Harms_20140409LCLSmeeting

Report
3.9 GHz Harmonic Linearizer
Elvin Harms
LCLS-II Accelerator Physics Meeting
9 April 2014
Introduction
•
•
•
•
•
ACC39 at FLASH
Proposal for LCLS-II LH
Parameters
Assumptions
Summary
2
Harms, LCLS-II Accelerator Physics 9 Apr 2014
The Linearizer module - What is it, What does it do?
• The 3.9 GHz module, ACC39, is installed in the DESY FLASH injector
just after the 1.3GHz ACC1 (first) cryo module.
• It is used in conjunction with ACC1 in order linearize the bunch energy
vs. time over the bunch length.
• This in turn makes “bunch compression” to very short bunches with
high peak currents more efficient,
a more controlled longer bunch
charge distribution.
3
3
Harms, LCLS-II Accelerator Physics 9 Apr 2014
Bunch Compression with 3.9 GHz Module
After acceleration w/o
After compression w/o
Short bunch
Flat bunch
~250 fs 
From M Dohlus et
al
With 3rd harmonic
4
4
From P Piot
Harms, LCLS-II Accelerator Physics 9 Apr 2014
Test Results
• Entire cryomodule was
first tested at DESY in
their CryoModule Test
Bench (CMTB) prior to
installation in FLASH
– Successful assembly
– Successful transatlantic
shipment
– Some rework at DESY
• Longitudinal realignment
• Instrumentation
terminations
5
5
Harms, LCLS-II Accelerator Physics 9 Apr 2014
ACC39 Performance
• Set up for lasing has gone from expert tune up (shifts) to nearly
immediate after turn-on
6
6
Harms, LCLS-II Accelerator Physics 9 Apr 2014
ACC39 Performance
•
•
ACC39 routinely operates at 18.9 to 19.7 MV/m
Capable of operation at 22 MV/m
– Limitation set by thermal interlocks – concern about compromising HOM’s
on Cavities 3 & 5 (trimmed 2-post style)
• Amplitude stability ≤ 2x10-5 pulse-to-pulse
• Phase stability ≤ 0.003° pulse-to-pulse
7
7
Harms, LCLS-II Accelerator Physics 9 Apr 2014
Experiences & Lessons Learned (and applied)
•
New infrastructure had to be built
–
Commissioning in parallel with qualifying components
•
•
•
•
•
•
HOM redesign
–
–
–
•
8
Prototyping and testing is vital before placing order
Engineering Notes & Operational Readiness Clearance
Transatlantic Transport of a complete cryomodule
Welding certification
–
•
Not simply a scaled version
Modeling and prototyping important
Robust design is the end result
HOM coupler antennae
–
•
•
•
Vertical test stand
Horizontal Test Stand
Coupler Conditioning
Clean rooms
…
Especially Helium vessels
‘New’ technologies explored and applied
Harms, LCLS-II Accelerator Physics 9 Apr 2014
Looking Ahead – CW Operation
• Heat Load/Cryogenic piping
– Limit total cryomodule heat load to <250 watts
– Dynamic Heat Load needs to be revisited
• Estimated heat load ~15 watts/cavity @ 15 MV/m
• Input coupler power capability/cooling
– Current design should be good to ~2kW cw, but needs to
be re-analyzed
• Studies are possible using coupler stand at A0 with a cw
amplifier
– Is a variable coupler necessary rather than current ‘fixed’
design
9
Harms, LCLS-II Accelerator Physics 9 Apr 2014
Looking Ahead – CW Operation (2)
• HOM coupler and feedthrough capability
– Current (single-post) ‘Formteil’ design
• multi-pacting lower bound ~25 MV/m
– Feedthrough – adequate cooling?
– Simulation needed to look at modes (THz?) present in the ends
• Absorbing material needed in beam pipe?
• Preliminary simulations begun
• Microphonics/Fast tuning
– Current tuner is slow (blade) tuner only
– Cavities are already fairly stiff – resistant to microphonics/LFD
– Consider addition of fast/piezo’s or passive microphonics
compensation
10
Harms, LCLS-II Accelerator Physics 9 Apr 2014
Looking Ahead – CW Operation (3)
• Cavity Global thermal limit is estimated to be of order
25 MV/m
• Alignment
– Offset = HOM power considerations
– Tilt = transverse emittance implications
• Other issues to consider
– Beam diagnostics, especially internal to module
– Quads, dipoles
– Cryogenic and vacuum interconnectability to adjacent
components
– Standalone vs. Integrated module
11
Harms, LCLS-II Accelerator Physics 9 Apr 2014
LCLS-II Layout
P. Emma, L. Wang,
C. Papadopoulos
M. Ross
12
E. Harms
Harms,
LCLS-II Accelerator
9 Apr2014
2014
APT
Seminar
14Physics
January
Proposed Parameter Set
13
E. Harms
Harms,
LCLS-II Accelerator
9 Apr2014
2014
APT
Seminar
14Physics
January
3.9 GHz Section Assumptions - 1
•
Cryomodule
–
–
–
–
–
–
–
8 cavities per Cryomodule with 2 Cryomodules
Total heat load ~200 W / cryomodule
No magnet, corrector, or BPM’s in 3.9GHz CM’s
Magnets/Quad/BPM will be downstream in warm section
Cavities spaced for Q0 > 2e9
Beam pipe tapering from 1.3 GHz CM to 3.9 GHz CM
Length of interconnection (distance between CM’s) is free
• no requirements in term of wavelength.
– Larger diameter 2-phase pipe/nozzle akin to 1.3 GHz
• will provide additional cooling than existing design
– Cryo interface will be CM2/3 upstream and 'standard' feedbox immediately downstream
– Design of cryomodule will strive to bring all signal cables out on the aisle side
– Cryomodule will be cooled with the proposed 2nd cryoplant capable of ~1kW at 2K
along with the other upstream (non-main Linac) CM’s
– Cavity-to cavity distance is short (as in FLASH)
– HOM (beam pipe) absorber likely placed between two 3.9 GHz CM’s
14
E. Harms
Harms,
LCLS-II Accelerator
9 Apr2014
2014
APT
Seminar
14Physics
January
3.9 GHz Section Assumptions - 2
•
Cavities
– Capable of 14.5 MV/m cw
– Power dissipation ~17 W / cavity, maximum HOM power ~13 W / CM or 1.6 W / cavity
• (>50% will dissipated at 70K or RT)
– Blade tuner design with 3 piezos incorporated.
– HOM coupler signals may be used for beam offset measurements
– Consider modest Q0 improvement development program (N2 doping), but low priority
and is not project critical
• 1- and 9-cell cavities are available for processing & testing
15
E. Harms
Harms,
LCLS-II Accelerator
9 Apr2014
2014
APT
Seminar
14Physics
January
3.9 GHz Section Assumptions - 3
•
Couplers
– Cavity orientation for placement downstream and every second rotated by 180°
– Dig tunnel wall if necessary to make room for coupler clearance
• “not out of the question”
– Waveguide from back side of CM will be pre-assembled and extend to top of CM for
connection
– Two options for klystron to CM feed: Waveguide or coax
• WG option is preferable
• coax is lossier, possible center conductor heating
– Coupler will require modifications for CW regime
– Fixed coupling (Q~1.e7) looks OK.
16
E. Harms
Harms,
LCLS-II Accelerator
9 Apr2014
2014
APT
Seminar
14Physics
January
Coupler Orientation Options
Fermilab Cryomodule Layout:
Arun Saini, N. Solyak
XFEL Cryomodule Layout:
Start here
FLASH Cryomodule Layout:
17
E. Harms
Harms,
LCLS-II Accelerator
9 Apr2014
2014
APT
Seminar
14Physics
January
Goal
1.30 GHz
Coupler Orientation Summary
•
Beam optics through harmonic section is studied.
•
•
•
•
•
Fermilab configuration
XFEL Configuration
ILC type cryomodule
Start here
Goal
1.30 GHz
Coupler RF kick results in significant emittance growth.
Local compensation of coupler kick is possible and emittance
dilution can be minimized using different orientations of
alternative cavity.
•
XFEL configuration provides best solution in terms of emittance
compensation.
Arun Saini, N. Solyak
18
E. Harms
Harms,
LCLS-II Accelerator
9 Apr2014
2014
APT
Seminar
14Physics
January
3.9 GHz Section Assumptions - 4
•
Testing
– Test cavities in VTS and HTS (subset)
• maybe ANL/Ostroumov team will do horizontal testing
– Cryomodules will be cooled down and tested prior to transport to
SLAC
– 3 kW cw 3.9 GHz in hand – needs to be made operational
– 3.9 GHz Infrastructure exists (or can be modified) to do most testing
• cavities
• couplers
• cryomodules
19
E. Harms
Harms,
LCLS-II Accelerator
9 Apr2014
2014
APT
Seminar
14Physics
January
3.9 GHz Section Actions & Dates
•
Actions
– Will try to start from XFEL CM design --> (Elvin and Tom)
– May want to run cavities in pairs to compensate coupler kicks - not
clear that this is necessary (Arun to evaluate)
– Andrei is evaluating wakefields but expect them to be small
– Need to specify alignment tolerances (Arun to evaluate)
– Need to test new 3.9 GHz FPC in CW operation --> (need schedule)
– In addition to testing FPC new design, task force needs to be formed
to look at and review design
•
Key Dates
– 1 June 2014: FRS submitted
– 6 October 2014: Design Review
20
E. Harms
Harms,
LCLS-II Accelerator
9 Apr2014
2014
APT
Seminar
14Physics
January
Summary
•
•
•
•
•
•
•
3.9 GHz (pulsed) operating at FLASH
XFEL building a similar module
Technology can be applied for LCLS-II
Issues related to cw operation identified - manageable
Parameter list compiled – needs to be verified
List of assumptions and actions for LCLS-II LH compiled
Simulation/modeling work begun
– Coupler design and orientation
• Effort needs further formalization
• Do as much work as possible now
21
Harms, LCLS-II Accelerator Physics 9 Apr 2014
Thanks
22
Harms, LCLS-II Accelerator Physics 9 Apr 2014

similar documents