Over a period of 40 years, the ITER project has provided many examples in large magnet structures, from pre-concept to design to manufacture to delivery and assembly, from which lessons can be learned for the future. This presentation is divided into three parts. The first, 'Making and Assembling Large (steel) Cryogenic Structures for Magnets' is based on ITER Coil Structure experience, particularly the Central Solenoid (CS) and Toroidal Field (TF) structures and particularly on large 316LN forgings weighing about 50t individually, requiring then high accuracy machining and deep (up to 0.3m) welds with controlled distortion. New materials development is often proposed. This is easy on a laboratory scale, but often not transferable to an industrial large scale. Examples are provided. The issues to be considered (and solved) are large scale quality (and repair of defects), joining, tolerances and assembly. The second 'Designing Large Steel Support Structures for Magnets' is concerned with structural design codes criteria, and modern analysis, especially of imperfect, but manufacturable structures. Points that are considered are typical critical issues with cyclically loaded thick components, analysis of partial penetration welds and other welds not permitted under normal American Society of Mechanical Engineers (ASME) welding procedures (including fillet welds and/or those volumetrically not inspectable by Non-Destructive Examinations (NDE)) and allowing for tolerance compensation (i.e. designing with overmetal or designing for shimming). The third 'Avoiding or Reducing Large Steel Support Structures for Magnets' shows how it is possible to choose other routes to include structural materials in magnets, for example (as in ITER TF coils) by using plates to contain the conductor, or by using a thick conductor jacket to at least share the loads with massive steel cases.

The ITER Pre-Compression Ring system provides the radial preload required to maintain structural integrity and coil-to-coil contact in the Toroidal Field magnet assembly. Each of the six 5.6 m fiberglass-reinforced composite rings operates at 4.2 K under sustained load for 20 years. Given the viscoelastic nature of fiber composites, two long-term phenomena are of concern: stress relaxation, which could reduce preload, and creep rupture. Earlier studies demonstrated that preload relaxation stabilizes quickly and does not threaten performance. However, initial assessments of creep rupture, based on conventional stress-rupture curves and extrapolated strength degradation, predicted a reduction in safety margins to levels below the design target, though not necessarily resulting in failure. To address this conservatism, a revised testing program focused on residual strength after sustained loading. Small-scale uniaxial tests were conducted by CERN (up to 5000 hours at room temperature), while ENEA carried out multiaxial tests on 1:5-scale PCR mock-ups. All results showed no measurable degradation, confirming that the composite rings can retain adequate mechanical performance for the full 20-year operational lifetime.

Central Solenoid (CS) components have an abundance of technical challenges ranging from material science to component interfacing. In manufacturing, a CS module is subjected to stringent high voltage insulation strength tests at multiple occasions. The engineering challenge is where a multitude of quench detection wires need to exit the bulk insulation and the materials boundary or compatibility become an issue. In assembly, each CS module terminals is jumped to its respective feeder line via an extension lead that is manufactured with interfacing constraints with the module terminal vector as well as the feeder terminal mating surface. Another critical aspect is the clearance available in assembly with other surrounding tokamak components. A summary of the issues found in manufacturing of large scale structural components is also given.

In the International Tokamak Experimental Reactor (ITER), six Poloidal Field (PF) ring coils are employed to initiate, stabilize, and shape the plasma. The Russian Domestic Agency for ITER is responsible for the supply of the PF1 Coil. The manufacturing process, which includes design, tooling procurement, building preparation, and component qualification, spanned over 10 years and concluded in 2022. This article primarily focuses on the final assembly activities following the vacuum pressure impregnation of the coil's winding pack, with specific emphasis on risk mitigation measures to prevent the failure of the coil's electrical insulation and instrumentation. Further, the factory acceptance tests of the completed coil and their outcomes are described and discussed in comparison with the established requirements. These tests include measuring the electrical insulation performance under partial vacuum conditions, checking for leak tightness of the pressurized hydraulic circuit, and conducting dimensional inspections of the coil's interfaces critical for integration into the ITER tokamak. Finally, the article addresses the transportation of the coil and the upcoming cold test to be conducted at the ITER site.

The ITER Pre-Compression Ring (PCR) system consists of two sets of 3 rings each, located on the top and the bottom inner region of the Toroidal Field (TF) coils providing a centripetal load and a radial constraint with beneficial effects on the force distribution in the TF magnet system. PCRs are large and thick rings with a diameter of 5 m manufactured by winding and bonding a flat pultruded tape made of fiberglass in epoxy matrix. Due to the particular composite material and to the innovative process an overall qualification phase has been planned including a dedicated testing campaign on reduced scale ring mock-ups with the aim to characterize their mechanical properties and to verify the compliance to requirements. The mock-ups were procured by F4E with a diameter of 1 m (1/5 of the full scale) and the tests were performed at ENEA Frascati by using a dedicated hydraulic testing facility consisting of 18 radial pulling actuators aiming at reproducing the loads and displacements occurring to the rings. The testing campaign consisted in a detailed sequence of Stress Relaxation (SR) tests performed at different stress levels/durations and followed by final ultimate tensile strength (UTS) tests at room temperature. Acoustic emission (AE) monitoring was also performed during testing of one ring mock-up. This paper describes the testing programme, the ring mock-ups, the test equipment and reports the main results of the testing campaign.

General Atomics (GA) is under contract to UT-Battelle c/o Oak Ridge National Laboratory for the fabrication of ITER Central Solenoid Modules (CSM). GA will provide seven modules to ITER Organization (IO), six of which will be assembled in a stack that forms the ITER Central Solenoid. All CSMs are required to pass factory acceptance testing (FAT) at General Atomics’ Magnet Technology Center (MTC) test facility prior to shipment to IO. Currently, CSMs 1, 2, and 4 have completed FAT and have been delivered to IO. CSM5 has completed FAT and CSM3 will repeat FAT after resuming its test campaign. FATs on CSMs 6 and 7 are planned as the modules complete fabrication. The MTC test facility was designed and built to conduct CSM factory acceptance tests. The test facility consists of the test chamber, high vacuum system, 4.5 K recirculating supercritical helium refrigeration plant, 50 kA DC power supply, capacitor-pulsed DC circuit breakers, 1 GJ discharge resistor, high-temperature superconducting current feeders, data acquisition system, supervisory control and magnet quench protection system, and a multitude of supporting equipment. Where possible, the test facility utilizes similar equipment planned for use at IO for magnet operation. The test facility operational lessons learned as a result of CSM 1-4 FATs are presented and discussed in this paper.

The ITER magnet system contains 50 GJ of energy in 3 main coil sets and feeders. Lesson learned from arc incidents in JT-60SA and CS module 3 cold test show the consequences of insulation failure. There are three major insulation challenges in ITER magnets: bulk insulation in the coil, ground insulation and high voltage wiring. The most issues are found at the vacuum interface with Paschen breakdown in poor vacuum leading to arcs from ground insulation and the risk of high voltage wire insulation cracking. High voltage arcing can potentially be a safety concern in ITER due to arcing to the vacuum vessel. The mitigation focuses in 3 main areas: Prediction: Simulation of arc development within the cryostat to assess potential damage and if necessary implementation of extra protection; Prevention: Minimisation of weak areas, notably high voltage (HV) wire exits from ground insulation. As a result of failure root cause analysis, chemical corrosion between coil resin and wire insulation is a concern being investigated, with if necessary quick "universal" solutions to repair or reinforce predicted weak areas. Vacuum monitoring in the cryostat is being improved to identify potential Paschen conditions in advance; Early Detection: Implementation of regular intermediate global electrical tests under controlled Paschen conditions at low temperature to detect breakdown with high sensitivity of leakage currents (similar to 10 mu A) which may in turn require room temperature water cooled busbars to be easily isolated. At least initially, in-cryostat cameras or other spark detection methods are needed to localise failures and repair methods need to be pre-qualified.

The Pre-Compression Ring (PCR) is an important component of ITER Magnet System. It is designed to generate a radial inward force on every Toroidal Field (TF) coil, putting in compression the shear keys, contrasting the bending of the inner leg and improving the stress distribution on the wedge surface. Stretching the PCR by tensioning 16 bolts on 4 counter flanges generates a total force of 47.8 MN on each coil. The application of such high preload, together with the large number of bolts to be tighten, the limited space available, and obviously the need of preserving the structural integrity of all the components, make this phase of the tokamak assembly far to be trivial. For these reasons, an in-depth analysis of the tightening procedure has been performed by means of FEM analysis, investigating the possible patterns and multi-step sequences to achieve the desired preload. The study led to a deep comprehension of the behavior of the system and permitted to identify and control the main critical aspects, i.e., the increase of stress in the PCR due to the ‘ripple’ and the variation of preload in some portion of the rings when the preload is applied in a different area, both due to the need of operating on a limited number of bolts at each time. The outcome of the investigation is the detailed definition of two multi-step tightening procedures suitable to be adopted for the assembly.

In the ITER machine, two sets of three pre-compression rings will be installed at the top and bottom of the 18 toroidal field coil structures. The rings will tightly hold the coils with a radial force of 4,800 t per coil against the operating forces. Four other pre-compression rings will serve as spares. Weighing 3.4 t each, with an inner diameter of 5 m, the rings made of pultruded S-fiberglass composite are possibly the largest and most highly stressed composite structures to be designed for a cryogenic environment. All 10 rings have been delivered to ITER and tested at the ITER pre-compression ring facility at a double nominal stress level. The facility is used to measure the stress at rupture of sub-size rings and for acceptance tests of the full size rings. These tests are supplemented by the stress relaxation test at nominal load on a ring equipped with acoustic sensors, to confirm that the creep performances are also adequate. The paper discuss the design of the rings, the manufacturing route, and the test results including the analysis of the acoustic emission signals.

The ITER Thermal Shields (TS) consist of actively cooled stainless steel panels. Their role is to minimise the radiation heat load from warm components, such as the Vacuum Vessel (VV) and the cryostat, hence contributing to insulating the magnet system operating at 4.5 K. The panels, cooled by 304L stainless steel pipes stitch welded to 10 mm to 20 mm thick 304LN plates, were coated by a low emissivity silver layer following welding of the pipes and bending operations. Pressurised He gas flows in the pipes at a temperature of 80 K and a pressure of 1.8 MPa at the inlet. These components have been manufactured and delivered to IO and their assembly had been started. Three leaks were detected in Helium leak tests during site acceptance test of the TS. A failure analysis was carried out based on advanced non-destructive and destructive examination techniques, including X-ray microtomography and Focused Ion Beam - Scanning Electron Microscopy (FIB-SEM), confirming crack initiations of different severities, in particular presence of multi-branched “lightning bolt” transgranular cracks through the thickness of the pipes and of corrosion residues featuring high chlorine content. The root cause of the leakages has been univocally associated to Stress Corrosion Cracking (SCC), initiated by halide residues from the silver coating process in combination with the stress induced by the attachment, by welding, of the pipes to the plates. The paper summarises the results of these investigations and the lessons learned, the remedial actions implemented and the planned repair/replacement solutions.

The ITER superconducting magnet electrical insulation is designed and manufactured to be Paschen proof. The ITER cryostat and the feeder vacuums are also designed to provide a level of vacuum where Paschen conditions should not occur in case of insulation failure. The high voltage insulation failure coincident with degraded vacuum has been identified as a risk with major consequences. The level of magnetic energy stored in the superconducting coils, up to a maximum value of about 50 GJ, could result in electrical arcs leading to severe damages on the magnet system. The present paper focus on the operation, control and protection to mitigate the risk of Paschen discharge: vacuum monitoring, detection and protective actions are discussed: what the baseline strategy is and what should be further developed and improved for the ITER magnet system based on the return of experience of large scale superconducting magnet tests.

This paper describes critical assembly process qualifications, acceptance criteria, and worker training of ITER Central Solenoid Assembly. Coaxial joint assembly was qualified and trained to achieve resistance below 4.1 nOhm. The electrical insulation process was qualified to 30 kV and 15 kV under Paschen conditions. Shear pin drilling and installation within specified tolerance was achieved. Modules were lifted and positioned within 2 mm. The lessons learned, and subsequent work execution onsite within the IO facility assembly hall are also presented.

In the ITER magnetic system 18 Toroidal Field coils (TF coils, or TFC) must be assembled so that the Current Center Line (CCL) of the TF magnets is correctly positioned within its tight tolerances. There are two major processes that cause large deformation of the TF coils: the load transfer toward the gravity support, occurring when the temporary inboard vertical support is removed, and the centripetal displacement due to the Pre-Compression Ring (PCR) tensioning. For those processes, specific procedures and tools are being developed, investigating at the same time the proper method to verify the procedures. Monitoring deformations and forces during the application of huge loads is the key for guaranteeing CCL quality, construction schedule, and health & safety matter. As results of the investigation, a part of the components and tools for TF PCR pre-tensioning has been changed from baseline plan: manual tightening multi-jackbolt tensioners (MJT) have been replaced with hydraulic tools and dedicated nuts. Current procedures including the tools design, the FE simulations, and the monitoring system plans are presented in this paper.

The early plasma ramp-up on ITER will be performed under a limiter configuration. In this early plasma phase, the possible misalignments between the plasma configuration and the ITER first wall panels (FWP) in the inboard regions can lead to an increase of the power flux density acting on the FWPs. This situation is considered as a safety risk for the machine: a significant increase of the flux density may exceed the design limits of the FWPs and damage the inboard modules. Therefore, in order to avoid an excessive increase of the power flux density on the FWPs, one of the main goals of the tokamak assembly strategy is to achieve the most symmetric magnetic field structure and the best blanket alignment. The symmetric toroidal field (TF) coils structure is key for achieving the symmetric magnetic field. In the ideal case, the assembly gap dimensions in the wedging interfaces of every two adjacent TF coils are designed as an uniform value, i.e., 2 mm. However, in reality the assembly gap tolerances can be generated during the manufacturing and assembly process of the TF coils, which can lead to the asymmetric TF coils structure configuration after energization. The purpose of this paper is to investigate and discuss the consequent displacement distortions of the TF coils by simulating the non-constant assembly gap distributions in a 360° mechanical FE model of the ITER TF system.

Fusion for Energy (F4E), the European Domestic Agency for the International Thermonuclear Experimental Reactor (ITER) is responsible for the manufacturing, test and delivery of 5 out of 6 Poloidal Field Coils of ITER. F4E has currently delivered 4 coils: PF2, PF4, PF5 and PF6, the last two are installed in the Tokamak pit in temporary positions while PF2 and PF4 are stored awaiting the Tokamak torus completion. At the PF coils manufacturing production site in Cadarache, France, few meters away from the ITER Tokamak assembly building, F4E is currently finalizing the manufacturing and testing of the largest coil: PF3, with 24 m diameter and a weight of 384 t , approaching to the end of an endeavor that started back in 2013. This article describes all electrical tests the PF coils underwent to meet the ITER technical requirements during their lengthy manufacturing process, and which lasted several years. Tests performed at key points of the manufacturing process avoided discovering non-conformities at later stages that could otherwise become critical for the project accomplition. The main electrical tests consisted of high voltage DC/AC, partial discharge tests, and local and global Paschen tests. The experimental setups and procedures are presented and discussed in more detail. We also summarize the acceptance test results for the finalized coils i.e., 80 K forced flow pressure drop testing, leak tightness in vacuum before, during and after the thermal cycle to 80 K .

The International Thermonuclear Experimental Reactor (ITER) magnet system is one of the most sophisticated superconducting magnet systems ever designed, with a stored energy of 51 GJ. The coils are wound from cable-in-conduit conductors made of superconducting and copper strands assembled into a multistage rope-type cable, inserted into a conduit of austenitic steel tubes. The ITER central solenoid (CS) works in pulsed mode, reaching a peak field of 13 T, thus allowing the induction of a high intensity current in the plasma of the ITER tokamak. This magnet consists of a stack of six modules which include around 125 t of Nb 3 Sn strands. The production of all CS conductors has been completed and module manufacturing is well underway; throughout the production phase, samples were cut at the extremities of the conductor unit lengths to undergo quality control tests. About 25% of the conductor short samples were tested in current and field at cryogenic conditions at the SULTAN facility in Villigen, Switzerland. This work reports the comparative analysis of the short samples set of test results.

Now that ITER magnets are nearing procurement completion, with many of them on site and being installed, there is growing interest in the next step to a tokamak fusion reactor. A big question in this next step will be the safety and reliability of the magnet system. Even before operation we have many lessons to be applied in the next step. ITER magnets are not largely safety-important components and have been produced without the appointment of an Authorised Notified Body(ANB), by the Safety Regulator. In future fusion reactors, reliability and reparability will have a higher importance than pure functionality, and construction cannot be approached with an R&D mentality. Methods and techniques for implementing quality assurance in ITER superconducting magnets started with design control, then extended to manufacturing quality control in procurement including control verification and methods for acceptance tests. Use of on-line tools such as a manufacturing database to provide the quality traceability during procurement control and non-conformities database applied to ensure conformity to fulfil the requirements over the ITER international suppliers. As a staged project, currently ITER is facing the installation process control including component identification and tracing, tool proofing, qualification of installation, inspection and test. Next will be the control of commissioning activities and verification functions until the hand over of superconducting magnets for operation. There has been remarkable progress from the ITER programme on developing and applying quality assurance, with many lessons learned. However, the examples given by incidents such as occurred in LHC, JT60-SA and ITER CS module test would not be acceptable for a nuclear reactor. We need now to review our approach to give a firm footing for DEMO QA/QC.

AC loss in the Nb3Sn cable-in-conduit conductors (CICC) usually decreases after the conductor undergoes electromagnetic cycling. This can be attributed to the increase of inter-strand resistance due to the detachment of the strand-bonding during cyclic loading. In the years 2018 to 2020, ITER launched a series of test campaigns in SULTAN test facility, whose aim was to study the Tcs degradation of TF conductors of all manufactures. Along the main scope of the project, an accompanying study of the AC loss evolution has been measured on two samples. The AC loss was measured prior to any electromagnetic loading, after 1, 5, 50 and 1000 cycles, and at the very end also after a thermal cycle to room temperature. Sinusoidal AC loss was measured in the frequency range of 0.1 to 1.0 Hz. The measured AC evolution will help to predict heat load generated in the TF coils during the initial phase of ITER operation. It may also serve as an input to analysts for deducing the evolution of inter-strand resistance during electromagnetic cycling. The main observation is that already after the first electromagnetic cycle the AC loss reduction is significant, namely 80-90% lower compared to the initial conductor state.