Fast Probe Results and Plans

1
Fast Probe Results and Plans
By J. Boedo For the UCSD and NSTX Teams
2
Probe Installed and Commissioned
Location is 6/8 below midplane
11 deg angle
6.8
3
New head and tips to accommodate curvature and
heat flux
Thermal Conductivity Thermalgraph Tips (parallel)
800 W/mK Copper 401 W/mK POCO graphite 72 W/mK
4
Existing Probe Head has 10 Tips

• Tips in blue will be active on day one, the rest
  implemented as upgrades (if funded)
• Fluctuations to 1 MHz
• Two Vf tips used for Epol (and fluctuations)
• Two Vf tips used as Er (and fluctuations) gtgt
  Reynolds Stress
• One tip as Isat gtgt ne
• Two tips as double probe (Te and Ne profiles)

Vf1
Epol
Isat
Vf2
Er
Vf3
Vf4
Te fluct
Bt
Mach1
Double Probe
Mach2
5
Data Taken at High Spatial and Temporal Resolution
Insertion completed in 100 ms. 30 ms in
plasma Time resolution for Te and Ne is 1
ms Time resolution for fluctuations is 1 micro
sec Spatial resolution is 1.5-2 mm (tip size
probe motion)
6
Within the limited shot space, density and power
were varied. So scaling is inconclusive So
far decreases in H mode with power 2.5gt1.5
cm Will obtain wider operating space in the
future and produce scalings
7
H-mode Radial Electric Field Flat in SOL
8
Fluctuation Quantities to Characterize in NSTX
Need to add Te fluct capability
9
Harmonic Diganostic
Probe
V
R1
Ampli- tude detector
Amp
C1
C3
fdrive 400 kHz
R2
V_fast
I?
C2
400 kHz Bandpass filter
Ampli- tude detector
Pearson coil
Amp
Amp
250 W RF amp
I2?
800 kHz Bandpass filter
Ampli- tude detector
Amp
Amp
I_fast
Dummy transmission line
I_min
V_min
I? /I2?
Analog divider
Compa- rator
Compa- rator
Adder
Tunable LPF
Analog multiplier
Ref
Ref
Analog multiplier
Te_slow
10
Harmonics Technique

• Current to a DC-floating probe driven by
  sinusoidal voltage
• can be expressed as a series of sinusoidal
  harmonics 1
• where - amplitude of mth harmonic
• - Bessel functions of
  integer order k
• For eU0/kTe ltlt 1
• Thus Te can be determined from the
  ratio
• of the amplitudes of 1st and 2nd
  harmonics
• The error of this approximation for eU0/kTe 1
  is only about 5 1

1 Boedo et al, Rev. Sci. Instrum. 70 (1999),
2997
11
Need to Include Magnetic component
Particle Flux
Reynolds Stress (neglecting ion pressure
fluctuations)
Heat Flux
Parallel Current Flux
Helicity Flux
12
Normalized N fluctuations very large
13
Radial Turbulent Flux Lower in H-mode
Value is 3-5 1017 cm-2 s-1
14
Intermittent Plasma Events Pervasive
15
Power Spectrum Comparison
Power spectrum vs f (in kHz) from probe data
during gas injection (solid circles) and
without(x) and comparison to that obtained form
the GPI system (open circles). Note that the
power spans y 4 orders of magnitude indicating
excellent S/N ratio.
16
Intermittency (Radial Convective Transport) Lower
in H-Mode
L-Mode 109033
H-Mode 109052
Intermittency is very strong in NSTX!
17
Intermittency (Radial Convective Transport) Lower
in H-Mode
L-Mode 109033
H-Mode 109052
Intermittency is very strong in NSTX!
18
L-Mode 109033
Density is highly intermittent Rate of bursts is
1E4 Poloidal field is much less
intermittent Coherent modes show in
Etheta Instantaneous flux is 1-3E15 cm-2s-1
19
H-Mode 109052
Density is intermittent (ELMs?) Rate of bursts is
2E3 Poloidal field is much less
intermittent Coherent modes show in
Etheta Instantaneous flux is 2E15 cm-2s-1
20
Initial Reynolds Stress and Bicoherency
Measurements
H-Mode
Codes Written and Tested
L-Mode
21
Bispectrum
H-mode
L-Mode
f3f1f2
f3f1-f2
22
Bicoherency
H-mode
L-Mode
23
Profile Conclusions
Edge/SOL profiles available with high spatial
(2-2.5 mm) resolution Limited to 1086xx, 1089xx,
109032-109062 Fully calibrated data for
109032-xxx062 NSTX profiles seem different from
larger aspect ratio devices Te profile in the
edge/SOL is flat at 25-30 eV Ne profile has a
very long decay length. Reduced by power! H-mode
Er profile flat in SOL How do flat Te profiles
are maintained? Do we have (more) anomalous
radial transport? (intermittency) Adding fast Te
diagnostic
24
Future Work and Goals
Data obtained in L and H-mode. SOL length scales
with power. Ongoing Many puzzles need
addressing Features usually associated with LCFS
vicinity are different in NSTX, such as a
potential well, steep H-mode profiles, etc
Ongoing Intermittency is quite strong in NSTX
(very strong radial transport?) Need to quantify
electrostatic turbulence and Intermittency
(enough to explain strong transport?) Codes just
ported for electrostatic turbulence, ongoing for
intermittency! Bicoherency and Reynolds Stress
codes just finished. Investigate energy cascading
and self-organized flows Need to quantify
electromagnetic components of transport Upgraded
head partially designed and to be built in FY04
25
(No Transcript)
26
(No Transcript)
27
(No Transcript)