Collimator Wakefield Issues

1
Collimator Wakefield Issues

• Outline
• Introduction
• Wake potentials
• Measurements
• Project Plan
• Conclusions

Nigel Watson CCLRC RAL/PPD
2
Introduction

• Success of LC requires
• High (integrated) luminosity
• Acceptable backgrounds
• Machine protection
• Detectors must be able to turn on and use
  recorded data
• High luminosity
• Minimise emittance growth, DR ? IP
• Control relative beam-beam motion at IP
• High beam currents

3
Spoiler/Absorber

• Thin (lt1X0) spoilers
• spread out beam, multiple Coulomb scattering,
  dE/dx
• Low z bulk, but require durable, low resistivity
  surface
• Damaged surfaces significant effect on WF
  behaviour
• Thick (20-30X0) absorbers, in spoiler shadow

SLC, 20mm Au coating on Ti Decker et al,
Linac96
4
Collimator Wakefields

• Conservative system design, to remove 0.1 of
  beam
• Avoid large amplitude component from entering FF
• Mechanical collimators close to beam
• Continuously scrape halo (10kW)
• Enlarge spot size of mis-steered beam by Coulomb
  scattering
• Beam excites wake potential in material, acts
  back on (tail of) bunch
• Transverse loss (kick) factor, ?yKy (near
  axis)
• Position jitter ? angle jitter
• Jitter amplification a la TRC AbKb
  p(1Ab2)1/2
• Emittance growth De/e ? A2b

5
Jitter Amplification Comparison
De/e ? A2b

• Vertical plane, unacceptable jitter
• NB with tail folding OFF, pessimistic/conservativ
  e
• Tail folding loosens collimation requirements
• Reduces significance of collim. wake fields
• Experimental verification?

6
Collimator Wakefields

• WF in vertical plane important even in error free
  machine
• Collim at betatron phase of FD most criticaL
• Contribute to position jitter of beam at IP
• Separate into to reduce effect use
• Geometric wake ? smooth tapers
• Resistive wall wake ? high conductivity
• Roughness impedance ? high quality finish
• Very difficult to calculate analytically -
  possible only for simple configurations
• Difficult to model, esp. for short bunches
  (sz300mm), shallow tapers (a20mrad), small ½
  gaps (b0.4mm), in reasonable time

7
Data

• Recent measurements using dedicated facility at
  SLAC, study geometric and resistive wakes
• Improvements to theory (Stupakov et al)
• Geometric wakes for tapered, rectangular
  collimators
• inductive (shallow tapers)
• intermediate regime
• diffractive (steep tapers)
• Resistive wakes (Piwinski et al)
• Analytic calculations used in TRC, assuming
• ? is Cu
• no tail folding
• near-axis wakes (linear, dipole region)
• Near-wall wakes (non-linear) possible machine
  protection issue

8
SLAC CollWake Expt.
At 1.19 GeV point in SLAC linac
sz 650mm
Magnet mover, y range ?1.4mm, precision 1mm
9
Near wall wakes

• Primarily study near axis wakes, dipole mode,
  linear region
• Add bump, study near wall region
• Non-linear, more important for machine protection

Kick angle (mrad)
Beam-collim. offset (mm)
From Tenenbaum, SLAC accel. seminar, Feb. 01
10
e.g. Resistive Wake Study

• Initial study was of geometric wake
• Second study compared Cu and graphite, same
  geometry
• Reasonable agreement with resistive wake theory

Kick angle (mrad)
Beam-collim. offset (mm)
From Onoprienko, Seidel, Tenenbaum, EPAC02
11
Third Set of Collimators

• Continue study of resistive wakes, compare Cu vs.
  Ti
• Thereafter, concentrate on geometric (perhaps
  two-step) tapers

12
Wakefield Reduction Methods

• Optimisation of collimator form need
  reliable/validated predictions
• Ideal case - infinite long taper, circular
• Realistic - include constraints from finite size,
  available longitudinal space, and adjustability
• 2-step tapers
• More complex shapes, non-linear tapers,
• Tail folding
• (and if all else fails) increase vxd radius at
  IP?

13
Project Plan

• Studying basic physics effects
• Not critically dependent on technology, but
  easier for cold machine
• Aim to design optimal spoiler jaw
  (material/geometry)
• Direct measurement of wakefields at SLAC
• Dedicated facility, single beam measurements
• Already one good collab. (Seidel/DESY) with
  Tenenbaum et al on graphite collimators
• Also previous UK involvement (Brunel)
• Progress slow need two interventions
  technical experts to install new collimators,
  plus operation time
• Turn around 1 set of 4 spoiler jaws per 1.5yr
  (3 between 99-04)
• SLAC willing to collaborate, situation evolving
  post ITRP decision

14
Project Plan

• Design/test optimal spoiler jaw profiles
  (material/geometry)
• Beam tests slow, need improved turnround to test
  new ideas
• RGC.s proposal (8/03) to use cold tests,
  building on groups expertise and existing h/w at
  DL/Lancaster
• Emulate beam by short pulse along thin wire
  stretched along axis of pipe with spoilers inside
• Microwave signal excites h.o. modes, then
  measure
• Distortion of current pulse on leaving structure
  in time domain
• Transmission parameter S21 as f(freq.)
• Integration of impedances -gt loss factors
• Study variation with displacement of wire from
  axis
• Challenging, but benefits from strong
  collaboration within CI (CCLRC/Lancaster)

15
Project Plan

• Set up cold test rig within CI (calibration,
  etc.)
• SLAC recently using coaxial wire in freq. domain,
  measure loss factors in NLC accelerating
  structures (Baboi, Jones et al)
• Do-able!
• Benchmark against known spoiler profile, large
  taper angle, MAFIA etc. simulation
• Carl has started to set up MAFIA simulations
• Extend to alternative designs (e.g. multi-step or
  non-linear tapers), extract at least relative
  performance
• Use cold test results to
• Provide data to assist development of improved
  e.m. modelling in problematic regimes
• Also within UK, build up expertise (non-LC
  applications)
• (Within EUROTeV) collaborate with TU Darmstadt
• Design one-off test proposal for direct beam
  measurement at SLAC
• With FP6 funding, extend into material damage
  studies

16
TDR impedance measurements

• Used for SRS components
• SRS bunch length at injection 20ps
• Fast step pulse ? gaussian via IFN, into vessel
• Receive pulse on sample scope, waveform analysis
  gives loss parameter

From HillPugh, EPAC94
17
TDR setup

• Carl B. examined mothballed h/w
• Many conical launch cones (need rectangular)
• Thin wire (250mm?)
• Consider vertical plane operation to avoid sag
• Critical issue
• Pulse speed!
• Time LC bunch sz
• 1 ps
• Quick survey, fastest available off the shelf
  pulse generator for TDR 10ps (Tek. 80E04
  modulePSPL module TDS8200 scope)
• Options
• build our own fast rise source
• scale up spoilers (but x10?!)
• alternative approaches?
• Lancaster freq. domain h/w?

1.7m
From HillPugh, EPAC94
18
SLAC

• Considering move of WF box to End Station A
• Significant improvement in access, share time
  with Mike Woods tests in ESA
• Access could allow 2 sets profiles/yr
• PTs initial estimates, kick resolution
  comparable to sector 2 location, even with higher
  beam energy
• Better single pulse resolution
• Longer lever arms for BPMs
• Timescale not fixed, but assuming no disasters
  crop up, expect to be done in current (US) FY
• Will help us, but does not remove need for cold
  tests

19
Improved Predictions
Yokoya formula

• Theoretical/Modelling
• Zagorodnov, Weiland ECHO
• Uniformly Stable Conformal approach very close
  to stable absolute error,
• indep. of collim. length
• Even in region close to origin
• Applicable for non-linear, near-wall wakes.
• Essential improvements, permit more reliable
  optimisation

transverse wake
20
From Zagorodnov and Weiland, TESLA Collab, Jan.
04
21
SummaryConclusion

• Understanding wakefields important for LC
• Calculations (Bane, Stupakov) and modelling
  (esp. near wall) much improved, e.g.
• constant (absolute) error, indep. of mesh
  ECHO (Weiland, Zagorodnov)
• More data will improve further
• More reliable LC spoiler design
• UK effort (Lancaster/CCLRC), starting programme
  of
• cold tests
• e.m. modelling
• beam test
• More ideas

Build up UK expertise