OVERVIEW OF THE LARGE HELICAL DEVICE PROJECT. The Large Helical Device (LHD) has successfully started running plasma confinement experiments after a long construction period of eight years. During the construction and machine commissioning phases, a variety of milestones were attained in fusion engineering which successfully led to the first operation, and the first plasma was ignited on 31 March 1998. Two experimental campaigns are planned in 1998. In the first campaign, the magnetic flux mapping clearly demonstrated a nested structure of magnetic surfaces. The first plasma experiments were conducted with second harmonic 84 and 82.6 GHz ECH at a heating power input of 0.35 MW. The magnetic field was set at 1.5 T in these campaigns so as to accumulate operational experience with the superconducting coils. In the second campaign, auxiliary heating with NBI at 3 MW has been carried out. Averaged electron densities of up to 6 × 10 19 m-3 , central temperatures ranging from 1.4 IAEA-F1-CN-69/OV1/4 2 to 1.5 keV and stored energies of up to 0.22 MJ have been attained despite the fact that the impurity level has not yet been minimized. The obtained scarling of energy confinement time has been found to be consistent with the ISS95 scaling law with some enhancement.

Transient transport experiments are performed in plasmas with and without internal transport barriers (ITB) on LHD and JT-60U. The dependence of χe on the electron temperature, Te, and on the electron temperature gradient, ∇Te, is analysed with an empirical non-linear heat transport model. In plasmas without an ITB, two different types of non-linearity of the electron heat transport are observed from cold/heat pulse propagation: the χe depends on Te and ∇Te in JT-60U, while the ∇Te dependence is weak in LHD. Inside the ITB region, there is none or weak ∇Te dependence both in LHD and JT-60U. Growth of the cold pulse driven by the negative Te dependence of χe is observed inside the ITB region (LHD) and near the boundary of the ITB region (JT-60U).

In the first four years of the LHD experiment, several encouraging results have emerged, the most significant of which is that MHD stability and good transport are compatible in the inward shifted axis configuration. The observed energy confinement at this optimal configuration is consistent with ISS95 scaling with an enhancement factor of 1.5. The confinement enhancement over the smaller heliotron devices is attributed to the high edge temperature. We find that the plasma with an average beta of 3% is stable in this configuration, even though the theoretical stability conditions of Mercier modes and pressure driven low-n modes are violated. In the low density discharges heated by NBI and ECR, internal transport barrier (ITB) and an associated high central temperature (>10 keV) are seen. The radial electric field measured in these discharges is positive (electron root) and expected to play a key role in the formation of the ITB. The positive electric field is also found to suppress the ion thermal diffusivity as predicted by neoclassical transport theory. The width of the externally imposed island is found to decrease when the plasma is collisionless with finite beta and increase when the plasma is collisional. The ICRF heating in LHD is successful and a high energy tail (up to 500 keV) has been detected for minority ion heating, demonstrating good confinement of the high energy particles. The magnetic field line structure unique to the heliotron edge configuration is confirmed by measuring the plasma density and temperature profiles on the divertor plate. A long pulse (2 min) discharge with an ICRF power of 0.4 MW has been demonstrated and the energy confinement characteristics are almost the same as those in short pulse discharges.

have started this year after a successful eight-year construction and test period of the fully superconducting facility. LHD investigates a variety of physics issues on large scale heliotron plasmas ͑Rϭ3.9 m, aϭ0.6 m͒, which stimulates efforts to explore currentless and disruption-free steady plasmas under an optimized configuration. A magnetic field mapping has demonstrated the nested and healthy structure of magnetic surfaces, which indicates the successful completion of the physical design and the effectiveness of engineering quality control during the fabrication. Heating by 3 MW of neutral beam injection ͑NBI͒ has produced plasmas with a fusion triple product of 8ϫ10 18 keV m Ϫ3 s at a magnetic field of 1.5 T. An electron temperature of 1.5 keV and an ion temperature of 1.4 keV have been achieved. The maximum stored energy has reached 0.22 MJ, which corresponds to ͗␤͘ϭ0.7%, with neither unexpected confinement deterioration nor visible magnetohydrodynamics ͑MHD͒ instabilities. Energy confinement times, reaching 0.17 s at the maximum, have shown a trend similar to the present scaling law derived from the existing medium sized helical devices, but enhanced by 50%. The knowledge on transport, MHD, divertor, and long pulse operation, etc., are now rapidly increasing, which implies the successful progress of physics experiments on helical currentless-toroidal plasmas.

The structure of the radial electric field and heat transport at the magnetic island in the Large Helical Device is investigated by measuring the radial profile of poloidal flow with charge exchange spectroscopy. The convective poloidal flow inside the island is observed when the n/m=1/1 external perturbation field becomes large enough to increase the magnetic island width above a critical value (15-20% of minor radius) in LHD. This convective poloidal flow results in a non-flat space potential inside the magnetic island. The sign of the curvature of the space potential depends on the radial electric field at the boundary of the magnetic island. The heat transport inside the magnetic island is studied with a cold pulse propagation technique. The experimental results show the existence of the radial electric field shear at the boundary of the magnetic island and a reduction of heat transport inside the magnetic island