper kick strength
A3, [450 GeV commissioning] - passiveCommissioning RF front-end (beam position module) of damper and check optics:Verify RF signals from RFLLCommission analog front-endCommission digitization and frev tagging of bunchCheck phase advance Q7->Q9->damper (both beams and planes) Verify beta functions.
A6, A7 RampCheck abort gap cleaning, test parameters & methodsCheck machine protection, BLM triggering fors low intensity bunch and the damper in anti-
damping. At 7 TeV, Again :
Measure open loop transfer function.Make necessary adjustments (gain, phase, delay), Close loop
Blow-up rate below the 24 h synchrotron radiation damping time.1 ps rms white noise just compensates synchrotron radiation damping. We measure 2.4 10-2 ps rms from DC to frev =11250 Hz .
Crossing the 50 Hz line during ramp: During ~ 1 min, 50 Hz falls inside fs band. Dangerous
0.2 % rms emittance increase If amplitude of 50 Hz line is increased by 10 linear, we get 27 % emittance increase with bunch centre reduced in population
Circulating beam at 450 GeV/c with 8 MV, 0.7 eVs (fs0=63 Hz).
50 Hz line multiples do not hit the populated synchrotron frequency band -> no significant effect observed in simulations1 ps rms white noise now gives 0.1 % loss after 1 hour.