In this work, a 3D FEM electro-thermal model of a slotted core HTS cable, developed at the ENEA Frascati Superconductivity Laboratory (Italy) in collaboration with the University of Bologna (Italy) and Eni S.p.A. (Italy), is presented. A benchmark of the model with experimental tests is provided, showing the temporal evolution of voltages and temperatures (monitored through fibre optic sensors) during quench. The model is also applied to compute the normal zone propagation velocity. The analysis finally provides critical information for the design of a quench protection system based on fiber optics.
Distributed Fiber Optic Sensors (DFOS) are being considered as an alternative solution for Quench Detection in HTS fusion magnets. Their successful application is not straightforward due to the simultaneous need for high spatial and temporal resolution over long-length conductors operating in demanding cryogenic environments. In this work, we explore the potential of two different commercially available distributed sensing technologies, respectively based on the analysis of Rayleigh-backscattering through Optical Time Domain Reflectometry (OTDR) and Optical Frequency Domain Reflectometry (OFDR). The two techniques were used first to characterize the cryogenic temperature response of similar fibers between room temperature and 4 K. The temperature sensitivities were computed and compared to those of Fiber Bragg Grating sensors, showing comparable results. The performance of the investigated techniques in detecting local quenches was then experimentally assessed in operating conditions on a 2-meter-long, aluminum-stabilized HTS sample cooled by Liquid Nitrogen flow. While OFDR needs improvements in temporal resolution, the limited scale of the hot spot represents a challenge for the OTDR technique due to its lower spatial resolution.
The primary Quench Detection System (QDS) of the Central Solenoid (CS) of the ITER magnet system must be able to distinguish the resistive component of the voltage, arising during quench, from the inductive one due to time-varying magnetic fluxes. This task is especially challenging for the six modules composing the CS (CSMs), with a coil self-inductance of the order of ∼ 1 H, subjected to current ramps up to ∼10 kA/s. Each CSM is equipped with stainless-steel tapes wrapped around the outer conduit of the conductor, referred to as Co-Wound Tapes (CWTs), linked to a magnetic flux approximating the one linked to the conductor itself. This work focuses on the definition of a model able to compute the voltages measured by the QDS with the high precision required. The analysis is based on a 2D axisymmetric FEM model of the entire CSM in a stand-alone configuration. A novel 3D approach based on an integral computation method allows one to account for the twisting of the conductor sub-cables of the last cabling stage (petals). The model is applied to quantify the impact of three main error sources contributing to the residual voltage signal measured by the QDS: the difference in self-field flux linked to the conductor and to the CWT, the inhomogeneity of the background field and the twisting of the petals.
In this paper, we report the results from two test campaigns conducted to evaluate the feasibility of implementing a commercially available optical fiber-based distributed sensing technology for quench detection purposes. We successfully characterized the temperature response and sensitivity achievable by this technique when using a bare single mode fiber within a cryogen free cooling system operating in the temperature range from 4 K to 300 K. A fiber was also incorporated into a sample featuring an HTS stack mounted onto an extruded Aluminum slotted-core cable. Quench-like events were triggered by locally lowering the critical current using the magnetic field produced by a set of permanent magnets placed near the center of the stack. The experiments were carried out at 77 K by cooling the sample through the central channel of the Aluminum core with a forced flow of liquid nitrogen. The test results indicate the viability of this technology as a suitable alternative for Quench Detection in long-length HTS cables and motivate further work to develop distributed sensing systems able to work in the conditions foreseen for HTS fusion coils.
In this work, we show that high spatial resolution (3-cm) distributed temperature measurements at cryogenic temperatures can be realized using the Brillouin Optical Frequency-Domain Analysis (BOFDA). A large increase of peak gain and narrowing of the Brillouin gain spectrum were observed at temperatures below approximate to 10K under high spatial resolution conditions, thanks to the pre-activation of the acoustic wave involved in the stimulated scattering process. Adopting a slope-assisted configuration, a temperature sensitivity less than 0.1 K is demonstrated over a maximum fiber length of 50 m and at an acquisition rate of 1.75 Hz.
Powerful electromagnetic transients characterize the operation of the ITER superconducting magnet system. This is especially the case for so-called plasma disruptions, but also during plasma current initiation and termination and during the fast discharge of the magnets for quench protection. To model these electromagnetic transients the CARIDDI code (Albanese and Rubinacci, 1988) has been used to calculate the induced voltages in several magnet quench detection loops during operational events of interest. The numerical model implemented in CARIDDI, based on an integral formulation discretized in terms of edge elements, is shown to be particularly well suited for analysing these transients which require a high level of precision. Note that, although we will mostly present results obtained with CARIDDI, all calculations were cross-checked with other codes and methods, most notably (ANSYS Maxwell, 2022).