Fault Analysis of ITER Coil Power Supply System

Similar documents
3.4 Poloidal Field Power Supply Systems for the EAST Steady State Superconducting Tokamak

A Design Study of Stable Coil Current Control Method for Back-to-Back Thyristor Converter in JT-60SA

MATEFU Insulation co-ordination and high voltage testing of fusion magnets

ELECTRICAL POWER TRANSMISSION TRAINER

Insulation Co-ordination For HVDC Station

EVALUATION OF DIFFERENT SOLUTIONS OF FAULTED PHASE EARTHING TECHNIQUE FOR AN EARTH FAULT CURRENT LIMITATION

AORC Technical meeting 2014

Consorzio RFX. Report RdS/2013/283

Tab 2 Voltage Stresses Switching Transients

Design and Implementation of Four 20 ka, 5 kv Hybrid Switching Networks for Plasma Ignition in the International Tokamak JT-60SA

Error Fields Expected in ITER and their Correction

CONTENTS. 1. Introduction Generating Stations 9 40

Design and R&D for an ECRH Power Supply and Power Modulation System on JET

2.3 PF System. WU Weiyue PF5 PF PF1

Magnets Y.C. Saxena Institute for Plasma Research. 1/16/2007 IPR Peer Review Jan

A new hybrid protection system for high-field superconducting magnets

Effects of Phase-Shifting Transformers, and Synchronous Condensers on Breaker Transient Recovery Voltages

Conventional Paper-II-2011 Part-1A

Simulations of W7-X magnet system fault scenarios involving short circuits

Impulse testing of coils and magnets: present experience and future plans

DYNAMICS OF NONLINEAR PLASMA-CIRCUIT INTERACTION *

Abstract. PEGASUS Toroidal Experiment University of Wisconsin-Madison

ICRF Physics in KSTAR Steady State

DESIGN OF INTEGRATOR SYSTEM FOR PLASMA REACTOR USING CADENCE

Enhancement of Fault Current and Overvoltage by Active Type superconducting fault current limiter (SFCL) in Renewable Distributed Generation (DG)

SINGLE-STAGE HIGH-POWER-FACTOR SELF-OSCILLATING ELECTRONIC BALLAST FOR FLUORESCENT LAMPS WITH SOFT START

OBICON. Perfect Harmony. Short overview. ROBICON Perfect Harmony. System Overview. The Topology. The System. ProToPS. Motors.

Simulations of open phase conditions on the high voltage side of YNd05-power plant transformers

POWER- SWITCHING CONVERTERS Medium and High Power

ECE 422/522 Power System Operations & Planning/Power Systems Analysis II 5 - Reactive Power and Voltage Control

A Pyrotechnic Fault Current Limiter Model for Transient Calculations in Industrial Power Systems

A Highly Versatile Laboratory Setup for Teaching Basics of Power Electronics in Industry Related Form

DC current interruption tests with HV mechanical DC circuit breaker

FGJTCFWP"KPUVKVWVG"QH"VGEJPQNQI[" FGRCTVOGPV"QH"GNGEVTKECN"GPIKPGGTKPI" VGG"246"JKIJ"XQNVCIG"GPIKPGGTKPI

Curso de Transmissão em Corrente Continua Rio de Janeiro, de Junho, 2007

EUROPEAN ORGANIZATION FOR NUCLEAR RESEARCH European Laboratory for Particle Physics

Validation of a Power Transformer Model for Ferroresonance with System Tests on a 400 kv Circuit

Over-voltage Trigger Device for Marx Generators

Study of Plasma Equilibrium during the AC Current Reversal Phase on the STOR-M Tokamak

H. Y. Lee, J. W. Lee, J. G. Jo, J. Y. Park, S. C. Kim, J. I. Wang, J. Y. Jang, S. H. Kim, Y. S. Na, Y. S. Hwang

COOPERATIVE PATENT CLASSIFICATION

Voltage Source Converter Modelling

The Fault Level Reduction in Distribution System Using an Active Type SFCL

25 MW SMES-BASED LONG-PULSE KLYSTRON MODULATOR

Lecture 4 ECEN 4517/5517

Stability Enhancement for Transmission Lines using Static Synchronous Series Compensator

Course No: 1 13 (3 Days) FAULT CURRENT CALCULATION & RELAY SETTING & RELAY CO-ORDINATION. Course Content

Research Article Survey of Induced Voltage and Current Phenomena in GIS Substation

ABB Power Systems AB Sweden

Design, Control and Application of Modular Multilevel Converters for HVDC Transmission Systems by Kamran Sharifabadi, Lennart Harnefors, Hans-Peter

Chapter -3 ANALYSIS OF HVDC SYSTEM MODEL. Basically the HVDC transmission consists in the basic case of two

Design, Control and Application of Modular Multilevel Converters for HVDC Transmission Systems by Kamran Sharifabadi, Lennart Harnefors, Hans-Peter

An Improvement in the Virtually Isolated Transformerless Off - Line Power Supply

Bus protection with a differential relay. When there is no fault, the algebraic sum of circuit currents is zero

COMPARATIVE PERFORMANCE OF SMART WIRES SMARTVALVE WITH EHV SERIES CAPACITOR: IMPLICATIONS FOR SUB-SYNCHRONOUS RESONANCE (SSR)

Design and Simulation of superconducting fault current limiter

Enhancement of Voltage Stability & reactive Power Control of Distribution System Using Facts Devices

I -limiter The world s fastest switching device

Analysis of Transient Recovery Voltage in Transmission Lines Compsensated with Tpcs-tcsc Considering Accurate Model of Transformer & Generator

10. High-Boost HAM. Design Guide & Applications Manual. Maxi, Mini, Micro Family DC-DC Converters and Configurable Power Supplies

NEO TELE-TRONIX PVT. LTD. 6/7 Bijoygarh, Kolkata , Tel : ; Fax :

Assessment of Saturable Reactor Replacement Options

High-Voltage Switch Using Series-Connected IGBTs With Simple Auxiliary Circuit

International Journal of Engineering Technology and Scientific Innovation

HIGH VOLTAGE ENGINEERING(FEEE6402) LECTURER-24

Calculation of Transient Overvoltages by using EMTP software in a 2-Phase 132KV GIS

Reducing the Fault Current and Overvoltage in a Distribution System with an Active Type SFCL Employed PV System

P. O. BOX 269 HIGHLAND, ILLINOIS, U.S.A PHONE FAX

3.10 Lower Hybrid Current Drive (LHCD) System

Status of the KSTAR Superconducting Magnet System Development

HVDC Transmission. Michael Muhr. Institute of High Voltage Engineering and System Performance Graz University of Technology Austria P A S S I O N

Residual Current Operated Circuit-Breakers (RCCBs)

Use of inductive heating for superconducting magnet protection*

Numbering System for Protective Devices, Control and Indication Devices for Power Systems

3.7 Grounding Design for EAST Superconducting Tokamak

G. KOEPPL Koeppl Power Experts Switzerland

Transformer Factory Testing

Faster, Hotter MHD-Driven Jets Using RF Pre-Ionization

Analysis of Short Circuit Fault for 4.6GHz/6MW LHCD High Voltage Power Supply

Error vs. Uncertainty Historical Perspective

Simulation of HTS saturable core-type FCLs for MV distribution systems

Earth Fault Protection

Shunt Reactor Switching

2. Composing and characteristics of EAST

IJREE - International Journal of Research in Electrical Engineering ISSN:

An Introduction to the CSCT as a New Device to Compensate Reactive Power in Electrical Networks

Fault Current Limiter Selection Considerations for Utility Engineers

INCREASING THE POWER SYSTEM SWITCH GEAR CAPACITY BY USING SUPERCONDUCTING FAULT CURRENT LIMITER

I. INTRODUCTION IJSRST Volume 3 Issue 2 Print ISSN: Online ISSN: X

High voltage engineering

Power Quality and Reliablity Centre

Grid Impact of Neutral Blocking for GIC Protection:

The Results of the KSTAR Superconducting Coil Test

Electrical Power Systems

Busbars and lines are important elements

HVDC High Voltage Direct Current

Design of an 80kV, 40A Resonant SMPS for Pulsed Power Applications

A Simulation of the COMPASS Equilibrium Field Power Supply PID Controller

1% Switchgear and Substations

DC Link. Charge Controller/ DC-DC Converter. Gate Driver. Battery Cells. System Controller

Transcription:

Fault Analysis of ITER Coil Power Supply System INHO SONG*, JEFF THOMSEN, FRANCESCO MILANI, JUN TAO, IVONE BENFATTO ITER Organization CS 90 046, 13067 St. Paul Lez Durance Cedex France *Inho.song@iter.org Abstract: - The verification of the high voltage design of the magnet system is an important issue for a reliable operation of the ITER machine. Transient peak voltages occur on the TF, CS and PF coils during the normal operation including SNU operation, in case of a fast discharge or if a failure appears in the components of the coil power supply circuit. The maximum voltages across a coil (terminal to terminal) or terminal to ground depends on the circuit configuration, on the and on the number and type of fault. The maximum coil voltage to ground strongly varies with the number of simultaneous failures and the stray parameters of the system contribute mainly on the transient peak voltages. For proper insulation co-ordination it is important to determine the peak transient and steady-state voltages applied to the coils. The maximum calculated voltages of all relevant cases will produce necessary basis for testing parameters for design and selection of insulation. All of coil power supply systems are modeled and stray parameters of key components are introduced for calculating the peak transient and steady-state voltage induced to the coils considering the normal and abnormal. Key-Words: - Superconductors, Coil power supply, Fault voltages, SNU, FDU, DC busbars, Grounding circuit 1 Introduction The magnet system for ITER consists of 18 Toroidal Field (TF) coils, a Central Solenoid (CS, six independent winding pack modules), six Poloidal Field (PF) coils and 18 Correction Coils (CCs), Fig.1 All coils are designed using superconductors with high current carrying capability [1]. The Toroidal Field (TF) coils, which provide the Toroidal field for confining charged particles in the plasma, operate in a steady-state mode with a current of 68 ka and discharge the stored energy in case of quench with using 9 interleaved Fast Discharge Units (FDUs). The Central Solenoid (CS) coils and Poloidal Field (PF) coils, which provide the change in Poloidal flux needed to initiate the plasma, generate the plasma current and maintain this current, operate in a pulse mode with currents of up to 55 ka and require fast variation of currents inducing more than 10 kv during normal operation on the coil terminals using Switching Network (SN) systems (all CSs, PF1 and PF6) and VS converter (8.1 kv at no-load) for PF2 to PF5, which are series and parallel connected to PF converters. SN and FDU systems comprise high current DC circuit breakers and resistors for generating high voltage for plasma initiation (SN) and dissipating magnetic energy in case of quench (FDUs). High transient voltages can arise due to the switching operation of SN and FD and the characteristics of resistors and stray components of DC distribution systems. Also, control faults and shorts or grounding faults can produce higher voltages between terminals and terminal to ground. Therefore, the design of the coil insulation, coil terminal regions, feeders, feedthroughs, pipe breaks, DC busbars, switches and instrumentation must take account of these high voltages during normal and abnormal. Fig.1. The ITER Coils. This paper describes the fault analysis of the TF, CS and PF coil power supply systems, taking into account of the stray parameter of the power supply and switching systems and inductively coupled super-conducting coil models. Resistor grounding systems are included in the simulation model and all fault of AC/DC converter, SNU, FDU, DC short circuits and single grounding are simulated. The occurrence of two successive faults is considered for the TF coil power supply and CS/PF coil power supply systems also take into account of double fault. The analysis results are discussed for transient and steady-state during normal and abnormal operations. Voltage insulation level can be defined and it is necessary to test the coils at ISSN: 1792-5088 153 ISBN: 978-960-474-233-2

higher voltages, to be sure of reliable performance during the lifetime of operation. This paper presents the proposed design solution of ITER AC/DC converters currently undergoing the ITER acceptance process. 2 Coil Power Supply System (CPSS) Function of the Coil Power Supply (CPS) is to provide controlled DC current and voltage to each coil for plasma confinement, plasma initiation and shape control. The CPSS will include nine systems to supply the coils: one for the 18 series connected TF coils; one for the CS1 upper and lower modules connected in series; four for the CS2 upper, CS2 lower, CS3 upper and CS3 lower modules; two for individual supply of the PF1 and PF6; one common system for the four outer PF coils (PF2, 3, 4, and 5) including VS converter In addition, nine relatively small PS systems with very similar configurations will supply the error field CCs allowing correction of error field harmonics due to position errors as well as from busbars and feeders. Fig.2 shows the configuration of CPS. to absorb their stored energy in case of other abnormal events which could potentially damage the coils. Protective Make Switch (PMS) bypasses the AC/DC converters and separates the magnets from the power sources. The whole CPSS will be interconnected between the AC/DC converters, PMS, SNU, FDU and magnets by a huge and complex DC busbar system. All TF coils are supplied with power by one thyristor converter (TF PS): a 12-pulse, 2-quadrant converter rated for 68 ka, 900 V no-load voltage. The converter is designed as two thyristor 6-pulse bridges connected in parallel via Interphase reactors. In case of a failure in the thyristor converter, in can be bridged by an external thyristor bypass and PMS. The coil fast discharge is provided by the 9 FDUs, which are interleaved with the pairs of the series-connected coils in order to limit the voltage between the coil terminals and to earth. TF AC/DC Converter EPMS2 EPMS1 TF-PMS Main Switch FDU4 FDU5 TF9 Main Switch PB FDU3 Feeder FDU6 FDU7 FDU8 TF10 TF11 TF12 TF13 TF14 TF15 TF16 Cryostat TF8 TF7 TF6 TF5 TF4 TF3 TF2 TF1 FDU2 FDU1 TF17 TF18 FDU9 Grounding R Fig.3. A schematic of the TF coil power supply. The TF system is earthed through a set of identical resistors associated with the 9 coil groups. The mid points of the resistors connected in parallel to each group are connected to the TF neutral. Due to this arrangement, the potentials of the two terminals of each group are balanced regarding the neutral/earth potential and, therefore, each terminal voltage to earth is reduced ½ of the voltage across the terminals. The maxim voltage across current leads at fast discharge is about 8 kv. Fig.2. Configuration of Coil Power Supplies. The loop voltage required for breakdown and plasma initiation is obtained in the PS systems for CS modules and the coils PF1 and PF6, by AC/DC converters and switching resistors into the circuits in series with the coils causing a very large amount of power (about 2 GW) to be extracted. These circuits, called switching networks, are made up of circuit breaker, thyristor switches, make switches and resistors and in the PF2-5 coils PS system (power to be extracted is less than 100 MW), by adding more thyristor AC/DC converter in series. Protection of the superconducting magnets in case of quench is necessary. Resistors, normally bridged by circuit breakers, are included in series with the coils Fig.4. A schematic of the CS1U&L coil power supply. ISSN: 1792-5088 154 ISBN: 978-960-474-233-2

The common feature of CSs, PF1 and PF6 is that all of them use AC/DC converters (2.7 kv no-load, 45 ka for CSs and 55 ka for PF1, PF6) and SNU (8.5 kv except for CS1). With the exception of the CS1 PSS all other systems supply individually on CS module (upper and lower) or PF coils. The two CS1 modules, upper and lower, are connected in series and are interleaved with the two SNUs (6 kv each) and two FDUs. CS1U&L coil PSS is shown in Fig.4. A relatively high voltage is needed for fast control of currents in the coils PF2 ~ PF5 as it is show schematically in Fig.5. The quasi symmetrical distribution of currents in these coils made it possible to use one fast response thyristor converter (Vertical Stabilize (VS) PS), which is connected to the coils PF2 ~ PF5 as it is shown in Fig.5. Each PF PS has 4.05 kv no-load and 55 ka current rating. A soft earthing via high impedance resistors, as described for the TF system, is foreseen. Fig.5. A schematic of the PF2 ~ PF5 coil power supply. Each FDU comprises two DC circuit breakers connected in series; first, the so-called Main Switch will open when the external trigger signal is received, second is the Pyrobreaker (PB) used as a backup if the Main Switch fails to open. The Main Switch can interrupt 70 ka DC current with the help of counterpulse circuit and consists of Bypass switch (BPS, mechanical switch) and VCB. Snubber circuit is used for limiting the overvoltages due to the stray parameters and switching. SNU will generate a high voltage (maximum 8.5 kv) required for breakdown at the beginning of each plasma operation. It consists of two resistors, Fast Opening Switch (FOS), Fast Disconnect Switch (FDS), thyristor switch and counterpulse circuit. Fig.6. Section view of TF DC busbar. DC busbars will connect the TF, CS, PF and CCs to the AC/DC converters, FDUs and SNUs. Aluminium will be used for all the busbars and it will be cooled by water. As for the example, drawing of TF DC busbar cross-section is shown in Fig.6 and earthed cases and separators between positive and negative poles will be provided to protect the busbars from damage and to decrease the probability of a pole-to-pole short circuit. The reference design of the grounding scheme is to provide for all the coils and power supply components a soft grounding via high impedance resistors. In case of a ground fault, the terminal voltage to ground increases by a factor two, up to the voltage across the terminals. The leakage current will be measured and used for ground fault protection. The grounding resistor is built with three resistors in series to reduce the risk of a short circuit between resistor terminals. Two of them, with the total resistance 1 kω, are always included in the grounding circuit. The third resistor (~10 kω) is normally bridged by a disconnector. When a ground fault is detected by the grounding fault detector, the disconnector opens and inserts the third resistor in the grounding circuit, further limiting the fault current. The factor of two of terminal to ground voltage increase in case of ground fault does not take into account the stray capacitance to ground. The ripple in the AC/DC converter and the transient voltage of SNU and FDU and the effect of stray parameters of the DC busbars may cause higher voltages to ground. A transient analysis is needed to being carried out. 3 Fault Analysis The Coil Power Supply (CPS) systems are designed to limit the high voltages for protecting the magnets, but under the certain and faults, voltages can rise above 22 kv in certain coils due to the stray parameters of system and switching operation. Therefore, the magnet must be designed to have certain insulation level and the high voltage tests (insulation test) of coils and CPS systems have to be conducted at higher voltages. For proper insulation of coils it is important to determine the peak voltages applied to the coils. For simulating the peak voltages, the CPS is modeled and all fault events are sorted according to probability of these events on the base of assumptions without the probabilistic analysis. Also, the redundancy on the electrical components is not taken into account when defining the single or multi failure cases (BPS and Pyrobreaker are considered as one switch). 3.1 Simulation model The PSIM software tool is used for the analysis and the models of transformer and thyristor converter in PSIM ISSN: 1792-5088 155 ISBN: 978-960-474-233-2

are used. Detailed models of SNU and FDU were built including all the stray parameters such as stray inductances and capacitances. Fig.7 shows the equivalent circuit for the calculation of the transient characteristics of FDS. BPS, VCB, counterpulse circuit, snubber circuit and resistors including cables are modeled with stray parameters. L, mkh/m 0,44 0,42 0,4 0,38 0,36 0,34 0,32 0,3 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 Fig.9. Dependence of inductance of DC busbar on the frequency f, Hz Fig.7. The equivalent circuit for the calculation of the transient characteristics of FDU. The SNU systems consists of two mechanical switch (FOS and FDS), two stage thyristor switch (TS and TH1), counterpulse circuit (TH2), Snubber circuit and resistors. The equivalent circuit is shown in Fig.9. Fig.8. The equivalent circuit for the calculation of the transient characteristics of SNU For the DC busbars, calculated parameters per meter from the design are compared with the measured values and the length of DC busbars included and the single π- scheme for DC busbars modeling applied. The calculated and measured parameters of CS/PF DC busbars are in Table1. For the inductance (µh/m), Fig.10 shows the characteristic. Table 1. Resistance and capacitance value Parameter Calculated Measured Resistance 1.13 µω/m 1.35 µω/m Capacitance 2.83 nf/m 1.90 nf/m Inductance 1.00 µh/m Fig.10 3.2 Normal operations and fault The maximum voltages introducing on coil terminals and coil terminal to ground in normal are simulated and operation are as follows; 1) FDU opening, 2) SNU opening, 3) jitter between switches in TF CPSS, 4) converter maximum output voltage. Several fault and combinations were considered to determine the maximum voltage inducing in fault. Considered fault are as follows; 1) FDU opening failure for TF FDUs, 2) converter control fault or bypass fault, 3) SN failure, which means that SN keeps opening when FDU opens and 4) single grounding fault. Transient and steady-state over peak voltages are measured and double faults were simulated as worst cases. Events with more than two independent faults were not considered. 3.3 Fault analysis of TF coil power supply The schematic diagram and grounding scheme for TF PSS is given in Fig.3. All the TF coil terminals are connected by resistors to the neutral busbar and the neutral busbar is connected to the Machine Ground by the grounding resistor. Table 2. Peak voltage of TF PSS Normal Abnormal Cases Maximal voltage to ground in transient Maximal voltage ground in steady-state FDU opening 5.2 kv 3.8 kv Jitter 7.3 kv 3.8 kv FDU opening failure FDU opening + FDU opening failure + Two FDU opening failure + 7.5 kv 7.0 kv 7.6 kv 7.6 kv 12.3 kv 14 kv 14.3 kv 16 kv ISSN: 1792-5088 156 ISBN: 978-960-474-233-2

Protection of nuclear safety boundaries and prevention of coil damage are both important concerns. The most complicated exist in TF PSS due to high amount of elements and huge stored energy. A fault scenario can develop in different sequence: fast discharge, shorting to ground (single ground fault), fault of one of the FDUs and jitters between FDUs. The first step is to simulate the normal operation including FDUs operation with and without jitters. All the fault cases listed above are simulated and the result is given in Table 2. Maximum fault voltage occurred when there is a ground fault and a FDU opening failure and is 14 kv. Peak value is reached in a steady-state and does not depend on stray parameters. Moreover, it does not depend on grounding resistance values. It seems to be acceptable with regard to the insulation of TF coils and the TF coil protection (12.3 s discharge time constant instead of 11 s). 3.4 Fault analysis of CS1U&L coil power supply Fig.4 shows the coil circuit with interleaved SNU + FDU and this scheme has the reduced terminal to ground voltage compare with the one FDU + SNU scheme. SNU and FDU could not be operated simultaneously in normal operation but it could be happen as fault case. Single ground fault and converter control fault are also simulated. Table 3 shows the peak voltage values in normal and abnormal. The highest peak voltage is shown in the fault condition that SNU and FDU open simultaneously and ground fault happen. It shows that a transient maximal peak voltage of about 20 kv and steady-state peak voltage of 13.4 kv. These transient peak values depend on the stray parameter and ground circuit. Triple fault condition shows highest peak voltages but it is not considered. Table 3. Peak fault voltage of CS1U&L PSS Abnormal Two SNUs open + One SNU opening failure + SNU and FDU opening + SNU and FDU opening + Converter set to maximum voltage + Ground fault Maximal voltage to ground in transient Maximal voltage ground in steady-state 9.7 kv 6.23 kv 9.7 kv 6.24 kv 20.0kV 13.4 kv 26.7 kv 17.6 kv 3.5 Fault analysis of PF2 ~ PF5 coil power supply In this schematic, the coils are symmetrically connected in series with their own PF PS and VS PS. Fig.10 shows the schematic of PF2 ~ PF5 power supply system using PSIM simulation software. The best DC grounding point has been set in the schematic of Fig.10. In this circuit, the terminal to ground voltage is around half the maximum coil voltage. Each PF converter comprises of three Main Converters (MCs) and each MC has 1.35 kv no-load and 55 ka current rating. Vertical stabilization power supply has six VS power supply system and the rating of VS is 1.35 kv no-load and 22.5 ka. Fig.10. Detail simulation model of PF2 ~ PF5 circuit Table 4. Peak voltage of PF2 ~ PF5 PSS Abnormal Plasma initiation (PF set to maximum voltage) + FDU opening + Maximal voltage to ground in transient Maximal voltage ground in steady-state 19.9 kv 12.5 kv 15.3 kv 12.0 kv PF set to maximum voltage + FDU open + Ground 22.1 kv 18.3 kv fault PF and VS set to maximum voltage + FDU open + 22.4 kv 18.4 kv Table 4 shows the peak voltage values in normal and abnormal. We can find a transient maximal ISSN: 1792-5088 157 ISBN: 978-960-474-233-2

peak value of about 22.1 kv and 13.4 kv for maximal steady-state when the converter and ground fault occur during the FDU open. The high difference between the values in transient and in steady-state is due to the busbar capacitances and it causes a transient overvoltage of about 58%. Normally when the fast discharge triggered, the control system bypasses the PF and VS converters and then opens the FDU switch. The triple fault condition of PF and VS converter s control fault and ground fault produces the highest peak fault voltages. Fig.11 shows the peak fault voltage waveforms when the maximum negative voltage of the converters and opening of FDU simultaneously with a single ground fault. acceptance tests of the systems must be carried out at higher voltages and high dielectric strength must be included in insulation design to minimize the risk of coil failure and increase the safety and availability. 5 Disclaimer The views and opinions expressed herein do not necessarily reflect those of the ITER Organization. References: [1] N. Mitchell, D. Bessette, R. Gallix, C. Jong, P. Libeyre, C. Sborchia and F. Simon, The ITER Magnet System, IEEE Trans. Applied Supercond., submitted for publication in 2008. [2] P.L. Mondino, T. Bonicelli, V. Kuchinskiy and A. Roshal, ITER R&D: Auxiliary Systems: Coil Power Supply Components, Fusion Eng. Des., Vol.55, p.325, 2001. [3] N. Mohan, T. M. Underland, and W. P. Robbins, Power Electronics: Converters, Applications, and Design, 2nd edition, New York: Wiley. [4] C. Neumeyer, Fast discharge options for tokamak physics experiment Toroidal field & Poloidal field superconducting magnets, 40-9308270, PPPL. Fig.11. The peak fault voltage when the maximum negative voltage of the converters and opening of FDU simultaneously with a single ground fault. Top: PF2~5 oil currents, Middle: PF2~5 termianl_1 voltages, Bottom: PF2~5 terminal_2 voltages. 4 Conclusion High voltages in the ITER coils can occur either due to plasma operation or fault. The highest voltages during plasma operation occur at plasma initiation phase and the highest fault voltages are shown when the hardware fault happens before/after single ground fault. The design voltages to be taken into account are simulated under all fault. The peak voltage is reaching 14 kv in TF coils and 20 kv in CS1U&L coils and 22.1 kv in PF2~5 coils. These results depend strongly on stray parameters, especially the FDU parameters (for example, it has been seen that the cable inductance between FDU and discharge resistor increases the voltage overshoot) and DC busbar stray capacitances. The coil power supply systems are designed to limit the high voltages, but under certain coil terminal to terminal voltage and terminal to ground voltage can increase above 22 kv. Therefore, the ISSN: 1792-5088 158 ISBN: 978-960-474-233-2