Advances on the high speed ignitor Pellet Injector (IPI)

The control of the density profile during the initial plasma current rise is a critical issue to optimize ohmic and fusion heating rates of Ignitor plasmas. Simulations performed with the NGS ablation model, for the reference ignition plasma parameters (n
_e0
≅ n
_i0
≅ 10
²¹
m
^-3
, T
_e0
≅ T
_i0
≅ 11 keV), indicate that deuterium pellets of a few mm (≤ 4 mm) in size injected at 3-4 km/s from the low field side should ensure adequate deep fuelling. ENEA and ORNL are collaborating on the development of a four barrel, two-stage pneumatic injector for the Ignitor experiment, featuring two innovative concepts: (i) the proper shaping of the propellant pressure pulse to improve pellet acceleration, and (ii) the use of fast closing (~9 ms) valves to eliminate the need of large expansion volumes for propellant gas removal. Two independent sub-systems have been built. The ENEA equipment, including four independent two-stage guns (TSG) and pulse shaping valves, the gas removal system, and the associated controls and diagnostics, has been built and thoroughly tested at CRIOTEC, prior to being shipped to ORNL. The ORNL apparatus consists of the cryostat and pellet diagnostics, with related control and data acquisition system. Integration of the two subsystems (except for the gas removal system) has been readily achieved. The present arrangement accommodates both a TSG and a standard propellant valve on each barrel, allowing seamless switching between standard and high-speed operation on any or all gun barrels. Previous joint experiments at ORNL, demonstrated that the two systems match properly, while their respective control systems interface correctly. The injector performed outstandingly, showing excellent repeatability. These preliminary results indicate that, for the same peak pressure, the IPI has the potential of achieving higher speed performance, as compared with those attained by previous high speed injectors, such as the Single Pellet Injector (SPIN) installed on the FTU in the early 90's. Launching sequences at moderate propellant pressure and speeds up to ~2.6 km/s were performed with all four barrels; however, it was not possible to observe intact pellets at speeds above 2 km/s. The analysis of experimental data indicate that, at high speeds, the pellets may spin and hit the wall of the too narrow conduit crossing the diagnostics. This hypothesis is corroborated by numerical simulations. Following this analysis, the inner diameter of this guide tube has been enlarged, and a new target diagnostic has been implemented to test the dispersion of pellet trajectories immediately downstream of the diagnostics. The results of the latest experimental campaign are reported.