ILC Damping Rings Mini-Workshop, KEK, Dec 18-20, 2007 Status and Plans for Impedance Calculations of the ILC Damping Rings Cho Ng Advanced Computations.
Post on 18-Jan-2016
213 Views
Preview:
Transcript
ILC Damping Rings Mini-Workshop, KEK, Dec 18-20, 2007
Status and Plans for Impedance Calculations of the ILC Damping Rings
Cho NgAdvanced Computations Department
Stanford Linear Accelerator Center
* Work supported by US DOE ASCR & HEP Divisions under contract DE-AC02-76SF00515
Outline
Damping Ring Vacuum Chamber Impedance
SLAC Parallel Modeling Suite
Simulation Status
Schedule & Plans to Facilitate Collaboration
Preliminary List of Vacuum Chamber Components
(Marco Venturini, LBNL)
Damping Ring Impedance Calculations
• Broadband impedance
- Identify major components that contribute to the impedance budget - Calculate short-range wakefields for single-bunch stability studies
• Narrowband impedance
- Identify trapped modes in cavity-type structures - Provide HOM parameters for coupled-bunch stability studies
Beam Heating & Engineering Analysis
• Beam Heating
- Identify sources of HOM heating - Investigate damping schemes to mitigate HOM effects
• Engineering Prototyping
- Contribute to integrated analysis including electromagnetic, thermal and structural effects - Include transfer impedances of pickup devices
SLAC Parallel Modeling Suite
Supported by US DOE SciDAC program, SLAC Parallel Finite Element codes can simulate large problems to high accuracy with near linear speedup using petascale computers at NERSC and NCCS. They include: Omega3P – nonlinear eigensolver to find resonant modes in damped RF cavities
T3P – time-domain solver to calculate transients due to external drive and wakefields generated by beam transit (implementation of indirect wakefield integration)
TEM3P – multi-physics analysis tool to simulate integrated electromagnetic, thermal and mechanical effects Resources: Zenghai Li, Cho Ng
Superconducting RF Cavity
• Cornell Model – 500 MHz
• KEK Model – 508 MHz
r= 92 mm r= 25 mm
f0= 650 MHz
ILC DR cavity scaled from Cornell model
(Sergry Belomestnykh, Cornell)
DR Cavity: sigma_z=6mm
-1.0-0.8-0.6-0.4-0.20.00.20.40.60.81.0
0.0 0.1 0.2 0.3 0.4 0.5s (m)
Wa
ke_
L
L Wake
Charge
Loss Factor = 1.455 V/pC
ABCI calculation
Damping Ring Cavity
= 6 mmDR Cavity (scaled Cornell): sigma_z=0.5mm
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07s (m)
W_
L,
Q
Long. Wake
Charge
Loss Factor = 16.17 V/pC
= 0.5 mm
• Need 20 points per sigma for convergence
• Used as pseudo-Green’s functionFurther studies
• Narrowband impedance and damping• Effectiveness of beampipe absorber
Damping Ring BPM
10 mm button
25 mm radius
Snapshots of beam transit from T3P simulation
Scaled model from PEP-II 15 mm button
BPM Longitudinal Wakefield
Loss Factor = 0.0015 V/pC
• Effects of trapped modes at the buttons need to be studied for coupled bunch instability and beam heating
= 6 mm = 1 mm
BPM Transfer Impedance• Field monitored at coaxial port as a function of time• Transfer impedance obtained by Fourier transform• Signal sensitivities in x- and y- directions determined by simulations
with offset beam excitations
TE Mode Propagation
• TE HOM power propagating in vacuum chamber can couple to BPM, and thus affecting processing signal
• Ante-chamber lowers the TE mode cutoff frequency
TE cutoff at 2.929 GHz TE cutoff at 2.389 GHz
Omega3P Calculation
(Andy Wolski, Cockcroft)
Damping Ring Bellows
Scaled model from PEP-II bellows(Preliminary)
• Loss Factor = 0.0168 V/pC • Dominated by step used to shield the bellows
= 6 mm
Trapped Modes in Bellows
19.95 GHz6.202 GHz 8.724 GHz
Examples of trapped modes from Omega3P calculation
• Trapped modes in bellows convolution are potential sources of excessive heating
• Excited by HOM power propagating in vacuum chamber
Impedance Budget
Impedance budget
Component Quantity Loss factor (V/pC)
RF Cavity 18 26.19
BPM 682 1.02
Resistive wall 12.08
Total 39.29breakdown
c.f.• PEP-II HER – 2.5 V/pC for 1 cm bunch length
• NLC DR – 7.67 V/pC for 4 mm bunch length
total
Schedule of Simulation Effort• Year 1
- Impedance modeling using scaled models - Determine longitudinal and transverse wakefields for single- bunch stability studies - Determine HOMs in rf cavity for coupled-bunch stability studies
• Year 2- Repeat calculations of broadband impedance using improved
models of technical designs - Investigate effectiveness of absorbers in damping HOMs in rf cavity
• Year 3 - Integrated analysis including rf, thermal and mechanical effects of ring components for optimized technical designs - Finalize impedance calculations using models of engineering prototypes
Multi-Physics Analysis for Prototyping
• Virtual prototyping on computers from CAD model
• Integrated EM, thermal and mechanical effects
• Augmented by additional physics - particle effects (emittance, multipacting) - transient and non-linear effects in superconducting rf cavities
CAD model of LCLS RF gun Electromagnetic
Thermal
Mechanical
TEM3P
Work Packages
WP5: Impedance Computation at ANL• Resources
– Xiaowei Dong (0.25 FTE), Yong-Chul Chae (0.1 FTE) – Linux cluster with 120 cores (4 core/node * 30 nodes) and 480 GB of total
memory– Parallelized 3D EM code GdfidL
• Experience in Computing Wake Potentials– Regular APS storage ring with bunch lengths z= 1, 2, 5 mm
• 8.4 cm x 4.2 cm– Reduced APS storage ring chamber with bunch length z= 5 mm
• 4.0 cm x 2.0 cm• All chamber components scaled by a factor of two in transverse
dimension without new design
• Deliverables– Assuming the APS components in the DR, we will deliver the total wake
potential of z=1 or 2 mm of the ring in the first year by July, 2008– Refine and update as the mechanical design changes
Courtesy of Yong-Chul Chae
Plans to Facilitate Collaboration
• Availability of models of vacuum chamber components from existing machines
• Standardized CAD format to facilitate information exchange among physicists and engineers
• Coordination of impedance calculations among different institutions
• Database to store CAD models and computational results
top related