A manhole of the tokamak vessel was closed on September 1. Leak tests of the vessel were completed in September, both before and after the vessel baking at 300C. Tests of power supply systems including motor generators and conditioning of NB and RF systems were carried out in October. The temperature of the tokamak vessel was kept 300C during the last two weeks in October for wall conditioning before the start of tokamak operation in November.
The following were done, in addition to usual maintenance, during the maintenance period from December 2004 to October 2005; (1) repair of the motor generator for heating systems, (2) installing ferritic steel to reduce field ripple, (3) restoration of an LH antenna, (4) modification of armor tiles in NB ports, (5) installing a supersonic molecular beam injection system, (6) modification of the pellet injector system, and (7) improvement of diagnostics systems.
(1) Repair of the motor generator for heating systems: The motor generator for heating systems (H-MG), which had been damaged in an accident in February 2004, has been repaired. In the operation from July to November 2004, the heating systems (NB and RF) were connected to the motor generator for toroidal field coils (T-MG) instead of H-MG and the power for the toroidal field coil was supplied directly from a commercial line solely. As a result, the strength of the toroidal field was limited to a few values below ~3 T. In the forthcoming operation, normal operation of the toroidal field up to 4 T will be possible.
(2) Installing ferritic steel to reduce field ripple: Ferritic steel tiles have been installed into the JT-60U vacuum vessel to reduce toroidal field ripple and the resultant loss of beam ions. The material is ferritic steel with 8% Cr, 2% W and 0.2% V. The saturated magnetization is 1.8 T at the room temperature and 1.7 T at 300C. The ferritic steel tiles of typical thickness of 23 mm have been mounted in place of graphite tiles, as plasma-facing components, on the upper low-field side beneath the toroidal field coils. The number of ferritic steel tiles is 1122 and they cover ~10% of area inside the vacuum vessel. The absorption power of NB is expected to increase by a factor of 1.3 for large volume plasmas with the toroidal field of ~2 T.
(3) Restoration of an LH antenna: Damage was observed in an LH antenna with a carbon grill after the last campaign (2003 to 2004). The damage, melting in the stainless base, was caused by discharge at the connection between the stainless base and the carbon grill. The carbon grill has been removed and the stainless base was restored for 6 rows out of the total 8 rows of the antenna. In the forthcoming operation, only 6 rows will be used, which will result in 25% reduction of available power and slight broadening of the power spectrum.
(4) Modification of armor tiles in NB ports: Damage was observed in bolts fixing the armor tiles in port ducts of tangential NBs, the beam duration of which was extended to 30 s, after the last campaign. This was caused by heat load in the long pulse operation due to reionized particles hitting the duct wall. The armor tiles have been modified so that the bolts are completely covered by carbon tiles. Additional thermocouples have been placed to monitor the temperature increase near the damaged bolts.
(5) Installing a supersonic molecular beam injection system: Preparation of a supersonic molecular beam injection system, which is able to launch a series of pulsed high-pressure gas jets, has been started in collaboration with Tore Supra, Association Euratom-CEA, CEA Cadarache. The injector heads and the compressor, major parts of the injection system, have been installed this summer. The injector heads are located both on the high-field side and the low-field side of the torus. The injection system will be completed and be used for efficient particle fuelling in 2006.
(6) Modification of the pellet injector system: The pellet injector system in JT-60U is a centrifugal acceleration type that injects up to 40 pellets at frequencies of 1 to 10 Hz. The pellet size is 2.1 mm cubic and the injection velocity is 0.1 to 1.0 km/s. Since the number of pellets was insufficient for long pulse experiments with 30 s NB heating, a new screw-type pellet generator has been prepared. The pellet system is under modification outside the torus hall, and it will be installed to JT-60U in 2006 after the test operation is completed.
(7) Improvement of diagnostics systems
(7.1) Edge fast TV camera: A fast TV camera has been prepared for measurement of plasma emission fluctuations near the plasma edge. The sampling speed is up to 120 kHz with 128x16 pixels.
(7.2) IRTV: an IRTV observing the divertor baffle plates and the equatorial regions of the first wall has been installed. This will be used to estimate the fast ion loss onto these regions and will be an important tool in the study of fast ion confinement under the reduced toroidal field ripple with ferritic steel.
(7.3) Repair of the reciprocating Mach probe on the high-field side: Three reciprocating probes were equipped for SOL plasma study in JT-60U; a low-field-side horizontal one near the equatorial plane, a low-field-side horizontal one near the X-point, and a high-field-side vertical one near the baffle. The last one had been unavailable during the last campaign due to damage of the probe head caused in 2002. It has been successfully repaired and will be used together with the other two reciprocating probes.
(7.4) Fast sampling of magnetic probe signals: Data acquisition system for magnetic probes, Mirnov coils (~30) and saddle coils (8), have been upgraded so that 1 μs sampling for 50 s (2 μs for 65 s) is available.
(7.5) Wide-spectral-band spectrometer for the divertor: A new visible spectrometer has been installed to observe two-dimensional distribution of emission from the divertor plasma using existent optical systems with 60 vertical viewing chords and 32 horizontal viewing chords. The spectrometer covers a spectral range of 350-800 nm simultaneously.