Technological Challenges for the LHC Upgrade W. Scandale CERN Accelerator Technology Department

1
Technological Challenges for the LHC Upgrade
W. ScandaleCERN Accelerator Technology
Department
Thanks to the valuable contributions of D.
Tommasini
CERN-CARE Workshop HHH2004, 8 November 2004
2
Outlook

• Present context
• A road map for the upgrade of the LHC luminosity
• Technological challenges
• High-field superconducting magnets for the LHC-IR
  
• Medium-field fast-cycling superconducting magnets
  for the LHC-injector complex
• SPL and RCS
• Conclusive remarks

3
Present Context

• LHC in operation within about 30 months
• GSI program based on SIS100 and SIS300 approved
• EU-CARE activities settled
• HHH-network investigating
• Possible scenarios for LHC upgrade
• New concepts for Interaction Regions design
• Possible use of high-field and for medium-field
  fast-pulsed magnets
• NED-Joint Research Activity (NED-JRA) launching
• RD for high-field Nb3Sn superconducting wire
• New concepts for the design of high-field
  superconducting IR magnets
• HIPPI-Joint Research Activity (HIPPI-JRA)
  launching
• RD for high-intensity pulsed linear accelerators
• Optimization of up to 200 MeV Linac
• Beam dynamics and RF component design for Linac
  up to the GeV energy

4
A Road Map for the LHC Upgrade
See LHC Project Report 626

• Baseline hardware ultimate performance -gt
  Lmax 2.3x1034/cm2 s-1
• Ultimate bunch intensity -gt Ib 1.71011
  protons per bunch
• Requires RF batch compression in the PS or
  Linac4
• Two collision points (instead of four) with f
  315 mrad (instead of 300)
• Luminosity increase by reducing b -gt Lmax
  4.6x1034/cm2 s-1
• IR quadrupole upgrade (higher aperture - higher
  pole field) -gt b0.25 m
• larger crossing angle -gt f 445 mrad (Crab
  crossing RF-cavities?)
• Beam density increase and LHC turn-around upgrade
• RF upgrade for bunch compression in the LHC
• Super-PS and super-SPS injecting at 1TeV (first
  step for future LHC energy upgrade)
• Beam energy increase
• Higher field dipoles (14 T) and higher gradient
  quadrupoles (500 T/m)
• Mass production of a new superconductor (most
  likely Nb3Sn)

5
A Time-Window for LHC-IR Upgrade
Radiation damage limit 700 fb-1

• Due to the high radiation doses to which they
  will be submitted, the life expectancy of LHC IR
  quadrupole magnets is estimated 5-7 years
• IR-quadrupoles will have to be replaced in
  2013-2015, thereby offering an opportunity of
  upgrading LHC IR optics to improve luminosity

Courtesy of F. Ruggiero and Jim Strait

• Mid-2010s is also the earliest time frame when
  one can expect to need final-focusing quadrupole
  magnets for any of the proposed projects of
  linear colliders. At least one needs very strong
  wide final triplets

6
IR based on High Fields Magnets with reduced b
New Interaction Regions beam dynamics versus
magnet technology and design
See PAC03 pp 42-44
blue DIPOLES red QUADRUPOLES green RF-CAVITIES
7
RD needed for High Field Magnets

• SC Cable
• High performance SC cable aiming at a non-Cu JC
  up to 1500 A/mm2 _at_15 T at a temperature of 4.2 K
  or 1.9 K
• Insertion and magnet design
• Simultaneous optimization of optics and magnet
  design
• 15 T dipoles and 12 T - 100 mm quadrupoles of
  accelerator type (reasonable quench margin and
  good field region, easy to build)
• Particle loss hardness
• Upgrade of the Heat Transfer in SC cables
• Comparative study among 4.2 K and 1.9 K solutions
  (imposing the constraint of the LHC cryogenic
  plant)
• Upgrade simultaneously radiation hardness (cable
  insulators and coil design) and local collimator
  layout

8
SC conductor for High Field Magnets
See CARE-HHH-AMT workshop WAMS 22-24 March 2004
Archamps http//amt.web.cern.ch/amt/activities/wor
kshops/WAMS2004/WAMS2004_index.htm

• High Temperature Superconductors (HTS) are not
  yet ready for large-scale applications requiring
  high current densities under high magnetic
  fields. It will take at least another decade
  before they become competitive in terms of
  performances, yield and cost
• The upper critical field of MgB2 is too low
• Nb3Al exhibits promising properties but there are
  serious manufacturing issues that have yet to be
  resolved
• At present, the only serious candidate to succeed
  NbTi, suitable for industrial production, is the
  intermetallic compound Nb3Sn (world production
  still rather low 15 t/year).

RD on Nb3Sn conductor started in the frame of
CARE-NED
9
High Field Magnets recent results
A series of record-breaking dipole magnet models,
opening the 10-to-15 T field range (however, not
yet of accelerator class)
D20 (cosq)
13.5 T at 1.8 K in a 50-mm bore (LBNL, 1997)
MSUT (cosq)
RD-3 (Racetrack)
11 T on first quench at 4.4 K in a 50-mm-bore
(Twente University, 1995)
14.7 T at 4.2 K in a 25-mm gap (LBNL, 2001)
10
The poor man way LHC-IR upgrade with new NbTi
quadrupoles -gt b0.25 m
See EPAC 04 pp 608-10
The quadrupole aperture is matched to the real
beam size
Comparison between NbTi, NbTiTa and Nb3Sn
conductors
11
The EU Joint Research Activity CARE-NED

• The main objective of the NED JRA is to develop a
  large-aperture (more than 88 mm), high-field (up
  to 15 T) dipole magnet model relying on
  high-performance Nb3Sn conductors (non-Cu JC up
  to 1500 A/mm2 _at_15 T and 4.2 K).
• Such magnet is aimed at demonstrating the
  feasibility of the LHC-IR upgrade scenarios based
  on high field dipole and quadrupole magnets and
  is meant to complement the US-LARP.
• In addition, the NED model could be used to
  upgrade the CERN superconducting cable test
  facility (presently limited to 10-10.5 T).
• The NED JRA proposal involves 7 collaborators
  (CEA/Saclay, CERN, INFN-Milan and Genoa, RAL,
  Twente University and Wroclaw University), plus
  several industrial sub-contractors.
• EU funding limited to 25 of the original
  request -gt new resources needed soon to complete
  the program

12
De-scoping CARE-NED

• Given the present State of the Art and the magnet
  requirements foreseen for LHC IR upgrade and for
  IRs of future linear colliders, we established
  the following road-map
• revisit magnetic and mechanical designs to
  achieve enhanced performances with coils made
  from brittle conductors,
• address coil cooling issue under high beam
  losses,
• keep promoting high-performance Nb3Sn wire
  development (to ensure the survival of multiple
  suppliers including in EU),
• improve mechanical robustness and assess
  radiation hardness of Nb3Sn conductor insulation,
• put into practice all of the above in magnet
  models and prototypes.

13
Beam Density Increase
The upgrade of the injector chain is needed
Poor-man way
RF upgrade for batch compression in the PS

• Up to 160 MeV LINAC 4
• Up to 2.2 GeV the SPL (or a super-BPS)

See CARE-HIPPI
Rich-man way

• The superconducting way
• Up to 60 GeV a SC super-PS
• Up to 1 TeV a super SPS
• SC transfer lines to LHC

• The normal conducting way
• Up to 30 GeV a refurbished PS
• Up to 450 GeV a refurbished SPS

See CARE-HHH and CARE-NED

• A 1 TeV booster ring in the LHC tunnel may also
  be considered
• Easy magnets (super-ferric technology?)
• Difficult to cross the experimental area (a
  bypass needed?)

14
Low Energy Injector Upgrade LINAC4 SPL
see CERN-AB-2004-21
0.91014 particles at 2 Hz for the PS booster
2.31014 particles at 50 Hz for the PS
15
Upgrade of the Injector RingsBooster, PS and
SPS

• Basic investigations still needed
• Main constraints
• Use the existing tunnels
• Increase the beam density and the beam intensity
  possibly by a large factor
• Fast repetition rate to speed-up the LHC
  injection process
• Expected challenges
• Fast-cycling SC magnets
• Powerful RF within a limited space
• Cryogenic, vacuum
• Ejection optimization, loss control, beam
  disposal, instrumentation

16
Recent Activity on Fast Cycling Dipoles

• SIS 200 (abandonned)
• 4 T central field, 1 T/sec ramp
• Design based on RHIC dipoles
• Costeta, Rutherford cable
• One phase He cooling
• BNL model optimize to higher ramp-rate
• Wire twist pitch 4 mm instead of 13 mm
• Stabrite coating instead of no coating
• Stainless steel core (2x25 microns)
• G-11 wedges instead of copper wedges
• Thinner yoke laminations (0.5 mm instead of 6.35
  mm), 3.5 silicon, glued with epoxy.

Cable inner edge
Courtesy A.Ghosh and P.Wanderer
17
The BNL Fast Cycling Dipole Model
Cross section of GSI-001 Prototype Magnet
Courtesy A.Ghosh
18
The SIS 300 Fast Cycling Dipole Model
Coil
Courtesy of G.Moritz

• SIS 300
• 6 T, 1 T/sec ramp, 100 mm bore
• Design based on UNK dipoles, bore from 80 mm to
  100 mm
• 2-layers Cos?, Rutherford cable
• One phase He cooling

Collars
Key
Iron yoke
Shell

• Challenges
• high operational field for 4.2 K, pulsed, high
  losses
• Activity on cable development
• Reduction of conductor AC loss adjusting filament
  hysteresis, strand matrix coupling current, cable
  crossover resistance Rc, and adjacent resistance
  Ra.
• A 3.5 micrometer filament diameter was chosen
  because it appears to be the minimum value that
  can be reached in a standard copper matrix strand
  without the onset of proximity coupling.
• The use of a Cu-0.5 Mn as an interfilamentary
  matrix material is also under consideration, to
  reduce both matrix coupling current losses (due
  to the high resistivity of CuMn ) and hysteresis
  losses.

C-Clamp
Staples
19
Cables for Fast Pulsed Dipoles
A.D. Kovalenko, JINR, 2004
20
RD Still Needed(a non-exhaustive list)

• Lowering losses in pulsed magnets
• Industrialize filament size 3.5 microns or
  smaller, reduce twist pitch
• Electromagnetic design for minimum amount of
  superconductor
• Optimize cable (cable size, keystone angle,
  number of strands)
• Cored cables and strands with resistive coating
  
• long term behaviour issues
• investigate limits of high Ra/Rc keeping
  acceptable current sharing
• Resistive matrix
• Alternatives to Rutherford cables, such as
  Nuclotron and CICC

• Other issues
• Thermal modelling of magnet cross section under
  helium flow
• Characterization of cable insulation schemes
  (dielectric/mechanical/thermal)
• Manufacture of a small scale prototype for
  thermal model/parameter validation, for cable
  testing/characterization, and as coil test
  facility
• Manufacture of an optimized prototype to prepare
  series production
• Field quality during the ramp modellization and
  experiments
• Develop dedicated magnetic measurement systems
  for fast varying magnetic fields

21
Pulsed Dipoles for PS and SPS?
Initial considerations based on known technology

• Upgraded PS and SPS may require two different
  types of pulsed magnets
• 3T 2T/s for the PS
• 5T 1.5 T/s for the SPS
• The quench limit performance ican be achieved
  with present technology
• Modified RHIC dipoles or Nuclotron/CICC cable
  based dipoles for PS
• Modified (lower losses)  SIS 300  type dipoles
  for the SPS

22
Technological Challenges

• Losses are a major concern -gt Vigorous RD
  program needed
• Study and evaluate different scenarios of beam
  losses in PS and SPS
• Study and evaluate a maximum allowed cryogenic
  budget
• Optimize the dipoles not only for good quench
  performance in condition of cable/iron losses,
  but also for cryogenic budget

• A SC dipole for the SPS may produce 70 W/m peak
  (35 W/m effective ? 140 kW for the SPS,
  equivalent to the cryogenic power of the LHC !)
• A rather arbitrary guess for beam loss is of
  about 1012px100GeV/10s 15 kW
• By dedicated RD magnet losses should be lowered
  to 10 W/m peak (5 W/m effective ? 20 kW ),
  comparable to expected beam loss power

Tentative SPS cycle
23
What about High Power Beams ?
seeH.Schonauer EPAC 2000 pp966-68

• High power beams what for?
• Improve LHC beam (yet to be seen)
• High flux of POT for hadron physics
• Feed n-factory

Main Ring Cycle
24
Possible parameters
see H. Schonauer, April 03
25
Technological Challenges in a 30 GeV RCS
see H. Schonauer, April 03

• Lattice and beam dynamics
• High gt needed but difficult to have
  dispersion-free SS at the same time
• Constraint on ?1 together with x0 and large
  dynamic aperture
• Potential coupled bunch instability during the
  long injection plateau
• RF
• Large RF voltage needed, but little space for
  RF-cavity in dispersion-free SS
• Injection capture in an accelerating bucket not
  truly an adiabatic process
• Demanding HOM damper
• Difficult adiabatic bunch compression at 30 GeV
  (too low synchrotron fr.)
• Capture loss versus injection energy
• Vacuum pipe and bean surroundings
• Large shielded ceramic chamber
• Tight impedance budget Z/N lt 2 ohms critical
• Dipoles and power supplies
• Large stored energy (some hundreds of kJ per
  dipole)
• Fast power supplies

26
Conclusion

• A staged roadmap for the LHC luminosity upgrade
  needs RD on
• High-field (up to 15 T) superconducting cables
  and magnets
• Powerful and sophisticated RF devices for beam
  manipulations
• Medium-field fast-pulsed superconducting cables
  and magnets
• Accelerator design and integration to existing
  constraints
• Upgrading LHC complex is a unique opportunity to
• Share technological developments with other
  communities such as
• Fusion (EFDA)
• Nuclear physics (GSI)
• NMR developers
• Boost the CERN accelerator complex for future
  applications such as
• High intensity hadron and neutrino physics at
  intermediate energy
• Injector developments for neutrino factory
• Initial resources for RD are presently provided
  by EU and CERN within the frame of CARE, in
  particular within the HHH-network and in the NED
  and the HIPPI JRAs (most likely, more support
  will be needed soon)

27
Thank-you for your attention