A detailed comparison of antenna impedance measurements on ASDEX Upgrade with the ion cyclotron range of frequencies antenna code TOPICA

Abstract

Abstract. New antenna diagnostics on ASDEX Upgrade, in the form of voltage and current probe pairs on the feeding lines of each ICRF antenna, close to the input ports, has made it possible to study in detail the behavior of the ASDEX Upgrade two-strap antenna under changing loading conditions, and compare these measurements with results of simulations using the TOPICA code. The present work extends previous studies by using the input impedance (more precisely, the complex voltage reflection coefficient Γ) on each antenna port for comparison, instead of the more commonly used loading resistance or coupled power. The electron density profiles used for the simulation were reconstructed from DCN interferometer and Lithium beam emission spectroscopy measurements, ELMsynchronized and averaged over time intervals from 10 to 200 ms depending on the case; 112 cases were compared from seven ASDEX Upgrade discharges with widely different plasma parameters and two operating frequencies (30 and 36.5 MHz). Very good agreement in |Γ| is found with the measurements on antenna 3 (< 3% averaged over a shot), good agreement is found with antennas 1 and 2 (< 10%); the code reproduces the correct trend in loading resistance R L in a significant majority of cases, although the discrepancies in the absolute values can be rather high (up to ∼ 50%) due to high reflection. Sources of discrepancy are discussed.PACS numbers: 52.50.Qt TOPICA study on ASDEX Upgrade 2

Validation of the ICRF antenna coupling code RAPLICASOL against TOPICA and experiments

“…At the time this experiment was done (March 2013), antenna 3 had a closed Faraday screen, which is consistent with higher reflection coefficients [28] (all antenna models in this paper have open Faraday screens). Recent measurements indicate that antenna 1 is a few mm closer to the plasma than antenna 3, which is also consistent with lower reflection (better coupling) at antenna 1, but [3] concludes that the difference in radial position is too small to explain the observed differences. Toroidal non-uniformity of the plasma and/or RF probe calibration uncertainties [3] may also play a role.…”

“…Recent measurements indicate that antenna 1 is a few mm closer to the plasma than antenna 3, which is also consistent with lower reflection (better coupling) at antenna 1, but [3] concludes that the difference in radial position is too small to explain the observed differences. Toroidal non-uniformity of the plasma and/or RF probe calibration uncertainties [3] may also play a role. We see that "TOPICA curved", "TOPICA flat", and "RAPLICASOL imported" give very similar results for the reflection coefficient (table 4).…”

“…The TOPICA calculations with which we wish to compare our RAPLICASOL results were originally done with a curved geometry [3,4]. A comparable flat TOPICA geometry exists whose results have been compared with those of the curved geometry [26], and were found to be in close agreement (on the order of 5% difference in coupled power).…”

“…The corresponding voltage reflection coefficients at the ports of antennas 1 and 3 were also measured using voltage and current probes. At the time, TOPICA calculations were performed [3,4] to compute the reflection coefficients. They match the experimental values closely: the RMS difference between the predicted and observed reflection coefficients is 0.05 for antenna 1 and 0.02 for antenna 3.…”

“…In general, the simplified flat 3D model in TOPICA can be considered as a rather good approximation of the real curved one, in particular for the estimation of antenna coupling [3,4,26,27]. However, for more local phenomena such as the radiated electric fields, there are significant differences between the flat and the more realistic curved model [27].…”