PERMANENT MAGNET QUADRUPOLE FOR THE LINAC 4 CCDTL

PERMANENT MAGNET QUADRUPOLE FOR THE LINAC 4 CCDTL Linac 4 Beam Coordination Committee Meeting 15/02/2011 Alexey Vorozhtsov, Evgeny Solodko, Pierre-Alexandre Thonet TE-MSC-MNC

Outline Ø Ø Ø Ø Permanent magnet solution Permanent magnet quadrupole general view Permanent magnet material Magnetic design Fabrication of the prototype Magnetic measurements Conclusions & Future actions Pierre-Alexandre THONET Linac 4 BCC - 15. 02. 2011 2

Permanent magnet solution Why a PM solution? • Only 204 mm available between 2 CCDTL accelerating tanks. • Beam energy will remain the same at each magnet emplacement. • Savings (power supply, cables, instrumentation and operation of the magnets). Requirements: Number of magnets : Integrated gradient (Max): Integrated gradient (Min): Inner diameter (Min): Outer diametre (Max): Length : Gradient integral error (rms): Magnetic versus geometric axis: Harmonic content at 15 mm radius: Bn/B 2 for n=3, 4, . . . : Yaw/pitch/roll: Each quadrupole has a different gradient Pierre-Alexandre THONET 14 installed + 3 spares 1. 6 Tesla 1. 3 Tesla 0. 040 m 0. 200 m 0. 100 m ± 0. 5 % < 0. 1 mm < 0. 05 1 mrad Linac 4 BCC - 15. 02. 2011 3

Permanent magnet quadrupole general view Return yoke C 10 R steel. Tuning blocks C 10 R steel, to compensate the possible p. m. inequalities and set the field gradient (up to a reduction of 6 % of the nominal value). Mechanically movable up to 6 mm, independently for each pole. The exact position of the blocks is assured by non-magnetic spacers of different thickness. Pole tip C 10 R steel, smooth the possible differences on the easy axis orientation of the permanent magnet blocks. Permanent magnet blocks Sm 2 Co 17, as a flux generator. Core aluminium, structure maintaining all the parts together, giving a high precision on the poles positioning. Advantage of this design: • Possibility to compensate the differences on the permanent magnet blocks. • Possibility to set the field gradient. 3 types of quadrupoles (low, medium and high gradient) cover an integrated gradient range from 1. 3 T to 1. 6 T. Pierre-Alexandre THONET Linac 4 BCC - 15. 02. 2011 4

Permanent magnet material Samarium Cobalt Sm 2 Co 17 • Maximum specific energy product fulfils with the magnet design. • Small temperature coefficient: 0. 035%/°C. • Good radiation resistance. • Acceptable corrosion stability without protective coating. Sm 2 Co 17 Recoma 30 S from “ARNOLD MAGNETICS” • High remanent field Br=1. 12 [T]. • Small deviation of the magnetic characteristics (maximum 3% from the nominal values on Br and Hc). • ± 2[deg] maximum error of easy axis orientation. Pierre-Alexandre THONET Linac 4 BCC - 15. 02. 2011 5

Opera 2 D model Pierre-Alexandre THONET Linac 4 BCC - 15. 02. 2011 6

Field quality 2 D model (ideal case) • Field gradient = 18 [T/m] which is 2 T/m higher than the requirements. • 3 D modeling suggested that, due to the short magnet length and relatively large aperture, the central gradient Grad(z=0) is not as high as in 2 D model (end effects domination). Normal relative field components bn/b 2 [10 -4] at R=15 mm 0. 4 bn*104/b 2 0. 3 0. 2 0. 1 0 -0. 1 -0. 2 B 4 Pierre-Alexandre THONET Linac 4 BCC - 15. 02. 2011 B 6 B 8 B 10 B 12 B 14 B 16 B 18 B 20 B 22 B 24 B 26 B 28 B 30 Harmonic number 7

Opera 3 D model ~1. 7 [T/m] difference due to the 3 D effects Pierre-Alexandre THONET Linac 4 BCC - 15. 02. 2011 8

Integrated relative field components at R=15 mm Absolute value of integrated relative field components 1. 00 E-03 abs[∫Bndz/∫B 2 dz] no chamfer 1 mm 1. 00 E-04 chamfer 2 mm chamfer 3 mm 1. 00 E-05 chamfer 4 mm chamfer 5 mm 1. 00 E-06 chamfer 6 mm 1. 00 E-07 B 4 B 6 B 8 B 10 B 12 B 14 B 16 B 18 B 20 Harmonic number B 22 B 24 B 26 B 28 B 30 Minimum dodecapole component is mandated for chamfer height between 5 and 6 mm Chamfer height [mm] B 6 0 1 2 3 4 5 6 Pierre-Alexandre THONET Integrated relative field components ∫Bndz/∫B 2 dz at R=15 [mm] B 10 B 14 -2. 8 E-03 -2. 28 E-03 -1. 54 E-03 -9. 14 E-04 -4. 43 E-04 -1. 12 E-04 1. 10 E-04 3. 57 E-05 -4. 66 E-05 -8. 59 E-05 -8. 71 E-05 -7. 99 E-05 -7. 42 E-05 -7. 30 E-05 Linac 4 BCC - 15. 02. 2011 -1. 31 E-05 -5. 27 E-06 -9. 3 E-06 -1. 21 E-05 -1. 29 E-05 -1. 32 E-05 -1. 28 E-05 9

Relative dodecapole component B 6(z) at R=15 mm Pierre-Alexandre THONET Linac 4 BCC - 15. 02. 2011 10

Fabrication of the prototype (1) • Fabrication of ≈80 mm long permanent magnet blocks and gluing by pair. • Cutting and magnetization of slices for each block to check the magnetic properties. • Gluing of the blocks together with the soft steel pole tip. • Final grinding. • Magnetization and measurement of magnetic properties of each pole assembly. Pierre-Alexandre THONET Linac 4 BCC - 15. 02. 2011 11

Fabrication of the prototype (2) • The aluminium core was made by EDM cutting. • All the soft steel parts were machined and ground. • The assembly of the magnet was done carefully by hand. Pierre-Alexandre THONET Linac 4 BCC - 15. 02. 2011 12

Magnetic measurements First Results: • An integrated gradient of 1. 77 T. m/m has been measured with a rotating coil which is exactly the value of the Opera 3 D model. • The average field harmonics have been measured without shims on the tuning blocks and they are already much lower than the requirements. n bn an Unit Tolerance 3 -5. 2 8. 5 units @15 mm 500 4 6. 1 0. 5 units @15 mm 500 5 -0. 3 -1. 3 units @15 mm 500 6 -2. 2 (-1. 12) 0. 8 units @15 mm 500 7 0. 1 0. 0 units @15 mm 500 8 0. 0 units @15 mm 500 9 -0. 1 units @15 mm 500 10 -0. 6 (-0. 7) 0. 1 units @15 mm 500 11 0. 0 units @15 mm 500 12 0. 0 units @15 mm 500 13 0. 0 units @15 mm 500 14 0. 0 units @15 mm 500 Field harmonics measured at 17 mm and corrected at 15 mm Next measurements: • We could improve field quality by moving tuning blocks (but already 100 times better than requirements). • Measurement of minimum quadrupole gradient by adjusting the tuning blocks. • Measurement of the magnetic center vs geometric center. Pierre-Alexandre THONET Linac 4 BCC - 15. 02. 2011 13

Conclusions & Future actions • The first results from the magnetic measurements confirm the excellent field quality provided by this design. • Definition of an adapted positioning system of the magnet function of measurements results of magnetic center vs geometrical center and tilt. • Writing of a technical design report once all the tests will be completed. • Production of the 17 magnets (14 installed + 3 spares) after the final validation (delivery time ≈ 12 months). Pierre-Alexandre THONET Linac 4 BCC - 15. 02. 2011 14

Thank you for you attention! Pierre-Alexandre THONET Linac 4 BCC - 15. 02. 2011 15

Additional slides Pierre-Alexandre THONET Linac 4 BCC - 15. 02. 2011 16

p. m. easy axis orientation errors Bn*104/B 2 Normal relative field components Bn/B 2 [10 -4] at R=15 mm 0. 5 0. 4 0. 3 0. 2 0. 1 0 -0. 1 -0. 2 Ideal case easy axis error case 1 easy axis error case 2 B 4 B 5 B 6 B 7 B 8 B 9 B 10 B 11 B 12 B 13 B 14 B 15 B 16 B 17 B 18 B 19 B 20 B 21 B 22 Harmonic number Scew relative field components An/B 2 [10 -4] at R=15 mm 6 An*104/B 2 5 4 3 Ideal case 2 easy axis error case 1 1 easy axis error case 2 0 -1 A 2 A 3 A 4 A 5 A 6 A 7 A 8 A 9 A 10 A 11 A 12 A 13 A 14 A 15 A 16 A 17 A 18 A 19 A 20 A 21 A 22 Harmonic number Pierre-Alexandre THONET Linac 4 BCC - 15. 02. 2011 17

Positioning of the magnet in space • The magnet will be positioned on the referential plate with a key (x axis). • The referential plate will be machined in order to have the magnet center well positioned. • An adjustment will be possible on the z axis. Referential plate y x Key Pierre-Alexandre THONET Linac 4 BCC - 15. 02. 2011 z 18

Budget Price for the prototype (CHF) Permanent magnet blocks Unit price for 19 magnets (CHF) 5000 2900 900 Magnetic and non magnetic parts 3800 2000 Tooling (machining and assembly) 2600 Magnet assembly Unit price 12300 Total price for 17 magnets Pierre-Alexandre THONET 5800 98600 Linac 4 BCC - 15. 02. 2011 19