Search

A pellet injection system operating at 20 Hz has been operated in KSTAR (Korea Superconducting Tokamak Advanced Research) since 2016. This system can inject pellets with various size, velocity, and frequency to the KSTAR vacuum vessel during plasma experiments. Various experiments such as peak density control, enhancement of confinement by increasing core beta, and NTM stabilization were carried out with the pellet injection system in 2018. Most of all, the characteristics of the pellet should be studied to better understand the interaction with plasma. The pellet trajectory is one of the interesting topics in KSTAR and therefore a related investigation was carried out outside of the tokamak first. We introduce the preliminary results of the trajectory test, the pellet velocity tendency, and a plasma experiment in this paper.

20 Hz pellet injection system was installed for improving a fueling system of KSTAR (Korea Superconducting Tokamak Advanced Research) in 2016. Several things of hardware such as a liquid helium transfer line, an extruder nozzle and an extension part of pellet guide tube are changed for 2017 campaign. For plasma experiments, the pellet injection system uses the smallest pellet (1.5 mm of length, 2 mm of diameter). The pellet velocity can be variable from 150 m/s to over 200 m/s by controlling of propellant gas pressure. The clear increase of plasma density observed during KSTAR experiments is also presented in this paper.

Since the first plasma was achieved in 2008, the quench detection system for the Korea Superconducting Advanced Tokamak Research (KSTAR) coils has been successfully operated. Simple bridge circuits and cowound voltage sensors were used for the first quench detectors. In order to enhance the reliability and stability of the quench detection especially for the pulse coils such as the KSTAR CS coil, a couple of schemes to actively reject inductive voltages have been added and tested. They have focused on rejecting as much inductive voltages as possible and increasing the noise rejection ratio (NRR) because the inductive voltage noise is a main factor to disturb the stable quench detection. One scheme is to use the voltage difference between two balanced bridge voltages or two voltages of cowound voltage sensors, where each voltage is the voltage across each of two coils having the same electrical currents and structurally symmetry. The other is to use the voltage difference between the measured voltage and the calculated voltage. With the additional schemes, inductive voltages were reduced so that the NRR could substantially increase up to 20 000. In this paper, the upgraded quench detection system was described for the KSTAR PF1 coil and the performance results were presented together with experimental data in the 2016 KSTAR campaign.

The primary quench detection system of the Korea Superconducting Tokamak Advanced Research (KSTAR) device has been operated in technical and plasma experiments for eight years. Cowound-type quench detectors have exhibited distinct performance of compensation for both self-induction and mutual induction; on the other hand, some cowound strips tended to break in the long-term operation. In contrast, Wheatstone bridge (WB)-type quench detectors had no wire breakage, and their signal wires are still accessible and repairable. Therefore, WB-type quench detectors are very important to be a backup method of quench detection for the coils. New WBs using a half-bridge at a fixed ratio of resistances of 1:1 were tested in the KSTAR operation. Upper-lower comparison of the new WBs performed good compensation for induced voltage. The new WBs clearly demonstrated that the bridge circuits may become very simple, and in situ calibration may be unnecessary if the arrangement of voltage taps and quench detectors is optimized with respect to the magnetic field of the tokamak device.