Download presentation
Presentation is loading. Please wait.
Published byAubrey Lane Modified over 9 years ago
1
Making the ATF into an ERL-FFAG May 27, 2015Stephen Brooks, eRHIC R&D retreat1 In relation to eRHIC risks addressed
2
Approximate Site Layout (Dejan) May 27, 2015Stephen Brooks, eRHIC R&D retreat2 Approx. 30m Up to 150MeV injector linac Up to 300MeV ERL linac Use the ATF linac repeatedly e.g. 4 times to obtain higher energies and test ERL-FFAG principle
3
Scaling type (by factor a) LengthAngleDipoleGradient Quad offset (=dipole/grad) & orbit excursion Momentum (~energy) 11aa1 Machine radiusa1a -1 a -2 a FFAG beta length (fixed bend radius, fixed cell tune) aa1a -2 a2a2 FFAG arc-to-straight (=row 3/row 2) 1aa1a FFAG radius with fixed orbit excursion and field (row 2*row 1/sqrt(row 3)) a 1/2 a -1/2 111 FFAG scaling* laws * not scaling FFAG laws May 27, 2015Stephen Brooks, eRHIC R&D retreat3 I’m going to use this one and the optimised 10.5m diameter C cell as a start.
4
C -like Cell (can use iron) May 27, 2015Stephen Brooks, eRHIC R&D retreat4 50 MeV injector200 MeV linac Energies (MeV)250, 450, 650, 850 Bending diameter30 m Orbit max range32.8 mm Max field on orbit0.732 T Quad gradientsQF: -42.54 T/m BD: 27.493 T/m Cell length55.76 cm TOF range (rel.)1.402 × 10 -3 cm grid Can extract 1.05 GeV in non-ERL mode
5
Superimposed on ATF area May 27, 2015Stephen Brooks, eRHIC R&D retreat5 Racetrack with matching section 84 cells per half turn Transverse exaggeration x128
6
Magnet Options May 27, 2015Stephen Brooks, eRHIC R&D retreat6 PM Iron Holger Witte’s quadrupole using iron poles Pure permanent magnet with rotatable rods (BD shown), or Halbach quad
7
High-field Cell Option May 27, 2015Stephen Brooks, eRHIC R&D retreat7 150 MeV injector300 MeV linac Energies (MeV)450, 750 … 1950 Bending diameter30 m Orbit max range32.7 mm Max field on orbit1.673 T Quad gradientsQF: -97.59 T/m BD: 63.07 T/m Cell length55.76 cm TOF range (rel.)1.403 × 10 -3 cm grid Can extract 2.25 GeV in non-ERL mode
8
Achievable with Halbach Magnets? No The B=B r ln(R out /R inner ) law is only for dipoles! – It was tempting to think we could get arbitrarily high-field quads using the log term As our FFAG magnets are predominantly quadrupoles, the ~1.6T pole tip fields do not look achievable using permanent magnets – Or at least that is what my code is telling me May 27, 2015Stephen Brooks, eRHIC R&D retreat8
9
Cost/Components Scaling The 850MeV 4-turn ATF design uses exactly the same magnet cross-sections as C – Beamline length increased by 30/10.5 = 2.86x – Also applies to PM volume, as packing factor same Rough cost scaling: C is $5-10M/10m diam. FFAG, eRHIC is $50-100M/1km, so sqrt law – Expect $8-17M for ATF FFAG loop Cheaper “just a line of magnets” expt possible May 27, 2015Stephen Brooks, eRHIC R&D retreat9
10
ATF Normal-Conducting Linac ATF linac cannot run CW – Maximum bunch train length so far: 3us (~6 turns) – 10us might be possible (~20 turns) – Limited by RF modulators (Marcus Babzien) – Need 8+ turns to establish multi-turn ERL pattern! Can have up to 100 bunches in train period Bunch length 100fs – 6ps (0.03–1.8mm) – No problems with charge, size etc. May 27, 2015Stephen Brooks, eRHIC R&D retreat10
11
My View of eRHIC Risk 3 chained amplifiers for errors and noise Messy, real-world inputs (surprise potential) May 27, 2015Stephen Brooks, eRHIC R&D retreat11 Permanent magnets Beam dynamics High-Q cavities Field errors Beam position errors E field kicks Alignment, construction and material errors, temperature dependence Ground vibration spectrum, BPM/corrector noise, halo from source, pipe impedances Correctors
12
Commonality of ATF-FFAG with C The linac and RF are already available – Major cost advantage over “green field” machine First ever ERL-FFAG First ever PM-based FFAG (of this size anyway) Record NS-FFAG energy range of ~4x – Similar to eRHIC FFAG1, similar tune range, similar orbit excursion, similar chromaticity per cell Real, complete tests of single particle optics May 27, 2015Stephen Brooks, eRHIC R&D retreat12
13
Differences of ATF-FFAG with C Main: ATF is not a CW superconducting linac – No high-Q cavity coupled to the system Can’t study BBU or any other modes coupled via cavity – Energy recovery unlikely to reach equilibrium state ATF-FFAG energy is higher (850 vs. 286MeV) – Might be able to measure synchrotron emission ATF FFAG cells are longer by 1.69x – Helps with engineering, less magnet interference May 27, 2015Stephen Brooks, eRHIC R&D retreat13
14
Funding Profile Risk May 27, 2015Stephen Brooks, eRHIC R&D retreat14 CD-0 Now Construction CD-1 $100Ms/year $10Ms/year on-project R&D But not enough time $1Ms/year C-AD R&D (not enough) CD-3CD-4 Spacing of CDs based on NSLS-II
15
Conclusion(?) This is really for group to decide ERL-FFAGs are a new class of accelerator – C=3.8km is quite large for the prototype – Building a small one first seems a sane approach C addresses more issues for eRHIC only ATF-FFAG has a great deal of synergy with eRHIC and might choose it if desire for ATF upgrade explicitly weighted into consideration May 27, 2015Stephen Brooks, eRHIC R&D retreat15
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.