Fig 1: ECEI setup in DIII-D
The Electron Cyclotron Emission Imaging diagnostic had been commissioned on the DIII-D tokamak (San Diego, CA) since 2010. Dual detector arrays provide simultaneous two-dimensional images of Te fluctuations over radially distinct and reconfigurable regions, each with both vertical and radial zoom capability. A total of 320=20 vertical x 16 radial channels (range from 75 -140 GHz) are available. The experimental results show that the acquisition of coherent electron temperature fluctuations can be as low as 0.1% with excellent clarity and spatial resolution.
Fig 2: Imaging of Alfvén eigenmodes on DIII-D illustrates significant improvements in the ECEI diagnostic technique
Fig 3: A sawtooth crash exhibiting complete reconnection is imaged on DIII-D with unprecedented clarity.
Since the commissioning of ECE-Imaging instrument on the DIII-D tokamak in 2010, many exciting physics results have been obtained, from identity of sawtooth precursor oscillations (Fig 4), to good match between observed Alfven instabilities eigenmode on DIII-D with modeling in numerical (Fig 5-6), to the identification of phase locking of magnetic islands (Fig 7), to the study of edge harmonic osillations (Fig 8-9), etc.
Fig 4: Two distinct m/n = 1/1 electron temperature perturbations are identified during the sawtooth ramp. The normalized electron temperature of each poloidal mode is reconstructed under an approximation rigid-body rotation in (a) and (b). When the contribution of slow (∼15-Hz) electron temperature profile evolution is removed, as in the corresponding plots in (c) and (d), the 2-D temperature fluctuations are found to be indistinguishable. The complete instantaneous view of an ECEI array is overlaid in (d) for reference.
Fig 5: A reversed shear induced Alfven eigenmode (RSAE) observed on DIII-D is compared to modeling in numerical eigenmode solvers. (a) In an ideal-MHD approximation, the n = 4 RSAE is predicted at 65 kHz for the experimentally obtained thermal plasma equilibrium. (b) A non-perturbative hybrid gyrofluid-MHD model implemented by TAEFL reproduces the same mode, but includes a distortion also observed by ECEI (c).
Fig 6: Eigenmode structures for three RSAEs excited near t = 550 ms are captured with ECEI and compared to simulated mode structures obtained with NOVA. The color bar given is applicable to the measured data from ECEI, while simulated modes are plotted on a normalized scale. The modes in (a) and (c) show a ‘‘fundamental’’ radial mode structure with a single maximum in the radial mode envelope. The mode shown in (b) exhibits a radial null in the fluctuation amplitude and is identified as the first radial harmonic of the n = 3 mode shown in (a).
Fig 7: (a) A spectrogram of magnetic fluctuations. Coalescence of 3/2 and 4/2 island frequencies occurs near t = 2580 ms. (b) and (c) Bicoherence estimates before and after phase-locking. (d) and (e) Local phase velocities near ρ = 0.4, measured by ECE-imaging.
Fig 8:Real (a) and imaginary (b) components of an n = 1 EHO, imaged by Fourier filtering techniques, from shot #145049, t = 2.708 s. Optically thick channels are mapped to the mean position of origin of the emission, as calculated using the synthetic diagnostic, while other data remains mapped to cold resonances. The solid contour marks ρ = 0.95, the outer boundary of the optically thick region for this case.
Fig 9: Comparison between M3D-C1 simulation (a), synthetic ECEI response (b), and ECEI measurement (c) of electron temperature fluctuations of an EHO (DIII-D, shot #157102 at 2420 ms).
Recently new horn array with System-on-Chip solution for ECEI diagnostic has developed and fabricated for DIII-D. The upgraded system is designed for improved sensitivity and enhanced EMI shielding to improve SNR for ELM imaging and thus has better shielding against RF heating and noise, greatly reduced noise temperature and compact system size. More details on System-on-Chip (SoC) solution can be found here.
Fig 10: System-on-Chip Based Horn Array for DIII-D