1

Solenoid-free Plasma Start-up In NSTX Using

Transient Coaxial Helicity Injection

Roger Raman, Univ. of Washington

For the NSTX Team

University of Maryland

9 May 2006

College Park, MD

Acknowledgments

T.R. Jarboe, D. Mueller, B.A. Nelson, M.G. Bell, M. Ono, M.J. Schaffer

Institutions

Univ. of Washington, PPPL, ORNL, GA, Nova Photonics

Supported byOffice ofScience

College W&MColorado Sch MinesColumbia UComp-XGeneral AtomicsINELJohns Hopkins ULANLLLNLLodestarMITNova PhotonicsNew York UOld Dominion UORNLPPPLPSIPrinceton USNLThink Tank, Inc.UC DavisUC IrvineUCLAUCSDU ColoradoU MarylandU RochesterU WashingtonU Wisconsin

Culham Sci CtrU St. Andrews

York UChubu UFukui U

Hiroshima UHyogo UKyoto U

Kyushu UKyushu Tokai U

NIFSNiigata UU Tokyo

JAERIHebrew UIoffe Inst

RRC Kurchatov InstTRINITI

KBSIKAIST

ENEA, FrascatiCEA, Cadarache

IPP, JülichIPP, Garching

ASCR, Czech Rep

2

Outline

• Motivation for solenoid-free plasma startup

• Helicity Injection & Implementation of CHI

• Requirements for Transient CHI

• Experimental results from NSTX

• Brief summary of HIT-II results

• Brief summary of other plasma-startup methods

• Summary and Conclusions

3

Solenoid-free Plasma Startup is Essential for the Viability of the ST Concept

• Elimination of the central solenoid:

– Simplifies the engineering design of tokamaks (Re: ARIES AT & RS)

– Allows access to lower aspect ratio (high β)

• CHI is capable of both plasma start-up and edge current in a pre-established diverted discharge

- Edge current profile for high beta discharges

4

160 kA of closed flux current produced in NSTX, without using a solenoid

• Transient Coaxial Helicity Injection (CHI), previously demonstrated on HIT-II* at U- Washington

• Important step in the production of a starting equilibrium for solenoid free operation

• Conventional tokamak uses solenoid – ARIES-AT has no solenoid

• ST has advantages of high β and good E

– ST reactor cannot use solenoid – Alternate method for plasma startup is essential for ST concept– Could also reduce the cost of a future tokamak reactor

*R. Raman, T.R. Jarboe, B.A. Nelson et al., Phys Rev. Lett. 90 075005 (2003)

5

6

To inject helicity one must apply voltage to a flux tube

7

NSTX incorporates toroidal insulation breaks to enable CHI operation

Transient CHI: Axisymmetric reconnection leads to formation of closed flux surfaces. Driven (Steady State) CHI: Non-axisymmetric modes needed for closed flux generation

Fast Camera (R. Maqueda, Nova Photonics & L. Roquemore, PPPL)

8

NSTX CHI Builds on Progress made on HIT-II

HIT-II (R/a = 0./0.2m) NSTX: (R/a = 0.86/0.68m, 30x HIT-II Vol)

9

Simultaneous Requirements for Transient CHI

• Bubble burst current*:

= injector flux

= flux foot print width

= current in TF coil

• Time needed to displace toroidal flux– For typical voltage at the injector after breakdown ~500V need ~1 ms to

displace 600 mWb

• Energy for peak toroidal current:

• Exceed Energy for ionization and heating to 20eV (~50eV/D)– For 2 Torr.L injected, need ~2kJ

inj

* T.R. Jarboe Fusion Tech. 15, 7 (1989)

1 12 22 2CV LI

2 2 22 /( )inj inj o TFI d I

d

TFI

10

Upgrades that enabled progress in Transient CHI

• Replaced rectifier PS with capacitor PS

• Added MOV and Snubber

• Gas & microwave injection from lower divertor region

• Improved Absorber insulator design

• Discharge resistor across capacitor bank

11

Improved pre-ionization to a level that results in injected gas 10 times less than in 2004

Location of center stack

Divertor gap ECH: T. Bigelow (ORNL)

Injected gas amount now same as that used for inductive discharges

12

Closed flux current generation by Transient CHI

Abs. Arc

R. Raman, B.A. Nelson, MG.Bell, et al., PRL 97, 175002 (2006)

13

Discharges without an absorber arc show high current multiplication ratios of 60

Zero injector current

14

Improvement in high current generation due to operation at higher voltage

LRDFIT (J. Menard)

•2006 discharges operated at higher toroidal field and injector flux •EFIT is done when no injector current is present •Magnetic sensors and flux loops used in reconstruction

15

Electron temperature and density profiles become less hollow with time

120814: Black: 8ms, Red: 12ms

Thomson Scattering (B. LeBlanc)

Profile becomes less hollow with time

Plasma and Injector current

8 ms

12 ms

16

Some discharges persist for as long as the equilibrium coil currents are maintained.

Fast camera: R. MaquedaThese discharges are sensitive to gas pressure

17

Movie of a high current discharge.

Fast Camera:R. Maqueda &L. Roquemore

18

19

Highest ever plasma current on HIT-II (300 kA) obtained using CHI start-up

• No transient coil currents– Discharges initiated &

maintained in diverted configuration

• CHI is more tolerant to field errors– Discharges can be initiated

when -VLoop is applied to CHI

20

Current multiplication needed for larger machines achieved on NSTX

• Attainable current multiplication (CM) is given as:

• For similar values of BT, toroidal flux,

• So CM in NSTX should be 10x HIT-II, which is observed

• CM is understood from HIT-II and NSTX results

• No reason to believe CM will get worse in a larger machine (may get better)

( / )P inj T injI I

~ 10NSTX HIT IIT T

21

Maximum injector currents determined by system voltage limits

• Assuming constant resistivity on field lines,

For similar values of , at the same voltage,

in HIT-II should be ~10 x NSTX

• Consistent with observed ~15-30kA in HIT-II vs ~2-4 kA in NSTX

• Also consistent with the bubble burst relation, which requires 10x more current in HIT-II than in NSTX

• With 10-20kA of injector current MA levels of startup current should be possible in larger STs

( )injinj inj

TI V

inj

injI

injI

22

Full 2kV capability in NSTX would increase Ip ~ 300kA.

HIT-II data: R. Raman, T.R. Jarboe et al., Nuclear Fusion, 45, L15-L19 (2005)

Voltage, flux optimization allowed HIT-II to increase closed flux current as capacitor charging voltage was increased

23

Plasma Startup Using Outer Poloidal Field Coils

Outer PFs has been used to startup the plasma:

• MAST (START) - poloidal field coils + radial compression

• JT-60U - Aggressive application of RF heating and current drive

Three important conditions for “Inductive” plasma startup:

1. Satisfy the “Lloyd condition” for plasma startup.

With strong ECH (in DIII-D), ET·BT / BP ≥ 0.12  kV/m.

2. The field null must be maintained for a sufficient duration

2 – 3 ms in the presence of wall eddy currents.

3. Sufficient available flux for subsequent current ramp up to ~ 500 kA.

24

Creation of High-Quality Field-Null with Significant Poloidal Flux is Possible with NSTX PF Coils

Radial profile of flux Mod-B contours (Gauss)

Flux contours

• ~0.15 Vs available on NSTX

• 1 MA achieved with ~0.3 Vs from solenoid

• Sugggests possibility of ~ 500 kA in NSTX

- ~20 kA produced in FY 2005

- Improvements to pre-ionization are needed

- e.g. plasma injectors, ECH

24 kA

- 24 kA

3.4 kA

W. Choe, J. Kim and M. Ono, NF 45 (2005) 1463

25

J. Menard

Improved preionization and PF coil waveforms to be optimized

Solenoid-free Start-up with Outer Poloidal Field CoilsThree scenarios being investigated

26

• At low IInj and/or high pitch angle, helical current filaments form

– M = Ip/Iinj = Geometric stacking

• Filaments merge into cylindrical sheet as IInj increased, pitch decreased

– M > Geometric stacking

• At low fields (B ≈ 0.01 T, B ≈ 0.005 T), tokamak-like relaxation occurs

– M >> Geometric stacking

Filaments Reconnected Sheet Relaxed

GDG, ISTW 2006, ChengduPEGASUS Toroidal Experiment University of Wisconsin-Madison

Pegaus – Plasma Gun Injector Produces Current FilamentsRelaxed plasmas can be formed

Greg Garstka (Pegasus)

27

160kA closed flux current in NSTX validates feasibility of CHI for high current generation

• NSTX has addressed the plasma startup issue for the ST by demonstrating a viable alternate plasma startup method

– World record for non-inductively generated closed flux current– Other Solenoid-Free plasma start-up methods are also being tried (Synergistic

effects)

1) demonstration of the process in a vessel volume thirty times larger than HIT-II on a size scale more comparable to a reactor,

2) a remarkable multiplication factor of 60 between the injected current and the achieved toroidal current, compared to six in previous experiments,

3) results were obtained on a machine designed with mainly conventional components and systems,

4) indicate favorable scaling with machine size 5) HIT-II results indicate efficient coupling to other current drive methods is

feasible

• Future experiments in NSTX to explore coupling to OH– 200kW ECH to heat the CHI plasma– Coupling to RF and NBI