Barrier Buckets in the Fermilab Recycler Ring
C.in
M.the
Bhat
Barrier Buckets
Fermilab Recycler Ring
MIIRR Department, Beams Division,
C. Batavia,
M. BhatIL,60510, USA,
Fermilab", P.O. Box 500,
MI/RR Department, Beams Division,
∗
Abstract. The Fermilab Recycler Ring Fermilab
is an 8GeV
permanent
storage ring. It is equipped with a
, P.O.
Box 500, magnet
Batavia, antiproton
IL,60510, USA,
wide-band rf system to produce barrier buckets of any shape. We report here on some beam dynamics simulation studies
and beam experiments
out using
the barrier
in the Recycler
Abstract. carried
The Fermilab
Recycler
Ring is rf
ansystem
8GeV permanent
magnet Ring.
antiproton storage ring. It is equipped with a
wide-band rf system to produce barrier buckets of any shape. We report here on some beam dynamics simulation studies
and beam experiments carried out using the barrier rf system in the Recycler Ring.
INTRODUCTION
The Fermilab RecyclerINTRODUCTION
Ring (RR) [1] has been
Fermilab
Recycler Ring
(RR) collider
[1] has been
built with theThe
goal
of increasing
ppbar
built
with
the
goal
of
increasing
ppbar
luminosity for Run-lib in the Tevatron fromcollider
for 2 Run-IIb
in the Tevatron from
1
SESlcnTWluminosity
to 2E32cm"
sec"1. The
RR will be the
5E31cm-2sec-1 to 2E32cm-2sec-1. The RR will be the
main pbar storage
andstorage
cooling
ring in ring
the Fermilab
main pbar
and cooling
in the Fermilab
accelerator complex.
Hence,
RR
plays
a
vital arole
accelerator complex. Hence, RR plays
vitalinrole in
future Tevatron
collider
operations.
future
Tevatron
collider operations.
particle motion in barrier rf buckets is presented in
reference
particle4.motion in barrier rf buckets is presented in
reference 4.
Presently,
we are commissioning
RR using
Presently, we
are commissioning
the RRtheusing
protons from the Fermilab Booster or pbars from the
protons from the Fermilab Booster or pbars from the
Accumulator Ring (AR). The beam transfer in both
Accumulatorcases
Ringhas
(AR).
Thedone
beam
in both
FIGURE 1. A schematic view of cooled, hot and transfer
to be
by transfer
separating
previously FIGURE
1. Ainschematic
viewusing
of cooled,
and transfer
beam regions
RR separated
barrier hot
buckets.
The
cases has to
be done
previously
transferred
beamby
fromseparating
the new transfer.
In future,
beam
regions
in
RR
separated
using
buckets.
transfer
region
has
2.5MHz
rf wave
withbarrier
rf bucket
matched The
transferred beam
from
the
new
transfer.
In
future,
unused pbars at the end of each collider run will be
to theregion
beam coming
from MI.rf wave with rf bucket matched
transfer
has 2.5MHz
firstofineach
the Tevatron
TeVbeto 150
unused pbarsdecelerated
at the end
collider from
run 1will
to
the
beam
coming
from
ML
GeV and
thenTevatron
in the MI from
GeV toto8 150
GeV and,
decelerated first
in the
from150
1 TeV
BARRIER
BUCKETS IN RR
finally, transferred to RR. The emittance of the pbars
GeV and then
in the MI from 150 GeV to 8 GeV and,
from the Tevatron at the time of injection to RR is
ABARRIER
wide-band RF BUCKETS
system capable IN
of RR
generating
finally, transferred
The eV-s
emittance
of the pbars
expectedtotoRR.
be <16
in longitudinal
plane and
many barrier buckets of any shape has been built[5]
from the Tevatron
at the
of injection
to RRpbars
is will
Aandwide-band
RF The
system
capable
installed in RR.
maximum
heightof
of agenerating
barrier
<20π-mm-mr
in time
transverse
planes. These
expected to have
be <16
eV-s in
longitudinal
and to the many
barrier
buckets
of anyeach.
shape
built[5]
considerably
larger
emittance asplane
compared
pulse
is set to
about +/-2kV
The has
widthbeen
of a pulse
from AR. Inplanes.
RR, we have
separatewill
the cold and can
be varied.
The pbar
from AR
will be
<207i-mm-mrpbars
in transverse
Theseto pbars
installed
in RR.
The beam
maximum
height
ofhaving
a barrier
pbars and
the hot
pbars in azimuth.
Thus, to
at any
rf structure
and thateach.
from The
RR towidth
MI can
have considerably
larger
emittance
as compared
the given pulse2.5MHz
is set to
about +/-2kV
ofhave
a pulse
time it is required that a maximum of three zones in
2.5MHz or 7.5MHz rf structure. The rf system can
pbars from AR.
In RR, we have to separate the cold
can
be
varied.
The
pbar
beam
from
AR
will
be
having
RR azimuthally separated, viz., transfer, stacked
produce 2.5MHz/7.5 MHz rf buckets in between
pbars and the(cooled)
hot pbarsand
in azimuth.
at any
given in 2.5MHz
structure
and2that
from
RR to MI
can
hot beam Thus,
regions
as indicated
barrierrfbuckets.
Figure
shows
a schematic
view
of have
a
time it is required
that
a
maximum
of
three
zones
in
2.5MHz
or
7.5MHz
rf
structure.
The
rf
system
can
Figure 1. All these specific needs can be met only by
barrier bucket in RR generated using square pulses.
RR azimuthally
viz., intransfer,
stacked
havingseparated,
barrier buckets[2]
the RR and
be able to produce 2.5MHz/7.5 MHz rf buckets in between
dynamically
change
their as
locations
in the
(cooled) and
hot beam
regions
indicated
in RR barrier buckets. Figure 2 shows a schematic view of a
Figure 1. Allazimuthally.
these specific needs can be met only by
barrier bucket in RR generated using square pulses.
having barrier buckets [2] in the RR and be able to
The barrier bucket rf technology was invented at
dynamically Fermilab[2]
change in
their
in the inRR
¥0
1983 locations
and used successfully
the pbar
azimuthally. Debuncher Ring. The shape of barrier pulses used in
ftis/ffi
the Debuncher Ring was sinusoidal. In the RR we
!(
D
adopted
bipolar
square barrier
Recently,
a
The barrier
bucket
rf technology
waspulses.
invented
at
multi-particle
2-D simulation
code[3]
Fermilab[2] in
1983 and beam
used dynamics
successfully
in the pbar
has been
study beam
dynamics
FIGURE 2. A schematic view of a barrier bucket in RR.
Debuncher Ring.
Thedeveloped
shape ofto barrier
pulses
used issues
in in
Equal Hamiltonian contours are also shown.
RR and a comprehensive Hamiltonian formalism of
the Debuncher Ring was sinusoidal. In the RR we
adopted bipolar square barrier pulses. Recently, a
by the
Departmentcode[3]
if Energy under contract No. DE-AC02-76CH03000.
multi-particle*Work
beamsupported
dynamics
2-DU.S.
simulation
has been developed to study beam dynamics issues in
FIGURE 2. A schematic view of a barrier bucket in RR.
RR and a comprehensive Hamiltonian formalism of
Equal Hamiltonian contours are also shown.
*Work supported by the U.S. Department if Energy under contract No. DE-AC02-76CH03000.
CP642, High Intensity and High Brightness Hadron Beams: 20th ICFA Advanced Beam Dynamics Workshop on
High Intensity and High Brightness Hadron Beams, edited by W. Chou, Y. Mori, D. Neuffer, and J.-F. Ostiguy
© 2002 American Institute of Physics 0-7354-0097-0/02/$ 19.00
229
previously transferred
transferred beam
beam is
is shown
shown in
in bottom
bottom right
right
previously
of
Fig.
3.
of Fig. 3.
Half-height
Half-height of
of aa barrier
barrier bucket
bucket shown
shown in
in Fig.2
Fig.2 is
is
given
given by
by [1,4],
[1,4],
We have
have conducted
conducted series
series of
of experiments
experiments with
with
We
protons
and
pbars
in
the
RR
barrier
buckets
to
protons and pbars in the RR barrier buckets to
understand
beam
dynamics.
We
have
used
pbars
from
understand beam dynamics. We have used pbars from
AR and
and protons
protons from
from Booster.
Booster.
AR
and,
and, bunch
bunch area
area is
is given
given by,
by,
E
the beam
beam and
and r|η is
is slip
slip
£"o0 is
is synchronous
synchronous energy
energy of
of the
factor (=-0.0087 for RR). ω
=2π/T0.o. Where
Where
oxo =2πf
=2nf0o=2n/T
β =v/c.
beam revolution period in RR T0o~ 11.12 jisec.
µsec. fi=vlc.
^A
The quantity ∆
A E is energy spread of the beam. The
remaining parameters are indicated in Fig.2.
PBAR STACKING IN BARRIER
BUCKETS
FIGURE 3. Longitudinal beam dynamics simulation results
for beam stacking. Shown are
are distribution
distribution of
of beam
beam particles
particles
for
in (∆E,∆φ)-space.
The
rf
wave
forms
and
buckets
are also
(AE,A(|))-space.
forms
shown. For details see the text.
RF manipulations involved in beam stacking in RR
have been simulated using the program ESME[3].
These simulations do not take into account any rf noise
or other instrumental effects
effects which may give rise to
additional longitudinal emittance growths. Therefore
we demand all rf operations carried out in simulation
to be fully adiabatic i.e., with no emittance dilution.
However, from practical point of view we allow up to
about 5% longitudinal emittance dilution.
stacking in RR
Experimental results for pbar beam stacking
are shown in Fig.4. The lower traces
traces in
in each
each of the
the
figures
4(a)-4(e)
show
RF
cavity
gap
monitor
voltage
figures
monitor voltage
outputs. The upper traces are wall current
current monitor
monitor
signals fed to a digital oscilloscope. The
The later
later results
results
projection of the phase space
are similar to the projection
distributions shown in Fig.3 on §φ axis (i.e, on
figures are as
horizontal axis). Descriptions of figures
four bunches on the
the left
left are
are newly
newly
follows: a) four
transferred pbar beam and the long
long bunch on the
the right
right
transferred
immediately
hand side is the stacked beam, b) beam immediately
after debunching in barrier bucket. In this case, notice
after
that the beam distribution is slanted
slanted indicating
indicating
asymmetry in barrier pulse area,
area. c) squeezed
squeezed beam in
after transferring
transferring the
the newly
newly stacked
stacked
barrier bucket, d) after
beam. e) rf
rf waveform
waveform just
just
beam to originally stacked beam,
before next beam transfer;
transfer; this time notice the
the stacked
stacked
before
far, the
bucket is slightly wider than in 4(a) to 4(d). So far,
about 30E9 and
and
maximum pbar stacked in RR was about
life
that of proton beam was about 750E9 with a beam life
hours.
time of a few hours.
Figure 3 shows phase-space distributions of
particles in (∆E,∆φ)-space
four stages of beam
(AE,A(|))-space for four
stacking. The upper left figure shows beam in
2.5MHz rf bucket in RR. Next, the beam is debunched iso-adiabatically in about a second (see Fig.3
top right) in a barrier bucket with pulse width set to
491 nsec. This width was not fully optimized during
the simulation. During the next two seconds the debunched beam particles are squeezed to match
momentum spread of the stacked beam (notice that the
shown here is a first transfer). There is a potential for
emittance growth during squeezing and transferring
the beam to stacked region. This emittance growth has
to be minimized by optimizing speeds for squeezing
and transferring the beam. The simulation showed that
if beam squeezing is carried out faster than about 0.22
radian/sec in azimuthal plane we see emittance growth
more than 5%. Some level of emittance growth may
be acceptable for hot beam, but, not for cooled beam
from AR. Fully squeezed beam is shown in bottom left
side of Fig. 3. Finally, the new transfer
transfer along with the
We have carried out some preliminary experiments
from a stacked
stacked beam
beam and
and tried
tried to
to
to un-stack the beam from
transfer the beam to the
the MI.
MI. The
The sequences
sequences of
of this
this
transfer
operation is essentially reverse of stacking.
.
230
buckets is not trivial. We plan to
to measure
measure itit by
by use
use of
of
buckets.
schottky signals of the beam in the barrier buckets.
MacLachlan
The author would like to thank J. MacLachlan
during the beam dynamics simulations.
simulations. Special
Special thanks
thanks
Dey, Consolato
Consolato Gattuso,
Gattuso,
are due to Brian Chase, Joe Dey,
Keith Meishner and John Reid for
for their help at
at various
various
stages of this work
REFERENCES
1. G.
Jackson, “The
"The Fermilab
Fermilab Recycler
Recycler Ring
Ring
Technical Design Report”
Report" ,, FERMILAB-TM-1991
FERMILAB-TM-1991
(1996).
2. G.E. Griffin et. al, IEEE, Trans.
Trans. Nucl.
Nucl. Sci.,
Sci., NS-30,
NS-30,
3502 (1983).
3. J. MacLachlan, ESME modified to
to handle
handle beam
beam
dynamics in barrier buckets
buckets (1996),
(1996), (private
(private
communications 2001.
4. S.Y. Lee and K. Y. Ng, Phys. Rev. E, Vol. 55,
5992 (1997).
5. J.E. Dey and D.W. Wildman, IEEE, Proc. the 1999
5.
1999
Part. Accel. Conf. New York, NY,
NY, 869,
869, 1999).
1999).
B. Fellenz
J. Crisp,
of timer
two timer
6. 6.
B. Fellenz
and J.and
Crisp,
a set aofsettwo
gates
gates
are
being
developed
for
each
of
the
are being developed for each of the barrierbarrier
system.
system.
The will
system
will integrate
total
areawall
under
The system
integrate
total area
under
wall
current
monitor
signal
to
determine
current monitor signal to determine beam beam
intensity
intensity (private
(private communication
communication 2002).
2002).
FIGURE 4.
4. Barrier
Barrier buckets
buckets in
in RR
RR (lower
(lower traces).
traces).
FIGURE
Transferred and
and stacked
stacked pbars
pbars in
in RR
RR (upper
(upper traces),
traces). (a)
Transferred
(a) and
and
(e)
also
show
2.5MHz
rf
wave
inside
transfer
bucket.
(e) also show 2.5MHz rf wave inside transfer bucket.
.
Measurement of
of beam
beam intensity
intensity and
and the
the
Measurement
longitudinal emittance
emittance of
of the
the beam
beam in
in the
the barrier
barrier
longitudinal
bucket isis vital
vital in
in understanding
understanding beam
beam life
life time
time and
bucket
and
emittance
growth
in
RR.
Presently,
we
are
developing
emittance growth in RR. Presently, we are developing
methods to
to measure
measure beam
beam intensity
intensity by
by using
using aa
methods
technique
very
similar
to
fast
bunch
integrator[6].
technique very similar to fast bunch integrator [6].
Longitudinal emittance
emittance measurements
measurements in
in barrier
barrier
Longitudinal
231