WakeFest Summary 2

WakeFest Summary 2
ILC Wakefest Workshop, Dec 11-13, 2007
Subject
j
Areas
• Wakefield with cavity imperfection
• Beamline HOM absorber
• Large system simulation
• Multipacting simulation
A total of 9 talks
Modeling Imperfection Effects on Dipole
M d iin TESLA Cavity
Modes
C it
Liling Xiao
Advanced Computations Department,
SLAC
1. TESLA Cavity Dipole Mode Measurement Data
TTF m odule 5: 1s t/2nd dipole band
1.E+07
1st band 6th pair
1.E+06
1.E+05
1.E+06
1.E+04
1.E+05
1.E+03
1700
Qext
2nd band 6th pair
1701
1702
1703
1704
1705
1706
1.E+04
1877
1878
1879
1880
1881
1882
1.E+05
1.E+04
1.E+03
1600
• Omega3P
o Measured data in colour
1650
1700
1750
1800
1850
1900
F (M Hz)
Dipole mode frequencies shift and Qext scatter
2. TESLA Cavity Imperfection Model
x
z
a) Cell length
error
b) Cell radius
error
Red: ideal cavity, Blue: deformed cavity
c)) Cell
C surface
f
deformed
d) Elliptical cell
shape
1st ILC Workshop at KEK
A Matheisen
A.
Di t t d mesh
Distorted
h for
f the
th cavity
it imperfection
i
f ti model
d l
Example 2: Cell shape elliptical deformed (dr=0.25mm)
- cause mode p
polarization change
g and mode splitting
p
g
TDR cavity with elliptical cell shape
ideal cavity
1.E+07
1.E
07
1st band
6th pair
2nd band: 6th pair
defor cell1 along x
defor cell4 along x
defor cell1 and 4 along x
idea cavity
1.E+05
1.E+05
defor cell1 along x
defor cell4 along x
defor cell1 and 4 along x
idea cavity
Qext
1 E+06
1.E+06
1 E 04
1.E+04
1.E+04
1704.0
1705.0
1706.0
1707.0
F (MHz)
Qexxt
2nd band 6th pair
1708.0
1.E+03
1787.0 1787.5 1788.0 1788.5 1789.0 1789.5 1790.0
Elli. deformed cavity
1.E+05
1.E+04
• ideal cavity
O
1.E+03
1.60E+09
elliptically deformed cell shape cavity
1.65E+09
1.70E+09
1.75E+09 1.80E+09
1.85E+09
x
1.90E+09
F (Hz)
y
Beamline HOM Absorber
Talks:
• Beamline HOM Absorber – Oleg Nezhevenko,
FNAL
• B
Beamline
li HOM absorber
b b simulation
i l ti using
i T3P
– Liling Xiao, SLAC
O.N.
T.Higo, et al, IId ILC Workshop
8
Absorption efficiency estimations.
O.N.
Simple model of HOM absorber: the ring
has the length of 50mm, internal
diameter of 90 mm, and the thickness of
10mm (DESY style)
style).
9
RF Unit Test Facility
O.N.
Existing Building
72 M
New ILC like tunnel
~ 22 M
3rd har
Gun
CC I,II
Laser
ILC RF unit
Diagnostics
Bunch
Compressor
2nd ILC RF unit
Test Area
New Building
RF Equipment
•Overall Plan: Test ILC RF units
•3 CM, Klystron, Modulator, LLRF
•Move A0 Injector to provide ILC like beam
•New bldg: diagnostic, new cryo plant
•ILC Twin tunnel design
g to allow 2nd RF unit and to
study tunnel layout and maintenance issues
S. Nagaitsev
Test Areas
new 300 W
cryo plant
10
10
L.X.
• Simulation Results
1) 3D single
i l cavity
it with
ith beamline
b
li absorber
b
b (ε
( r=15,
15 σeff=0.6s/m)
06 / )
A Gaussian bunch with σz=10mm, Q=3.2nc on axis.
TM0n modes
Copper coated wall
3.5million mesh elements, 2nd basis function run on Franklin at Nersc.
512 processors 24000 time
ti
steps
t
within
ithi 12 h
hours
Energy stored in cavity
6.0E-05
Absorption
@6.6ns,
@6
6ns exciting bunch
just left the structure
5.5E-05
Left beampipe
E nergy (J )
Right beampipe
5.0E-05
Up stream HOM
Down stream HOM
4.5E-05
4.0E-05
3.5E-05
3.0E-05
0
100
200
30
t (ns)
400
50
Monopole Single Passage Losses
One bunch Q
Q=3.2nc,, bunch length=10mm
g
Loss factor (V/pc)=3.566V/pc
Lossyy dielectric conductivityy σeff=0.6(s/m)
( )
Dielectric constant εr=15, Within 45ns
Total Energy Generated by Beam (J)
3.65e-5
Energy stored in cavity (J)
3.25e-5
Energy leaked out HOM coupler ports (J)
4.05e-7
Energy propagated into beam pipe (J)
2.11e-6
Energy dissipated in the absorber (J)
2.4e-6
Energy loss on the copper absorber beampipe wall (J)
6.6e-10 (cold copper conductivity=350ms/m )
(FM mode energy=2.06e-5J)
Energy absorption/Total HOM energy
Energy leaked into Left beampipe/Total HOM energy
Energy leaked into right beampipe /Total HOM energy
K loss 6.86v/pc
K-loss=6
86v/pc for 10mm
10mm,
Total Energy=7.02e-5 (J)
absorber (εr=15, σeff=0.6s/m)
Large
g System
y
Simulation
Talks:
• Multi-cavity trapped mode simulation – Cho
N SLAC
Ng,
• Gl
Globalised
b li d scattering
tt i matrix
t i simulation
i l ti iin ILC
cavities and modules – Ian Shinton,
Manchester University
Uni ersit
• Large scale 3D wakefield simulations with
PBCI – Sascha Schnepp, T.U. Darmstadt
3rd Dipole-Band Trapped Modes in Cryomodule
C.N.
Cryomodule 3rd Dipole-Band Mode - Qext and Kick
TDR 8-Cavity Module
1.0E+14
Kick Factor (V/C
C/m/module)
1.0E+07
Qext
1.0E+06
1.0E+05
1.0E+04
1.0E+03
2.577
2.5775
2.578
F (GHz)
2.5785
K_X
K_Y
1 0E+13
1.0E+13
Kx[y]=Ky[x]_amp
1.0E+12
Ky[0]_amp
Kx[0]_amp
1.0E+11
1 0E+10
1.0E
10
1.0E+09
1.0E+08
1.0E+07
1.0E
07
1.0E+06
1.0E+05
2.5770 2.5772 2.5774 2.5776 2.5778 2.5780 2.5782 2.5784
F (GHz)
• Modes above cutoff frequency
q
y are coupled
p
throughout
g
8 cavities
• Modes are generally x/y-tilted & twisted due to 3D end-group geometry
• Both tilted and twisted modes cause x-y coupling
C.N.
Trapped Mode using Omega3P
Electric field
T
Trapped
d mode
d
• TM-like mode localized in beampipe
between 2 cavities
• Frequency = 2.948 GHz, slightly higher
than TM cutoff at 2.943 GHz
• R/Q = 0
0.392
392 Ω; Q = 6320
• Mode power = 0.6 mW (averaged)
TM mode
C.N.
S.S.
Introduction
There is an actual demand for:
1. Wake field simulations in arbitrary 3D-geometry
3D--codes
3D
2. Accurate numerical solutions for high frequency fields
(quasi--) dispersionless codes
(quasi
3 Utilizing
3.
Utili i llarge computational
t ti
l resources ffor ultra-short
lt
h t bunches
b
h
parallelized codes
4. Specialized algorithms for long accelerator structures
moving window codes
S.S.
ILC-ESA collimator
ILC-ESA collimator #8
Convergence vs. grid step
60
40
σ / Δz = 2
2.5
5
σ / Δz = 5
σ / Δz = 10
σ / Δz = 15
bunch size
300μm
no. of g
grid points
p
~450M
no. of processors
24
simulation time
85hrs
Moving window:
3 mm length
g
Wz / [V
V / pC]
20
0
-20
20
-40
-60
60
-80
-100
-4
-3
-2
-1
0
s/σ
1
2
3
4
S.S.
TESLA / HOM coupler
TESLA 9-cell cavity
bunch
-15
-10
-5
0
Ez / [kV /m]
5
15
PBCI
ECHO2D
5
bunch length
10
1mm
bunch charge
1nC
cavity length
1.5m
Wz / (V
V / pC)
0
-5
no. of grid points
~760M
-10
no. off processor
cores
408
-15
simulation time
-5
~40hrs
-4
-3
-2
-1
0
s/σ
1
2
3
4
5
Multipacting
p
g Simulation
Talks:
• Multipacting simulation for the SNS cavity
using
i A
Analyst
l t – I.
I Gonin,
G i FNAL
• M
Multipacting
lti
ti simulation
i l ti using
i parallel
ll l code
d
Track3P – Lixin Ge, SLAC
• TTF-III coupler processing and multipacting
simulation using MAGIC – Faya Wang
Wang, SLAC
I.G & L.G.
MP IN SNS CAVITY
Analyst
Enhanced Counter Function (20 impacts)
Track3P
1,E+05
1,E+04
1069
1,E+03
1,E+02
0
5
10
15
20
25
30
35
Eacc(MV/m)
Models simulated using
Analyst and Track3P are
different!
MP in ILC HOM coupler: notch gap
HOM coupler
Emax ~ 48.1
48 1 V/m
Emax ~ 3.8 V/m
I.G.
Enhanced Counter Function (20 impacts)
2500
1-D
2000
1500
1000
500
1.73mm
Egap ~ 3.8 V/m;
kV/m
0
0
50
100
150
200
250
300
350
Enhanced Counter Function
for gap 1.73mm and 1.3GHz
α = Egap / Eacc ~ 0.16
Z94_02_Aug24_06
EMP_analytical ~ 200 kV/m
Eacc= EMP / α ~ 1.25 MV/m
Enhanced Counter Function (20 impacts)
1.E+04
1.E+03
2.0E+10
3-D
1.E+02
1.8E+10
1.E+01
1 6E+10
1.6E+10
1.E+00
MP in the gap
1.4E+10
1.2E+10
FE mesh, ~1 million
second
d order
d elements
l
t
1.E-01
1.E-02
Test: Q vs. Eacc
1.0E+10
0.0E+00 5.0E+06 1.0E+07 1.5E+07 2.0E+07 2.5E+07 3.0E+07
Eacc (MV/m)
1.E-03
0
0.5
1
1.5
2
2.5
Analyst results
Good correlation between 1D, 3D&measurements
3
L.G.
Cold Coax in TTF-III Coupler
6.25mm
300mm
20mm
Delta as a function of RF input power and Multipacting order
F.W.
Ion Pump
Ion Pump
PM
e-pickup
PM
F.W.
TTF-III coupler processing
600mm long straight SS
coax section test results
600mm long straight Copper
plated SS coax section test results
150
-50
electron signal amplitude
100
-100
50
0
0
Strong MP around 300kW!
vacuum level
200
400
600
800
1000
Forward power: kW
1200
1400
Electron signal:mV
Current of Ion Pume: uA
C
0
0
0 01
0.01
-20
0.02
Electron ssignal: -V
elecctron signal amplitude from e-pickkup: mV
F.W.
RF processing result of ramping RF from 20us pulse to 400us
0
-40
-60
20us
50us
100us
200us
400us
-80
-100
0
200
400
600
800
1000
Forward RF power to coax: kW
0.03
0 04
0.04
0.05
20us RF pulse
50us
100us
200us
400us
0.06
0 07
0.07
1200
0.08
0
7
seymax=1.5
6
seymax=1.3
5
4
3
2
1
0
100
200
300
400
RF power: kW
Magic Simulation
500
1400
Copper plated S/S Coax Tube
Relative Am
mplitude of Electron
n Signal: arb.u.
Space Cha
arge at Satura
ation: arb.u.
Stainless Steel Coax Tube
200
400
600
800
1000 1200
Power forward into straight coax section: kW
600
1
08
0.8
0.6
0.4
0.2
0
200
400
600
RF Input Power (kW)
800
1000
Track3P simulation (indicated by dashed
lines) is independent of SEY.