IEEE TRANSACTIONS ON APPLIED SUPERCONDUCTIVITY, VOL. 22, NO. 3, JUNE 2012 4702404 Investigations About Quench Detection in the ITER TF Coil System M. Coatanea, J.-L. Duchateau, S. Nicollet, B. Lacroix, and F. Topin Abstract—Due to the large stored energy (40 GJ) and the small allowed number of fast discharges (50), severe requirements have been put in the ITER project for the Toroidal Field (TF) coil system in comparison with the Poloidal Field (PF) and Central Solenoid (CS) systems, aiming at avoiding any fast discharge not related to a quench. A very important point, which has to be examined, is whether a quench detection based on a co-wound tape, recommended in the ITER project, and located inside the conductor insulation is compulsory to ensure the inductive voltage compensation. Another possible solution is the balance of the voltage of TF coils or TF coils subcomponents. The sensitivity of the quench detection systems is examined ac- cording to the different types of flux variations experienced in the machine. During plasma discharge, the sensitivity to poloidal flux variations and plasma paramagnetism is highlighted, the last effect being illustrated for Tore Supra. It is eventually recommended to select the quench detection by balancing coils or coils subcomponents. This solution has been adopted for Tore Supra, KSTAR and in JT-60SA project. The level of refinement should be adjusted during commissioning, as a function of the required voltage detection level (0.4 V) and to the required holding time (1 s). Index Terms—ITER, pancake, quench detection, voltage detection. I. INTRODUCTION A T variance with the PF and the CS systems [1]–[3] the TF coil system of a tokamak operates in steady state system. Except during the ramping phase of the current, which should occur roughly weekly (1000 TF charging cycles in the DDD), the TF current is kept nearly constant during normal operation. Inductive voltages should therefore be less important across the TF coils than across the PF and the CS coils and thus the quench easier to be discriminated. Each fast discharge of the TF causes a significant aging of the components due to the associated thermal ramp up and down. Re-cooling of the TF coil system and related components requires significant time and effort. Consequently, the total Manuscript received September 09, 2011; accepted October 30, 2011. Date of publication December 09, 2011; date of current version May 24, 2012. This work was supported by ITER Organization in the framework of contract ITER/CT/09/4300000014. The views and opinions expressed herein do not necessarily reflect those of the ITER Organization. M. Coatanea, J.-L. Duchateau, S. Nicollet, and B. Lacroix are with CEA, IRFM, F-13108 St-Paul-lez-Durance Cedex, France (e-mail: marc.coatanea@cea.fr). F. Topin is with IUSTI, Marseille 13453 Cedex 13, France (e-mail: Frederic. Topin@polytech.univ-mrs.fr). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TASC.2011.2178979 Fig. 1. Simplified electrical scheme for the ITER TF coil system. amount of allowed fast discharges should not exceed 50, and any fast discharge not related to a quench has to be avoided. Compared to the CS and PF coils, a false quench detection in the TF coil system has heavier consequences, hence a special care about quench identification must be brought. A key point which has to be decided is whether or not a quench detection based on a co-wound tape is compulsory to ensure the inductive voltage compensation (EAST system) or whether the balance of the voltage of two TF coils is sufficient (Tore Supra (TS), KSTAR). The major aspects of this work are presented in this paper. The inductive voltage cannot be estimated as easily as for the PF and the CS coil systems. For them, it is possible to calculate the voltage across the different components during for instance a plasma discharge [4]. The voltage is more difficult to predict for the ITER TF coil system. II. REACTION OF TF COILS TO INDUCTIVE FLUX VARIATIONS A very simplified scheme of the ITER TF coil system elec- trical circuit is presented in Fig. 1 (18 TF coils in series). If we call the magnetic flux which is collected by a surface where the contour is given by the center line of one petal of a TF coil, (1) stands for the 18 coils (noted k), subject to the external flux variations , L being the equivalent inductance of each of the 18 coils supposed identical and the mutual inductance of coils k and . All the currents are identical as the coils are in series. (1) (2) average flux across the coils: . 1051-8223/$26.00 © 2011 IEEE