1. GO3.00001: Progress toward fully non-inductive operation in NSTX Jonathan Menard, PPPL For the NSTX Team 47 th Annual Meeting of the DPP Monday–Friday, October 24–28, 2005 Denver, Colorado Culham Sci Ctr U St. Andrews York U Chubu U Fukui U Hiroshima U Hyogo U Kyoto U Kyushu U Kyushu Tokai U NIFS Niigata U U Tokyo JAERI Hebrew U Ioffe Inst RRC Kurchatov Inst TRINITI KBSI KAIST ENEA, Frascati CEA, Cadarache IPP, Jülich IPP, Garching ASCR, Czech Rep U Quebec College W&M Colorado Sch Mines Columbia U Comp-X General Atomics INEL Johns Hopkins U LANL LLNL Lodestar MIT Nova Photonics New York U Old Dominion U ORNL PPPL PSI Princeton U SNL Think Tank, Inc. UC Davis UC Irvine UCLA UCSD U Colorado U Maryland U Rochester U Washington U Wisconsin Supported by Office of Science

2. 2 NSTX Plasma Current Sustainment Goals:  pulse >>  CR, 100% non-inductive, extend to high  Longest I P flat-top durations nearly doubled during this run –TF upgrade in 2005 allowed operation to I 2 dt limit at 4.5kG for 1.5s pulses –Operate at I P = 700-800kA to reach OH & TF heating limits simultaneously Utilize rotational stabilization of RWM + recently enhanced shaping –Sustain high  N,  P, and bootstrap fraction –See contributed orals this session to hear more about NSTX RWM physics 2005  2004 I P flat-top duration in seconds Time-averaged  T versus shaping factor S  q 95 I P /aB T

3. 3 New divertor poloidal field coils have significantly enhanced the plasma shaping capabilities of NSTX Highest   2.75 now obtained at highest   0.8, S  q 95 I P /aB T  37 –Record stored energy = 430kJ at I P =1.4MA,  N =5.3,  T = 29%, q*=2.3 20042005 New divertor coilOld divertor coil Small ELM regime recovered at high  > 2.5 with new divertor coils –Previously observed onset of large ELM-like events when  > 2.2   All values at MAX(  T ) > 20% Record stored energy = 430kJ at I P = 1.4MA 20042005 See D. Gates invited talk poster

4. 4 Record discharge pulse-lengths have been achieved by operating with sustained H-mode and high  N H-mode with small ELMS  reduced flux consumption, slow density rise  N > 4 for  t > 1s at high  P > 1 increases bootstrap fraction, lowers V LOOP NN I P (MA) V SURFACE (V)  T = 17%  P = 1.5, l i = 0.6  CR Time (s) n e / n GW EE H 89P = 2-2.2 H 98(y,2) = 1-1.1 116318  = 2.4,  L = 0.77,  R SEP = -1cm

5. 5 MSE data indicates low loop-voltage phase ends at onset of saturated n=1 mode when q MIN  1 Saturated n=1 mode persists for 0.5s late in discharge evolution Central rotation drops by factor of 3 at mode onset –Edge f  maintained –T i / T e  1 (not shown)  N = 6 decreases to 4 –  N = 6 above no-wall limit –  N = 4 near no-wall limit No RWM observed… q MIN sustained near 1 –No sawteeth observed –Discharge runs out of OH flux and TF flat-top –Possible “hybrid” mode Time (s) Core f  (kHz) Edge f  NN q MIN q MIN without E r correction (Nova Photonics) P NBI = 6MW

6. 6 Longest duration discharges exceed 60% non-inductive current fraction during high-  phase 85% of non-inductive current is  p-driven = BS + Diamagnetic + PS Neutron rate comparison (normalized) TRANSP agrees with measured neutron rate to within ± 15% during high-  phase Normalize at high   TRANSP over-predicts neutron rate early and late in shot –Low-f MHD is present at these times  fast-ion diffusion and/or loss likely –Assessing impact of MHD on J NBI profile and q-profile evolution 10 14 s -1 Diam+PS f NI > 60%

7. 7 MSE diagnostic enables testing of models of inductive and non-inductive current drive sources Compute V LOOP distribution/evolution directly from MSE-constrained fits –Long pulse-length and quiescent discharges needed for analysis Fit T, p, Z eff to , compute  NC, J OH & J BS (Sauter model), add TRANSP J NBI Sauter collisional NC model consistent with experimental I P and J || 116313 Plasma Currents Equilibrium profile (dashed) I P =750kA, f NI = 55% Comparing Sauter to NCLASS models to assess role of aspect ratio, impurities, etc... NC model (solid)

8. 8 Neoclassical current profile analysis consistent with J || -profile peaking and q(0)  1 at end of discharge Under-predict total current during last 0.5s – due to model, or n=1 MHD? 116313 Plasma Currents Before late n=1 mode During late n=1 mode Equilibrium Profile

9. 9 TSC simulations indicate 100% non-inductive operation is possible with density control and confinement improvement Requires larger CD from NBI –Reduce n e to 3.5  10 19 m -3 –Increase T e from 0.8 to 1.5keV  Need increased H 98(y,2) = 1.4 Reversed q profile predicted –Evolves to monotonic q profile –q(t) decreases monotonically NI BS NBI PS+Diam IPIP q(0) q min Stationary q profile  need broader J || (r) –Broader J NBI from fast ion diffusion? –Off-axis Electron Bernstein Wave (EBW) CD NBI BS JB B2JB B2 See C. Kessel’s invited talk Friday = UI2.00004

10. 10 NSTX is integrating high  and high confinement with a high fraction of non-inductive current-drive High beta –High  at high   record pulse lengths >>  CR, sustained  N / l i = 9 –When q(0)  1, saturated n=1 can flatten   profile, degrade confinement High confinement –H-mode provides broad pressure profile  low-l i < 0.7 from J BS High non-inductive current fraction –Achieved  60% NI current fraction – dominated by J BS –Density control + enhanced confinement  f NI = 100% –Off-axis current drive needed for stationary J profile  EBW See contributed orals this session for NSTX Li & EBW research