Title: ILC Beam physics simulation Nikolay Solyak for ILC beam simulation group
1ILC Beam physics simulationNikolay Solyak(for
ILC beam simulation group)
- ILC main Linac lattice design from ILC BCD-like
to Realistic Realization - Single-bunch emittance preservation in ILC Main
Linac using Dispersion Free Steering - Static Tuning Local and Orthogonal Tuning Bumps
- Dynamic Alignment Ground Motion and Adaptive
Alignment - New Codes CHEF - DFS Implementation
ML Lattice Design
Emittance preservation in ILC
- Lattice defined by cryo segmentation (realistic
cold warm space drifts are used) - Linac curved along earth surface
- Two version of segmentation
- 8-8-8 scheme (Nov.2006)
- 9-8-9 scheme (Dec. 2006)
- Basic segmentations
- ILC Main linac will accelerate e-/e from 15
GeV? 250 GeV - Upgradeable to 500 GeV
- Emittance preservation is major issue (Luminosity
frontier) -
-
- Vertical plane - more challenging
- Large aspect ratio (xy) in both spot size
emittance (4001) - 2-3 orders of magnitude more difficult
Cryogenic Unit
RF Unit
SC String
Electron Main Linac (9-8-9)
Small normalized vertical emittance
Luminosity Scaling
undulator
Linac 15-150 GeV
Linac 150 -250 GeV
Orthogonal Correctors
BPM / Corrector Failure
Local Bumps
- Global Orthogonal Correctors
- Vector of particles final state (after DFS)
-
- Each corrector changes final state
-
- Minimum emittance growth when
- Perform SVD (orthogonal correctors)
- Columns of U are weights of correctors in
orthogonal correctors - Singular values of the matrix Ksvd are quickly
fall down. Only few first orthogonal correctors
do the job.
Dispersion and Wakefield bumps can reduce
emittance dilution significantly
5 randomly chosen vertical correctors not
working. Adjusted the adjacent two correctors to
guide the beam on to the design orbit
Orthogonal Correctors in TESLA-like (1Quad/4CM)
lattice
After DFS
Bump 1
Corrected emittance (nm)
Bump 2
Orbit at the YCOR (m)
bump Corrector 3 36
Nominal misalignments
BPM index
Dispersion only No Wakefields, No Quad roll (X/Y
coupling)
Faulty corrector not used in DFS
Emittance dilution (nm)
Corrected emittance (nm)
zoom
Seed number
- Locations of the dispersion bump is crucial to
get the optimal result - No significant improvement in the emittance
growth was observed after using three bumps - Wakefield bumps not help much after dispersion
bumps
BPM index
Exclude faulty correctors from the DFS and the
results are fine
Dynamic Tuning Adaptive Alignment and Ground
Motion
CHEF for LET simulations
- CHEF is C based Framework
- Successfully benchmarked against other codes in
FY06/FY07 - Study of static tuning of the misaligned ILC Main
Linac - Preliminary Study of the steering performance
with dynamic effects - CHEF on FermiGrid computer
Local method BPM readings (Ai) of only 3 (or
more) Nneighboring quads are used to determine
the necessary shifting of the central quad (Dyi)
(V.Balakin, 1991)
BPM res0 um
Normalized emittance
Linac with misaligned quads and BPMs.
Beam Jitter 1um NUMI Ground motion
--After GM only -- After AA (1hr step)
Emittance growth vs. time (1month scale) in
presence of ground motion (models A,B,C). Dynamic
Adaptive Alignment Tuning keeps emittance under
control.
conv Speed of convergence of algorithm Ai
BPM reading of the central quad and so on Ki
Inverse of quad focusing length L
Distance between successive quads DE
Energy gain between successive quads E
Beam Energy at central quad
BPM res0.2um
Future plans (EDR)
- Complete Static tuning Studies for realistic
curved Lattice - Complete Adaptive Alignment as Dynamic tuning
method for RDR lattice - Dynamic tuning Design FeedBack system
- Start-to-end simulation
Normalized emittance
Ground Motion models (Technical noise incl.) A
quiet (CERN-like) B moderate (SLAC-like) C
Noisy (DESY-like) FNAL between A and B
Normalized vertical projected emittance vs. time
in a dispersion-free steered linac. AA (100
iterations, gain0.3) is implemented after every
one hour of GM of model C.
Emittance growth for two BPM resolutions (GM
model B). Required resolution available after
averaging in train