IEEE TRANSACTIONS ON PLASMA SCIENCE, VOL. 31, NO. 4, AUGUST 2003 745 Study of the Effects of ECH Power and Pulse Length on Preionization in the KSTAR Tokamak Young-soon Bae, Won Namkung, Member, IEEE, Moo Hyun Cho, and Alan C. England Abstract—In this paper, we present the results of a study of the preionization effects for the Korean Superconducting Tokamak Advanced Research (KSTAR) tokamak ( m, m, T, MA, s) that is under construction by the Korea Basic Science Institute (KBSI). The preionization will be given by the Electron Cyclotron Heating (ECH) System with an 84-GHz 500-kW gyrotron tube being made by Communications and Power Industries. The ECH preionization effects are investigated by a 0-dimensional code (TECHP0D 1 ) that includes the operational scenarios of KSTAR tokamak. The code is now improved and advanced with carbon, oxygen, and iron im- purity effects, and with the self and mutual inductances of seven pairs of superconducting poloidal coils for the KSTAR tokamak. Index Terms—Electron cyclotron heating (ECH), gyrotron, Korean superconducting tokamak advanced research (KSTAR), poloidal, preionization. I. INTRODUCTION E LECTRON cyclotron heating (ECH) preionization has been successfully applied in a variety of tokamaks and is normally used to produce the plasma in contemporary stellarators. The general conclusion of these experiments is that ECH was effective in producing a good plasma which would: (a) reduce the startup runaway electrons; (b) reduce the voltage required to start the plasma current; and (c) somewhat reduce the Volt-s expenditure from the transformer needed to establish the plasma. The motivation for many of those experiments was the pioneering work of Peng et al. [1]. While some of the predictions of the theory were not born out by the experiments, nevertheless, the theory was important in showing the value of preionization for voltage reduction. On the basis of the experimental work by Kulchar et al. [2], a model was produced which was able to adequately describe the experiments in ISX-B. Later experimental work included the large tokamak experiments in T-10 [3] and DIII-D [4]. The Modeling efforts of Fidone and Granata [5] showed that for a large tokamak, energy deposition would not occur at the electron cyclotron resonance heating (ECRH) layer. Maroli and Petrillo [6] added plasma radial growth and impurities. Lloyd et al. [7] produced a more advanced model and added the effect of impurities, but Manuscript received October 28, 2002; revised May 5, 2003. This work was supported by the Korea Basic Science Institute (KBSI) and by the Korea Atomic Energy Research Institute (KAERI). Y. S. Bae, W. Namkung, and M. H. Cho are with the Pohang University of Sci- ence and Technology, Pohang 790-784, Korea (e-mail: ysbae7@postech.ac.kr). A. C. England is with the Korea Basic Science Institute, Daejon 305-333, Korea. Digital Object Identifier 10.1109/TPS.2003.815474 1 0-Dimensional Tokamak ECH Preionization code. did not consider the error field effects and poloidal field coil circuit equations. Because of the thick vacuum vessel wall, superconducting poloidal field coils, and the concomitant limited current ramp rates, the generated loop voltage for breakdown may be too low to provide breakdown reliably in the Korean superconducting advanced research (KSTAR) tokamak [8]. As will be shown in the following, the current buildup time is more than 1 s. Even- tually, the plasma duration will be 300 s. For KSTAR tokamak, preionization was studied earlier using the 0-dimensional (0-D) code (TECHP0D) for many other initial conditions [9]–[11] considering the effects of the error field and impurities. How- ever, KSTAR has seven pairs of superconducting poloidal field coils that are controlled independently, we made a more ad- vanced code to which the additional circuit parameter are added with circuit parameters of self inductances and mutual induc- tances. These parameters were calculated by KBSI [12]. There- fore, in the present set of calculations, the effect of the additional circuit equations is examined. This paper presents subsequent results of the preionization simulation using this advanced code that is quite adequate for describing the physical process in KSTAR tokamak of major radius m and minor radius m. The code is 0-D so that the elongation and triangularity are not relevant parameters. Similarly, the effect of variation of the initial minor radius is not examined. The present version of the code allows the minor radius to grow but the radius is held constant for these studies. The variation of the minor radius would introduce an artificiality into the results that is not justified by the 0-D char- acter of the code. Similarly, whether the energy deposition is at the ECR layer or the upper hybrid layer is not relevant to the re- sults. The complexity of the plasma expanding in minor radius can be the subject of future studies. Previously, it has also been found [2] that the best results are obtained with a parallel index of refraction, of 0.5 and 100% -mode polarization. This implies that the radiation incident on the plasma either from the high-field side or from the low-field side is transformed by a polarization rotator on the inside wall and reflected back to the plasma as -mode. II. PREIONIZATION MODEL The model of the preionization was first applied to the ISX-B tokamak for early preionization experiments [2], [13]. As we already mentioned, more advanced models have appeared since that time [5]–[7]. The model in use here includes the circuit equations of seven pairs of controlled PF coils and the effects of hydrogen ions and impurities. Except for the impurity effects 0093-3813/03$17.00 © 2003 IEEE