The Japan Atomic Energy Research Institute has been involved in developing Nb3Al conductors since the middle of 1980s based on the consideration that Nb3Al conductors are capable of producing a higher magnetic field than Nb3Sn conductors owing to the extremely high critical current density in high magnetic fields. At the beginning of the development work, a fabrication technique for Nb3Al strands using a Jelly-roll process was established. This process requires heat treatment at 750°C for 50 h instead of a temperature of more than 1,800°C as required by the conventional method. Using this technique, about 1 ton of strands was produced and a 150-m Nb3Al cable-in-conduit conductor was fabricated. For the next step, to demonstrate the applicability of the Nb3Al conductor to a large coil, a coil 1.5-m in diameter, called the Nb3Al Insert, was manufactured. A react-and-wind method was tried for the production process as it simplifies the fabrication of large coils such as an ITER-TF coil. Performance tests of the Nb3Al Insert were conducted in 2002. The Nb3Al Insert could be charged to the designed point of 13 T and 46 kA without showing any instability. Thus, the world's first large superconducting coil using a Nb3Al conductor was successfully developed, thus indicating the possibility of producing fusion magnets that can operate in higher magnetic fields than those used with Nb3Sn conductors.
Nb3Al superconductor has the potential capability of producing a higher magnetic field than that by Nb3Sn and is less sensitive against strain, which will simplify the fabrication of a large coil. From this point of view, development work for a Nb3Al conductor (cable-in-conduit conductor with stainless steel conduit) was started at the Japan Atomic Energy Research Institute and the Nb3Al Insert was successfully developed and tested in 2002. In the tests, a nominal current of 46 kA was achieved in a 13 T magnetic field and critical currents were measured in terms of magnetic field and temperature in order to clarify the strain state of the cable, which determines the overall performance of the conductor. From the results, the thermal strain of the Nb3Al conductor was estimated to be around −0.4%, corresponding to a 10% reduction in the critical current. This is much smaller than that observed in the same type of Nb3Sn conductor (−0.7% strain and 50% reduction). In addition, there was no effect of the electromagnetic forces on critical current observed in the Nb3Al conductor during coil charge while an evident reduction has been reported for Nb3Sn conductors. The higher rigidity of the Nb3Al strands and smaller sensitivity of the critical current against strain can account for these characteristics. Furthermore, although the insert was fabricated using a react-and-wind method, this did not produce any observable effect on the strain of the cable. These results demonstrated the excellent performance of the Nb3Al conductor to be used for a high-field, large coils that will experience large electromagnetic forces.
The ITER Central Solenoid (CS) model coil, CS Insert and Nb3Al Insert were developed and tested from 2000 to 2002. The AC loss performances of these coils were investigated in various experiments. In addition, the AC losses of the CS and Nb3Al Insert conductors were measured using short CS and Nb3Al Insert conductors before the coil tests. The coupling time constants of these conductors were estimated to be 30 and 120 ms, respectively. On the other hand, the test results of the CS and Nb3Al Inserts show that the coupling currents induced in these conductors had multiple decay time constants. In fact, the existence of the coupling currents with long decay time constants, the order of which was in the thousands of seconds, was directly observed with hall sensors and voltage taps. Moreover, the AC loss test results show that electromagnetic force decreases coupling losses with exponential decay constants. This is because the weak sinter among the strands, which originated during heat treatment, was broken due to the electromagnetic force, and then the contact resistance among strands increased. It was found that this exponential decay constant was the function of a gap (i.e., a mechanical property of the cable) created between the cable and conduit due to electromagnetic force. The gap can be estimated by pressure drop, measured under the electromagnetic force. The pressure drop can easily be measured at an initial trial charge, and then it is possible to estimate the exponential decay constant before normal coil operation. Accordingly, it is possible to predict promptly how many times the trial operations are necessary to decrease the coupling losses to the designed value by measuring the coupling losses and the pressure drop during the initial coil operation trial.
The Japan Atomic Energy Research Institute (JAERI) has demonstrated the superconducting performance of a large coil using a newly developed cable-in-conduit Nb3Al conductor comprised of 1,152 superconducting strands bundled on a central cooling channel. The conductor was developed as one of the R&D activities of the International Thermonuclear Experimental Reactor (ITER). The hydraulic performance of the Nb3Al conductor under electromagnetic force was studied to clarify the relation between pressure drop under electromagnetic force and cable stiffness. Under an electromagnetic force of 500 kN/m, the pressure drop of the Nb3Al conductor decreased by 5%, which is smaller than the 13% decrease measured for a Nb3Sn conductor. This characteristic was well simulated by taking into account the high stiffness of the Nb3Al cable as related to the high strength and stiffness of Nb3Al strand. Furthermore, the pressure drop of several Nb3Sn conductors tested in previous R&D studies was investigated by means of the same evaluation method to study the effect of cable types. It was concluded that the pressure drop under electromagnetic force can be explained by taking the cable stiffness of any type of CIC conductor into account. This paper describes the measurement results, detailed evaluation method, and considerations that led to this conclusion.
The Central Solenoid Model Coil (CS model coil) program was established in 1992 as one of the projects for the Engineering Design Activities (EDA) of the International Thermonuclear Experimental Reactor (ITER). In the year 2002, we carried out several kinds of tests using a CS model coil and Nb3Al insert coil (Nb3Al Insert). In the experiments, we measured the acoustic emission (AE) signals emitted from these coils. In this paper, we focus our discussion on the AE signals emitted from the coils. Two kinds of data acquisition methods for the AE signals were adopted during the series of excitations: whole waveform recording and AE envelope recording. Furthermore, two kind of measuring methods for AE envelope recording were selected: full-time recording and constant interval recording. In this paper, we mainly discuss the AE signals emitted from these coils using envelope recording at a constant interval. Observation of the AE signals showed that disturbances in both coils decreased within a few times of excitation iteration. Even though the Nb3Al Insert fully quenched at the time of inductive heating, we confirmed that the coil returned to an unaltered stable condition after removal of the heat. Furthermore, we confirmed that the average AE voltage kept a constant distribution pattern during the 1,000-time cycle test under a background field of 13 T using the CS model coil.
Superconducting fault current limiters have been researched to protect electric power systems. We have developed a superconducting fault current limiter. As the active element, we choose Bi-2212 thick film for the superconductor cylinder. In order to attain practical use of the superconducting fault current limiter, larger scale superconducting cylinders must be developed. Processing energy increases as the size of the limiter is increased. If there are local defects in the superconducting cylinder and the processing energy is concentrated on them, there is a possibility that the cylinder will be damaged. We therefore investigated the thermal runaway propagation of the Bi-2212 thick film from a local heat point and how local defects affect the current-carrying capacity. The result is that the velocity of the thermal runaway propagation in the thick film from the local heat point is very slow. Thermal runaway doesn't propagate within a few milliseconds. If there are local defects such as cracks in the superconducting cylinder, the current will concentrate at the tip of the defect and the Joule temperature rises. The cracks then grow as the result of thermal stress. When a crack runs from end to end on the cylinder, superconductor on the cylinder melts due to arc discharge.