Upgrades to the
ISIS Facility
John Thomason
ISIS Accelerator Division
ISIS Accelerators
• H ion source (17 kV)
• 665 kV H RFQ
• 70 MeV H linac
• 800 MeV proton
• Extracted proton
beam lines
The accelerator produces a
pulsed beam of 800 MeV
(84% speed of light) protons
at 50 Hz, average beam current
is 230 A (2.9× 1013 ppp) therefore
184 kW on target (148 kW to TS-1 at 40
pps, 36 kW to TS-2 at 10 pps).
ISIS Upgrades
• Present operations for two target stations
Operational Intensities: 220 – 230 μA (185 kW)
Experimental Intensities of 31013 ppp (equiv. 240 μA)
DHRF operating well: High Intensity & Low Loss
Now looking at overall high intensity optimisation
• Study ISIS upgrade scenarios
0) Linac and TS1 refurbishment
1) Linac upgrade leading to ~0.5 MW operations on TS1
2) ~3.3 GeV booster synchrotron: MW Target
3) 800 MeV direct injections to booster synchrotron: 2 – 5 MW Target
4) Upgrade 3) + long pulse mode option
ISIS MW Upgrade Scenarios
1) Replace ISIS linac with
a new ≈ 180 MeV linac
(≈ 0.5MW)
2) Based on a ≈ 3.3 GeV
RCS fed by bucket-to-bucket
transfer from ISIS 800 MeV
synchrotron (1MW, perhaps
3) RCS design also
accommodates multi-turn
charge exchange injection to
facilitate a further upgrade
path where the RCS is fed
directly from an 800 MeV
linac (2 – 5 MW)
Power / Benefit / Cost
Upgraded TS1
£ + Risk
Existing TS1
ISIS Upgrades, Developments and R&D Work
• We have on-going research and studies to
develop and fully exploit the machine
map out the best development routes
define principle upgrades
undertake basic R&D into physics of high intensity beams
• Main focus presently ~180 MeV Injector Upgrade
summarised in the following pages
holistic optimisation including targets, neutronics, … “at the user”
• Next steps
Exploring the possibilities for optimistic & less optimistic funding scenarios
Mapping out the best options for a 1-2 MW short pulse neutron source
Development and research on present machine
ISIS Injection Upgrade
• A New 180 MeV Injector
Update old linac
Increase beam power ~0.5 MW
70 MeV Linac
800 MeV
• Advantages
Reduces Space Charge (factor 2.6)
Qinc 
rp N
2   B
2 3
Chopped, Optimised Injection & Trapping
• Challenges
Injection straight
Activation (180 MeV)
Space charge, beam stability, ....
ISIS Injection Upgrade Ring Physics Study
• Snapshots of the work: challenges of getting 0.5 MW in the ISIS Ring
Injection Straight Modelling
Injection Straight
Longitudinal Dynamics
Simulation Results
Test Distribution
Analytical Work
Foil temperatures
Injected distributions in (x,x’),(y,y’),(,dE)
Transverse & Full Cycle 3D Dynamics
Predicted Space Charge Limit
Single particle tune shift
distributions at 0.5 MW
RF Bucket
Evolution of bunch
Variation of key parameters
Other Essentials: Activation, Diagnostics
Activation vs Energy
Activation Measurements
Coherent Tune Shift and Resonance
Electron Cloud Monitor
Accelerated distributions in (x,x’),(y,y’),(,dE)
Strip-line Monitor/Kicker
Possible ≈ 3.3 GeV RCS Rings
Bucket-to-Bucket Transfer
5SP RCS Ring
0.8 – 3.2 GeV
Rep Rate
50 Hz
C, R/R0
367.6 m, 9/4
frf sweep
6.1-7.1 MHz
Peak Vrf
≈ 750 kV
Peak Ksc
≈ 0.1
εl per bunch
≈ 1.5 eV s
Grahame Rees,
Ciprian Plostinar (
800 MeV, Hˉ Linac Design Parameters
Ion Species
Output Energy
Accelerating Structures
Beam Current
Repetition Rate
Pulse Length
Duty Cycle
Average Beam Power
Total Linac Length
H800 MeV
DTL/SC Elliptical Cavities
324/648 MHz
43 mA
30 Hz (Upgradeable to 50 )
0.75 ms
2.25 %
0.5 MW
243 m
Design Options
Capacity upgrade scenarios
“Traditional” 3-stage MW upgrade scenario
could be extended so 3.2 GeV RCS includes
multiple extraction straights (or switchyard in
EPB), with or without 800 MeV linac.
Stacked rings (as at CERN PSB) could
be implemented as part of AC magnet
replacement programme. Would
require increased linac performance,
but otherwise it is an engineering
challenge to minimise off time during
installation rather than an accelerator
physics challenge, and would be a
very predictable upgrade.
One synchrotron with several extraction straights?
Target station #1
Target station #2
“Efficient” footprint
Maximises total
number of neutron
beam lines
Easy extraction
of proton beams
of different
intensities and
repetition rates
to suit wide
range of neutron
Would need to drive trim
quads. and steerers
differently for different
energies and intensities,
but trim quads. and
steerers are pulsed
anyway, and so changing
trim magnet current
profiles from acceleration
cycle to acceleration cycle
should raise no
Target station #4
Target station #3
Ring High Intensity Beam Studies on ISIS
• Some of our R&D Studies
Half-integer intensity limit in proton rings
Using the ISIS ring to study halo formation
Y profile
Y profile
Higher order loss effects and images
Investigating complex loss mechanisms
Loss vs Q
Model losses, benchmark on ISIS
Image driven resonance
Head-tail instability
Key for high intensity proton rings
Vertical dipole
motion along
bunch on
successive turns
Vertical difference signal
(along bunch, many turns)
Samples along bunch
New simulation code: Set 3Di
Turn 
Necessary Hardware R&D
To realise ISIS upgrades and generic high power proton driver development, common
hardware R&D will be necessary in key areas:
• High power front end (FETS)
• RF Systems
• Stripping Foils
• Diagnostics
• Targets
• Kickers
• etc.
• In the neutron factory context SNS and J-PARC are currently dealing with
many of these issues during facility commissioning and we have a watching
brief for all of these
• Active programmes in some specific areas

similar documents