ATLAS The Worlds Largest Magnet System Roger Ruber Contents: - - PowerPoint PPT Presentation

▶
atlas the world s largest magnet system
SMART_READER_LITE
LIVE PREVIEW

ATLAS The Worlds Largest Magnet System Roger Ruber Contents: - - PowerPoint PPT Presentation

Sep 28, 2023 20 likes •500 views

ATLAS The Worlds Largest Magnet System Roger Ruber Contents: ATLAS & LHC Magnet System Cryogenics & Vacuum Current & Controls Conclusions presented at Uppsala University, 8 th December 2006 LHC: The Large Hadron Collider

slide-1
SLIDE 1

ATLAS The World’s Largest Magnet System

Roger Ruber Contents: ATLAS & LHC Magnet System Cryogenics & Vacuum Current & Controls Conclusions presented at Uppsala University, 8th December 2006

slide-2
SLIDE 2

2 Roger Ruber - Uppsala, 8 December 2006

LHC: The Large Hadron Collider

  • Circular accelerator and collider in the 27 km LEP tunnel

– 10x higher energy – 100x higher luminosity then previous proton-proton colliders

  • General purpose machine to study the universe

– Unexplored aspects of the Standard Model

  • search for mass-generating mechanism: Higgs boson
  • search for origin of matter/antimatter asymmetry: CP-violation

– Supersymmetry: a new frame- work for matter & interactions

  • many new particles within

the mass scale of LHC

slide-3
SLIDE 3

3 Roger Ruber - Uppsala, 8 December 2006

LHC Experimental Areas

ATLAS CMS LHCb ALICE

slide-4
SLIDE 4

4 Roger Ruber - Uppsala, 8 December 2006

LHC Underground Installation

  • 9135 magnets:

– 1232 main dipole (twins) – 392 lattice quadrupole (twins)

  • 1000 main dipoles installed
  • 1 sector fully completed

(interconnections)

slide-5
SLIDE 5

5 Roger Ruber - Uppsala, 8 December 2006

LHC Current Distribution Box

  • cryogenic distribution

line completed

  • current distribution:

– 1182 HTS leads: 600 – 13,000 A – 2104 copper leads: 60 – 120 A

slide-6
SLIDE 6

6 Roger Ruber - Uppsala, 8 December 2006

ATLAS Surface Buildings

Point 1 & the Globe: in front of CERN main entrance

slide-7
SLIDE 7

7 Roger Ruber - Uppsala, 8 December 2006

ATLAS Underground Installation

2 caverns 2 main shafts give access to 50,000m3 cavern for detector

55m 32m 35m

  • 92.5m

detector cavern services cavern

slide-8
SLIDE 8

8 Roger Ruber - Uppsala, 8 December 2006

Cavern Preparation

June 2003 March 2004 Ready for detector installation Jan. 2003

slide-9
SLIDE 9

9 Roger Ruber - Uppsala, 8 December 2006

The ATLAS Detector

25 m diameter 46 m length 7000 tons Toroids: 26 m length 1320 tons

CERN Bat.40

slide-10
SLIDE 10

10 Roger Ruber - Uppsala, 8 December 2006

The Inner Tracking Detectors

~6m long, 1.1 m radius

  • Pixels
  • Silicon Strip Tracker (SCT)
  • Transition Radiation Tracker

(TRT)

TRT SCT Pixels Beam Pipe

slide-11
SLIDE 11

11 Roger Ruber - Uppsala, 8 December 2006

The Liquid Argon Accordion Calorimeter

E-M calorimeter (>22X0)

  • LAr as active material

inherently linear

  • hermetic coverage

(no cracks)

  • longitudinal segmentation
  • high granularity

(Cu etching)

  • inherently radiation hard
  • fast readout possible

tdrift =450 ns

slide-12
SLIDE 12

12 Roger Ruber - Uppsala, 8 December 2006

The Hadronic Tile Calorimeter steel absorbers & plastic scintillators

  • tiles perpendicular to beam
  • staggered in depth
  • 7.2λ thick
  • 10k channels
slide-13
SLIDE 13

13 Roger Ruber - Uppsala, 8 December 2006

The Forward Hadronic Calorimeters

Forward Calorimeter (FCAL)

  • 1st wheel: Cu matrix (2.6λ, 28X0)
  • 2nd,3rd wheel: W matrix (2x3.6λ)

Hadronic End-Cap Calorimeter (HEC)

  • share cryostat w/ 1 wheel LAr EMcal
  • 2 wheels (10 λ):

– Cu absorber (25/50mm) – 4x LAr filled 1.85mm gap

slide-14
SLIDE 14

14 Roger Ruber - Uppsala, 8 December 2006

The Muon Spectrometer

Track & trigger μ trajectory

  • 6 points
  • precision 50μ (each point)
  • maximum 4 T toroidal field
  • background of γ & n
  • follow-up position of every

measuring element with a 30μ precision 2 technologies:

  • MDT - Monitored Drift Tubes
  • RPC - Resistive Plate Chambers

(trigger)

Trigger chambers (RPC) rate capability required ~ 1 kHz/cm2

slide-15
SLIDE 15

15 Roger Ruber - Uppsala, 8 December 2006

The Superconducting Magnets

Barrel Toroid + 2 End-Cap Toroids + Central Solenoid – 4 magnets provide magnetic field for the inner detector (solenoid) and muon detectors (toroids) – 20 m diameter x 25 m long – 8200 m3 volume – 170 t superconductor – 700 t cold mass – 1320 t total weight – 90 km conductor – 20.5 kA at 4.1 T – 1.55 GJ stored energy – conduction cooled at 4.8 K – 9 years construction 98-07

The largest superconducting magnet in the world !

slide-16
SLIDE 16

16 Roger Ruber - Uppsala, 8 December 2006

Why Superconducting Magnets? Technology Drivers

  • momentum resolution

– depends on sagitta term

  • transparency

– reduction of material – choose low X0 materials

  • detector configuration

– determines magnet configuration

  • cost

– construction – operation

Solutions

– high field – large volume – superconducting – aluminium alloys – dipole spectrometer – solenoid or toroid (forward/backward symmetry) – conductor, cryostat, iron yoke – water or cryo cooling

p qBL s 8

2

≈

slide-17
SLIDE 17

17 Roger Ruber - Uppsala, 8 December 2006

Solenoid Magnet

  • Resolution

– inside solenoid: dp/p ~ {B·R2

solenoid}-1

– outside solenoid: dp/p ~ {B·Rsolenoid}-1

  • Field & Symmetry

– axial and uniform – but field lines parallel to particle path at small angles

  • Installation

– self supporting structure – iron yoke required to contain stray field (improves bending power at small angles)

slide-18
SLIDE 18

18 Roger Ruber - Uppsala, 8 December 2006

CMS: Compact Muon Solenoid

  • 16 m diameter x 21 m long
  • 12,500 tonnes total weight
  • 6 m diameter x 12 m long solenoid
  • 4 T at 19.5 kA
  • 2.7 GJ stored energy
  • 220 t cold mass, 4 layers, 5 segments
slide-19
SLIDE 19

19 Roger Ruber - Uppsala, 8 December 2006

Toroid Magnet

  • Resolution

– inside toroid: dp/p ~ sinθ {Bφ·Rin·ln(Rin/Rout)}-1

  • Field & Symmetry

– tangential field (∝1/r) – field lines perpendicular to particle path – closed field: centred on and circulating around beam (no influence on beam) – no stray field: no iron yoke required

  • Installation

– support required to keep self balance

θ

CLAS/CEBAF 1995

slide-20
SLIDE 20

20 Roger Ruber - Uppsala, 8 December 2006

1992: Proposal for LHC Experiments

EAGLE

  • 2 – 4 m thick warm iron toroid
  • total weight 26,400 tonnes!
  • SC central solenoid (r=1.2m)
  • combined cryostat with

liquid argon calorimeter ASCOT

  • 12 coil SC air core toroid

with muon spectrometer

  • 2x twin iron core end cap toroid
  • separate cryostat solenoid/LAr

(ATL-TECH-92-003/4) (ATL-TECH-93-008)

slide-21
SLIDE 21

21 Roger Ruber - Uppsala, 8 December 2006

1996: ATLAS

Central Solenoid 2 T in centre hadron calorimeter as return yoke Barrel Toroid 8 coils 4T on conductor want the magnetic field light, low density materials, for enhanced transparency 2 End-Cap Toroids, 8 coils each 4T on conductor

1992

slide-22
SLIDE 22

22 Roger Ruber - Uppsala, 8 December 2006

ATLAS Toroid Magnet

  • ptimization of field uniformity &

access vs. cost: 6 / 8 / 10 / 12 coils

  • same ampere-turns
  • less cryostats
  • high peak to operating field ratio
  • large field volume: ~7000m3
  • open structure for detector:

cryostat occupies ~2% of total volume

  • good resolution at small forward

angles

1 2 3 4 5 6 7 8 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25 2.5 2.75 3

Pseudorapidity Mean Integral B.dl T.m

8 degrees

η = -ln tan(θ/2)

slide-23
SLIDE 23

23 Roger Ruber - Uppsala, 8 December 2006

ATLAS Superconductors

90 km aluminium stabilized superconductor in 3 versions

  • Toroids (BT/ECT): 65 kA at 5 T

– 40 x 1.25 mm NbTi/Cu strand, 2900 A/mm2 at 5 T (~1700 A/strand) – co-extrusion with high purity aluminium: high RRR > 1500 – intermetalic Cu-Al bonding for current and heat transfer – size: BT = 57 x 12 mm2, 56 km ECT = 46 x 12 mm2, 25 km

  • Solenoid (CS): 20 kA at 5 T

– 12 x 1.22 mm NbTi/Cu strand, 2750 A/mm2 at 5 T – co-extrusion with Ni-doped aluminium RRR ~ 500; improved yield strength – size: 40 x 4.2 mm2, 9 km

slide-24
SLIDE 24

24 Roger Ruber - Uppsala, 8 December 2006

The Barrel Toroid

  • 8 coils, 25 x 5 m2
  • 20 kA, 4 T peak field
  • 16 support rings
  • mounted on 18 feet

& 6 bedplates

  • services via top feed

box and cryo-ring

  • 2 rails for calorimeter
slide-25
SLIDE 25

25 Roger Ruber - Uppsala, 8 December 2006

Barrel Toroid Cold Mass Integration

Scale 8 coils in separate cryostats –

  • pen structure

Force transfer ~ 1100t/coil Cold mass ~ 450t

slide-26
SLIDE 26

26 Roger Ruber - Uppsala, 8 December 2006

Barrel Toroid Cryostat Integration

Challenge Scale of components and integration accuracy Tolerances << 1 mm in 26m < 40 parts per milllion

slide-27
SLIDE 27

27 Roger Ruber - Uppsala, 8 December 2006

Barrel Toroid Coil Installation

slide-28
SLIDE 28

28 Roger Ruber - Uppsala, 8 December 2006

Barrel Toroid Assembly

  • release BT: 830 tonnes → sag 18 mm
  • 350 tonnes muon chambers → sag 27 mm

Sag ~ 30mm

slide-29
SLIDE 29

29 Roger Ruber - Uppsala, 8 December 2006

Barrel Toroid Completed

slide-30
SLIDE 30

30 Roger Ruber - Uppsala, 8 December 2006

Commissioning up to Full Current (21 kA)

Cool down 6 weeks (Jul-Aug) Powering step-by-step to 21 kA (9 Nov’06) At each step: slow dump, fast dump & re-cool down Full Current 21 kA Ramp-up / down = 2 h + 2 h E = 1.1 GJ Fast dump Tmax = 55 K Cryo-recovery = 84 h

Barrel Toroid Ramp, Slow-Dump, Fast Dump, Cryo-recovery times

10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 5 10 15 20 25

Current (kA) Ramp, SD & FD times (min)

10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

Cryo-recovery time (hrs)

SD time FD time Ramp time (2.5A/s) min CryoRecovery time hrs

slide-31
SLIDE 31

31 Roger Ruber - Uppsala, 8 December 2006

The End-Cap Toroids

  • 2 x 8 coils,

4 x 4.5 m2

  • 20 kA, 4 T peak
  • torus assembly
  • 8 keystone boxes
  • hanging on bore

tube

slide-32
SLIDE 32

32 Roger Ruber - Uppsala, 8 December 2006

End-Cap Toroid Cold Mass Assembly

2x + KSB coil winding

slide-33
SLIDE 33

33 Roger Ruber - Uppsala, 8 December 2006

End-Cap Toroid Cryostating

slide-34
SLIDE 34

34 Roger Ruber - Uppsala, 8 December 2006

End-Cap Toroid Vacuum Vessel

thermal shield bore tube + gravity rods

slide-35
SLIDE 35

35 Roger Ruber - Uppsala, 8 December 2006

End-Cap Toroid Ready for Closure

slide-36
SLIDE 36

36 Roger Ruber - Uppsala, 8 December 2006

The Central Solenoid

2 T at 7730 A serving the inner tracking detector

9 km conductor (NbTi/Cu + Al-stab.) pure-Al quench prop. 5 tonnes cold mass 2.4 m bore x 5.3 m long 39 MJ at 2 T, 7.73 kA 0.66 radiation lengths

slide-37
SLIDE 37

37 Roger Ruber - Uppsala, 8 December 2006

To provide high field with minimizing wall material

  • Develop high strength superconductor

Ni-doped Aluminium-stabilizer:

– mechanical reinforcement with keeping quench stability

  • Integrate CS in common cryostat

with LAr calorimeter

  • pure Al-strip quench propagator
  • Sophisticated current &

cryogenics feeding

– 3D chimney design – Full integration at CERN

How to meet the Requirements?

50 100 150 200 1975 1980 1985 1990 1995 2000 200

Yield Strength [MPa] Year

(Pure-Al) ASTROMAG (Al-Si) SSC/SDC (Al-Zn/Si) LHC/ATLAS (Al-Ni) BESS-Polar (Al-Ni)

Ordinal Copper

slide-38
SLIDE 38

38 Roger Ruber - Uppsala, 8 December 2006

To provide high field with minimizing wall material

  • Develop high strength superconductor

Ni-doped Aluminium-stabilizer:

– mechanical reinforcement with keeping quench stability

  • Integrate CS in common cryostat

with LAr calorimeter

  • pure Al-strip quench propagator
  • Sophisticated current &

cryogenics feeding

– 3D chimney design – Full integration at CERN

How to meet the Requirements?

slide-39
SLIDE 39

39 Roger Ruber - Uppsala, 8 December 2006

To provide high field with minimizing wall material

  • Develop high strength superconductor

Ni-doped Aluminium-stabilizer:

– mechanical reinforcement with keeping quench stability

  • Integrate CS in common cryostat

with LAr calorimeter

  • pure Al-strip quench propagator
  • Sophisticated current &

cryogenics feeding

– 3D chimney design – Full integration at CERN

How to meet the Requirements?

2x field connection Barrel Cryostat Control Dewar chimney cryogenics distribution box

slide-40
SLIDE 40

40 Roger Ruber - Uppsala, 8 December 2006

Central Solenoid Installation

October 2004: going down 100 m …

in the shaft fit into TileCal

slide-41
SLIDE 41

41 Roger Ruber - Uppsala, 8 December 2006

4 November 2005: in position

Solenoid

slide-42
SLIDE 42

42 Roger Ruber - Uppsala, 8 December 2006

Excitation to full field: 8 kA (1 August)

after closing TileCal End-Caps

  • Ramp in steps:

7730 A = 2 T

  • repositioning

accuracy ±0.1 mm

  • final position

0.0 ± 1.4 mm (relative IP) 7730 A 2.00 T 7980 A slow dump voltage (blue) current (red) Coil length shrinkage, linear to I2

slide-43
SLIDE 43

43 Roger Ruber - Uppsala, 8 December 2006

The Magnetic Field

  • Excitation reproducibility

– field: < 0.1x10-4 T – current: < 5 ppm

  • No hysteresis effect iron

– iron contribution ~3.5 %

  • Accurate measurements:

– possible to identify details in winding structure

  • After first corrections:

– +7x10-4 ~ -13x10-4 T

  • Improvements (8 Nov.’06)

– ±4x10-4 T (RMS)

Higher winding density

Conductor joint = missing 1 turn to be studied

Fit using simple helical conductor model (large radius) [courtesy S.Snow] Fit using detailed conductor position [courtesy S.Snow]

slide-44
SLIDE 44

44 Roger Ruber - Uppsala, 8 December 2006

Helium Cryogenics

  • shield refrigerator:

20 kW at 40~80K

  • helium liquefier:

6 kW at 4.5 K

  • toroid helium circulation pump

1.2 kg/s at 0.4 b

  • local valve boxes to regulate

helium flow in each magnet In case of power failure:

  • thermo-syphon solenoid
  • PCS for toroid pumps
slide-45
SLIDE 45

45 Roger Ruber - Uppsala, 8 December 2006

Experience of a General Power Cut

Cryogenics Operation, 29 July 2006 (Central Solenoid)

05:50 UTC: (6:50 Geneva) CERN wide power cut

  • Auto-switch to thermo-syphon mode,
  • Cooling kept for ~ 2 hours, (Sufficiently long for safe slow discharge)

07:40: LHe empty

  • Coil start to warm up

10:00: restart refrigerator 11:30: coil re-cooled down 19:00: LHe level recovered

power cut cold restart empty

~ 13 hours

LH Level

Coil Temp.

slide-46
SLIDE 46

46 Roger Ruber - Uppsala, 8 December 2006

Current & Control

  • 24 kA/24 V (T) + 8 kA/8V (S)

power converters

  • current run down unit

solenoid / toroids

  • 150 m Al. bus-bars
  • magnet safety system

quench detection and protection

  • magnet control system
slide-47
SLIDE 47

47 Roger Ruber - Uppsala, 8 December 2006

Connections ECT Services

Tower and services chains installation in cavern

slide-48
SLIDE 48

48 Roger Ruber - Uppsala, 8 December 2006

Summary

CS & BT commissioned

  • Operation without problems
  • Field reproducible in 10-5
  • Safe operation confirmed

CS & BT completed

  • Functioned as a front runner

– First to deal with various challenges in construction and operation, – The first superconducting magnet

  • perated in LHC underground areas!

ECT on fast track to completion Many thanks to all collaborators in ATLAS, CERN, KEK, CEA, RAL, cooperation of BNL, and a lot of companies, small and big.