LIU-PSB Beam Dynamics WG #1 (kick-off, Longitudinal BU, Triple Harmonics)

Europe/Zurich
864/2-B14 - SALLE J.B.ADAMS (CERN)

864/2-B14 - SALLE J.B.ADAMS

CERN

30
Show room on map

Participants: Jeroen Belleman, Alan James Findlay, Klaus Hanke, Jocelyn Tan, Christoph Hessler, Heiko Damerau, Chiara Bracco, Andrea Santamaria Garcia, Wolfgang Bartmann, Fanouria Antoniou, Greta Guidoboni, Matthew Alexander Fraser, Adrian Oeftiger, Pangiotis Zisopoulos, Voncenzo Forte, Alexander Huschauer, Ana Garcia-Tabares Valdivieso, Danilo Quartullo, Simon Albright, Gian Piero Di Giovanni, Joël Repond, Alexandre Lasheen, Maria Elena Angoletta

 

Introduction and Objectives (Gian Piero Di Giovanni)

  • Introduction to the nature of the working group and its objectives.

  • LIU-PSB baseline requires the brightness to increase by a factor of two, major contributors:

    • Higher energy

    • New injection scheme (H- stripping)

    • Transverse and longitudinal painting

  • Many ongoing studies into different aspects of the problem, the working group is intended to provide a forum to discuss both LHC and non-LHC beams and provide feedback to LIU management

  • There are many studies in both longitudinal and transverse plane on going and planned for the future, as well as “other” studies such as transfer and injection

  • The intention is to meet every 3 or 4 weeks to discuss ongoing projects and ensure dissemination of ideas, with strong encouragement to document work

  • A LIU-PSB GitLab Repo has been created by A. Santamaria Garcia and a Twiki page is under development to help share code and best practice for different types of Mds

 

Overview of Longitudinal Beam Dynamics Measurements in the PSB (Simon Albright)

  • Main longitudinal changes:

    • New magnetic cycle

    • New RF systems (Finemet cavities) with large broadband impedance

    • Notch filter to reduce impedance of Finemet cavities

  • Finemet systems will allow distributed voltage, phase misalignment between cavities will cause a phase shift and a reduction in total voltage seen by the beam

  • Using the change in synchronous phase and bunch length with two cavities at h=1 it is possible to determine their relative phase, allowing the required time delay and azimuth compensation to be measured – reasonable experimetal results have been achieved

  • Longitudinal emittance blow-up to 3 eVs is the baseline for LIU-PSB, at current extraction energy this scales to 2.8 eVs

  • Using fixed bucket area leads to more effective emittance blow-up with a high harmonic RF system, current acceptance necessitates sequential stages of blow-up but the target emittance can be achieved

  • RF phase noise has also been used to reach 2.8 eVs, and a selection of other emittances have also been provided demonstrating flexibility in the approach

  • Space charge at injection can be reduced using three harmonics instead of the usual two

  • Numerical simulations suggest the footprint will be reduced, which should allow for higher brightness beams

  • At nominal LHC25 intensity a 10% brightness increase was seen, for MTE approximately 0.5 micron reduction in the vertical emittance was seen

  • Longitudinal painting has so far been studied assuming standard synchrotron operation, but fixed frequency injection changes the beam dynamics

  • Injection whilst assuming standard operation will lead to uncontrolled blow-up during first milliseconds of the cycle as the true separatrix is for a decelerating bucket

  • Experimental confirmation provided by S. Hancock in 2016 using phase space tomography

  • More studies are required including:

    • Longitudinal painting

    • Longitudinal emittance blow-up

    • Improvements to relative phase alignment between cavities at the same harmonic

    • Relative phase alignment between different harmonics

 

Overview of Longitudinal Beam Dynamics Simulations in the PSB (Danilo Quartullo)

  • Introduction to Beam Longitudinal Dynamics (BLonD) code and an explanation of interesting features, including multiple RF stations and harmonics, and LLRF features

  • Variety of examples of benchmarking studies with BLonD and comparisons with measurements

  • Significant work to give an accurate estimation of the space charge impedance, current value obtained using LSC code of SLAC, 633.14 Ω Z/n rescaled with energy through the ramp

  • Full impedance model currently includes beam pipe step transitions, Finemet cavities, kickers, space charge and some smaller sources

  • Effect of Finemet fast feedback loops to suppress induced voltage shown to have a large effect in simulation as expected

  • Simulations performed starting with a matched bunch at C300 to avoid injection, full impedance model, and phase and radial loops

  • Explanation of phase noise and impact of intensity effects on synchrotron frequency distribution, due to phase loop shaping the noise spectrum might lead to better results

  • LHC beams at 2 GeV with 8 kV only work without intensity effects, with intensity effects high voltage is required, 16 kV is sufficient and blow-up can be achieved with noise from C450 to C600

  • Additional simulations at a selection of voltages and intensities also gave acceptable results

  • For very high intensities (~1.6E13 ppp) an instability was seen around C500, possibly due to the peak in the Finemet impedance around 20 MHz

  • In current machine phase noise can be used to reach 2.8 eVs using quadrupolar excitation (twice synchrotron frequency), and simulations and measurements are in good agreement

  • Studies ongoing to replicate current performance of C16 for a selection of operatinal beams

  • Initial simulations of injection with fixed RF frequency give qualitatively similar results to measurement, including reduced losses

  • BLonD simulations have been used to study longitudinal beam dynamics issues for the PSB and are showing good agreement with measurements, reaching LIU-PSB baseline parameters appears to be possible with more studies ongoing

 

 

Questions for both talks saved until the end due to the overlap between the topics

 

Questions:

  • G. P. Di Giovanni asks about the instability seen in simulations of high intensity and if the intensity threshold is known: Threshold not yet clear, instability appears to be due to Finemet impedance so may be avoidable with more feedback loops and careful design of voltage program

  • H. Bartosik asks about simulations of different cycle types: Changing from entirely single harmonic to double harmonic improves the result, but instability is still seen. Work is ongoing to design and simulate more realistic voltage programs.

  • G. P. Di Giovanni asks if the impedance model is up to date: The current model is about two years old so more accurate information would be useful, especially at higher frequencies for the Finemet cavities.

  • G. P. Di Giovanni asks if the simulations including LLRF feedbacks are accurate or if more work is needed: Further validation work is planned, but current results look good.

  • G. P. Di Giovanni asks if a reliability run is planned for the phase noise: Reasonable stability with MD beams has been seen, but a full reliability run on an operational beam is desired.

  • G. P. Di Giovanni asks if the PS has considered using phase noise: H. Damerau says it would be possible to try

  • G. P. Di Giovanni asks if triple harmonic operation is beneficial to all beams: A benefit has been seen for some cycles, it is not yet clear which other cycles will benefit as there are different limitations

  • A. Findlay asks if longitudinal painting studies are planned for all considered combinations of harmonics and voltages: The plan is to study as wide a variety as possible

  • M. E. Angoletta asks how many harmonics are desired for bunch flattening: Adding additional harmonics follows a law of diminishing returns, it is not clear how many would be of benefit. An additional consideration is the total available voltage as sufficient voltage is required at h=1 for acceleration. It is worth studying more harmonics in the long term, but three is definitely sufficient for the medium term.

  • G. P. Di Giovanni asks how long it takes to measure the relative alignment between Finemet at h=1 and C02: It is in principle quite fast, the limit is down to the number of measurments desired. One measurement per RF frequency is technically sufficient, the more are taken the better the results are expected to be.

  • G. P. Di Giovanni asks about short term priorities: These include documenting existing results and improving numerical work.

 

Next meeting: Tentatively scheduled for 11-Dec-17 dedicated to impedance modelling

There are minutes attached to this event. Show them.