corrector development

The HiLumi LHC Design Study (a sub-system of HL-LHC) is co-funded by the European Commission within the Framework Programme 7 Capacities Specific Programme, Grant Agreement 284404.

Corrector Development

Status of the Corrector R&D in the Frame of the Phase I Upgrade Project

M. Karppinen CERN TE-MSC

M. Karppinen CERN TE-MSC 2

AcknowledgementsSTFC-RAL, UK

A. Brummitt, M. Courthold, S. Jones

CIEMAT Spain

P. Abramian, F. de Aragón, J. Calero, J. de la Gama, L. García-Tabarés, J. L. Gutiérrez, T. Martínez, E. Rodríguez, L. Sánchez, F. Toral, C. Vázquez

CERN

N. Dalexandro, N. Elias , L. Favre, O. Gumenyuk, A. Kuzmin, M. Karppinen, J. Mazet, L. Oberli, J-C. Perez, D. Smekens, V. Sytnik, G. Trachez, G. Villiger

14/1

1/20

12


Outline• IR Corrector layout & parameters for Phase I

upgrade

• Radiation environment

• Orbit correctors

• Skew quadrupole

• Higher order multipoles

• Summary

14/1

1/20

12


IR Corrector Layout for Phase ICorrector Package (CP)

14/1

1/20

12


Corrector Package (CP)

Current Integrated strength (field) Coil Aperture

MCXB (B1/A1) +/- 2.4 kA (6 Tm 2.5 Tm ) 1.5 Tm 140 mm

MQXS (A2) +/- 2.4 kA (0.8 Tm ) 0.55 Tm@40 mm 140 mm

MQXS MCXBH

~1 m~0.9 m

MCXBV

~0.5 m

MCXT (B6 ) +/- 120A 0.075 Tm @ 40 mm 140 mm

MCXO (B4 ) +/- 120A 0.035 Tm @ 40 mm 140 mm

MCXSO (A4 ) +/- 120A 0.035 Tm @ 40 mm 140 mm

MCXSS (A3 ) +/- 120A 0.055 Tm @ 40 mm 140 mm

MCXS (B3 ) +/- 120A 0.055 Tm @ 40 mm 140 mm

IPMCXSSMCXS

MCXSOMCXO

MCXT

~0.5 m~0.5 m ~1 m

14/1

1/20

12


Correctors in Q2

• Base-line (HV and VH) orbit corrector scheme allows controlling the orbit to a level 3 times larger that then BPM resolution.

• To reach the same level as the effective BPM resolution :• Provide 1.5 Tm (1.8 Tm) in H&V-plane in BOTH locations.

• Feasibility study was initiated on combined H/V-corrector that meets the reliability requirements

• An extra H/V pair means:• Magnet R&D, material R&D, design, component & tooling procurement

• Additional powering and protections circuits

7 m7 m

MCXBH/V

(H&V)

1..1.3 m

MQXC MQXC

Q2aQ2b Q1

MQXC

10 m

Q3

MQXC

10 m 1..1.3 m

MCXBV/H

(H&V)

REF: S. Fartoukh, R. Tomas, J. Miles: “Specification of the Closed Orbit Corrector magnets for the NEW Inner Triplets”, sLHC Project Report 030

14/1

1/20

12


Expected Radiation Levels CP & Q2 • Luminosity: 2 L0 = 2 ×1034 cm-2 s-1 & 1000 fb-1

• Peak dose CP:~50..65 MGy ø120 mm, no shield~30..35 MGy ø140 mm, no shield~10 MGy ø140 mm, 10 mm SS

• Peak dose in Q2 (with 13 mm liner in Q1):~28 MGy, ø120 mm, no shield~ 8 MGy ø140 mm, 10 mm SS

• Base-line for IR correctors: ø140 mm coil aperture with 10 mm Stainless steel shielding (i.e. ø120 mm free aperture)

Courtesy of F. Cerrutti & A. Mereghetti EN-STI, FLUKA-team

14/1

1/20

12


Requirements..• Very hostile environment• Material selection (insulation, head spacers, shielding etc..)

• Spare policy

• “Intervention friendly” design of cryo-magnets

• Radioprotection

• Reliability:• MCXB must work, no redundancy

14/1

1/20

12


Unit

Integrated field Tm 6

Nominal field T 4.0

Mag. length m 1.50

Nominal current A 2438

Stored energy kJ 233

Self inductance mH 78

Working point <75%

Cable width/mid-height mm 4.37 / 0.845

Total length m 1.8..2

Aperture mm Ø140

Total mass kg ~2700

MCXB Initial 2-Layer Design (Bint = 6 Tm)

14/1

1/20

12


MCXB and MQXS(1: Strand & CableStrand parameters

Cu:Sc 1.75

Strand diameter 0.48 mm

Metal section 0.181 mm2

No of filaments 2300

Filament diam. 6.0 µm

I(5T,4.2K) 203* A

jc 3085* A/mm2

Cable ParametersNo of strands 18

metal area 3.257 mm2

cable thickness 0.845 mm

Cable width 4.370 mm

cable area 3.692 mm2

metal fraction 0.882

Key-stone angle 0.67 degrees

Inner Thickness 0.819 mm

Outer Thickness 0.870 mm

275 km SC-strand in stock at CERN Polyimide Insulation: 2 x 25µm + 55 µm (in stock at CERN)

14/1

1/20

12

*) extracted strand March -09

1) 3.5–mm-wide 14-strand cable was developed for the initial MQXS design


Single Plane MCXBH/V Design OptionsTwo-layer design, 6 Tm / 4 T:• Conceptual magnetic and mechanical design completed.

• Sensitive for the HX-hole location and diameter (saturation effects).

• Overall length ˜1.8 m.

• Based on 18-strand cable successfully produced at CERN.

• Over-designed for the updated strength requirements. For 2.5 Tm could run at reduced current and/or shorten to ˜1.5 m

• Not enough SC strand stock (275 km) for the total no. of magnets (with 8 MQSX).

Single-layer design 1.5..2.5 Tm / 2.3 T:• Engineering design completed.

• Less sensitive for the HX-hole dimensioning (as long as in 45 degrees…).

• Overall length ˜1.1..1.4 m

• Better adapted for stay-clear collars.

• Better adapted for the new “spec”.

• Existing SC strand stock (275 km) would be sufficient for 45 MCXB coils and 9 MQSX magnets.

14/1

1/20

12


Unit

Integrated field Tm 1.5 (2.5)

Nominal field T 2.3

Mag. length m 0.65


Stored energy kJ 28


Working point 50%


Total length m ~1.1 (1.4)

Aperture mm Ø140

Total mass kg ~2000

MCXB Single-Layer Design

Ø140

Ø570

14/1

1/20

12


MCXB 3D harmonics (2 x return end)

14/1

1/20

12

˜260 mm

B1= 0.34 Tm x 2 + 2.3T x 0.36 m = 1.5 Tm

ENDS STRAIGHTb3= 0.74 unitsb5= 2.86 unitsb7= -0.41 unitsb9= -1.65 unitsb11= -0.55 units

Coil length = 0.9 mTotal length = ˜1.1 m


MCXB 4-Block Design Quench (3kA)

14/1

1/20

12

Rd = 0.16 ΩWarm diodeNo heatersTmax < 90 K


MCXB 150 mm Mechanical Model• All components for the

model magnet in stock

14/1

1/20

12


Combined H/V-Dipole Development PlanSingle plane model magnet #1

Single layer coils Porous polyimide insulationStatus: winding trials done

Single plane model magnet #2 Same coil designResin impregnated coilsBraided S2-glass insulationStatus: insulated cable characterization started

Combined H/V magnet #3 Nested H/V-dipolePotted or porous coils?Status: Mechanical Concept & FEA in started.

Today we do not have a design concept for nested MCXB meeting the requirements for IR.

14/1

1/20

12


Nested MCXB Conceptual Design• Magnetic optimization of the nested

design has been done for a few possible configurations

• The analysis of possible mechanical concepts based on nested collaring initiated

Torque: 90’000 Nm/m Shear stress at coil interface ~2.5 MPa.

14/1

1/20

12


MQXS Skew Quadrupole 2-Layer Design

Unit

Nominal gradient T/m 40

Mag. length m 0.5


Stored energy kJ 19.1


Working point <55%

Cable width/mid-height mm 3.40 / 0.845(*

Cu/Sc 1.2

Total length m ~0.8

Aperture mm ø140

Total mass kg ~500

14/1

1/20

12*) 14-strand cable


MQSX Single-Layer DesignUnit

Nominal gradient T/m 28

Mag. length m 0.69


Stored energy kJ 19.4

Self inductance mH 4.3

Working point 48%


Cu/Sc 1.2

Total length m ~0.8

Aperture mm ø140

Total mass kg ~500

14/1

1/20

12


MQSX 3D harmonics (2 x return end)

14/1

1/20

12

˜120 mm

A2= 2.36 Tm/m x 2 + 28.3 T/m x 0.54 m = 20 Tm/m

ENDS STRAIGHTa6= -0.39 unitsa10= 0.01 unitsa14= 1.35 units

Coil length = 0.78 mTotal length = ~0.9 m


MQSX Single Layer Design Quench (3kA)

14/1

1/20

12

Rd = 0.16 ΩWarm diodeNo heatersTmax < 50 K

Higher Order Corrector Development

22

• In the framework of the SLHC Collaboration, CIEMAT has developed, constructed and tested two superconducting corrector magnets: the MCXS sextupole and MCXO octupole.

• The design is made in view of the very high radiation dose.• Superferric design is sufficient to produce the required field

strength1. The superconducting coils are placed further out and, to some

extent, shielded by the iron poles. 2. The simple race-track coils are easy to make and the field

quality is defined by the precision of the iron poles.

P. Abramian, F. de Aragón, J. Calero, J. De la Gama, L. García-Tabarés, J. L. Gutiérrez, T. Martínez, E. Rodríguez, L. Sánchez, F. Toral, C. Vázquez

14/1

1/20

12


Magnetic design: MCXS

23

Nominal current 100 ABare wire diameter 0.5 mmInsulation thickness 0.02 mmCu/Sc 1.55Filament size 4 μmNumber of turns 228Aperture 140 mmEffective length 137 mmOverall length 160 mmIntegrated strength (r=40 mm)

0.055 T.m

Integrated b9 0.504 1e-4Integrated b15 0.127 1e-4Integrated b21 -0.001 1e-4Non-linearity in the load line

3 %

Coil peak field 2.02 TWorking point @ 1.9 K 33.5 %Iron outer radius 140 mmSelf inductance 192 mHStored magnetic energy 960 J

14/1

1/20

12


Magnetic design: MCXO/MCXSO

24

Nominal current 100 ABare wire diameter 0.5 mmInsulation thickness 0.02 mmCu/Sc 1.5Filament size 5 μmNumber of turns 165Effective length 0.161 mIntegrated strength (r=40mm)

0.035 T.m

Integrated b12 0.052 1e-4Integrated b20 0.016 1e-4Integrated b28 -0.001 1e-4Non-linearity in the load line

2.2 %

Coil peak field 1.87 TWorking point @ 1.9 K 30.6 %Iron outer radius 125 mmSelf inductance 152 mHStored magnetic energy 758 JOverall length 180 mm

14/1

1/20

12


MCXS Sextupole Engineering Design

25

Wet impregnated race-track coils

Standard Araldite resin

Laminated ARMCO iron yoke

Alignment by stainless steel keys

Radiation resistance:Coils located further out and partially shielded by the iron poles.

14/1

1/20

12


MCXS Sextupole Fabrication

26

14/1

1/20

12


MCXO Octupole Engineering Design

27

Vacuum impregnated race-track coils

Laminated ARMCO iron yoke

Alignment by stainless steelkeys

Radiation resistance: Polyimide insulated NbTi

wire CTD 422B: a blend of

cyanate ester and epoxy resin

Stainless steel coil spacers Duratron 2300 PEI

connection plate and ancillary pieces

Insulating sleeves made of polyurethane and glass fiber

14/1

1/20

12



MCXO Octupole fabrication

14/1

1/20

12

Test results: Sextupole

29

Quench Current (A) Coil Coil peak voltage (V)

1 189 6 164 2 191.5 2 163 3 203.6 1 166 4 217.3 2 167 5 228.3 1 165

Warm magnetic measurements at CELLS

(Barcelona)Cold training test (CIEMAT)

• Training test was done in a vertical cryostat at 4.2 K.

• The first quench was at 189 A, a working point about 76% on the load line, reaching 89% at quench number 5, where we ran out of helium, without any detraining.

14/1

1/20

12



Test Results: OctupoleCold training test (CIEMAT)• Training test was done in a vertical

cryostat at 4.2 K.• The first quench was relatively low, at 48%

on the load line, although well above the nominal current.

• Afterwards, quench current was increasing slowly, with some slight detraining, till currents around 200 A, where coil 8 was repetitively triggering quench.

• After thermal cycle the first quench happened at high current, but second quench showed a significant detraining.

• Most of the quenches triggered by coil 7, were followed by a quench in coil 8. Afterwards, it was checked that some screws of the common support wedge were loose.

Quench Current (A) Coil Coil peak voltage (V)

1 120.0 1 69 2 138.3 8 92 3 153.5 2 109 4 184.0 8 114 5 174.3 5 112 6 188.1 2 115 7 189.4 2 113 8 186.7 7 113 9 195.3 8 113

10 203.1 8 114 11 203.9 8 112

Retraining 1 182.7 6 102 2 136.5 7 60 3 207.8 2 106 4 199.4 7 79 5 192.9 7 113 6 198.5 7 78 7 217.5 8 100 8 215.1 7 84 9 225.0 7 88

10 203.6 7 79

14/1

1/20

12


Summary Engineering design of the single plane orbit correctors

(MCXB) and skew quarupole (MQXS) have been completed meeting the requirements of the Phase I upgrade.

Components in stock for single-layer MCXB model magnets based on porous and vacuum impregnated coils.

MQXS and MCXB trial coils have been successfully made with polyimide insulation.

150-mm-long MCXB instrumented mechanical model was successfully assembled.

Nested MCXB conceptual design work was initiated, but halted since 3 years. Today we do not have design of a nested MCXB magnet meeting the requirements of the HiLUMI upgrade.

Superferric MCXS sextupole and more radiation resistant superferric MCXO octupole have been succesfully constructed and tested at CIEMAT.

Magnetic design of Cos-6Θ dodecapole MCXT was made.

14/1

1/20

12

corrector development

Documents

ka6 tm

corrector rd

combined hvcorrector

mm140 mmmqxsmcxbh

closed orbit corrector

vh orbit corrector scheme

mm140 mmmcxo b4 120a0

mm140 mmmcxso a4 120a0