Abstract Details
Abstracts
Author: Klissman H. Franco
Requested Type: Poster
Submitted: 2025-03-14 12:28:10
Co-authors: C. R. Sovinec
Contact Info:
University of Wisconsin-Madison
1500 Engineering Dr
Madison, 53706
United States
Abstract Text:
The nonlinear visco-resistive MHD code NIMROD is applied to study global-scale fluctuations for edge safety factor 0<q(a)<2, with a self-consistent evolving profile. This regime lies between conventional tokamaks and RFPs and is usually impossible to access in most devices due to disruptive kink instabilities. In the Madison Symmetric Torus (MST) this regime is achieved due to a feedback-controlled programmable power supply (PPS) driving plasma current, along with a close-fitting shell whose resistive wall time is longer than typical discharge duration [N. C. Hurst, et al., Phys. Plasmas 29, 080704 (2022)]. Relatively poorly understood fluctuation phenomena arise under these conditions. Our NIMROD simulations complement the experimental data by providing insights on relaxed profiles. These computations begin with vacuum field, and proportional-differential loop voltage control induces plasma current, modeling the PPS drive in experiments. The inclusion of Ohmic heating and anisotropic thermal conduction enables self-consistent investigation of transport in these self-organized states. Normalized computations with constant thermal diffusivities show qualitative agreement with observations, where 1/1 sawtoothing dominates in most of the 1<q(a)<2 regime, while in 0<q(a)<1, profiles are flattened by the resistive tearing fluctuations in the vicinity of well-separated rational surfaces with fluctuation amplitudes depending on the dominant helicity. Transition from 1/2 to 2/5 happens with a core minimum pitch, which moves to the edge when crossing the 1/3 rational surface. The inclusion of temperature-dependent thermal coefficients reveals that computations using Braginskii coefficients underestimate thermal transport. More realistic results emerge when incorporating nonzero normal magnetic field at boundary, hence resonant field errors, allowing heat loss to the wall through parallel conduction. Work supported by US DOE award DE-FG02-85ER53212.
Characterization: 1.0
Comments: