Design and Upgrade of a Compact Imaging
System for the APS Linac Bunch Compressor
B. Yang, E. Rotela, S. Kirn, R. Lill, and S. Sharma
Argonne National Laboratory, 9700 South Cass Avenue, Argonne, IL 60439
Abstract. We present the design, performance, and recent upgrade of a high-resolution, highcharge-sensitivity imaging camera and beam position monitor (BPM) system for the APS linac
beam profile measurement. Visible light is generated from the incoming electron beam using
standard YAG or optical transition radiation (OTR) converter screens. Two CCD cameras share
the light through a beam splitter, each with its own imaging optics. Normally, one camera is
configured with high magnification and the other with large field of view. In a different lens
configuration, one of the cameras focuses at the far field, allowing the measurement of beam
divergence using an OTR screen, while the other camera simultaneous measures the beam size.
A four-position actuator was installed recently to provide the option of two screens, a wakefield
shield, and an in situ calibration target. A compact S-band beam position monitor electrode was
designed to mount directly on the flag. The BPM rf circuit was fabricated from a machinable
ceramic (MACOR) cylinder substrate, and the copper electrodes were deposited on the substrate.
The new design and precision fabrication process make it viable to explore more complex
microstrip components printed on the substrate and higher frequency applications. The proximity
of the BPM and the camera (< 5 cm) will provide a precise calibration platform to study shot-toshot jitter, long-term stability of both systems, and the dependence of BPM signal on beam
properties (size, charge distribution) due to nonlinearity.
INTRODUCTION
A chicane bunch compressor was designed and implemented at the Advanced
Photon Source (APS) in 2000 [1] to increase the peak current of the bunch and
improve the performance of the low-energy undulator test line (LEUTL) free-electron
laser (PEL). It is expected to operate at -200 MeV with normalized emittance in the
range of 1 n to 4 n mm-mrad. Coherent synchrotron radiation (CSR) effects are
expected to be significant at these emittance levels. Their study calls for accurate
emittance measurement at the level of several percent or better.
TABLE 1. Electron Beam Parameters at the Three Screen Section______________
ELECTRON ENERGY (MEV)
200 ( y= 400)
Single bunch charge (nC)
0.2 -- 1 .0 x
Normalized emittance (n mm-mrad)
4.0
1.0
Emittance 8 (71 mm-mrad)
0.010
0.0025
Beta function at beam waist (3 (m)
1 .00
Rms beam size, ^£/3 (urn)
100
50
Rms beam size, e/j3 (urad)
100
50
CP648, Beam Instrumentation Workshop 2002: Tenth Workshop, edited by G. A. Smith and T. Russo
© 2002 American Institute of Physics 0-7354-0103-9/02/$19.00
393
A compact, modular imaging system [2] was designed and implemented to support
studies of the low-emittance beams, with the high resolution, reliability,
reproducibility, and accuracy needed. In this paper, we report several implemented
and planned upgrades to the system.
A new screen mount was designed and installed. It includes a wakefield shield, a
YAG screen mount, an optical transition radiation (OTR) screen mount, and a
calibration target. A new set of optics was designed to focus in the far field, or image
the angular divergence of the electron beam when the OTR screen is employed.
A compact electronic beam position monitor (BPM) was designed to mount directly
on the flag. The BPM design uses a new fabrication process, which is expected to
make it more viable to explore complex electrode configurations and for higher
frequency applications.
FOUR-POSITION SCREEN MOUNT
Due to high-energy spread used for chirping the beam, the rms beam size in the
horizontal direction can reach several millimeters in the midsection of the chicane.
This is comparable to the aperture of the vacuum vessel of 25 mm. A smooth
transition of the beam pipe is desired to reduce wakefield effects to the electron bunch.
Figure 1 shows the design of the wakefield shield. Figure 2 shows an ACAD rendering
of the screen mount assembly.
The wakefield shield / transition is mounted at position no. 1 on a four-position
vacuum feedthrough, powered by a pair of pneumatic actuators connected back-toback. The stroke of the short actuator is 38.4 mm and that of the long actuator is 76.4
mm. Four screen positions (38.4 mm spacing) can be reached by extending or
contracting these two pneumatic actuators.
FIGURE 1. Four position screen mount assembly: (1) Long-stroke actuator, (2) short-stroke actuator,
(3) mounting block and linear bearing housing, (4) vacuum bellow, (5) wakefield shield / transition,
(6) OTR screen mount, (7) YAG screen mount, (8) mirror, (9) calibration target, and (10) viewport for
target illumination.
394
FIGURE
FIGURE 2.
2. ACAD
ACAD 3-D
3-D rendering
rendering of
of the
the four-position
four-position screen
screen mount
mount assembly.
assembly.
At
At position
position no.
no. 2,
2, aa 45-degree
45-degree mirror
mirror is
is mounted
mounted as
as an
an OTR
OTR screen.
screen. ItIt isis used
used for
for
high-intensity
high-intensity (charge
(charge density)
density) electron
electron beams.
beams. At
At position
position no.
no. 3,3, aa YAG
YAG scintillator
scintillator
screen
screen (20
(20 mm
mm ×x 15
15 mm
mm ×x 0.1
0.1 mm)
mm) isis mounted
mounted with
with aa mirror.
mirror. ItIt isis used
used for
for lowlowintensity
intensity electron
electron beams.
beams.
At
At position
position no.
no. 4,
4, an
an aluminum-coated
aluminum-coated prism
prism isis used
used as
as aa first
first surface
surface mirror,
mirror, in
in
conjunction
conjunction with
with aa calibration
calibration target.
target. The
The target
target isis designed
designed to
to have
have aa dark
dark metallic
metallic
coating
coating with
with an
an etched
etched dot
dot matrix
matrix at
at 0.25
0.25 mm
mm spacing.
spacing. The
The illumination
illumination light
light isis fed
fed
from
outside
of
the
vacuum
through
a
viewport.
Images
of
these
bright
dots
in
from outside of the vacuum through a viewport. Images of these bright dots inthe
the dark
dark
background
background are
are convenient
convenient for
for the
the video
video digitizer
digitizer to
to pick
pick out
out and
and perform
perform profile
profile
analysis.
The
center
coordinates
of
the
dots
are
used
to
calibrate
pixel
sizes,
analysis. The center coordinates of the dots are used to calibrate pixel sizes, while
while the
the
µm-diameter dots
rms
rms image
image size
size of
of these
these 1010-//m-diameter
dots are
are used
used to
to obtain
obtain optical
optical resolution
resolution of
of
the
the camera.
camera. Since
Since the
the optical
optical distance
distance from
from the
the calibration
calibration target
target isis the
the same
same as
as from
from
the
the YAG
YAG screen,
screen, the
the target
target is
is also
also used
used as
as aa focusing
focusing aid.
aid.
OPTICS
OPTICS FOR
FOR BEAM
BEAM DIVERGENCE
DIVERGENCE MEASUREMENT
MEASUREMENT
By
By design,
design, the
the camera
camera module
module has
has two
two cameras
cameras sharing
sharing the
the light
light with
with aa beam
beam
splitter.
Normally,
each
camera
has
its
own
imaging
optics,
one
is
used
splitter. Normally, each camera has its own imaging optics, one is used for
for highhighresolution
resolution imaging
imaging but
but covers
covers only
only aa part
part of
of the
the screen,
screen, while
while another
another isis used
used for
for lowlowresolution
resolution imaging
imaging and
and covers
covers the
the full
full view
view of
of the
the screens.
screens. By
By reconfiguring
reconfiguring the
the
optics,
optics, however,
however, one
one camera
camera could
could be
be equipped
equipped with
with optics
optics for
for the
the far
far field,
field, i.e.,
i.e., with
with
object
object space
space at
at infinity.
infinity.
395
Measurement Using OTR Angular Distribution
To measure relatively large beam divergence (crx< ~ 0.1/j}, we can use the features
in the angular distribution of the optical transition radiation. A linear polarizer
installed in the turn table can be used to select the direction of polarization to be along
the x or y-axis. When the x-polarization is selected, the OTR angular distribution is,
where (ft, 8y) is the angle from the specular direction, and A is a constant. For an
electron beam with Gaussian divergence distribution, with rms divergence of (a^, 0y),
the observed spatial distribution is
]]g(ex-elt',ey-ey'}ix(ex',ey'}delt'dey'.
(3)
-n -n
If we integrate over the y-coordinates to improve statistics, the resultant profile is
For y» 1, we could change the limits of the integration to infinity. Inserting Eqs. (1)
- (3) into (4), one obtains the following expression,
where
p = y0x,andcrp=ycrx,.
(6)
For very small beam divergence, Eq. (5) gives
^(0)DAcJ/,(c7 p «l).
(7)
The entire profile can be calculated with a given beam divergence ox> (Fig. 3).
However, for a quick estimate of the beam divergence, the ratio of the center
minimum and the maximum of the angular distribution can be used,
.
Fmax Fmax
The ratio for Gaussian beams with ap « 1 can be easily modeled. The results can be
summarized in the following relations,
(9)
This fitted expression has an accuracy of 1% or better for ax < 0.8/7(Fig. 4).
396
INTENSITY(ARB.
(ARB.UNITS)
UNITS)
INTENSITY
0.5
0.5
0.5
CO 0.4
0.4
0.4
CD
0.3
0.3
0.3
0.2
j= 0.2
0.2
CO
σx=0.5/γ
σx=0.5/γ
z
LU
t 0.1
0.1
0.1
σ =1/γ
σxx=1/γ
σx=0.2/γ
σx=0.2/γ
σx=0
σx=0
0.0
0.0
0.0 0
0
0
1
1
1
2
22
3
33
4
44
5
55
OBSERVATION
ANGLE
(γθ
OBSERVATION ANGLE
ANGLE (γθ
(y0xx)))
OBSERVATION
x
FIGURE
3. Minimum/maximum
ratio of
the
1-D
OTR
profile
for
Gaussian
electron
beam.
FIGURE
of the
the 1-D
1-D OTR
OTR profile
profile for
foraaaGaussian
Gaussianelectron
electronbeam.
beam.
FIGURE 3.
3. Minimum/maximum
Minimum/maximum ratio
ratio of
For
compressor,
the
dynamic
range
of
the
system
For
our
application
at the
the APS
APS bunch
bunch compressor,
compressor,the
thedynamic
dynamicrange
rangeof
ofthe
thesystem
system
For our
our application
application at
at
the
APS
bunch
is
limited
by
the
video
digitizer,
which
has
an
8-bit
resolution
at
zero
noise.
If
we
set
isis limited
which has
has an
an 8-bit
8-bit resolution
resolutionatatzero
zeronoise.
noise.IfIfwe
weset
set
limited by
by the
the video
video digitizer,
digitizer, which
the
noise
level
at
2
bit,
the
usable
dynamic
range
is
about
6
bit.
Best
measurements
the
noise
level
at
2
bit,
the
usable
dynamic
range
is
about
6
bit.
Best
measurements
the noise level
dynamic
range
is
about
6
bit.
Best
measurements
-6
can
lowest
measurable
beam
divergence
given
by
can
obtain
ratio
of ηηr/ >> 222"-66,, hence
lowest measurable
measurablebeam
beamdivergence
divergenceisis
isgiven
givenby
by
can obtain
obtain aa ratio
ratio of
hence the
the lowest
−6
−6
2
1
σ
≥ 2 === 1—,, ,
(10)
(10)
σvxxx>^—
≥
(10)
γγ
88γγ
400.
with Table
Table
shows
that
higher
resolutionwill
will
be
or
or
0.3
mrad
for γy=
Comparison with
Table 111shows
shows that
that higher
higher resolution
resolution
willbe
be
γ == 400.
400. Comparison
or 0.3
0.3 mrad
mrad for
needed
for
the
measurements
at
the
APS
bunch
compressor.
needed
APS bunch
bunch compressor.
compressor.
needed for
for the
the measurements
measurements at the APS
7
%y
LU
o
11
0.1
0.1
0.1
0.1
0.1
0.1
0.01
0.01
0.01
0.01
0.01
0.01
LU
s
LU
RELATIVE
RELATIVE ERROR
ERROR
BEAM DIVERGENCE
DIVERGENCE(γσ
(γσxx))
BEAM
11
OC
O
DC
DC
LU
LU
1
LU
DC
CO
0.001
0.001
0.001
0.0001
0.0001
0.0001
0.001
0.001
0.01
0.01
0.01
0.1
0.1
0.1
0.001
0.001
0.001
11
MINI-TO-MAX
RATIO
ηη))
MINI-TO-MAX RATIO
RATIO (((rj)
FIGURE
1-D
OTR
profile
for
Gaussian
electron
beam,
FIGURE 4.
4. The
The minimum/maximum
minimum/maximum ratio
FIGURE
4.
The
minimum/maximum
ratio of
of the
the 1-D
1-D OTR
OTR profile
profile for
foraaaGaussian
Gaussianelectron
electronbeam,
beam,
circles:
curve:
fit
by
Eq. (9),
(9),
and
triangles:
the
relative
circles: calculated
calculated ratio
ratio for
for aa given
circles:
calculated
ratio
for
given beam
beam divergence,
divergence, curve:
curve: fit
fit by
by Eq.
Eq.
(9), and
and triangles:
triangles:the
therelative
relative
errorin
inbeam
beam divergence
divergence between
between the
calculation and
error
between
the exact
exact calculation
and
the fit.
fit.
error
in
beam
divergence
and the
the
fit.
397
Measurement Using OTR Interferometer
To
divergence measurements
measurements ((o^
0.1/j^,
we
To improve
improve the resolution of the beam divergence
σx' <<0.1/
γ), we
could
foil is
is to
to be
be inserted
inserted at
at the
the first
first station
station
could use
use the
the OTR
OTR interferometer [3]. One thin foil
of
mid-station,
of the
the three-screen
three-screen section, and an OTR screen/mirror is inserted at the mid-station,
1.0 m downstream from the foil. The
The far-field
far-field camera
camera located
located atat the
the second
second station
station
records
records the
the interferogram.
interferogram.
To
at this
this geometry,
geometry, we
we calculated
calculated the
the
To illustrate
illustrate the resolution
resolution of the interferometer at
OTR
beam σcrxx', ==σO"yy', ==a.
The result
result isis
σ . The
OTR intensity
intensity distribution for circular electron beam
given
given by
by aa simple
simple integral,
π 2π
3
π (1 + q 2 )
q
2
K (θ ) = 4 A∫ ∫
cos
G (γθ , q ) dq ,
(11)
(11)
2
2
pλ 2
0 0 (1 + q )
where
where a
σpp =
= ya
γσ , pλ = γ 2λ / L and

(12)
+ q 2 − 2 pq cos φ  dφ .
(12)
πσ p 0
p


Assuming we
we have
have 1 meter
meter foil-mirror spacing, and an observation
Assuming
observation wavelength
wavelength of
of 0.63
0.63
µm, we
we show
show several
several sample
sample calculations
calculations in
Fig. 5,
including the
|Lim,
in Fig.
5, including
the beam
beam conditions
conditions
listed in
in Table
Table 1.
1. From
From the
the simulated
simulated interferogram,
interferogram, we
listed
we conclude
conclude that
that the
the geometry
geometry is
is
γ.
suitable for
for the
the measurement
measurement of
of beams
suitable
beams with
with divergence
divergence less
less than
than 0.1/
Q.l/y.
=
G ( p, q ) =
G(p,q)

π
1
2

∫ exp − 2σ ( p
1
)
2
2
INTENSITY (ARB. UNITS)
0.30
0.30
0.25
CO 0.25
σx=0.04/γ
σx=0.1/γ
0.20
σx=0.02/γ
0.15
σx=0
0.10
0.05
0.00
0.00
0
1
2
3
4
OBSERVATION ANGLE
OBSERVATION
ANGLE (γθ)
(76)
FIGURE
5.
The
OTR
interferometer
light
intensity
as
a
FIGURE 5. The OTR interferometer light intensity as a function
function of
of observation
observation angle
angle (in
(in far
far field).
field).
generate intensity
Beam with
with lower
lower emittance
emittance (0^=0)
(σx=0) generate
Beam
intensity oscillations
oscillations that
that extends
extends to
to higher
higher angles,
angles, while
while
generates oscillations
beam with
with higher
higher emittance
emittance (o^=0.1/^
(σx=0.1/γ) generates
beam
oscillations that
that damps
damps in
in aa short
short range.
range.
COMPACT ELECTRONIC
ELECTRONIC BEAM
COMPACT
BEAM POSITION
POSITION MONITOR
MONITOR
The shorted
shorted A/4
λ/4 wave
wave S-band
S-band BPM
The
BPM electrode
electrode shown
shown in
in Fig.
Fig. 66 was
was designed
designed as
as aa
test
vehicle
for
evaluating
the
performance
of
a
new
compact
BPM
electrode.
The
goal
test vehicle for evaluating the performance of a new compact BPM electrode. The goal
398
was
was to
to reduce
reduce the
the size
size and
and assembly
assembly complexity
complexity for
for the
the BPM,
BPM, so
so aa flag
flag and
and BPM
BPM
combination
can
be
used
in
a
limited
length
of
beam
line.
This
concept
of
combining
combination can be used in a limited length of beam line. This concept of combining
diagnostics
diagnostics also
also has
has the
the advantage
advantage of
of using
using aa cross
cross calibration
calibration scheme
scheme that
that can
can be
be
implemented
to
verify
the
performance
of
each
of
the
instruments.
The
proximity
implemented to verify the performance of each of the instruments. The proximity of
of
the
micron
the BPM
BPM and
and the
the camera
camera (<
(< 55 cm)
cm) will
will give
give us
us aa good
good opportunity
opportunity to
to study,
study, to
to micron
precision,
precision, the
the dependence
dependence of
of BPM
BPM signal
signal on
on beam
beam properties
properties (size,
(size, charge
charge
distribution)
due
to
nonlinearity.
It
also
provides
a
precise
calibration
distribution) due to nonlinearity. It also provides a precise calibration platform
platform to
to
study
study shot-to-shot
shot-to-shot jitter,
jitter,and
andlong-term
long-term stability
stability of
ofboth
bothsystems.
systems.
Several
Several design
design features
features can
can be
be seen
seen from
from Fig.
Fig. 6:
6: (1)
(1) The
The rf
rf microstrip
microstrip circuit
circuit and
and
vacuum
vacuum feedthrough
feedthrough connectors
connectors are
are replaceable.
replaceable. This
This isis achieved
achieved by
by welding
welding the
the
feedthrough
feedthrough connectors
connectors to
to aastandard
standard removable
removable vacuum
vacuum flange.
flange. The
The center
center conductors
conductors
make
make contact
contact to
to the
the stripline
stripline via
via aa beryllium
beryllium copper
copper spring
spring contact.
contact. This
This design
design
simplifies
assembly
of
the
BPM
and
enables
the
replacement
of
a
connector
simplifies assembly of the BPM and enables the replacement of a connector in
in the
the
event
of
damage.
(2)
The
BPM
electrodes
are
fabricated
on
a
machinable
(MACOR)
event of damage. (2) The BPM electrodes are fabricated on a machinable (MACOR)
ceramic
ceramic cylinder
cylinder substrate
substrate with
with aa dielectric
dielectric constant
constant of
of 4.88.
4.88. The
The rfrf circuit
circuit isis etched
etched on
on
the
copper/ceramic
substrate
with
a
process
similar
to
printed
circuit-board
the copper/ceramic substrate with a process similar to printed circuit-board
technology.
technology. The
The ceramic
ceramic cylinder
cylinder isis captivated
captivated in
in the
the BPM
BPM housing
housing by
by aa threaded
threaded
transition
sleeve.
The
new
design
and
precision
fabrication
process
make
transition sleeve. The new design and precision fabrication process make itit viable
viable to
to
explore
explore more
more complex
complex microstrip
microstrip components
components printed
printed on
on the
the substrate
substrate and
and higher
higher
frequency
frequency applications.
applications. The
The fabrication
fabrication costs
costs are
are expected
expected to
to be
be much
much lower
lower than
than the
the
34
mm
BPM
currently
in
use
in
the
APS
LINAC.
34 mm BPM currently in use in the APS LINAC.
The
The connectors
connectors followed
followed by
by aa bore
bore in
in the
the metal
metal housing
housing form
form 50-ohm
50-ohm transmission
transmission
lines
lines to
to each
each of
of the
the four
four blades.
blades. The
The wall
wall current
current isis intercepted
intercepted by
by the
the 26.24-mm
26.24-mm long
long
electrode
electrode that
that forms
forms 50
50 ohms
ohms impedance
impedance with
with the
the vacuum
vacuum chamber.
chamber. The
The gap
gap that
that is
is
formed
formed between
between the
the vacuum
vacuum chamber
chamber and
and the
the stripline
stripline isis 3.48
3.48 mm.
mm. The
The total
total length
length of
of
the
the test
test BPM
BPM stripline
stripline assembly
assembly is
is 54.61
54.61 mm
mm and
and the
the inner
inner electrode
electrode radius
radius 35
35 mm,
mm,
which
which isis the
the same
same as
as that
that for
for the
the linac
linac vacuum
vacuum chamber.
chamber. The
The downstream
downstream end
end of
of the
the
transmission
transmission line
line isis electrically
electrically shorted,
shorted, which
which also
also provides
provides the
the mechanical
mechanical strength
strength
and
and rigidity
rigidity to
to support
support the
the striplines.
striplines.
FIGURE
FIGURE 6.
6. Test
Test BPM
BPM assembly.
assembly.
399
The preliminary electromagnetic modeling indicates that the ceramic actually
The preliminary electromagnetic modeling indicates that the ceramic actually
improve
pincushion distortion performance as compared in Fig. 7. The thickness of
improve pincushion distortion performance as compared in Fig. 7. The thickness of
the ceramic has not been optimized and is presently set to 3.4 mm for mechanical
the ceramic has not been optimized and is presently set to 3.4 mm for mechanical
reasons. The overall performance is expected to be similar to the standard S-band [3],
reasons. The overall performance is expected to be similar to the standard S-band [3],
which
is presently used in all APS linac applications.
which is presently used in all APS linac applications.
FIGURE7.7.Vertical
Verticalplane
planeequipotential
equipotential lines
lines with
with ceramic
FIGURE
ceramic (left)
(left) and
andwithout
withoutceramic
ceramic(right),
(right),showing
showing
largerlinear
linearregion
regionininthe
thecase
caseof
ofceramic
ceramic housing
housing (field
(field lines
larger
lines being
being parallel
paralleland
andequal
equalspacing).
spacing).
ACKNOWLEDGMENTS
ACKNOWLEDGMENTS
Wewish
wishtoto thank
thank G.
G. Decker,
Decker, S.
S. Milton,
Milton, M.
We
M. Borland,
Borland, and
and O.
O.Singh
Singhfor
forhelpful
helpful
discussions,
encouragement,
and
support.
discussions, encouragement, and support.
REFERENCES
REFERENCES
1. Borland, M et al., A Highly Flexible Bunch Compressor for the APS LEUTL FEL, Proc. LINAC
1. Borland, M et al., A Highly Flexible Bunch Compressor for the APS LEUTL PEL, Proc. LINAC
2000.
2.2000.
Yang, B. X., et al., Design and Performance of a Compact Imaging System for the APS Linac Bunch
2. Yang,
B. X., etProc.
al., Design
Performance
of a Compact
Imaging
System for the APS Linac Bunch
Compressor,
Particleand
Accelerator
Conference
2001, 2335
(2002).
Particle
Accelerator
Conference
2001, 2335
3.Compressor,
R. B. FioritoProc.
and D.
W. Rule,
“Optical Transition
Radiation
Beam(2002).
Emittance Diagnostics”, in AIP
3. R.Conference
B. Fiorito and
D. W. Rule,
Transition
Proceedings
No. "Optical
319, R. Shafer,
editorRadiation
(1994). Beam Emittance Diagnostics", in AIP
No. 319, R.
Shafer,
editor (1994).
4.Conference
A. Gorski,Proceedings
R. Lill, “Construction
and
Measurement
Techniques for the APS LEUTL Project RF
4. A.Beam
Gorski,
R. Lill,
"Construction
and Measurement
Techniques
for the
APS LEUTL
RF
Position
Monitors,”
Proceedings
of the 1999 Particle
Accelerator
Conference,
New Project
York, pp.
Beam
Position
Monitors," Proceedings of the 1999 Particle Accelerator Conference, New York, pp.
1411-1413
(1999).
1411-1413(1999).
400