Status and Plans of the SPL Study at CERN
Status and Plans of the SPL Study at CERN
R. Garoby for the SPL Study Team
R. Garoby for the SPL Study Team
CERN, Geneva, Switzerland
CERN, Geneva, Switzerland
Abstract. The study of the SPL (Superconducting Proton Linac), a 4 MW / 2.2 GeV H"- linac, began at CERN in 1999.
Abstract.
of theof
SPL
Proton Linac),
a 4the
MW
/ 2.2
GeV Electron
H linac, Positron
began at collider),
CERN in it
1999.
Based
on theThe
largestudy
inventory
RF(Superconducting
equipment decommissioned
from
LEP
(Large
was
Based onproposed
the largeas
inventory
of RF
LEPaccelerators.
(Large Electron
collider),
it was
originally
an upgrade
to equipment
the injectordecommissioned
complex for the from
high the
energy
SincePositron
that time,
the proposal
as an
to the injector
for communities
the high energy
Since
that time,
the proposal
hasoriginally
attractedproposed
the interest
ofupgrade
an increasing
number complex
of physics
andaccelerators.
the design has
evolved
in consequence.
haspresent
attracted
the interest
of an
number
of physics
communities
and the
evolved in consequence.
The
design
of the SPL
is increasing
presented in
this paper,
together
with a proposal
fordesign
a stagedhasrealization.
The present design of the SPL is presented in this paper, together with a proposal for a staged realization.
150m radius storage ring is used to accumulate the
150 m radius storage ring is used to accumulate the
linac beam. After accumulation, a 3 |is beam burst of
linac
Afterisaccumulation,
a 3 µs
beam burstring
of
14
2.27xl0beam.
protons
sent to a bunch
compression
2.27×1014 protons is sent to a bunch compression ring
before being delivered onto the pion production target.
before being delivered onto the pion production target.
The resulting requirements for the SPL are shown in
The resulting requirements for the SPL are shown in
Table 1, and its overall layout is sketched in Figure 1.
Table 1, and its overall layout is sketched in Figure 1.
TABLE 1. Main SPL characteristics.
TABLE 1. Main SPL characteristics.
H
Accelerated
ions
Accelerated ions
H2.2
Kinetic
energy
GeV
Kinetic energy
2.2
GeV
Mean
beam power
4
MW
Mean beam power
4
MW
Repetition
rate
Hz
50
Repetition rate
50
Hz
Pulse
2.8
ms
Pulse duration
duration
2.8
ms
mA
Mean
13
Mean current
current during
during the
the pulse
pulse
13
mA
Number
2.27X101414
Number of
of H"
H- per
per pulse
pulse
2.27×10
352.2
MHz
Bunch
Bunch frequency
frequency
352.2
MHz
Chopping
duty
cycle
61.6
%
Chopping duty cycle
61.6
%
Bunch
5/8
Bunch pattern
pattern
5/8
(nb.
(nb. of
of bunches
bunches // nb.
nb. of
of buckets)
buckets)
0.4
Norm,
Norm. r.m.s.
r.m.s. transverse
transverse emittances
emittances
0.4
πTcmmmrad
mm mrad
Longitudinal
0.3
deg MeV
MeV
Longitudinal r.m.s.
r.m.s. emittance
emittance
0.3
π71 deg
INTRODUCTION
INTRODUCTION
The SPL [1,2] was first proposed as a way of
The SPL [1,2] was first proposed as a way of
profiting
from the decommissioned LEP RF
profiting from the decommissioned LEP RF
equipment, to upgrade the characteristics of the beam
equipment, to upgrade the characteristics of the beam
delivered by the CERN complex of proton accelerators
delivered by the CERN complex of proton accelerators
at low cost. The SPL would then replace the 50 MeV
at low cost. The SPL would then replace the 50 MeV
proton linac (linac 2) and the 1.4 GeV PS Booster, and
proton linac (linac 2) and the 1.4 GeV PS Booster, and
inject
directly into the PS. Moreover, thanks to the
inject directly into the PS. Moreover, thanks to the
high
flux
high fluxofofprotons
protonspotentially
potentiallyavailable
availablefrom
from such
such aa
superconducting
linac,
other
users
could
easily
superconducting linac, other users could easily be
be
accommodated.
accommodated.For
Fora asecond
secondgeneration
generationISOLDE-like
ISOLDE-like
facility
facility[3],
[3],the
theSPL
SPL beam
beam isis directly
directly usable.
usable. For
For
neutrino
physics
(first
a
"Neutrino
super-beam"
neutrino physics (first a “Neutrino super-beam”
facility
facility[4]
[4]and
andultimately
ultimatelyaa"Neutrino
“NeutrinoFactory"
Factory” [5]),
[5]),
accumulator
and
bunch
compression
rings
accumulator and bunch compression ringshave
havetoto be
be
added.
added. The
The design
design ofof the
the accelerator
accelerator set-up
set-up has
has
consequently
consequentlyevolved.
evolved.
SPL
SPLDESIGN
DESIGN
The
Theproton
protondriver
driverofofaaneutrino
neutrinofactory
factoryisisthe
themost
most
demanding
application
for
the
SPL
[5].
For
demanding application for the SPL [5]. For this,
this, aa
33MeV
MeV
45keV
45 keV
6m
-
H
120
120MeV
MeV
62 m
40MeV
2.2 GeV
584 m
237MeV 383MeV
RFQ chopping RFQ
DTL
1 chop.
CCDTL
RFQ2 β
RFQ1
0.52 chop.
β 0.7RFQ2 β 0.8
SourceLow
LowEnergy
Energysection
section
Source
DTL
Superconducting section
Superconducting
dump
Debunching
Dehunching
Stretching and
and
collimation
collimation line
li
666 m
PS / Isolde
Accumulator Ring4
Accumulator
FIGURE1.
1. SPL
SPL synoptic
synoptic
FIGURE
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
96
Room
temperature
accelerating
structuresare
areused
used
Room
temperature
accelerating
structures
are
used
Room
temperature
accelerating
structures
in
the
first
69
m
of
the
linac
(120
MeV
kinetic
energy).
in
the
first
69
m
of
the
linac
(120
MeV
kinetic
energy).
in the first 69 m of the linac (120 MeV kinetic energy).
Above
this
energy
and
for
most
thelength
lengthofof
ofthe
the
Above
this
energy
and
for
most
the
length
the
Above
this
energy
and
for
most
ofofofthe
accelerator,
superconducting
structures
are
employed.
accelerator,
superconducting
structures
are
employed.
accelerator, superconducting structures are employed.
The
parameters
ofthe
the
sections
are
detailed
in
Table2.2.
2.
The
parameters
sections
are
detailed
Table
The
parameters
ofofthe
sections
are
detailed
inin
Table
TABLE
2.SPL
SPL
sections
parameters.
TABLE
sections
parameters.
TABLE
2.2.SPL
sections
parameters.___________
SectionFinal
FinalNb.
Nb.
Peak KlyKly- TeTe- Length
Length
Section
Final
Nb.
Peak
KlyTeLength
Section
ofofofPeak
cavities
(m)
Energy
RF
strons
trodes
cavities RF
(m)
Energycavities
RF strons
stronstrodes
trodes (m)
Energy
(MeV)
Power
(MeV)
Power
(MeV)
Power
(MW)
(MW)
(MW)
Source
0.045
Source
0.045
11
Source
0.0453
12.4
RFQ
1
0.5
1
RFQ
3
0.5
1
2.4
1 13 0.5
RFQ
2.4
Chopper 3 3 3
0.06 1
3.6
Chopper
3
0.06
33
3.6
Chopper
3
3
0
.
0
6
3
3.6
DTL
120
13
11.8
15
62
DTL
120
13
11.8 1515
62
DTL
236 13 42 11.8
1.5
42 62
101
β=0.52 120
236
42
1.5
101
β=0.52
4242
236
P=0.52
383 42 32 1.51.9
32 10180
β=0.7
383
32
1.9
32
80
β=0.7
32
32
383
80
1.9
P=0.7
1111
52
9.5
13
166
β=0.8
1111
52
9.5
13
166
β=0.8
166
p=0.8
2235 52 76 9.514.6 1319
237
β=0.8 1111
2235 7676
14.6 1919
237
β=0.8
237
p=0.8
Debunch.2235
2235
4 14.6
1
13
Debunch. 2235
2235
1
13
Debunch.
44223
Total
39.9 1 49
77 13666
Total
223 39.9
39.9 4949 7777 666
666
Total
223
FIGURE 2. Chopper prototype: double 100 Ω meander-line
FIGURE2.2.Chopper
Chopperprototype:
prototype:double
double100
100QΩmeander-line
meander-line
FIGURE
on ceramic
substrate.
ceramicsubstrate.
substrate.
ononceramic
By subsequently
adding
second
By
subsequentlyadding
adding the
thesecond
secondRFQ
RFQsegment,
segment,
By
subsequently
acceleration
will proceed upthe
to 5 MeV. RFQ
Beamsegment,
tests up
accelerationwill
willproceed
proceedup
uptoto55MeV.
MeV.Beam
Beamtests
testsup
up
acceleration
to 100 mA are foreseen, in pulsed mode with chopping
to
100
mA
are
foreseen,
in
pulsed
mode
with
chopping
toand
100inmA
foreseen,
in pulsed
mode
with chopping
CWare
without
chopping.
Beam
measurements
are
andininCW
CW withoutchopping.
chopping. Beammeasurements
measurementsare
are
and
planned towithout
fully characterise Beam
chopping efficiency and
planned
to
fully
characterise
chopping
efficiency
and
planned
to fully characterise
halo development.
The test chopping
stand will efficiency
be locatedand
at
halo
development.The
Thetest
test stand
standwill
willbebe located
located atat
halo
development.
CEA-Saclay.
CEA-Saclay.
CEA-Saclay.
RFQ and Chopping line
Drift Tube Linac
RFQand
andChopping
Choppingline
line
RFQ
DriftTube
TubeLinac
Linac
Drift
The RFQ will be of the four vane type, and its
TheRFQ
RFQ
willisbebe
thefour
fourvane
vane
type,from
anditsthe
its
The
will
ofofthe
type,
and
precise
design
now
expected
to
benefit
precise
design
is
now
expected
to
benefit
from
the
precise
design is of
nowthe
expected
benefitactually
from thein
development
IPHI todevice
development
of the
the
IPHI
device
actually inin
development
the
IPHI
device
construction of
for
CEA
and IN2P3
[6].actually
constructionfor
forthe
theCEA
CEAand
andIN2P3
IN2P3[6].
[6].
construction
The proposed SPL chopper structure consists of a
Theproposed
SPL
chopper
structure
consists
pair
ofproposed
deflecting
plates
withstructure
a meander
delay-line
The
SPL
chopper
consists
ofofaa
pair
of
deflecting
plates
with
a
meander
delay-line
on alumina
(Fig.with
2), without
separating
ridges
pairprinted
of deflecting
plates
a meander
delay-line
printed
onalumina
alumina(Fig.
(Fig.
2),without
withoutseparating
separating
ridges
[7]. onAttenuation
and2),
dispersion
measured
on
printed
ridges
[7].
Attenuation
and dispersion
dispersionwith
measured
prototypes
are in and
good
agreement
computation.
[7].
Attenuation
measured
onon
prototypes
areinmeander
ingood
goodagreement
agreement
with
computation.
The printed
on alumina
has
very good
prototypes
are
with
computation.
The
printed
meandereasy
alumina has
has very
very
good
vacuum
properties,
implementation
andgood
good
The
printed
meander
onon alumina
vacuum
properties,
easy
implementation
and good
good
radiation
resistanceeasy
and implementation
heat
transfer (water
cooled
vacuum
properties,
and
radiation
resistance
and
heat
transfer
(water
cooled
metal ground
plane).
The
high
permittivity
(ε)
of the
radiation
resistance
and
heat
transfer
(water
cooled
ceramic
permits
a
meander
width
below
25
mm
for
metalground
groundplane).
plane).The
Thehigh
highpermittivity
permittivity(e)
(ε)ofofthe
the
metal
ceramic
permits
a
meander
width
below
25
mm
for
particle
velocitya of
β=0.08, width
and gives
the 25
possibility
ceramic
permits
meander
below
mm forto
particle
velocity
β=0.08,
andquadrupoles.
givesthe
thepossibility
possibility
installvelocity
the deflectors
inside
The 100toto
Ω
particle
ofofp=0.08,
and
gives
characteristic
impedance
helps
reduce
the
driver
install
the
deflectors
inside
quadrupoles.
The
100
Ω
install the deflectors inside quadrupoles. The 100 Q
power. A prototype
500 Vhelps
chopper
pulse amplifier
has
characteristic
impedance
helps
reduce
the driver
driver
characteristic
impedance
reduce
the
been
realised,
and
tests
are
in
progress
[8].
power.
A
prototype
500
V
chopper
pulse
amplifier
has
power. A prototype 500 V chopper pulse amplifier has
beenrealised,
realised,
andtests
tests
are
inprogress
progressof
[8].
been
and
[8].
The
chopper
lineare
is incomposed
two 1 m long
The DTL section is based on a conventional
The DTL
DTL section
section
is MeV.
basedFrom
on athis
a conventional
conventional
The
based
on
Alvarez
structure
up to is
40
energy, the
Alvarez
structure
up
to
40
MeV.
From
this
energy,the
the
Alvarez
upincreased,
to 40 MeV.
From the
thislongitudinal
energy,
focusingstructure
period is
keeping
focusing
period
is
increased,
keeping
the
longitudinal
focusing
period
is
increased,
keeping
the
longitudinal
phase advance below 65° to avoid emittance exchange.
phase
advancebelow
below
65°totoavoid
avoid
emittance
exchange.
A Cell-Coupled
DTL65°
design
(CCDTL
[9]) at exchange.
352
MHz
phase
advance
emittance
Cell-Coupled
DTLdesign
design
(CCDTLcoupling
[9])atat352
352
MHz
isCell-Coupled
adopted,
characterised
by identical
cells.
A
AA
DTL
(CCDTL
[9])
MHz
adopted,
characterised
by3)
identical
coupling
cells.
12-cell
cold
model (Fig.
has been
realised
and
isis
adopted,
characterised
by
identical
coupling
cells.
AA
12-cell
cold model
model (Fig.
(Fig. 3)3) has
has been
been realised
realised and
and
tested. cold
12-cell
tested.
tested.
double
FODO line
sections,
for transverse
matching
Thechopper
chopper
composed
two11m
mlong
long
The
line isiscomposed
ofoftwo
betweenFODO
a fast phase
advance
in transverse
the accelerators
and a
double
sections,
for
matching
double
FODO
sections, forplustransverse
matching
slow one
in phase
the chopper,
a 1.6
m long FODO
between
afast
fast
advanceininthe
the
accelerators
anda a
between
a
phase
advance
accelerators
and
period
in in
between.
The latter
houses
themchopper
inside
slow
one
the
chopper,
plus
a
1.6
long
FODO
slow one in the chopper, plus a 1.6m long FODO
the quadrupoles
and
the 90°
advance
at
period
between.The
Theprovides
latterhouses
houses
thephase
chopper
inside
period
ininbetween.
latter
the
chopper
inside
the
dump
placed
at
its
end,
needed
for
the
separation
thequadrupoles
quadrupolesand
andprovides
providesthe
the90°
90°phase
phaseadvance
advanceatat
the
of dump
chopped
and unchopped
beam.
the
placed
at
its
end,
needed
for
the
separation
the dump placed at its end, needed for the separation
of
chopped
and
unchopped
beam.
A
test
of
the
chopping
line
using
a
3
MeV
segment
of chopped and unchopped beam.
ofAthe
352
MHz
IPHI
RFQ,
presently
being
assembled,
test ofthe
thechopping
choppingline
lineusing
usinga a33MeV
MeVsegment
segment
A test
is
in ofpreparation
with CEA
and CNRS-IN2P3
of
the
352
MHz
IPHI
RFQ,
presently
being
assembled,
of the
352 MHz IPffl RFQ, presently being assembled,
(France).
preparation with
with CEA
CEA and
and CNRS-IN2P3
CNRS-IN2P3
isis inin preparation
(France).
(France).
FIGURE 3. 3D drawing of a fraction of the CCDTL cold
FIGURE 3. 3D drawingmodel.
of a fraction of the CCDTL cold
FIGURE 3. 3D drawing of a fraction of the CCDTL cold
model.
The advantages of this
structure are easy access and
model.
alignment
for
the
quadrupoles,
loware
construction
cost,
The
advantages
of
this
structure
easyaccess
access
and
The advantages
of this structure
are easy
and
stable
π/2
mode
operation,
continuous
focusing
lattice,
alignment
for
the
quadrupoles,
low
construction
cost,
alignment
forRFthedistribution
quadrupoles,
low
cost,
and simple
with
one construction
klystron
per lattice,
tank.
stable
π/2mode
modeoperation,
operation,continuous
continuous
focusing
stable
Ti/2
focusing
lattice,
However,
atRFthis
relatively with
low one
frequency,
realand
simpleRF
distribution
klystronthe
pertank.
tank.
and
simple
distribution
withCCDTL
one klystron
per
estate
shunt impedance
of the
remains
similar
However,
at
this
relatively
low
frequency,
the
realHowever,
this relatively
low frequency, the realto that of aatconventional
DTL.
estateshunt
shuntimpedance
impedanceofofthe
theCCDTL
CCDTLremains
remainssimilar
similar
estate
to
that
of
a
conventional
DTL.
to that of a conventional DTL.
97
Superconducting Linac
The superconducting part of the SPL begins at
120 MeV kinetic energy. Up to 383 MeV, multi-cell
cavities optimised for p=0.52, 0.7 are used. To ease
stabilization of the field in the cavities and minimize
the energy fluctuation of the beam, each cavity is
driven by its own tetrode amplifier. Above 383 MeV,
p=0.8 5-cell cavities are used. Unmodified LEP
cavities are no longer employed, even at the highest
energy. The additional cost is compensated by the
higher accelerating gradient and transit time factor of
the new cavities which allow to reduce the linac
length. Housed in LEP cryostats, these 5-cell p=0.8
cavities are prepared at minimal cost. Four of them are
driven by a single LEP klystron.
The beam dynamics design of the SC section is
optimised for minimum emittance exchange,
maximum stability against mismatch and simplified
layout for minimum cost [9]. The length of the
focusing periods, each containing a quadrupole
doublet of 120 mm aperture diameter, increases along
the linac to a maximum of eight cavities (two
cryostats) per period above 1.1 GeV. This corresponds
to 13 to 21 p?l per period and keeps the maximum
longitudinal phase advance below 65°. The relatively
low longitudinal phase advance allows the full current
tune ratio (Oi/at) to be kept below 0.8 thus avoiding
emittance exchange between the longitudinal and the
transverse planes [10]. At the same time, the
maximum transverse phase advance can be held below
85° to avoid particle lattice instabilities. A smooth
phase advance per metre in both planes ensures a
minimum mismatch in the transition areas between
sections, which are matched with existing beam line
elements. Mismatch simulations with 50 M particles
show only moderate emittance growth even for strong
initial mismatch (30 % radial) [9].
A recent study having underlined the difficulty to
properly control 4 cavities simultaneously [11], high
power phase and amplitude modulators are now felt to
be necessary to stabilize the field in each of the p=0.8
cavities individually (specifications in Table 3). The
ferrite loaded waveguides and their external bias
system are the only components that must be bought
(Figure 4). Prototypes have been ordered and the first
experimental results are expected by the end of 2002.
TABLE 3. Modulators specifications.___________
352.2 MHz
RF frequency
Peak power
350 kW
Phase modulation depth
+ 25 degrees
Amplitude modulation depth
+ 10%
Rise-time (10-90%)
1 ms
adjustable H
circuit
ferrite loaded /
wove guide
500mm long
Phase /Amplitude Modulator Type B
FIGURE 4. Phase/Amplitude modulator
STAGED REALISATION
A staged realisation is necessary to comply with the
limited resources available at CERN during the
construction of the LHC. Effort is concentrated on the
low energy part of the machine (up to a few MeV) in
close collaboration with the CEA and IN2P3 [6]. In a
second stage, the realisation of the 120 MeV room
temperature front-end is envisaged as an improved
injector for the PSB.
REFERENCES
1. R. Garoby, M. Vretenar, "Proposal for a 2 GeV Linac
Injector for the CERN PS", CERN PS/RF/Note 96-27.
2. M. Vretenar (editor), CERN 2000-012.
3. http://www.jganiLfr/eurisol/index.html
4. http://muonstoragerings.cern.ch/NuWorkshopQ2/
5. R. Garoby, CERN/PS 2001-055 (RF).
6. J.-M. Lagniel et al., "IPHI, the Saclay High-Intensity
Proton Injector Project", PAC 97, Vancouver, pp. 11201122.
7. F. Caspers, A. Mostacci, S. Kurennoy, CERN/PS 2002027 (RF).
8. M. Paoluzzi, CERN/PS 2002-026 (RF).
9. F. Gerigk, M. Vretenar, R.D. Ryne, "Design of the
Superconducting Section of the SPL Linac at CERN",
PAC 2001, Chicago, pp. 3909-3911.
10. F. Gerigk, I. Hofmann, "Beam Dynamics of NonEquipartitioned Beams in the case of the SPL Project at
CERN", PAC 2001, Chicago, pp. 2872-2874.
11. Tiickmantel, "Mathematical analysis of spontaneous
symmetry breaking in a multi cavity RF system with
vector sum feedback and Lorentz detuning", CERN-SLNote-2001-023.