DIII-D RESEARCH OPPORTUNITIES FORUM FOR THE 2013 EXPERIMENTAL CAMPAIGN Review | Direct submission with log-in | Request submission without log-in
Print this page

Print this page
Title	240: Optimization of Applied Error Field Spectrum for NTV
Name:	Nikolas Logan ncl2128@columbia.edu	Affiliation:	Columbia University
Research Area:	Stability & Disruption Avoidance	Presentation time:	Requested
Co-Author(s):	E.J. Strait, J.K. Park, C. Paz-Soldan	ITPA Joint Experiment :	No
Description:	This proposal is the experimental optimization of NTV torque using DIII-D 3D field coil sets. The goal is to test the hypothesis that a single spectrum can be found to best couple to the plasma in such a way as to induce NTV torque.	ITER IO Urgent Research Task :	No
Experimental Approach/Plan:	The basic plan is to perform a scan of combined I and C coil currents to optimize the poloidal spectrum for n=3 non-resonant breaking. Assuming no (or corrected) intrinsic error field contributions and monotonic increase of NTV with applied amplitude of a given spectrum, the degrees of freedom are reduced to relative phases and amplitudes. Due to the limited phasing options of the six-coil coil sets applying n of 3 fields, the full operating space can thus be represented on a 2D grid of I vs. C coil current amplitude (where negative I-coil current implies odd parity, and negative C-current implies a similar parity with respect to the bottom I-coil). The trick here is the ability to follow a contour of constant area (or equivalently energy) norm for the applied plasma surface mode. These contours will need to be created in the two dimensional I and C-coil amplitude space prior to the experiment. This is not challenging using DCON. If the theory is valid, the trace of a single such contour will provide the optimum configuration. To validate this, however, a minimum of 3 shots should be used: 1) Establish equilibrium and shape identical to selected NTV reference shot (ex. 131861). 2) Energizing n=3 I-coils in odd parity, scan from positive to negative amplitude adding C-coil field as necessary to follow contour of applied energy norm spectra. During this stage NBI and gas/pumping should be on feedback to maintain as closely as possible plasma rotation, beta, density and shape. The NTV torque as a function of relative coil amplitudes will thus be available from the required NBI torque, and directly from the electromagnetic torque measured at the vessel wall by the new 3D magnetic diagnostics (see relevant devoted proposals). 3) Repeat step 2 for multiple initial amplitudes so as to effectively confirm the maximum torque occurs at a consistent value of I vs C current. If scans find optima with differing spectra, an empirical optimum must be found by filling the space more completely. If scans find optima with same spectra and time permits, non-resonant error field effect should be checked by comparing the two available phasing of the n=3.
Background:	Recent success of NTV experiments on DIII-D have provided a tool to effectively control plasma rotation, improving access to QH mode. They have also inspired significant advances in the theory and modeling of these non-resonant torques. This experiment will seek to optimize NTV torque as an experimental tool by choosing the optimal spectrum. This (as far as the author knows) has not been attempted beyond taking advantage of the n^2 dependence and choosing between even or odd parity based on pitch/kink resonance calculations. Further, the experiment seeks to validate qualitative details of the NTV models used in support of experiment. The Ideal Perturbed Equilibrium Code (IPEC) can currently be used to calculate the dominant (and ith-dominant) external field (with energy norm) for a given equilibrium maximizing the sum of the squared pitch resonant field at each rational surface. The code is being upgraded to perform a nonlinear optimization of the normalized external field for inducing NTV torque with the hypothesis a dominant mode structure will be found. The experiment would thus validate or invalidate the theories ability to predict optimized I/C-coil configurations for NTV torque experiments (impacting lower n non-resonant EFC). Measurement of the Maxwell stress tensor (and thus integral electromagnetic torque) have never before been available for n=3 error field experiments in DIII-D. If they are successful in measuring torque from non-resonant electromagnetic fields this will be an ideal candidate to showcase the new ability. The author expects that low m modes will dominate the NTV spectrum, allowing clear reconstruction of the plasma response across a large area of the vessel wall in the optimal configuration.
Resource Requirements:	Requires careful pre-determined mixing of I and C-Coil amplitudes, along with ability to phase shift at least 2 sets (i.e. C and top I coils). Also requires co and counter beam mixing and feedback to ensure profile / plasma shape changes do not obscure the NTV dependence on the poloidal profile.
Diagnostic Requirements:
Analysis Requirements:	The experiment requires some theoretical preparation: DCON analysis of energy norm contours in coil-amplitude space, and IPEC prediction of optimized coil fields with the corresponding falloff (ratio of NTV for the ith-dominant modes) to guide experiment in choosing intelligent scans through the operating space. Control room analysis of NTV (especially EM torque measurements) would guide the experiment as to how densely each operating space needed to be filled.
Other Requirements: