R. A. Pitts et al., O-161 of 12PSI17, Heifei Anhui, China, 22-26/05/2006 Parallel SOL flow in TCV R. A. Pitts, J. Horacek, W. Fundamenski 1, A. Nielsen.

R. A. Pitts et al., O-16 1 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

Parallel SOL flow in TCVParallel SOL flow in TCVR. A. Pitts, J. Horacek, W. Fundamenski1, A. Nielsen2,

O. E. Garcia2, V. Naulin2, J. Juul Rasmussen2, M. Wischmeier Centre de Recherches en Physique des Plasmas, Association EURATOM-Confédération

Suisse, École Polytechnique Fédérale de Lausanne, CH-1015, Switzerland 1UKAEA/Euratom Fusion Association, Culham Science Centre, Abingdon, UK

2Association-Euratom Risø National Laboratory, Roskilde, Denmark

Centre de Recherches en Physique des Plasmas

Brief introduction to flowsExperimental arrangementParallel flow in FWD-B and REV-B density variationEffect of location in the outboard midplane vicinityUnderstanding the flowsConclusions

OutlineOutline

R. A. Pitts et al., O-16 2 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

B

BxB

ErxB, pxB

Ballooning

Pfirsch-SchlüterDivertor sink

ExB

Simplified – flow components in poloidal plane only

Poloidal

Parallel

Motivation – understanding SOL flowsMotivation – understanding SOL flows

FWD B

B

BxBREV B

More in Review paper by N. Asakura, Fri. morning

R. A. Pitts et al., O-16 3 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

Magnetic configurationsMagnetic configurations#26092 #27585 #27582 #27588

BxBxBB BxBxBBOnly ohmic diverted plasmas, with emphasis on direction of B, configuration and density (|B| = 1.43 T)B and Ip always reversed together to preserve helicity

Mach probe

R. A. Pitts et al., O-16 4 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

Mach probeMach probeFast reciprocating probe with Mach probe head mounted on the machine midplane

Mach No. defined in the usual way:

Postive flow defined UPWARDSAll data mapped to outer midpl.

sat,2sat,1 II0.4l||M n 12

10 mm

Two separate heads used to account for differing poloidal contour as plasma displaced downwards

12

Non-Mach pins used to measure profiles of ne, Te, Vplasma

R. A. Pitts et al., O-16 5 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

Case:Case: REV-B, SNL, +340 kA, n REV-B, SNL, +340 kA, nee scanscanFlow always towards

outer target in REV-B

High flow (M|| ~ 0.6) near separatrix at low density

Flow decreases with increasing density

Very similar in He

OU

TE

R d

ivertor

en (1019m-3)3.14.86.5

9.08.4

And in ohmic H-mode (Type III)

OU

TE

R d

ivertor

en (1019m-3)3.14.86.5

9.08.4

OU

TE

R d

ivertor

en (1019m-3)3.14.86.5

9.08.4

6.0

R. A. Pitts et al., O-16 6 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

Case:Case: FWD-B/REV-B, FWD-B/REV-B, 260 kA, n260 kA, nee scanscan

OU

TE

R

diverto

r

en (1019m-3) 1.7 2.5 4.2 7.36.3

REV-B

Very similar to 340 kA in REV-B

Switch direction with B

Always co-current

Decrease with increasing ne

Behaviour consistent with Pfirsch-Schlüter

Slight negative offset (i.e. towards outer divertor)

OU

TE

R

diverto

r

en (1019m-3) 1.7 2.5 4.2 7.36.3

OU

TE

R

diverto

r

INN

ER

div

erto

r

en (1019m-3) 1.7 2.5 4.2 7.36.3

REV-B

FWD-B

R. A. Pitts et al., O-16 7 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

Field independent component (260 Field independent component (260 kA)kA)Assume flow made up

of two components:

B dependentB independent

mean of flows in FWD/REV-B gives B independent term

OU

TE

R

diverto

r

en (1019m-3)1.7 2.5 4.2

7.46.3

TCV data support small, offset term (M|| ~0.05 – 0.1) with indication of density dependence

Good candidate for “ballooning” driven flow

R. A. Pitts et al., O-16 8 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

Case:Case: REV-B, SNL, z-scan, 260 kA REV-B, SNL, z-scan, 260 kAO

UT

ER

dive

rtor

en = 4.2 x 1019m-3

Change in M|| with z consistent with a ballooning drive

Probe to outer target connection length increases as plasma is lowered vertically

+10 cm0 cm-10 cm

OU

TE

R d

iverto

rO

UT

ER

dive

rtor

R. A. Pitts et al., O-16 9 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

Case:Case: REV-B, SNL, z-scan, 260 kA REV-B, SNL, z-scan, 260 kA

en = 4.2 x 1019m-3

Good match in local SOL pe and Vp profiles

Ion pressure and Er main drivers for neoclassical flows

Not the cause of the flow offset

+10 cm0 cm-10 cm

R. A. Pitts et al., O-16 10 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

Return to previous 260 kA density scan in FWD/REV-B:

Main flow drive is Pfirsch-SchlüterMain flow drive is Pfirsch-SchlüterO

UT

ER

d

ivertor

INN

ER

div

erto

r

en (1019m-3) 1.7 2.5 4.2 7.36.3

REV B

OU

TE

R

diverto

r

INN

ER

div

erto

r

en (1019m-3) 1.7 2.5 4.2 7.36.3

REV B

FWD B

Choose radial band in the main SOL:8 < r-rsep < 12 mm

Take mean exptl. M|| and plot versus density

OU

TE

R

diverto

r

INN

ER

div

erto

r

OU

TE

R

diverto

r

INN

ER

div

erto

r

Compare with predicted Pfirsch-Schlüter flow 2B

Benp

Ec e

rs

2qcosMPS

||

R. A. Pitts et al., O-16 11 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

Return to 260 kA REV-B, high density case:

What drives the offset What drives the offset component?component?

Strong candidate is enhanced parallel pressure due to cross-field motion of interchange driven filaments in the outboard midplane region

SOLPS5 modelling without drifts shows Ti gradients develop above the X-pt. which can drive flows – open divertor geometry

OU

TE

R

diverto

rO

UT

ER

d

ivertor

ESEL #115, Density

Sep

aratrix

Wall

O. E. Garcia et al., PPCF 48 (2006) L1

See also P3-8, Thurs. afternoonEstimate M|| with simple Ansatz relating the transient filament overpressure to parallel flow (ESEL code 2D only)

W. Fundamenski et al., submitted to NF

OU

TE

R

diverto

r

R. A. Pitts et al., O-16 12 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

ConclusionsConclusions

First parallel particle flow measurements in the outboard midplane region of TCV have shown:

Flows can be large (M|| = 0.5-0.6) at low density

Decrease as density increases – almost stagnant at high ne

Reverse nearly symmetrically with B reversal

Are consistent with Pfirsch-Schlüter return as main drive

Clear, small (M|| ≤ 0.1) residual offset component towards outer target below and above midplane is consistent with enhanced, interchange driven transport in the midplane region

Further experiments necessary to eliminate divertor sink as a possible contributor

R. A. Pitts et al., O-16 13 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

Reserve slidesReserve slides

R. A. Pitts et al., O-16 14 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

Case:Case: REV-B, SNL z = 0, 260 kA, REV-B, SNL z = 0, 260 kA, nnee scanscan

en (1019m-3)3.04.3 6.7

4.8

OU

TE

R d

iverto

r

Flows on the midplane same direction and similar magnitude to those between midplane and X-point

Only REV-B data available

Very sensitive to ne

R. A. Pitts et al., O-16 15 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

Example probe and edge TS profilesExample probe and edge TS profiles

Good quality measurements for estimates of drift flowsClear flattening of density profile as density increasedProbe and edge TS scattering agree well in Te (not ne – understood)

R. A. Pitts et al., O-16 16 of 12 PSI17, Heifei Anhui, China, 22-26/05/2006

Return to 260 kA REV-B, high density case:

What drives the offset What drives the offset component?component?

Strong candidate is enhanced parallel pressure due to cross-field motion of interchange driven filaments in the outboard midplane region

SOLPS5 modelling without drifts shows Ti gradients develop above the X-pt. which can drive flows – open divertor geometry

OU

TE

R

diverto

r

ESEL #115, Density

Sep

aratrix

Wall

O. E. Garcia et al., PPCF 48 (2006) L1

See also P3-8, Thurs. afternoonEstimate M|| with simple ansatz:<M||> 0.5 t(p > <p>)/t assuming a “sub-sonic” transient flow of M|| = 0.5

W. Fundamenski et al., submitted to NF

OU

TE

R

diverto

r