Eu CARD HTS insert design construction and standalone

Eu. CARD HTS insert design, construction, and stand-alone test results CERN 01/10/2018 Maria Durante CEA Paris-Saclay

• the test of the insert alone • • 01/10/2018 • The objective of the project was to demonstrate the possibility of using a REBCO type HTS ceramic tape to generate 6 -T field in the 13 -T field of the Nb 3 Sn FRESCA 2 dipole, for a total field of 19 T. • It was the first step towards the use of HTS for accelerator magnets, the second one being the Eu. CARD 2 dipole under construction. • It started in the framework of Eu. CARD European program High Field Magnet, WP 7. 4, and continues within the first CERN/CEA Collaboration agreement for the development of superconducting magnets for future programs beyond the LHC, No. KE 2275/TE. • 2 steps foreseen: Eu. CARD HTS Insert Goal of the Eu. CARD Insert realized during Summer 2017 the test of the insert into FRESCA 2 dipole magnet • to be discussed this afternoon 2

Eu. CARD HTS Insert 01/10/2018 REBCO Tape YBa. Cu. O Tape (Super Power) • Special tape produced by Super. Power for the Eu. CARD Insert : 200 µm HTS 1 µm 3 Double YBCO layer tape SP BCSC 12050 -AP 2 G (Super Power)

• Je= 250 A/mm 2 Layer 1 Layer 2 Layer 3 • FRESCA 2 bore 100 mm insert external diameter 99 mm • 3 double layers coils Eu. CARD HTS Insert 01/10/2018 Insert Design No more welded 3 External pads, 316 L 15 100 -mm long parts 2 50 -mm long parts 2 1 External tube, 5. 5 mm thick Nitronic 40 - 8 parts, 100 mm long - Thermal shrinkage - TIG welding Glass epoxy inter-coil and external insulation Iron pole Compression plates, 316 L 4

Hastelloy+supra (65 µm) CONDUCTOR Cu. Be 2 (100µm) + polyester insulation (30 µm) 130 µm CABLE (made of 2 conductors) 920 µm copper (70 µm) Eu. CARD HTS Insert SC TAPE SP BCSC 12050 -AP 2 G HTS 01/10/2018 Insert Conductor 12 mm 200 µm 460 µm • 6 co-woud tapes : 2 sc, 4 Cu. Be 2 • SC tapes from Super. Power • Cu. Be 2 tapes from Beryl. Co, annealed at Saclay, insulated by High Precision Foil in UK. 5

01/10/2018 • To prevent the generation of imbalanced currents in the central coil caused by imbalanced inductances of the two co-wound conductors, the conductors are transposed one time at the center of the magnet (pole to pole) Eu. CARD HTS Insert Tape transposition Cable-1 Cable-2 6 Cable-1 Cable-2

Fy Bext Couche 3 12 mm 0 T background field 13 T background field Fx (k. N/m) σx* (MPa) Layer 3 Layer 2 Layer 1 47 77 91 4 7 8 447 1024 1290 37 85 108 01/10/2018 Fx Eu. CARD HTS Insert Design - Mechanical structure *σx = Fx/cable_width, with cable width = 12 mm. Couche 2 Fy Layer 3 Fy Layer 2 Fy Layer 1 Couche 1 Unit k. N/m 0 T -55 -93 -38 13 T -55 -93 -38 Fy is the same in 0 T and 13 T background field because the background field is oriented vertically (along y) Fx (k. N/m) Fy (k. N/m) Total (Bext = 0 T) 215 -186 Total (Bext = 13 T) 2761 -186 Mechanical structure must support 276 ton/m! 7

• Evolution Of Magnet Design (Nominal Current of 2800 A) Layer 1 Layer 2 Layer 3 First Design Optimized Design (Mecanics) Built Magnet Unit Outer tube thickness 3 5. 5 mm # of turns central coil layer 1 36 33 30 turns # of turns external coils layer 2 29 25 24 turns # of turns external coils layer 3 14 11 10 turns Computed magnetic field, @In* 6. 1 5. 6 5. 4 T Engineering current density 250 235 A/mm 2 PARAMETER *Not taking into account screening current induced field (SCIF) Eu. CARD HTS Insert 01/10/2018 Insert Design Evolution 8

Persistant Inter-filament max By= +25 m. T 5 -25 seconds • • • max Bx = + 20 m. T max By = + 36 m. T time = long Persistant due to perp field • • max By= - 700 m. T time = long Eu. CARD HTS Insert • • Imbalanced 01/10/2018 • Impact of persistent currents (B⊥ and B//) and of imbalanced currents on the central field have been computed. Eu. CARD HTS Insert Design – Induced field • The self-field at nominal current is expected to be around 4. 7 T instead of 5. 4 T 9

Eu. CARD HTS Insert Background field T 0 13 Nominal current A 2800 Central field T 5. 4 T* Temperature K 4. 2 Stocked energy k. J 12. 5 Inductance m. H 3. 2 Temperature margin K 29 12 Load line margin % 47 32 *The self-field is expected to be around 4. 7 T due to SCIF 01/10/2018 Insert Final Design Layer 1 Layer 2 Layer 3 10

[m] Eu. CARD HTS Insert [Pa] 01/10/2018 Insert Design - Mechanical structure Maximum stress 726 MPa [Pa] Radial deformation due to magnetic forces < 0. 5 mm [m] 11 The displacement is magnified 60 times.

01/10/2018 • Phase I, stand-alone mode : demountable mechanical structure (assembly less risky and faster, possibility to adjust the prestress, possibility to quickly and easily open the magnet) • Phase 2, insert mode in Fresca 2 : demountable structure replaced by the 5. 5 mm thick tube (heat-shrink method) Eu. CARD HTS Insert Design – Mechanical Structure 130 mm ø 99 mm 130 mm 12 Phase 1 Phase 2

01/10/2018 • Phase I, stand-alone mode : demountable mechanical structure (assembly less risky and faster, possibility to adjust the prestress, possibility to quickly and easily open the magnet) Eu. CARD HTS Insert Design 800 mm 13

01/10/2018 Eu. CARD HTS Insert • Iseult vertical-axis winding machine • In preparation of the winding, each tape is split onto two reels, respectively with the tape for the upper layer and for the lower one. • The six upper layer reels were fixed on a support mounted on top of the coil pole, while the six lower layer reels were placed on six independent magnetic breaks in line with the winding table Eu. CARD HTS Insert Coil fabrication – Conductor preparation 14

• No hard way bending on the tape in layer jump • Special wedges featuring a small angle 3 D printed at CERN in Accura® Bluestone glued on iron pole ends before and after the first turn of each layer 01/10/2018 Eu. CARD HTS Insert • Top and bottom coil of the magnet are classical double layer racetrack coils. Eu. CARD HTS Insert Coil fabrication – Top/Bottom coils 15

• First Coil 2 -3 A : Delamination of one of the four SC layers observed 01/10/2018 Eu. CARD HTS Insert • 3 coils fabricated Eu. CARD HTS Insert Coil fabrication – Top/Bottom coils 16

• First Coil 2 -3 A : Delamination of one of the four SC layers observed • Use of margin on central coil conductor length to build a third external coil • Coil 2 -3 B and 2 -3 C OK 01/10/2018 Eu. CARD HTS Insert • 3 coils fabricated Eu. CARD HTS Insert Coil fabrication – Top/Bottom coils 17

• The two insulated conductors are transposed in the layer jump area 01/10/2018 • Fabrication process almost identical to that of the top and bottom coils. Eu. CARD HTS Insert Coil fabrication – Central coil Eu. CARD HTS Insert • The relatively long pole length (0. 7 m) makes it unnecessary to use wedges in the ends for the layer jump. 18

01/10/2018 A special tooling was designed to • put the three coils together, along with all other magnet components, • turn the assembly 90° to put in place the pads • always keeping the coils under compression to prevent them from unwinding. Eu. CARD HTS Insert Magnet assembly 19

Eu. CARD HTS Insert Magnet supporting rods 01/10/2018 Demountable intercoil connexions Current lead clamps Current measurement device Quench detector 20

01/10/2018 • GN 2 & LN 2 77 K • GHe 4 -10 K • LHe 4 K Eu. CARD HTS Insert Test facility in Saclay – ex Séjos 21

01/10/2018 • The Magnet Safety System (MSS) was based on the same boards and crate used in the JT 60 -SA Cold Test Facility. • Detection is based on the comparison of drop voltage with a threshold voltage during a holding time Eu. CARD HTS Insert Magnet protection (DIS) • differential quench detection on the coils (between top and bottom coils and between external and central coils), • single ended (intercoil joints, superconducting lines, CL). Current Lead connexions Intercoil splices 22 Top coil Central coil Bottom coil

• 3000 A, 3 V • Ripple, current and tension oscillations, MSS triggering à 100, 200 A • Resistive behavior observed on the central coil 01/10/2018 • Hazemeyer power supply, Tests @ 77 K Eu. CARD HTS Insert Power supplies • Sorensen power supply, 77 K and 4 K • 400 A, 10 V • Powering up to 320 A (75% LL @ 77 K) without problems • TDK Lamda power supplies, 4 K • 2 x 1000 A, 10 V + 3 x 600 A, 8 V connected in parallel • Powering up to I nom = 2800 A then up to 3200 A without quench 23

Eu. CARD HTS Insert 01/10/2018 Powering tests 24

• Current plateaus 30 min @ 200 A Measurements and simulation at 77 K The markers represent the measurements, the lines represent the simulations • Good accordance between simulation and measurements • Decay of the magnetization with logarithmic rate in time • Initial conditions before the plateau at 200 A have a strong effect on the magnetic field and its relaxation. Same phenomenon than in the ’overshoot’ technique used to remove the effects of the screening currents Ph. Fazilleau et al. , IEEE Transactions on Applied Superconductivity ( Volume: 28 , Issue: 4 , June 2018 ) Eu. CARD HTS Insert 01/10/2018 Screening current studies - 77 K 25

• Current plateaus 30 min @ 200 A, 400 A and 1000 A Measurements and simulation at 4 K Eu. CARD HTS Insert 01/10/2018 Screening current studies - 4 K The markers represent the measurements, the lines represent the simulations • The accordance between simulations and experiments is not as good as at 77 K. The impact of intercoil resistance, not taken into account in simulations, is more important at 4 K than at 77 K. • Some measurements (and simulation) shows a decrease of the total magnetic field. This can be explained by a redistribution of current between the two non transposed SC layers in each conductor. Ph. Fazilleau et al. , IEEE Transactions on Applied Superconductivity ( Volume: 28 , Issue: 4 , June 2018 ) 26

• Standalone tests results analysis • Current redistribution, ramp rate impact… • Phase 2 – Insert mode (to be discussed this afternoon) 01/10/2018 Eu. CARD HTS Insert Next steps – Phase 2 • Preparation for tests in FRESCA 2 dipole • Insert structure Assembly mockup • Insert Structure Assembly • Intercoil connections to fit inside FRESCA 2 • Tests in FRESCA 2 before end of 2019 27

Thank you for your attention