Current holes at ASDEX Upgrade Presented by O. Gruber for D. Merkl, J. Hobirk, P.J. McCarthy, E. Strumberger, ASDEX Upgrade Team - hardware upgrades for.

Current holes at ASDEX Upgrade

Presented by O. Gruber

for D. Merkl, J. Hobirk, P.J. McCarthy, E. Strumberger, ASDEX Upgrade Team

- hardware upgrades for improved control

- integrated advanced scenarios

- ion ITB with current hole: equilibrium, current diffusion

- electron ITB with current hole

- summary

LT WS (W56) on Physics of Current Holes, Mito, Japan, 3-4 Feb 2004

EURATOM Association

ASDEX Upgrade: flexible heating and fuelling systems

Pellet Centrifuge:240 - 1000 m/srepetition rate 80 Hz

Neutral Beam Injection: - perp. heating 15 MW / 70-100 keV- tang. off-axis beams 5 MW/ 100 keV

Ion Cyclotron Resonance Heating: - 8 MW / 30-60 MHz- variable deposition

Electron Cyclotron Resonance Heating: 2 MW / 140 GHz / 2 s

4 MW / 105 -140 GHz / 10 s - on-line steerable mirrors

R= 1.65 ma = 0.5 mIp 1.4 MABt 3 T

ASDEXUpgrade

- extended pulse length to 10 s flattop (= 2-3R even at Te(0) = 10 keV)

- extended PF coil operational window to run <> = 0.55 discharges

- developed ICRH to routinely deliver > 5 MW in ELMy H-mode

- increasing W coverage of inner wall

First wall materials need minimum erosion & low Tritium retention

recent hardware upgrades improved control

● stepwise towards C-free interior - in 2004 campaign 70% of first wall covered

- W-divertor in upper SN C-divertor in lower SN

● Up to now: - all plasma scenarii still accessible usually W concentration below 10-5

- machine has been more ‚delicate‘ to run central RF heating & ELM control by pellets suppresses impurity accumulation at improved core confinement

W

W

W

W

W

• Towards a C-free first wall:

- W coating of LFS poloidal limiters (actively cooled) 2004

- W coated bottom divertor 2005

• Off-axis CD - upgrading of ECRH started (4 MW / 10 s / steerable mirrors / 105–140 GHz )

- LHR system (3.7 GHz) in discussion

Next hardware extensions

• Stabilizing shell for external kinks & active RWM control


TNBI A9DCN

CX/LHR

NI IMSE,...

YAG

ECRH

shell time constant for n = 1above 30 – 40 ms (d=3 cm, steel)

passiveshell

current feeders ?

2 sets of 8 active toroidal coils (n = 1, 2)

• Stabilizing shell for external kinks & active RWM control


passiveshell

current feeders ?

TNBI A9

DCN NI I

ECRHYAGMSE,...

A9 CX/LHR

ECRH

CAS-3D (P. Merkel): - shell currents for n=1 kink- extension to resistive wall- benchmarking with resistive 2d-wall code

More compact pulsed reactor / steady-state operation

H89-P N / q95 > 0.8

Aim to achieve these conditions in steady state:

- energy and particle exhaust need nenGW

- tolerable ELMs

bootstrap current fraction > 50% in stationary advanced H-mode: prime candidate for ´hybrid´ ITER scenario

BS fraction > 80% for continuous reactor operation: strong ITBs with reversed shear needed

2

Performance beyond H-mode: integrated "advanced" scenarios

N

H89-P

ITER

• ITBs produced in the current ramp-up with strong reversed shear at JET (using LHCD) and JT-60U (using NBI) showed the existence of an extended core region with zero toroidal current: current holes

• open questions: equilibrium, stability, transport, sustainment

• influence of size: duration limited by skin effect ?

ITB driven bootstrap current sufficient? full non-inductive current drive needed for sustainmen

t?

Motivation for current hole investigations

(0.8 MA, 2.7 T)

Ion ITB discharge with current hole

barrier extends over the qmin region

ITB driven bootstrap current and shear profile can be aligned

Ion ITB discharge with current hole: MSE results

geometry formula of MSEat ASDEX Upgrade

current hole lostwhen third source switched off (ITB lasts longer)

DTM

magnetic axis

current hole: equilibrium reconstruction

Cliste:- solves Grad-Shafranov equation using external magnetics and MSE data- cubic spines for basic functions prevents sharp „current hole“ edge- uses poloidal flux as main coordinate

- for very low central current densities, pol = as a function of

spacial coordinates is poorly defined convergence problems

- new version: mid tor

weighted sum of previous solutions for (R,z) and j(R,z)

during iteration to improve convergence (successive ´over-relaxation´)

NEMEC:- modified 3-d stellerator equilibrium code (S.P. Hirshman)- energy minimizing fixed / free boundary code assuming nested flux surfaces- uses toroidal flux as main coordinate


- good agreement of the q-profiles except of the current hole edge- measured position of the (2,1) DTM from SXR & ECE


- colored points are the MSE observation points- shaded area labels the current hole

Comparison of equilibrium reconstruction with MSE measurements

current hole: current diffusion

- ITB driven off-axis bootstrap current not sufficient to maintain current hole- initial current hole taken from CLISTE at 0.3 s vanishes within 100 ms

- fast diffusion of beam current density ?- high fast particle content may contribute to BS current

current hole: confinement

- reversed magnetic and velocity shear improve heat insulation in core T driven transport suppressed internal transport barriers (ITBs)

- stored energy of ion ITBs increases linearly with heating power

ITB scenario with counter-ECCD pre-heating

# 17542

Electron and ion ITB

- no MSE available- sawtooth-like crasches in Te due to collapsing electron ITBs during ctr-ECCD:

indicates strong reversed shear (previous AUG results) or current hole (JET)

Ctr-ECCD

Combined electron and ion ITB

• early ctr-ECCD produces electron ITB

• at delayed NBI onset, ion ITB develops combination of electron and ion ITB

• foot of electron ITB sits at smaller radius

Small current hole during on-axis ctr-ECCD (ASTRA)

TORBEAM: IECCD=70 KA

Summary

current holes in ion ITB discarges (early NI heating) observed :

- current hole diameter up to 25% of minor radius

- equilibrium reconstruction with CLISTE (convergence up to q0 40) and

NEMEC (q0 > 1000) possible

- ASTRA current diffusion simulations show no sustainment by off-axis BS current - anormal beam driven current diffusion & fast particle BS needed:

- off-axis co-CD supports: current hole lost with switch-off of tangential off-axis beam• current holes with on-axis ctr-ECCD: - electron ITBs

- combined electron and ion ITBs with both ECCD and NBI (Ti Te 8 - 10 keV)

- ASTRA simulations indicate small current hole during central ctr-ECCD

extended control tools for all scenarii:

- operation at high shaping

- variable schemes for profile control (pressure, momentum, density, j, impurities)

- variety of methods for NTM suppression

- ELM control via shaping (type II ELMs), QH-mode and pellet pacemaking

- kink and RWM control envisaged

Ion ITB discharge with current hole: SXR results

H98(y,2)

0.5

1.0

1.5

ne/nGW

0.2 0.4 0.6 0.8 1.0 1.2

Improved H-mode

High N

q95 = 3.3 - 4.3

ne/nGW

0.2 0.4 0.6 0.8 1.0 1.2

N

0

1

4

3

2

Improved H-mode

High N

N = 1.8

q95 = 3.3 - 4.3

Advanced H-modes: performance

* reactor relevant at medium densities : H89-P= 2.8, N= 3.2 (IAEA1998)

optimum exhaust close to Greenwald : H89-P= 2.4, N = 3.5 (H-mode WS 2001)

(at q95 = 3.5) - continuous transition

ITER

advanced

ITER

advanced

Advanced H-modes:

progress towards steady state & adv. performance

Steady conditions for many current redistribution times:

low *

- tripple product 1020 m-3keV s-1

- QDT(equivalent) 0.2

high Greenwald fraction

best combination of confinement, stability and density

at high > 0.4 and q95 3.5

higher q95 over-compensated

by enhanced performanceN H98-P / q95 = 0.35

(0.2 in conv. ITER)

2

ITBs: missing stationarity due to MHD events

-

• ITBs with early heating and RS

- limited by coupling of infernal (at qmin 2)

and extrnal kinks to N< 2

• ITBs with delayed heating - highest performance achievable - high performance terminated by ELMs

• Combined electron and ion ITBs - high performance terminated by central 2/1 MHD

Decisive influence of scenario:

sustained only with L-mode edge or poor H-mode edge at better performance discharges short compared with current diffusion time high control efforts required: p, j, MHD modes

a self-consistent scenario with reduced control requirements exists

N

H89-P

ITER

• high pressure gradient needed to get 80% bootstrap current fraction (Q >30)• reversed magnetic and velocity shear improve heat insulation in core

T driven transport suppressed internal transport barriers (ITBs) • ITB driven bootstrap current and reversed shear profile can be aligned • optimise MHD stability – high p-gradient at q(min) leads to global MHD modes

• combination of electron and ion ITB scenarii needed

Can tokamaks be optimised towards continuous reactor?

- foot of ITB at = 0.6

• reversed magnetic and velocity shear improve heat insulation in core

T driven transport suppressed internal transport barriers (ITBs)

• high pressure gradient needed to get 80% bootstrap current fraction (Q > 30)

• ITB driven bootstrap current and reversed shear profile can be aligned • optimise MHD stability: high p-gradient at q(min) leads to global MHD modes

• combination of electron and ion ITB scenarii needed



• at delayed NBI onset, ion ITB develops


Ion ITBs: barrier position and q profile aligned

- MHD modes trigger ITBs relation with rational q values

- strong barriers only in connection with reversed magnetic shear

• barrier extends over the qmin region

ITB driven bootstrap current and

shear profile can be aligned

(qmin)

pol

Ion ITBs: route to very high bootstrap fractions

ITB scenario with delayed heating:

- heating of 15 MW late in the current ramp- lower SN with high triangularity- transition to H-mode

0 1 t(s)

Ion ITBs: route to very high bootstrap fractions

800 kA:

ne/nGW=0.45

No-wall limit reached !? N = 4.0

H89-P = 3.2

Tio = 14

keV• first large ELM destroys ITB !

ITB scenario with delayed heating:

- heating of 15 MW late in the current ramp- lower SN with high triangularity- transition to H-mode

1 MA: Tio > 20 keV

1.5.1020 m-3keV s-1

≥ 60 % BS current

Motivation



• at delayed NBI onset, ion ITB develops


Highest performance achieved in Ion ITBs with reversed shear

- scenario extended to high confinement H89-P= 3.4 and high beta N= 4

- Ti Te 8 - 10 keV with ctr-ECCD and NI

- duration limited by strong ELMs, core and edge MHD modes

- up to now transient max performance not sustainable

- benchmark is advanced H-mode scenario

Summary (1)

Extended control tools for all scenarii:

- 10 s flat-top pulses allow current profile relaxation

- operation at high triangularities close to DN (= 0.55 achieved)

- variable heating / CD schemes for profile control (p, momentum, density, j, impurities)

• Active MHD control: - variety of methods for NTM suppression - ELM control via shaping (type II ELMs), QH-mode and pellet pacemaking reduced target loads, impurity control - kink and RWM control envisaged - disruption mitigation (not covered)

Summary (2)

Advanced H-mode scenario: a basis for ITER hybrid operation(even steady-state or ignition possible)

- relaxed low shear q-profile (long sustainment compared to res. diffusion) - control of density peaking & impurity accumulation with tailored heat dep.

- enhanced confinement H98-P= 1.1 - 1.5 and beta N> 3 (up to no-wall limit)

over substantial operational range of q95 , and density

- integration of type II ELMs close to Greenwald density and double null

- despite high densities, > 60% non-inductive current drive achieved• stepwise towards C-free interior (reduced erosion, T retention)

- all advanced plasma scenarii accessible with W concentration below 10-5

- impurity accumulation at improved core confinement suppressed

with central RF heating & ELM control by pellets

Current holes at ASDEX Upgrade Presented by O. Gruber for D. Merkl, J. Hobirk, P.J. McCarthy, E. Strumberger, ASDEX Upgrade Team - hardware upgrades for.

Documents