Êtes-vous un étudiant de l'EPFL à la recherche d'un projet de semestre?
Travaillez avec nous sur des projets en science des données et en visualisation, et déployez votre projet sous forme d'application sur Graph Search.
The low aspect ratio of the mega amp spherical tokamak (MAST) allows differentiation between different forms of the H-mode threshold scaling. With optimized fuelling using inboard puffing, and a connected double null divertor (DND) magnetic configuration, the H-mode power threshold data lie about 1.7 times higher than recent scaling laws. Slight magnetic configuration changes, of the order of the ion Larmor radius, around a connected DND significantly influence H-mode access. H-mode confinement in discharges with low frequency edge localized modes (ELMs) is generally consistent with international scaling laws, e.g. IPB98(y,2). Strong indications of both particle and energy internal transport barriers have been seen. Normalized beta values beta(N) > 5 have been obtained, approaching the ideal n = I no wall external kink stability limit. Sawtooth triggered neo-classical tearing modes have been observed; numerical modelling of the island evolution reproduces mode behaviour well and confirms the significance of stabilizing field curvature effects. Divertor power loading studies, including transient effects due to ELMs, show a strong bias of power efflux to the outboard targets, where it is more easily handled. ELM energy losses, DeltaW(ELM), are less than 4% of the stored energy in all regimes explored so far, but ELM effluxes extending 30 cm outside the outboard separatrix have been measured. Toroidally asymmetric divertor biasing resulted in significant broadening of the D-alpha profile on the biased components and a reduction in the total power to the unbiased components. Halo current magnitudes and asymmetries are generally small compared with conventional tokamaks; recent measurements show that the plasma behaves more as a voltage source than a current source. Initial neutral beam current drive experiments indicate non-inductively driven current values (I-NBI similar to 0.3I(p)) comparable with code predictions.
Basil Duval, Stefano Coda, Joan Decker, Umar Sheikh, Claudia Colandrea, Luke Simons, Jean Arthur Cazabonne, Bernhard Sieglin, Gergely Papp