Search for a command to run...
A series of observations is presented concerning divertor heat flux, qdiv, in the DIII-D tokamak, and it is shown that many features can be accounted for by assuming that the heat flux flows preferentially along field lines because τ|| < τ⊥ in the scrape-off layer (SOL). Exceptions to this agreement are pointed out and the discrepancies explained by means of two dimensional (2-D) effects. About 80% of the discharge input power can be accounted for. The power deposited on the target plate due to enhanced losses during edge localized modes (ELMs) is less than 10% of the total target power in most cases. X point height scans for lower single null (LSN) diverted discharges show that the peak heat flux variation is primarily due to flux expansion and secondarily due to transport of energy across the magnetic field in the divertor. At the outer strike point qdiv,peak ∝ Pin(Ip - Ip,0)G(gin)(1/Bt)4/9(Bdiv/Bmp)f(Ldivχ⊥), where G is a linear function of the inner gap, gin, over a specified range and f describes cross-field energy transport in the divertor. Evidence of radial in-out asymmetries (comparing the outer strike point with the inner strike point or centre-post) and toroidal asymmetries in qdiv is presented and the heat flux peaking due to tile gaps and misalignment of tiles is examined. For magnetically balanced double null (DN) discharges with downward ∇B ion drift, it is found that qdiv is inherently higher in the lower divertor than in the upper divertor, having a 3:1 downward bias. Examples of heat flux reduction by gas puffing deuterium or neon in LSN and DN discharges are given. At least a threefold reduction of the peak heat flux in both the upper and lower divertors of a DN discharge, using D2 puffing, is reported.