Beam Delivery System - PowerPoint PPT Presentation

Loading...

PPT – Beam Delivery System PowerPoint presentation | free to download - id: 632b0-YjExO



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Beam Delivery System

Description:

... Fermilab and consequent visit of John Tompkins and Vladimir Kashikhin to SLAC on ... Discussed this with Daniel Schulte. e.g. Daniel asked to find ... – PowerPoint PPT presentation

Number of Views:109
Avg rating:3.0/5.0
Slides: 64
Provided by: Andrei79
Category:
Tags: beam | delivery | system

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Beam Delivery System


1
Beam Delivery System
  • Deepa Angal-Kalinin, Hitoshi Yamamoto, Andrei
    Seryi
  • for BDS Area
  • review video meeting
  • March 28, 2006

2
Agenda questions suggested by RDR management
  • Optics and layout status (35') Optics, general
    layout, technical systems, etc
  • CFS for BDS (15') alcoves, tunnel size, access
    shafts, IR halls, etc
  • Options for cost reduction (15') of tunnels,
    system length, energy upgrade, of Irs, etc
  • Operations reliability issues (10') beam
    losses, dumps, diagnostics, tuning etc
  • What are the design criteria that have driven
    your choices?
  • What performance overheads are included in the
    design?
  • Are their outstanding (cost-relevant) design
    choices?
  • Is all the cost-relevant information available to
    the TS groups?
  • What is the status of the TS groups (with respect
    to the BDS)?
  • Are there identified 'costing bottlenecks'?
  • Can you foresee cost saving modifications, and
    what would their performance impact be?

3
Baseline layout 20mrad IR and 2mrad IR
  • Two longitudinally separated IPs, two independent
    collider halls for two experiments
  • Entry to entry 5.5km (grid size 100m 5m)
  • Shafts service tunnel not shown

4
Details of layouts and corresponding optics
EFF1
EDL1
undisrupted beam size does not destroy beam dump
window without rastering. Rastering to avoid
boiling of water
spoilers survive 2 bunches
5
EBSY
skew correction
betatron collimation
polarimeter chicane
EIRT1
EIRT2
MPS betatron collimators
kicker, septum
to IP
EFF2
4-wire 2D e diagnostics
EBSYD
Energy diag. chicane
High bandwidth horiz. bending system
EFF2
to IP
to tune-up dump
EFF1
  • Updated optics (ILC2006a). August-October 2005
    optics put together. No change with respect to
    what is described in BCD, except
  • new sacrificial MPS betatron collimation at the
    entrance
  • some length (diag. kicker) increased (according
    to BCD)

EBSY
1um beam at laser wires
Still working on EBSY EBSYD and will have new
version soon
6
betatron collimation
energy spectrometer
energy collimation
EIRT2
EFF2
to IP
EIRT1
ISR in 11mrad bend
EFF1
EIRT2
to IP
7
Layouts with captions (fragment of EFF2 PDL2)
http//www-project.slac.stanford.edu/ilc/acceldev/
beamdelivery/rdr/components/bds_layout_feb10_06.pd
f
muon wall
tail folding octupoles
absorbers
crab cavity in 2mrad
sextupoles
grid dx100m
8
E and polarization diagnostics
EDL2
PDL2
EFF2
PFF2
aberration correction section
EDL2
final transformer
Length needed to provide 3m separation of the
dump from incoming beamline
EFF1
disruption beam capture
tail folding octupoles
E and polarization diagnostics
EDL1
PDL1
9
Optics organization
  • Optics ILC2006ahttp//www.slac.stanford.edu/mdw
    /ILC/Beam_Delivery/2006a/
  • The MAD optics files
  • Survey files Sub-beamlines

ebds0.mad ebds1.mad ebds2.mad pbds0.mad pbds1.mad
pbds2.mad
ebds0_survey.tape EBSYD ebds1_survey.tape
EBSY EIRT1 EFF1 EDL1 ebds2_survey.tape
EIRT2 EFF2 EDL2 pbds0_survey.tape
PBSYD pbds1_survey.tape
PBSY PIRT1 PFF1 PDL1 pbds2_survey.tape
PIRT2 PFF2 PDL2
10
Functional description
  • EBSY (Electron Beam Switch Yard)
  • BEGIN exit of last ELIN2 cryomodule
  • MPS betatron collimation
  • skew correction section
  • 4-laser wire 2D e diagnostic section with
    trajectory feedback
  • energy diagnostic chicane (with MPS energy
    collimator)
  • END entrance of first extraction bend magnet
    (kicker)
  • EFF1 (Electron Final Focus 1)
  • BEGIN exit of last EIRT1 polarimeter chicane
    bend magnet
  • primary betatron collimation section
  • primary energy collimation section
  • energy spectrometer chicane
  • tail folding octupoles
  • aberration correction section
  • final transformer
  • final doublet
  • END IP1 (20 mrad crossing angle)
  • EBSYD (Electron Beam Switch Yard to Dump)
  • BEGIN entrance of first extraction bend magnet
    (kicker)
  • Bend (kicker) for slow (fast) extraction
  • septum
  • high bandwidth horizontal bend system
  • vertical bending system
  • beam sweeping system
  • tune-up dump
  • END entrance of tune-up dump window
  • EDL1 (Electron Dump Line 1)
  • BEGIN IP1 (20 mrad crossing angle)
  • disrupted beam capture
  • energy spectrometer chicane
  • polarimeter chicane
  • collimation
  • rastering and final drift to dump
  • END dump window
  • EIRT1 (Electron Interaction Region Transport to
    ir 1)
  • BEGIN entrance of first EBSY extraction bend
    magnet (kicker)
  • matching section
  • stretch (for longitudinal interaction region
    separation)
  • polarimeter chicane
  • END exit of last polarimeter chicane bend magnet

11
Tables of elements (example)
http//www-project.slac.stanford.edu/ilc/acceldev/
beamdelivery/rdr/components/
  • Aperture is as seen by beam, vacuum chamber
    thickness not included
  • SC magnets indicated by SC in eng. type column
  • Beam energy in tables is 250GeV. Magnets and PS
    should work for 500GeV
  • Additional description required for kickers,
    septa, IR magnets
  • Optics has info where movers are. It needs to be
    transferred to tables.

12
Counts of elements in subsystems
  • Red numbers indicate possible inconsistencies
    that need to be fixed

13
Tech specs being prepared
http//www-project.slac.stanford.edu/ilc/acceldev/
beamdelivery/rdr/
http//www.linearcollider.org/wiki/doku.php?idrdr
rdr_asrdr_as_home
ILC BDS Area. Materials for RDR Layouts,
apertures, tables of components, etc. (tentative)
UPDATED Feb 11, 2006 BDS Optics files (tentative)
UPDATED Feb 11, 2006 Technical specs for
feasibility study and cost estimations    2mrad
IR Final Doublet Posted Feb 22    Crab cavity
system Posted Feb 26to come     Collimators
(spoilers, absorbers, PCs)    Beam dumps   
Muon walls    20/(14)mrad FD etc
14
Tech systems progress
  • Well developed communication and started work
    with Tech.Syst. Glob.sys
  • Magnets
  • Instrumentation
  • Installations
  • Controls
  • Conv. facility
  • Need to be improving
  • Vacuum
  • Dumps collimators
  • Mid April plan to have phone/video meeting with
    Tech.systems to discuss their first cost
    estimates

15
(No Transcript)
16
Technical system
  • Magnets tech system
  • Meetings at Fermilab and consequent visit of John
    Tompkins and Vladimir Kashikhin to SLAC on March
    2-3
  • Discuss cost estimations, design of magnets,
    movers, etc.
  • Issues identified that need further work or
    definition in BDS
  • low field magnet design, field quality, etc.
  • stringing rules for 500GeV CM
  • going down in energy by switching strings off
  • Need guidance on both Max Energy and also Min
    Energy from Exec.Comm.
  • should we assume 91GeV CM to 250 GeV CM for 1st
    stage?
  • Missing bend strategy is being developed (50 or
    less installed at 500GeV)
  • IR magnets
  • Fermilab focused on 2mrad IR magnets, BNL is
    working on 14/20mrad IR, Saclay will hopefully
    also consider 2mrad IR magnets
  • Met and discuss this further with Sugahara-san at
    Bangalore

17
Missing bend strategy at 500GeV CM
Example shows one of possible configuration in
the FF part (Y.Nosochkov)
  • At 500GeV CM will install 50 or even less number
    of bends to cut cost. These bends need to be
    installed after the energy upgrade.
  • Bends will be split and arranged in strings so
    that decreasing the energy much below the nominal
    500GeV CM will be done by switching off the
    strings.
  • Splitting bends (12m presently) will be
    determined by the energy range (should we assume
    91GeV 500GeV CM?) by 5 to 2.4m?

18
14(20)mrad, BNL
19
2mrad IR Fermi and possibly Saclay will look on
feasibility and cost estimates
Shared Large Aperture Magnets
Disrupted beam Sync radiations
Q,S,QEXF1
SF1
QF1
SD0
QD0
60 m
Beamstrahlung
Incoming beam
pocket coil quad
Rutherford cable SC quad and sextupole
20
Muon walls
  • BCD two walls, 9m and 18m per branch, to reduce
    muon flux to less than 10muons/200bunches if
    collimate 0.001 of the beam
  • Predictions of halo 1e-6 - 1e-5
  • Min one 5m wall needed for PPS
  • Approach consider installing single 5m wall,
    space in tunnel for full set
  • The 5m wall allow to collimate 2e-5 before
    reaching 10m/200bunches
  • Before the CCR can be considered gt
  • ask Accel. Phys. Tech. System to evaluate
    predictions of halo population
  • ask Installation Tech. System to evaluate
    possibility to install additional wall if muon
    rate will exceed the limit
  • ask MDI panel to evaluate the 10m/200bunches
    detector tolerance

Muon Momenta from PC-3 Hitting 6.5 m Detector
Lew Keller
250 GeV beam
No magnetic wall 5 m wall at S -321m 18 m
wall at S -321m
21
Technical system (contd.)
  • Accel. Phys. tech system
  • Video mtg on Feb 24. Defined urgent Acc. phys.
    questions from BDS
  • Vacuum chamber material (Cu, Al, SS, wakes etc),
    tens of M difference
  • Prediction of halo population (collimate 1E-5
    instead of 0.001?) Dtens of M for muon walls
  • Evaluate BDS vacuum requirements (10nTorr to
    50nTor w.r.to e-cloud in e branch, etc)
  • Evaluate e and dE/E of the beam with large b and
    E errors
  • Evaluate effect of HOMs and LOMs of the crab
    cavity on the beam
  • Discussed this with Daniel Schulte
  • e.g. Daniel asked to find resources in BDS group
    to investigate the issue of material for vacuum
    chamber
  • plan to study if SLAC ARDA can do this

22
Technical system (contd. 2)
  • Beam dumps and collimators tech system
  • Meeting at DESY on Feb.16 devoted to beam dump
    design with participation of DESY, UK and SLAC
    colleagues, and with video link to SLAC.
  • Review design of water beam dump and its cost
    estimate
  • Cost differences of earlier estimates are
    understood much better
  • Cost of dump vessel itself or plumbing and
    cooling systems are not very different in DESY
    and SLAC early cost estimates. Civil construction
    gives most of the difference. Decommissioning
    cost.
  • DO NOT include decommissioning cost???
  • DOE or other rules may require us to consider
    plan for decommissioning?
  • There are still questions of maintenance,
    approval, etc, that need to be carefully
    addressed
  • For rd, the window remains the most important
    item
  • Tech.System is just starting the work may be a
    concern
  • Collimators system need a fresh design look as
    well. Tech.system requests help in collimation
    system design

23
Beam dumps collimators
  • ILC Beam dump design based on design of MW dumps
    established at SLAC in sixties by D.Walz
  • Adjusted design for 18MW
  • Working with collaborating labs (UK) on eng.
    design and cost estimate for RDR
  • Collimators experience in design of survivable
    and renewable spoilers
  • Plan survivable spoilers for ILC, will build
    renewable for LHC (experience may then be
    applied to ILC)

Picture of beam dump from NLC 1999 study
Collimators (spoilers, absorbers). Spoiler
1r.l. of Cu, and Cu plated Be to min wakes
24
Beam dumps and collimations
  • Full power dumps (18MW) (6)
  • Removing tune-up dumps will be considered
  • Photon (1-3?MW) dumps (2)
  • May consider reducing power of tune-up dumps when
    cost will be provided by Tech. system
  • (if cost would be weak function of power, such
    change may not make sense)
  • Fixed aperture protection collimators (60)
  • Adjustable spoilers and absorbers (60)
  • Passive devices to limit betatron aperture

25
Technical system (contd. 3)
  • Instrumentation controls tech system
  • Following discussions at Fermilab, will look at
    the list of instrumentation in BDS and include
    what is missing (still to be done)
  • Based on Excel spreadsheets of BDS elements, the
    list of controls requirements was generated by
    ANL colleagues
  • Number of DAQ channels, cables, etc.
  • Started detailed discussion of these lists
  • Discussed with John Carwardine and Marc Ross at
    Bangalore
  • Discussed with Marc Ross Grahame Blair last
    week at KEK
  • Added into BDS instrumentation (Marc Ross, Feb.
    RDR mtg)
  • Imagers profile/SR/XSR (2/0/4)
  • LOLA cold (4) to measure sz and y/x/E z
    correlations

26
Model of 3.9GHz deflecting (crab) cavity designed
by Fermilab
L.Bellantoni, et al
HFSS Mesh from FNAL
SLAC Model Meshing is next
K.Ko, Z.Li, et al
Collaboration is being formed to work on ILC crab
cavity systems Fermilab, Daresbury, SLAC, The
3.9GHz deflecting cavity designed at Fermilab.
Several simplified models have been manufactured.
Complete design with all couplers exist now.
Design is being verified with various tools
including parallel codes (see next page), and
then a prototype will be built. The phase
stabilization system is being designed.
27
Vacuum system
  • Total length 12km, 30 in bends with SR losses
  • Vacuum requirements 10nTorr in 0.5km within IP
    (based on beam-gas scattering), and 50nTor (tbc)
    elsewhere
  • Fast valves, optical windows,
  • Apertures shown in Fig. gt (region of crotches
    show oscillation, special design attention
    required)

28
Vacuum system (contd.)
  • Material choice is not finalized (cost driver)
  • Cu, Al and SS considered for NLC. Finally, SS was
    costed in 1999.
  • TESLA design SS chamber with Cu plated bellows
    and ports to reduce wakes
  • Asked Accel.Physics Tech.System to evaluate
    effect of wakes in BDS, define requirements on
    conductivity and other parameters of vacuum
    chamber
  • SS vacuum chamber concerns
  • may affect field quality in magnets (mgt1,
    experience of KEK-B) to be evaluated
  • SR losses in bends areas at 1TeV CM up to
    several kW/m in some chicanes and septa gt may
    require copper and/or antechamber
  • Information on NLC and TESLA vacuum chamber
    design

http//www-project.slac.stanford.edu/ilc/acceldev/
beamdelivery/rdr/docs/vacuum_intro.html
29
Conv. facility Transverse space
  • Beamline approximately 1m from the floor
  • Min 1m from the wall (beamline is curved)
  • Space for low energy positrons
  • Space for optional tune-up line to the main dump
    on the same side (if decide later to include it
    into design)

Low E e
1m
30
Shafts and path of access in BDS
  • Locations of beamlines
  • Close to external walls, to allow access from IR
    halls without crossing the extraction lines
  • The service tunnel in the linac should be moved
    to the side of larger angle IR
  • Shafts maximum number

Low E e
Where this shaft is located and can it provide
access to all three branches?
31
Baseline layout 20mrad IR and 2mrad IR
  • Grid size 100m 5m
  • (Beamline is not placed near external walls, as
    suggested above)
  • Shafts and places for power supplies not specified

32
Max number of shafts in BDS (8)
  • Shafts near all full power dumps, IR halls and
    Ys.

33
Additional tunnels for power supplies
service tunnel
  • 1) Extend linac service tunnels to BDS for 500m
    to house power supplies and beam diagnostics
    lasers
  • 2) Connect two collider halls by an accessible
    tunnel to place power supplies

34
Min number of shafts in BDS (2 large and 1 small)
  • Two large shafts for detectors and one small shaft

35
Issues of no service tunnel in BDS
  • Long cables up to 1.5km
  • Power supply rating is likely to double for
    cables
  • Cable plant cost is noticeable (still seems
    smaller than service tunnel)
  • Temperature stability in tunnel is very serious
    issue
  • cables will probably dissipate several MW(?)
    i.e. several hundred W/m gt cables need to be
    placed in water cooled enclosure (?) to provide
    the needed temperature stability
  • T stability is to be defined. In NLC specified
    ?2?C/24hr and ? 0.25 ºC/1hr, however, this appear
    not stable enough. Would like to have
    ?0.5?C/24hr and ? 0.1 ºC/1hr
  • Laser transport up to several hundred meters may
    be needed (vacuum pipe) probably feasible
    (cost)
  • BPM performance with long cables

36
Issues of min shafts
  • Installation, access, maintenance of beam dumps
    and other equipment
  • Min number of shafts may be mitigated by a full
    length service tunnel
  • In general, decision of shafts and service
    tunnel should come from consideration on space
    needed, access and installation needed requested
    by Tech. Systems. In many cases such
    consideration havent started yet

37
CF BDS layouts example of e transport options
e tunnel
suggested baseline solution
abandoned version
F.Asiri, C.Corvin, et al.
abandoned version
38
IR hall sizes
  • Large range (3.5 times in volume)
  • Assumptions for RDR Largest size
  • also weight of detector, crane, etc.

SiD 48m x 18m x 30m SiD Collab. Mtg. 16-17
December 2005
GLD 72m x 32m x 40m Snowmass data
39
Discussed in details in LCWS06/MDI panel
  • Upgrade from ee- to gg and back to ee-
  • assume that 25mrad is needed for gg to attain its
    max potential
  • reversibility of upgrade may be essential
  • e.g. ee- run gt gg run gt E upgrade and next
    ee- run
  • consider concept for 20mrad or 14mrad IR upgrade
  • The suggested upgrade scenarios were accepted as
    a guideline need to write a corresponding
    description for BCD
  • For single IR, if two detectors are planned, then
    push-pull is needed
  • discussion started of the technical requirements
    that need to be solved to enable push-pull with
    rapid switch over time
  • Anything else than the baseline two IRs has met
    very strong resistance of detector/physics
    community at LCWS06
  • Updates on losses and radiation levels in 20mrad
    and 2mrad extraction lines
  • SR from disrupted beam is important

40
20mr IR
41
20mr gt 25mr
  • See notes on next page

42
Upgrade of 20mr to gg 25mr and back
  • Install bend after energy collimator of modify
    E-coll bend to create additional 2.5mrad
  • Will need to study if this affects collimation
    efficiency and whether this is an issue for gg
  • Pink area show tunnels that need to be built for
    gg in the upgrade or from start
  • Detector and IP moved by about 1.8m, FF elements
    moved
  • Build new 0.25km gas dump followed by water dump
    for gg (next slide) in a new tunnel, do not
    dismantle either the water dump for ee- or
    extraction line
  • If the gg beam dump and new tunnel need to be
    much longer, the positron transfer tunnel should
    go above the projected path of gg dump
  • In back conversion gg to ee-, move FF beamline
    and detector back, continue to use ee- water
    dump.

43
Assumed this dump for gg (feasibility to be
studied)
44
14mr gt 25mr
  • additional angle is 5.5mrad and detector need to
    move by about 4.2m

45
Upgrade of 14mr to gg 25mr and back
  • Additional comments
  • Tunnel in FF area may need to be wider
  • The gg dump is in separate tunnel may ease
    radiation issues and upgrade back to ee-

46
Discussion of requirements to enable push-pull
  • Detailed discussion happened on MDI panel/LCWS
    side (presented by Tom Markiewicz)
  • Technically, rapid (1-2 day every 3month) switch
    may require
  • Detailed engineering for push-pull from inception
  • Part of Final Doublet carried with detector
    during move
  • Break point in optics with double valves cold
    warm transitions
  • Sufficient embedded steel in the floor
  • Reference alignment network in the hall to
    monitor detector and floor deformation jacks
    under detector minimize detector deformation
    during move. Vertex and tracker may be on its own
    independent alignment system
  • Detector is self shielded and no shielding wall
  • SC solenoid of detector may need to stay cold and
    energized
  • etc.
  • Single IR (even with push pull) met strong
    resistance at LCWS06. It was strongly suggested
    that any such change go through ILCSC

47
Radiation Loads in 20-mrad Extraction Line
  • Dynamic heat loads up to 500 W/m
  • Power density above quench limit
  • Peak dose in coils up to 270 MGy/yr
  • 2-mrad x-ing 0.76 MW synch rad loss

48
Approach suggested by DCB focus on cost drivers
  • Decision cost drivers
  • One or two IRs
  • Max energy (design for 250GeV/beam or
    500GeV/beam)
  • Independent or single collider hall
  • Push pull detector or not, etc.
  • Components cost drivers
  • tunnels (12km of beamlines), shafts and halls
  • vacuum system
  • magnets (large count)
  • muon walls (material cost)
  • IR regions (unique and complex)
  • beam dumps (CF cost)

49
Possible cost saving strategies
  • Need to know the costs to understand which
    strategies to implement
  • Singe IR. Resistance from physics/detector
    community will be high.
  • Plan to organize WBS in such a way that would be
    able to cost one IR or another (removing cost of
    bends and stretches not needed for single IR)
  • If the design of single IR would need to be
    chosen, unlikely it would be 2 or 20mrad, but
    more likely something in between

50
Discussion of WBS organization
  • Plan to organize in such a way to allow easy
    evaluation of either of IRs
  • E.g., larger xing IR would include
  • BDS_Larger_Xing
  • BDS_Common
  • CF need to go to BDS
  • (Numbers arbitrary)

51
Other possible cost saving strategies
  • Need to know the costs to understand which
    strategies to implement
  • Install only fraction (half) of bends at 500GeV
    CM stage
  • this will increase difficulty and cost of the
    energy upgrade
  • Design all quads as consisting from two halves
    and install only one half at 500 GeV CM
  • same difficulty with upgrade also double movers
    BPMs?
  • Replace high power tune-up dumps with low power
  • do not consider additional beamline from BDS
    entrance to main dump
  • reduce band-path and shorten tune-up extraction
    line
  • Consider staging construction and installation of
    muon walls (e.g. start with min 5m wall). Install
    more if muon rate is too high
  • May be difficult to install the wall in
    operational tunnel
  • Consider alternative muon spoilers
  • Undisrupted beam size at dump window - rely not
    on drift but more on rastering - shorten
    extraction lines (MPS)

52
Length and energy
  • ILC BDS presently designed for up to 1TeV and
    assume that geometry (angle of bends) do not need
    to be decreased in energy upgrade. (Assume we do
    change FD in upgrade)
  • If we would assume decrease of bend angles at the
    energy upgrade to 1TeV, then the BDS would have
    larger overhead and can go to E somewhat higher
    than 1TeV without much dilution
  • If assumed this change of geometry strategy, and
    limit E to 1TeV CM, can say that BDS is longer
    than needed
  • Even though FF and energy collimation could be
    shortened (if limit E strictly to 1TeV CM), with
    two IPs the limit is in transverse distance
    between detectors, i.e. the sum of crossing
    angles
  • A radical way to attempt to reduce cost but keep
    two IRs may be to shorten ff and E-coll by 500m
    per side, increase sum of crossing angles to
    35mrad, and use single IR hall
  • BDS entry to entry will be 4.5km or a bit less
  • Length of beamlines and vac. chamber reduced by
    (also shorten extraction lines) by 3km

53
ILCFF9 (current BDS for 20mrad IR) vs beam energy
with 1TeV nominal emittances and IP beta and zero
energy spread. Without change of FD at higher
energy. No change of bend angles.
54
500GeV/beam
  • Bends are OK until 500GeV/beam, at that E there
    is a minor growth

55
  • FD is OK until 250GeV/beam, after that assume it
    is replaced

56
Backup slides
57
IR design and Radiation safety design criteria
  • Started the work on ILC radiation guidance
    document
  • beam containment policies and devices, conditions
    for occupancy
  • For IR region, in particular, defines
  • Normal operation dose less than 0.05 mrem/hr
    (integrated less than 0.1 rem in a year with 2000
    hr/year)
  • This number 0.05mrem/hr is ten times less than
    what was discussed at Fermilab RDR meeting.
    Reason to allow non-rad workers.
  • Accidents dose less than 25rem/hr and
    integrated less than 0.1 rem for 36MW of maximum
    credible incident (MCI)
  • The team initially included N.Mokhov, D.Cossairt,
    L.Keller, S.Rokni, A.Fasso
  • The document sent for discussion to colleagues
    from CERN, DESY, KEK, SLAC, Fermilab
  • Alberto Fasso, Lewis Keller, Nikolai Mokhov,
    Sayed Rokni, Tom Himel, Nobuhiro Terunuma,
    Eckhard Elsen, Hideo Hirayama, Syuichi Ban, Hans
    Menzel, Norbert Tesch, Albrecht Leuschner, Andrei
    Seryi

58
Updates on IR radiation safety design
  • Considered cases
  • non self-shielding detector, with shielding wall
    in the hall
  • self-shielding detector
  • Loss of the beam inside detector
  • loss of the beam in the drift in the pacman
  • loss of the beam near tunnel entry to the hall
  • These study allow to define
  • thickness and distance to the shielding wall, or
  • thickness of packman,
  • design of the tunnel plug
  • design of instrumented muon gaps in the detector

59
IR radiation safety1) Hall with shielding wall
18MW loss on Cu target 9r.l \at s-8m. No
Pacman, no detector. Concrete wall at 10m. Dose
rate in mrem/hr.
  • For 36MW MCI, the concrete wall at 10m from
    beamline should be 3.1m

Alberto Fasso et al
Wall
10m
60
2) Self-shieldingdetector
18MW on Cu target 9r.l at s-8m Pacman 0.5m iron
and 2m concrete
  • A proper pacman can reduce dose below 25rem/hr
  • Desired thickness is in between ofthese two
    cases

18MW on Cu target 9r.l at s-8m Pacman 1.2m iron
and 2.5m concrete
color scale is different in two cases
18MW at s-8m Packman
dose at pacman external wall dose at r7m
Fe 0.5m, Concrete2m 120rem/hr
(r3.5m) 23rem/hr Fe 1.2m,
Concrete 2.5m 0.65rem/hr (r4.7m)
0.23rem/hr
61
Muon gaps
  • 18MW loss on a Cu target (9r.l) placed at s-8m
  • Pacman is 0.5m iron and 2m concrete
  • With muon chambers (with gap 5cm)
  • Dose rate shown is in mrem/hr

62
Self-shielding (cont.)
Illustrate two designs os the tunnel plug In
the first one there is not enough overlap of
shielding in the tunnel-hall transition next
iteration of design show satisfactory performance
Proper tunnel plug can be designed
0.5-0.7m of Fe for last 5m of tunnel 0.5m
transverse Fe shielding in Pacman
Iron
color scale is different in two cases
63
Preliminary conclusions from all cases
  • Self-shielding detectors
  • issue of gaps (muon chambers) need to be solved
  • entrance of tunnel should be covered by iron as
    well
  • distance helps (fencing out the working detector)
  • if issue of gaps can be solved, 3m pacman is
    needed to meet 25rem/hr MCI requirement
  • accurate model of detector is very important
  • studies will continue
  • Non self shielding detector
  • concrete wall at 10m from beamline should be 3.1m
About PowerShow.com