Radial Ion Pump, BPMs,

1
Radial Ion Pump, BPMs, HOM Bellows
Machine Advisory Committee Meeting December 14,
2004 Nadine Kurita
2
Outline

• Radial Ion Pump
• Beam Position Monitors
• HOM Bellows
• Q4/Q5 Bellows absorber
• Straight HOM Bellows
• Q2 HOM Bellows

3
Contributers/ Upgrade Staff

• Physicists
• Michael Sullivan
• John Seeman
• Stan Ecklund
• Sasha Novokhatski
• Stephen Weathersby
• Cho K. Ng
• Artem Kulikov
• Uli Wienands

• Designers
• Ho Dong
• Manual Trigos
• Michael Kosovsky

• Engineering
• Nadine Kurita
• Dan Wright

4
B1 Radial Ion Pump

• Pump modeled after PEP-I, SPEAR and the Damping
  Ring.
• Detail design of cell arrays engineered by C.
  Perkins 1998.

5
B1 Radial Ion Pump
Prodec Anode cell structure
Ceramic Standoff
Tantalum Cathode Plates

• Modified to Ta from Ti to increase noble gas
  pumping and eliminate the argon instability.

Additional BPM set
Reduced to 4 cell arrays from 6.
Baffles

• Shorten pump to add BPM set.

?2mm holes
6
B1 Radial Ion Pump
Create two independent pumping cells

• Pro if a cell fails you still have another
  operational unit
• Con you are 2 times as likely to have a
  failure.

Standard pump feedthrough

• Current feedthrough rated fo 6 kV
• Pump operates at 5.5 kV
• Standard feedthrougs rated for 12 kV, 10A

7
B1 Radial Ion Pump

• Pump cell array is unchanged
• Cell diameter optimized for pumping speed and
  operating pressure
• ?.36 cm for a Penning cell in a 15 kilogauss
  field at 1 x 10-9 Torr pressure.
• Speed versus diameter curve is flat above ?.2 cm
  so the hole size is driven by manufacturability.
• 21 cell height to cell diameter ratio gives
  optimum surface coverage for sputtering on the
  anode.

• ?.48 cm
• Need to review depth

8
Ion Pump

• Holes baffles unchanged
• ?.094 x .245 deep
• Baffles to prevent SR from striking anodes or
  cathodes.
• No direct line of sight

9
Ion Pump Milestones

• Final Design Review 1/05
• Order long lead items 1/05
• Tantalum plates
• Complete piece detail part drawings 3/05
• Receive piece parts 5/05
• Assemble, bake 7/05

10
Beam Position Monitor (BPM)

• Upgrade improvements
• Added BPM set at each radial ion pump.
• The new set is separated in z by f 7.9 cm from
  the BPMs in the B1 chamber, where f 2.
• 7.9 cm corresponds to a quarter wavelength of 952
  MHz, the BPM procesing frequency.
• In the electronics they can then synthesize
  independent linear combinations of the signals
  which correspond to the two beams moving in
  opposite directions.

27.9
11
BPMs in Replacement Chambers

• HER Q4 Q5 Chambers
• Located at the outboard end of Q4 and outboard
  end of Q5.
• Use spare PEP-II BPMs for Al chambers (LER arc
  style).

Q5L looking downbeam
Q4R looking downbeam
12
Improvements HER Q4/Q5 BPMs

• Bellows allows the Q5 BPM to be rigidly supported
  (xx, y z). Q4 BPM is held in x, yy
• Greater thermal stability
• Lowered thermal gradients
• Support BPM to Quad magnet
• No calibration required QMS/BBA

• Place BPMs on flat surfaces.

• BPMs are centered on the beam in the
  x-direction,
• BPM center to BPM center x- spacing determined by
  (R. Johnson, S. Smith (2004).

13
BPMs (cont.)

• Spares
• HER Arc Style (CuNi Housing) 3 units
• LER Arc Style (Tin seal housing) 44 units
• Straight Style (SS Housing) 2 units
• Total quantity needed
• Radial ion pump 8
• HER Q4/Q5 16
• LER Q4/Q5 16
• Could use LER Arc Style, but HER Arc style
  preferred
• Total 44 BPMs
• Equals spare quantity of LER arc style
• No loss of units, no additional sets if possible

14
BPM History

• PEP-II - purchased alumina borosilicate glass
  feedthroughs from Kaman Instrumentation.
• 2002 Meggit Safety Systems purchased Kaman
  Instrumentation.
• 2003 Meggit produced spare BPM's for SPEAR3
• XPS analysis shows product to be incompatible
  with vacuum.
• 2002 Times Microwave starts up a new division
  with the Kaman engineers to produce borosilicate
  connectors and feedthroughs.
• Times has no rights to our PEP-II or SPEAR3
  design.
• New process and ceramic to produce the seal.
  This technology is better for vacuum cleanliness,
  but we have no history on the integrity of the
  seal.
• 2004 Bejing receives BPM's from Times that
  leaked after welding.

15
BPM Vendor Selection

• Meggit
• Pros They have detailed drawings and procedures
  to fabricate our BPM's.
• Cons They are not as responsive as Kaman was.
• Cons They have not successfully built a clean
  vacuum component.
• Times
• Pros They have the original engineers that
  helped develop the PEP-II BPMs.
• Pros They are responsive.
• Cons Unproven design and manufacturing of the
  seals. We would require RD funds to validate
  their sealing technology and connector
  reliability.
• Cons It would be beneficial to develop another
  company that could produce BPM's for the lab in
  the future.
• Cons Long term viability of the RF
  instrumentation division.

16
BPM Future Tasks

• Clean the Meggit SPEAR3 BPMs with a non-corrosive
  solution, bake and RGA scan.
• Re-develop with Times a comparable BPM's.
• These BPM's should be electrically identical to
  the PEP-II BPM's and they must meet our technical
  specification.
• Testing per the SLAC specification
• Estimated lead time for fabrication is 10 weeks
  from Times. Potentially longer lead time for
  Meggit.

17
Q4/Q5 Bellows Absorber

• Major Requirements
• HER 2.2A, 9Gev
• Beam stay clear
• 12 ? 0 mm in X
• 9 ? 0 mm in Y
• Luminosity Cone 6.24?
• Synchrotron Radiation
• No SR power strikes the bellows module
• Mis-steer
• RF fingers protected by chambers
• 1 mrad in X requirement
• 2 mrad in Y requirement
• Forward gt 5 mrad
• Backward gt 25 mrad
• HOM power, Scattered SR, Ohmic
• Engineering estimate 1 KW/m

18
Q4/Q5 Bellows Requirements (cont.)

• Modular Design 4.25
• Operating Temperatures
• Tmax Finger lt 100ºC
• 500 C _at_ 10 hrs w/ minimal stress relaxation
• 0ºC - 100ºC, Installed
• 200ºC Bake Out, Manufacturing
• Chamber Operating Temperatures
• Cold Day 0?C
• ?Tave 45?C
• Allows for misalignment and manufacturing
  tolerances of mating chambers.
• Allows for thermal expansion of mating chambers.
• Installation space for chambers.
• Load bolts from bellows.
• Space is tight may need to remove corrector

19
Q4/Q5 Bellows Layout

• Q4 side, 10 flange

Inconel Spring Finger
Absorbing Tile
GlidCop Stub
GlidCop RF Shield Finger
Welded Bellows
Q5 side 12 flange
Cooling not shown
20
Q4/Q5 Blws Detail Design

• HER Arc Bellows concept with absorber
• Ensure failure does not result in the RF shield
  falling into beam tube
• Shield fingers slide on outside of chamber stub
• Keep high stress areas away from high heat areas
• Keep steps to a minimum, reduce impedance
• Plating to minimize wear, decrease cold welding,
  solid lubrication

21
Q4/Q5 Blws Aperture
.080 step at stub

• SR passes both directions
• Stub cant protect thin RF shield fingers
• Backward side
• Mask on chambers protect bellows from large
  misteers
• Forward side
• Chamber walls protect bellows from 5 mrad misteer

e- Forward

• BSC grows in Q5
• No taper step at stub only

e- Backward
e Backward
22
Q4/Q5 Blws - Absorber

• Three options for absorber placement.
• 1 - Directly above RF shield fingers
• 2 - Above the Spring Fingers
• 3 - In the bellows cavity space
• Tile is located in the HOM cavity ?
• Creates another vacuum joint ?
• Makes GlidCop stub a mechanical braze not a
  vacuum braze. ?
• Latest design uses option 3.
• All options probably absorb the trap mode between
  the RF shield fingers and the welded bellows
• Sasha/Stephen have a model of option 1. Option
  3 next week.

Option 1
Option 3
23
Q4/Q5 Blws - Absorber Analysis

• Tile
• Actual size and quantity TBD. Engineering
  evaluation assumes optimal tile volume.
• Size .4 x .47 x .5
• 14 tiles in module
• Ceralloy 13740
• K 30 W/m-C
• Flexural strength 43.5 ksi
• HOM power
• 2 KW assumed
• Ansys Results
• Tcool 51?C
• Tmax tile 240?C
• ?tile z25 ksi

24
HOM Absorbing Bellows
25
HOM Absorbing Bellows

• New bellows designs that also function as
  beamline HOM absorbers.
• LER arc bellows
• Straight bellows
• Q2 bellows
• New bellows designs that have absorbers that
  protect themselves from modes that leak behind
  their RF shields.
• Vertex bellows
• Q4/Q5 bellows

26
Straight HOM Blws -Design Details
Absorbing Tile
2.75 long by .24 wide HOM Trapping Slots
Inconel Spring Finger
Modes in the chamber propagate through the slots
are absorbed by the AlNiSiC.
GlidCop Stub
GlidCop RF Shield Finger
Bellows Cavity
Welded Bellows
Modes that leak past the RF shield finger and are
trapped in this area still see the absorber
27
Straight Section HOM Bellows

• Prototype of the HOM absorbing bellows
• Simple round geometry
• Locate near isolation valves to tests its impact
  on HOMs in neighboring components.
• Conceptual design near completion
• HOM calculations are being done to optimize tile
  size and slot dimensions.
• Initial HOM analysis shows that the concept
  works.
• Reduces monopole absorption while optimizing
  dipole and quadrupole field absorption.

28
Near IR Layout
29
Q1/Q2 HOM Bellows

• FY2003 added 4 layers of tiles per module.
• Absorbing 10 KW presently
• Predict 50 KW in 2007
• Numerous iterations on HOM absorbers have been
  analyzed by S. Weathersby and A. Novokhatski (38
  runs).
• Goal Create a HOM absorber that doesnt
  generate 50 of its absorption power.
• Reduce monopole without significantly reducing
  dipole and quadrupole modes
• Most effective design requires at minimum 4
  slots as in the Straight HOM Bellows.
• The optimized design for various modes must be
  chosen by February 2005.
• A few more design/analytical iterations will be
  performed
• Reduce power absorption, but still reduce HOM
  power at the vertex ends, vertex bellows and
  radial ion pump.
• Vertex bellows will have HOM tiles
• Gold plating will be extended on the vertex ends.

30
Q1/Q2 Blws - HOM Analysis

• 4 long tile sets
• Suppresses the monopole mode without reducing the
  dipole and quadrupole mode
• Sasha calculated the set back of the tiles

• Focusing on 2 long tile sets
• Reasonable length for the 5 bellows module

31
Q1/Q2 Blws - Design Status

• New concept developed
• based on best information available.
• Maximum Tile/slot length
• 2.4

• Absorbing tiles is open to the convolutions
• No additional tile set needed in bellows cavity.
• HER Arc Style Bellows
• Spring
• Stub
• RF shield

• Possibly reduce further the travel and offset
  requirements to increase length.

32
Q1/Q2 Blws - Major Milestones

• Finalize Physics/HOM Reqs Feb 05
• Conceptual Design Review Mar 05
• Final Design Review Apr 05
• Long Lead Procurements Apr 05
• Detail Drawings Complete Jun 05
• Receive Parts Aug 05
• Final Assembly Sep 05
• Ready for installation Sep 05