Figure 4

(a) Halves of two neighbouring heat shield DEMO divertor cassettes from Fig. 1 bottom with the hole in between, magnetically shielded by fish-scale toroidal shaping (sketched on top left, according to⁴⁷). AMC is connected to the divertor cassette, nested into the blanket module. The vertical electrical joints stop the AC magnetic field from returning back to the coil. The bottom part of AMC is unnecessary (its removal does not significantly change the desired influence on the plasma), serves only to protect the surrounding diagnostics from the kHz B-field. The simulation shows that 2 cm toroidal gaps between the heavy cassettes (as on ITER⁴⁸ due to engineering reasons) and 2 mm cuts in AMC (feasible as it’s a light tube) yields the strongest magnetic field penetration into the plasma. (b) COMSOL Multiphysics calculated magnetic field lines (for 1.6 kHz) penetrating through the divertor cassette hole. The induced eddy currents in the divertor targets and AMC stop its magnetic field returning straight back to the solenoid further end, thus penetrating deeper into the plasma. (c) Poloidal cross-section of a magnetic field line passing through the divertor region. Plasma progressively deviates outwards while flying through each swept coil (each with the maximum \(N_\mathrm {turns}I_\mathrm {coil}=1\) MegaAmperTurns), such that the strike point integrated shift is \(\lambda _{\mathrm {swp}}=9~\)cm above the unperturbed (\(I_\mathrm {coil}=0\)) (thin) line. For \(-I_\mathrm {coil}\), the shift is the same in opposite direction. Field lines outside the strike point shift by a similar distance. (d) Experimental profile of magnetic field B along coil axis inside the Alternating Magnetic Conductor. Sketched on right, the AMC shows \(L^{-1}\) decay with distance, whilst in free space far away from the coil it decays as \(L^{-3}\)⁴⁹.