Project code: C5-05

Project acronym: TomoModelSupra

Field: Fundamental research on Energy

Topic: Nuclear fusion and fission


Project title

Characterisation and modelisation  of superconductor cables:  Examination of a selection of Superconductor Cables by X-ray microtomography and full reconstruction of strand trajectories; Application of analytic model of AC losses to in-situ real geometries and subsequent experiments with analyses of screening currents distribution;



RO team

CEA team

Project leader Name


Louis ZANI




Tel :
Fax :

Head of laboratory

Plasma Physics and Nuclear Fusion

National Institute for Lasers, Plasma and Radiation Physics (INFLPR)
Atomistilor Str. 409, P.O. Box. MG-36 
077125, Bucharest-Magurele, ROMANIA


Institut de Recherche sur la Fusion par confinement Magnétique (IRFM)
Commissariat à l’Energie Atomique et aux Energies Alternatives (CEA)
Centre de Cadarache
13108 St Paul Lez Durance FRANCE

Dr. Alain Becoulet


Project proposal

Cable-in-Conduit Conductors (CICCs) are employed in the construction of superconducting fusion grade magnets and tokamak of different configurations. Due to their mechanical strength and ability to withstand the large electromagnetic force applied to the superconductors in large magnets during excitation CICC type superconductors will be applied in the next stage of fusion magnets. Thus, the Poloidal Field Coil in ITER (International Thermonuclear Experimental Reactor), Toroidal Field Coil in JT-60SA and the Wendelstein 7-X magnet system would rely on twisted multifilament NbTi-based composite strands bundled together with copper strands [1-2].

Optimizing the design of a CICC is an extremely complex task coupling of multi-physics issues (electromagnetic, metallurgical, mechanical, thermal and hydraulic) and therefore needing deep knowledge in each area as it should ensure the conditions for a safe, stable and reliable operation of the superconducting magnet. This is a difficult task taking into account the real uncertainties in the strand performance, the fluctuation of the operating temperature, and the impact of AC losses (mostly coupling losses) on cable stability. The common technique to reduce the coupling losses is to twist the wire and the filaments during the manufacture. This also reduces the time-independent proximity effect between the filaments and its associated losses. The whole bundle is jacketed in a stainless steel sheath. The strong and ductile jacket provides protection of the relatively weak, brittle superconducting strands from the global forces generated in the coil.
The tokamak operation undergoes many magnet current possible scenarii where rapid fluctuations of the magnetic field are imposed (e.g. plasma breakdown) therefore rending the issue of AC losses possibly critical. As a matter of fact the screening currents induced by rapid field changes can strongly vary with respect to the cable geometry and architecture and can induce local heating while screening current transfer from one strand to another, leading to loss of superconducting properties into quench propagation. In the present framework a comprehensive modelling is constructed and developed under an analytical approach and sized to be accommodated with the actual geometry. In this regard valuable inputs are needed to be defined to feed the model such as inter-strand contacts statistic. This parameter would be extracted from a high accuracy examination of strand bundles inside the CICC and the inter-bundle resistances would be accordingly mapped. The application of this numerical approach would be backed with ad-hoc experimental tests led at CEA led on JT-60SA-like short samples. Once consolidated, the modelling approach would be generalized to a diversity of other types of CICCs and parametric studies conducted to explore possible optimization of cables design in tokamak operation conditions.

In the last years X-ray micro-tomography emerged as powerful tool for the non-destructive analysis of superconducting materials (bulk, strands and cables). Using 3D X-ray computed micro-tomography (μXCT) we can investigate in a non-destructive manner, various complex structures, for defects classification and analysis, void analysis, porosity, crack or other internal features. The volumetric models generated by μXCT analysis (3D volume rendering) are very useful in visualizing the shape, distribution and evolution of the local macro-density-regions. With adequate software, it is possible to navigate inside the 3D reconstructions and perform sectioning of the volume with arbitrary planes. According to our knowledge, in the field of CCIC analysis the first results have been recently reported [1-3]. It worth to be mentioned that two of these contributions are provided by the team proposing this project [2, 3]. It was shown that parameters such as cable twist pattern (TP) and void fraction (VF) that may have a substantial impact on CCIC performances, can be accurately determined. However, supplementary research and development both in CT setup and data processing are needed in order to take full benefit from μXCT analysis and to allow extensive analysis.


Task 2017 objectives:
Validation of the applicability regarding the X-ray micro-tomography instrument as highly efficient and reliable in obtaining necessary information for 3D model reconstruction of superconductor cables type CICC:   

  • Validation of the optimal X-ray microtomography experimental setup;
  • Establish image processing algorithm that ensures a 100% detection rate for strands positioning.
  • Validation of complete trajectory reconstruction for strand cable;
  • The implementation of an high resolution detector (TDI type) and obtained image reconstructions for quality evaluation;
  • The implementation of high energy X-ray source and preliminary investigations on high dimension CICC (ITER, DEMO);
  • First high spatial resolution examination on Nb3Sn components;

Summary on task 2017

The main specific objectives in this task project are presented by the optimization and validation of the X-ray microtomography experimental setup. This instrument was developed in NILPRP, represents unique equipment in the European scientific community and permits complex investigation of strands and high performance superconductor cables that are involved in the most advanced nuclear facilities such as: ITER, JT-60SA, Wendelstein-7 and DEMO. The acquired high resolution images permits post algorithm processing that delivers 100% strands detection rate obtained on standard and upgraded CICC. In this same task, a high resolution detector type TDI was implemented in order to obtain high quality images. Thus is presented high spatial resolution examination on a new concept of high performance Nb3Sn strand components. The implementation of a high energy X-ray source and new results on CICC of greater dimensions is to be reported in the near future.

Summary on task 2016

Photometric analysis on tomography images was conducted in order to determine strand trajectories based on a bi-variant Gaussian function that fits well the strand 2D section profile. The fitting method proves to be powerful for the detection process, even for the particularly case when the strands are joined.
Even though the analysis algorithm is fast, it needs minor optimizations in order for the detection process to be precise and stable over the all CT acquired images. In the present work, numerous nonlinear fitting methods were applied in the presence of background noise. Also, multiple Gaussian fitting were involved in order to optimize the centroids detection that are based on functional formulations or on a cumulative distribution (example Voight profile).






Back to tomography.inflpr.ro

Site designed and maintained by Cosmin Dobrea