T. Hamajima

Homogeneous current distribution in a coaxial superconductor with and without return current path

The authors have developed a theoretical method based on magnetic flux conservation between adjacent layers. One of the advantages of this method is that one can directly obtain twisting pitch and radius for realizing homogeneous current distribution in coaxial multi-layer superconductor. A set of the obtained twisting pitch and radius was employed in a sample three-layer conductor comprised of silver-sheathed multi-filamentary BSCCO-2223 tapes and the current distribution was measured by a Rogowski coil. Agreement between the experiment and the theory on current distribution is quite remarkable. Using this theory, the authors analytically investigated the influence of the manufacturing error of twisting pitch and radius on current distribution. The results revealed that the manufacturing errors of twisting pitch and radius have large effect on current distribution and a suitable set of twisting pitch and radius against manufacturing error can be found. They also investigated the relationship between twisting pitch and current distribution in coaxial six-layer conductor with return current path. The characteristics of twisting pitch in the conductor with return current path are different from those of the conductor without return current path.

Influence of current distribution on conductor performance in coaxial multi-layer HTS conductor

We have developed a simulation method based on magnetic flux conservation between two filaments of adjacent layers to estimate current distribution in coaxial multi-layer HTS conductor. Using this method, we have demonstrated homogeneous current distribution and verified that current distribution was controllable directly by the conductor parameters of layer radius, twisting pitch and twisting direction. Although it has been considered that homogeneous current distribution is effective for reducing AC loss, the most suitable conductor parameters and operating condition have not been investigated sufficiently yet. Therefore, we improved our developing method to estimate current distribution more rigorously considering the nonlinear voltage-current characteristic of HTS tape. To verify the validity of the simulation method, we measured current distribution using coaxial two-layer conductors. Agreement of current distribution between the experiment and the analysis was good.

AC loss performance of the 100 kWh SMES model coil

AC loss tests of the SMES model coil for 100 kWh SMES pilot plant were carried out at Lawrence Livermore National Laboratory (LLNL) in 1998, in collaboration between Japan and the USA. The AC loss results at LLNL were in good agreement with those obtained at the Japan Atomic Energy Research Institute (JAERI) in 1996. The coupling loss in the coil could be described by two components with a short time constant (0.22s) and a long time constant (30s). The short time constant was in good agreement with that measured in a short sample. The signals of Hall probes, mounted on the surface of the coil, revealed that the induced loop currents in the conductor decayed with long time constants. At least two long time constants were observed: about 4s and 100s. The long time constant was also identified by the observation of voltage decay after the coil discharge. These loops result in the additional AC loss in the coil. Effect of lateral force in the cable on losses was studied as well. An improved conductor aiming to reduce the AC loss was designed, fabricated and wound in a small coil. The measured AC loss in the small coil made of the improved conductor was about 1/6 of the SMES model coil per strand volume.

A mechanism causing an additional AC loss in a large CICC coil

A large superconducting coil wound with cable-in-conduit (CIC) conductor caused an additional AC loss which cannot be estimated from short conductor sample test results. It was confirmed that the additional AC loss was generated by long current loops in the CIC conductor. Magnetic field decays of the loops with various long time constants were observed through Hall probes. We propose a mechanism forming the long loops. The CIC conductor is composed of several staged sub-cables. If one strand on the surface of a sub-cable contacts with the other strand on the surface of the adjacent sub-cable, the two strands must encounter each other again at LCM (least common multiplier) distance of all staged cable pitches and thereby result in forming a pair of a long loop. We traced each strand in the CIC according to a method that the sub-cables at all sub-stages rotate around a center of inertia. The long time constants were calculated and their results can explain the data measured in a large SMES coil. The proposed mechanism is effective for estimating the additional AC loss in the coil.

Dependence of current carrying capacity and AC loss on current distribution in coaxial multi-layer HTS conductor

We had developed a simulation method of current distribution in coaxial multi-layer HTS conductor and investigated influence of the nonlinear voltage-current characteristic of HTS tape on current distribution. It had been reported that homogeneous current distribution, especially the same layer current, is effective in terms of reducing AC loss. There are, however, many sets of cable parameters to achieve homogeneous current distribution in such the coaxial multi-layer cables. Therefore, using our developed evaluation method, we numerically investigated the relationship between AC loss and the cable parameters such as twisting pitch, radius, and direction in coaxial three- and four-layer conductors. We evaluated both hysteresis loss and flux flow loss as AC loss using the Norriss model and V-I characteristic of HTS tape, respectively. The critical current of whole cable and current density of each tape are key parameters in terms of reducing AC loss. The larger twisting pitch is better for increasing the critical current of cable due to the greater number of usable tapes and the shorter tape-length per unit length of cable in longitudinal direction. Alternate twisting pitch, however, is ineffective for increasing the critical current due to small twisting pitch and small number of tapes for realizing homogeneous current distribution. There is no effect of the degradation of the critical current caused by magnetic field generated by the other layers on AC loss in the cable with the current carrying capacity of the order of at least 1 kA.

Periodic artificial pinning centers for high critical current density of tape conductor

For high magnetic field in superconducting magnet, superconducting tape conductor with effective pinning centers is required for high critical current density. We studied periodic artificial pinning centers for use in HTS tape conductor. Periodic patterns were introduced as artificial pinning centers on a superconducting Nb film. 2 and 4 /spl mu/m period groove patterns were formed on the Nb film with various depths. Magnetization and critical current density of the films with the groove patterns were larger than those of the film without the pattern, as a result of effective artificial pinning. After regarding immanent magnetization caused by defects, films with 2 /spl mu/m period grooves had about 2 times as much magnetization as the films with 4 /spl mu/m period grooves. Ratio of pinning force density of 2 /spl mu/m period groove pattern to that of 4 /spl mu/m period groove pattern was calculated to be 2. The magnetization of both films increased with groove depth until it reached a half of film thickness. It is considered that this increase of magnetization with groove depth is mainly caused by an increase of elementary pinning force for individual flux line.

Irregular AC losses with long time constants in large cable-in-conduit conductors

A large superconducting coil wound with Cable-in-Conduit (CIC) conductor causes both irregular AC losses that cannot be estimated from short conductor sample test results, and regular AC losses that are proportional to cable twisting pitch squared. We proposed a mechanism forming loops that generated the irregular losses. The CIC conductor is composed of several stages of sub-cables. If one strand on the surface of a sub-cable contacts another strand on the surface of the adjacent sub-cable, the two strands must encounter each other again at the LCM (Least Common Multiplier) distance of all staged cable pitches, and thereby a long loop is formed. We orderly labeled all strands in CIC conductors for the SMES and the LHD. It was found that strands in a triplet were widely displaced from their original positions on one cross section, but contacted each other tightly on the other cross section. This fact suggests that the loop with the large displaced strand links irregularly with external field so that the loops cause the irregular AC losses. Moreover, it indicates that a contacting length of the large displaced strands can be quite long, giving rise to a low contact resistance for the loop, and leading to the long time constants. It is believed that the widely displaced strand are inherent in a CIC conductor. It was demonstrated that the strand surface coated with CuNi was effective to suppress the irregular AC losses.

Irregular Loops With Long Time Constants in CIC Conductor

AC losses consist of both regular losses that are proportional to cable twisting pitch squared and irregular losses that could not be estimated from short conductor sample test results. It was explained from our previous works that irregular loops in conductor which are caused by asymmetric strand positions as a result of low void fraction of CIC conductor, produce the losses with long time constants up to several hundred seconds. The observed long time constant indicates that the typical loop length should be about LCM (Least Common Multiplier) of all sub-staged cable pitches, and that contact conditions between the two strands forming the loop should be line contact. In order to investigate the contact conditions in detail, we traced 81(=3times3times3times3) strands every 11 mm of CIC sample conductor with 1 m in length whose strands are NbTi/Cu without any surface coating. The measured traces of 81 strands show that asymmetric strand positions, in other words, large displacements of strands from their original positions due to compressing the conductor provide many line contacts. It is found that the averaged line contact length reaches about 10 mm that is three order of magnitude larger than the 10-2 mm of point contact length

Superconducting properties and rapid heating condition in transformed jelly-roll Nb/sub 3/Al multifilamentary wires as a function of maximum ohmic-heating temperature

Jelly-roll Nb/sub 3/Al wires with a Nb matrix were ohmically-heated to maximum temperatures ranging from 1800 to 2300/spl deg/C in vacuum in order to optimize the ohmic-heating conditions and to investigate the superconducting properties for this transformation method. The diameters of these wires were 0.80-1.34 mm. Surface temperature at the central point of the sample was measured by a photodiode during rapid ohmic-heating. After the surface temperature reached a maximum, the sample was quenched in liquid gallium. All the samples were annealed at 800/spl deg/C for 3-25 hours after the rapid heating process to transform the bcc-phase to the A15 phase. Critical currents were measured up to 23 T. The samples heated to 2000/spl deg/C showed a maximum critical current density of 64 A/mm/sup 2/ at 20 T. The critical current density decreased with increasing maximum temperature during rapid heating. This paper describes the superconducting properties, the rapid heating conditions and the achievement of high critical current density at high magnetic fields.

AC loss reduction of coaxial multi-layer HTS cable

Reducing AC losses, cable size and cost of HTS tapes are required for the practical application of HTS power transmission line. We have theoretically investigated the relationship between AC losses and cable parameters such as layer radius and twisting pitch and direction in coaxial multi-layer HTS cable using our developed simulation method and revealed that the larger number of HTS tapes and twisting pitch are desirable for reduction of AC losses. In this paper we investigated influence of twisting pitch and direction on cable size and total amount of HTS tapes under a condition of constant AC losses. The diameter of each layer and total amount of HTS tapes in the cable with the same twisting direction are much smaller than those of alternate twisting direction. The cable with the same tape current would be better than that of the same layer current because of the smaller difference of the number of tapes per layer. In the cable with the same tape current, cable size decreases with the number of layers in spite of almost the same amount of total HTS tapes in the whole cable. In cable design, we cannot handle the restrictions of AC loss and load factor individually; it would be difficult to satisfy both the conditions of low AC loss and high load factor. AC losses could be reduced by increasing critical current of HTS tape without the restriction of load factor, while AC losses increases with the critical current with the restriction of load factor.