Progress on the MICE RFCC Module

1
Progress on the MICE RFCC Module
MuCool Test Area RF Workshop October 15, 2008

• Mike Zisman/Steve Virostek
• Lawrence Berkeley National Laboratory

2
Progress Summary

• The MICE cavity design is heavily based on the
  successful MuCool 201-MHz prototype RF cavity and
  lessons learned
• Fabrication and post processing
• Cavity conditioning and operation
• Engineering design of the cavity is complete
• CAD model of the cavity, tuners, support and
  vacuum
• Fabrication schemes and vendor identification
• RFCC module engineering design nearing completion
• Possible operation at LN temperature?
• Design accommodates LN operation, but is very
  challenging

3
Muon Ionization Cooling Experiment
RFCC
RFCC
Spectrometer Solenoid 1
AFC
AFC
Spectrometer Solenoid 2
Absorber and Focusing Coil (AFC)
4
2 RFCC Modules, 8 Cavities
5
RF Cavity Summary

• Fabrication of the prototype cavity was
  successful
• A slight reduction in cavity diameter to raise
  the frequency has been specified and analyzed
• The fabrication techniques used to produce the
  prototype will be used to fabricate the MICE RF
  cavities
• A detailed WBS schedule for the design and
  fabrication of the MICE RF cavities has been
  developed

6
Cavity Design Parameters

• Cavity design parameters
• Frequency 201.25 MHz
• ß 0.87
• Shunt impedance (VT2/P) 22 M?/m
• Quality factor (Q0) 53,500
• Be window diameter and thickness 42-cm and
  0.38-mm
• Nominal parameters for MICE and cooling channels
  in a neutrino factory
• 8 MV/m (16 MV/m) peak accelerating field
• Peak input RF power 1 MW (4.6 MW) per cavity
• Average power dissipation per cavity 1 kW (8.4
  kW)
• Average power dissipation per Be window 12 watts
  (100 watts)

7
Recent Progress RF Cavity Design

• 3-D Microwave Studio RF parameterized model with
  ports and curved Be windows
• Hard to reach the design frequency by spinning
• Frequencies between cavities should be able to
  achieve within ? 100 kHz
• Approaches
• Modification the spinning form
• Targeting for higher frequency
• Fixed tuner to tune cavity close to design
  frequency (deformation of cavity body)
• Tuners are in push-in mode ? lower frequency

8
Numerical Study with B Field

• Preliminary studies, in collaboration with
  colleagues at SLAC using Omega-3P and Track-3P
  codes
• Cavity with flat windows 5 MV/m on axis 2-T
  uniform external magnetic field scan of a few
  points from one cavity side

Trajectories with external B 2-T field
Trajectories without external B field
E field contour
9
High Power RF Tests on Prototype Cavity

• The cavity was first tested with flat copper
  windows and reached 16 MV/m quickly and quietly
  
• The cavity then was tested with thin and curved
  Be windows and again reached to 19 MV/m quickly
• Cavity frequency had to be retuned
• Cavity frequency was stable during the operation,
  however, we did observe frequency shift due to RF
  heating on the windows
• Frequency shift of 125 kHz (from 0 to 19
  MV/m, 150-micro-second pulse, 10-Hz repetition
  rate) in 10 minutes, well within the tuning
  range (230 kHz/mm per side, ? 2-mm range)
• With a few hundred Gauss stray field from Lab-G
  magnet, the cavity gradient can be maintained at
  19 MV/m
• To test with stronger external magnetic fields
• Move the cavity closer to Lab-G magnet
• SC coupling coil for MuCool

10
Tests with Stronger Magnetic Fields

• New vacuum pumping system
• Separation of the nearest curved Be window from
  the face of Lab-G magnet is 10-cm (before was
  110-cm)
• Maximum magnetic field near the Be window 1.5
  Tesla (at 5 Tesla in magnet)
• Test Results
• Multipactoring was observed over the entire
  magnetic field range up to 1.1-T at nearest Be
  window
• A strong correlation exists between cavity vacuum
  and radiation levels
• We have achieved 14 MV/m at 0.75-T to the
  nearest curved thin Be window
• Recent test results show plateau at 10 MV/m

11
RF Cavity Assembly
12
RF Cavity Fabrication Overview
The same fabrication techniques used to make the
prototype cavity will be used for the MICE RF
cavities
Engineering CAD Model of the RF Cavity

• Cavity half-shells to be formed by metal spinning
• Precision milling of large parts (1.2 m diameter)
  
• Precision turning of mid-sized parts (0.6 m dia.)
• Precision manufacture of smaller parts
• CMM measurements throughout the process
• To limit any annealing and maintain cavity
  strength, e-beam welding will be used for all
  cavity welding
• cavity equator, stiffener rings, nose rings, port
  annealing and port flanges
• Cavity inside surfaces are finished by
  electro-polishing

Prototype RF Cavity
13
Recent Progress

• RFCC engineering CAD model refined
• RF and engineering design of 201-MHz RF cavity
  complete
• Integration and interface issues addressed
• Vendor identification and qualification near
  complete

14
Recent Progress (procurement fab)

• The large copper sheets used in the fabrication
  of the cavity shells have been ordered and will
  arrive at LBNL in mid-December
• A series of vendor qualification visits has been
  conducted
• Applied Fusion - San Leandro, CA (e-beam welding,
  machining)
• Meyer Tool Mfg., Inc. - Chicago, IL (machining)
• Sciaky, Inc. - Chicago, IL (e-beam welding)
• Roark Welding Engineering - Indianapolis, IN
  (e-beam welding, machining)
• ACME Metal Spinning Minneapolis, MN (cavity
  shell spinning)
• Other vendors have been identified
• Midwest Metal Spinning, Inc. Bedford, IN (cavity
  shell spinning)

15
Other Module Components

• Beryllium window design is complete windows are
  in the process of being ordered (8 per module
  needed)
• Design and analysis of the cavity frequency
  tuners is complete, drawings to be done soon
• A hexapod cavity suspension system has been
  incorporated in the design
• The RF coupler will be based on the SNS design
  using the off the shelf Toshiba RF window
• The vacuum system includes an annular feature
  coupling the inside and the outside of the cavity
  (further analysis of vacuum rupture scenarios
  TBD)

16
Other Module Components
Cavity Suspension
Dynamic Tuners
RF Coupler
Beryllium Window
Vacuum System
17
201 MHz Beryllium Windows

• Each cavity has two Be windows
• 42-cm diameter and 0.38-mm thick
• Window is formed at high temperature and later
  brazed to copper frames
• Thin TiN coatings on both sides of the window
• One window curves into the cavity and one curves
  out
• Already tested up to 5 MW in 201-MHz cavity at
  MTA, FNAL

42-cm
18
Cavity Tuner Design Features

• Six tuners spaced evenly around each cavity
  provide individual frequency adjustment through a
  feedback loop
• Layout is offset by 15º from vertical to avoid
  conflict with cavity ports
• Tuners touch cavity and apply loads only at the
  stiffener rings
• Tuners operate in push mode only (i.e.
  squeezing)

19
Tuner component Details
Actuator bellows assembly
Fixed arm
Pivoting arm
Forces are transmitted to the stiffener ring by
means of push loads applied to the tuner lever
arms by the actuator assembly
20
Hexapod Strut Arrangement

• Analysis of a hexapod strut system is complete
• Each cavity will contain a dedicated set of 6
  suspension struts arranged in a hexapod type
  formation
• This system spreads the gravity load of the
  cavity across several struts

Example of a hexapod stage
21
Hexapod Strut Cavity Mounting

• Copper mounting block will be e-beam welded
  directly to the RF cavity
• The cavity has very little deformation at the
  mounting block location

22
Vacuum System

• A NEG pump has been chosen because it will be
  unaffected by the large magnetic field
• A vacuum path between the inside and outside of
  the cavity eliminates the risk of high pressure
  differentials and the possible rupture of the
  thin beryllium window

23
Vacuum Vessel and Support Summary

• Engineering 3D CAD model of the vacuum vessel
  mechanical design is complete
• Standard machining and manufacturing methods will
  be used
• A plan for attaching the coupling coil and the
  vacuum vessel together has been developed
• Conceptual design of the support stand is
  complete (analysis will need to be performed)
• A method for assembling the cavities into the
  vacuum vessel has been formulated
• A conceptual design of the cavity water cooling
  feedthrough system is finished

24
Vacuum Vessel Fabrication

• Vacuum vessel material must be non-magnetic and
  strong therefore 304 stainless steel will be used
  throughout
• The vacuum vessel will be fabricated by rolling
  stainless steel sheets into cylinders
• Two identical vessel halves will be fabricated
  with all ports and feedthroughs

25
Vacuum Vessel and Coupling Coil
26
Vacuum Vessel/Coupling Coil Integration

• LBNL will weld in gussets that fit between the
  coupling coil and the vacuum vessel
• Sixteen special gussets are welded between the
  coupling coil magnets cold mass support tubes
  and the magnet housing
• The gussets will transfer the magnetic loads
  between the coupling coil and the vacuum vessel

27
RFCC Attachment to Support Stand

• The vacuum vessel is bolted to a saddle made up
  of small plates welded to the support stand
• Stainless steel bars are welded onto the vacuum
  vessel for attaching bolted gusset plates

Bolted gusset mounting bars
28
RFCC Support Stand

• RFCC support stand must withstand a longitudinal
  force of 50 tons transferred from the coupling
  coil
• Bolted gussets and cross bracing provide shear
  strength in the axial direction (analysis will
  confirm the stand design)

29
Liquid Nitrogen Cooling Considerations

• Suspension of cavities on struts provides low
  heat leak from cavity to vacuum vessel
• Beryllium window FEA thermal analysis will need
  to be performed with new parameters
• The cavity frequency will be shifted
  (approximately 600 kHz), therefore tuning system
  or RF power source modifications will be needed
• Insulators will need to be added to the RF
  couplers
• Coaxial LN feedthrough tubes will be needed to
  insulate the connection outside of vacuum

MICE RF Cavity Mechanical Design and Analysis
Page 29
Allan DeMello - Lawrence Berkeley National Lab
- June 4, 2008
30
Module Assembly Sequence
Beryllium windows are installed onto the cavities
Bare cavity
31
Module Assembly Sequence
Assembly of the tuners onto the cavities (w/o
actuators)
Install struts onto the cavity
32
Module Assembly Sequence

• Slide the inner cavity into vacuum vessel using
  spacer/alignment blocks for support
• Shim cavity to align tuner and coupler vacuum
  feedthroughs with tuner mounts and cavity ports
• After adjusting their lengths, secure the struts
  to the vacuum vessel mounting blocks

33
Module Assembly Sequence
Install tuner actuators
Install RF couplers
Install vacuum couplers
Install cooling water feedthroughs
34
Module Assembly Sequence
Repeat the same process for the other cavities
Install vacuum valves and pumps
Two cavities are installed from each end of
vacuum vessel
35
RFCC Module Shipping Configuration
Cryocoolers removed Cavities removed
Couplers removed Vacuum pumps removed

• Module short stand used for
• Initial module assembly
• Shipping to RAL
• Moving into MICE Hall

36
MICE Hall Access
37
Installation in MICE Hall

• Module assembled on lateral tracks
• Module aligned and shimmed to correct height
• Bellows on either end of module are pulled back
  into open position and O-ring in place
• Moved into position w/rails for final alignment
• Bellows are released and flanges are sealed
• Bellows are locked out using bridging bolts
• RF and utilities are connected to module

38
Module Flange Connection
End flange O-ring seal
Formed bellows
Flange through holes
39
Flanges and Seals

• Flanges designed to mate with AFC module
• AFC flanges are flat w/no grooves or bellows
• RFCC modules have bellows and O-ring grooves on
  flanges at either end
• Bellows have gt1 cm of total travel
• Bellows are locked out after installation to
  provide a means of transmitting forces between
  modules

40
Module Mounting Provision

• Six mounting plates welded to the support base
  for installation on rails
• Same as spectrometer solenoid mounts
• All magnetic loads carried to the floor through
  the mounting plates

41
Cavity RF Feeds

• Eight RF feeds/module use standard 4 RF coax
• Cooling water for loop is required (ltlt1 lpm each)
• Adapter on end of couplers isolates ceramic RF
  windows from forces during installation
• Location of coax interface w.r.t. module center
  TBD

RF feeds
42
Schedule Overview

• RFCC design and fabrication project originally
  expected to be a 3year project (10/06 to 10/09)
• Coupling coil effort began in 2006 at ICST
  (Harbin)
• Design and fabrication of other RFCC module
  components was scheduled to begin 10/07
• Start was delayed due to lack of availability of
  qualified manpower
• Earlier this year, mechanical engineer A. DeMello
  joined MICE to work on RFCC module design (FTE)
• Additional manpower required to make up schedule

43
Manpower Summary

• Primary Manpower
• Allan DeMello lead ME for RFCC Module design
  fab
• 3D engineering CAD model, cavity analysis, design
  fab
• Steve Virostek engineering oversight for MICE at
  LBNL
• Mechanical Engineer (TBD) design/fab of
  subcomponents
• Cavity tuners, support structure, large
  procurements
• Mechanical Designer (TBD) generation of
  fabrication dwgs
• Other
• Derun Li cavity physics design and oversight
• Mike Green coupling coil design interface with
  module

44
Schedule Summary