The design and fabrication of reliable, low-loss joints between high current superconductors remains one of the most difficult tasks in fusion magnet system design. Understanding of design options is urgent because of the International Thermonuclear Experimental Reactor (ITER) and the Korean Superconducting Tokamak Advanced Research (KSTAR) programs. The Pulsed Test Facility at MIT was constructed to test half a dozen joint design options for ITER, before the completion of the ITER Engineering Design Activities (EDA) period. The KSTAR project must make decisions in the near future concerning the design and test of joints for its all-superconducting coil system.

This study compares analytical calculations and experimental results for two different joint topologies. The first is that of the US ITER Preprototype joint and the second that of the KSTAR conceptual design joint, based on that of the cancelled Tokamak Physics Experiment (TPX) magnet system design. This joint is similar to those used in the National High Field Magnet Laboratory (NHFML) 45 T Hybrid Outsert magnets.

The US ITER Preprototype joint has the unique advantage for loss calibration that all five of the independent loss mechanisms (DC, parallel pulsed field, transverse-transverse pulsed field, transverse-parallel pulsed field, and dI/dt) have been tested. Experimental results are compared with numerical predictions, using the commercial electrodynamics code ANSYS. The numerical results are also compared with predictions using analytical expressions. To the best of our knowledge, this is the first time that experimental results have been compared with numerical predictions for all possible orientations. The latter two comparisions are also done for the KSTAR joint topology. Scalings compare the two topologies for actual pulsed and steady-state requirements for tokamak operation.

* Research supported by the Office of Fusion Energy, U.S. Department of Energy under contract DEFC-0293-ER-54186