Drive Beam generation with collector ring

1
Progress in the design of a damped an tapered
accelerating structure for CLIC
Jean-Yves Raguin CERN AB/RF
2
The Tapered Damped Structure

• 2p/3 travelling-wave mode, 150 cells Av. Acc.
  Grad. of 150 MV/m
• Strong damping
• Each cell coupled to four rectangular identical
  waveguides
• Cutoff frequency of the fundamental waveguide
  mode between the p-mode frequency of the
  fundamental passband and the lowest frequency of
  the first dipole band
• Trapped fundamental mode whereas the E-M energy
  of the higher modes propagates out of the cells
  and is absorbed by SiC loads terminating the
  waveguides
• Light detuning
• Linear variation of the irises radius from 2.25
  mm at the head of the structure to 1.75 mm at the
  end, each cell being tuned to have the same
  fundamental frequencies
• Dipole frequency spread (5.4 ) which contributes
  to a further reduction of the transverse
  wakefields
• Demonstrated suppression of the transverse
  wakefields by two orders of magnitude within 0.67
  ns (time between two consecutive bunches) on a 15
  GHz-scaled version (ASSET experiment)

3

Accelerating mode electric field magnitude (Log)
SiC load
Dipole mode Poynting vector
4
Performance limits and peak surface electric
fields
L 50 cm Pin/section 248 MW h 23.8
5
Pulsed surface heating
Temperature increase for the middle cell of the
TDS with 150 MV/m and tp130ns
Due to the configuration of the cell-waveguide
coupling iris (3.3 mm wide) the local magnetic
field is too high, leading to excessive maximum
temperature rise
6
Design of a new 2p /3 damped and tapered structure
Solutions
Criteria to design new CLIC damped accelerating
structure

• To prevent electric breakdown, low peak surface
  electric field
• To prevent excessive pulse heating, low peak
  surface magnetic field
• Good damping of higher order modes and, in
  particular, of the first and second dipole modes

• Use of an elliptical profile for irises
• Appropriate profile of the cells wall
• Damping of higher order modes by coupling each
  cell to four T-cross waveguides

7
(No Transcript)
8
Topology of the cell and of the damping waveguides
9
Field pattern of the first two waveguide modes
Fundamental mode Damping for the first dipole
band
Second mode Damping for the second dipole band
Electric field
Electric field
Perfect Elec. wall
Perfect Mag. wall
Magnetic field
Magnetic field
10
First cell Iris thickness 0.55 mm
3.792 mm
2.0 mm
Q3744 with s 5.80.107 Mhos/m (Cu) R/Q
24.5 kW /m vg/c 8.1 fcutoff,TE10-e 32.1
GHz Epeak / Eacc 2.55 Hpeak / Eacc 4.50 mA/V
3.0 mm
5.25 mm
11
Hsurf/Eacc (mA/V) on the walls of the first cell
12
Last cell Iris thickness 1.00 mm
1.5 mm
3.661 mm
Q3373 with s 5.80.107 Mhos/m (Cu) R/Q
30.0 kW /m vg/c 2.6 fcutoff,TE10-e 32.4
GHz Epeak / Eacc 1.75 Hpeak / Eacc 4.38 mA/V
3.0 mm
5.25 mm
13
(No Transcript)
14
(No Transcript)
15
For 84 cells (L 28 cm) Epeak400
MV/m Pin/section 130 MW h 26.1
DT154 K
DT120 K
16
Transverse wakefields analysis
First cell
Middle cell
Last cell
Real part of the transverse impedance vs.
frequency computed with GdfidL for the first,
middle and last cells
Transverse wakefields
W t 90 V/pC/mm/m at the 2nd bunch
Q dip,first 52 Q dip,middle 51 Q
dip,last 44 Dfdip /f dip,first 3.3
17
THERE IS ROOM FOR IMPROVEMENT
18
Design of a damped structure - b d 110 deg.
Working at a lower phase advance allows to
decrease Epeak / Eacc Decrease iris thickness
along the structure Increase the coupling
cell- waveguide along the structure for better
damping
19
First cell Iris thickness 0.80 mm
2.0 mm
3.870 mm
Q3387 with s 5.80.107 Mhos/m (Cu) R/Q
24.1 kW /m vg/c 7.7 fcutoff,TE10-e 32.3
GHz Epeak / Eacc 2.21 Hpeak / Eacc 4.50 mA/V
3.0 mm
5.25 mm
20
Last cell Iris thickness 0.55 mm
1.5 mm
3.539 mm
Q3365 with s 5.80.107 Mhos/m (Cu) R/Q
33.2 kW /m vg/c 3.8 fcutoff,TE10-e 32.3
GHz Epeak / Eacc 2.00 Hpeak / Eacc 4.17 mA/V
3.2 mm
5.32 mm
21
(No Transcript)
22
(No Transcript)
23
For 83 cells (L 25.4 cm) Epeak355
MV/m Pin/section 128 MW h 24.8
DT122 K
DT122 K
24
Transverse wakefields analysis
First cell
Middle cell
Last cell
Real part of the transverse impedance vs.
frequency computed with GdfidL for the first,
middle and last cells
Transverse wakefields
W t 58 V/pC/mm/m at the 2nd bunch
Q dip,first 53 Q dip,middle 31 Q
dip,last 24 Dfdip /f dip,first 4.9
25
There is room for improvement (2)
26
Design of a damped structure - b d 110 deg. (2)
First cell Iris thickness 0.80 mm
2.0 mm
3.867 mm
Q3266 with s 5.51.107 Mhos/m (Cu) R/Q
24.0 kW /m vg/c 7.6 fcutoff,TE10-e 32.3
GHz Epeak / Eacc 2.20 Hpeak / Eacc 4.51 mA/V
3.0 mm
5.25 mm
27
Last cell Iris thickness 0.55 mm
1.5 mm
3.532 mm
Q3252 with s 5.51.107 Mhos/m (Cu) R/Q
32.5 kW /m vg/c 3.8 fcutoff,TE10-e 32.3
GHz Epeak / Eacc 1.95 Hpeak / Eacc 4.10 mA/V
3.2 mm
5.32 mm
28
(No Transcript)
29
(No Transcript)
30
For 83 cells (L 25.4 cm) Epeak355
MV/m Pin/section 130 MW h 24.4
DT119 K
DT128 K
31
For 77 cells (L 23.5 cm) Epeak348
MV/m Pin/section 125 MW h 23.8
DT121 K
DT122 K
32
Transverse wakefields analysis 83-cell
structure
First cell
Middle cell
Last cell
Real part of the transverse impedance vs.
frequency computed with GdfidL for the first,
middle and last cells
Transverse wakefields
W t 45 V/pC/mm/m at the 2nd bunch
Q dip,first 43 Q dip,middle 27 Q
dip,last 21 Df dip/f dip,first 5.4
33
Shall we dare to lower the accelerating gradient?
34
Average accelerating gradient of 125 MV/m
For 75 cells (L 22.9 cm) Epeaklt 300
MV/m Pin/section 89 MW h 27.2
DT88 K
DT87 K
with the same beam current
35
Conclusions

• Design of copper structure with average
  accelerating gradient of 150 MV/m, peak surface
  electric field lower that 300 MV/m and maximum
  temperature rise lower than 100 K seems
  unrealistic would lead to smaller iris radius
• Copper structure with average accelerating
  gradient of 125 MV/m, peak surface electric field
  lower that 300 MV/m and maximum temperature rise
  lower than 100 K seems feasible
• Investigation of new materials
• Design of the latest structure with molybdenum
  irises calls for a reassessment of the
  fundamental mode characteristics
• Material for the cell outer walls which would
  solve the pulsed surface heating problem?
• Transverse wakefields suppression
• Need for tuning the dipole frequencies