Page 5 - ntegrated diagnostic and data analysis system of the C-2W advanced beam-driven field-reversed configuration plasma experiment

10K114-4 Thompson et al.
Rev. Sci. Instrum. 89, 10K114 (2018)
the jet Thomson system is designed to accommodate electron temperatures from 10 eV to 2 keV.
Doppler spectroscopy measurements of the jet impu-
rity ion temperature and velocity profile are also collected
in the mirror region at z = 3.06 m. The system will have
multiple chords facing in both the positive and negative z
directions and initially focus on the O4+ impurity line at
34
by D-TAQ Solutions Ltd. and National Instruments Corpora- tion. Most signals are acquired in one of the four screened rooms, but some custom DAQ solutions provide at-the- vacuum-flange digitization of sensitive signals. Digitizing at the flange essentially eliminates issues of ground loop noise and reduces cabling down to three connections: fiber optic trig- ger cable, Ethernet cable, and power cord. A series of custom, TAE Technologies developed, DAQ systems based on Altera and Xilinx FPGAs are in use on C-2W.
Currently, C-2W generates ∼4200 individual raw data sig-
nals, typically voltage as a function of time V(t) arrays. This
is roughly half of the expected total when the diagnostic suite
39
C-2U prototypes.
37
Each ELA package consisted of a pyro-
Expected values, from measurements on C-2U,
278.1 nm.
are oxygen impurity T i ∼ 400 eV and oxygen impurity velocity
32
The jet plasma flows out through the CV magnetic mirror region and into the divertors. Magnetic field lines are designed to flare out within both the inner and outer divertors, Fig. 1, and terminate on either biasable metal electrode plates or a plasma gun. Results on the Gas Dynamic Trap (GDT) device35 indicate that appropriate configuration of the plasma regime in the divertors can greatly reduce electron thermal losses in the system. Divertor diagnostics are significantly expanded on C-2W in support of an extensive program to reduce electron heat loss.
Sixteen sets of end loss analyzers (ELAs) are deployed
36
􏰀i ∼ 10 km/s.
C. Divertor diagnostics
on C-2W.
The units are improved versions of the successful
electric crystal bolometer to measure the total particle power
density and a gridded ion energy analyzer that measures ion
current density and can also measure ion energy distribution.
Two sets of these devices are placed 180◦ apart on each of
the four biasing rings in one inner and one outer divertor.
The density of detectors provides a degree of radial resolution
and the ability to detect gross azimuthal asymmetries in the
plasma outflow. Early results from the outer divertor ELAs (ion
current density ∼5 kA/m2 and power density ∼1 MW/m2)36
37
A versatile insertable probe platform with ∼2 m of travel
will position Langmuir probes within the divertors to provide
information on the plasma density, potential, and electron tem-
38
tem using microwave sources operating at λ ∼ 4 mm.
A series of over 60 post-shot programs automatically draw metadata from the MSDB and use it to calculate physical quan- tities from diagnostic raw data after every plasma discharge. These programs are written in a standard form and style in the Python programming language. The ∼4200 raw data sig- nals currently collected on C-2W reduce to ∼800 physics data signals (e.g., line integrated plasma density from the FIR inter- ferometer ∫ nedl), which are also stored in MDSplus. The physics data signals are further combined and reduced into profiles of the plasma density ne(r, t) and other quantities of interest using Abel inversion11 and Bayesian inference.
ACKNOWLEDGMENTS
We thank our shareholders for their support and trust and all fellow TAE staff for their dedication, excellent work, and extra efforts.
1M. Tuszewski, Nucl. Fusion 28, 2033 (1988).
2L. C. Steinhauer, Phys. Plasmas 18, 070501 (2011).
3M. W. Binderbauer et al., AIP Conf. Proc. 1721, 030003 (2016).
4H. Gota et al., Nucl. Fusion 57, 116021 (2017).
5M. W. Binderbauer et al., Phys. Plasmas 22, 056110 (2015).
6 M. C. Thompson, H. Gota, S. Putvinski, M. Tuszewski, and M. Binderbauer,
Rev. Sci. Instrum. 87, 11D435 (2016).
7H. Gota, M. C. Thompson, M. Tuszewski, and M. W. Binderbauer, Rev. Sci.
Instrum. 85, 11D836 (2014).
8M. C. Thompson, J. D. Douglass, P. Feng, K. Knapp, Y. Luo, R. Mendoza,
V. Patel, M. Tuszewski, and A. D. Van Drie, Rev. Sci. Instrum. 83, 10D709 (2012).
are broadly consistent with the C-2U results.
to implement thermographic imaging of the divertor biasing plates.
Nonperturbing measurements of ne within the inner
perature.
divertor will be available from a 5 chord interferometry sys-
34
wavelength imagining system25 aimed at providing nonper-
turbing Te measurements in the inner divertor is also under
development along with a high throughput spatial heterodyne
34
by a Edgertronic camera (typically operated at 256 × 256 pix- els and 6148 fps) and a Vision Research Phantom v5.2 camera (typically operated at 256 × 256 pixels and 15 564 fps), respec- tively. Additional cameras are available for visualizing other areas as needed.
III. DIAGNOSTICS INFORMATION MANAGEMENT
C-2W diagnostic signals are digitized using data acquisi- tion (DAQ) electronics both developed in-house and provided
spectrometer for measuring ion velocity.
Imaging of one inner and one outer divertor is provided
We also plan
A dual
is fully deployed. Raw data are stored using MDSplus.
this scale, thousands of signals from hundreds of individual diagnostics, keeping track of the metadata necessary to inter- pret each raw signal into physically meaningful information is a challenge. For example, to convert one signal from one magnetic probe from V(t) to Bz(x, y, z, t) requires knowledge of at least the coil effective area, location on the machine in three dimensional space, effective resistance of the coil and cabling, termination resistance, active integrator time constant, and which DAQ channel the signal connects to. Our expe- rience on C-2U6 and C-27 vividly illustrated the significant difficulties that arise when this metadata is managed on an ad hoc basis. Therefore, we created a custom Machine State Database (MSDB) for C-2W built upon a PostgreSQL database with a Django web framework and RESTful application pro- gram interface (API). The MSDB, which currently contains ∼48 000 rows of data, serves as the central repository of diag- nostic metadata and the overall C-2W device configuration. It also tracks changes as the machine evolves over time.
On