DUAL FEED RF GUN DESIGN FOR LCLS Liling XIAO, Zenghai LI Advanced Computations Department Stanford Linear Accelerator Center Nov.3 2004, SLAC-LCLS Injector

DUAL FEED RF GUN DESIGN FOR LCLS

Liling XIAO, Zenghai LI

Advanced Computations Department

Stanford Linear Accelerator Center

Nov.3 2004, SLAC-LCLS Injector RF System Internal Review

L. Xiao and Z. Li LCLS Injector RF System Review

2

Outline

1. Introduction

2. Design via Simulation

3. Pulsed Heating

4. Quadruple mode

5. Dual Feed Gun Parameters

6. Summary


3

1. Introduction

• Requirements and Parameters

- 2.856GHz operating frequency mode field imbalance within 10% Coupling factor around 2 T < 50oC due to pulsed heating

for Ecathode=120MeV, t-pulse = 3s with 120Hz pulse repetition rate


4

D.T.Palmer, et al., “Microwave Measurements of the BNL/SLAC/UCLA 1.6 cell Photocathode RF Gun “

SLAC-PUB-95-6799

• Previous Study

R.Boyce, et al., “Design Considerations for the LCLS RF Gun”,

LCLS TN 04-4, April 2004

J.Hodgson


5

• New Design Features

Dual RF feeds

- to eliminate the dipole modes

Larger rounding of coupling iris

- to reduce the pulsed heating

Racetrack cell shape

- to minimize the quadruple modes


6

A A

BB

Results: f=2.8625GHz, β=2.28 (E0:E1=1:0.908), T=150oC for the single feed gun dimensions. (comparing the data on Slide 15)

Laser Port

RF Coupling Aperture

2. Design via SimulationOriginal Design


7

2b0 2b1lchc

r2

New Optimized Design- Adjust labeled dimensions by modeling with ACD’s parallel eigensolver Omega3P and parallel S-parameter solver S3P


8

S12

GHz

mode

Model is generated and meshed using Cubit.

Typical transmission curves calculated with S3P (solid curve) and fitted using Lorentzian profiles (dotted curve)

• Modeling with S3P


9

b0, b1,

lc, hc(inch)

1.6335,1.6435, 0.868, 0.056

(starting point)

1.6370, 1.6470, 0.868, 0.056

1.6370, 1.6482, 0.82, 0.056

1.6370, 1.6475, 0.868, 0.09

F0 (GHz) 2.86247 2.85605 2.85622 2.85617

Q0 11384 13214 13337 13307

beta 2.28 2.45 1.93 1.93

E0/E1 1:0.908 1:0.908 1:0.985 1:0.96

Iteration to optimized dimensions(Pulsed heating and quadruple correction not considered)


10

3. Pulsed Heating

Previous Results from ANSYS on Temperature Distribution in the LCLS RF Gun (LCLS TN 04-4):


11

Breakdowns due to the pulsed heating had been observed in the Fermilab RF Gun G4

J.P.Carneiro, et al., TESLA Note 2003-13


12

• The temperature rise at the end of RF pulse is

heatspecificmD

tyconductivithermalCmWK

R

tt

dttH

D

K

RT

s

ss

t

p

ss p

sec/10*132.1

//360

2

1

2

)(2

1

24

0

0

2

maxmax

Hsmax(t)

Maximum surface magnetic field located on the coupling aperture edge.

D.T.Pritzkau, “RF Pulsed Heating”, SLAC-Report-577


13

40

60

80

100

120

140

160

0 0.05 0.1 0.15 0.2 0.25r2 (inch)

dT d

egre

e C

Temperature Rise vs. Rounding Radius of RF Aperture

Assuming: Ecathode =120MV/m, f = 2.856GHz, β = 2,Q0 =13300, tp = 3s

Start point

Final point


14

Starting point: hc=0.056”, r2=0.022” - Tmax=150oC

Final point: hc=0.2165”, r2=0.185”- Tmax=42oC


15

hc=0.2165 inch

Coupling Factor vs. Length of Aperture

1.8

2

2.2

2.4

2.6

2.8

3

0.81 0.82 0.83 0.84 0.85 0.86 0.87 0.88 0.89

lc (inch)

beta

hc=0.056 inch

0.5

1

1.5

2

2.5

3

0.86 0.88 0.9 0.92 0.94 0.96 0.98 1 1.02

lc (inch)

beta

before racetrack

after racetrack

Field imbalance for all the points are within 3%.

Final pointStart point

hc=0.2165 inch

4. Quadruple Mode in Coupler Cell


16

βr/

mm

Quadruple Moment in Cylindrical Cavity

-0.008

-0.006

-0.004

-0.002

0.000

0.002

0.004

0.006

0.008

-180 -120 -60 0 60 120 180

rf phase (degree)

cylindrical (lc=0.9744")

cylindrical (lc=0.868")

4th order finite element basis functions in Omega3P have to be used to compute the quadruple fields accurately in the models. And the fields from Omega3P are being used

in PARMELA emittance calculations.


17

βr/

mm

b1 b1d

-0.008

-0.006

-0.004

-0.002

0.000

0.002

0.004

0.006

0.008

-180 -120 -60 0 60 120 180

rf phase

cylindrical cavity

racetrack cavity withd=0.124"racetrack cavity withd=0.14"racetrack cavity withd=0.134"

Quadruple Moment in Racetrack Cavity


18

Effect of Laser Ports

1: max T=36oC

βr/

mm

2: Quadruple moment due to the laser ports is about 5% of that in the cylindrical cavity. -0.008

-0.006

-0.004

-0.002

0.000

0.002

0.004

0.006

0.008

-180 -120 -60 0 60 120 180

rf phase

cylindrical coupler cell

half cell with laser port


19

βr/

mm

-0.006

-0.004

-0.002

0.000

0.002

0.004

0.006

-180 -120 -60 0 60 120 180

rf phase

quad in the cylindrical cavity

quad in the racetrack cavitywith d=0.134"

skew quad in the half cell withlaser ports

The racetrack corrects only the quadruple in the coupler cell.

The skew quadruple effect in the half cell due to laser ports needs to be evaluated; racetrack shape can be added to the half cell if necessary.

βr/

mm


20

5. Dual Feed “Corrected” RF Gun

b1 b1

d


21

Parameter Value

Race track arc radius b1 1.5999 inch (original:1.6435”)

Race track arc separation d 0.134 inch (original: 0)

Race track cell length l1 1.28 inch

Half cell radius b0 1.6361 inch (original:1.6335”, not consider the laser ports)

Half cell length l0 0.896 inch

RF coupling hole size (slot length) lc 0.95 inch (original:0.868”)

RF coupling hole radius of curvature rc 0.1875 inch

RF coupling hole size (slot width) 2rc 0.375 inch

RF coupling hole thickness hc 0.2165 inch (original: 0.056”)

RF coupling hole rounding radius r1 on up side 0.022 inch

RF coupling hole rounding radius r2 on down side 0.185 inch (original: 0.022”)

Cell iris radius a 0.492 inch

Disk thickness t 0.868 inch

Disk rounding radius r 0.375 inch

Laser port hole size (slot length) ll 0.433 inch

Laser port hole radius of curvature rl 0.125 inch

Laser port hole size (slot width) 2rl 0.250 inch

Laser port hole rounding radius r3 0.030 inch

Laser port offset the cathode plate 0.531 inch

Laser port angle 18 degree

Waveguide 2.840 inch*1.340inch

RF-GUN Final Dimensions


22

“Corrected” RF GUN Parameters:

RF Parameters

F0 (GHz) 2.856022

Q0 13325

β 2.07

Mode Sep. f (MHz) 3.4

T max (oC) 44

E0:E1 1:0.962


23

On-axis Accelerating Field Profile in RF Gun Cavity

-1.0

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

0 50 100 150 200 250 300

mesh in z-axis

norm

aliz

ed E

z al

ong

z-ax

is

E0:E1=1:0.962


24

Dimension Freq. error per 0.0001”

Field error per 0.0001”

Full cell radius b1 113KHz 5.1%

Half cell radius b0 70KHz 5.2%

Racetrack arc separation d 45KHz 2%

Cell iris radius a 40KHz 0.3%

Disk rounding radius r 24KHz 0.2%

Frequency and Field Sensitivity to Major Gun Dimensions


25

The original RF gun design has been optimized to:

- eliminate the dipole modes

- reduce pulsed heating

- minimize the quadruple moment

A “corrected” design has been generated using SLAC’s parallel codes Omega3P & S3P with 4th order elements

The steady-state thermal properties needs to be studied again for the new dual feed gun design.

6. Summary

Documents

DUAL FEED RF GUN DESIGN FOR LCLS Liling XIAO, Zenghai LI Advanced Computations Department Stanford Linear Accelerator Center Nov.3 2004, SLAC-LCLS Injector