Measurements and Simulations of CSR Effects in the APS

ADVANCED PHOTON SOURCE Operations Analysis Grouphttp://www.aps.anl.gov/asd/oag Michael Borland [email protected]

tions

r

May 31, 2001

Measurements and Simulaof CSR Effects in the

APS Bunch Compresso

Michael Borland

Argonne National Laboratory


, J. Lewellen,

ma,

. Soliday,

Contributors

• Simulation: M. Borland, J. Lewellen

• Design: V. Bharadwaj, M. Borland, P. EmmaS. Milton, K. Thompson

• Diagnostics: G. Decker, B. Lill, B. Yang

• Experiments and support: M. Borland, P. EmJ. Lewellen, Y. Li, S. Milton

• Software: N. Arnold, M. Borland, Y. Chae, RJ. Stein

• Management: S. Milton


Outline

• APS linac and bunch compressor overview

• Beam requirements and limitations

• CSR simulation methods

• Design simulation results

• Comparison of experiments and simulation

• Plans


D)

toFEL

ance measurement

tance measurement

CTRdiagnostic

APS Linac Schematic

TC RF guns

L1L2 (SLEα magnets

PC gun

L3

L4 (SLED)

emitt

L5 (SLED)

B1

B2 B3

B4

Q1 Q2emitmatching quads

BPMvertical bend diagnosticscraper flag


EL

0A wasas seen.

s achieved withollowed shortly.

Charge

nC

>0.5

>0.5

Beam Specifications for LEUTL F

• Prior to bunch compressor installation, ~15achieved with ~0.7nC. Modest FEL gain w

• With the bunch compressor, up to 600A wa~0.2nC. Saturation at 530nm and 385nm f

Energy PeakCurrent

RMSEnergySpread

NormalizedEmittance

MeV A % um

217 300 <0.1% 5

457 600 <0.15% 5


ll

ressor

ωrf ∆to

E--------------------

δ

∆t

∆to2⟨ ⟩ 1

R56VωcE

-----------------+

2

Bunch Compression in a Nutshe

four-dipolechicane

precomplinac

δ δo

Vrf---------+=∆t ∆to δR56

c---------+=

∆t ∆toR56

c--------- δo

Vω∆toE

-----------------+ += ∆t2⟨ ⟩ ≈


ffects.

icane.

to model.

Limits on Injector Performance

• Space charge and rf focusing in the gun.

• Wakefields in accelerating tanks.

• Jitter of rf systems and laser.

• Rf curvature and other nonlinear transport e

• Coherent synchrotron radiation (CSR) in ch

With the exception of CSR, these are “easy”


ctory allowsm tail to catchhead.

bunch radiatet wavelengths

han the bunch

n produces aendent energy

along the

Qualitative Explanation of CSR

retardedposition

presentposition

(t=-s/c) (t=0)

s

Curved trajeradiation froup with the

Particles incoherently amuch less tlength.

This radiatioposition-depmodulationbunch.


dent energyin the energy

and inside thees of particle

ittance in the

e can be seenr the chicane.

How CSR Affects the Bunch

• CSR imposes a longitudinal-position-depenmodulation on the bunch. This will show upspectrum.

• This modulation is imparted inside a dipolechicane, producing a modulation of the sloptrajectories.

• This results in a growth of the projected embending plane.

• CSR also introduces x-p correlations. Theson a vertical bend (“Dowell diagnostic”) afte


m of Saldin, et

into the bend,

Simulation of CSR Effects

• Inside dipoles, use free-space, 1-D formalisal., in NIM A 398 (1997):

where R is the bend radius, is the angleand . The two terms are

and

dE s R φ, ,( )cdt

--------------------------- T1 s R φ, ,( ) T2 s R φ, ,( )+=

φs ct=

T1 s R φ, ,( ) K λ′ z( ) s---z

---13

------zd

s---sl

s

∫=

T2 s R φ, ,( ) Kλ s---sl( )---λ s---4sl( )

sl1 3/

------------------------------------------=


ngth, and is

R wake ise end of eachach simulation

λ(s)
where , is the slippage le
the longitudinal density of the bunch.

• Dipoles are cut into ~100 slices and the CScomputed from the longitudinal density at thslice. This is used to modify the energy of eparticle.

K---2e2

3R2( )1 3/------------------------= sl

Rφ3

24----------=


ole

ssian beam= 1.5m

50µm=

Examples of CSR Wakes in a Dip

GauR

σz


s

the radiation

ke propagatesis is confirmed

vertaking

for an idealizedhese are useduates.

Simulation of CSR in Drift Space

• CSR effects are not confined to dipoles, ascontinues to propagate with the beam.

• After dipoles, assume the terminal CSR wawith gradual attenuation but fixed shape. Thby detailed simulations (Dohlus et al.).

• Attenuation length is roughly given by the “o

length,” .

• Saldin et al. give equations for this radiationrectangular beam distribution. In elegant , tto determine how quickly the radiation atten

24σzR2( )1 3/


Attenuation of CSR in Drift


ssor

tribution to

Simulation Method for APS Compre

• Photoinjector (J. Lewellen) uses PARMELA

• Simulate a nominal 1nC bunch.

• Scale the charge and emittance of this dismatch what we actually run.

• Linac-to-FEL uses elegant , including

• longitudinal and transverse matching

• longitudinal wakes

• exact rf curvature effects

• second-order matrix for quadrupoles

• symplectic integrator for chicane dipoles

• CSR in dipoles and drifts


R Effect
Example of Bunch Compression and CS


itions
Predictions for Design Operation Cond


ber problem),d rf phase.

irp and thus the

ergy fixed.

chicane)

nd line)

tem)

g)

Bunch Compressor Experiments

• Bunch compressor is fixed in position (chamso we basically have two knobs: charge an

• Rf phase scans allow varying the energy chbunch length.

• We vary the rf voltage to keep the beam en

• Measurements include:

• “incoming” energy spectrum (at center of

• energy spectrum after chicane (vertical be

• emittance after chicane (three screen sys

• bunch length after chicane (rf zero-phasin


1/01)
Phase Scan for 200pC Beam (3/3


ion
Comparison with “Nominal” Simulat


nt

pace enteringemittance

asured bunchinal properties.

to have thetance.

Matching Simulation to Experime

• Knowledge of the initial longitudinal phase sL2 is critical to getting the bunch length andcorrect.

• Used elegant without CSR to match the melength data by adjusting initial RMS longitud

• “Morphed” the PARMELA-generated bunchsame RMS longitudinal properites and emit


e-Space

l

Nominal and Morphed Longitudinal Phas

morphed

nomina


imulation
Comparison with Bunch-Length-Matched S


amatic
Effect of Using a Gaussian Beam is Dr


s Match
...Even Though the RMS Bunch Length


nt

bend only

OL model

S54 model

.

CSR Model for Drifts is All-Importa

OL model uses Overtaking Length forCSR attenuation in drifts.

S54 model uses equations 53-54 fromSaldin et al. for CSR attenuation in drifts

Standard 0.4nCdesign simulations.

Bend only means no CSR in drifts.


30/01)

ase to vary the

the bunchal bend (“DD”

4.3µm and

Impact of CSR on Energy Spectra (3/

• In this experiment, we again varied L2 rf phbunch length.

• We recorded beam profiles at the center ofcompressor (“BCC” flag) and after the verticflag).

• Charge was 270pC, normalized emittances3.1µm for x and y.


Bunch Length Data


ssion
Apparent Energy Spectra in Decompre


est
Apparent Energy Spectra Near Cr


ompression
Apparent Energy Spectra Near Maximum C


ompression
Apparent Energy Spectra Near Maximum C


ing

tive of the

the tail.

wake that lookserating leading

w in energy andoves them

linearize (or

Why CSR Causes Energy Clump

• The CSR wake has the shape of the derivalongitudinal density.

• CSR accelerates the head and decelerates

• A temporal clump produces a feature in thelike the derivative of the local density, accelparticles and decelerating trailing particles.

• In compression, the leading particles are lothe trailing particles are high. Thus, CSR mtogether into an energy clump.

• To reduce the impact of CSR, one needs tothermalize) the longitudinal distribution.


y Spreads
Measured and Simulated Apparent Energ


tra
Simulated Apparent Energy Spec


tra


tra


c) Images
Vertical Bend Diagnostic (Dowell Diagnosti





es
Simulation Produces Similar Imag




ing et al.)

ipoleis used

5nCrom theest Facility

Data from CTF II Experiments (L. Groen

0

0.25

0.5

0.75

1

1.25

1.5

1.75

2

0

0.25

0.5

0.75

1

1.25

1.5

1.75

2

5 10 15

Mom

entu

m L

oss

[MeV

/c]

σl = 1.25 mm

αl = 7.58

Exp CTF II hc = 15 mm

TraFiC4 hc = inf

ELEGANT hc = inf

σl = 1.00 mm

αl = 6.67

γεl = 350 µm

αl = 6.67

σl = 0.76 mm

αl = 5.63

R56 [mm]

σl = 0.53 mm

αl = 3.55

5 10 15

A four-dchicanewith a 1beam fCLIC TII.


pressor

beam to match

ement.

e incoming

t dipole.

n wakefields.

symmetry.

er.)

Ideas for Future Work on APS Bunch Com

• New PARMELA simulations of photoinjectoractual conditions.

• Improve resolution of bunch length measur

• Perform tomography to accurately determinlongitudinal phase-space.

• Quantify effect of horizontal beamsize in las

• Verify that effects are due to CSR rather tha(Requires high charge.)

• Conduct experiments with variable and

(Requires completion of telescoping chamb

R56


Documents

Measurements and Simulations of CSR Effects in the APS