In order to determine the effectiveness of the B-Stark diagnostic in measuring the magnetic field line pitch, detailed comparisons need to be made with MSE over a range of fields and currents. These comparisons would be improved by taking measurements using a hydrogen neutral beam. A hydrogen beam will have sqrt(2) the wavelength separation of the Stark components as compared to deuterium, potentially improving our spectral fits. Characterization done with hydrogen will also be relevant to ITER, where a greater Stark separation will be seen as compared with DIII-D. The fitting model used for the B-Stark spectra uses the measured plasma density and temperature. Scans of these parameters will aid in validating the fitting model.
Experimental Approach/Plan:
Produce and maintain plasmas during which current, field, and density scan cans be performed. In addition to the scans we will also use current diffusion during the startup to obtain a q scan. Field to be scanned between 1.40 - 2.16 Tesla. Current scan to be performed over the maximum possible range consistent with q95 < 3, < 2.5 preferred. Density scan over the range 2x10^19 - 6x10^19. For these scans either L or H mode shots may be used. While dedicated shots duplicating scans taken with deuterium beams are desired, these measurements can likely be done piggyback.
Background:
A new diagnostic, B-Stark, has been developed for the DIII-D tokamak to measure the magnitude and direction of the internal magnetic field. This diagnostic relies on the relative line intensities and spacing of Stark split emission from the neutral beams. This technique may have advantages over a Motional Stark Effect (MSE) diagnostic in future devices, such as ITER. In ITER, films deposited on first surface mirrors are expected to cause changes in the polarization direction, interfering with the MSE diagnostic. The B-Stark diagnostic is not sensitive to changes in the polarization direction, only to polarization dependent transmission. As part of fitting the Stark spectrum we will be utilizing an atomic code to predict the population levels. Comparisons with MSE will aid in validating this code.