MSC/MME Seminar HTS Demonstrator Mechanics · MSC/MME Seminar HTS Demonstrator Mechanics GaToroid Project Jérôme Harray May 12, 2020

MSC/MME Seminar

HTS Demonstrator Mechanics GaToroid Project

Jérôme Harray

May 12, 2020

Jérôme Harray | HTS Demonstrator Mechanics - GaToroid ProjectMay 12, 2020

GaToroid Project • Demonstration of HTS

technology

• Development of a prototype in scalable conditions

• Potential mechanical impact of the presence of a pole

• Optimization of impregnation stress-state

• Minimization of cable stress-state

• Preload scenario investigation

GaToroid Project • LTS gantry design in

collaboration with hadron therapy centers

• Demonstration of HTS technology

• Design of a prototype in scalable conditions

• Mechanical evidence of design choices by numerical approach

• Rigorous analysis to enhance behavior understanding

2

Prototype designSetting the baseline

3


Prototype Layout

4


Prototype Layout

Grades

5


Prototype Layout

Spacers

Grades

6


Prototype Layout

Spacers

Grades

7

Grade jumps


Prototype Layout

Spacers

Grades

Pole

8

Grade jumps


Prototype Layout

Spacers

Grades

Outer Rim

Pole

9

Grade jumps


Prototype Layout

Bolts

Spacers

Grades

Outer Rim

Pole

10

Grade jumps


Development sequence

Winding Impregnation Cool-down LorentzAssembly

11




How to jump from one grade to another?

12




How to bolt the whole assembly?


13






How should we impregnate?

14





How to select materials?



15





How to jump from one grade to another? How to validate grade stress-state?

16

How to select materials?


ApproachNumerical implementation

17


Multi-step static analysis

t

t

t

F

T

F

(1) Bolt pretension

(2) Cool-down

(3) Lorentz forces

18



t

t

t

F

T

F

(1) Bolt pretension

(2) Cool-down

(3) Lorentz forces

19



t

t

t

F

T

F

(1) Bolt pretension

(2) Cool-down

(3) Lorentz forces

20

Jérôme Harray | HTS Demonstrator Mechanics - GaToroid ProjectMay 12, 2020 21

• Tape material composition 70 % SS & 30 % Cu

• Stacked layout anisotropic orthotropic

∼

→ ∼

HTS modeling

ref: Barth, Christian & Mondonico, Giorgio & Senatore, Carmine. (2015). Electro-mechanical properties of REBCO coated conductors from various industrial manufacturers at 77 K, self-field and 4.2 K, 19 T. Superconductor Science and Technology.

Anisotropic

Tape




• HTS cable with insulation composition 19 % SS & 81 % Cu

∼

→ ∼

∼

HTS modeling


Anisotropic Anisotropic

Tape Cable




• HTS cable with insulation composition 19 % SS & 81 % Cu

• Lack of input data simplified material model

∼

→ ∼

∼

→

HTS modeling


Anisotropic Anisotropic Isotropic

Tape Cable Model


Material thermal properties

• Consideration of different alloys for structural components

• Importation of thermal properties from database

T [K]

CTE [K−1] CTE(K)

Coefficient Thermal ExpansionCTE →





• Integrated thermal properties from K to K300 4T [K]

CTE [K−1]

4 K 300 K

CTE(K)






• Integrated thermal properties from K to K

• averaged value as input for static simulations

300 4

CTE

Al SS Ti HTS mixture

1,44E-05 9,9E-06 5E-06 1,10E-05CTEeq [C−1]


T [K]

CTE [K−1]

4 K 300 K

CTEeq

CTE(K)


Contact definition

• Type of contact between components

27

A. Bonded where resin withstand

B. Frictional with

→

μ = 0.3


Contact definition

28

A. Bonded where resin withstand

B. Frictional with

→

μ = 0.3• Type of contact between components

Contact breakdown


Contact definition

• Contact between grades and spacers

Assumed as bonded for all configurations→

29


Contact definition

• Contact between the most inner grade and the pole

Assumed as either bonded or frictional depending on the configuration→

30


Contact definition

• Contact between the most outer grade and the outer rim

Assumed as either bonded or frictional depending on the configuration →

31


Contact definition

• Contact between the cover/intermediate plate and grades/spacers/pole/outer rim

Assumed all time as frictional→

32


Contact definition

• Contact between the cover/intermediate plate and grades/spacers/pole/outer rim

Assumed all time as frictional→

33

If bonded simplified computation provides evidence of delamination→

Mechanical conceptFrom theory to practice

34


Ideally Reality

35

• Lorentz forces try to open the coil

• Load converted into hoop stress

• HTS cable works in tension

• HTS not under a critical mode of failure


Ideally Reality

• Lorentz forces try to open the coil

• Load converted into hoop stress

• HTS cable works in tension

• HTS not under a critical mode of failure

• Bizarre grade shapes

• Cool-down with difference

• Grade jumps are single points of failure

• Resin can give birth to delamination

CTE

36

ResultsPreload strategy

37


Motivations

• Importation Lorentz forces from magnetic simulations



• Lorentz forces expand the coil outward

39

Motivations




• Balance of this trend by applying some preloads inward

40

Motivations




• Balance of this trend by applying some preloads inward

• Advantage taken from cool-down phase

• Use of difference between material

• Materials with larger CTE considered for outer components

• Materials with smaller CTE considered for inner components

CTE

41

Motivations


Material configurations

HTS tape Cu/SS

42



HTS tape Cu/SS

43

Cover

Outer rim

Pole

Spacers

Intermediate plate



HTS tape Cu/SS

44

Cover

Outer rim

Pole SS

Spacers

Intermediate plate

SS

A



HTS tape Cu/SS

45

Cover

Outer rim

Pole SS SS

Spacers

Intermediate plate

SS Al

BA



HTS tape Cu/SS

46

Cover

Outer rim

Pole SS SS Ti

Spacers

Intermediate plate

SS Al Al

A B C


Material choice

• Overall stress-state dominated by the cool-down phase

• Lorentz forces have a marginal effect on the overall stress-state

• Large configurations lead to dangerously high stress-state in grades

• difference between components may not be desired

ΔCTE

CTE

1 2 30

100

200

300

1 2 30

50

100

0 0.2 0.4 0.6 0.8 10

0.5

1

1 2 30

100

200

300

1 2 30

50

100

0 0.2 0.4 0.6 0.8 10

0.5

1

1 2 30

100

200

300

ABC

1 2 30

20

40

60

80

10022


Material choice

• Overall stress-state dominated by the cool-down phase

• Lorentz forces have a marginal effect on the overall stress-state

• Large configurations lead to dangerously high stress-state in grades

• difference between components may not be desired

ΔCTE

CTE

1 2 30

100

200

300

1 2 30

50

100

0 0.2 0.4 0.6 0.8 10

0.5

1

1 2 30

100

200

300

1 2 30

50

100

0 0.2 0.4 0.6 0.8 10

0.5

1

22

Full Stainless Steel configuration preferred to match HTS cable → CTE

1 2 30

100

200

300

ABC

1 2 30

20

40

60

80

100

ResultsImpregnation strategy

49


1 2 3-30

-20

-10

0

1 2 30

50

100

1 2 3-5

0

5

10

• When bonded to the outer rim, free shrinkage is prevented resin mostly working in tension

• Delamination at the end of the cool-down

• When frictional to the outer rim, presence of the pole increases the work in compression during cool-down

• When Lorentz forces are applied, delamination occurs if bonded to the pole

• Beneficial effect of Lorentz forces if not bonded to to any pole

→1 1.5 2 2.5 3

-30

-20

-10

0

w/o - bondedw/o - frictional rimw/ - bonded

w/ - frictional rimw/ - frictional rim & poledebonding limit

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

50

100

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3-5

0

5

10

50

Contact stress-state


1 2 3-30

-20

-10

0

1 2 30

50

100

1 2 3-5

0

5

10






→1 1.5 2 2.5 3

-30

-20

-10

0



1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

50

100

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3-5

0

5

10

51



1 2 3-30

-20

-10

0

1 2 30

50

100

1 2 3-5

0

5

10






→1 1.5 2 2.5 3

-30

-20

-10

0



1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

50

100

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3-5

0

5

10

52



1 2 3-30

-20

-10

0

1 2 30

50

100

1 2 3-5

0

5

10






→1 1.5 2 2.5 3

-30

-20

-10

0



1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

50

100

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3-5

0

5

10

53



1 2 3-30

-20

-10

0

1 2 30

50

100

1 2 3-5

0

5

10






→1 1.5 2 2.5 3

-30

-20

-10

0



1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

50

100

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3-5

0

5

10

54



1 2 3-30

-20

-10

0

1 2 30

50

100

1 2 3-5

0

5

10






→1 1.5 2 2.5 3

-30

-20

-10

0



1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

50

100

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3-5

0

5

10

55



1 2 3-30

-20

-10

0

1 2 30

50

100

1 2 3-5

0

5

10






→1 1.5 2 2.5 3

-30

-20

-10

0



1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

50

100

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3-5

0

5

10

56



1 2 3-30

-20

-10

0

1 2 30

50

100

1 2 3-5

0

5

10







• No noticeable difference in terms of contact shear

→1 1.5 2 2.5 3

-30

-20

-10

0



1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

50

100

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3-5

0

5

10

57

ResultsVon-Mises stress

58


Grades stress-state

• Presence of the pole or bonding to the outer rim both lead the overall stress-state during cool-down

• Application of Lorentz forces has a stronger impact on stresses when there is no pole

• Larger overestimate due to safety factor over J × B

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

50

100w/o - bondedw/o - frictional rimw/ - bondedw/ - frictional rimw/ - frictional rim & pole

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

20

40

60

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

0.5

1

59

1 2 30

50

100

1 2 30

20

40

60

0 0.2 0.4 0.6 0.8 10

0.5

1

1 2 30

50

100

1 2 30

20

40

60

0 0.2 0.4 0.6 0.8 10

0.5

1





1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

50


1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

20

40

60

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

0.5

1

60

Grades stress-state1 2 30

50

100

1 2 30

20

40

60

0 0.2 0.4 0.6 0.8 10

0.5

1

1 2 30

50

100

1 2 30

20

40

60

0 0.2 0.4 0.6 0.8 10

0.5

1





1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

50


1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

20

40

60

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

0.5

1

61

Grades stress-state

1 2 30

50

100

1 2 30

20

40

60

0 0.2 0.4 0.6 0.8 10

0.5

1

1 2 30

50

100

1 2 30

20

40

60

0 0.2 0.4 0.6 0.8 10

0.5

1





1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

50


1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

20

40

60

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

0.5

1

62

Grades stress-state1 2 30

50

100

1 2 30

20

40

60

0 0.2 0.4 0.6 0.8 10

0.5

1

1 2 30

50

100

1 2 30

20

40

60

0 0.2 0.4 0.6 0.8 10

0.5

1





1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

50


1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

20

40

60

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

0.5

1

63

Grades stress-state

Where does the peak stress occur?

Under which stress type the cable is subjected to?

→

→

1 2 30

50

100

1 2 30

20

40

60

0 0.2 0.4 0.6 0.8 10

0.5

1

1 2 30

50

100

1 2 30

20

40

60

0 0.2 0.4 0.6 0.8 10

0.5

1


Stress distribution

• Worst case scenario for stresses during nominal conditions

• Peak stress remains below MPa

• Peak stress occurs far from jumps

• Grades perfectly work in tension

• Configuration seems viable in terms of grades overall stress-state

100

64

ResultsGrade jumps

65


Jump layouts

• Singular point of failure in winding

• Double pancake configuration

• One pancake winded clockwise

• One pancake winded anti-clockwise


Jump layouts





• Local disturbance in field homogeneity

• Jumps avoided in beam zone


Jump layouts





• Local disturbance in field homogeneity

• Jumps avoided in beam zone

• Jump design implemented in first prototype carefully analyzed

+ A couple of pairs are investigated and validated


Jump layouts

• Criteria to characterize jump validity


Jump layouts


• If spacers and grades are bonded

• Reaction force at the jumps interface

• Contact pressure

• Contact shear

p

τ


Jump layouts


• If spacers and grades are bonded

• Reaction force at the jumps interface

• Contact pressure

• Contact shear

• If resin breaks and contacts all turn frictional

• Sliding distance

• Deformation and interaction between components

p

τ

ResultsBolts assessment

72


• Assembly of components numerically modeled

• Bolt stress-state available by FEA

• Safety factor verification according to norm ISO3506

• Equivalent stress computation:

σred,B = σ2zb,max + 3 (kt ⋅ τmax)2

Bolt

73


• Assembly of components numerically modeled

• Bolt stress-state available by FEA

• Safety factor verification according to norm ISO3506

• Equivalent stress computation:

• Safety factor on each bolts of at least during all stages:

σred,B = σ2zb,max + 3 (kt ⋅ τmax)2

1.5

Bolt

74

σred,B

Rp0.2min> 1.5

ConclusionDesign guidelines summary

75


Conclusion (1)

• No preload induced by difference is desired components in Stainless SteelCTE →

76


Conclusion (1)

• No preload induced by difference is desired components in Stainless Steel

• Impregnation guideline:

• Bonding of the coil pack with the cover and the intermediate plate leads to global risk of delamination surface treatment before impregnation

• Bonding of the coil pack with the outer rim and the pole leads to local risk of delamination design of specific tooling for impregnation

CTE →

→

→

77


Conclusion (1)





• Overall stress-state in grades verified and viable

CTE →

→

→

78


Conclusion (1)






• Grade jumps implemented in the first demonstrator version are validated

CTE →

→

→

79


Conclusion (1)






• Grade jumps implemented in the first demonstrator version are validated

• Components of assembly assessed safety factor on bolts all time greater than

CTE →

→

→

→ 1.5

80


Conclusion (2)

• Two remaining viable configurations:

A. With pole - frictional with pole & outer rim

B. Without pole - frictional with outer rim

81


Conclusion (2)

• Two remaining viable configurations:

A. With pole - frictional with pole & outer rim

B. Without pole - frictional with outer rim

A B

Could reach higher mean MPa

Smaller local deformation

σVM ∼ 40 Could reach smaller mean MPa

Larger local deformation

σVM ∼ 30

Stainless Steel structural components

Work similarly well in compression at resin/resin contacts

82

What’s next?Let’s move forward

83


• Major importance of accurate determination numerically demonstrated

• Experimental characterization of cable stacked samples

• measurements in different directions

CTE

CTE

What’s next? (1)

84

stre

ss v

aria

tion

[%]

-100

-50

0

50

100

CTE variation [%]

-10 0 10

stre

ss v

aria

tion

[%]

-1,5-0,75

00,75

1,52,25

3

E variation [%]

-10 0 10

coils mean stresscoils peak stress


• Numerical model with anisotropic cable model already defined

• Layered material

• Local coordinates system at the element level

• Experimental data waited as input for anisotropic modeling

What’s next? (2)

85

• Enrico Felcini and Tuukka Lehtinen as well as the whole Engineering Unit for their constant support and availability

• Luca Bottura and Diego Perini for their trust and supervision in these turbulent times

• Glyn Kirby, Daniel Schoerling, Gijs de Rijk, Juan Carlos Perez, Jacky Mazet, Nicolas Bourcey, Francois Olivier Pincot for their wise advices as well as all the insightful interactions and discussions

• EN/MME, TE/MSC and CERN-KT for their warm welcome

Acknowledgements

Je te l’avais promis!

Additional informationsBack-up slides

87


1 2 30

5

10

15

1 2 30

2

4

0 0.2 0.4 0.6 0.8 10

0.5

1

88

Contact shear

• All solutions mainly lead by the difference in

• Lorentz forces tend to curve/bend the bonded components increase of shear

• Solutions cannot be distinguished according to the behavior in shear

• Peak values seem acceptable for all solutions

CTE

→

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

50


1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

20

40

60

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

0.5

1

1 2 30

50

100

1 2 30

20

40

60

0 0.2 0.4 0.6 0.8 10

0.5

1


1 2 30

0.02

0.04

0.06

1 2 30

0.005

0.01

1 2 30

0.01

0.02

0.03

89

Outer rim gap

• The pole limits the overall structure shrinkage

• Bonding to the pole prevents structure expansion during Lorentz forces application

• Presence of the pole delays the contact with the outer rim

NB: Gap magnitude of the order of few tens of m

• If structure seen as an equivalent cylinder:

m

μ

deq = ΔCTE × Req × ΔT ∼ 75 μ

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

0.02

0.04

0.06

w/o - frictional rimw/ - frictional rimw/ - frictional rim & pole

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

0.005

0.01

1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 30

0.01

0.02

0.03


1 2 3-150

-100

-50

0

1 2 30

200

400

1 2 3

-20

-10

0

90

Contact pressure

• If bonded to the cover and the intermediate plate response dominated by the difference between HTS & SS

• Peak in compression reaches significant magnitude

• Work mainly in tension and above the debonding limit

Inevitable delamination at each grade

→ CTE

→

1 1.5 2 2.5 3-150

-100

-50

0

w/ - fully bondeddebonding limit

1 1.5 2 2.5 30

200

400

1 1.5 2 2.5 3

-20

-10

0


Grade jump layoutsv1.1

v1.2

v2.1

v2.2


home.cern 93

MSC/MME Seminar HTS Demonstrator Mechanics · MSC/MME Seminar HTS Demonstrator Mechanics GaToroid Project Jérôme Harray May 12, 2020

Documents