NSTX IAEA FEC 2006 PD: S.A. Sabbagh 1
Supported byOffice ofScience
S.A. Sabbagh1, R. E. Bell2, J.E. Menard2, D.A. Gates2, A.C. Sontag1, J.M. Bialek1, B.P. LeBlanc2, F. Levinton3, K. Tritz4, H.
Yu3, and the NSTX Research Team
Resistive Wall Mode Active Stabilization in High Beta, Low Rotation Plasmas
Columbia UComp-X
General AtomicsINEL
Johns Hopkins ULANLLLNL
LodestarMIT
Nova PhotonicsNYU
ORNLPPPL
PSISNL
UC DavisUC Irvine
UCLAUCSD
U MarylandU New Mexico
U RochesterU Washington
U WisconsinCulham Sci Ctr
Hiroshima UHIST
Kyushu Tokai UNiigata U
Tsukuba UU Tokyo
JAERIIoffe Inst
TRINITIKBSI
KAISTENEA, Frascati
CEA, CadaracheIPP, Jülich
IPP, GarchingU Quebec
21st IAEA Fusion Energy Conference
16 – 21 October, 2006Chengdu, China
1Department of Applied Physics, Columbia University, New York, NY, USA2Plasma Physics Laboratory, Princeton University, Princeton, NJ, USA
3Nova Photonics, Inc., Princeton, NJ, USA4Johns Hopkins University, Baltimore, MD, USA
v1.0
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 2
Experiments in 2006 examined RWM physics and stabilization at ITER-relevant rotation
RWM active stabilization New control system installed RWM control demonstrated RWM actively stabilized in
slowly rotating plasmas
Plasma rotation control Sustained rotation by real-
time reduction of amplified error field
Reduced rotation by non-resonant magnetic braking
Passive plates Blanket modules PortControl Coils
Control Coils
ITER vessel
ITERplasma
boundary
0 1 2R(m)
Z(m
)
0
-1
-2
1
2
NSTX / ITER RWM control
Advantage: low aspect ratio, high provides high leverage to uncover key tokamak physics for ITER (e.g. RWM control, momentum dissipation)
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 3
2.01.51.00.50.0R (m)
-2
-1
0
1
2
120047t = 0.745s
Bru
Bpu
Br ext
n = 1
Brl
Bpl
controlcoils
Z (
m)
stabilizer plates Stabilizer plates for kink
mode stabilization
External midplane control coil closely coupled to vacuum vessel Similar to ITER port plug
designs
Internal sensors can detect n = 1 – 3 RWM Unstable n = 1 – 3 RWMs
already observed in NSTX (Sabbagh, et al., NF 46 (2006) 635.)
n > 1 RWM studied during n = 1active stabilization
RWM Active Feedback System Installed on NSTX
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 4
Dynamic error field correction (DEFC) increases pulse length in strongly rotating plasmas
No-wall limit
Rotating mode onset
control offopen-loop
Open+closed loop
Open-loop control of error field amplified by stable RWM yields higher rotation yields longer pulse
Combination of open + closed loop control yielded best result Rotation increase or
saturation at long pulse lengths - first time in NSTX
see OV/2-4 Menard
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 5
RWM stabilized at ITER-relevant rotation for ~ 90/RWM
First such demonstration in low A tokamak Exceeds DCON N
no-wall for n = 1 and n = 2
n = 2 RWM amplitude increases, mode remains stable while n = 1 stabilized
• Multi-mode research – connection to RWM stabilization in RFPs
(Sabbagh, et al., PRL 97 (2006) 045004.)0.40 0.50 0.60 0.70 0.80 0.90
t(s)
0.00.51.01.52.0
05
10152005
1015200.00.51.01.5
02468
Shot 120047
0
2
4
6
N
IA (kA)
Bpun=1 (G)
Bpun=2 (G)
/2 (kHz)
N > N (n=1)no-wall
120047
120712
< crit
92 x (1/RWM )
6420840
1.51.00.50.02010
02010
0
t(s)0.4 0.5 0.6 0.7 0.8 0.9
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 6
RWM stabilized at ITER-relevant rotation for ~ 90/RWM
First such demonstration in low A tokamak Exceeds DCON N
no-wall for n = 1 and n = 2
n = 2 RWM amplitude increases, mode remains stable while n = 1 stabilized
• Multi-mode research – connection to RWM stabilization in RFPs
n = 2 internal plasma mode seen in some cases
(Sabbagh, et al., PRL 97 (2006) 045004.)
t(s)0.4 0.5 0.6 0.7 0.8 0.9
0.40 0.50 0.60 0.70 0.80 0.90t(s)
0.00.51.01.52.0
05
10152005
1015200.00.51.01.5
02468
Shot 120047
0
2
4
6
N
IA (kA)
Bpun=1 (G)
Bpun=2 (G)
/2 (kHz)
N > N (n=1)no-wall
120047120712
< crit
92 x (1/RWM )
1207176420840
1.51.00.50.02010
02010
0
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 7
RWM stabilized at ITER-relevant rotation for ~ 90/RWM
First such demonstration in low A tokamak Exceeds DCON N
no-wall for n = 1 and n = 2
n = 2 RWM amplitude increases, mode remains stable while n = 1 stabilized
• Multi-mode research – connection to RWM stabilization in RFPs
n = 2 internal plasma mode seen in some cases
Plasma rotation reduced by non-resonant n = 3 magnetic braking Non-resonant braking to
accurately determine RWM critical rotation
(Sabbagh, et al., PRL 97 (2006) 045004.)
t(s)0.4 0.5 0.6 0.7 0.8 0.9
0.40 0.50 0.60 0.70 0.80 0.90t(s)
0.00.51.01.52.0
05
10152005
1015200.00.51.01.5
02468
Shot 120047
0
2
4
6
N
IA (kA)
Bpun=1 (G)
Bpun=2 (G)
/2 (kHz)
N > N (n=1)no-wall
120047120712
< crit
92 x (1/RWM )
1207176420840
1.51.00.50.02010
02010
0
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 8
0.9 1.0 1.1 1.2 1.3 1.4 1.50.0
0.1
0.2
0.3
0.4
Dotted curve: 120038, t = 0.535s
/
A
passively stabilized by rotation(120038, t = 0.535s)
experimental critical rotation profile
(120712, t = 0.575s)
activelystabilized
(120047t = 0.825s)
R (m)
crit/A
(axis)
(q=2)
Rotation reduced far below RWM critical rotation profile
Rotation typically fast and sufficient for RWM passive stabilization Reached /A = 0.48|axis
Non-resonant n = 3 magnetic braking used to slow entire profile The /crit = 0.2|q = 2
The /crit = 0.3|axis
Less than ½ of ITER Advanced Scenario 4 /crit (Liu, et al., NF 45 (2005) 1131.)
Rotation profile responsible for passive stabilization, not just single radial location see paper EX/7-2Rb Sontag
(120717, t = 0.915s)
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 9
0.9 1.1 1.3 1.5R (m)
TN
TV (
N m
)Observed rotation decrease follows NTV theory
3
4
2
1
0
measured
theoryt = 0.360s116931
axis
n = 3 applied fieldconfiguration
First quantitative agreement using full neoclassical toroidal viscosity theory (NTV) Due to plasma flow through
non-axisymmetric field Computed using experimental
equilibria Trapped particle effects, 3-D
field spectrum important
Viable physics for simulations of plasma rotation in future devices (ITER, CTF, KSTAR) Scales as B2(Ti/i)(1/A)1.5 Low i ITER plasmas will have
higher rotation damping(Zhu, et al., PRL 96 (2006) 225002.)
see EX/7-2Rb Sontag
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 10
0.40 0.50 0.60 0.70 0.80 0.90t(s)
0
10
20
30
400
2
4
6
8
100
2
4
6
Shot 119755
45o (119761)
315o (119755) 290o (119764)
225o (119770)
negativefeedbackresponse
f
6
4
2
086420
30
20
100
N
/2 (kHz)
Bpun=1 (G)
0.50.4 0.6 0.7 0.8 0.9t (s)
feedback on RWM active feedback on n = 1 Control current relative
phase, f
Phase scan shows superior settings for negative feedback Pulse length increases Internal plasma mode seen
at f = 225, damped feedback system response
Gain scan also performed Sufficiently high gain
showed feedback loop instability
Varying relative phase shows positive/negative feedback
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 11
Poloidal n =1 RWM field decreases to near zero Radial RWM field increasing
Subsequent growth of poloidal RWM field Asymmetric above/below
midplane
Midplane radial sensor shows RWM bulging Upper/lower radial sensors show
decrease, while midplane sensor increases
Theory: may be due to stable ideal n = 1 modes becoming less stable (e.g. q evolution)
RWM may change form and grow during active control
0.55 0.59 0.63 0.67 0.71 0.75t(s)
02
4
6802468
10120
10
20
300
2
4
6
Shot 120048
0
2
4
6
Bp n = 1 lower
Br n = 1 lower
Shot 120048
0.590.55 0.63 0.67 0.71 0.75t (s)
6420
840
2010
0
420
6420
N
/2 (kHz)Bpu,l
n=1 (G)
Bru,ln=1 (G)
Br extn=1 (G)
N collapse
120048bulge
Future research will assess usingcombined sensors for optimization
= crit
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 12
Clear differences between RWM and internal plasma modeNo active stabilization
(RWM disrupts plasma)Active stabilization
(fast N drop, plasma recovers)
0.56 0.60 0.64 0.68
20
40
0
t(s)
0.5
1.0
1.5
0.0
Bp
n=
1 (
G)
20
40
0
Bp
n=
1 (
G)
US
XR
(a
rb)
edge
core 0.5
1.0
1.5
0.0
US
XR
(a
rb)
edge
core0.7240 0.7241 0.72420.7239
t(s)
0.66 0.70 0.74 0.78t(s)
Internal mode ~ 25 kHz Plasma rotation ~ 12 kHz (n = 2)
5 ms
20 s
120717
negativefeedbackresponse
120705
(USXR: K. Tritz JHU)
mode strongest at edge
0.56 0.60 0.64 0.68t(s)
0100
200
300
Shot 120705
010
20
3040
0.66 0.70 0.74 0.78t(s)
0100
200
300
Shot 120717
010
20
3040
0.56 0.60 0.64 0.68t(s)
0100
200
300
Shot 120705
010
20
3040
0.66 0.70 0.74 0.78t(s)
0100
200
300
Shot 120717
010
20
3040
n = 2
RWM n = 1
n = 1n = 2
n = 2 internalmode
for toroidal mode number:
0.2 0.4 0.6 0.8 1.0Time (s)
0
20
40
60
80
100
Freq
uen
cy (
kH
z)Shot 120717B() spectrum
1 2 3 4 5
0.6 0.8
20
40
0t(s)
(kHz)
n = 1n = 2
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 13
NSTX begins RWM active stabilization research relevant to ITER and beyond
First demonstration of RWM active stabilization in high , low A tokamak plasmas with significantly less than crit
In the predicted range of ITER Plasma response to feedback control demonstrated
Stability of n = 2 RWM demonstrated during n = 1 RWM stabilization n = 1,2 plasma mode sometimes observed; fast collapse, recovery
Plasma rotation reduction by non-resonant applied field; follows NTV theory Full NTV calculation yielding quantitative agreement to experiment Key component of RWM stability physics and dynamics; general
momentum transport relevance
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 14
Additional slides for poster follow
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 15
Work slides follow
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 16
NSTX begins RWM active stabilization research relevant to ITER, KSTAR, CTF
Passive stabilizers
(IVCC) (top)Passive stabilizers Blanket modules Port
Control Coils
Control Coils
ITER vessel
ITERplasma
boundary
0 1 2R(m)
Z(m
)
0
-1
-2
1
2
NSTX ITER KSTAR
KSTAR physics design supported by past design studies, present experiments in NSTX, DIII-D
In-Vessel Control Coils
NSTX IAEA FEC 2006 PD: S.A. Sabbagh 17
Close connection to present experiments in NSTX impacts the KSTAR stability physics study
RWM active stabilization demonstrated in low rotation (ITER-relevant) plasmas (Sabbagh, et al., PRL 97 (2006) 045004)
KSTAR with co-NBI should have rotation control for experiments Precise plasma rotation control through neoclassical
toroidal viscosity (Zhu, et al., PRL 96 (2006) 225002)
n = 2 non-resonant magnetic braking possible rotation control option for KSTAR
Unstable resistive wall mode with toroidal mode number n > 1 observed (Sabbagh, et al., NF 46 (2006) 635)
May need to address n > 1 unstable modes in KSTAR at the highest beta or for certain equilibrium profile shapes
RWM critical rotation speed (H. Reimerdes, et al., PoP 13 (2006) 056107)
Dependence on aspect ratio, Alfven speed, ion collisionality – key research topic
Top Related