TESLA DAMPING RING RF DEFLECTORS DESIGN

1

• TESLA DAMPING RINGRF DEFLECTORS DESIGN
• F.Marcellini D. Alesini

2

• SUMMARY
• THE CTF3 RF TW DEFLECTOR
• FROM DESIGN TO OPERATION WITH THE BEAM
• COMPARISON OF DEFLECTION EFFICIENCY OF p/2, 2p/3
  AND 4p/5 MODES, SCALING RESULTS OBTAINED BY
  LENGELER FOR 2.855GHz STRUCTURES
• HFSS AND MAFIA SIMULATIONS OF THE SINGLE CELL
  DEVELOPED FOR BOTH p/2 AND p/3 MODES
• CHARACTERISTIC PARAMETERS AND DISPERSION CURVES
  OBTAINED.

3
OUR EXPERIENCE WITH RF DEFLECTOR FOR CTF3
2. MECHANICAL DRAWING
1. STUDY AND NUMERICAL SIMULATIONS
4. MEASUREMENTS
3. CONSTRUCTION
4
5. RESULTS OF RECOMBINATION TESTS
FOUR STEPS BUNCH TRAIN RECOMBINATION
ENLARGING THE IRIS APERTURE, CURRENT INCREASES
FROM FIRST TO LAST TURN OF RECOMBINATION WITH
NEGLIGIBLE LOSSES.
5
In this paper, for the first time, TW deflecting
structures were studied. Three different modes
(p/2, 2p/3, 4p/5) tuned at the same RF frequency
(2.855GHz) were completely characterized as a
function of the cell dimensions.
Typical dispersion curve of a periodically loaded
structure. The three modes considered by the
Lengeler analysis are pointed out. The p/2 mode
presents the higher group velocity vg (higher
tangent slope).
6
As a first step we have done an analogous
analysis by scaling the Lengeler results with
frequency (1.3 GHz in our case). For each
considered mode we have fixed the energy of the
beam, the angle of deflection and the RF power
feeding the structure. Two different values have
been supposed (9 MW is the max output power of
the klystrons developed for TESLA). Therefore
the length of the structure (L) is directly
linked to its efficiency (shunt impedance per
unit length). L, together with the group velocity
(vg), determines the filling time according to
the formula The power dissipated along the
deflector due to resistive losses are also
evaluated. In the following plots, the behaviour
of each of these quantities as a function of the
iris radius (a), are plotted. The same
evaluations have been repeated, for each of the
three modes, varying the thickness (t) of the
iris. However only results for t11.53 mm have
been shown.
t/2
7
p/2 MODE Deflection 0.5 mrad fRF 1.3
GHz Disk thickness 11.53 mm Cell length 57.65
mm
8
2p/3 MODE Deflection 0.5 mrad fRF 1.3
GHz Disk thickness 11.53 mm Cell length 76.9
mm
9
4p/5 MODE Deflection 0.5 mrad fRF 1.3
GHz Disk thickness 11.53 mm Cell length 92.2
mm
10
Reduction of the effective kick when the
deflector is fed by a 3 MHz detuned excitation.
11
INPUT GEOMETRY FOR SIMULATIONS
E
H
E.M. FIELDS ON THE AXIS OF THE SIMULATED
STRUCTURE (HFSS)
12
p/2 MODE PARAMETERS FROM SIMULATION RESULTS

Equivalent deflecting voltage P
RF power pd rms dissipated power
per unit length w rms stored energy
per unit length
13
p/3 MODE PARAMETERS FROM SIMULATION RESULTS

Equivalent deflecting voltage P
RF power pd rms dissipated power
per unit length w rms stored energy
per unit length
14
Dispersion curves calculated by MAFIA

2D simulations have been performed to evaluate
the dispersion curve of both the considered
modes. Their slopes at 1.3 GHz indicate that the
group velocity is close to its maximum reachable
value.
15
Magnitude of E field Fields distribution in the
volume of the cell has also been evaluated from
simulations. In particular, the peak values for
the electric field are localized in
correspondence of the irises, as it is shown in
the plot. The resulting values are listed below
for input power of both 9 MW (single frequency
input mode) and 27 MW (triple frequency input
mode). p/2 Epeak 10 MV/m _at_ PRF 27 MW
5.7 MV/m _at_ PRF 9 MW p/3 Epeak 9.3 MV/m _at_
PRF 27 MW 5.4 MV/m _at_ PRF 9 MW