Search

The central solenoid (CS) coil system of the KSTAR device consists of four pairs of coils with up-down symmetry and a set of support structure. The CS coil system should endure various loads including thermal and electromagnetic load during operation. One of the most important roles of support structure is to preserve the structural stability of coil system from vertical forces by applying the pre-compression on the coil stack. The main assembly procedure of the coil stack is as follows: 1) coil subassembly and stacking, 2) structure assembly, 3) preloading process by heating shells in the support structure, and 4) measurement on the final dimension. The special jigs and heating system for the preloading were designed to apply pre-compression on the coil stack during the assembly at room temperature. Moreover, a few tens of thermocouples and strain gauges were installed for measuring temperature and thermal expansion of preloading structures. Since the final assembly of the CS coil was successfully completed with about 7.3 MN of preloading in 2006, the CS coil system has been experienced four thermal cycles from 300 K to 5 K until 2011. This paper describes the main results of applying the pre-compression on the coil stack and mechanical behaviors of preloading structures during thermal cycles. [ABSTRACT FROM PUBLISHER]

The Cable-In-Conduit Conductor (CICC) for the KSTAR buslines is made of NbTi superconducting (SC) strands. A busline consists of several electrical joints, which are the major heat load contributors to the busline cryo-system. In the poloidal field (PF) busline helium circuit, the supercritical helium is fed to the electrical joint of current lead end and comes out to the magnet terminal joint. This helium flow configuration has been verified to maintain the cryogenic stability of the buslines through the KSTAR operation. During the normal operation of the KSTAR PF coil, the heated helium coming out to both the coil and the busline meets at the magnet terminal and exchange heat, but the busline outlet temperature still remained less than magnet outlet temperature. As the buslines for the electrical connection in series of the upper and lower coils for PF1 and PF2 have the helium path through the two terminal joints of magnet, they experience higher temperature than the other buslines mainly due to the larger heat exchange. In this case, the connection buslines are considered to have very low safety margin and have the strong possibility of quenches. [ABSTRACT FROM PUBLISHER]

Superconducting magnets of the Korea Superconducting Tokamak Advanced Research (KSTAR) are cooled by supercritical helium with 4.5 K, which was supplied and recovered by the 9 kW of the Helium Refrigerator System (HRS). While current is being charged, the supercritical helium expands to both side of the helium inlet and the outlet of the magnets due to the generated AC losses. To maintain the pressure gradient, both the supply and the return pressures of the HRS are increased at the same time and the differential pressure of the HRS was reduced after the event. However, the pressure rising in the magnets may block the helium flow or create reversal flow of the helium. During unipolar experiment of PF1 magnet up to 2 kA with 1 kA/s of ramp-up rate, the mass flow rate was decreased at the PF1 cooling tube (manifold) in the helium distribution system (HDS), whereas the pressure is increased and the temperature is to be increased or decreased according to compression and expansion of the heated helium in the magnets. For the bipolar experiment of PF1 up to \pm2 kA with 1 kA/s ramp-up rate and 2 kA/s ramp-down rate, the conditions in the helium flow were drastically changed, especially the mass flow rate was measured to be maintained at zero for a few second (more than 4 s). This behavior could decisively affect the cryogenic stabilities in the magnet, and may impose a major limit on the long pulse operation of KSTAR. In this paper, we investigated this behavior and analysed by using 1-dimentional thermo-hydraulic code, GANDALF. [ABSTRACT FROM AUTHOR]

The AC loss in a large superconducting magnet coil shows a tendency to be changed after assembly. For the ac loss measurement of the Korea Superconducting Tokamak Advanced Research (KSTAR) superconducting coils after assembly, several current scenarios, trapezoidal pulses and a DC offset sinusoidal pulses, were applied to the PF1 Upper (U) and Lower (L) coils during the commissioning. The measurement was done once before and once after the plasma discharge experiments. All coil data were obtained by the tokamak monitoring system and helium distribution system which were designed to measure temperature, pressure, and mass-flow at both inlets and outlets of the coils. The PF1 coils of the cable-in-conduit conductor type were made of Nb3Sn superconducting strands, whose winding scheme is 20 layers with each layer having 9 turns. Each has 10 cooling channels for the heat removal by the supercritical helium at 4.5 K. For the trapezoidal pulse tests, the current was increased up to 4 kA with several different ramp rates and a 2 kA DC offset and 0.5 kA sine waves with different frequencies from 0.1 Hz to 0.3 Hz were applied to the coil. According to the analyses, the AC loss was slightly decreased for the same condition after 700 plasma shots. It was believe that such a result was due to reduced inter-strand resistances which changed the transverse resistance between the inter-strands. The coupling time constant was estimated to be 32 ms in the trapezoidal tests and 13.6 ms for the DC offset sinusoidal pulses. The former is larger than the latter because of the effect of the jacket eddy current loss due to Incoloy 908 which is a ferromagnetic material. Considering the jacket eddy current losses, the coupling time constant was recalculated and the value estimated to be about 13 ms for all current wave forms during first commissioning. [ABSTRACT FROM AUTHOR]

The KSTAR current lead system (CLS) was commissioned during the first campaign of the plasma experiments. The current lead system is divided into a toroidal field (TF) system and a poloidal field (PF) system. Two pairs of current leads for the TF system and 7 pairs for the PF system were installed. Instead of the copper leads, vapor-cooled type brass current leads (CLs) were mounted in order to minimize the cryogenic load. The CLs were installed after the verification of the components through helium leak tests, low current charging tests at room temperature, and high voltage tests. The stability of the CLS was inspected through LN2 cool down tests, helium leak tests, flow control tests, and temperature control tests using a ceramic heater after the installation all of their components. During the first campaign, it was cool down until liquid helium temperature after vacuum pumping and the helium flow rates of the CLs were controlled successfully at the optimum values. In case of the TF CLs were 0.20 g/s and 0.26 g/s at zero current and 15 kA respectively. On the other hands the PF CLs kept 0.2 g/s during PF coil operation at 4 kA and zero current. And a ceramic heater was reliably operated to prevent water condensation at the warm terminal of the CL. [ABSTRACT FROM AUTHOR]

The KSTAR superconducting magnetic coils, which are made of cable in-conduit conductor (CICC), maintain a superconducting state with forced-flow supercritical helium (4.5 K, 5.5 bar). During current changing of the superconducting magnetic coils, AC losses are generated in the CICC due to dI/dt, and the heat generated from the loss is removed by high heat capacity supercritical helium. At the same time, reversed flow of the helium occurs due to a rapid increase of the helium temperature and momentary changing of the pressure inside the CICC. This phenomenon has been detected in all of the poloidal field (PF) coils, especially in the upper (U) and lower (L) PF1 ~ PF4 coils. The maximum change of the magnetic field in the PF1UL ~ PF4UL coils is located near the inlet and outlet of the helium cooling channels, and that of the PF5UL ~ 7UL coils is located at the center of the cooling channel. The temperature variation at the helium inlet was always measured to have a time delay after each shot. In the PF1 coil tests, it was measured to have a delay of 26 sec. During the first plasma campaign, this phenomenon was more severe in the case of all PF coils operating together than for a single PF operation. In this paper, we investigated the thermal-hydraulics of this phenomenon. [ABSTRACT FROM AUTHOR]