ILC prototype undulator project

ILC positron source meeting 31stJan – 2rd feb 2007

@IHEP Beijing

James RochfordFor the

HeLiCal Collaboration

HeLiCal Collaboration

CCLRC Technology Rutherford Appleton Laboratory:D.E. Baynham, T.W. Bradshaw, A.J. Brummitt, F.S. Carr, Y.

Ivanyushenkov, A.J. Lintern, J.H. Rochford

CCLRC ASTeC Daresbury Laboratory and Cockcroft Institute:A. Birch, J.A. Clarke, O.B. Malyshev, D.J. Scott

University of Liverpool and Cockcroft Institute: I.R. Bailey, P. Cooke, J.B. Dainton, L.J. Jenner, L.I. Malysheva

University of Durham, CERN and Cockcroft Institute :G.A. Moortgat-Pick

DESY:D.P. Barber, P. Schmid

Talk outline

• Specification of an undulator for the ILC• 3d modelling of the undulator

– Defines period and bore for operating point 80%

• Why operate at 80% short sample • Benefits of operating at 90%• Latest test results from experimental prototypes• Summary

Undulator specification

• Undulator period: as close as possible to 10 mm• Field on axis: to produce 10 MeV photons (first

harmonic)• Field homogeneity: ≤1% • Vacuum bore: to have beam stay clear of 4 mm

=> about 5 mm for vacuum bore and about 6 mm for magnetic bore

• Superconductor (NbTi) working point : about 80% of short sample critical current.

• Modular design -module length: 4 m

Modelling-predictions

3d model of undulator with iron poles

High mesh density Iron

Conductor

Bore

To deliver 10MeV photons assuming 7 wire wide

1:1 NbTi winding with operating point of 80%

10.0

10.5

11.0

11.5

12.0

12.5

13.0

13.5

14.0

2.0 4.0 6.0 8.0

Bore(mm)

Pero

id (

mm

)

3d models (wire peak)

3d models (block peak)

2d models

Linear (3d models(wire peak))Linear (2d models)

Linear (3d models(block peak))

Conclusion for NbTi a period of 10 mm means very small bore -

unpractical!Realistic figures:

Beam stay clear – 4 mmVacuum bore - 5 mmWinding bore - 6 mmPeriod - 11.5 mm

Modelling predicts relationship between the bore and period

Why operate at 80% 1

The RAL design is based on operation the magnet at 80% along the magnet peak field load line

This choice is one of the factors which limits the minimum period that can be obtained for a given bore

A design based on operation at 90% could allow some reduction in period or an increase in the bore

The justification for operation at 80% is now given

The gains from operation at 90% are quantified

Factors defining operating point/margin

Variation in operating temperature•Local variations within the magnet

•Global due to operation from a refrigerator at higher To or higher pressure

Variation in Jc superconductor value ? 1-2% mabye as much as 5%

Variation in Cu:Sc

Figure typically quoted (+/-10%)

Why operate at 80% 2

Magnet quenching•Enthalpy margin

•simplest criterion – energy to raise winding from Top to Tc (critical temperature)

•Minimum propagating zone (MPZ) – minimum quench energy (MQE)

•Defines the size of normal (non-superconducting) zone which can exist without developing into a magnet quench

•Characteristic length L~((2k(Tc-To))/(Jc^2xrho))^0.5

•k=thermal conductivity,Tc= critical temp, To= Op temp,Jc= critical current density, rho =resistivity

•For the undulator typically – MPZ length ~ 10mm Energy to Quench ~ 50microjoules

•Reliable operation is the balance between energy releases in the winding (wire movement-resin cracks) and stability margin

•In the undulator the energy releases will probably be small ( stresses are small) – very difficult to estimate energy release spectrum

Why operate at 80% 3

Magnet Op points

0

200

400

600

800

1000

1200

1400

1600

0 0.5 1 1.5 2 2.5 3 3.5

Field Tesla

Win

din

g C

urr

en

t De

nsi

ty

Jc wind@4.2K

Jc wind @Top

Jwind vs Bp

Jop

This is calculated for Prototype 4

operating at 80% at 4.2k

It equates to a temperature margin

of 1.2K i.e it equivalent to

operating at 100% at 5.4K

J cNbTi(5T,4.2) A/mm 2̂ 2824

Top K 5.4

Cu:Sc 1.35wire overall diameter mm 0.44Ic 5T 4.2K Amps 151.01Axis Field B(T) 0.79Bp operating field B(T) 2.3J w operating current density A/mm 2̂ 855Iop Amps 166J w/J c along load line(4.2K) % 80Temp margin K 1.2

Why operate at 80% 4

What does the operating point mean in terms of temperature margin?

% %Operating Point 80 90Operating Temperature K 4.2 4.2Critical temperature K 5.4 4.8Temperature margin K 1.2 0.6Enthalpy Margin relative 539 220Cu:Sc 0.9 0.9tolerance = -0/+0.1Cu:Sc max 1 1Operating Temperature K 4.2 4.2Critical temperature K 5.2 4.6Temperature margin K 1 0.4Enthalpy Margin relative 420 137

J c NbTiDesign(NbTi,5T,4.2K) A/mm 2̂ 2824 2824Actual - 5% 2542 2542Operating Temperature K 4.2 4.2Critical temperature K 5 4.3Temperature margin K 0.8 0.1Enthalpy Margin relative 314 31

Operating TemperatureOperating Temperature design K 4.2 4.2Operating Temperature Operation K 4.5 4.5Critical temperature K 5.4 4.8Temperature margin K 0.9 0.3Enthalpy Margin relative 440 121

Operating point

80% to 90% halves the enthalpy dt 1.2k- 0.6K

Why operate at 80% 5

Sensitivity analysis

Aim here is to quantify effect of parameter variation on temperature margin and enthalpy margin. Note the enthalpy margin is a relative value

Variation in Cu:SC

@80% dT reduced by 0.2K

@90% dT reduced by 0.2K

Variation in operating temp

@80% dT reduced by 0.3K@90% dT reduced by 0.3K

Variation in JC in NbTi@80% dT reduced by 0.4K@90% dT reduced by .5K

Benefit operating at 90% 1

Period Vs bore for diff erent operating points Assumed

J c=2850@5T,4.2K, f rom Vac

10.0

10.5

11.0

11.5

12.0

12.5

4.0 5.0 6.0 7.0 8.0Bore(mm)

Peri

od (

mm

)

7 wire 1:1 NbTi ribbon operating at 80% of SS

7 wire 1:1 NbTi ribbon operating at 90% of SS

Linear ( 7 wire 1:1 NbTi ribbon operating at80% of SS)Linear ( 7 wire 1:1 NbTi ribbon operating at90% of SS)

Gains

Reduction in magnet period estimated for operation at 90% of short sample rather than 80% ~ 0.25mm

Increase in bore ~ 0.5mm

Only one of these is available

Reduction in manufacturing cost is insignificant

Magnets should operate in a long string without quenching

Mass production in industry – variations in fabrication quality

Need allowances for

•variations in operating temperature (pressure)

•variations in wire properties Jc(NbTi) and Cu:Sc ratio

•for enthalpy margin/stability

All the effects have been quantified in previous slide

If these effects are assumed to be additive – which they can be – An operating point at 90%of short sample wire performance leaves magnet operation vulnerable to very small variations in fabrication and operating parameters.

Whilst the benefits of increasing the operating point to 90% are marginal

So we consider 80% to be the correct design choice for the undulator

Operating point – conclusion

Prototypes programme 1

Experimental programme at RAL:

•To verify modelling and prove technology

•Series of five prototypes

•Last one has just been tested

Final 500mm long

prototype-

Prototypes programme 1

Experimental programme at RAL:•To verify modelling and prove technology

I II III IV V

Former material Al Al Al Iron Iron

Pitch, mm 14 14 12 12 11.5

Groove shape rectangular trapezoidal trapezoidal trapezoidal rectangular

Winding bore, mm 6 6 6.35 6.35 6.35

Vac bore, mm 4 4 4 4.5

(St Steel tube)

5.23*

(Cu tube)

Winding 8-wire ribbon,

8 layers

9-wire ribbon,

8 layers

7-wire ribbon,

8 layers

7-wire ribbon,

8 layers

7-wire ribbon,

8 layers

Sc wire Cu:Sc 1.35:1 Cu:Sc 1.35:1 Cu:Sc 1.35:1 Cu:Sc 1.35:1 Cu:Sc 0.9:1

Status Completed and tested

Completed, tested and sectioned

Completed and tested

Completed and tested

Test is in progress

1st results from prototype V

Prototype V fi eld profi le

-1

-0.5

0

0.5

1

0 100 200 300 400 500

Z, mm

Fiel

d on

axi

s, T

Prototype V details

Period : 11.5 mmMagnetic bore: 6.35 mmConfiguration: Iron poles and yoke

Measured field at 200A (RAL PSU):

0.822 T +/- 0.7 %

Prototypes programme 2

Prototypes programme 3

Prototype V training

0

50

100

150

200

250

300

350

0 2 4 6 8 10 12

Magent runup

Que

nch

curr

ent (

A)

23/01/2007

25/01/2007

Prototype V:•Did not go straight to full field before

quenching•Magnet warmed

between 23-25th to ~150K before

testing recommenced

•Reason for training is not

known may be due to impregnation

problem

Quench current 316A

Equates to a field of 1.1T in bore

Prototypes programme 4

Prototype V:•Achieved a higher critical field than

that predicted from manufacturer

nominal values •The implication is

the wire we are using is at pretty good in terms of super conductor

content. •If we assume the

upper limit then the observed quench current agrees

exactly with the calculated value

0

50

100

150

200

250

300

350

0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0

Field (T)

Curr

ent

(A)

Bp modelled (T)

Bo modelled (T)

Jcrit measured (A)

American Scond' data: Type 56S-53 ,d=0.4, nom' cu:sc (0.9 +/-.1)

American Scond' data: Type 56S-53,d=0.4,max cu:sc 0.8

American Scond' data: Type 56S-53,d=0.4 ,min cu:sc 1

Field on axis vs. Undulator period

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

6 8 10 12 14 16 18

Period, mm

Fie

ld o

n a

xis,

T

K=1

10.7 MeV -photons

Achieved with Al-former(Prototype I)

Achieved with Al-former(Prototype III)

Achieved with iron former(Prototype IV)

Achieved with iron formerand iron yoke (Prototype IV)

Achieved with Prorotype V(iron former and iron yoke)

Field is measured at 200A

Prototypes programme 5

Conclusions

Experimental programme at RAL:

•Modelling of a undulator capable of satisfying the ILC requirements has been completed

•The chosen the technology is to use NbTi ribbon operating at 80% of short sample

•The testing of the series of five prototypes has been completed

•Work on the design and manufacture of a full scale 4m prototype is now the priority and is well

underway.