1. Permanent Magnet Option for ILC Final Doublet Quads and Tail Folding Octupole
Y. Iwashita, T. Mihara, S.Nakamura,
M. Ichikawa, Y. Tajima, Kyoto University
M.Kumada, NIRS
C.M. Spencer, SLAC 1

2. Adjustable Permanent Magnet Quadrupole Prototype for 20mr X’ing Angle
(NEOMAX38AH) 2

3. Prototype Magnet Extra beam hole Inside View of Outer Ring Inner Ring
Bore radius
1cm
Inner ring radii
In 1cm out 3cm
Outer ring radii
In 3.3cm out 5cm
Outer ring section length
Physical length
1cm, 2cm, 4cm, 8cm
23cm
Pole material
Permendur
Magnet material (inner ring)
NEOMAX38AH
Magnet material (outer ring)
NEOMAX44H
Integrated gradient (strongest)
24.2T
Integrated gradient (weakest)
3.47T
Int. gradient step size
1.4T
Extra beam hole
Inside View of
Outer Ring
Inner Ring
Base plate
Before assembly 3

4. Magnetic Center Movement
The cm values show the Switched-On-Length
The center moves several µm for 20% strength change.
See (LINAC’04) 4

5. Recent Modification
Demonstrate a higher field gradient by reducing the bore size from ø20mm down to ø15mm.
Temperature compensation of the inner ring; 1 mm (left) and 1.6 mm (right) MS-1 trapezoidal plates are seen in a 2cm space between magnets. 5

6. Longitudinal Distribution
Vertical component By measured along Z by a
Group3 Tesla meter.
dz = 2mm,
z = 0 ~ 380mm
dx, dy = ±2mm
Bore: ø15mm 6

7. Longitudinal Distribution
Bore ø15mm
All ON
All OFF
T/m
T/m
Centroid
Centroid
Z [mm]
Z [mm]
Almost flat: Gmax ~１９０T/m
Gmin ~ ２９T/m 7

8. Gradient is high at ON region.
SWL = 4cm
SWL = 8cm
T/m
Centroid
T/m
Centroid
Centroid
Centroid
Z [mm]
Z [mm]
Gradient is high at ON region.
Magnet gaps of the inner ring affects the distribution. 8

9. Integrated Strength (GL): ∫G dz40 35 30 25 20 15 10 5
GL value is proportional to the SWL:
GLmax ~ 38 T
GLmin ~ 5.9T
@ø15,L=23cm 5 10 15
SWL [cm] 9

10. Magnetic field centroid50
Cases of
only ８ｃｍ (entrance) ON or
only ４ｃｍ(exit) ON
shows the maximum and minimum. mm
-50 5 10 15
Centroid location as a function of SWL
The excursion of the centroid is almost half of its length;
it can be reduced by using a couple of PMQ’s back to back. 10

11. Required spec’s for Final Focus
Eff.L m
Radius
PoletipB kG
Gradient
kG/m
IntG.dl
QF1
2.0
0.010
8.0
803
1605
QD0
2.2
-14.2
-1416
-3116
QDEX1
1.1
0.015
-15.0
-1000
-1060
Variable PMQ will generate 1400kG/m,
but 14mrad. case needs more study.
Brett Parker's sketch of how they be placed as sc. quads 11

12. Demagnetization by Radiation
Energy deposit
Demagnetization by 14MeV neutron
GLD
SiD
SiD (by
Takahashi)
neutron
BeamCAL
17mW
13mW
29mW
QD0
94mW
97mW
147mW
105
[n/cm2s]
SD0
11mW
QF1
16mW
18mW
15mW
SF1
0.4mW
0.3mW
1mW
T. Kawakubo, et al., The 14th Symposium on Accelerator Science and Technology, Tsukuba, Japan, November 2003, pp , in Japanese,
very preliminary results by T.Abe (university of Tokyo),
in private communication
Continuous 1mo.(2.6x106s) operation may cause about 0.01[%] of (reversible?) demagnetization on NEOMAX 32EH.
(1% for 10 years) ... needs more info. 12

13. PM-Octupole for Tail Folding13

14. 14

15. PM-Octupole for tail-folding
Bore ø32, Size ø150 for ATF2
ILC: Baseline Sc design: tip, 1TeVCM
PMO can be strong; shorter in length. 15

16. Rotation of the ring may be analog not digital way.
Mutlipole components
Rotation of the ring may be analog not digital way. 16

17. Reduced skews
Skew components can be canceled if two outer rings are incorporated and rotated in opposite direction. 17

18. Summary Adjustable PMQ realizes 140T/m@ø20mm
The excursion of lens center was measured.
New config. may be possible for smaller cross section (Needs more idea and study including lattice design)
Demagnetization seems small (1% for 10 years?) maybe OK even if 10 times more, but needs more info.
Octupole for tail folding?
Will fabricate one model (Quad or Octupole) — may be tested in ATF2 18