Development and Assessment of “X-point limiter” Plasmas M. Bell, R. Maingi, K-C. Lee Coping with both steady-state and transient (ELM) heat loads is a critical issue for ITER Is there an alternative to the “conventional” poloidal divertor? –Divertors are used because they are associated with the H-mode, but –H-mode can be obtained reliably without an X-point on the boundary e.g. JET in its early investigation of H-modes (1980s) – best  E Critical factor seems to be high magnetic shear in the edge e.g. H-mode (“pesky” - RJG) in TFTR with I p rampdown, high  P Rebut has suggested that the “X-point limiter” would a better approach (e.g. Alfvén prize address at EPS Conference, Rome, 2006) –There is a separatrix but it is just outside the LCFS –Contact point of limiter with LCFS is close to and just inside the X-point –Flux expansion near poloidal field null and tangential contact spread heat load Experiment in NSTX would be relevant both to ITER and ST development

Discharges Produced with X-Point Close to Inboard Divertor Plate , 0.5s Example: 1MA, 0.45T, 4MW NBI –H-mode transition at 0.15s –From XP-820 “EBW coupling” on 4/4/07 Advantageous to bring contact point of LCFS onto outer lower divertor plate –Avoid loading edge of inner plate –Discharges would contact at proposed location of LLD –Triangularity will be lower

Use PF2 Coils Rather than PF1A to Produce X-Point Filaments produce reasonable approximation to reference shot With same plasma current distribution, shifting divertor current from PF1AL to PF2L produces close to desired condition Increasing PF2L (and PF2U) raises X-point above outboard plate

Experiment Expected to Require ~1 Run Day Develop target at 1MA, 0.45T, 4MW using PF2 coils rather than PF1A –Use rtEFIT control of outer gap and preprogrammed I PF2L /I p ratio –May need to use PF1A coils to compensate for OH fringing field Scan X-point through outboard limiter surface –Adjust HFS gas to promote H-mode transition Uncertainty is state of “conditioning” of new contact area –Repeat shots to assess whether this is evolving Assess H-mode threshold and confinement scaling in NB power scan –Assemble full kinetic data for analysis –Measure heat loading on both divertors Consider repeating some conditions when LITER operating Experiment provides proposed X-point height scan (K-C. Lee) Also provides useful data for milestone: R(08-3) Study variation and control of heat flux in SOL

Execution Plan 1.Rerun shot (1.0MA) with 2 NBI sources & compare H-mode access and performance. (1) 2.If the flattop is too short, decrease the plasma current to 0.9MA (shot ) (1) 3.Reduce PF1AL, PF1AU/IP current ratio in the flattop progressively to zero, and set the ratio of PF2L to plasma current to 4kA/MA and PF2U to 2.5kA/MA. Run a shot with 2 NB sources. If necessary adjust outer boundary control parameters. Assess need for small programmed PF1A currents to compensate for the time-varying OH leakage field. (3) 4.Adjust the vertical position (downward) and, if necessary reduce the PF2L current control ratio to 3.5kA/MA, to bring the X-point close to, and if possible, through the lower divertor plate. (3) 5.Decide whether to lower PF2L current control ratio further to 3kA/MA depending on equilibrium shape achieved and plasma performance. (3 additional shots possible) 6.At lowest X-point achieved, assess whether conditioning of the new contact point is occurring (3) 7.When conditions stabilize, assess H-mode access (L-H and H-l) either by adding source C at 0.12s (as in reference shot) or by delaying the second NB source (B) by 50ms on successive shots. (2) 8.Assess whether to apply PWM to the final source to determine the threshold power more finely. (4) 9.Return to the original shape at step 2 and perform the same assessment of H-mode access. (3)