California Energy Comission 500-2013-134 DEVELOPMENT OF FAULT CURRENT CONTROLLER TECHNOLOGY Prototyping, Laboratory Testing, and Field Demonstration

Energy Research and Development Division
FINAL PROJECT REPORT
DEVELOPMENT OF FAULT CURRENT
CONTROLLER TECHNOLOGY
Prototyping, Laboratory Testing, and
Field Demonstration
JUNE 2011
CEC-500-2013-134
Prepared for: California Energy Commission
Prepared by: University of California, Irvine

PREPARED BY:
Primary Author(s):
Keyue Smedley
Alexander Abramovitz
University of California, Irvine
Contract Number: UC MR-064
Prepared for:
California Energy Commission
Jamie Patterson
Contract Manager
Fernando Pina
Office Manager
Energy Systems Research Office
Laurie ten Hope
Deputy Director
ENERGY RESEARCH AND DEVELOPMENT DIVISION
Robert P. Oglesby
Executive Director
DISCLAIMER
This report was prepared as the result of work sponsored by the California Energy Commission. It
does not necessarily represent the views of the Energy Commission, its employees or the State of
California. The Energy Commission, the State of California, its employees, contractors and
subcontractors make no warranty, express or implied, and assume no legal liability for the
information in this report; nor does any party represent that the uses of this information will not
infringe upon privately owned rights. This report has not been approved or disapproved by the
California Energy Commission nor has the California Energy Commission passed upon the
accuracy or adequacy of the information in this report.

ACKNOWLEDGEMENTS
The University of California, Irvine gratefully acknowledges the financial support from the
California Energy Commission, and the administrative support of the California Institute for
Energy & Environment (University of California, Berkeley), which made this project possible.
We sincerely thank our collaborators at Southern California Edison Co., Zenergy Power plc, the
Electric Power Research Institute (EPRI) and Silicon Power Corporation for their technical
expertise, teamwork and support which made this project successful. We also express our
sincere appreciation to our knowledgeable and supportive Project Advisory Group for their
valuable advice, which has been of critical importance in every phase of the project.
Contributing Authors:
Franco Moriconi, Francisco De La Rosa, Alonso Rodriguez, Bert Nelson, Amandeep Singh,
Nick Koshnick – Zenergy Power plc
Ram Adapa – Electric Power Research Institute
Christopher R. Clarke, Ed Kamiab, Syed Ahmed – Southern California Edison Co.
Mahesh Gandhi, Swapna Bhat, Simon Bird – Silicon Power Corporation
Project Advisory Group:
Jamie Patterson – California Energy Commission
Lloyd Cibulka – California Institute for Energy and Environment
Bert Nelson, Alonso Rodriguez – Zenergy Power plc
Ram Adapa – Electric Power Research Institute
Alfonso Orozco, Katie Speirs – San Diego Gas & Electric Co.
Bob Yinger, Christopher R. Clarke, Russ Neal, Anthony Johnson, Raj Vora,
Ed Kamiab – Southern California Edison Co.
Ron Sharp – Pacific Gas & Electric Co.
Ed Muljadi – National Renewable Energy Laboratory
Jim McHan – California Independent System Operator
Ken Edwards – Bonneville Power Administration
Ron Meyer – Santa Margarita Water District
UC–Irvine Technical Contributors:
Jun Wen
Franco Maddaleno
Liang Zhou
In Wha Jeong
Marco Tedde
Sadigh Khoshkbar Arash
Chaitanya Vartak
Wensheng Song
Pengju Sun
ZhengSheng Wu
Wenchao Xi
Zhuo Zhao
Fei Gu
i

PREFACE
The California Energy Commission Energy Research and Development Division supports
public interest energy research and development that will help improve the quality of life in
California by bringing environmentally safe, affordable, and reliable energy services and
products to the marketplace.
The Energy Research and Development Division conducts public interest research,
development, and demonstration (RD&D) projects to benefit California.
The Energy Research and Development Division strives to conduct the most promising public
interest energy research by partnering with RD&D entities, including individuals, businesses,
utilities, and public or private research institutions.
Energy Research and Development Division funding efforts are focused on the following
RD&D program areas:
• Buildings End-Use Energy Efficiency
• Energy Innovations Small Grants
• Energy-Related Environmental Research
• Energy Systems Integration
• Environmentally Preferred Advanced Generation
• Industrial/Agricultural/Water End-Use Energy Efficiency
• Renewable Energy Technologies
• Transportation
Development of Fault Current Controller Technology is the final report for the Development of Fault
Current Controller Technology project (contract number UC MR-064) conducted by University
of California, Irvine. The information from this project contributes to Energy Research and
Development Energy Systems Integration Program.
For more information about the Energy Research and Development Division, please visit the
Energy Commission’s website at www.energy.ca.gov/research/ or contact the Energy
Commission at 916-327-1551.
ii

ABSTRACT
Fault current controller technology, also frequently referred to as fault current limiter
technology, has been identified as a potentially viable solution for expanding the capacity of the
transmission system and its service life to meet the growing demand for electricity by
addressing the impacts of the resulting higher fault currents. This report discusses the
development and field demonstration of distribution-class fault current controller technology.
The project focused on prototyping, test plan development, laboratory testing, and field testing
and demonstration, with a view toward potential further technical development and
application of the technology in transmission-level systems.
A full-size three-phase distribution-level high-temperature superconducting fault current
controller prototype was designed, built, and field-tested. The prototype fault current controller
went through several iterations of extensive testing prior to field installation. It was then
installed for field demonstration from March 2009 through October 2010.
The research team completed an initial design for a fault current controller based on solid-state
(power electronics) technology. Initial analysis indicated that design changes to the prototype
were necessary to improve its thermal management, the immunity of the control circuits to
noise and interference and to address mechanical issues. The additional cost for these items was
beyond the scope and budget of this project so this prototype design did not advance to the
laboratory and field test stages during the project period.
This report provided a survey of fault current controller technology development status,
including both the saturable-core and solid-state types represented by this project, followed by
detailed design considerations, laboratory test procedures and results, field test installation and
metering, and field demonstration outcomes. The report concluded with a summary of the
lessons learned and recommendations for future fault current controller research efforts and
commercial industry applications.
Keywords: Fault Current, Fault Current Controller, FCC, Fault Current Limiter, FCL, Short
Circuit Current, Power System Protection, Saturable-core Reactor FCL, Solid-State FCL, High
Temperature Superconductivity, HTS
Please use the following citation for this report:
Smedley, Keyue; Alexander Abramovitz. (University of California, Irvine). 2011. Development
of Fault Current Controller Technology. California Energy Commission. Publication
number: CEC-500-2013-134.
iii

TABLE OF CONTENTS
Acknowledgements...................................................................................................................................i
PREFACE................................................................................................................................................... ii
ABSTRACT..............................................................................................................................................iii
TABLE OF CONTENTS......................................................................................................................... iv
LIST OF FIGURES................................................................................................................................. vii
LIST OF TABLES......................................................................................................................................x
EXECUTIVE SUMMARY........................................................................................................................1
Introduction ........................................................................................................................................1
Project Purpose...................................................................................................................................1
Project Results.....................................................................................................................................2
Project Benefits ...................................................................................................................................7
CHAPTER 1: Introduction ......................................................................................................................9
1.1 The Short Circuit Problem in Power Systems........................................................................9
1.2 The FCC Solution .....................................................................................................................10
1.2.1 FCC Features and Benefits..............................................................................................10
1.2.2 FCCs in the Power Distribution System .......................................................................12
1.2.3 FCC Applications – Simulation Example .....................................................................13
1.3 FCC Development Objectives ................................................................................................16
CHAPTER 2: Fault Current Controller Technology Survey...........................................................17
2.1 Solid-State Fault Current Controller Technologies.............................................................17
2.1.1 Series Switch-type FCCs..................................................................................................17
2.1.2 Bridge-type FCCs.............................................................................................................18
2.1.3 Some Practical Difficulties of the Bridge-type FCCs...................................................22
2.1.4 Resonant-type FCCs ........................................................................................................23
2.2 Saturable-core Fault Current Controller Technologies.......................................................26
2.2.1 Basic Saturable-core FCC ................................................................................................27
2.2.2 Saturable Reactor Limiter for Current ..........................................................................30
iv

2.2.3 Magnetic-controlled Switcher-type Fault Current Controller...................................31
2.2.4 Improved Saturable-core FCC with Magnetic Decoupling........................................33
2.2.5 Saturable-core FCC with Parallel Permanent Magnet Bias........................................34
2.2.6 A Series-biased Permanent Magnet FCC......................................................................34
2.2.7 A Three-Material Passive dI/dt Limiter ........................................................................36
2.2.8 Saturated Open-Core FCC..............................................................................................38
2.2.9 Bias Power Supply Issues ...............................................................................................39
2.3 Summary ...................................................................................................................................41
CHAPTER 3: EPRI/Silicon Power Solid-State Current Limiter Development.........................42
3.1 EPRI/Silicon Power Solid-State Current Limiter Concept..................................................42
3.2 SSCL Design..............................................................................................................................43
3.2.1 Standard Building Block and Standard Power Stack..................................................43
3.2.2 Final Assembly .................................................................................................................44
3.2.3 Thermal Design ................................................................................................................44
3.2.4 Controls Design................................................................................................................45
3.2.5 SSCL Design Changes .....................................................................................................46
3.3 SBB Testing ...............................................................................................................................48
3.4 Summary ...................................................................................................................................48
CHAPTER 4: Zenergy Power First-generation HTS FCL Development ......................................52
4.1 Zenergy Power HTS FCL Concept ........................................................................................52
4.2 First-Generation “Spider” HTS FCL Design and Prototyping ..........................................53
4.3 HTS FCL Modeling..................................................................................................................56
4.3.1 Nonlinear Inductance Model..........................................................................................56
4.3.2 Finite Element Model and Analysis ..............................................................................57
4.3.3 Experimental Model Validation.....................................................................................60
4.4 Summary ...................................................................................................................................64
CHAPTER 5: Zenergy Power HTS FCL Laboratory Testing ..........................................................66
5.1 Standards...................................................................................................................................66
v

5.2 Pre-connection Testing............................................................................................................67
5.3 Normal State Performance Testing........................................................................................68
5.4 Fault Condition Testing...........................................................................................................72
5.4.1 Single-fault Test................................................................................................................72
5.4.2 Double-fault Sequence.....................................................................................................74
5.4.3 Endurance Test.................................................................................................................74
5.4.4 AC to DC Coil Coupling Test.........................................................................................74
5.5 Summary ...................................................................................................................................78
CHAPTER 6: Zenergy Power HTS FCL Field Testing.....................................................................79
6.1 Setup and Metering .................................................................................................................79
6.1.1 Avanti “Circuit of the Future” .......................................................................................79
6.1.2 Metering ............................................................................................................................79
6.2 FCL Key Technical Events ......................................................................................................82
6.2.1 Pre-Installation Events ....................................................................................................82
6.2.2 Post-Installation Events...................................................................................................84
6.3 Analysis of Major Events ........................................................................................................87
6.3.1 Loss of DC Bias Current, March 16, 2009......................................................................87
6.3.2 Downstream Short Circuit Fault – Jan 14, 2010 ...........................................................93
6.4 Summary ...................................................................................................................................97
CHAPTER 7: Development of Zenergy Power Second-generation “Compact” HTS FCL .......98
7.1 “Compact” FCL Concept ........................................................................................................98
7.2 Comparison with Other Saturable Core FCL Designs......................................................100
7.3 Distribution-Level “Compact” FCL Prototyping ..............................................................101
7.4 Transmission-Level Compact FCL Development .............................................................106
7.5 Summary .................................................................................................................................107
CHAPTER 8: Lessons Learned ...........................................................................................................110
CHAPTER 9: Potential Applications of FCC Technology ............................................................113
9.1 Fault Current Limiter in Substation Bus Tie ......................................................................113
vi

9.2 Fault Current Limiters in the Incoming Feeders ...............................................................113
9.3 Fault Current Limiters in the Outgoing Feeders ...............................................................114
9.4 Grid Coupling.........................................................................................................................114
9.5 Power Plant Auxiliary Power Feeder..................................................................................115
CHAPTER 10: Installation of the Zenergy Power HTS FCL ........................................................117
10.1 Site Preparation to Accommodate the FCL........................................................................117
10.1.1 Pre-installation High-Power and High-Voltage Testing..........................................117
10.1.2 On-site Commissioning Test ........................................................................................117
10.1.3 Special Bypass Switch....................................................................................................118
10.1.4 Civil Engineering ...........................................................................................................118
10.1.5 High Voltage Connections and Grounding................................................................119
10.1.6 Auxiliary Power .............................................................................................................119
10.1.7 Noise Ordinance Compliance ......................................................................................119
10.1.8 Connection to High Voltage Side.................................................................................120
10.1.9 Metering and Interconnection Cabinet .......................................................................120
10.1.10 Communication Interconnection Point...................................................................121
10.1.11 Bypass Relay and Alarm Cabinet ............................................................................121
CHAPTER 11: Summary and Recommendations...........................................................................122
11.1 Summary .................................................................................................................................122
11.2 Development of a Solid-State Current Limiter..................................................................122
11.3 Development of a Saturable-core High Temperature Superconducting Fault Current
Limiter .................................................................................................................................................123
11.4 Recommendations..................................................................................................................127
GLOSSARY............................................................................................................................................128
REFERENCES ........................................................................................................................................131
APPENDIX A: List of Attachments...................................................................................................A-1
LIST OF FIGURES
Figure 1: Possible FCC Locations in the Power Grid..........................................................................13
Figure 2: PSCAD Model of Hypothetical 230 kV Substation with FCC Protection........................14
vii

Figure 3: Comparison of PSCAD Simulation Waveforms of the Fault Current of a Hypothetical
230 kV Substation, with: No Protection Devices; Air Core Inductor; FCL. .....................................15
Figure 4: Basic Topology of the Series Switch-type FCC....................................................................17
Figure 5: Bridge-Type FCC.....................................................................................................................19
Figure 6: Single-reactor Rectifier Bridge FCC......................................................................................20
Figure 7: Transformer-isolated Three-phase Rectifier Bridge FCC...................................................21
Figure 8: Two-Reactor SCR Bridge-type FCC......................................................................................22
Figure 9: AC Line and DC Reactor Currents for a Bridge-type FCC................................................23
Figure 10: Topology of the Resonant-type FCC...................................................................................24
Figure 11: Types of Resonant-circuit FCCs ..........................................................................................25
Figure 12: Inverter-based FCC ...............................................................................................................26
Figure 13: Conceptual Structure of a Saturable-core FCC..................................................................27
Figure 14: Connection of a Saturable-Core FCC..................................................................................28
Figure 15: Saturable-core FCC Operating Points.................................................................................29
Figure 16: Saturable-Core FCC Inductance as a Function of AC Current .......................................29
Figure 17: Structure of a Saturable Reactor Limiter (SLR) .................................................................30
Figure 18: Magnetic-controlled Switcher-type FCL ............................................................................32
Figure 19: Saturable-Core FCL with Magnetic Decoupling...............................................................34
Figure 20: A Parallel-Biased Permanent Magnet FCC........................................................................35
Figure 21: A Series-Biased Permanent Magnet FCC...........................................................................35
Figure 22: A Three-Material Passive dI/dt Limiter..............................................................................37
Figure 23: A Three-material Magnetic Current Limiter .....................................................................38
Figure 24: A Saturated Open-core FCC ................................................................................................39
Figure 25: DC Magnetization System for a 35 kV, 90 MVA Superconducting Saturated Iron-Core
FCL.............................................................................................................................................................40
Figure 26: Parallel Overvoltage Suppression Winding Configurations...........................................41
Figure 27: SSCL Schematic......................................................................................................................42
Figure 28: Simulated Waveforms of the SSCL Circuit........................................................................43
Figure 29: Modular Structure of the SSCL............................................................................................44
Figure 30: The SSCL Assembly and Its Components..........................................................................45
Figure 31: SSCL Control Architecture...................................................................................................46
Figure 32: SEL Industrial Panel-mount Controller..............................................................................47
Figure 33: Current Interruption Simulation Results for the Standard Building Block (SBB)
Module.......................................................................................................................................................50
Figure 34: Conceptual Structure of Zenergy Power Single-Phase Saturable-Core Reactor FCL..52
Figure 35: Three-phase Zenergy Power Saturable-core Reactor HTS FCL......................................53
Figure 36: Cutaway Views of the Distribution-Level Three-Phase HTS FCL Assembly...............54
Figure 37: Experimental Zenergy Power HTS FCL Prototype ..........................................................55
Figure 38: Magnetization Curve of M-6 Steel Used in the Finite Element Model ..........................58
Figure 39: Finite Element Model of a 15 kV Three-Phase FCL..........................................................59
Figure 40: Total Flux Density Due to DC Magnetization at 80,000 Amp-Turns.............................59
Figure 41: FEM Simulated Results and Model Fitting........................................................................60
Figure 42: Zenergy HTS FCL High-Power Short-circuit Test Setup.................................................61
viii

Figure 43: Zenergy HTS FCL Short-Circuit Tests: Measured and Modeled Current as a Function
of Time.......................................................................................................................................................62
Figure 44: Measured vs. Modeled Parameters for the Zenergy HTS FCL.......................................63
Figure 45: Measured vs. Updated Modeling Results for the Zenergy HTS FCL............................64
Figure 46: Experimental Waveform of FCL Full-load Current in the Normal State ......................68
Figure 47: Measured FCL Impedance (Ω) and Voltage Drop (%) as a Function of Load Current69
Figure 48: Measured Normal-State FCL Power Loss (%) as a Function of Load Current.............70
Figure 49: Measured Temperature on AC Coil During Temperature Rise Test with 750 A (rms)
Load Current per Phase ..........................................................................................................................71
Figure 50: FCL Insertion Impedance as a Function of DC Bias MMF ..............................................72
Figure 51: Experimental Results of Zenergy HTS FCL Clipping a 23 kA Fault..............................73
Figure 52: Actual vs. Simulated Fault Current Clipping (%) as a Function of Prospective Fault
Current Level............................................................................................................................................74
Figure 53: Experimental Bus Voltage Sag Under a 20 kA Fault Condition .....................................75
Figure 54: FCL Current and Voltage Waveforms for a 20 kA Symmetric Double Fault with 2
Seconds Spacing.......................................................................................................................................76
Figure 55: Experimental FCL Current During the 20 kA, 1.25-second, Extra-long Fault..............77
Figure 56: Coupling of the AC-to-DC Coils in the Normal State and Under Fault Conditions ...77
Figure 57: Diagram of the Avanti “Circuit of the Future” .................................................................80
Figure 58: Zenergy Power HTS FCL Prototype in Operation at SCE’s Avanti Circuit of the
Future.........................................................................................................................................................81
Figure 59: Aerial View of Zenergy Power HTS FCL Installation in SCE’s Shandin Substation,
San Bernardino, California, March 2009...............................................................................................81
Figure 60: Location of the Data Recorders at Shandin Substation....................................................82
Figure 61: Diagram of the Avanti Circuit Prior to the March 16, 2009 Loss of DC Bias Event.....88
Figure 62: HTS Coil Current and Coil Voltage for the March 16, 2009 Loss of DC Bias Event ....89
Figure 63: Line-to-line Voltage between A and B Phases at CB #1 for the March 16, 2009 Loss of
DC Bias Event...........................................................................................................................................90
Figure 64: Line-to-ground Phase Voltages and HTS Coil Current for the March 16, 2009 Loss of
DC Bias Event...........................................................................................................................................91
Figure 65: Simplified PSCAD Model of the FCL Installed in the Avanti Circuit............................92
Figure 66: Simulated FCL Inductance for the Loss of DC Bias Event...............................................92
Figure 67: Comparison of Simulated vs. Measured Line-to-ground Voltage for the Loss of DC
Bias Event..................................................................................................................................................93
Figure 68: Voltages and Currents Measured Downstream of the FCL During the Multiphase
Fault Event of Jan. 14, 2010.....................................................................................................................94
Figure 69: PSCAD Simulation Results of the Multiphase Fault of Jan. 14, 2010.............................95
Figure 70: Cryogenic System Behavior During the Jan. 14, 2010 Fault ............................................96
Figure 71: The Architecture of a Single-phase Saturated-Core “Compact” FCL............................98
Figure 72: Saturated-core FCL Architectures: (a) “Spider” FCL; (b) Open-core...........................100
Figure 73: Physical Structure of a Rectangular-frame, Three-Phase Compact FCL Prototype...102
Figure 74: Path to Commercialization for FCL Prototypes..............................................................105
ix

Figure 75: Design and Layout of a Commercial 15kV-class Circular-frame Compact FCL
Assembly.................................................................................................................................................105
Figure 76: Three-phase Distribution-level, 15kV-class, Compact FCL Prototype with “Dry-
cooled” HTS Magnets Surrounding an Oil-filled Dielectric Tank with AC Coils........................106
Figure 77: Perspective View of a Transmission-level Single-phase Compact FCL.......................107
Figure 78: Perspective View of a Compact FCL Substation Installation........................................107
Figure 79: FCL Installation in Substation Bus Tie Application........................................................113
Figure 80: FCL Installed in Incoming Feeders ...................................................................................114
Figure 81: FCL Installation in Outgoing Feeders...............................................................................115
Figure 82: Two Sub-Grids Coupled by an FCL..................................................................................115
Figure 83: Location of FCLs in Power Plant Auxiliary Systems: (a), (b) Medium Voltage; (c) Low
Voltage.....................................................................................................................................................116
Figure 84: Bypass Switch for Zenergy Power FCL Installation.......................................................118
Figure 85: Installation of Voltage and Current Monitors for the Zenergy Power HTS FCL.......121
LIST OF TABLES
Table 1: Design Parameters and Main Features of the Zenergy Power Prototype HTS FCL........56
Table 2: Zenergy FCL Testing Criteria..................................................................................................66
Table 3: Comparison of Fault Currents – FCL Active vs. FCL Bypassed.........................................96
Table 4: Comparison of the “Spider,” Open-core and Compact FCLs...........................................101
Table 5: Comparison of the Main Features of the “Spider” and Compact FCL Designs.............102
Table 6: Specifications of Prototype Compact FCLs .........................................................................103
Table 7: Possible Characteristics of a Future Transmission-level Compact FCL ..........................108
x

EXECUTIVE SUMMARY
Introduction
Overall electric current loading on the transmission system has been rapidly increasing to meet
the growth in electricity demand. The resulting higher electrical energy levels have increased
the potential fault current magnitudes at locations throughout the transmission system. Fault
currents in many instances may exceed the capability of existing protection systems such as
circuit breakers to interrupt the faults safely and reliably, and represent an imminent threat to
electrical equipment and the safety of utility workers and the public. Consequently, a utility
must either upgrade breakers and equipment or reconfigure its system to reduce the potential
fault current. Both solutions are costly and frequently reduce system reliability and power
transfer capability. Application of fault current controller (FCC) technology, also frequently
referred to as fault current limiter (FCL) technology has been identified as a potentially viable
solution for managing fault currents to keep them within the existing short circuit capacity
ratings of the system in order to expand and extend the transmission system’s capacity and
service life while minimizing capital costs. This approach allows utilities to continue to meet the
growing demand for electricity reliably and cost-effectively.
Project Purpose
The California Energy Commission (CEC) funded this project to develop and evaluate reliable,
cost-effective, and environmentally acceptable technologies for the control of high fault
currents. The goals of the proposed research, directed by Principal Investigator the University
of California at Irvine were to establish the desired criteria for FCC performance and to test two
different leading FCC technologies against those criteria in controlled laboratory testing and
field demonstration in a commercially operating Southern California Edison Company (SCE)
distribution system. The overall objective of this project was to facilitate the improved, safe and
reliable operation of the power system by advancing FCC technology as a cost-effective and
environmentally-preferred option to breaker upgrades or system reconfiguration, and to
evaluate the realistic potential of this technology to mitigate fault current levels at higher
voltages in the electric system through real-world utility testing. The ultimate program goal was
to enable the commercialization of FCC technology for the benefit of California and the United
States.
The objectives of the proposed research were to:
1. Establish an FCC test plan including test criteria, test protocol, test site selection, test
scheduling, data collection plans, and interface requirements for both the laboratory
testing and field demonstration.
2. Prototype two FCCs, one saturable-core type developed by Zenergy Power plc, and one
solid-state type developed by the Electric Power Research Institute (EPRI) and Silicon
Power Corporation team.
1

3. Conduct laboratory testing of the two FCCs against the criteria, including pre-
connection high-voltage insulation tests, normal operation tests and fault current
limiting tests.
4. Install the FCCs on the SCE distribution system and perform field demonstration for a
minimum of six months.
5. Complete the evaluation of the technologies on the basis of performance, respective
strengths and weaknesses, costs, reliability, installation, operation and maintenance
issues and potential for development to high-voltage design.
Project Results
The project started in September 2007. This was not the first attempt in the United States (U.S.)
to develop a distribution-level FCC and demonstrate it on an operating power system.
However, none of the earlier efforts had led to a successful field demonstration in the U.S. The
obstacles to overcome were challenging. There were no industry standards for such a device,
complicating the development of testing protocols. There were also significant engineering
challenges in developing the prototypes to withstand the currents and voltages in the field. In
addition, the prototype FCCs had to meet SCE’s specifications for field demonstration on their
system. Finally, the FCCs had to demonstrate the ability to withstand live circuit events and
severe environmental conditions.
An important step taken in the early stages of the project was a focused effort to develop a test
plan that was acceptable to SCE and workable for the teams. This test plan needed to
incorporate to the extent possible the relevant engineering standards that apply to electrical
apparatus. Zenergy Power worked closely with the Georgia Institute of Technology’s (Georgia
Tech) National Electric Energy Testing, Research and Applications Center (NEETRAC) and
several of its member utilities, including SCE to implement a detailed FCC test program based
on selected Institute of Electrical and Electronics Engineers (IEEE) and the Council on Large
Electric Systems (CIGRE) standards and protocols for transformers and reactors. The
EPRI/Silicon Power team created their plan based on American National Standards Institute
(ANSI) C39.09-1999 and ANSI C37.06-2000, covering the entire spectrum of possible tests that
needed to be carried out on their FCC, including component level factory tests, system level
factory tests, acceptance tests and system field tests. Many technical issues related to laboratory
test and field demonstration interfacing were resolved via a series of meetings with
knowledgeable SCE engineers. Both teams completed their test plans, which provided valuable
design and testing guidelines.
Zenergy Power completed the design, construction, and testing of an FCC based on a saturable-
core concept. This type of FCC is basically a coil wound on an iron core and connected in series
with the power line. In the normal state the core is biased into magnetic saturation by a
dedicated direct current (DC) electromagnet so that the inductance of the magnetically
saturated coil and the corresponding voltage drop across the coil terminals are negligible and
the coil has no deleterious effect on the system during normal operation. The magnetic core is
driven out of saturation by the fault current in the event of a fault. Consequently, the coil
2

becomes highly inductive, effectively inserting a large impedance into the power circuit and
limiting the fault current without any need for active sensing or switching. A key enabling
feature of the Zenergy Power high-temperature superconducting (HTS) FCL was the use of a
HTS winding for the DC bias circuit, which reduced bias circuit power loss while increasing the
intensity of the bias field. The Zenergy Power device is referred to as an HTS FCL. As
previously noted, while fault current controller (FCC) is the preferred terminology for the class
of technologies that address the management of fault currents, fault current limiter (FCL) is also
commonly used. Both Zenergy Power and EPRI/Silicon Power Corporation (Silicon Power)
used the term FCL when referring to their devices and their terminology was used in this report
in such references.
A full-scale three-phase 12 kilovolt (kV) distribution level HTS FCL prototype was built for a
rated load current of 1200 amps (A) (three phase), less than one percent normal voltage drop at
maximum load, and 20 percent fault current clipping capability. Zenergy also developed a
comprehensive test plan for laboratory testing of their HTS FCL.
The prototype HTS FCL went through a series of rigorous laboratory tests in BC Hydro’s
Powertech Laboratory in Surrey, British Columbia, Canada. A total of 65 separate tests were
conducted, including 32 full-power fault tests with first peak fault current levels up to 59
kilowamps (kA) at the rated voltage. Fault tests included individual fault events of 20-30 cycles
duration, as well as multiple fault events in rapid sequence to simulate automatic re-closer
operation and extended fault events of up to 82 cycles duration simulating primary protection
failure scenarios. The HTS FCL passed the test criteria in all cases. Only minor modifications to
the high-voltage isolation layout were needed. The unit was then transferred to the SCE
Westminster High-Voltage Test Facility for acceptance tests by SCE per IEEE Standard C57-
12.01-2005.
The HTS FCL was installed in SCE’s Avanti Circuit of the Future and operated from March 2009
through October 2010 after the successful completion of the acceptance tests. The unit was
integrated into SCE’s supervisory control and data acquisition (SCADA) system and operated in
real time to provide protection to the distribution circuit during its field demonstration. A
monitoring and data acquisition system was installed to archive performance data of the HTS
FCL throughout the field testing period.
Several deficiencies in the original design and construction were revealed during the course of
the field demonstration: a programmable logic controller (PLC) random-access memory (RAM)
overflow programming issue; a heating, ventilation, and air conditioning (HVAC) shutdown
due to excessive ambient temperature; helium leaks in the cryogenic coolers; nitrogen pressure
instabilities associated with the liquid nitrogen cooling system; and a terminal block short.
These events provided valuable learning experiences for the research team to identify design
weaknesses and make necessary corrections along the way, resulting in significant design and
implementation enhancements. The FCL also experienced an in-service multiple-fault event
during the live grid demonstration and it was able to limit the fault current throughout the
individual faults.
3

The FCL survived and thrived during its 19-month field service demonstration, which included
nine months on line and 10 months in standby model. The standby period was due to a back-
ordered bypass switch required before the unit could be put back on line after a maintenance
outage. The immediate research benefit obtained from the Zenergy Power HTS FCL‘s
operational experience was validating the importance of a more compact, more reliable, and
easier to maintain non-HTS design for the FCL, and one which was scalable to transmission
voltages.
The field demonstration was instrumental in furthering the development of FCL technology
and contributing to Zenergy Power’s first commercial sale of a medium-voltage device. Zenergy
Power extended the research results to prototype development and applications at the
transmission voltage level.
The EPRI and Silicon Power team completed an initial design for an FCL based on solid-state
circuit technology employing Super Gate Turn-Off (SGTO) thyristor switches. The FCL design
consisted of a set of standard building blocks (SBB), each containing an SGTO-based circuit
designed for 5 kV voltage blocking and 2000 A continuous current ratings. The SBBs were used
in multiples to form standard power stack (SPS) assemblies rated for 50 kV voltage blocking
and 2000 A continuous current ratings. One SPS per phase was required for the 15 kV-class
system. Three SPSs were then housed in an oil-filled tank to form a complete three-phase FCL,
referred to as the Solid State FCL (SSFCL) unit. In the normal state the SGTO switches were
turned on, which allowed continuous rated current to flow through the circuit. The control
circuits opened the SGTO switches to insert a current limiting reactor (CLR) into the circuit to
limit the fault current when a fault was detected.
The 15kV, 1200 A SSFCL was designed for outdoor use according to the C57.12.00-2010 IEEE
Standard for General Requirements for Liquid-Immersed Distribution, Power, and Regulating
Transformers. The overall SSFCL package was physically similar to a typical substation
transformer tank with an external radiator bank for the cooling system. A comprehensive test
plan was also developed for the SSFCL.
Less than optimal thermal performance was predicted during initial simulation of the thermal
performance of the SSFCL design, which could have led to undesirable effects on the stability
and reliability of the final, manufactured SSFCL. The designs of the thermal management
system and the control boards were therefore modified during the course of the project to
provide a more efficient thermal management system and to improve the noise immunity of the
controls. An unfortunate result of this necessary re-design was a much higher projected
construction cost than allowed by the available project budget; therefore, this part of the project
was terminated at the conclusion of the design stage. EPRI was able to incorporate this and
other lessons learned into a newer, separately-funded SSFCL design, which they intended to
develop and offer for demonstration in the future.
The research in this project led to the successful field demonstration of the Zenergy Power HTS
FCL, marking a milestone event in the history of FCL development in the U.S. The experience
gained from the research contributed to a more reliable controller, a dramatic reduction of the
4

FCL’s size (with a slight increase in weight), and the replacement of liquid nitrogen cryogenic
refrigeration by a low-maintenance, dry (non-HTS) cooling system for increased reliability and
reduced maintenance requirements.
The Zenergy Power HTS FCL was first laboratory tested successfully against FCL specifications
and test criteria developed by U.S. utilities, including the host utility SCE. The HTS FCL was
then successfully demonstrated in actual field operation for an extended period, during which it
experienced several system events and performed as designed. The HTS FCL unit was rated at
15 kV, 1200 A, 110 kV (Basic Insulation Level), steady-state insertion impedance of less than one
percent and a fault current reduction capability of 20 percent at 23 kA maximum fault current.
The first-generation “Spider-core” HTS FCL design had a dry-type transformer (air dielectric)
structure, total overall dimensions of 19 feet by 19 feet by seven feet, and weighed
approximately 50,000 pounds. The second-generation “Compact” HTS FCL that followed from
this demonstration employed innovative core architecture and oil dielectric transformer
construction techniques that led to a much more compact size of approximately eight feet by 10
feet by 11 feet and a somewhat increased weight of approximately 67,000 pounds. The Compact
FCL had a power rating similar to the Spider-core FCL, but with a much higher fault current
limiting performance. The new Compact FCL design was an improved option for applications
where real estate was limited, albeit with somewhat increased weight.
The first-generation device had a relatively low power rating on the order of 25 megavolt-
amperes (MVA) so the present manufacturing cost may not represent a high value proposition.
This issue was being addressed on two fronts: HTS FCL technology was being improved to
reduce basic device manufacturing cost, and it was being scaled up for higher voltages and
currents. The latter may be more important in terms of cost-effectiveness and value proposition
considering the fact that higher voltage substations can often be more crowded and have fewer
options for expansion and that higher voltage components such as switchgear, insulators,
transformers and bus-work were exponentially more expensive than their low-voltage
counterparts. Furthermore, the larger renewable power generators will most likely desire to
connect to the transmission system and high-voltage tie-lines have become increasingly more
common as more power is wheeled from long distances and as grid interconnections occur to
improve reliability and better control power flows. All of these factors were projected to
continue to increase energy levels within the grid, leading to potentially unsafe fault current
levels. In some cases higher-rated components cannot be retrofitted in the available space,
leading to lengthy and costly major upgrades of grid infrastructure. Economic studies and
performance models showed that FCC technologies can be very cost-effective compared to
major upgrade projects at high voltages at current performance levels and price points.
The HTS FCL demonstration project also brought to light a reliability issue. While the HTS FCL
was maintained in good operating condition and performed satisfactorily as a demonstration
unit, an improvement is needed for a commercial product. An FCL ideally needs to target “five-
nines” reliability (0.99999 availability) with only a single, scheduled, short-duration annual
maintenance outage. Even better would be extending the maintenance interval to two years or
longer, but an annual outage of a day or less would be acceptable if the FCL were otherwise
5

reliable and required no additional maintenance periods. Bypassing the FCL from service
temporarily for maintenance could be a risk for the circuit if the FCL was to be relied on for
protection. Possible temporary solutions for a scheduled bypass could be accomplished by
installing an air-core reactor, splitting buses, or otherwise re-configuring the system for short
durations, but these methods might not be feasible or acceptable for long-term operation. An
FCL should be designed to the maximum extent possible to allow essential routine
maintenance, such as cryogenic system maintenance to be performed while the unit is
energized. Zenergy Power was aggressively modifying their HTS FCL technologies for reduced
maintenance requirements, higher mean time between failure (MTBF) and lower mean time to
repair (MTTR).
A key area for improvement identified during the field testing was the liquid nitrogen cooling
system, which required periodic maintenance of the cryostat including drying, establishing
vacuum and refilling the liquid nitrogen. This may not be practical since the unit would
typically be installed in a high-voltage substation area. However, this issue was resolved in the
second-generation FCL design by replacing the liquid nitrogen HTS system with a cryogen-free,
“dry-type” conductive cooling system, available as a commercial off-the-shelf unit, resulting in
a more reliable and robust system requiring less maintenance.
Overall, the Zenergy Power HTS FCL project was considered to have been an important success
and it has already led to a scaled-up design for a transmission-level 138 kV FCL application.
The EPRI/Silicon Power SSFCL represented another potentially cost-effective solution to the
rapidly increasing fault current levels in utility systems. One advantage of this type of FCL was
the flexibility to be configured as either interruptive (i.e., to act as a “solid-state circuit breaker”)
or simply limiting the fault current (and leaving the fault interruption to existing protective
devices), with only minor design differences. Also, a solid-state FCL may be used to limit the
current of superconducting cables to enable the use of smaller cable sizes. The solid state FCL
also had a unique capability to limit inrush currents, even for capacitive loads.
EPRI/Silicon Power identified some potential thermal management issues in their initial design.
An improved system design was completed and the major technical design challenges such as
the thermal management system and the control circuit architecture and timing issues were
resolved. However, the resulting cost increase to actually construct the device was constrained
by the project budget. Thus, this FCL was not able to advance to the laboratory test and field
demonstration stages under this project.
According to the design predictions, the system size and weight would be 6.5 feet by 12 feet by
12 feet and the weight would be 62,000 pounds with oil cooling. These size and weight
specifications would have been considered acceptable for utility applications.
The SSFCL employed a modular and scalable design. The SSFCL designed for the 12 kV line in
this project period was composed of 10 SBBs in series. A future consideration with this design is
that voltage sharing to maintain all the SBBs under the blocking voltage under dynamic
conditions may be a technical challenge for transmission level applications where the voltage
level is above 100 kV.
6

Overall, the EPRI/Silicon Power team made a substantial first effort in the design, development
and improvement of the SSFCL concept. Many engineering challenges were identified and a
new design was completed during the project period.
The research team accumulated extensive experience from the development of FCC technology
and from the operation and support of the field demonstration at SCE’s Avanti Circuit of the
Future. Further fast-paced and more advanced development of FCC technology was strongly
recommended. Substantial investments in system upgrades would have to be made on a fast-
track basis in order to maintain the required levels of electric system availability and reliability
as the demand for electrical energy rapidly increases, particularly in response to renewable
power and “green technology” initiatives, and in the absence of better alternatives. Significant
amounts of these investments may be potentially avoidable if suitable FCC technologies were
available to California utilities and if California utilities were given reasonable incentives to
deploy the new technologies. The research team recommended that investments be made in an
accelerated program of FCC technology focused on reducing the cost, improving the reliability
and increasing the voltage and current ratings of FCC.
Project Benefits
Electricity is a vital force in our economy. It is an important goal of the California Institute for
Energy and Environment and the California Energy Commission to support the technologies
required for providing a reliable, safe and environmentally responsible electrical energy supply.
As such, FCC technology is a potentially cost-effective alternative to the capital-intensive
upgrades of the power system that would ordinarily be required to meet growth in electrical
demand. These demonstration projects have already resulted in two test plans and two full FCC
designs, and have contributed to one commercial sale and one migration to a transmission-level
application. The recommended enhancements to FCC technology will help California utilities
and ratepayers reap the benefits. New and promising technologies are also important for
economic vitality and job creation.
7

California Energy Comission 500-2013-134 DEVELOPMENT OF FAULT CURRENT CONTROLLER TECHNOLOGY Prototyping, Laboratory Testing, and Field Demonstration

CHAPTER 1:
Introduction
The overall electric current loading on the transmission system has been increasing due to
continuing growth in electricity demand. These higher power currents have increased the
potential fault current magnitudes at locations throughout the transmission system. The fault
currents, in many instances, may exceed the capability of existing protection systems (circuit
breakers) to interrupt the faults safely and reliably, as well as threaten the safety of electric
equipment and utility personnel. Consequently, a utility must upgrade these breakers and
equipment or reconfigure its system to reduce the available fault current. Both solutions are
costly, and frequently reduce system reliability and power transfer capability. Fault current
controller (FCC) technology has been identified in the research community as a potentially
viable solution for managing the fault currents within the rated short circuit capacity
constraints, in order to meet the growing demand for electricity without the costly capital
upgrades that would otherwise be required.
1.1 The Short Circuit Problem in Power Systems
Among the numerous faults occurring in power distribution systems, the short circuit fault is,
probably, the most destructive one [1]. The short circuit fault can generate fault current more
than 20 times the rated current. Fault currents are normally interrupted by the protection
circuits, but serious consequences can still result, such as loss of service to customers, under-
voltage or over-voltage transients, and/or loss of system synchronization. In extreme cases, fault
currents can melt conductors, destroy insulation, cause fires, or induce strong mechanical
forces, all of which can damage equipment. Serious safety, reliability and economic
consequences will result from such incidents.
Distribution protection systems mainly rely on two customary and proven protection devices.
The first, a fuse, is a simple, rugged, small and reliable protective device that can be used to
interrupt fault currents as high as 200 kA. It is a self-triggering device and requires neither
sensors nor actuators. A fuse can open in less than half a cycle at current levels that are well
under the peak fault current [2], [3]. The major disadvantage of a fuse is that it is a single-use
device, which must be manually replaced in order to restore electric service.
The second device is a circuit breaker (CB), a more sophisticated but still proven and reliable
protection device [4]. A CB can be “tripped” by monitoring and instrumentation systems and
then reset either automatically or remotely. However, CBs with high current interrupting
capabilities are typically bulky and expensive. Furthermore, it is recognized that CBs require
periodic maintenance and calibration and have a limited number of cycles of operation before
teardown and rebuilding is required.
World economic growth poses a fast increasing demand for electrical power. Higher electrical
loads, new consumers, and new distributed generation plants are constantly being added to the
power grid. Higher fault current is a logical result of the growing capacity of the power grid;
there is an increasing possibility that the expected fault current levels will exceed the
9

interrupting capability of the existing protection systems in the near future, with the problem
only getting more urgent as time goes on. Failure of protection equipment to interrupt the fault
current may cause extensive damage and put at risk the integrity and stability of the power
system. Replacement or upgrading of a great number of CBs is a possible, but costly, solution to
cope with the rising fault current levels.
However, the problem extends beyond the potential costs of new CBs. An upgraded CB allows
higher fault current to flow through the rest of the now-underrated electrical equipment, which
would be subjected to a much higher stress than originally designed for. The high stresses may
incur damages even in the case that the CB interrupts the higher fault currents as designed.
Furthermore, upgrading the CB to higher currents allows higher energy to the fault. Thus, in
the event of the fault, the hazards to electrical installations may become increasingly higher.
High fault current can also induce higher voltage in the grounding conductors and reduce the
effectiveness of the substation grounding grid protective function. Thus, safety to utility
personnel as well as the general public becomes an issue to consider. An additional concern is
that, in certain cases, mechanical supports may need to be reinforced to withstand higher
stresses from the intense magnetic fields induced by the fault currents. Perhaps even the
clearances between the conductors will need to be increased. Therefore, along with upgrading
the CBs there may be a need to upgrade some other parts of the system as well. However,
limited capital budgets may not allow such a massive upgrade of the existing installations.
There are several traditional approaches to suppress the fault current in electrical systems [5].
System reconfiguration and bus splitting might be preferred for the fast growing areas.
However, this approach comes at the cost of added grid complexity as well as additional
switchgear. Other methods frequently used include the use of high-impedance transformers, a
reactive impedance inserted between transformers and the electrical ground, or air-core
reactors. All of these approaches, unfortunately, tend to reduce system efficiency and/or
degrade system voltage regulation.
An alternative solution to the problem, which has gained much attention lately, is the
application of a Fault Current Controller (FCC) [6]-[10], also known as a Fault Current Limiter
(FCL). These terms usually refer to more recent or advanced technologies to control fault
currents without the drawbacks of previous approaches. Funded by both government and
private companies, research and development activities of various types of FCCs have been
reported by many research institutions and private companies in many countries around the
world [11], [12]. The recent trends of deregulation, competition and restructuring of the power
grid have invoked a renewed interest in innovative FCC technologies for implementation of
reliable and economically feasible commercial devices for the grid.
1.2 The FCC Solution
1.2.1 FCC Features and Benefits
Application of FCCs has been identified as one potentially viable approach for limiting the fault
currents in power systems to safe levels. An FCC is typically installed in series with the
10

equipment to be protected. Under normal operating conditions, the FCC should display
negligible impedance so that the power flow is unobstructed. In the event of a fault, however,
the FCC’s impedance should rapidly increase, acting to limit the fault current to an appropriate
level.
Insight into FCC concepts and technologies can be found in several excellent references [13]-
[15]. In general, FCCs may be classified by their principle of operation and key technological
components used. As reported, FCCs can be implemented with passive non-linear elements,
inductive devices, vacuum switches, semiconductor switches, superconductors, as well as
combinations of these devices. FCCs may also be classified as either an interrupting type or a
non-interrupting type. Interrupting FCCs are designed to interrupt the entire fault current,
thereby acting as a circuit breaker. The task of a non-interrupting FCC is to limit the fault
current magnitude to an acceptable level by absorbing or diverting a portion of the current; the
limited current can then be safely interrupted by the existing protection systems. Technical
feasibility of various types of FCCs has been already demonstrated with full power prototypes
or scaled models.
From review of the literature and discussions with utilities and FCC developers, it becomes
apparent that for an FCC to be commercially successful, it should have the following desired
features.
In the normal (non-faulted) system condition, the FCC is expected to:
a) have a low insertion impedance;
b) be able to withstand distribution and transmission level voltages and currents
(continuous rating);
c) have a low voltage drop;
d) have a low power loss;
e) have a low distortion and provide high downstream power quality; and
f) have low electromagnetic emissions.
In case of a fault, the FCC is required to:
a) be capable of limiting the first fault current peak;
b) display a large increase of impedance;
c) have a sufficient fault-condition impedance;
d) have an acceptable power loss;
e) tolerate the mechanical stresses;
f) endure the temperature rise;
g) withstand the fault condition for a sufficient time;
11

h) endure a sequence of recurring faults; and
i) be capable of fast transition from normal to faulted state and vice-versa.
On a system level, the FCC is desired to:
a) be a fully autonomous system;
b) be remotely accessible;
c) operate with minimum down time;
d) require no special current sensors or actuators;
e) at minimum, present no interference with existing protection schemes, and ideally have
the ability to coordinate with them for improved system protection;
f) have a high level of reliability; and
g) exhibit “fail-safe” operation in case of malfunction.
For wide-scale commercialization, the following attributes of an FCC would be desirable:
a) acceptable size, weight and cost;
b) simple installation procedures;
c) low maintenance requirements;
d) long service life;
e) tolerance for harsh field conditions and weather extremes; and
f) compliance with the basic insulation level (BIL) standards, lightning tests, and other
appropriate standards and specifications.
1.2.2 FCCs in the Power Distribution System
An FCC can serve different purposes in the power system, depending on where and how it is
installed. Figure 1 illustrates three possible configurations appropriate for FCC installation [16].
The FCCs in positions marked A in Figure 1 can help reduce generator infeed fault currents,
prevent transformers from damage, and help reduce the voltage sag on the upstream high-
voltage bus during a fault on the medium-voltage bus. Also, transformers with low impedance
can be used to maintain voltage regulation at a higher power level and to meet increased
demand on a bus without upgrades to circuit breakers or other equipment. Alternatively, an
FCC could be installed in the bus tie position B as shown in Figure 1. In the event of a fault on
one of the buses, the FCC can help maintain the voltage level on the un-faulted bus. High
capacity is available to both buses resulting in better transformer rating utilization. Smaller and
less expensive FCCs can be installed in the feeder position, as shown in position C, to provide
protection to overstressed equipment that is difficult to replace, such as underground cables or
transformers in vaults.
12

Figure 1: Possible FCC Locations in the Power Grid
L2
T2
B1 B2
R1 R2 R3
FCL
FCL
B
C
FCL A
L1
T1
FCLA
G
FCL A
Source: Larbalestier, et. al. [16]
1.2.3 FCC Applications – Simulation Example
Many California and US substations are experiencing a steady increase in electrical loading due
to economic growth, introduction of electric vehicles, and proliferation of renewable and
distributed generation. In many cases, the substation fault current is approaching or exceeding
the capabilities of the existing protective equipment, especially circuit breakers (CBs) with
ratings of 63 kA. Although CBs with ratings up to 83 kA are also available, they are much
larger, more expensive and sometimes difficult to obtain. Changing the power flow by splitting
substation buses or reconfiguring circuits is possible, but can negatively impact system capacity
and reliability. FCC technology offers a potentially cost-effective solution without degrading
system capacity and reliability.
A PSCAD model of a hypothetical 230kV substation is shown in Figure 2. An FCC, modeled
after the Zenergy Power HTS FCL, is installed in the feeder position with a bypass switch.
PSCAD simulation was performed with the following parameters: line voltage 230 kV, Ipeak =
173 kA, Isymm = 63 kA, K=173/63 = 2.75 and X/R = 55.06. The simulation results are given in
Figure 3. According to the simulation study, the prospective fault current, with FCC bypassed,
has 100 kA first peak and 63 kA (rms) fault levels. Figure 3 shows the effect of an FCC that trims
about 18 percent of the fault current. With such a device the fault current is controlled to a safer
55 kA level.
This simulation example provides insight into the required specifications of a transmission level
FCC. It is shown that a high-voltage FCC, i.e., 66 kV and up to 230 kV, with 10-15 percent
reduction of prospective fault current, is a feasible solution to the problem of excessive fault
current. If an FCC proves to be commercially successful, the lifetime of a range of substations
13

with the 63 kA existing CB limit can be extended. Keeping these substations operational for
another decade or two has significant economic implications.
Figure 2: PSCAD Model of Hypothetical 230 kV Substation with FCC Protection
RL RRL
#1 #2
1000.0 [MVA]
500 [kV] / 230.0 [kV]
IN
OUT
HTS_FCL
BRK
500/ 230kV
Transformers
14.75%Z
63kA
Breakers
#1 #2
1000.0 [MVA]
500 [kV] / 230.0 [kV]
#1 #2
1000.0 [MVA]
500 [kV] / 230.0 [kV]
#1 #2
1000.0 [MVA]
500 [kV] / 230.0 [kV]
#1 #2
1000.0 [MVA]
500 [kV] / 230.0 [kV]
BRK
BRK
BRK
BRK
BRKBRK
BRK
BRK
BRK
BRK
ABC->G
Timed
Fault
Logic
1750[MW] 500 [MVAR]
Vbus_an
BRK
NC NC NC NO
NC
NC
NC
NC
NC
NC
NC
NC
Vfcl_a
500kV
ACSource
Source: Zenergy Power plc
14

Figure 3: Comparison of PSCAD Simulation Waveforms of the Fault Current of a Hypothetical 230
kV Substation, with: No Protection Devices; Air Core Inductor; FCL.
(a) Top graph: instantaneous line currents. Bottom graph: rms currents.
(b) Expanded view of fault onset. Top Graph: Instantaneous line currents.
Bottom Graph: rms currents.
15

1.3 FCC Development Objectives
The objectives of this project were to develop and demonstrate reliable, cost-effective, and
environmentally acceptable technologies to equip the transmission grid to handle higher fault
currents. The goals of the research were to establish the desired criteria for fault current
controller performance and to test two different leading technologies against the criteria via
controlled laboratory testing and field demonstration in an SCE distribution circuit. The
research tasks were as follows:
1. Establish an FCC test plan including test criteria, test protocol, test site selection, test
scheduling, data collection plans, and interface requirements for both the laboratory
testing and field demonstration.
2. Develop two prototype FCCs, a saturable-core type developed by Zenergy Power plc,
and a solid-state type developed by the Electric Power Research Institute (EPRI) and
Silicon Power Corp. team.
3. Conduct laboratory testing of the two FCCs against the criteria, including pre-
connection high-voltage insulation tests, normal operation tests, and fault current
limiting tests.
4. Install the FCCs on the SCE distribution system and perform field demonstration for a
minimum of six months.
5. Evaluate the technologies on the basis of performance, respective strengths and
weaknesses, costs, reliability, installation, operation, and maintenance issues, and
potential for development to high-voltage design.
The intended outcome of this project is to facilitate improved, safe and reliable operation of the
power system by advancing FCC technology as a cost-effective and environmentally-preferred
option to circuit breaker upgrades or system reconfiguration, and by evaluating the potential of
this technology to mitigate fault current levels at higher voltages in the electric system through
real-world utility testing. The ultimate program goal is to enable the commercialization of FCC
technology for the benefit of California and the United States.
16

CHAPTER 2:
Fault Current Controller Technology Survey
This chapter presents an overview of past research activities, recent advances, and emerging
technologies of FCCs, and provides brief discussions about the operating principles and main
features of several representative FCCs based on solid-state and saturable-core concepts.
2.1 Solid-State Fault Current Controller Technologies
The advancement in high-power semiconductor technologies such as new thyristors, high-
power IGBTs and IGCTs, and emerging SiC devices, makes it possible to implement a
commercially viable FCC using solid-state electronic designs. In the following discussion, the
solid-state FCC topologies are classified into three major groups: the Series Switch, the Bridge
and the Resonant types.
2.1.1 Series Switch-type FCCs
The Series Switch-type FCC, illustrated in Figure 4, is composed of a bidirectional controlled
semiconductor switch, Sss, and a bypass network. The bidirectional switch may be implemented
with various semiconductor devices, whereas the bypass network can be a combination of
several parallel branches: the normal-state bypass, Sbp; fault current bypass, Zf; a zinc oxide
(ZnO) overvoltage protection bypass device; and a snubber network. Depending on the FCC’s
fault response algorithm, the bypass network can be a simple one or sophisticated one.
The normal-state bypass is usually implemented by means of an electromechanical switch, the
purpose of which is to provide a low-resistance conduction path in order to avoid the
semiconductor conduction losses and waveform distortion in the normal state.
Figure 4: Basic Topology of the Series Switch-type FCC
ZnO
Sss
Sbp
Zf
Snubber
Source: University of California – Irvine
17

The fault current bypass is employed by the non-interruptive types of FCCs to restrict the fault
current flowing in the power circuit and to allow existing protection schemes to take
appropriate action during a fault. The fault current bypass can be implemented with either
resistive or inductive components. The inductive bypass is the preferred solution due to its
reduced thermal management requirement. The interrupting FCCs do not need fault current
bypass. Some schemes simply turn off the switches to interrupt the current, while other designs
control the semiconductor switches to modulate the fault current and keep it within the
acceptable limits.
While the fault current bypass is optional, the overvoltage protection bypass is a must, which is
usually implemented by a high-voltage, high-power ZnO varistor (variable resistor), also called
an arrester. As the semiconductor switch is commanded to turn off, the varistor provides an
alternate current path, while limiting the voltage across the semiconductor switch and
absorbing some of the energy stored in the line inductance. In the case where the semiconductor
switch is a composite one, the varistors can also be placed across each individual semiconductor
device to limit the voltage across the series-connected devices and prevent overvoltage
breakdown due to turn-off or turn-on delays.
The snubber network is an important part of the solid-state FCC. The task of the snubber is to
limit the voltage rise across the switch at the turn-off instant and keep the dv/dt (rate of change
of voltage) below the maximum allowed value according to the manufacturer specifications.
The semiconductor switch in Figure 4 can be implemented by a number of solid-state
technologies, e.g., SCR [17], ETO [18], GTO [19], or IGBT [20], [21], [22].
2.1.2 Bridge-type FCCs
Bridge-type FCCs are implemented using a current-fed full bridge arrangement as shown by
the basic schematic in Figure 5(a). This topology is inherently well-suited to using diodes as
well as advanced thyristors as line-commutated switches. Bridge-type FCCs do not have a
normal state bypass, and may or may not have a fault current bypass, but they do need an
overvoltage protection bypass.
In the normal state all the bridge elements are “on,” providing an unrestricted conduction path
for the AC current, as shown in Figure 5(b). The bridge-type FCC relies on the insertion of a DC
current source in series with the line to limit the AC fault current, as illustrated in Figure 5(c). In
practice, a bridge-type FCC would likely use reactors to emulate the current source action. If the
bridge is implemented by thyristors or semiconductor-controlled switches, AC current
interruption or diversion to a fault current bypass is possible, as shown in Figure 5(d).
Figure 6(a) shows the schematic of an interesting rectifier bridge-type FCC, proposed by Boenig
and Paice [23] and patented in the early 1980s; this circuit has evolved into many derivatives
over the years.
Prior to start-up, the DC bias supply connected in series with L1 charges the inductor to a DC
current level that is higher than the expected AC peak current. All bridge diodes are in full
conduction and free-wheel the reactor current. Hence, the AC terminals of the bridge appear
18

“short circuited” and present a low insertion impedance. The AC line current splits equally
between the bridge legs. Depending on the line current polarity, one diode pair carries the
summation current ½(Idc+|iac|), and the other diode pair carries the differential current ½(Idc–
|iac|).
The equivalent circuits for the normal state and the fault condition state are illustrated in
Figures 6(b) and 6(c), respectively. When a downstream short-circuit fault occurs, the rising AC
fault current exceeds the DC current level. Consequently, the diode pair for the differential
current turns off when its current reaches zero. The diode pair for the summation current
remains on and steers the line current through reactor L1. Consequently, the fault current is
limited by L1 allowing the circuit breaker to take a protective action. Upon interruption, the
current in the DC reactor freewheels through the bridge and the energy absorbed during the
fault is then dissipated in the diodes and wiring resistance.
Figure 5: Bridge-Type FCC
(a) Basic Topology (b) Normal State
(c) Fault State (d) Thyristor Bypass
The advantage of this type of FCC is that neither controlled devices nor a control circuit are
needed. Available power diodes have rather high blocking voltage and current ratings. The
current rating of the diodes and limiting reactor L1 is dictated by the peak fault current. Since
the current limiting reactor L1 is on the DC side of the rectifier, during a fault condition the
inductor is subject to high DC voltage, which causes inductor current build-up and eventually
19

saturates the inductor. Once the inductor is saturated, the current increases sharply and the
FCC may lose its current-limiting capability. Therefore, this type of FCC cannot sustain the fault
condition indefinitely; rather, it limits the fault current build-up rate, helps to safely ride
through faults of short durations and, in the event of a major fault, it buys just enough time for
the circuit breaker to take action. Also, the diodes should be rated to withstand the full fault
current magnitude. A clear disadvantage of this FCC is the rather significant conduction losses
in normal state operation caused by the constantly flowing high DC current. To alleviate the
conduction losses Boenig and Paice [23] used a superconducting coil as a DC reactor. For this
reason the rectifier bridge FCC is sometimes referred to as a “superconducting” type; however,
the superconducting reactor is not a necessary feature in terms of the basic operating principle.
Figure 6: Single-reactor Rectifier Bridge FCC
Line
Load
D1
e(t)
L1
D2
D4D3
Vdc
(a) Circuit schematic
Line
Load
D1
e(t)
L1
D2
D4D3
I dc
(I - i )dc ac
1
2
(I + i )dc ac
1
2
(I + i )dc ac
1
2(I - i )dc ac
1
2
Vdci ac
(b) Equivalent circuit for the normal state
Line
Load
D1
e(t)
L1
D4
iac Vdc
i =Idcac
(c) Equivalent circuit for the faulted state
Source: Boenig and Paice [23]
Another example of a three-phase rectifier-type FCC is described by Nomura et al. [25] and
shown in Figure 7. This FCC uses a full-bridge three-phase diode rectifier with a
20

superconducting coil acting as a DC reactor at the secondary winding of the isolating
transformer. The transformer’s primary three-phase coils are inserted in series with the line to
provide the FCC function. The distinct feature of this FCC is the absence of a DC bias supply; it
relies on the rectified grid voltage of the secondary to charge the DC reactor. Otherwise, the
FCC operating principles are similar to that of the FCC previously described. The transformer
isolation makes this FCC suitable for higher voltage applications. Alternatively, diodes with
lower voltage ratings can be used.
Figure 7: Transformer-isolated Three-phase Rectifier Bridge FCC
D1
D2
D4
D3
L2
D6
D5
Ta
Tb
Tc
Idc
a
a'
b
c
b'
c'
Source: Nomura et al. [25]
A single-phase SCR bridge-type FCC is shown in Figure 8 (Liu, Jiang and Wu [26]). At start-up
the SCRs are gated to charge the L1 reactor with a DC current equal to or higher than a pre-set
peak line current. Alternatively, the charging may be accomplished by an auxiliary DC supply
connected across L1. All the SCRs are then gated to remain in full conduction. When a
downstream short-circuit fault occurs, the rising AC fault current reaches the DC current level.
Consequently, the SCR pair, which is carrying the differential current, turns off at zero current,
whereas the SCR pair carrying the summation current remains on and steers the line current
through L1 in order to limit the fault current. Meanwhile, the gate signals to SCRs T1 and T2 are
removed so that the conducting SCRs extinguish at the zero crossing of the fault current. The
gating of T3 and T4 remain on to provide a freewheeling path for the L1 current. In the next half
cycle, the fault current flows through the fault current bypass reactor L2.
Due to the fact that the fault bypass inductor is subject to an AC voltage, there is no current
build-up in the L2 reactor. Consequently, this type of FCC can endure rather long faults. The
fault current bypass reactor L2 should have a greater current rating than the DC reactor L1. The
latter should be designed to withstand only a half cycle of the line voltage. In practice the
21

current in the AC reactor L2 only lasts for a few cycles until the circuit breaker trips; therefore L2
can be relatively small. Also, since the fault current is diverted to the fault current bypass, the
switch current stress is lower. Thus, this kind of FCC generally exhibits high reliability. In
addition, since no cryogenics systems are required, both the initial cost and the maintenance
cost of this FCC are expected to be low.
The zinc oxide (ZnO) arrestor prevents over-voltages from developing across the bridge and
also provides a discharge path for the L2 reactor at the instant of circuit breaker opening.
Figure 8: Two-Reactor SCR Bridge-type FCC
Source: Lu, Jiang and Wu [26]
Many other approaches are reported with half-controlled IGCT switches [27], [28]; single switch
IGCT bridge [29], [30]; transformer isolated GTO bridge [31]; saturable DC reactor bridge [32];
GTO bridge with emergency power source function [33]; superconducting magnetic energy
storage (SMES) system [34]; and switched-resistance bridge [35], among others.
2.1.3 Some Practical Difficulties of the Bridge-type FCCs
In each of the circuits described above, there exist equivalent resistances Rdc and Rac
representing the losses of the AC and DC circuits, respectively. These are always present and
help dissipate the fault energy. The resistance Rac incurs a power loss only during the fault
period, whereas the resistance Rdc incurs the loss during the normal state operation as well as
during the fault period. Due to the high DC reactor current the DC power loss can be very
significant. This lowers the FCC normal-state efficiency and poses an additional thermal
management problem. By adopting a superconducting DC reactor, the DC losses can be
significantly reduced, but at the cost of a cryogenic system which requires periodic
maintenance.
Furthermore, for proper operation of the Bridge-type FCC, the DC inductor current has to be
maintained above a preset peak AC current. Thus, in practice, the inductor has to be recharged
in order to compensate for the current decay. Because of this, a self-charging approach is
22

preferred to one using an external DC supply. In this approach, as the differential current
ceases, one pair of devices turns off and the inductor is inserted into the line for recharging.
Since this happens near the peak of the line voltage, the instantaneous voltage drop across the
inductor “shaves the peak” of the line voltage and causes voltage distortion; the larger the DC
inductor, the greater the distortion. Moreover, with a large DC inductor, the Bridge type FCC
will have difficulty in accommodating a rapid load increase [35]. Figure 9 shows a simulation of
the line current, Iac, and the DC reactor current, Idc, during the transient and the steady state
conditions. The simulation was performed with the PSIM software package (by Powersim,
www.powersimtech.com) for the circuit shown in Figure 19 for the improved saturable-core
FCC with magnetic decoupling (Cvoric, de Haan and Ferreira [48]; see section 2.2.4) for 12 kV
line voltage and 1 kA load, summing DC reactor of 20 mH with 0.1 Ω stray resistance.
Recharging of the DC reactor introduces a clearly noticeable distortion. The design tradeoff
between the power quality in the normal state and fault current limiting performance can be
alleviated by introducing an AC by-pass inductor in addition to the DC reactor but with a
smaller value.
Better power quality can be obtained by using a DC bias supply, which pumps the DC reactor
current above the peak AC line current. This bias supply has to provide a rather significant DC
current at low voltage. Several proposals can be found in the literature for DC bias supply
implementation, e.g. [26], [36], [38].
Figure 9: AC Line and DC Reactor Currents for a Bridge-type FCC
2.1.4 Resonant-type FCCs
Instead of having distinct normal-state and fault current bypass elements, the resonant types of
FCCs use switches to reconfigure their networks either into the normal state (NS) or into the
fault condition (FC) sub-topologies (see Figure 10). These FCCs employ a series resonant circuit
(also called a resonant tank) as their normal-state bypass. To achieve near-zero series
23

impedance, the resonant circuit is fine-tuned to the line frequency. Under fault conditions, the
circuit is switched to the fault state sub-topology and is then out of resonance. Therefore, a
much higher impedance is presented to the line. Accordingly, resonant FCCs can reduce the
fault current but do not have interruption capability.
In practice, a bypass switch should be included in the installation in order to maintain service in
case of FCC malfunction, by allowing the device to be taken off-line for maintenance.
Several interesting FCCs based on resonant circuits and thyristor switches have been proposed,
as in [38]-[41]. The FCC topologies shown in Figure 11 are generally representative of the
circuits found in the literature sources. In the normal state the resonant circuit is tuned to the
line frequency, presenting negligible impedance to the line. In an event of a fault, the thyristors
are turned on either to introduce a short circuit or to connect an additional component to the
resonant circuit, which takes the circuit out of resonance. Subsequently, the fault current is
suppressed by the high impedance of the mistuned resonant circuit.
Figure 10: Topology of the Resonant-type FCC
NS
FC
CL
24

Figure 11: Types of Resonant-circuit FCCs
(a) (b)
(c)
Source: Lanes et al. [38]
The key advantages of a resonant FCC include better power quality and a desirable zero-
voltage or zero-current switching operation. Another advantage is that the switches are
activated in fault condition only and remain off in the normal state. This contributes to better
efficiency and reliability and reduces thermal management requirements. However, the
resonant circuit may impose higher peak voltages or current stresses on the semiconductor
devices. Fault transients also create over-voltages or voltage sags. Moreover, the inductor and
capacitor size and cost may be considerable, and require precise tuning.
Another example of FCC, proposed by Hojo et al. [42] (see Figure 12), consists of a series
inductor and a self-commutated voltage source inverter connected in series with the line by an
isolating transformer. (It should be noted that this FCC may also be classified as a resonant
type.) In the normal state, the inverter is operated in a capacitive mode to compensate for the
voltage drop across the series inductor. During a fault, the inverter is commanded to operate in
inductive mode in order to impede the fault current.
25

Figure 12: Inverter-based FCC
Source Load
Source: Hojo et al. [42]
Such FCCs can also be beneficially used in the normal state to improve the power quality of
downstream voltage by compensating for line voltage distortion and voltage sags [43], [44].
The disadvantage of this approach is that the inverter has to process high currents and must
operate at a relatively high frequency using pulse-width modulation (PWM) control. Also,
conduction and switching losses are considerable, and the inverter’s DC link capacitors have to
be of a relatively large value to handle the large amplitude currents while maintaining a
moderate DC bus voltage swing.
2.2 Saturable-core Fault Current Controller Technologies
Saturable-core FCCs exploit the nonlinear characteristics of ferromagnetic materials in order to
achieve a variable inductance. In essence, such an FCC is simply a coil of copper wire wound
around an iron core and connected in series with the electric circuit to be protected. In the
normal state, the iron core is saturated by a dedicated bias circuit, and the inductance of the
saturated coil and the resultant voltage drop across the coil terminals are both negligible. Thus,
the coil has no obvious effect on the electric circuit. However, in the event of a fault, the high
AC fault current drives the magnetic core out of saturation, and the resulting high inductance of
the coil limits the fault current. This basic principle of “passive” inductive switching can be
employed in a variety of ways to implement a saturable-core FCC. The various designs mainly
differ in terms of the core shape, magnetic circuit configuration, and core bias arrangement.
Many saturable-core FCCs also incorporate superconducting technologies and are frequently
referred to as superconducting devices. However, the key physical principle of a saturable-core
FCC is the inherent mechanism of impedance change due to magnetic core properties, whereas
the superconducting part is the assisting technology which aims to enhance performance, such
as higher current limiting capability, higher magnetic field intensity, reduced power loss, more
compact design, etc. Saturable-core FCCs can also be successfully designed using non-
superconducting components.
26

Saturable-core FCC technologies tend to exhibit physical and electrical characteristics that are
similar to other distribution equipment, such as transformers and power reactors. Particular
mention should be made of their ruggedness and inherent ability to quickly react to a fault
without any detection circuitry. Furthermore, a saturable-core FCC presents an inductive
impedance to the line during the faulted state. The absorbed energy is stored in the magnetic
field and mostly recycled back to the system, which reduces the power dissipation and can help
to alleviate thermal management issues.
Several feasible ideas for implementation of saturable-core FCCs have been proposed.
According to the literature, the major contributors to the cost of saturable-core FCCs are the
superconducting wire, the cryostat and the associated cooling system, and the materials costs of
the iron core and copper wire. It is a significant engineering challenge to minimize the cost of
the components in the design such that the FCC is an affordable and commercially viable device
for distribution and transmission systems. Another engineering challenge is to minimize the
voltage surge induced in the DC bias circuit during the fault.
In the following sections, several concepts of saturable-core FCCs are reviewed, considering the
aforementioned characteristics.
2.2.1 Basic Saturable-core FCC
A basic single-phase saturable-core FCC device was patented in the early 1980s as described by
Raju, Parton and Bartram [45]. The fundamental structure of this type of FCC is illustrated in
Figure 13.
Figure 13: Conceptual Structure of a Saturable-core FCC
aci
dcI
dcV
acL
dcL
acL
Φdc
Φac
dcL
Φdc
Φac aci
Source: Raju et al. [45]
This type of FCC is constructed around two iron cores, each core designed in what is termed an
“EE” shape, and with AC and DC coils wound around the center limbs. The AC coils are
conventional copper coils connected in series differentially and inserted into the AC line in
series with the protected load, as shown in Figure 14. The DC coils comprise the bias circuit,
wound on top of the AC coils and fed by a low-voltage, high-current DC power supply. With
superconducting (SC) wires, the DC coils impose zero resistance to the bias circuit. Hence, the
bias current is limited only by the internal resistance of the power supply. This allows
27

attainment of high DC current, generation of a strong magnetic field and efficient saturation of
both cores.
Figure 14: Connection of a Saturable-Core FCC
Source Load
Bias Supply
FCL
Ldc
acLacLaci
dcI
The principle of operation of this device is described by considering the magnetomotive force
(mmf)-flux relationships. By regulating the DC coil current, the mmf-flux operating point of
both cores is established in the vicinity of point B (see Figure 15). As the AC line current flows
through the AC coils, the core operating point is shifted. Due to the differential connection of
the AC coils, the AC-induced mmf in one core reinforces the DC mmf, whereas in the other core
the AC mmf weakens the DC mmf. Therefore, the operating point of one core shifts into a
deeper saturation (point A), while the operating point of the other core moves toward a shallow
saturation (point C), closer to the hysteresis knee point. In the normal state both cores remain in
saturation so that their combined inductance is low. When a fault occurs, the abnormal
amplitude of the fault current is capable of driving the core with the counteracting mmf out of
saturation into region C-D of the hysteresis curve. Here, the permeance (ability to conduct
magnetic flux) of the desaturated core sharply increases, resulting in a considerable increase in
coil inductance and, consequently, higher impedance, which limits the fault current. Depending
on the AC line current polarity, the coils alternate in and out of saturation to limit the current in
each half cycle.
The FCC inductance may be characterized as a function of the AC current amplitude, as shown
in Figure 16. The current threshold value, Iknee, corresponds to the operating point C, whereas
the value Imax corresponds to the operating point D. As the current amplitude is increased above
Imax the core enters the reverse saturation region to the left of D. As a result, at current levels
above Imax, FCC inductance falls back to the low value. Therefore, Iknee defines the highest
normal current amplitude above which the FCC current limiting commences, whereas Imax
defines the current level above which the current-limiting effect is lost.
28

Figure 15: Saturable-core FCC Operating Points
AB
mmf
C
D
0
Φ
Figure 16: Saturable-Core FCC Inductance as a Function of AC Current
mmf0
Φ
D
0
Z
I ac
maxZ
minZ
I knee I max
Un-Biased
Core
Positively
Biased Core
Negatively
Biased Core
Combined
Structure
C
D’
C’
-Iknee-Imax
The advantages of this approach include a simple and symmetrical core design, ruggedness and
fail-safe operation. However, the designer’s choice of two “E” cores per phase, each with coils
wound on the center limb, makes it difficult to implement a compact distribution-level three-
29

phase FCC, since it requires a battery of six cores, positioned sufficiently far apart from each
other to provide safe clearances. In addition to the drawback of the resultant bulky size, the
design of the cryogenics system is challenging. The original design used a quite bulky liquid
nitrogen cryostat to chill the superconducting DC coils. FCCs of higher power rating may have
larger dimensions so that each core may have to be provided with a dedicated cryostat and
superconducting DC coil. Another problem with this magnetic design is the transformer
coupling between the AC and DC coils in the faulted condition. The high voltage across the AC
coil may induce a large voltage in the DC coil and cause possible damage to the DC power
supply. As a protective measure, a resistor in series with the DC coil was used, at the cost of
higher losses in the bias circuit.
2.2.2 Saturable Reactor Limiter for Current
A “saturable reactor limiter (SRL) for current” as proposed by Oberbeck, Stanton and Stewart
[46] actually preceded that of Raju et al. [45]. The structure of a bipolar, single-phase SRL is
shown in Figure 17, constructed using a core with four limbs, the first of which has an air gap.
The DC bias coil is wound on the second limb. The AC coils are wound on the third and fourth
limbs and are connected in electrical opposition in series with the protective load. The current in
the DC bias coil induces a DC flux that drives the core into saturation. In the normal state, the
AC current is insufficient to desaturate the core so the AC coil inductance is low. In the fault
condition, the high fault current generates mmf of sufficient intensity to overcome the DC bias
mmf and desaturate the AC limb carrying counteracting mmf. The increased AC coil inductance
during the fault condition limits the fault current.
Figure 17: Structure of a Saturable Reactor Limiter (SLR)
aci
dcI
dcV
acLacL
ΦdcΦac
dcL
Φac
aci
gl
Source: Oberbeck et al. [46]
The idea behind this magnetic structure approach is that in the faulted state the AC flux
desaturates one limb and then flows through the gapped limb rather than through the strongly
saturated DC coil limb. In consequence, the AC flux is diverted from the DC coil. This results in
30

significant reduction of the magnetic coupling of the AC coil to the DC coil. Therefore, this type
of SRL may have a reduced voltage surge in the DC winding under fault conditions.
The major disadvantage, however, is the significant power loss of the bias circuit, since an
ordinary DC coil was used. The resulting DC coil temperature rise restricts the bias current level
and, accordingly, the attainable mmf. Furthermore, assuming that the AC coil limbs and the
gapped limb are of the same cross-section, in order to keep the AC limbs in deep saturation
with the same flux density, the cross-section of the horizontal segments, between the DC limb
and the AC limb, is required to be twice the cross-section of the AC limb; but the cross-section
of the DC limb must be about 2-3 times the cross-section of the AC limb. As a result, this design
of FCC has a complicated core structure that is difficult to build, and the achieved DC-AC coil
decoupling comes at the cost of additional core volume and weight as well as production
complexity and cost.
No suggestion was made by Oberbeck et al. [46] regarding how to construct a three-phase SRL.
One possible solution is simply to arrange three single-phase units together. Another solution
might be to extend the magnetic core so that two additional pairs of AC limbs could be fitted for
each phase; in this case, only a single DC bias coil is needed, but the core structure becomes
even more complicated.
Another disadvantage that can result from the asymmetry of the core is unequal Iknee, Imax and
Zmax values for the positive and negative current paths.
2.2.3 Magnetic-controlled Switcher-type Fault Current Controller
A single-phase FCC as shown in Figure 18 was proposed by Pan and Jiang [47]. This FCC was
constructed with EE cores with the AC coils wound on the outer limbs, which have a smaller
cross section than the rest of the core. The DC bias coil was wound on the central limb. Another
added feature is the inductor Ldc in series with the DC bias supply. Owing to these two features
the operating principle is somewhat different from the aforementioned FCCs.
31

Figure 18: Magnetic-controlled Switcher-type FCL
aci
dcI
dcV
dcN
acN
Φdc
Φac
aci
acN
Φac
dcL
Source: Pan and Jiang [47]
The FCC is connected to the line in the manner previously illustrated by Figure 14. In the
normal state, the outer limbs are biased to saturation, while the rest of the core remains in the
linear region. This takes place due to the lesser cross-section of the outer limbs. As a result of the
saturation of the outer limbs, the inductance of the AC coils and their resulting impedance is
very low so that the normal power flow is unobstructed.
In the faulted state, the large amplitude of the fault current drives one outer limb into deeper
saturation, while the other limb is desaturated. Consequently, a closed magnetic path with a
high permeance is established providing high magnetic coupling between the desaturated AC
coil and the DC coil. The operation of the magnetic structure therefore resembles that of a linear
transformer: the desaturated AC coil exhibits high inductance and starts acting as a transformer
primary, and, by virtue of the transformer action, the impedance of the DC circuit is reflected to
the line.
The advantage of selective saturation of the core segments is that the inductance of the DC coil
remains relatively large and requires lower bias current to keep the AC limbs saturated. This
feature allows an ordinary, i.e., non-superconducting, DC bias coil to be used. Thus, the
production and maintenance cost of the auxiliary support systems can be reduced. This
advantage is offset by the added volume, weight and cost of the core and the DC-side inductor
Ldc.
The fault impedance of this FCC design is determined by the impedance of the inductor Ldc,
reflected according to the turns-ratio of the AC and DC coils, in parallel with the inductance of
the desaturated AC coil. The high impedance introduced into the line helps limit the fault
current. However, since the reflected and magnetizing impedances appear in parallel, their
32

equivalent impedance is lowered. Therefore, the effectiveness of the method is somewhat
reduced, requiring a higher number of turns of the AC and DC coils. This results in a structure
with larger window area, larger core, and higher copper weight. In addition, the normal-state
impedance is higher and therefore causes a higher voltage drop across the FCC terminals.
Saturation of the FCC core requires high mmf, which is achieved by a sizeable number of turns
of the non-superconducting DC coil. Therefore, the turns ratio between the DC and AC coils is
high, and a very high voltage is induced in the DC coil under the fault condition. Proper
isolation is necessary for the DC coil, which further increases the thickness and volume of the
DC coil and consequently the window area of the core and the overall core size. Additional
concerns are related to Ldc: though the power supply current is limited by the DC inductor, Ldc,
a high voltage at the DC bias circuit poses a safety issue. Moreover, the DC bias supply has to
carry the sum of DC plus the reflected AC fault currents. The bottom line is that an effective
design for this type of FCC poses numerous engineering challenges.
2.2.4 Improved Saturable-core FCC with Magnetic Decoupling
The SRL proposed by Oberbeck et al. [46] was further improved by Cvoric, de Haan and
Ferreira [48] to achieve a better decoupling between the DC and AC coils. The key idea is to
interchange the positions of the DC and AC coils, as shown in Figure 19. The bias coil is wound
on an outer limb, whereas the AC coil is wound on a central limb. Otherwise the principle of
operation is identical to [46].
The advantage of this magnetic structure is that in the faulted state the AC magnetic flux flows
through the desaturated gapped limb, positioned further away from the DC coil. Thus, the
coupling between the AC and DC coils is weaker than that in the original structure of [46],
resulting in a greatly reduced voltage surge induced in the DC coil. Note that in the faulted
state, the permeance of the AC coil magnetic path is determined mainly by the air gap: the
shorter the air gap, the higher the permeance. The AC coil can then be constructed with a lower
number of turns, resulting in lower copper weight, lower normal state inductance, lower
voltage drop and reduced conduction losses.
33

Figure 19: Saturable-Core FCL with Magnetic Decoupling
aci
dcI
dcV
acL
Φdc
dcL
Φac
gl
Source: Cvoric et al. [48]
However, the need for an additional core limb, the asymmetric core structure, the additional
iron volume, and the overall cost remain the main disadvantages.
Another idea for implementing an FCC with reduced core volume was presented by Cvoric et
al. [49]. This single-phase FCC was constructed on a pair of EE cores having a deeply gapped
middle limb.
2.2.5 Saturable-core FCC with Parallel Permanent Magnet Bias
The FCC bias circuit as reported by Raju et al. [45] utilized a superconducting DC coil to allow
higher bias current and reduced DC bias losses. However, superconducting wire is a major
factor contributing to the overall FCC cost. The cryostat and DC bias supply both require
extensive maintenance and may adversely affect overall system reliability.
A permanent magnet (PM) bias circuit as proposed by Iwdiara et al. [50] and Liu et al. [51] can
potentially alleviate these problems. Several other derivatives of this idea have also been
reported in the literature. A proof-of-concept laboratory prototype was reported in [51] to verify
the idea. The PM biasing structure was installed as the center limb of the core structure, as
shown in Figure 20. Since the PM flux is split between two parallel branches, this bias
arrangement is referred to as “parallel PM bias.” The advantages of this approach include lower
cost, lower volume of the device, ruggedness, and higher reliability. The disadvantages of
parallel PM bias are the lower field intensity and PM demagnetization.
2.2.6 A Series-biased Permanent Magnet FCC
A single-phase 400V/63kA FCC using a permanent magnet (PM) for series biasing was
experimentally demonstrated by Rasolonjanahary et al. [53]. The basic structure of this type of
FCC is shown in Figure 21(a). In this design the PMs were inserted as segments of a “C” iron
core.
34

Figure 20: A Parallel-Biased Permanent Magnet FCC
aci
acL
Φdc
Φac
aci
acL
Φac
S
N
PM
Source: Liu et al. [51]
The principles of operation of such a device exploit the hysteresis property of the PM material.
In the normal state, due to the lower current amplitude, the magnetic field strength is decreased
before the flux density reaches Bs, so the operation follows the minor B-H loop within the
saturation region (see Figure 21(b), upper right quadrant). Therefore, in the normal state the
inductance of the saturated coil, Lac, is low. Regardless of the polarity, a high fault current in the
AC windings will cause large variations in magnetic field, following the major B-H loop.
Normal operation (following the minor B-H loop) therefore incurs low flux density variations
and low losses, whereas during faulted conditions greater flux variations occur and much
energy is dissipated when following the major loop. Thus, in the faulted state, the FCC’s
inductance and the equivalent series resistance increase considerably. The design challenge of
such an FCC is to minimize the losses and voltage drop across the FCC terminals in the normal
state, while attaining a large voltage drop under the fault condition.
Figure 21: A Series-Biased Permanent Magnet FCC
aci
acL
Φac
aci
acL
ΦacSN
PM
Φdc
S N
PM
Φdc
B
H0
s+B
s-B
(a) Physical Structure (b) B-H Operating Curve
Source: Rasolonjanahary et al. [53]
35

The main advantages of a series-biased PM FCC are the relative simplicity of construction and a
bias circuit that does not require a DC power source. The result is a more compact, lower
weight, all passive, and more reliable FCC that, potentially, can operate in harsh weather
conditions with minimum or no maintenance.
The disadvantages of the approach include higher normal-state voltage drop and power
dissipation due to the PM’s core hysteresis and eddy current losses. To minimize the eddy
currents the PM was constructed by stacking several insulated disks [53]. Another disadvantage
of this FCC is that it cannot limit the first peak fault current. This is because the initial fault
current surge may take the PM material from a point of normal saturation along the minor loop
towards the deepest saturation point of the major loop lying in the same quadrant of the B-H
curve (see Figure 21(b)). Along such a trajectory the relative permeability is low. As a result, the
FCC’s AC coil inductance remains low, and may not clip the first peak of the fault current. Also,
after a relatively long fault event, the PM may need some time to cool down before resuming a
normal state function.
Additional handling and mechanical construction problems will arise due to the natural
tendency of the PMs to repel each other and attract other metal objects.
2.2.7 A Three-Material Passive dI/dt Limiter
Another concept for a PM-biased FCC was proposed by Young, Dawson and Konrad [54], in
which only some parts of the core segments are saturated, as shown in Figure 22.
The core of the device consists of four AC coils connected in series. Each coil is wound on a low-
saturation-density core material segment and biased by a PM. Both high permeability and high
saturation flux density core segments are placed at the corners so as to provide a closed
magnetic path.
36

Figure 22: A Three-Material Passive dI/dt Limiter
aci
PM
S
NS
N
S
N
S
N
PM
PM
PM
aci
aci
aci
Source: Young et al. [54]
In the normal state, the PMs have sufficient intensity to saturate the low saturation density
segments. With cores saturated, the inductance of the coils is low. The high saturation flux
density material at the corners remains unsaturated, and it allows flux bending. As a result, the
flux is confined to the device and the flux leakage is minimized. This results in a more uniform
flux distribution in the saturated segments of the core. When a fault occurs, the larger current of
an appropriate polarity induces an opposing mmf that reduces the flux density in the core.
Consequently, the low saturation flux density segments are desaturated and the coil presents a
higher inductance to the electric circuit, limiting the fault current.
However, in view of the previously mentioned drawbacks of PM bias and a relatively large
number of components and the complexity of this design, this FCC structure appears as a less
attractive option compared to most others of its type.
Another PM-biased FCC exploiting the same concept, but with a simplified design, was
proposed by Young, Dawson, Iwahara and Yamada [55], as shown in Figure 23. Its “EE”-
shaped magnetic structure has a center limb made of a low saturation density material with a
PM inserted in the center of the limb. The rest of the core segments are composed of a high
saturation density material. The AC coil is wound on the center limb. With a reduced number of
components, this design can be constructed more easily, and is better suited for implementing a
practical device than the previously described device. However, it displays the common
weakness of the series PM bias FCCs in that the resultant mmf always passes through the PM,
causing elevated losses and demagnetization of the PM core.
37

Figure 23: A Three-material Magnetic Current Limiter
Source: Young et al. [55]
2.2.8 Saturated Open-Core FCC
The structure of a Saturated Open-Core FCC as proposed by Rozenshtein et al. [56] is shown in
Figure 24. In this design, the DC bias coil is wound on the narrow segment of a closed, strongly
elongated “CC”-shaped magnetic core. The AC coil is wound around the core so as to engulf
both elongated segments of the core. Owing to this magnetic design the core provides a closed
magnetic path for the DC bias flux; however, to the AC coil it appears as an open core.
The high-permeability closed magnetic path can be easily saturated by moderate-amplitude DC
bias current. Notice that the “go” and “return” DC fluxes are flowing in opposing directions
through the midst of the AC coil. Regardless of the AC current polarity, the induced AC flux
reinforces the DC flux in one segment and counteracts the DC flux in the other segment of the
core. In the normal state the AC flux is low, so all the segments of the core remain saturated and
the AC coil acts as an air core inductor with low inductance. In the event of a fault, the large AC
fault current desaturates the core segment carrying the counteracting flux. Thus, the AC coil
appears as having a high permeability open core and, consequently, presents a larger
inductance to the fault current.
A clear advantage of this FCC is that only one magnetic core and one AC coil per phase are
required. A three-phase FCC requires a bank of three cores, roughly half the size of the
previously mentioned FCCs. A single DC coil may serve all three cores. Furthermore, making
the DC coil superconducting helps improve the bias circuit performance and reduce losses.
Such an approach enables decreasing the volume, weight and cost of the FCC and allows for a
compact design with tunable limiting factors. Another important advantage of this FCC design
is the decreased coupling between AC and the DC coils, resulting from the orthogonal
arrangement of the AC and DC coils.
38

Figure 24: A Saturated Open-core FCC
aci
dcL
acL
Φdc
Φac
Φac
dcI
Φdc
Source: Rozenshtein et al. [56]
2.2.9 Bias Power Supply Issues
Low-voltage, high-current DC bias supply is required by most of the Saturated Core FCCs, the
exception being the PM-biased FCCs. The bias circuit can be either a simple unregulated or
phase-controlled rectifier bridge, or a switched-mode power supply. A clear advantage of the
regulated supplies is the option to preset for the anticipated fault current magnitude by
adjusting the bias current accordingly.
A switched-mode power supply is the preferred solution due to its much higher efficiency
compared to other approaches. An example of the switched-mode bias circuit is shown in
Figure 25. The circuit can be recognized as an inverted Buck chopper fed by a three-phase
rectifier, as proposed by Hong et al. [57]. Here, the superconducting DC coil is employed as the
buck inductor. The large DC coil inductance helps in attaining low bias current ripples. The
chopper is duty-cycle-controlled to provide precise regulation of the average DC coil current.
A serious concern, common to DC-biased saturated-core FCCs, is the voltage surge across the
DC bias coil (He et al. [58]). The surge may occur as the AC coil comes out of saturation and
establishes a linear operating regime in the core, thereby inducing a very high voltage surge in
the DC coil. Since the number of turns of the DC coil is greater than that of the AC coil, the
problem is quite serious and may result in damaging the electrical insulation of the DC coil,
damaging the bias supply itself, burning out the superconducting wire, or destroying the
cryostat.
One possible solution to the problem was proposed by Raju et al. [45] using an ordinary series
resistor. This, however, incurs a large DC power loss. A ZnO resistor installed in series with the
DC coil was proposed by Hong et al. [57], as shown in Figure 25. In the normal state the energy
released by the ZnO resistor is bypassed by a switch to help improve the efficiency of the bias
39

circuit. When a fault occurs, the bypass is opened and the DC coil current is diverted to the ZnO
arrestor, which absorbs the energy while keeping a constant voltage across the DC coil
terminals. This helps to protect the coil and bias supply circuitry from the induced voltage. The
disadvantage of this approach is that fast active control of the bypass switch is required.
Figure 25: DC Magnetization System for a 35 kV, 90 MVA Superconducting Saturated Iron-Core
FCL
Energy Release
C1
e(t) DS b
DC coil
Controller
Q
ZnO
Source: Hong et al. [57]
Another approach was proposed by Yin, Zhang and Gong [59], in which a low-turns, non-
superconducting suppressive winding, Laux, was wound on the DC bias limb and electrically
connected in parallel with the superconducting bias winding, Ldc, as shown in Figure 26(a). In
the normal state the DC bias current flows mainly through the superconducting winding due to
its zero resistance. Under fault conditions, however, the terminal voltage of the two parallel
windings is determined by the low-turn number suppressive winding so that the surge voltage
across the coil terminals is effectively reduced. The disadvantage of this approach is that though
the resulting terminal voltage is low, still, through the transformer-like action, the AC induced
voltage contributes to elevated current in the bias circuit to some degree.
Another interesting idea was proposed by Oberbeck et al. [46]. In order to protect the DC coil
from the induced AC flux, a shorted winding was added on the DC bias limb (see Figure 26(b)).
The shorted winding has no effect on DC operation; however, it reacts on the AC flux. A current
is established in the shorted winding which, by Lenz’s Law, creates a counteracting flux and
reduces the undesirable AC flux in the DC limb. This appears to be a good solution to the
problem. However, Oberbeck et al. [46] provide no report of experimental verification of the
effectiveness or limitations of the proposed method.
40

Figure 26: Parallel Overvoltage Suppression Winding Configurations
Φac
dcL
auxL
Φac
dcL
auxN
(a) Parallel (b) Shorted
Source: Hong et al. [57] Source: Oberbeck et al. [46]
2.3 Summary
This chapter surveyed solid-state and saturable-core Fault Current Controller technologies for
AC power systems and reported on recent research work done in this area. Notwithstanding
the fact that the research and development of FCCs has been going on for many years, diversity
of FCC concepts is still rather limited. Seemingly, much of the effort has been dedicated to
optimization of components, improving the efficiency and reducing the cost of a limited
number of basic ideas.
Experimental FCC systems described in this survey indicate continuing progress in this field.
However, a practical, efficient, reliable and economically feasible device, suited to utility needs,
has remained elusive.
The main challenge for Solid-State FCCs is related to the insufficient voltage and current ratings
of existing semiconductor devices, which are much lower than needed for operation in
transmission and distribution power systems. Thus, stacking of a number of solid-state modules
is typically required, which gives rise to voltage and current sharing issues.
The difficulties with the Saturable-Core FCCs are related to the magnetic coupling and, more
importantly, to the typically large size and high weight of the devices, which pose difficult
challenges to the commercial viability of these FCCs.
Nevertheless, innovation in FCC technologies is continuing. Two promising technologies, a
saturable-core, passive-switching type of FCC by Zenergy Power and a solid-state, active-
switching type by EPRI/Silicon Power, were the focus of this project and intensively
investigated for further development. The following chapters provide a detailed report on the
results of these investigations.
41

CHAPTER 3:
EPRI/Silicon Power Solid-State Current Limiter
Development
3.1 EPRI/Silicon Power Solid-State Current Limiter Concept
EPRI/Silicon Power based their FCC concept on a Solid-State Current Limiter (SSCL) design
using the Super GTO (SGTO) thyristor technology with a modular approach [60], [61]. A
Standard Building Block (SBB) was developed that can be stacked in series and in parallel to
attain the desired voltage and current ratings. The power train of a single SBB is illustrated in
Figure 27. SBB includes a main SGTO switch, auxiliary commutation circuit, fault bypass
current-limiting inductor, and an overvoltage bypass varistor/arrestor. The main SGTO switch
is composed of anti-parallel connected SGTOs. The commutation circuit consists of an auxiliary
SGTO switch and a series resonant commutation network.
Figure 27: SSCL Schematic
Source: Electric Power Research Institute
In the normal state the main GTO switches are gated “on” to carry the line current with only a
negligible voltage drop across the switches. In the event of a fault, the auxiliary circuit is
activated in order to provide low-stress turn-off of the main switches. The high switching speed
and superior di/dt capability of the auxiliary SGTO allows the commutation circuit to apply a
high-amplitude, short-duration resonant current pulse to the conducting main switch in order
to reduce the current of the main switch to zero at the turn-off instant. Thus, a reliable turn-off
of the main GTO switches is obtained which significantly reduces the switching losses of the
main switches. After the main switches turn off, the commutation capacitor resonates with the
line inductance and provides a soft dv/dt transient across the main switches. The switch voltage
is allowed to rise until it is clamped by the varistor at 4 kV. The varistor then sees the full fault
current. However, with 4 kV across the current-limiting inductor, only a few tenths of a
millisecond is sufficient for the inductor to take over the current, and the varistor becomes
42

inactive. Henceforth, the fault current is limited by the inserted inductor. PSCAD software was
used to simulate the SSCL. Principal waveforms of the simulated circuit are shown in Figure 28.
Figure 28: Simulated Waveforms of the SSCL Circuit
Source: Electric Power Research Institute
3.2 SSCL Design
3.2.1 Standard Building Block and Standard Power Stack
In order to facilitate manufacturing and to reduce the maintenance and ownership costs of the
unit, a modular approach with standard building blocks (SBB) is used. Figure 29 shows the SBB
design. One SBB contains the complete circuitry represented in Figure 27, including the main
switches, auxiliary switches, varistors and the current-limiting inductor, in addition to the
control boards. Each of these building blocks is a complete SSCL switch rated up to 2 kA and
about 5 kV blocking, resembling a fully functional SSCL. The number of series SBBs required is
determined by the breakdown voltage of the main switch modules and the arrestor voltage. The
current-limiting inductor is chosen based on the fault current requirement of a particular unit.
43

Ten SBBs will be connected in series to form a standard power stack (SPS) for a complete single-
phase unit rated for 15 kV and 1200 A. A conceptual view of an SPS with 10 SBBs in series is
shown in Figure 29. The inductor size chosen for the required 9 kA let-through current rating is
142 µH.
Figure 29: Modular Structure of the SSCL
(a) Standard Building Block (b) Standard Power Stack
Source: Electric Power Research Institute/Silicon Power Corp.
3.2.2 Final Assembly
The final design for manufacture of a complete distribution-level SSCL is as follows: Multiple
Standard Power Stacks are packaged into a three-phase unit as shown in Figure 30. The housing
encloses the power modules, current limiting reactors, bus bars and manifolds, assembled in a
tank that has cover-mounted primary and secondary bushings and a provision to connect
directly to the overhead utility power lines. The tank is made of 3/8” steel and will have
sufficient mechanical strength to withstand environmental conditions for the expected service
duration. The tank also includes accessories such as a liquid level gauge, liquid temperature
gauge, pressure vacuum gauges, pressure-relief device, control cabinet mounting, and cooling
radiators.
3.2.3 Thermal Design
The 15 kV, 1200 A SSCL is designed for outdoor substation use and follows the IEEE Standard
for General Requirements for Liquid-Immersed Distribution, Power, and Regulating
Transformers, C57.12.00-20061
. The overall package is similar to a standard substation
transformer tank with an external radiator bank. Immersing the system in a mineral oil
1
Since superseded by C57.12.00-2010.
44

dielectric fluid provides a high degree of electrical insulation. The radiator bank is connected to
a series of cold plates that are located directly underneath the power electronics devices. The
cold plates provide highly efficient, directional cooling with a relatively low coolant flow rate
and small pump size. The coolant chosen for the system is the same mineral oil as for the bulk
tank. This reduces the chance of incompatible fluids mixing. The tank design is such that all of
the power electronics will be immersed in Crosstrans 206 mineral oil as a dielectric. The radiator
bank provides cooling of the tank by use of forced-air cooling from external fans.
Figure 30: The SSCL Assembly and Its Components
3.2.4 Controls Design
The SSCL is designed to operate primarily from a remote operator interface. It also has the
ability to be operated locally. The controls can provide ON/OFF switching, equipment
protection, display and monitoring of operating parameters, a fault log, access protection, e-
Tagout, and safety padlocking, among others. The SSCL controls will also provide a trip-free
feature such that any “close” signal will not inhibit the SSCL from opening upon command.
The control hardware is divided into two groups as shown in Figure 31. First is the device level,
incorporating all device level controls like gate drives, sensing and device protection. Above
that is the system level master controller, consisting of all system level controls like display,
data acquisition, Local Operator Interface or human-machine interface (HMI), supervisory
controls, data storage and retrieval. The communication between the two tiers is via fiber optic
cables, which provide galvanic isolation and noise immunity; both are critical for proper
functioning of the controls. The system controller is built around a Schweitzer Engineering
Laboratories (SEL) controller. The selection of this particular controller technology is based on
the performance requirements, complexity, reliability, and cost.
45

The Schweitzer system controller is an off-the-shelf controller (see Figure 32), which provides
manual ON/OFF controls, equipment protection, display and monitoring of operating
parameters, local and remote interface and access protection. It also facilitates the IEC-61850
standard protocol communication. The monitoring functions are performed by using the inputs
from the standard CT and PT instrumentation in the substation.
3.2.5 SSCL Design Changes
The design of the SSCL evolved over the course of the project. Some of the design changes were
required for proper operation of the system. Others were to accommodate a particular mode of
operation or to use a specific mechanism to fulfill a given role. The main areas affected by these
design changes were the thermal management system and the control boards. These design
changes, and other material cost changes, led to an overrun in the estimated material budget for
construction of the SSCL. Details of these design changes are provided below.
Figure 31: SSCL Control Architecture
3.2.5.1 Thermal Management System
The original design for the SSCL envisioned the SGTO modules being mounted on heat sinks
with dielectric oil being forced over the fins of the heat sinks before returning to an external
radiator system to exchange heat with the ambient air. Simulation studies showed that in this
design the SGTO modules would have been operating at different temperatures and would
46

have experienced different thermal stresses at the two ends of the stack, a less than optimal
situation that could increase the probability of unforeseen failure modes.
The design was therefore adjusted to locate the SGTO modules on cold plates containing
internal serpentine paths for oil flow. Oil can be pumped into the individual cold plates to
remove the heat generated by the SGTO modules. With the new design the SGTO modules all
see the same temperature differential and each SGTO module experiences the same stress,
improving the reliability of the system.
Figure 32: SEL Industrial Panel-mount Controller
Source: Schweitzer Engineering Laboratories
The change to a cold-plate cooling system also reduces the volume of oil required for the closed
circuit cooling system and allows for the ability to direct cooling fluid from the manifold to the
current limiting inductors if desired.
3.2.5.2 Electrical and Control System
The original design concept used a single control board for the standard building block,
combining both power and controls. As the project progressed, the decision was made to place
the control and power functions on separate boards to provide better noise immunity. This in
turn led to custom designs for the microcontroller-based visual logic controller (VLC) board, the
floating power supply board and the capacitor trickle charge board. This separation allows a
less complex design for the individual boards. The floating power supply board components
now include EMCO™ high voltage power supplies, which are rare and expensive components.
The EMCO™ power supplies are required in order to maintain the charge on the commutation
capacitor. The VLC board communicates with the upstream FPGA (field-programmable gate
array) board via fiber optics, which also helps to improve noise immunity.
The high-speed current sensor design also underwent some modifications. The initial design
performed well in a certain subset of the measurement range, but the response exhibited an
undesirable non-linear behavior over other parts of the range. The design was modified to
improve the response across the entire measurement range.
47

The control cabinet design also evolved to accommodate the new control components. The
control cabinet now includes the FPGA board, the master SEL controller, the required control
power circuits and relays, fan and pump starters and fiber-optics interface board.
3.2.5.3 SGTO Modules
The SGTO modules required for the SSCL have overcome some manufacturing issues during
the course of the project. During the initial manufacturing stages, an unacceptably high number
of the SGTO wafers were becoming cracked or chipped. The problem was found to be the result
of the sawing process used to dice the silicon wafers and produce the individual SGTO die. The
saw was hitting small pieces of nickel on the top and bottom of the wafers. A layer of protective
material has been added to stop the nickel buildup; the cracks have been virtually eliminated,
and the wafer yield has improved to acceptable levels. The protective layer also had the
unexpected benefit of improving the voltage capability of the SGTOs.
The final design of the auxiliary module includes two diodes in series, to handle the expected
current and voltage conditions. There is the possibility of removing the second diode in the
future but this cannot be verified without further testing.
3.2.5.4 Material Cost Changes
The estimated cost to construct the new and improved SSCL design has significantly increased,
more than double over the original design. The principal reasons are changes to the functional
requirements of the unit for improved performance, and also increases in materials costs since
the original estimate was developed.
3.3 SBB Testing
Testing of the critical building block components such as the current limiting inductor has been
undertaken. The control boards and auxiliary power supply components have been tested in
the laboratory. The standard building block tests have verified the ability of the SGTO modules
to turn off in the required time and to block the necessary voltage. A series of tests was
conducted to demonstrate the maximum current that a single SGTO module could interrupt
(Figure 33). It was found that a single SGTO module was capable of interrupting over 6.5 kA in
a matter of microseconds. Exact time taken is a function of the auxiliary circuit, not the SGTO
module itself. It should be noted that the 6.5 kA per module is well in excess of the value
required for operation of the SSCL in the proposed design. A second series of tests measured
the current sharing between modules in the same standard building block. The current sharing
was observed to agree to within 5 percent.
3.4 Summary
The Solid-State Current Limiter (SSCL) design presented here is the result of pioneering work in
the high-power electronics equipment industry. The SSCL employs a modular and scalable
design, applicable to a range of voltage classes. Such an SSCL can provide a solution to the
rapidly rising available fault currents seen in utility systems. The advantages of the emerging
EPRI/Silicon Power SSCL approach are immediate recovery after a fault, no voltage or current
48

distortion, no cryogenics, low losses and reduced size and weight. In addition, this type of FCC
can be designed for interruptive capability as well as management of excess fault current.
Some major technical issues have been resolved and the detailed design of the system has been
completed. The design of the standard building block, control system, power stack and tank
were all finalized.
The main technical issues encountered have been related to the thermal management system
and the SGTO modules. The thermal management system design now features a forced oil
system, with oil passing through cold plates under the SGTO devices and cooled via an external
radiator system. This thermal management design has been verified by simulations performed
at the Novatherm Lab of Villanova University.
The results of the hardware design and elemental testing are promising and warrant further
research. The SGTO module and standard building block testing has demonstrated several
important functional requirements, including current interruption and module-to-module
current sharing. Control board testing has demonstrated the key algorithm functional
requirements.
49

Figure 33: Current Interruption Simulation Results for the Standard Building Block (SBB) Module
(a) Fastest Time (b) Largest Current
(c) Current sharing during fault operation.
50

A summary of the tests that have been completed so far is as follows: SGTO module current
interruption test; Standard Building Block topology demonstration; Standard Building Block
control boards and gate drive functionality test; Standard Building Block power supply test;
auxiliary power supply inverter and transformer test; and current-limiting inductor testing. The
tests have contributed to increased robustness of the design and have provided valuable
feedback to the improved performance of the system.
The scope of work for a continuation of this development effort would include the construction
of the Standard Building Blocks and the manufacture and testing of the SSCL. The possible
future steps for the 15 kV, 1200 A, three-phase SSCL could include SBB construction and testing
at Silicon Power, power stack assembly, assembly of full size SSCL, testing at an independent
high voltage lab and field demonstration at a utility site.
51

CHAPTER 4:
Zenergy Power First-generation HTS FCL
Development
4.1 Zenergy Power HTS FCL Concept
The Zenergy Power FCL is based on their High-Temperature Superconducting Fault Current
Limiter (HTS FCL) approach using the Saturable-Core Reactor concept, following the ideas
proposed by Raju et al. [45], Keilin et al. [62] and Xin [63]. The proposed HTS FCL is a three-
phase device. Each phase of the proposed HTS FCL is constructed around a dual iron core with
three coils wound around each limb, as shown in Figure 34. The DC bias coil is wound around
the combined center limb. A pair of AC coils is wound on the outer legs of the cores. The idea is
similar to that of [45], however, in this design, the AC coils are implemented using “CC” cores,
rather than “EE” cores. This approach has the advantage of requiring just a single DC bias coil
to be shared by the two cores.
Figure 34: Conceptual Structure of Zenergy Power Single-Phase Saturable-Core Reactor FCL
ΦdcΦdcΦac Φac aciaci
dcI
dcV
acLdcLacL
In similar fashion to the saturable-core reactor FCCs described in the previous chapter, the AC
coils are connected in series, differentially, and are inserted into the AC line in series with the
load as shown in Figure 14. The mmf-flux operating point of all the cores is established by the
DC coil in the vicinity of point B (see Figure 15). As the AC line current flows through the AC
coils, the core operating point is shifted. Due to the differential connection of the AC coils, the
AC-induced mmf in one core reinforces the DC mmf, whereas in the other core it weakens the
bias mmf. Therefore, the operating point of one core shifts into a deeper saturation (point A),
whereas the operating point of the other core moves towards shallow saturation, at point C,
closer to the hysteresis knee point. In the normal state the cores are designed to remain in
saturation so that they impose low impedance to the power system. When a fault occurs, the
52

high amplitude of the fault current drives one of the cores out of saturation into the steep region
of the hysteresis curve. The unsaturated AC coil is capable of supporting a large voltage across
its terminals, and thus limits the fault current. Depending on the AC line current polarity, the
coils alternate in and out of saturation to reduce the current in each half cycle.
4.2 First-Generation “Spider” HTS FCL Design and Prototyping
The Zenergy Power first generation prototype incorporates three pairs of single-phase FCLs in a
spider-like structure featuring a hexagonal iron core with a thick central limb as shown in
Figure 35(a). The three pairs of AC coils are wound on the outer legs of the cores using non-
superconducting conventional copper coils. Each pair of the AC coils are connected in series,
differentially, and inserted into the appropriate phase. The DC coil is made of first generation
High Temperature Superconducting (HTS) wire wound around the center limb. The HTS coil
forms a strong electromagnet fed by a low-voltage, high-current DC power supply. The HTS
coil provides an efficient method of saturating all six cores with minimum DC power losses.
It is worth noting that during normal operation the cores operate almost symmetrically, so that
the AC flux in the center leg is mostly canceled out. As a result, there is no AC voltage induced
in the DC coil. During the fault, however, the cores operate asymmetrically, with one or more of
the cores partially desaturated. Therefore, in a practical device, coupling between the AC and
DC coils can be a problem. However, the Zenergy Power design ensures that the coupling is
only a few percent and, therefore, of no significant consequence. This reduces the risk of high
voltage shock to the DC coil and power supply.
Figure 35: Three-phase Zenergy Power Saturable-core Reactor HTS FCL
(a) Physical Structure (b) Vertical View
53

Zenergy constructed a full-size, three-phase, distribution-level FCL prototype. The prototype
FCL employs cast-epoxy AC coils and a closed-loop cryogenic cooling system that uses sub-
cooled liquid nitrogen at approximately 68º K to increase the working current of the DC HTS
bias coil to maximize the DC magnetic bias flux. The basic design parameters of the prototype
FCL are shown in Table 1. The cutaway views of the distribution-level three-phase HTS FCL
assembly are shown in Figure 36. The view of the internal structure and physical view of the
experimental prototype HTS FCL unit are shown in Figure 37.
Figure 36: Cutaway Views of the Distribution-Level Three-Phase HTS FCL Assembly
The prototype FCL was designed for a modest fault current limiting capability and was
intended to limit a 23 kA (rms) potential steady-state fault current by 20 percent. During the
design, great emphasis was placed on accurately modeling and predicting the performance of
the FCL and its associated electrical waveforms, simulation analysis of different line conditions,
54

and predicting the interaction of the FCL with other power system components. Table 1
provides a list of the key electrical and mechanical characteristics of the prototype HTS FCL.
Figure 37: Experimental Zenergy Power HTS FCL Prototype
(a) Main Components
(b) Physical View: Outer Panels Temporarily Removed
55

Table 1: Design Parameters and Main Features of the Zenergy Power Prototype HTS FCL.
Line voltage 12.47 kV
Maximum load current 800 A (3-phase 60 Hz)
Voltage drop at max. load <1% (70 V rms)
Prospective fault current 23 kA rms symmetrical
Asymmetry (X/R) 21.6
Fault limiting capability 20%
Fault type 3 phase to ground
Fault duration 30 cycles
Recovery time Instantaneous
Iron Core Weight (lbs) 52,000
Cost of Iron @ $3.00/lb $156,000
Size 19' x 19'
4.3 HTS FCL Modeling
4.3.1 Nonlinear Inductance Model
Saturable-core HTS FCLs consist of a set of coils wound around one or more ferromagnetic
cores. A superconducting magnet is coupled to the core region in such a way that the DC
magnetization force can saturate the magnetic material. In the Zenergy Power FCL, each AC
phase consists of two coils connected in series and wound around two core regions. The
windings are oriented such that, in one core, positive AC current counteracts (bucks) the
superconducting DC bias, while in the other core, negative AC current assists (boosts) the
superconducting DC bias.
The nonlinear model that describes the behavior of the device is based on the physical principle
described in a previous chapter and was described by Zenergy in a paper presented at the 2010
IEEE PES Transmission and Distribution Conference and Exposition [64]. The core material has
a B-H curve that can be approximated by an inverse tangent function, as follows:
Here, iac is the instantaneous AC line current, and Imax is the line current that fully saturates the
cores, at which point the average magnetic field is Bsat, and parameter K determines the range of
(1)
56

line currents where the magnetic state of the cores are actively changing from saturated to
unsaturated. The equation is scaled in such a way that B = 0 at the bias point, i.e., with iac = 0,
and the majority of the change in field occurs just before iac = Imax. The induced voltage, or back
emf, across this section of the FCL is V= L
~
∂i/∂t, where L
~
is the differential inductance defined
as:
Here, nac is the number of turns in the AC coil, which carries the line current, and Acore is the
cross-sectional area of the paramagnetic core material. In some cases, this model is improved by
imposing two additional conditions. First, L
~
must be greater than or equal to an additional
parameter called Lair. This value represents the insertion impedance as an equivalent air-core
inductance. Furthermore, if iac is greater than Imax, then L
~
is constrained to be Lair. This accounts
for the fact that, when the line current is very large, the FCL’s magnetic core is reverse saturated
and the impedance is once again approximately equal to the insertion impedance of an
equivalent air-core inductor. The equations above provide a general framework for describing
the behavior of an FCL in an electric circuit. This framework requires four input parameters:
Imax, Bsat, K, and Lair. As described in the following sections, finite element modeling (FEM)
methods were used in simulations to calculate the average magnetic flux for various static
values of DC and line currents. A least-squares fitting procedure was used to determine the
above parameters. To validate the model, PSCAD electrical simulation software was used to
implement the FCL nonlinear inductance model and the simulation results were compared to
the experimental results from the extensive tests performed on the 15 kV FCL device.
4.3.2 Finite Element Model and Analysis
4.3.2.1 Finite Element Model
The finite element method was originally developed for structural analysis, and it has since
proven to be a versatile method of analysis and adapted to other fields, such as heat transfer
analysis and, in this case, magnetic field analysis. The objective of finite element analysis is to
find an approximate solution of a given boundary-value problem, by representing a region of
calculation by finite subdivisions where a spaced grid of nodes replaces a conduction region.
These nodes are the location where the solution is computed. The finite subdivisions are called
“finite elements” and they can have arbitrary shapes, resulting in a very versatile analysis
method. The collection of finite elements and nodes is called a finite element mesh. Continuous
variables, such as mechanical stress, temperature or magnetic flux, can be represented over the
finite element by a linear combination of polynomials called interpolation functions [80].
Precise finite element model (FEM) results require a number of inputs, including physical
geometry, accurate descriptions of the electromagnetic properties of each material, and the
current densities in the superconducting and AC coils. Calculations were performed using the
ANSYS FEM code, developed by Ozen Engineering, Inc. The geometric parameters were either
input directly from mechanical CAD drawings or by command files. Because the designs have
many elements, calculation speed is a significant constraint. In most cases the models make use
(2)
57

of geometric symmetries to improve the calculation speed. The B-H curve for the magnetic
material was estimated from measured data shown in Figure 38. All calculations were
performed under the static approximation. This assumption is justified by the fact that physical
devices do not display significant hysteretic behavior.
Figure 38: Magnetization Curve of M-6 Steel Used in the Finite Element Model
The LMATRIX subroutine of ANSYS provides the differential inductance matrix for the number
of coils present in the model, and the total flux linkage in each coil, under a given set of current
density conditions. The FEM program provides graphical outputs that allow easy visualization
of geometric properties (see Figure 39). The iron cores (gray) serve as flux links between the
superconducting magnet in the center of the device (blue) and the AC coils (orange). This three-
phase device has six AC coils in total: one boosting coil and one bucking coil for each phase at
any given time.
Graphical outputs also provide insight for future designs, allowing analysis of magnetic field
densities in each region (see Figure 40). After using the graphical outputs to verify the model is
working well, either the flux or the inductance outputs can be used to calculate the parameters
for the nonlinear inductor model.
4.3.2.2 Finite Element Model Analysis
FEM analysis allows modeling of the performance of devices prior to their construction. First, a
series of flux or inductance values over a range of AC and DC currents are calculated. The data
points shown in Figure 41 were calculated for a superconducting bias current of 100 amps, and
800 turns, resulting in 80,000 amp-turns. The process can be repeated for other superconducting
DC bias points if necessary. Figure 41(a) shows the average magnetic field linking the boosting
and the bucking coils. The values were calculated from flux values, Φ(iac), with the equation:
Bansys =Φ(iac)/nacAcore (3)
58

Figure 39: Finite Element Model of a 15 kV Three-Phase FCL
Figure 40: Total Flux Density Due to DC Magnetization at 80,000 Amp-Turns
In this case, nac = 20 turns and Acore = 300 cm2. Figure 41(b) shows the sum of these magnetic
fields. The inductance values shown in Figure 41(c) were calculated from Figure 41(b) with the
relation:
L
~
= nac Acore B(iac)/iac (4)
59

The dotted red lines in Figure 41(b) show the model described above applied to both the
boosting and bucking cores. A similar fit was obtained for the inductance values shown in
Figure 41(c). This inductance is related to the differential inductance by:
and was performed numerically to account for the conditional statements applied to our
differentially-defined inductance when iac > Imax and L ~< Lair. Next, the best-fit values from the
inductance fit were used to compare this model to the measured behavior of the FCL device.
Figure 41: FEM Simulated Results and Model Fitting
4.3.3 Experimental Model Validation
Short circuit tests of Zenergy’s 15 kV HTS FCL took place at Powertech Laboratories in Surrey,
British Columbia, Canada. The three-phase FCL was connected to the test source voltage as
shown in Figure 42, and auxiliary breakers (AUX) provided the fault current by closing all three
phases to ground. Source resistance and reactance were selected to provide the necessary
prospective fault current levels with the required asymmetry factor.
(5)
60

Figure 42: Zenergy HTS FCL High-Power Short-circuit Test Setup
Source: Zenergy Power plc/Powertech Laboratories
Figure 43 shows several tests performed with two cycles of load current followed two cycles
under faulted conditions. Figure 43(a) shows a 6.5 kV line-to-line system with two 830 A load
cycles followed by a 3kA fault; the symmetrical part of the fault current was limited by 17
percent. Figure 43(b) shows the same 6.5 kV line-to-line system with a 12.5 kA fault; the
symmetrical part of the fault current was limited by 30 percent in this case. Figure 43(c) shows a
13.1 kV line-to-ground system with two 780 A load cycles followed by a 20 kA fault; the
symmetrical part of the fault current was limited by 20 percent. All tests were performed with
an Xs/Rs ratio greater than 20. The symmetrical fault limiting performance of the model is
provided in Figures 44 and 45.
The experimental results were analyzed using a simplified lump-sum circuit that consists of a
voltage source, the nonlinear voltage FCL model, and a time-dependent load resistor. The
61

idealized source has the voltages listed above and a source impedance of 0.8 mH and 14 mΩ.
The FCL includes the current-dependent inductance described above, along with a resistance of
1 mΩ. The simplified model’s load impedance was chosen to generate the correct current during
the loading cycles. This impedance was switched to zero during the fault cycles.
Figure 43: Zenergy HTS FCL Short-Circuit Tests: Measured and Modeled Current as a Function of
Time
It was found that the nonlinear FCL voltage model quantitatively reproduced the measured
results. The FEM analysis provides a concrete method for estimating the model parameters for
the saturated-core FCL design. The estimated parameters were used for the modeled results
shown in Figures 43 and 44. Figure 44(a) shows the 12.5 kA symmetrical fault current after 21
cycles (350 ms) of continuous operation. At this point, the estimated parameters provide
excellent predictions for overall fault current reduction. The qualitative features of the FCL
voltage response are also estimated well, as shown in Figure 44(b). Figure 44(c) shows the
average magnetic field defined by the measured flux change (i.e., the integrated bushing-to-
bushing measured voltage with respect to time) divided by nacAcore.
While the qualitative features of the magnetic field are reproduced well by the model, the
quantitative agreement is not precise. In order to increase the accuracy of the model the input
parameters were adjusted to match the measured response. Figure 45 demonstrates the
62

qualitative agreement that is possible after the model updating, measuring the same parameters
as in Figure 44.
Figure 44: Measured vs. Modeled Parameters for the Zenergy HTS FCL
63

Figure 45: Measured vs. Updated Modeling Results for the Zenergy HTS FCL
4.4 Summary
This chapter reported on the recent development of a Saturable-core FCL by Zenergy Power plc.
Successful controlled and field testing of the FCL proved the technological feasibility of the
device. The FCL prototype was tested at Powertech Labs, and the revised prototype was tested
by the Southern California Edison Co. at their Westminster engineering laboratory, and
subsequently was installed in and underwent field testing at SCE’s Avanti Circuit of the Future
from March 2009 to October 2010. Although the design was still not optimized at this stage, the
FCL prototype operated successfully and provided an extensive learning experience for the
developer, Zenergy Power, for SCE’s engineers and technicians, and for UCI researchers.
This prototype FCL has the advantages of a passive design, the ability of limiting the first peak
fault current, automatic recovery under load, and fail-safe operation. Moreover, the DC bias
magnet showcased an excellent application of superconducting technology, where the
superconductor experienced no quenching during fault events, and thus no recovery time, for
control of multiple faults. In addition, the adiabatic operating regime of the HTS coil presents a
relatively small heat load to the cryocoolers.
State-of-the-art HTS technologies use liquid nitrogen to achieve high-temperature
superconductivity, and provide significant advantages over their low-temperature, liquid-
helium counterparts, since the liquid nitrogen systems are less complex compared to the liquid
helium systems, and offer significant savings in cryogenic equipment costs and a reduction in
operational costs. Furthermore, in the Zenergy design, only one HTS DC coil is needed, albeit a
64

relatively large-diameter one. An additional advantage of this FCL design is the placement of
the HTS coil on the center limb, which gives a sufficient electrostatic clearance to the high-
voltage AC coils wound on the outer limbs. The isolated high-voltage and cryogenic systems
lead to a simplified high-voltage design.
A systematic approach to modeling of the electrical behavior of the saturated-core HTS FCL
was also developed. The empirically-derived model was supported by FEM electromagnetic
simulations as well as circuit analysis, and then fine-tuned to match the experimental results.
The nonlinear FCL modeling, with the FEM-updated parameters, provided an excellent tool for
predicting the overall fault current reduction of a saturated-core HTS fault current limiter. The
FCL model was used by Zenergy in designing a superior “second-generation” 15 kV FCL with
the objective of developing a transmission-level FCL in the near term.
65

CHAPTER 5:
Zenergy Power HTS FCL Laboratory Testing
5.1 Standards
At the time of this writing, there were no official certification procedures or standards in place
for testing of FCLs. The testing procedure for the prototype FCL was developed under a joint
effort of NEETRAC, SCE, Zenergy Power, and the University of California – Irvine (UCI). The
procedure was derived from the existing IEEE Standards C57.16-1996 [76] and C57.12.01-2005
[77]. Zenergy Engineering Specification ZP/ES-2008-05 (Attachment I), comprises the test
protocol developed for this project; it has been submitted to the appropriate IEEE and CIGRE
Working Groups for their consideration in development of new or revised standards and
certification procedures covering FCC technologies.
The prototype was shipped to BC Hydro’s Powertech Laboratory in Surrey, BC, Canada for
laboratory testing, which was performed in three categories: basic pre-connection tests, normal
state performance, and fault condition testing. FCL testing criteria are summarized in Table 2.
Table 2: Zenergy FCL Testing Criteria
No. Test Reference
1 Winding Resistance IEEE Std. C57.16-1996
2 Impedance IEEE Std. C57.16-1996
3 Total Losses IEEE Std. C57.16-1996
4 Temperature Rise IEEE Std. C57.16-1996
5 Applied Voltage IEEE Std. C57.16-1996
6 Insulation Power Factor IEEE Std. C57.12.01-2005
7 Insulation Resistance Measurement IEEE Std. C57.12.01-2005
8 Fault Current Engineering Spec. ZP-ES-08-05
9 Turn-to-Turn IEEE Std. C57.12.01-2005
10 Lightning Impulse @ 110 KV IEEE Std. C57.12.01-2005
11 Chopped Wave Impulse IEEE Std. C57-12.01-2005
12 Audible Sound Engineering Spec. ZP-ES-08-05
13 Partial Discharge IEEE Std. C57.16-1996
14 Seismic Verification IEEE Std. 693-2005
Source: Zenergy Power plc/NEETRAC/SCE/UC–Irvine
66

5.2 Pre-connection Testing
A series of pre-connection dielectric and high-voltage tests were designed to verify the integrity
of insulation of the FCL device, bushings and supporting structures, and also to measure the
AC coil resistance.
During the pre-connection testing, the FCL prototype was exposed to a series of high voltage
lightning impulse tests. These were meant to ensure that the device could cope with anticipated
adverse conditions on the distribution grid, in a manner similar to other existing distribution
equipment.
The Partial Discharge test is designed to test the insulation of the device. To conduct the test
both terminals of each phase of the FCL were shorted. Then, each phase was connected to 11.3
kV line-to-ground (L-G) for 10 seconds. The voltage was then reduced to 9.5 kV, held for 60
seconds and then the partial discharge was measured. After the first partial discharge test, each
phase was energized at the applied potential test level of 34 kV for 60 seconds. The voltage was
then reduced to 9.5 kV, held for 60 seconds and then the partial discharge level was measured
again. Partial discharge levels at 9.5 kV fell below the recommended 100 pC (picoCoulomb)
value in the three phases; the FCL therefore passed the test.
The objective of the Lightning Full Impulse test is to simulate a traveling wave due to lightning
strikes. This test is a good indicator of quality of insulation, design and manufacture. With the
terminals of each phase under test connected together, each phase was subjected to one reduced
full wave and three full waves of positive polarity, with a crest voltage of 110 kV for the full
waves. Initially, the FCL had flashovers. Reconfiguration of the connecting cables and added
insulation sleeves resolved the problem and all three phases withstood the lightning impulse
tests.
The goal of the Chopped Wave Impulse test is to emulate a sudden flashover of line insulation.
A chopped wave imposes a high rate of change in voltage and generates oscillations which
result in high internal voltage stress. Thus, the Chopped Wave Impulse test is more challenging.
To conduct the test both terminals of each phase of the FCL were shorted. Then, high-voltage
impulses were applied to each phase of the FCL in the following order: one reduced full wave,
one full wave, one reduced chopped wave, two chopped waves, followed by two full waves,
with a crest voltage of 110 kV for the full waves and 120 kV for the chopped waves. Initially,
phases A and B passed the test, but phase C failed. The FCL passed the repeated testing after
reconfiguration of the connectors and additional insulation was added.
The goal of the Turn-to-Turn test is to detect inter-turn insulation breakdown in the copper
coils. The turn-to-turn test consisted of a series of high frequency, exponentially decaying
(“ringing”) voltages between the terminals of each winding, implemented by repeatedly
charging a capacitor and discharging it through sphere gaps into each terminal of the FCL,
which generated ringing. Other terminals of the FCL were grounded during the tests. One
reduced and three full-wave impulses of positive polarity were applied to each terminal of the
fault current limiter. The peak of the full waveform was 95 kV. The discharge waveforms were
67

oscillating at the same frequency and damping, which meant that no inter-turn fault occurred,
and the FCL had passed the test.
The purpose of the Applied Voltage test is to ensure that the device can withstand the required
power frequency voltage from the bushing to ground. This test also stresses the insulation in
between the different windings. During the test a potential test level of 34 kV was applied for 60
seconds on the supporting structure, including insulators. The FCL passed this test.
5.3 Normal State Performance Testing
The sequence of the normal state performance tests was designed to measure the steady-state
voltage drop, impedance, and power losses, and to verify the thermal design.
The FCL was connected to a three-phase bus voltage of 12.4 kV and loaded as shown in Figure
42. The test was conducted under a set of predetermined load currents. The load power factor
was close to 0.9. The voltage drop across each phase was measured and insertion impedance of
the FCL was found. The FCL passed this test with less than a 1 percent drop in the nominal
voltage.
In the normal state the voltage drop across the FCL terminals remains low, so the terminal
voltage distortion has a negligible effect on line currents and the power quality remains good.
As shown in Figure 46, the full load currents appear to be of satisfactory quality.
Figure 46: Experimental Waveform of FCL Full-load Current in the Normal State
The FCL insertion impedance and rms voltage drop, across a single phase of the experimental
prototype under different load current conditions and DC bias currents, are shown in Figures
47(a) and (b), respectively. Both the FCL insertion impedance and the voltage drop have a
nonlinear dependence on the load current and increase as the FCL prepares to assume the
current limiting function.
68

The FCL’s power losses are caused primarily by the AC coil resistance. Figure 48 shows the
normal state power losses of the FCL measured as a function of the load current.
The temperature rise test was conducted according to the test setup in Figure 42. Temperature
was measured on the AC coils with thermocouples on different parts of the AC coils. Load
current was increased until the nominal current of 750 A was reached. The test was conducted
for nearly 18 hours. The measured temperature of the hottest thermocouple on the AC coil
during the temperature rise test was compared to the analytical thermal model in Figure 49. The
thermal time constant was found to be equal to 3 hours. A final temperature rise of 35º C above
the ambient 22º C was recorded.
Figure 47: Measured FCL Impedance (Ω) and Voltage Drop (%) as a Function of Load Current
(a) FCL Impedance (Ω)
69

(b) Voltage Drop (%)
Figure 48: Measured Normal-State FCL Power Loss (%) as a Function of Load Current
70

Figure 49: Measured Temperature on AC Coil During Temperature Rise Test with 750 A (rms) Load
Current per Phase
The non-linear core characteristic makes the FCL a nonlinear device with sharply varying
insertion inductance, Xins. The insertion impedance (Zins = Rins + Xins) depends on the core
magnetic properties, number of turns in the AC coil, and the DC bias mmf. The experimentally-
measured impedance of a single pair of AC coils as a function of DC bias is shown in Figure 50.
The measurement was performed at low AC current level. The dashed arrow shows the
minimum DC bias mmf required to saturate the core and keep the FCL in the low impedance
state.
71

Figure 50: FCL Insertion Impedance as a Function of DC Bias MMF
The FCL insertion impedance Zins is a function of the AC current magnitude. Generally, the
insertion impedance increases with the magnitude of the AC current. Therefore, the DC bias
mmf was optimized to attain the largest insertion impedance change in response to the increase
in the AC current.
5.4 Fault Condition Testing
The objective of the fault condition testing was to evaluate the fault current limiting capabilities
of the FCL under laboratory conditions.
5.4.1 Single-fault Test
The prototype FCL was subjected to a series of faults with different prospective fault current
levels. Experimental results of a 23 kA prospective fault current are shown in Figure 51. Figure
51(a) shows the comparison of the prospective current magnitude with no FCL to the clipped
(i.e., limited by FCL) AC line currents during the fault experiment. Clearly, the FCL is able to
successfully limit the first peak of the fault current in addition to limiting the successive current
peaks. Figure 51(b) shows the expanded view of FCL terminal voltage during the fault.
A comparison of the predicted (by the FCL’s simulation model) to the measured percent of fault
current clipping versus prospective fault current level is shown in Figure 52. The prototype FCL
achieved 20 percent clipping of 20kA prospective fault current.
72

Figure 51: Experimental Results of Zenergy HTS FCL Clipping a 23 kA Fault
(a) Prospective vs. Clipped Fault Current
(b) Expanded View of Terminal Voltage
73

The measured bus voltages under fault condition with the FCL bypassed and with the FCL
active are shown in Figures 53(a) and (b), respectively. Comparison of the waveforms clearly
shows that due to the FCL’s protective clipping the upstream bus voltage sag was significantly
limited.
Figure 52: Actual vs. Simulated Fault Current Clipping (%) as a Function of Prospective Fault
Current Level
5.4.2 Double-fault Sequence
The prototype FCL also withstood a 20 kA, double-fault sequence lasting for 20 cycles within a
two-second time interval. The FCL succeeded in limiting both faults. Inspection of the
experimental FCL current and voltage waveforms, shown in Figures 54(a) and (b) respectively,
reveals that on clearance of the fault the prototype FCL had successfully recovered under load.
5.4.3 Endurance Test
The prototype FCL was also subjected to an endurance test and withstood a 20 kA extra-long
duration fault for 1.25 seconds, which is 82 cycles long. The FCL successfully clipped the fault
without damage to itself. The experimental waveforms of the fault current during the
endurance test are shown in Figure 55 and show no signs of abnormal activity.
5.4.4 AC to DC Coil Coupling Test
One of the major concerns in implementing the Saturable Core FCL is the AC voltage that may
be induced by the AC coil current in the DC coil. High voltage induced across the HTS DC coil
may, potentially, destroy both the DC coil and the bias power supply and, therefore, should be
controlled.
During normal operation the cores operate almost symmetrically, so that the AC flux in the
center leg is mostly canceled out. As a result, the AC voltage induced in the DC coil is
negligible. During the fault, however, the cores operate asymmetrically, with one of the cores at
74

times out of saturation. By proper geometry of the coils the coupling was minimized. Still, in a
practical device some residual coupling, about 5 percent, between the AC and DC coils was
observed as shown in Figure 56. As can be seen, higher DC bias results in lower coupling
between the AC and DC coils. This reduces the risk of high-voltage shock to the DC coil and
power supply.
Figure 53: Experimental Bus Voltage Sag Under a 20 kA Fault Condition
(a) FCL Bypassed
(b) FCL Active
75

Figure 54: FCL Current and Voltage Waveforms for a 20 kA Symmetric Double Fault with 2
Seconds Spacing
(a) Current
(b) Voltage
76

Figure 55: Experimental FCL Current During the 20 kA, 1.25-second, Extra-long Fault
Figure 56: Coupling of the AC-to-DC Coils in the Normal State and Under Fault Conditions
77

5.5 Summary
The prototype HTS FCL went through a series of rigorous laboratory tests in BC Hydro’s
Powertech Laboratory in Surrey, BC, Canada. A total of 65 separate tests were performed,
including 32 full-power fault tests with first-peak fault current levels up to 59 kA at the rated
voltage. Fault tests included individual fault events of 20 to 30 cycles duration, as well as
multiple fault events in rapid sequence (to simulate automatic re-closer operation) and
extended fault events of up to 82 cycles duration, simulating primary protection failure
scenarios.
The experimental FCL prototype passed all the pre-connection, normal state, and fault
condition testing procedures. The prototype FCL proved to have low normal-state insertion
impedance, was able to withstand real fault conditions, endured long duration faults, tolerated
the resulting magneto-mechanical forces and thermal stresses with no damage, and
demonstrated instantaneous recovery under load.
Successful laboratory testing of the prototype Zenergy Power FCL demonstrated, to the
satisfaction of both Zenergy Power and the host utility SCE, that the unit was ready to be
connected to a commercial electric utility circuit with commercial, industrial and residential
loads in the subsequent field trials at SCE’s Avanti Circuit of the Future.
78

CHAPTER 6:
Zenergy Power HTS FCL Field Testing
The experimental Zenergy Power HTS FCL prototype was installed by Southern California
Edison Co. (SCE) in the Avanti “Circuit of the Future” in San Bernardino, CA, and was in
operation from March 2009 until October 2010 for field testing and data collection. The Circuit
of the Future (COF) is a key part of SCE’s program for performance evaluation of new and
emerging grid technologies and systems, as well as for gaining experience with novel “smart
grid” equipment. As such, it was an ideal test bed for the field trial of FCC technology that was
the primary objective of this research project, and SCE was an enthusiastic partner and host
utility, as they were keenly interested in the evaluation and, hopefully, the commercial uses of
FCCs. The prototype FCL was integrated into SCE’s SCADA system and was operated in real
time, providing protection to the distribution circuit by limiting any fault currents that might
occur on the COF. This comprehensive field demonstration was the first-of-its-kind instance of
an HTS FCL being used in commercial service in the US electric grid.
6.1 Setup and Metering
6.1.1 Avanti “Circuit of the Future”
The Avanti Circuit, shown in Figure 57, is a 12.47 kV feeder serving residential, commercial and
light-industry customers in San Bernardino, California. The Avanti Circuit is specially
commissioned and instrumented to assess new “smart grid” technologies. The main feeder is
about 7 miles long and composed of about 20 percent overhead and 80 percent underground
construction.
The experimental FCL prototype (Figure 58) was installed inside the Shandin substation fence
line (Figure 59), immediately beyond the main feeder circuit breaker. The FCL was connected in
series between the feeder loads and the 12 kV side of the substation‘s transformer bank. The
FCL was also equipped with a bypass switch so that it could be taken off-line if needed.
The purpose of the installation was to demonstrate the FCL’s functions and gain operational
and maintenance experience under real-world conditions [66], [67]. The operating environment
at the substation is a severe, southwest US desert environment with hot, dry, dusty, high-wind
conditions. Summer daytime temperatures were frequently well in excess of 40° C (104° F).
Additionally, a residential neighborhood located nearby imposed a very restrictive ambient
noise requirement on the operation of the unit.
6.1.2 Metering
Figure 60 shows the schematics of the Shandin substation and installation of data recorders 1
and 2. Data recorder 1 is installed in the control room. Data recorder 2 is installed in the
immediate proximity of the Zenergy Power FCL.
Dranetz™ Power Xplorer PX5 data recorders (http://dranetz.com/portable/powerxplorer-px5)
were used, having the following basic characteristics:
79

• Data volume: 1 GB memory card (upgraded to 2 GB).
• Data rate: 256 samples per cycle (60 Hz line).
• Recording: Adjustable. Depends on setting; is triggered by current level.
• Available recording time: Adjustable. Depends on number and duration of recorded
events.
These recorders have four independent channels. Three channels were monitoring the line
currents and one channel was unused.
Figure 57: Diagram of the Avanti “Circuit of the Future”
Source: Southern California Edison Co.
80

Figure 58: Zenergy Power HTS FCL Prototype in Operation at SCE’s Avanti Circuit of the Future
Figure 59: Aerial View of Zenergy Power HTS FCL Installation in SCE’s Shandin Substation, San
Bernardino, California, March 2009
81

Figure 60: Location of the Data Recorders at Shandin Substation
Source: Southern California Edison Co./University of California–Irvine
6.2 FCL Key Technical Events
6.2.1 Pre-Installation Events
6.2.1.1 Controller Assembly and Software Testing – May 2008
This development task performed by Zenergy in preparation for the field prototype’s
implementation was a critical milestone in the field trial. The basic monitoring and control
software necessary to operate the FCL as designed was thoroughly tested and debugged in the
lab prior to sending the FCL to the demonstration site. From this initial testing, the most
important conclusions, derived in consultation with SCE’s Advanced Engineering Group, were
that redundant and fail-safe hardware and software designs as well as secure
telecommunications were necessary in order to ensure reliable operation with a minimum of
required maintenance interventions.
6.2.1.2 Seismic Analysis per IEEE Standard 693 – June 2008
Seismic modal analysis of the FCL unit was performed by Zenergy to determine any structural
requirements to withstand a 0.5 g zero-period acceleration, and to provide the necessary
information for SCE to design the anchorage system and the stress release for the electrical
connection according to the reaction forces. Finite element modeling and spectrum analysis up
to 40 Hz with 2 percent damping indicated that the iron core of the AC coils required extra
supports to limit stresses and deflections to within the allowable values, per IEEE Standard 693
[78].
82

SCE’s Civil Engineering Group was consulted to determine the best option for the physical
installation of the FCL. One option considered for the foundation was to use railroad ties over
the existing crushed gravel in the substation’s yard. Another alternative was to build a
traditional concrete foundation to ensure that the FCL would remain level in this seismically
active area. The civil engineering team’s recommendation was to set the FCL on compacted and
leveled crushed gravel without any additional support because the FCL’s enclosure was welded
to steel I-beam skids, providing adequate seismic rigidity.
6.2.1.3 FCL Assembled and Continuity and Temperature Rise Tests Performed – September
2008.
As there were no industry-accepted standards for FCL testing, NEETRAC, in close collaboration
with several of its member utilities, including SCE, cooperated with Zenergy Power to develop
a detailed FCL test protocol (Attachment I) based on IEEE and CIGRE standards for
transformers and reactors. The FCL was rated at 15 kV, 1,200 A, 110 kV BIL, and designed to
limit a prospective 23 kA symmetric fault by at least 20 percent, with less than 1 percent voltage
drop at maximum load current in the normal state. The FCL was first subjected to heat runs at a
load current of 750 A and full DC bias current to verify the maximum temperature rise of the
AC coils and the high voltage terminations. All measurements were within the limits specified
in IEEE Standard C57.16-1996 [75].
6.2.1.4 High-power and High-voltage Testing Completed at BC Hydro – September 2008.
Prior to high-power testing, continuity and insulation resistance measurements were performed
to verify that no movement of the device’s components occurred during shipping. Source and
load-side bushing terminations were measured separately. A comprehensive series of high-
power tests was then conducted at BC Hydro’s Powertech Laboratories. A total of 65 separate
tests were performed, including 32 full-power fault tests with first-peak fault current levels up
to 59 kA, all at rated voltage. Fault tests included: individual fault events of 20 to 30 cycles
duration; multiple fault events in rapid sequence to simulate operation of automatic circuit
reclosers; and extended fault events of up to 82 cycles duration to simulate primary protection
failure scenarios. The FCL’s measured performance agreed closely with the engineering
calculations in all cases.
The FCL was then tested under full lightning impulse tests at Powertech’s high voltage
laboratory. During the first round of tests, flashovers occurred between the HV jumper cables
connecting the AC coils. All jumpers were replaced and rearranged with more separation.
Other HV insulation enhancements made in order to pass the more severe chopped impulse
tests included: application of heat shrink insulation over the exposed terminations; larger lugs
and wall openings; molding mastic rubber to cavities, sharp objects, and connectors; and solid
connection of all ground leads to a common ground point.
6.2.1.5 High-voltage Testing Performed at SCE’s Westminster HV Lab – December 2008
High-voltage acceptance testing was also conducted by SCE at its laboratories in Westminster,
California, as follows:
• One reduced (1.2 x 50 μs) full wave – 50 percent or 55 kV peak wave
83

• One full (1.2 x 50 μs) wave – 100 percent or 110 kV peak wave
• One reduced chopped wave – 50 percent or 60 kV peak wave (chopped at 2 μs)
• Two full chopped waves – 100 percent or 120 kV peak waves
• Two full (1.2 x 50 μs) waves (within 10 minutes after the last chopped wave)
All the above tests, as well as the partial discharge tests, were successfully passed by the FCL
per IEEE Standard C57-12.01-2005 [77].
6.2.1.6 Cold-head Replacement – February 2009
One of the two cold heads of the helium refrigeration sub-system was replaced as a routine
preventive maintenance task that was due. From this experience it was learned that this
hardware needs better accessibility when performing maintenance in the field.
6.2.2 Post-Installation Events
6.2.2.1 FCL Commissioned on the Avanti Circuit of the Future – March 9, 2009
On-site commissioning tests performed by SCE engineers were successfully passed. These tests
consisted of coil resistance, power factor, and impedance measurements of all phases. These
measurements were performed as a safety and reliability check every time work was conducted
in the HV compartment to verify that the HV insulation had not been disturbed.
The change of the FCL’s insertion inductance as a function of the DC bias current was also
measured by SCE when the HTS DC magnet bias system was first energized and de-energized.
6.2.2.2 Damped Resonance Event Triggered by Controller Memory Overflow – March 16, 2009.
The FCL experienced a loss of DC bias current, caused by a reset of the Programmable
Automation Controller that was triggered by a RAM overflow. This shutdown occurred at a
time when the Avanti Circuit’s load current was approximately 120 A. The FCL remained in the
circuit for approximately 45-50 minutes, producing an initial 300 V line voltage rise, followed by
a 400 V line voltage drop on the 12 kV circuit bus. During this period, two large automatic
capacitor banks operated to regulate the voltage on the circuit according to their programmed
limits. The combination of the FCL’s high inductance due to the absence of DC bias current and
the high shunt capacitance caused an intermittent damped resonance condition. The FCL was
bypassed after 50 minutes and the circuit returned to normal operations. The FCL remained
bypassed and was operating in the standby mode until December 9, 2009, when an automatic
bypass was installed.
The event occurred as follows: At 10:10 am the DC bias current source shut off, initiating the
feeder’s voltage rise. At 10:13 am, three minutes into the event, the system voltage had risen
above the upper dead-band limit (12.45 kV) of the 1.8 MVAR capacitor bank. After 60 seconds
above the dead-band limit the 1.8 MVAR capacitor opened. Without a capacitor bank, the
system voltage dropped to about 11.65 kV at 10:14 am. At approximately 10:17 am the 1.2
MVAR capacitor bank switched in, due to the system voltage being below the lower limit of the
dead-band for three minutes. This action caused the feeder to re-enter a resonance condition. As
a result the feeder voltage increased to 12.4 kV, remaining there until the FCL was manually
84

bypassed at approximately 11:35 am. After the FCL was bypassed, the feeder voltage dropped
to approximately 12.2 kV, which was approximately the same voltage as before the event. There
were no negative consequences, but as a result of this event, SCE installed an automatic bypass
switch and set it in its automatic mode whenever the FCL was in service.
Due to delays in acquiring and installing the bypass switch, the FCL was maintained in a ready
condition, but was off-line (not installed in the circuit in active service) until December, 2009.
6.2.2.3 HVAC Shutdown Due to High Ambient Temperature – June 2009
An ambient temperature of 108° F was reached at the site in June 2009. The refrigeration
systems of the device had operated almost continuously since its installation. During a hot
summer week, the FCL experienced venting of the cryogenics fluid and required replenishment
of the liquid nitrogen due to a shutdown of the HVAC unit in the compressor compartment.
The solution was to upgrade the 3-ton HVAC to a 5-ton unit rated at 125° F ambient working
temperature. An extension to the existing compressor enclosure was added to provide sufficient
air flow to the three faces of the heat-exchanger coils. A shade structure for the 5-ton HVAC and
the compressor enclosure was also built to protect the FCL from direct sun exposure. No
changes were needed to the station’s power and light supply and there were no more HVAC
interruptions through the rest of the summer, which saw even hotter days than before.
6.2.2.4 Controller Upgraded, Including Watchdog Function – June 2009
During the outage taken to install the higher rated HVAC unit, the controller was also
upgraded. A more robust industrial-grade controller was installed with software and
communications to perform the monitoring and control functions being performed by the
original controller but with higher reliability to withstand the harsh environment found at
Shandin Substation.
At this time a watchdog function was also added at the request of SCE to bypass the FCL in the
event of loss of any of the critical control functions or loss of auxiliary and/or backup power.
6.2.2.5 Helium Leak Detected and Fixed – June 2009
A slow helium leak was detected by observing the pressure and temperature trends of the
cooling system. This did not cause any malfunction or shutdowns, but the leak was located and
corrective action was taken.
6.2.2.6 Automatic Bypass Switch Installed and FCL Returned to Service – December 9, 2009.
Once the automatic bypass switch was installed, the FCL was returned to active service on the
COF. Since the unit had been in off-line mode for 9 months, it was not possible to collect fault or
performance data on the FCL. The FCL not entered its official field trial and data collection
period.
6.2.2.7 Multiple-Fault Event – January 14, 2010
One month after being returned to active service, the FCL experienced its first in-service fault.
Measured data by SCE and Zenergy Power confirmed that the event evolved from a phase-to-
phase fault, to a three-phase fault, to a temporary recovery, to a phase-to-phase fault, to another
three-phase fault, and ended with a line opening and fault clearing. This multi-fault event
85

occurred over a three-second period. The sequence initiated when phases A and B of the
overhead section of the Avanti Circuit came together during high wind conditions. This phase-
to-phase fault lasted about 250 milliseconds, when Phase C also faulted, thus evolving into a
three-phase fault, lasting about half a second. The air recovered its dielectric strength and
extinguished the fault arc for about one second, but then the conductors of phases A and B
again flashed over to each other for about three quarters of a second. The arc was about to be
extinguished at this time, but again Phase C also flashed, resulting in a three-phase fault. The
event ended about a quarter of a second later when the Phase B conductor separated and
dropped to the ground, and the circuit’s protection cleared this solid ground fault.
The FCL limited the fault current throughout the various faults and sent the corresponding
monitoring and control signals as designed.
6.2.2.8 Nitrogen Pressure Stability Problem – February 2010
The cause of this problem was that the cold-head heaters’ power changed in steps that were too
large, causing overshoot in the system’s temperature control. As with the slow leak problem,
this was detected through long-term observations.
An improvement with finer resolution of the power increments to the heater elements and a
feedback loop stabilized the pressure and temperature and solved the problem.
This problem did not have any negative impact on the operation of the FCL, but the associated
engineering and maintenance needed to fix the problem inspired Zenergy to pursue the option
of using dry (i.e., non-superconducting) cooling systems in future FCL designs.
6.2.2.9 Shutdowns Due to Loss of Auxiliary Power from Transmission Fault Events – March
2010.
The FCL experienced three outages of substation auxiliary power due to faults on the high-
voltage side of the substation or on the transmission lines. In one of these events, a fault on the
115 kV high-voltage side caused a voltage dip and a two-minute outage on the 12 kV side of the
substation.
As required by SCE, the FCL’s controller immediately issued a bypass command, sent out alerts
to SCE and to Zenergy, and initiated an orderly shutdown of the DC bias current and cooling
systems. It sent the correct alarms and bypass commands to SCE. Upon resuming AC power,
the cooling system compressor was restarted and the cryogenic parameters returned to normal,
with the HTS coil ready to be re-energized.
The FCL is now equipped with an uninterruptible DC power supply, which allows it to ride
through these types of outages without downtime.
6.2.2.10 Shutdown Due to Shorted Terminal Block – August 2010
This event initiated tripping of the auxiliary power from the substation’s light and power panel.
The FCL’s controller again safely bypassed the unit, alerted SCE of the event, and shut down
the unit as designed.
86

The problem was caused by a loose connection at one of the supply cables on the terminal block
where the auxiliary power to the FCL is received. A higher rated terminal was installed and the
proper torque value to use when installing this terminal was included in the FCL installation
manual.
6.2.2.11 Decommissioning of the Zenergy HTS FCL – December 2010.
The Zenergy FCL successfully completed its intended period of field testing by October 30,
2010. Per the agreement with SCE, the unit was subsequently decommissioned, and removed
from service on December 16, 2010.
6.3 Analysis of Major Events
6.3.1 Loss of DC Bias Current, March 16, 2009
6.3.1.1 General Overview
A loss of DC bias event is a severe failure of the FCL. The event that occurred on March 16, 2009
was analyzed in detail in order to study the implication of such an event on the power system,
and was described in a paper presented at the 2010 IEEE PES Transmission and Distribution
Conference and Exposition, April 2010 [68].
The Avanti Circuit is a 12 kV distribution feeder. There are four shunt capacitor banks installed
on the Avanti Circuit for voltage support. At the time of the event, two of the four capacitor
banks were disabled. The capacitor banks remaining in service (see Figure 61) are rated at 1.2
MVAR and 1.8 MVAR. Both banks are connected in an ungrounded wye configuration and are
capable of switching automatically. Prior to the event the 1.8 MVAR bank was closed, and the
1.2 MVAR bank was open.
87

Figure 61: Diagram of the Avanti Circuit Prior to the March 16, 2009 Loss of DC Bias Event
The switchable capacitor banks operate on a time-biased voltage control. The switched capacitor
banks open or close if the measured voltage is outside of a programmed dead-band for a set
duration. The capacitor banks also were set to perform an emergency open or close if the
voltage is measured at 130 V or 110 V (on a 120 V base), respectively, for five seconds. The 1.8
MVAR capacitor bank was programmed to open on 124.5 V and close on 120.0 V, and had a
delay of 60 seconds. The 1.2 MVAR bank was programmed to open on 125.0 V and close at 121.0
V and had a delay of 180 seconds. It is important to note that the capacitor controls are set with
a five-minute lockout delay before a capacitor bank can perform another open or close
operation. Data was sampled at CB 1 and CB 2 where SEL-351 relays from the Avanti Circuit’s
advanced protection scheme were monitoring and reporting A-B phase voltage and current
every 5 seconds.
6.3.1.2 Description of the Event
Prior to DC bias current loss on March 16, 2009, the FCL was operating normally. The FCL’s
inductance is not a constant number even during normal operating conditions, but is a function
of instantaneous AC current.
As shown in Figure 62, at 10:09 am the DC bias current source shut off, initiating the feeder’s
voltage rise. Current in the DC coil took approximately 5 minutes to decay to trace levels.
Figure 62 shows the HTS coil’s DC current (blue) and its voltage (red) during the event. The coil
voltage dropped from 0.12 V (with predominantly resistive coil impedance) to -2.0 V
88

(predominantly inductive coil impedance), and decayed to zero after several minutes as the
HTS current also settled to zero.
Figure 62: HTS Coil Current and Coil Voltage for the March 16, 2009 Loss of DC Bias Event
At 10:13 am, four minutes into the event, the system voltage had risen above the 1.8 MVAR
capacitor bank’s upper dead-band limit (12.45 kV). After 60 seconds above the dead band limit,
the 1.8 MVAR capacitor opened. Without a capacitor bank, the system voltage dropped to about
11.65 kV at 10:14 am (see Figure 63(a), with time scale of 10 minutes before the event to 25
minutes after event onset, and Figure 63(b), with expanded time scale of the first 18 minutes of
the event). At approximately 10:17 am the 1.2 MVAR capacitor bank switched in due to the
system voltage being below the dead-band lower limit for three minutes. This action caused the
feeder to re-enter a resonance condition. As a result, the voltage increased to 12.4 kV, where it
remained until the FCL was manually bypassed at approximately 11:31 am. A loud humming
sound in the FCL was heard, due to the high magnetostriction (mechanical stresses and
vibration due to large magnetic flux variations) of the iron cores. After the FCL was bypassed,
the feeder voltage dropped to approximately 12.2 kV, which was approximately the same
voltage as before the event.
89

Figure 63: Line-to-line Voltage between A and B Phases at CB #1 for the March 16, 2009 Loss of
DC Bias Event
(a) 10 Minutes Before Event to 25 Minutes After
(b) Expanded Time View of Fault
Figure 64 shows the line-to-ground voltages, as measured by the three potential transformers
installed at the load side of the FCL. The HTS current (black) also appears for reference. Both
feeder voltage and feeder current as measured at the FCL agreed with SCE’s field data.
6.3.1.3 Observations
The measured bus voltage tended to increase due to resonance between the 1.8 MVAR capacitor
and the increasing FCL inductance of de-saturation due to loss of the DC bias in the FCL. Note
that the 1.8 MVAR capacitor bank switched out before the DC coil’s current decayed to trace
amounts. If the DC coil had decayed to trace amounts before the 1.8 MVAR capacitor bank
90

opened, simulations show the line voltage would have risen to 12.65 kV. Simulations also
showed that, under normal system conditions, the voltage rise due to the 1.8 MVAR and the 1.2
MVAR capacitor banks would be about 0.2 kV and 0.12 kV, respectively. When the 1.8 MVAR
bank opened, the resonance condition ceased and the system voltage dropped to 11.87 kV. This
voltage drop from the pre-event 12.2 kV was due to the FCL’s voltage-drop under de-saturating
conditions. This measurement allows comparison of the line voltage for the saturated FCL with
that for the de-saturating FCL without a masking effect due to resonance. Furthermore, the DC
current was still decreasing, causing the FCL’s voltage drop to increase, resulting in a further
decrease to 11.65 kV.
Figure 64: Line-to-ground Phase Voltages and HTS Coil Current for the March 16, 2009 Loss of DC
Bias Event.
6.3.1.4 Simulation Study of the Event
The PSCAD model shown in Figure 65 was adapted to simulate FCL interaction with the Avanti
Circuit during the DC bias current loss and the resulting resonance phenomenon. The event
was simulated by estimating the transient decay of the DC bias current and calculating the
transient increase of the FCL inductance. As the FCL’s DC bias current collapses, simulated FCL
inductance shows increasing swings riding upon a quasi-exponential rise, as shown in Figure
66. The steady increase in inductance occurs as the operating point of the FCL’s core swings
over to steep slopes in the B-H magnetization curve due to the DC bias current loss. The
increased FCL inductance creates a resonance condition with the 1.8 MVA capacitor bank
resulting in a rise in bus voltage.
Comparison of the simulated and measured line to ground voltages during the event is shown
in Figure 67. Good agreement of the simulated and measured results suggests that the Zenergy
91

Power FCL model is capable of closely reproducing the loss of DC bias current event and the
resonance condition that followed.
Figure 65: Simplified PSCAD Model of the FCL Installed in the Avanti Circuit
Figure 66: Simulated FCL Inductance for the Loss of DC Bias Event
6.3.1.5 Recommendations for Resonance Suppression
i. Design
In the design process, simulation of the resonance should be conducted, including capacitor
banks in the line where the FCL is to be installed at different loading conditions, to check for
extraordinary events in the controller’s logic. This entails an increased FCL inductance under
de-saturation conditions.
92

Figure 67: Comparison of Simulated vs. Measured Line-to-ground Voltage for the Loss of DC Bias
Event
ii. Operational
The ideal protection against resonance in this case is to immediately bypass the FCL. Given that
the event evolves relatively slowly, an automated switch that can operate on the order of
seconds is adequate.
Fixed capacitor banks should be carefully evaluated, and if they are integral to the FCL
installation, operational planning should account for their presence. This measure is
recommended because if the automated bypass switch should fail to close, the capacitor banks
could be disconnected automatically. A reactive compensation device, such as EPRI’s solid-state
STATCOM, should be a better alternative since it has the ability to regulate the circuit voltage
within the design range. Lastly, base loading of the circuit should be increased, if operating
conditions allow, because a more heavily loaded FCL is less likely to remain in a potential
resonance condition.
6.3.2 Downstream Short Circuit Fault – Jan 14, 2010
6.3.2.1 Description of the Event
Prior to the fault event on January 14, 2010, the FCL was operating normally. The pre–event
circuit load was approximately 100 A measured at the Avanti CB with system voltage
approximately 12.1 kV near the fault. High winds were reported in the area of the overhead-line
section of the Avanti Circuit, approximately 5.7 miles from Shandin Substation.
On January 14, 2010, at 9:58 am, a sequence of faults occurred on the Avanti feeder. The voltage
data recorded at the load side of the FCL is shown in Figure 68. The data suggest the following
scenario: The event evolved from a phase-to-phase fault, to a three-phase fault, to a temporary
recovery, to a phase-to-phase fault, to another three-phase fault, ending with protection systems
93

clearing the line and fault. This very unusual, mostly symmetrical, multi-fault event occurred
over a three-second period. Physically, the event was initiated by the overhead conductors of A
and B phases slapping together during high wind conditions near the end of the Avanti Circuit.
This phase-to-phase fault lasted about 250 milliseconds, when Phase C also faulted, thus
evolving into a three-phase fault, lasting about half a second. The air insulation recovered its
full dielectric strength for about one second, but then the conductors of phases A and B again
arced to each other for about three quarters of a second. The arc was about to be extinguished at
this time; however, Phase C also arced, becoming another three-phase fault. The event ended
about a quarter of a second later when Phase B conductor separated and dropped to the
ground, and the circuit was eventually cleared by system protection.
Figure 68: Voltages and Currents Measured Downstream of the FCL During the Multiphase Fault
Event of Jan. 14, 2010
6.3.2.2 Simulation Study of the Event
The FCL performance during the fault event was evaluated using the Zenergy Power FCL
PSCAD simulation model previously described. As a first step, the FCL model performance was
verified by comparison with realistic test results obtained during the FCL’s laboratory testing.
The comparison of both simulated and real fault current clipping vs. the prospective fault
current is shown in Figure 52. This comparison shows that the accuracy of the simulation model
is satisfactory.
94

As the next step, the FCL model was embedded into a PSCAD model of the Avanti Circuit to
perform a full-circuit simulation of the event. To resolve the difficulty of establishing the exact
arcing impedances during the fault, the fault scenario was approximated by several linear
segments of appropriate duration. Fault impedance for each segment was adjusted so as to
generate the average fault current within the segment. PSCAD simulation results of the Avanti
Circuit fault with the FCL bypassed are shown in Figure 69(a). Next, the FCL’s bypass switch
was opened and the FCL was introduced into the circuit, affecting the current flow. Since the
FCL is a highly nonlinear device, during the normal state its impedance is very low; however,
under fault conditions the FCL’s inductance increases sharply. PSCAD simulation results of the
fault event with the FCL in the circuit are shown in Figure 69(b).
Figure 69: PSCAD Simulation Results of the Multiphase Fault of Jan. 14, 2010
Main:Graphs
0.0 1.0 2.0 3.0 4.0 5.0
0.0
0.5k
1.0k
1.5k
2.0k
2.5k
3.0k
3.5k
y(kA)
I Avanti C I Avanti B I Avanti A
5.60
5.80
6.00
6.20
6.40
6.60
6.80
7.00
y(V)
Avanti CB VCN Avanti CB VBN Avanti CB VAN
5.00
5.25
5.50
5.75
6.00
6.25
6.50
6.75
7.00
7.25
y(V)
FCL VCN FCL VBN FCL VAN
Main:Graphs
0.0 1.0 2.0 3.0 4.0 5.0
0.0
0.5k
1.0k
1.5k
2.0k
2.5k
3.0k
y(kA)
I_Avanti_C I_Avanti_B I_Avanti_A
5.60
5.80
6.00
6.20
6.40
6.60
6.80
7.00
y(V)
Avanti CB VCN Avanti CB VBN Avanti CB VAN
5.00
5.25
5.50
5.75
6.00
6.25
6.50
6.75
7.00
7.25
y(V)
FCL_VCN FCL_VBN FCL_VAN
(a) FCL bypassed (b) FCL active
Comparison of the fault current levels for both cases, that is, with the FCL bypassed and with
the FCL active, is summarized in Table 3. Clearly, during this sequence of events, the FCL
operated as designed and limited the fault current. However, the prototype FCL was designed
for clipping fault current up to 20 kA. Since the fault impedance was relatively high the fault
current magnitude during the event was below 3 kA. For such current levels, the experimental
FCL provided partial clipping as designed.
Another valuable observation was made regarding how the cryogenics and bias circuit
responded during the fault. The measured performance is shown in Figure 70. The upper screen
95

shows HTS magnet current and voltage; the lower screen shows the line currents. The fault
instant can be recognized by the sudden decrease in load current due to the CB clearing the
affected branch. There was no noticeable change in the DC bias and it was not affected by the
fault current, and no measurable effects on the cryogenics system, cold-head temperature or
liquid nitrogen were observed.
Figure 70: Cryogenic System Behavior During the Jan. 14, 2010 Fault
Table 3: Comparison of Fault Currents – FCL Active vs. FCL Bypassed
Segment
Active
FCL Bypassed FCL
%
Clipping
F1 2554 2748 7.1%
F2 2782 2979 6.6%
F3 1568 1592 1.5%
F4 normal normal normal
F5 2643 2861 7.6%
F6 1968 2035 3.3%
F7 2782 2979 6.6%
96

6.4 Summary
The prototype “Spider” saturable-core HTS FCL was installed by SCE on the Avanti “Circuit of
the Future (COF)” in San Bernardino, California, for field testing in a “live,” actual utility
distribution feeder environment. The FCL was energized in March 2009, decommissioned in
October 2010, and removed from Shandin Substation in December 2010. The Avanti COF is a
dedicated 12.47 kV feeder serving actual residential, commercial and light-industrial customers
of SCE. The objective of the demonstration was to obtain service experience with state-of-the-art
fault current limiting equipment and operating procedures that could result in increased system
reliability and lower costs. The FCL was integrated into the utility’s SCADA system, and
operated in real time to provide fault current protection to this distribution circuit of the
California power grid.
The field demonstration provided a living laboratory to study the HTS FCL design, engineering,
and operation. The experience gained from this demonstration has been invaluable to the
evaluation and improvement of the FCL, contributing to a second-generation “compact” design,
commercial sales, and migration to the development of a transmission-level HTS FCL.
97

CHAPTER 7:
Development of Zenergy Power Second-generation
“Compact” HTS FCL
Despite the success of the “Spider” FCL design, the large size and weight of the air-dielectric,
closed-core FCL architecture relative to its electrical power rating challenges the economic value
proposition at distribution voltages. At the same time, the air-dielectric design complicates
scaling-up the device to higher voltages. In light of these issues, Zenergy Power conceived an
open-core FCL design that uses oil dielectric and conventional transformer design and
manufacturing techniques to create a much smaller FCL with greater fault limiting performance
and much higher voltage and power ratings. This new design is referred to as the “Compact”
FCL.
7.1 “Compact” FCL Concept
The architecture of a single-phase Compact FCL is shown in Figure 71. The FCL is built around
two I-shaped iron cores. The AC coils are wound in opposite polarity and are connected in
series with the load. Two DC coils are wound on top of the AC coils around both iron cores. The
DC coils are wound with first-generation HTS wire. The HTS coils form a strong magnetic field
fed by a low-voltage, high-current DC power supply. Utilization of the HTS magnet allows
reduction of the DC bias circuit resistance, which helps to achieve high bias current and high
magnetic field intensity, and also to reduce the losses in the DC bias circuit.
Figure 71: The Architecture of a Single-phase Saturated-Core “Compact” FCL
The principle of operation of the device is as follows: The two HTS bias coils are axially aligned
and constitute a Helmholtz coil, which generates an almost uniformly distributed magnetic
98

field in the cores with a higher magnetic field intensity than would be possible using only a
single DC magnet with the same amount of ampere-turns. By regulating the DC coil current, the
mmf-flux operating point of both cores is established in the vicinity of point B (see Figure 15),
which is located in the saturation region of the core’s B-H curve. As the AC line current flows
through the AC coils, the core operating point is shifted. Due to the counter-directional winding
of the AC coils, the AC induced mmf in one core reinforces the DC mmf, and weakens the DC
mmf in the other core. Therefore, the operating point of one core shifts into a deeper saturation
(Figure 15, point A), whereas the operating point of the other core moves towards a shallower
saturation (Figure 15, point C), closer to the hysteresis knee point. In the normal state, the shift
is designed to be small enough such that both cores remain in saturation; as a result their
combined impedance is low, with less than 1 percent voltage drop across the AC coils.
When a fault occurs, the abnormal amplitude of the fault current is capable of driving the core
with the counteracting mmf out of saturation and into the steep region C-D of the hysteresis
curve. Here, the permeance of the desaturated core sharply increases, resulting in a
considerable increase in coil inductance and, consequently, higher impedance, which helps to
limit the fault current. Depending on the AC line current polarity, the coils alternate in and out
of saturation to reduce the current in each half cycle.
The advantage of the proposed concept is that it has a rather wide potential operating range.
There are several variables that can be adjusted, such as the iron core length, core cross-section
area, number of AC coil turns, number of DC coil turns, and DC bias current. These parameters
can be tuned to obtain the desired steady-state impedance Zmin, the percent of fault current
reduction in the worst case, the faulted-state impedance Zmax, and the knee point current Iknee at
which the limiting effect commences.
A unique feature of the compact FCL is that there is no physical interface between the AC side
of the device (the line current and voltage in the circuit being protected) and the DC side of the
device (the HTS DC bias coil and its associated power supply). The AC coils and buswork are
located inside a dielectric tank, and the HTS DC bias magnets are placed in cryostats that
surround the tank.
Another important design innovation is the use of “dry-type” cryogenics to conductively cool
the HTS coil without the use of liquid nitrogen. Dry-type cooling allows the HTS coil to be
cooled below the nitrogen freezing point, thereby improving the performance of the HTS wire
and reducing the amount used in the magnets. The dry-type conduction cooling method also
avoids the problem of having large volumes of liquid cryogens in confined spaces, creating
potential pressure problems. Hence, maintenance is significantly simplified.
Section 7.3 below provides a more detailed description of the second-generation Zenergy Power
HTS FCL This device has voltage, current and fault current limiting ratings similar to the
“spider” HTS FCL, but is 75 percent smaller and only slightly heavier, representing a
considerable saving in size, cost and maintenance requirements for the same or better electrical
performance.
99

7.2 Comparison with Other Saturable Core FCL Designs
In the first generation spider-core FCL, each phase is constructed using dual CC iron cores as
shown in Figure 72(a). A pair of ordinary AC coils is wound on the outer legs of the cores. The
HTS DC bias coil is wound around the combined center limb. This has the advantage of
requiring only a single HTS DC coil for the two cores. Another advantage is the closed magnetic
path for both AC and DC coils, which facilitates relatively easy saturation of the core. A major
disadvantage of this approach is the large volume and weight of the core. Another
disadvantage is the large diameter of the HTS DC coil and, accordingly, the required length and
cost of the HTS wire.
The structure of a saturated open-core FCL [56] is shown in Figure 72(b). This FCL uses a
closed, strongly elongated “C” type magnetic core. This magnetic design provides a closed
magnetic path for the DC bias flux, but an open magnetic path to the AC flux.
Figure 72: Saturated-core FCL Architectures: (a) “Spider” FCL; (b) Open-core
(a) (b)
Source: Zenergy Power plc (Rozenshtein et al. [56])
A clear advantage of this type of FCL is that the iron core has low volume and weight, owing to
its narrow and compact shape. Moreover, the narrow diameter of the HTS DC coil results in
shorter HTS wire length and, accordingly, lower cost. Furthermore, lower HTS magnet volume
requires lower cooling power. Another important advantage of this FCL is the orthogonal
arrangement of the DC and AC coils, which minimizes the AC-to-DC coupling.
The most significant advantage of the Compact FCL as shown in Figure 71 is that the I-cores
have the lowest volume, weight and cost. Another major advantage is that the I-cores are well
suited for oil-filled dielectric design. Thus, the Compact FCL can be adopted for high-voltage
applications without over-sizing the device. The main disadvantage of the Compact FCL is the
low-permeability magnetic path, which requires stronger DC bias mmf to saturate the core.
Therefore, two HTS magnets of an appropriate diameter and two sizable cryostat units are
needed. Comparison of these three devices is summarized in Table 4.
100

7.3 Distribution-Level “Compact” FCL Prototyping
A view of a three-phase Compact Saturable-core Reactor prototype FCL assembly is shown in
Figure 73. A rectangular supporting frame design was used. The FCL is composed of six I-cores
combined in a single bundle. An AC coil is wound on each core. A pair of AC coils is needed for
each phase. In order to attain an electrostatic isolation sufficient for high-voltage applications,
the cores are housed in an oil-filled tank. SF6 gas may also be used as an insulating medium, just
as it typically is used in circuit breakers and compact switchgear applications. The tank is
placed inside the supporting frame to which the DC bias coils are also attached.
Table 4: Comparison of the “Spider,” Open-core and Compact FCLs
Spider FCL Open-core FCL Compact FCL
Core Volume Large Medium-Small Small
HTS coil diameter Large Small Large
Copper coil diameter Small Large Small
Cryostat Medium Small Large
Dielectric type Air Air Oil
101

Figure 73: Physical Structure of a Rectangular-frame, Three-Phase Compact FCL Prototype
The prototype Compact FCL’s main specifications are as follows: line-to-line voltage 12.47 kV;
line current 1.25 kA; prospective fault current 20 kA; line frequency 60 Hz; normal state voltage
drop 1 percent; and fault current reduction 25 percent. The FCL uses an oil-filled dielectric
design and cryogen-free, conduction-cooled HTS magnets, resulting in a reduced coil size.
A comparison of weight, size and cost between the first-generation Spider FCL and the second-
generation Compact FCL designs with similar electrical characteristics is given in Table 5.
Table 5: Comparison of the Main Features of the “Spider” and Compact FCL Designs
Spider FCL Compact FCL
Iron Core Weight, lb. 52,000 67,000
Cost of Iron @ $3.00/lb. $156,000 $201,000
FCL Size 19' x 19' 10' x 7'
102

As a learning experience, several additional three-phase and single-phase full-scale prototypes
for medium-voltage applications were built. The prototypes were constructed as modular units.
The supporting frame was designed to hold the Helmholtz HTS coils aligned at their position.
Together with the cryostat, the power supply and the control electronics, the structure formed a
standard test bed. One of the advantages of the proposed FCL structure is that there are no
physical interconnections between the AC coils and DC magnets. This allowed the tank with the
AC coils to be lifted and removed from the frame and replaced with a differently designed AC
coils module. By such an approach, the experimental FCL could be reconfigured as a single-
phase or three-phase device with a different set of parameters. Four full-scale Compact FCL
prototypes were designed, built and tested. The specifications of these FCLs are summarized in
Table 6.
Table 6 shows all four of the compact HTS FCL prototypes that were built and tested. These
prototypes have the same nominal 15 kV design voltage and 110 kV BIL rating, but differ in
their steady-state AC current ratings, targeted steady-state AC current insertion impedance, and
AC fault current limiting capability. The designed steady-state AC current levels ranged from
1,250 to 2,500 A (rms), and the targeted AC fault current reduction levels ranged from about 30
percent to more than 50 percent of a 25 kA potential steady-state fault current with an
asymmetry factor yielding a first-peak fault current approaching 50 kA.
The compact FCL prototypes underwent full-power load and fault testing at Powertech
Laboratories in July 2009 using essentially the same comprehensive test plan that was employed
for the Spider FCL. All 118 separate tests were performed on the four compact FCL prototypes,
including 55 calibration tests, 12 load current only tests, and 51 fault tests. In many cases, the
measured performance exceeded expectations, and the test program fully validated both the
performance potential of the compact FCL design and the efficacy of Zenergy Power’s design
protocol.
A particularly important result from the compact FCL testing program was the fact that the AC
coils and the DC HTS coil exhibited very little electromagnetic coupling. The DC current in the
HTS bias coil varied by only about 5 percent as the compact FCL was subjected to peak fault
currents up to 30 kA. These results were typical for all of the compact FCL devices during fault
current testing.
Table 6: Specifications of Prototype Compact FCLs
Parameter Units FCL #1 FCL #2 FCL #3 FCL #4
Line-to-Line Voltage kV 12.47 12.47 12.47 13.8
Number of Phases # 3 3 1 1
Line Frequency Hz 60 60 60 60
Prospective Fault Current kA 35 46 80 25
Limited Peak Fault kA 27 30 40 18
103

Parameter Units FCL #1 FCL #2 FCL #3 FCL #4
Prospective Fault Current (symmetrical) kA 20 20 40 11
Limited Symmetric Fault Current kA 15 11.5 18 6.5
Load Current Steady-State kA 1.25 1.25 1.25 2.5 – 4.0
Voltage Drop Steady-State Maximum % 1 1 1 2
Line-to-Ground Voltage kV 6.9 6.9 6.9 8.0
Asymmetry Factor # 1.2 1.6 1.4 1.6
Source Fault Impedance Ohms 0.346 0.346 0.173 0.724
Fault Reduction % 25 43 55 41
Also, the steady-state voltage drop of the compact FCL typically remained low with increasing
AC currents, and the AC power was very “clean” with total harmonic distortion (THD) levels
well within the requirements of IEEE Std. 519-1992 [79].
The rectangular design of the experimental FCLs was adopted for the reason that rectangular
shaped HTS magnets of an appropriate size as well as their cryostats were readily available and
allowed rapid prototyping. The rectangular shape, however, is not the best choice for a practical
device, because it requires more structural components. Considering the electromagnetic forces,
a better structural form for a coil is a round shape. Hence, circular HTS coils, a circular
supporting frame and a circular oil tank are more natural and easier to build. Therefore, a
prospective compact FCL having a circular design was conceived.
The FCL evolution path to commercial product is shown in Figure 74. On the left is the first-
generation Spider FCL that was tested, installed and operated in a commercially operating grid
to verify performance and reliability. In the middle is the rectangular prototype Compact FCL,
which was built and tested to validate a smaller and more efficient FCL. And on the right is the
round Compact FCL for distribution-class applications that takes advantage of all lessons
learned throughout the prototyping.
104

Figure 74: Path to Commercialization for FCL Prototypes
A perspective view of a circular frame design three-phase compact FCL is shown in Figure 75.
Such an FCL can be installed in a vault shielding it from the elements as well as from possible
acts of vandalism. The auxiliary equipment such as the cryocoolers, chillers, power supplies and
control electronics are installed in an adjacent control compartment having a safe access for
maintenance personnel. Figure 76 shows the assembled three-phase distribution-level compact
FCL.
Figure 75: Design and Layout of a Commercial 15kV-class Circular-frame Compact FCL Assembly
105

Figure 76: Three-phase Distribution-level, 15kV-class, Compact FCL Prototype with “Dry-cooled”
HTS Magnets Surrounding an Oil-filled Dielectric Tank with AC Coils
7.4 Transmission-Level Compact FCL Development
A perspective view of a transmission-level single-phase compact FCL is shown in Figure 77. A
high-voltage three-phase compact FCL can be built using a bank of three single-phase compact
FCLs and installed in a substation as shown in Figure 78. Possible specifications of a prospective
transmission-level compact FCL are summarized in Table 7. Such a device is feasible. Per a
preliminary design, a single-phase FCL could require four HTS magnets. The estimated device
dimensions are approximately 8.3 meters in length and 2.3 meters in diameter.
Zenergy Power has entered into an agreement with American Electric Power (AEP), Columbus,
Ohio, to partner for the demonstration of a 138 kV three-phase FCL as a part of Zenergy
Power’s ongoing DOE-sponsored FCL development program. A single-phase 138 kV FCL
prototype will be built and tested in 2011, and a three-phase FCL demonstration unit is slated to
be built, tested and installed in AEP’s Tidd substation located near Steubenville, Ohio in 2012.
This device’s specifications call for it to operate at 1.3 kA in the normal state, reduce an
approximate 20 kA prospective fault by 43 percent, and recover instantaneously under load.
The installation will be on the low side of the 138/345 kV transformer to protect the 138 kV
feeder.
106

7.5 Summary
This chapter reported on the recent development of a second-generation “Compact” saturable-
core FCL for medium-voltage distribution systems. This device is an evolution of the first-
generation Spider FCL.
Figure 77: Perspective View of a Transmission-level Single-phase Compact FCL
Figure 78: Perspective View of a Compact FCL Substation Installation
107

Table 7: Possible Characteristics of a Future Transmission-level Compact FCL
Parameter Units Value
Line-to-line Voltage kV 230
Line Frequency Hz 60
Rated Current (rms) kA 2
Prospective Fault Current as
Maximum Symmetric Fault Current (rms)
kA 63
Asymmetry Factor # 2.75
Desired Limited Current (rms) kA 52
Maximum Voltage Drop at Rated Current
and Line Voltage
kV
%
0.5
0.37
Prospective Peak Fault Current kA 173
Percent Fault Limiting Capability % 17.5
Line-to-ground Voltage kV 132.8
Cryostat ID m 1.34
Magnetic Core Volume m3 2.5
Magnetic Core Weight kg 19,964
The compact FCL has several advantages, notably: reasonably small core size, completely
passive design, the ability to limit first-peak fault current, sufficient fault current reduction,
automatic recovery, and fail-safe operation. The decoupled high-voltage and cryogenic systems
allow simplified oil-filled high-voltage design. This approach presents a relatively small heat
load to the HTS magnet; therefore, cryogen-free, low-maintenance, and commercial off-the-shelf
cryogenic equipment can be used.
The compact FCL prototype is the second superconducting FCL to be successfully tested for US
utilities (the first-generation Spider FCL being the first). The compact FCL prototype is an
advanced full-scale experimental model, which has helped to build confidence in the new
technology. While the development and prototyping of distribution-level FCLs proved
successful, the ultimate objective is to develop commercially feasible, full size transmission-
level FCLs.
108

Research and development of FCLs has been conducted for many years. The compact FCL
system described here is a promising, practical, efficient and economically feasible device that
meets utility needs. The emerging compact FCL is a viable candidate to make a breakthrough
into commercialization.
109

CHAPTER 8:
Lessons Learned
While the first-generation FCC demonstration program was a technical success, the objectives of
the program, specifically with respect to the electrical performance of the FCC devices, were
modest. The devices had a relatively low electrical rating, with maximum potential ratings on
the order of 25 MVA, while the high duty-cycle ratings are significantly lower; and the overall
costs, as might be expected for development prototypes, were high. The development and
demonstration results clearly point out that the FCC technologies must be expanded to higher
voltage and higher current capabilities in order to be commercially useful and cost-effective for
high-power applications.
The cost effectiveness can be addressed in two directions. First, FCC technologies are being
improved to reduce basic device manufacturing and maintenance costs. Second, the
technologies are being improved to work reliably at higher voltages and currents. The latter
consideration is arguably more important in terms of FCL technology cost-effectiveness and
value proposition, especially in the short term, given the acute and growing needs of US
utilities to mitigate the impacts of increasing fault currents on their systems, and the drawbacks
of conventional solutions.
Higher voltage substations can often be more crowded and have fewer options for expansion,
due to electrostatic clearance issues, and higher voltage components (switchgear, insulators,
transformers, buswork, etc.) are exponentially more expensive than their low-voltage
counterparts. Furthermore, it is becoming increasingly the case that large renewable power
generators desire to connect to the electric grid via the high-voltage lines and substations, and
high-voltage tie-lines have proliferated as more power is being wheeled from long distances
and as grid interconnections become more necessary to improve system reliability and to better
control power flows. Economic studies and performance models show that, at the current
performance levels and price points, FCC technologies can be very cost-effective compared to
major upgrade projects at high voltages.
The FCC demonstration project also clearly highlighted the shortcomings of the current Spider
FCL with respect to reliability. While the Spider FCL performed satisfactorily while in active
service in the grid, a large number of interventions were needed to keep the device completely
operational. The demonstration suggested that cryogenics are a key shortcoming, and it is clear
that cooling the HTS coil must be more reliable, less costly, and less intrusive. At a minimum,
the cryogenics need to target “5-nines” reliability (0.99999 availability) with only a single,
scheduled, short-duration annual maintenance outage. Even better would be extending the
maintenance interval to two years or longer.
If an FCC is to be relied upon for protection (which is essential to the economic value
proposition), it must be installed such that it can be bypassed and removed from service
temporarily for maintenance, while fault currents will remain acceptable. This may be
accomplished by splitting buses or otherwise configuring the electric grid in a manner that is
110

acceptable for short durations or at times of known reduced load, but which may not be optimal
for long-term operation.
If such an alternate grid configuration is not acceptable or available, then the FCL must be
designed, to the maximum extent possible, to allow essential routine maintenance, such as
cryogenic maintenance, to be performed while the unit is energized, through measures such as
locating the cryogenic coolers remote from the FCL itself. Also, the FCL must be designed to
remain in the protected circuit indefinitely in the failed state, assuming the resulting higher
voltage drop is tolerable, until replacement or repair can be arranged.
Ultimately, the FCC must be as reliable as a transformer of similar voltage class, and exhibit a
similar maintenance profile. Then, even though major outages will occur, they will be
infrequent and tolerable, as corresponding transformer failures are accommodated today. FCC
developers and manufacturers are taking these reliability issues to heart and are aggressively
modifying their FCC technologies for reduced maintenance requirements, and higher MTBF
and lower MTTR metrics. They are also working with key industry partners and suppliers such
as cryogenic cooler manufacturers to reduce the cost and improve the performance of enabling
technologies.
One important weakness identified during the field test was that the nitrogen cooling system
could be problematic for maintenance since it requires periodic maintenance of the cryostat,
including drying the volume of, or replacing, the liquid nitrogen coolant. This may not be
practical considering the fact that the unit is typically installed in a high-voltage area. However,
this weakness was resolved in subsequent FCC designs by replacing the nitrogen cooling
system with a liquid-cryogen-free, “dry-type” conductive cryogenic cooling system, which is
commercially available off-the-shelf, resulting in a more reliable and robust system.
Overall, the Zenergy HTS FCL is considered as an important successful FCC development
milestone, and it has already led to a scaled-up design for transmission-level applications at 138
kV. Further fast-paced and more advanced development of FCC technology to higher voltage
levels is strongly recommended. This project has demonstrated the benefits of FCL technology,
but has also shown that realizing those benefits will require accumulating significant
operational experience in the field with beta hardware that is essentially the same as eventual
series production items and extrapolating the FCC technology to considerably higher power
levels.
While the private sector will probably achieve these objectives on its own, this trajectory will
take too many years, most likely not until many US utilities have experienced considerable
negative impacts from increased fault currents that they must manage with undesirable,
conventional methods. Tremendous amounts of capital investment in general grid upgrades
will have to be made on a fast-track basis in order to maintain required levels of reliability.
Significant amounts of this investment are potentially avoidable and represent substantial value
to ratepayers, if suitable FCC technologies are available to California utilities, and if California
utilities are given reasonable incentives to deploy new technologies. The CEC and CPUC can
perform a valuable service for California ratepayers and workers, and the economy in general,
111

if an accelerated program of high-voltage, high-power FCC technology is implemented under a
consortium involving the CPUC, the CEC, California’s electric utilities, and private industry.
112

CHAPTER 9:
Potential Applications of FCC Technology
This chapter suggests several possible high-value applications of FCL technology in the power
transmission and distribution systems (see references [15] and [72]).
9.1 Fault Current Limiter in Substation Bus Tie
A substation bus tie (“coupling”) is used to connect two previously separate substation buses,
for the purpose of flexible and reliable operation. However, doing so can increase the available
fault current, possibly beyond the capacity of the existing breakers and protection schemes. For
this application, an FCL can be installed in the position as shown in Figure 79 [72]. In case of a
fault, the FCL reduces the peak short-circuit current at the first current rise. The advantages of
this application are: the parallel connection of the transformers (two systems) will result in an
even distribution of the currents supplied; reduction of the required short circuit capability of
the system; and reduction of the network impedance. No disconnection of the feeder
transformers is required after tripping (bypass) of the FCL.
Figure 79: FCL Installation in Substation Bus Tie Application
G
Fault Current Limiter
Embedded
Generation
Incoming
Feeder
Bus Tie
Source: CIGRE [72]
9.2 Fault Current Limiters in the Incoming Feeders
As an alternative to the previously described scheme, FCLs can be installed in each of the
incoming feeders as shown in Figure 80 [72]. In case of a fault, the FCL reduces the peak short-
circuit current at the first current rise. The advantages of this application are: the parallel
connection of the transformers (two systems) will result in an even distribution of the currents
supplied; reduction of the required short circuit capability of the system; reduction of the
network impedance; and possibly smaller and less expensive FCLs than one large device.
113

Figure 80: FCL Installed in Incoming Feeders
G
Embedded
Generation
Incoming
Feeder
Bus Tie
Source: CIGRE [72]
9.3 Fault Current Limiters in the Outgoing Feeders
Another approach is to install FCLs in the outgoing feeder positions as shown in Figure 81 [72].
In case of a fault, the FCL reduces the peak short-circuit current at the first current rise. The
main bus must be capable of carrying the full fault current. The advantages of this application
are: reduction of the required short circuit capability of the feeders on which the FCLs are
installed; and reduction of the network impedance.
9.4 Grid Coupling
In the scheme shown in Figure 82 [72], two subgrids are fed via transformers from the 220 kV
and 380 kV voltage level. The transformer capacity feeding into a 110 kV subgrid is
approximately 1000 MVA taking into account the n −1 reliability criterion commonly used in
utility applications. By connecting the grids with FCLs in this manner, the available surplus
transformer capacity can be utilized without exceeding the admissible fault current level. This
leads to the same technical benefits as in distribution levels, but with a considerable additional
economic benefit by elimination of a 110/380 kV transformer and the associated switchgear
components. In the specific case shown in Figure 82, this application removes the need for two
transformers: the first one because of the introduction of the FCLs and the second one because a
small increase in the loading capacity of the other transformers is permissible.
114

Figure 81: FCL Installation in Outgoing Feeders
G
Embedded
Generation
Incoming
Feeder
Bus Tie
Source: CIGRE [72]
Figure 82: Two Sub-Grids Coupled by an FCL
Source: Noe and Steurer [15]
9.5 Power Plant Auxiliary Power Feeder
Power plant auxiliary systems are usually connected to the power plant generator via a
transformer. In general this will be a location with high short-circuit capacity because of the
vicinity to the power plant generator. Hence, an FCL should be inserted at one or more of the
points as shown in Figure 83 [15].
The main advantage of the FCL in this application is that the energy of the fault arc is reduced
significantly, which means that additional measures to protect against arcing faults may no
longer be needed. Savings can also be obtained from reduced stresses on switchgear
115

components, mainly at the medium-voltage level, and from the possibility of using cables or
conductors with reduced cross section. Because of the fault current limitation it would no longer
be necessary to use separate cables for each phase, resulting in much faster and cost-effective
cable installations.
Figure 83: Location of FCLs in Power Plant Auxiliary Systems: (a), (b) Medium Voltage; (c) Low
Voltage
Source: Noe and Steurer [15]
116

CHAPTER 10:
Installation of the Zenergy Power HTS FCL
10.1 Site Preparation to Accommodate the FCL
As there were no industry-accepted standards for FCL testing, the National Electric Energy
Testing Research and Applications Center (NEETRAC) in close collaboration with several of its
member utilities, including SCE, developed a detailed FCL test program based on Institute of
Electrical and Electronics Engineers (IEEE) and Conseil International des Grands Réseaux
Électriques (CIGRE) standards for transformers and reactors.
The following measures were taken in order to test the FCL for acceptance by SCE and to
prepare for the FCL’s installation at the site.
10.1.1 Pre-installation High-Power and High-Voltage Testing
The FCL was first subjected to heat runs at full prospective normal load current and full DC
bias current to verify the maximum temperature rise of the AC coils and the HV terminations.
All measurements were within the limits specified in IEEE Standard C57.16-1996.
A total of 65 separate tests were performed at PowerTech Labs, Vancouver BC, including 32
full-power fault tests with first peak fault current levels up to 59 kA at rated voltage. Fault tests
included individual fault events of 20-30 cycles duration, as well as multiple fault events in
rapid sequence (to simulate automatic re-closer operation) and extended fault events of up to 82
cycles duration (to simulate primary protection failure scenarios).
The FCL was tested under full lightning impulse for:
o One reduced (1.2 x 50 μs) full wave – 50 percent or 55 kV peak wave
o One full (1.2 x 50 μs) wave – 100 percent or 110 kV peak wave
o One reduced chopped wave – 50 percent or 60 kV peak wave (chopped at 2 μs)
o Two full chopped waves – 100 percent or 120 kV peak waves
o Two full (1.2 x 50 μs) waves within 10 minutes after the last chopped wave
10.1.2 On-site Commissioning Test
Continuity and insulation resistance measurements were performed to verify that the device’s
internal components were not disrupted during shipping. Source-side and load-side bushing
terminations were measured separately.
Commissioning tests at the site consisted of coil resistance, power factor, and megger
measurements of all phases. These measurements were performed as a safety and reliability
check every time work was conducted in the HV compartment to verify that the HV insulation
had not been disrupted. The change of the FCL’s insertion inductance as a function of the DC
117

bias current was also measured when the cryogenic system was first energized and de-
energized.
10.1.3 Special Bypass Switch
To help assure service reliability, an innovative FCL bypass/isolation switch conceptualized by
SCE and custom built by a leading circuit breaker supplier was installed. The design includes a
three-phase pad-mount device with fixed input and output elbow connections for both line and
load sides of the circuit. The device has two sets of elbow connections for insertion of any three-
phase device (device input & device output). On the FCL itself there are three independent
switches which can be operated either manually or remotely to open and close the vacuum
bottles on the device-input and device-output connections, and a switch to control the bypass
(or parallel the device-input and device-output) connections, as depicted in Figure 84.
The purpose of the bypass switch is to easily and quickly reconfigure the electrical arrangement
consisting of the source transformer, the Avanti Circuit, and the FCL without the need to take a
circuit outage.
The bypass switch has redundant controls and can be programmed to operate either in
automatic mode, via remote control, or manually under local control (overriding the remote
control for safety reasons). Physically the bypass switch is similar to a pad-mounted gas switch.
Figure 84: Bypass Switch for Zenergy Power FCL Installation
By means of the bypass/isolation switch the FCL can be connected in series with the source
only, thus applying 12 kV to the FCL’s AC coils, but without any load current flowing through
the unit. A second arrangement connects the FCL in series with both the source and the load,
which is the normal mode. The third arrangement completely isolates the FCL from the SCE
system.
10.1.4 Civil Engineering
Seismic modal analysis of the FCL unit was performed by the manufacturer to determine any
structural requirements necessary to withstand a 0.5 g, zero-period acceleration. Finite element
118

modeling and spectrum analysis up to 40 Hz with 2 percent damping were performed to verify
that stresses and deflections were limited to within the allowable values, per IEEE Standard 693
[78].
SCE’s civil engineering group was consulted to determine the best option for the physical
installation of the FCL and to design the anchorage and stress release for the electrical
connection according to the estimated reaction forces.
One alternative considered for the foundation was to use railroad ties over the existing crushed
gravel in the substation’s yard. Another option was to build a traditional concrete foundation to
ensure that the FCL would remain level in seismically active area. The civil engineering team’s
recommendation was to set the FCL on compacted and leveled crushed gravel without any
additional support because the FCL’s enclosure is welded to steel I-beam skids.
10.1.5 High Voltage Connections and Grounding
The next element in the site preparation was to interface the high-voltage (HV) power
terminations allowing the FCL to be easily connected to the circuit via the bypass switch.
Elbow terminations were selected. The advantage here is that elbow connection at 12 KV would
reduce cost and allow dead-front installation as opposed to live-front installation, which would
entail costly civil engineering work to enable overhead installation.
Dead-break pre-molded cable connectors penetrating the FCL’s enclosure three feet above
ground level were installed for the 12 kV cable connections from the FCL to the bypass switch.
This facilitated the physical connection and disconnection of the FCL and also eliminated cable
splicing. The connector ratings are 25 kV class, 125 kV BIL, 900 A continuous, and 25 kA
momentary. Stainless steel flanges were specified for these connectors to reduce induced eddy
currents. The flanges and cable shields were grounded to the FCL’s common grounding circuit,
which in turn was connected to the substations ground mat.
10.1.6 Auxiliary Power
The SCE and Zenergy team worked together closely to define the auxiliary power needed to
supply the FCL’s electrical load. First, the most practical voltage to use was determined. Then
the total power requirement for the FCL’s peak load scenario was calculated. This reduced the
cost and additional time requirements should the device control voltage need to be at a non-
standard voltage value that is not typically supported at SCE substations.
The available voltage at the site was 120/240 volts, three-phase delta, and the FCL’s low-voltage
components were designed accordingly. To meet the FCL’s 100 kVA requirement, the site’s
station light and power were upgraded, allowing the FCL to operate without affecting the
substation’s normal load. A 100A, three-phase circuit breaker was installed as a dedicated
source for the FCL power. Six hundred feet of 2/0 copper conductor were pulled from the AC
panel to a connection box next to the FCL.
10.1.7 Noise Ordinance Compliance
A local city ordinance allows no more than 55 db total sound energy emitted during the
daytime and 45 db at night, as measured at the perimeter of the substation. Sound
119

measurements with the unit both on and off were performed. To comply with the stated limits,
the FCL was sound insulated to better contain the sound emission of the compressors and cold
heads. The FCL enclosure was also oriented so as to reduce the noise projected to the closest
neighbors.
10.1.8 Connection to High Voltage Side
The design of the HV connection, dependent on the voltage class of the FCL, should be carefully
discussed before the test unit is built for field demonstration. Initially Zenergy Power had
planned to have overhead potheads installed for the HV interconnection, but SCE provided
timely input to change to a better concept. SCE suggested that the input and output be
equipped with 15 kV-class elbow connections. The resulting advantage was that elbow
connection at 12 KV reduced the cost of installation and provided quicker and safer dead-front
connection and disconnection, which was done more frequently because of the development
nature of this project. In contrast, pothead/live-front installation would have necessitated
additional costly civil engineering work to enable a safe overhead installation and more time to
make the actual connections.
For future demonstration units the interconnection means should be carefully discussed with
the host site to determine the optimal connection method.
10.1.9 Metering and Interconnection Cabinet
This interconnection cabinet was not part of the original installation, but was added later for
monitoring purposes. The location where terminal blocks were added to provide 3-phase
output voltage was limited to the side bypass alarm cabinet.
It is highly desirable to have relay-class PTs and CTs installed on both the input and the output
of the FCL device, as illustrated in Figure 85. This makes possible the installation of digital fault
recorders and power quality (PQ) instrumentation on the input and the output of the device,
operating at the required bandwidths to monitor the expected transient events and to provide
stand-alone high-speed PQ monitoring. This arrangement would allow the utility to capture
fault profiles before and after the device operates, producing more accurate data retrieval for
analysis and validation of the unit’s operation.
The secondary of the monitoring voltage and current transformers will need to be in a weather-
resistant enclosure mounted on the outside of the FCL unit, giving unrestricted safe access to
utility personnel. An auxiliary 120 V AC outlet should also be made available in this cabinet.
120

Figure 85: Installation of Voltage and Current Monitors for the Zenergy Power HTS FCL
FCL
Current
Shorting
Block
A
B
C
Input side
Potential TB
A B C N
A B B CA C
Current
Shorting
Block
Output
Potential TB
A B C N
A B B CA C
For PQ monitoring
input side
For PQ monitoring
output side
10.1.10 Communication Interconnection Point
For alarms and possible remote control (not used by SCE in this project), there should be some
communication means to provide a secure and accessible point from which the utility can
receive alarms and data acquisition system signals, if needed by operations personnel. During
the FCL field test, one single alarm point to the control screen at the control center was
necessary. It was set up to provide an alarm to the operator in case of an FCL abnormal
condition.
10.1.11 Bypass Relay and Alarm Cabinet
As part of the operations requirements, the host utility at minimum should receive a single
bypass condition alarm to safely isolate the FCL device from service and generate other
appropriate alarms accordingly. This alarm should be in the form of a latching relay to maintain
the condition until an operator can investigate the event. The bypass switch can also be
automated to expedite isolation of the device during abnormal conditions.
121

CHAPTER 11:
Summary and Recommendations
11.1 Summary
The objective of this research project was to evaluate Fault Current Controller (FCC)
technologies, also called Fault Current Limiter (FCL) technologies, through targeted
development efforts and field trials at a California utility. Two different types of FCCs, one
solid-state type and one saturable-core type, were selected for the development and field
demonstration. After three years and 9 months of teamwork, the project is complete, from
conceptual designs, to prototyping, to laboratory tests, to field trials and data evaluation.
From the earliest stage of the FCC development work, an important guideline was to “design
for the test”: given that the primary project objective was to be able to perform a comprehensive
field test at a utility site, every aspect of prototype design was examined to ensure the ability to
monitor and collect the necessary data in the field installation. Equally as important, significant
efforts were dedicated to the development of the test plans that would be acceptable to SCE and
workable for both the developer and utility teams. Zenergy worked closely with NEETRAC and
several of its member utilities, including SCE, and developed a detailed FCC test program
based on IEEE and CIGRE standards for transformers and reactors (the most applicable
standards extant). EPRI/Silicon Power created their plan based on ANSI C39.09-1999 and ANSI
C37.06-2000 standards, covering the entire spectrum of possible tests that need to be carried out
on solid-state equipment, including component-level factory tests, system-level factory tests,
acceptance tests, and system field tests. Many technical issues related to laboratory test and field
test interface were resolved during a series of meetings with knowledgeable SCE engineers who
provided valuable design and test guidelines. Both Zenergy and EPRI teams developed
comprehensive test plans, which provided invaluable design and test guidelines and protocols
for the field trials.
11.2 Development of a Solid-State Current Limiter
The Solid-State Current Limiter (SSCL) was developed by EPRI/SiliconPower Corp. The SSCL
employs a modular and scalable design intended for 12 kV-class applications, but potentially
applicable to a range of voltage classes and fault current ratings. The advantages of the
emerging SSCL approach are: immediate recovery after a fault, no voltage or current distortion,
no cryogenics, low losses and reduced size and weight. However, voltage sharing to maintain
all the SBBs under the rated blocking voltage under dynamic conditions will likely be the key
technical challenge for transmission level applications where the voltage level is above 100 kV.
The major technical issues with the SSCL design have been resolved and the detailed
specifications of the system have been completed. One of the lessons learned from the first-
generation SSCL was that the auxiliary circuit was the cause of many timing problems and
control complications, requiring a multi-channel, high-voltage, isolated charging power supply
for acceptable operation. This increased the SSCL cost and did not provide the desired level of
122

operational reliability. Therefore, in designing the second-generation SSCL, EPRI/Silicon Power
opted to remove the auxiliary thyristors and resonant current-shaping inductor. Instead, a
simple capacitive snubber was fitted to soften the main SGTO hard turn-off.
According to design predictions, the system size is estimated to be about 6.5’x12’x12’ and
weight about 62,000 lb. with oil cooling. These specifications are considered to be acceptable for
a unit with the stated performance parameters.
The second-generation, single-phase SSCL was constructed under the scope of a subsequent
research effort, separate from this Energy Commission-funded project. As of this writing, the
results of the SSCL hardware design and elemental testing are encouraging, and the prototype
underwent testing at KEMA Powertest Laboratories, Chalfont, PA, in May 2011. The SSCL had
successfully passed the normal state testing at 15 kV and 1200 A, and successfully reduced a 23
kA prospective fault current to 9 kA. In the near future, construction of the full-size three-phase
15 kV, 1.2 kA SSCL for a 23 kA fault current rating will be initiated. This will include an
extension of the existing design by duplication of the already tested power stack. Some
additional tasks to be performed are to extend the control from a single-phase to a three-phase
system; include additional diagnostics; design a full-size three-phase thermal management
system; and build a full-size enclosure. Then the complete SSCL will be field tested by a host
utility.
The EPRI/Silicon Power development team identified some thermal management issues during
the simulation tests on their initial design. An improved system design was completed, and the
major technical design challenges, particularly an improved thermal management system and
more robust control architecture, have been resolved. However, the resulting increase in
projected costs to build the prototype exceeded the available project budget; thus this FCC
technology did not advance to the laboratory test and field demonstration stages under this
program. EPRI continued with their development efforts at a later date, with additional funding
from internal and other sources.
Notwithstanding the initial development setbacks, the EPRI/Silicon Power SSCL represents a
potential cost-effective solution to the rapidly rising available fault currents seen in utility
systems. One advantage of this type of FCC is the flexibility to be interrupting (i.e., act as a
solid-state circuit breaker in addition to limiting fault current levels) or non-interrupting (fault
current limiting only, coordinated with existing breakers and protection schemes). Furthermore,
the SSCL may be used to limit the current of high-temperature superconducting cables,
enabling the use of smaller cable sizes. It also has the unique capability of limiting inrush
currents, even for capacitive loads.
11.3 Development of a Saturable-core High Temperature
Superconducting Fault Current Limiter
The first-in-its-class 15 kV saturable-core type High Temperature Superconducting Fault
Current Limiter (HTS FCL) prototype was designed and built by Zenergy Power Inc. It
underwent successful laboratory testing, and was connected in the Southern California Edison
Co. (SCE) Avanti Circuit of the Future for field demonstration. The HTS FCL was energized on
123

March 9, 2009 and decommissioned October 2010 after successfully completing its field testing
period. This superconducting FCL was the first to be placed in commercial service in the United
States [81].
This report presents the design, testing, and application issues as well as a systematic approach
to modeling the electrical behavior of this type of fault current limiter. The derived computer
model provided an excellent tool for predicting the overall fault current reduction of a
saturated-core HTS fault current limiter. The model was useful in designing and operating an
innovative distribution-class 15 kV saturated-core FCL.
The important results of this project were the development of the simulation model, design
tools, operations and maintenance experience gained, and valuable protocols for the design,
installation, control, monitoring, and employment of FCLs in an electric power system.
Also Zenergy Power and SCE have gained valuable experience in:
• Designing, building, installing and operating a fault current limiter in the electrical grid
including integration with the SCADA system.
• Operating an HTS FCL in real-world conditions through all four seasons.
• Learning to address unplanned events, such as loss of station power and loss of FCL
bias current.
• Observing the performance of the FCL as it responded to real fault events.
• Understanding the requirements of preventive maintenance, improving maintenance
procedures and reducing their costs.
• Turning lessons learned into design improvements for future FCL devices.
The FCL performed well throughout the field demonstration, but several significant
modifications were made in collaboration with SCE. Significant observations from the
demonstration project include:
• The FCL’s enclosure environmental conditions must be maintained within a narrow
band to maintain the reliability of the control and monitoring instrumentation, cryogenic
refrigeration, and DC power supply equipment. Additional air conditioning was
installed in the FCL enclosure to meet hot-day cooling requirements.
• Air-cooled cryogenic refrigerators were satisfactorily employed (with sufficient
enclosure air conditioning), but water-cooled cryogenic refrigerators may be more
robust in the field.
• A sub-cooled liquid nitrogen re-condensing cryogenic system was successfully
employed. Zenergy Power’s current FCLs will use a “dry-type” conduction-cooling
cryogenic system free of liquid cryogens, which is more reliable and easier to maintain.
• The FCL control system was able to shut itself down in an orderly and safe sequence
under loss of station power events. Several substation blackouts were experienced
124

during the field trial period. The FCL is capable of automatically restarting when station
power is restored, including the restoration and stabilization of cryogenic refrigeration,
but this feature was disabled for safety purposes.
• A UPS was installed to provide control, instrumentation and DC bias current for up to
two hours during station outages.
• While multiple channels for monitoring the FCL’s systems and overall condition were
provided, SCE desired a “watch-dog” monitoring system that provided a “Go” (all
systems normal) or a “No-Go” (some system off-specification) with an automatic bypass
switch to remove the FCL from service in the event of a “No-Go” condition.
• The FCL control system was successfully integrated through a secure Ethernet
connection into the SCE SCADA system; a secure physically separated and independent
MODBUS control scheme enabled secure, real-time monitoring and operation from the
SCE Control Room.
• If the FCL is being relied upon for system protection, the system must be configurable
(through temporary buss-splitting, for example) to control fault duty levels in the event
the FCL must be removed for maintenance or repair.
• Saturated-core FCLs, seen as almost transparent inductors during steady-state
conditions and large inductors during fault conditions, must be evaluated for
compatibility with existing circuit resistance and capacitance values to ensure voltage
stability on the protected circuit (see Zenergy’s 2010 IEEE PES T&D paper [68]).
• The effects of saturated-core FCLs on circuit breaker transient recovery voltage must be
considered to assure adequate rating of circuit breakers and determination of maximum
allowable interruption currents (see Zenergy’s paper presented at the 21st International
Conference on Electricity Distribution (CIRED) [66]).
The research in this period has led to the successful field demonstration of the Zenergy Power
HTS FCL, marking a milestone event in the history of FCL development in the United States.
The experience gained from the research led to a more reliable controller, a dramatic reduction
of the FCL’s size and weight, and the replacement of liquid nitrogen cryogenic refrigeration by
a low-maintenance dry cooling system.
The Zenergy Power HTS FCL was first laboratory tested successfully against FCL specifications
and test criteria developed by US utilities, including the host utility Southern California Edison.
The HTS FCL was then successfully demonstrated in actual field operation. The HTS FCL unit
is rated at 15 kV, 1200 A, 110 kV (BIL), steady-state insertion impedance < 1 percent, and a fault
current reduction capability of 20 percent of 23 kA prospective fault current.
The first-generation Spider-core HTS FCL had a dry-type transformer (air dielectric) structure,
total overall dimensions of about 19’x19’x7’, and weighed approximately 50,000 lb. The second-
generation “Compact” HTS FCL that followed from this research employed innovative core
architecture and oil dielectric transformer construction techniques that led to a much more
125

compact size of approximately 8’x10’x11’ (including electric bushings for AC circuit connection,
which protrude from the side in a customer-required air-filled cable box) and a weight of
approximately 67,000 lb. The Compact FCL had a similar power rating but much higher fault
current limiting performance, compared to the Spider-core FCL. The new Compact FCL design
is an attractive option where real estate requirements are tight.
Regarding cost-effectiveness, since the first generation device has a relatively low electrical
rating, with maximum potential ratings on the order of 25 MVA and high duty-cycle ratings
significantly lower, the present manufacturing cost may not represent a high-value proposition.
This issue has been addressed, in two directions: FCL technology is being improved to reduce
basic device manufacturing cost, and FCL technology is being scaled up for higher voltages and
currents. The latter may be more important in terms of cost-effectiveness and value proposition,
considering the fact that higher-voltage substations can often be more crowded and have fewer
options for expansion, due to electrostatic clearance issues, and higher voltage components
(switchgear, insulators, transformers, buswork, etc.) are exponentially more expensive than
their low-voltage counterparts. Further, as more and more large renewable power generators
desire to connect to the electric grid, high-voltage tie-lines are becoming increasingly strained as
more power is wheeled from long distances and as grid interconnections increase. All of these
factors are projected to continue to increase energy levels within the grid, leading to potentially
unsafe fault current levels. In some cases, higher-rated components cannot be retrofitted in the
available space, leading to lengthy and costly major upgrades of grid infrastructure. Economic
studies and performance models show that at current performance levels and price points, FCL
technologies can be very cost-effective compared to major upgrade projects at high voltages.
The FCL demonstration project also brought to light an issue with respect to the device’s
reliability and maintenance requirements. While the FCL device was kept operable and
performed satisfactorily while actively operating in the SCE grid, a number of interventions
were needed to keep the FCL completely operational. For a research prototype, this was
acceptable, but as a commercial product, it will need to be, and is being, substantially improved.
A desirable FCL needs to target “5-nines” reliability with only a single, scheduled, short-
duration annual maintenance outage to fit within today’s utility maintenance requirements.
Even better would be extending the maintenance interval to two years or longer, but an annual
outage of a day or less would be acceptable if the FCL were reliable and required no other
maintenance.
If the FCL is to be relied upon for protection, then bypassing it from service temporarily for
maintenance can be a risk for the circuit. Perhaps a scheduled bypass can be accomplished by
splitting buses or otherwise reconfiguring the electric grid in ways acceptable for short
durations, but it may not be optimal for long-term operation, or not acceptable at all in some
cases. Thus the FCL should be designed, to the maximum extent possible, to allow essential
routine maintenance, such as cryogenic maintenance, to be performed while the unit is
energized. Zenergy Power is aggressively modifying their FCL technologies for reduced
maintenance requirements, higher MTBF, and lower MTTR, and these considerations are also
an integral part of the EPRI/Silicon Power team’s development efforts.
126

One important issue identified during the field test was that the nitrogen cooling system could
be problematical for maintenance since it requires periodic maintenance of the cryostat,
including drying and/or replacing the volume of liquid nitrogen. This may not be feasible in
practice, considering the fact that the unit is typically installed in a high-voltage zone. However,
this issue was addressed in the second-generation HTS FCL design by replacing the liquid
nitrogen cooling system with a cryogen-free, “dry-type” conductive cryogenic cooling system,
available as a commercial off-the-shelf unit, resulting in a more reliable and robust system with
significantly reduce cryogenic maintenance requirements.
Overall, the Zenergy Power HTS FCL project is considered as an important success, which has
already led to a scaled up design for a transmission level (138kV) commercial application.
11.4 Recommendations
Electricity is a vital force in our national and state economies. It is an important goal of the
California Energy Commission to support the technologies required satisfying a reliable, safe
and environmentally responsible electrical energy supply. As such, FCC technology is a
potentially cost-effective alternative to the capital-intensive upgrades to the power system that
would ordinarily be required to meet growth in electrical demand while maintaining the
requisite level of electric service reliability. The development and demonstration efforts in this
research project have already resulted in two FCC test plans, two full FCC designs, valuable
information for developing the needed industry standards for FCCs, and have contributed to
one commercial sale and one migration to a transmission-level FCC application. Further RD&D
efforts building upon these results will promote additional progress toward achieving
commercial application of FCCs.
The research team has accumulated extensive experience in the development of FCC
technology, and in the operation and support of FCC field demonstration at Southern California
Edison’s Avanti Circuit. Further fast-paced and more advanced development of FCC
technology is strongly recommended. As the demand for electrical energy rapidly increases,
particularly in response to renewable power and “green technology” initiatives, substantial
investment in capital-intensive system upgrades will have to made on a fast-track basis in order
to maintain the required levels of system availability and reliability in the absence of advanced
and cost-effective new technologies such as FCCs. Significant amounts of these investments are
potentially avoidable, if suitable FCC technologies are available to California utilities, and if
California utilities are given reasonable incentives to deploy the new technologies. An
accelerated program of FCC technology focused on reducing the cost, improving the reliability,
and increasing the voltage and current ratings of FCC technology is needed in order for
California utilities and ratepayers to reap the benefits. California utilities have stepped up to the
challenge by sponsoring an ambitious attempt to provide a real-world test bed and testing
protocol for proving these technologies and showing how they can be implemented in the
utility grid. Additional development efforts and demonstration projects on newer FCC
technologies can play an important role in accelerating the acceptance of these technologies in
the California and US utility systems.
127

GLOSSARY
A Ampere, Amp (unit of current)
AC Alternating Current
ANSI American National Standards Institute
BIL Basic Insulation Level
C Celsius (temperature)
CB Circuit Breaker
CAD Computer-Aided Design
CEC California Energy Commission
CES China Electrotechnical Society
CIGRE Conseil International des Grands Réseaux Electriques (International
Council on Research for Large Electric Systems)
CLR Current-Limiting Reactor
CT Current Transformer
DC Direct Current
EPRI Electric Power Research Institute
ETO Ethylene Oxide
FCC Fault Current Controller
FCL Fault Current Limiter
FEM Finite Element Model
FPGA Field-Programmable Gate Array
g Gravitational Constant
GB Gigabyte
GTO Gate Turn-Off [Thyristor]
H Henry (unit of inductance)
HMI Human-Machine Interface
HTS High-Temperature Superconducting
HV High-Voltage
HVAC Heating, Ventilating and Air Conditioning
128

I Current (Amperes)
IGBT Insulated Gate Bipolar Thyristor
IGCT Integrated Gate-Commutated Thyristor
IEE Institution of Electrical Engineers (UK)
IEEE Institute of Electrical and Electronics Engineers (USA)
IET Institution of Engineering and Technology
IPEC International Power Engineering Conference
j Square Root of -1 (represents 90º phasor displacement)
kA Kiloamp (Amp x 103)
kV Kilovolt (Volt x 103)
kW Kilowatt (Watt x 103)
lb. Pound
mH MilliHenry (H x 10-3)
mmf Magnetomotive Force
MOV Metal-Oxide Varistor
MTBF Mean Time Between Failures
MTTR Mean Time to Repair
MVA MegaVolt-Amperes
MVAR MegaVolt-Amperes, Reactive
MW MegaWatt (Watt x 106)
NC Normally Closed
NEETRAC National Electric Energy Testing, Research and Applications Center
NO Normally Open
Ω Ohm (unit of resistance, reactance or impedance)
pC picoCoulomb (Coulomb x 10-12: unit of electrical charge)
PAG Project Advisory Group
PG&E Pacific Gas & Electric Co.
PLC Programmable Logic Controller
PM Permanent Magnet
PQ Power Quality
129

PSCAD Power System Computer-Aided Design
PT Potential Transformer
PWM Pulse-Width Modulation
R Resistance
RAM Random-Access Memory
RD&D Research, Development and Demonstration
rms Root Mean Square
ROM Read-Only Memory
SBB Standard Building Block
SC Superconducting
SCADA Supervisory Control and Data Acquisition
SCE Southern California Edison Co.
SCR Silicon-Controlled Rectifier
SEL Schweitzer Engineering Laboratories
SGTO Super GTO
SiC Silicon Carbide
SMES Superconducting Magnetic Energy Storage
SPS Standard Power Stack
SRL Saturable Reactor Limiter
SSFCL Solid-State Fault Current Limiter
THD Total Harmonic Distortion
UCB University of California – Berkeley
UCI University of California – Irvine
UPS Uninterruptible Power Supply
V Volt
VLC Visual Logic Controller
W Watt
X Reactive Impedance (Ohms)
Z Complex Impedance (R + jX) (Ohms)
ZnO Zinc Oxide
130

REFERENCES
[1] “Anatomy of a Short Circuit,” Siemens Tech Topic no. 44, rev1, December 12, 2003.
[2] L. A. Kojovic, S. P. Hassler, K. L. Leix, C. W. Williams, E. E. Baker. “Comparative Analysis
of Expulsion and Current-limiting Fuse Operation in Distribution Systems for Improved
Power Quality and Protection,” IEEE Transactions on Power Delivery, vol. 13, no. 3, pp. 863-
869, July 1998.
[3] “Using Current-limiting Fuses to Increase Short Circuit Current Ratings of Industrial
Control Panels,” Littelfuse Application Note, 2007.
[4] F. Pinnekamp. “The Circuit Breaker,” ABB Review, Jan. 2007. Available at:
http://library.abb.com/global/scot/scot271.nsf/veritydisplay/737de0b7f522f9b2c125728b004
74780/$File/75-78%201M720_ENG72dpi.pdf
[5] T. Taylor, A. Hanson, D. Lubkeman, M. Mousavi. “Fault Current Review Study,” ABB
Report No. 2005-11222-1-R.04, pp. 41-45, December 22, 2005.
[6] P. G. Slade, R. E. Voshall, J. L. Wu, J. J. Bonk, E. J. Stacey, W. F. Stubler, J. Porter, L. Hong.
“The Utility Requirements for a Distribution Fault Current Limiter,” IEEE Transactions on
Power Delivery, vol. 7, no. 2, pp. 507-515, April 1992.
[7] M. Noe, B. R. Oswald. “Technical and Economical Benefits of Superconducting Fault
Current Limiters in Power Systems,” IEEE Transactions on Applied Superconductivity, vol. 9,
no. 2, June 1999, pp. 1347-1350.
[8] A. Neumann. “Application of Fault Current Limiters,” Department for Business,
Enterprise and Regulatory Reform (BERR) report, pp. 3-14, 2007. Available at:
http://www.ensg.gov.uk/assets/dgcg00099rep.pdf
[9] S. Eckroad. “Utility Needs Survey for Fault Current Limiters,” International Workshop on
Coated Conductors for Applications (CCA08), Electric Power Research Institute, p. 7,
December 4-6, 2008.
[10] D. Haught. “Fault Current Limiters,” DOE newsletter, Nov. 6, 2009. Available at:
www.oe.energy.gov/DocumentsandMedia/hts_fcl_110609.pdf.
[11] J. C. Krause. “Short-Circuit Current Limiters – Literature Survey 1973-1979,” TH-Report
80-E-109, ISBN 90-6J44-109-9, Eindhoven University of Technology, July 1980.
[12] A. J. Power. “A Look at Tomorrow,” IEE Colloquium on Fault Current Limiters, pp. 1-5, Jun.
8 1995.
[13] E. M. Leung. "Superconducting Fault Current Limiters," IEEE Power Engineering Review,
vol. 20, pp. 15-18, 30, 2000.
[14] C. Meyer, S. Schröder, R. W. De Doncker. “Solid-State Circuit Breakers and Current
Limiters for Medium-Voltage Systems Having Distributed Power Systems,” IEEE
Transactions on Power Electronics, vol. 19, no. 5, pp. 1333-1340, Sep. 2004.
131

[15] M. Noe, M. Steurer. “High-temperature Superconductor Fault Current Limiters: Concepts,
Applications, and Development Status,” Superconductor Science & Technology, vol. 20, pp.
R15-29, January 15, 2007.
[16] D. Larbalestier, R. D. Blaugher, R. E. Schwall, R. S. Sokolowski, M. Suenaga, J. O. Willis.
“Power Applications of Superconductivity in Japan and Germany,” World Technology
Evaluation Commission (WTEC) Report, pp. 45-48, 1997. Available at:
http://www.wtec.org/loyola/scpa/toc.htm.
[17] M. M. Lanes, H. A. C. Braga, P. G. Barbosa. “Limitador de Corrente de Curto-Circuito
Baseado em Circuito Ressonante Controlado por Dispositivos Semicondutores de
Potência,” IEEE Latin America Transactions, vol. 5, no. 5, pp. 311-320, September 2007.
[18] C. Meyer, P. Köllensperger, R. W. De Doncker. “Design of a Novel Low Loss Fault Current
Limiter for Medium-Voltage Systems,” Nineteenth Annual IEEE Applied Power Electronics
Conference and Exposition (APEC04), pp. 1825-1831, 2004.
[19] B. Chen, A. Q. Huang, M. Baran, C. Han, W. Song. “Operation Characteristics of Emitter
Turn-Off Thyristor (ETO) for Solid-State Circuit Breaker and Fault Current Limiter,”
Twenty-First Annual IEEE Applied Power Electronics Conference and Exposition (APEC06), pp.
174-178, 2006.
[20] T. Genji, O. Nakamura, M. Isozaki, M. Yamada, T. Morita, M. Kaneda. “400V-class High
Speed Current Limiting Circuit Breaker for Electric Power System,” IEEE Transactions on
Power Delivery, vol. 9, issue 3, pp. 1428-1435, July 1994.
[21] M. M. R. Ahmed, G. Putrus, Li Ran, R. Penlington. “Development of a Prototype Solid-
State Fault-Current Limiting and Interrupting Device for Low-Voltage Distribution
Networks,” IEEE Transactions on Power Delivery, vol. 21, no. 4, pp. 1997-2005, Oct. 2006.
[22] M. M. R. Ahmed, G. A. Putrus, L. Ran, L. Xiao. “Harmonic Analysis and Improvement of a
New Solid-State Fault Current Limiter,” IEEE Transactions on Industry Applications, vol. 40,
no. 4, pp. 1012-1019, July/August 2004.
[23] H. J. Boenig and D. A. Paice. “Fault Current Limiter Using a Superconducting Coil,” IEEE
Transactions on Magnetics, vol. 19, no.3, pp. 1051-1053, May 1983.
[24] M. T. Hagh, M. Abapour. “DC Reactor Type Transformer Inrush Current Limiter,” IET
Electric Power Applications, vol. 1, issue 5, pp. 808-814, 2007.
[25] T. Nomura, M. Yamaguchi, S. Fukui, K. Yokoyama, T. Satoh, K. Usui. “Single DC Reactor
Type Fault Current Limiter for 6.6 kV Power System,” IEEE Transactions on Applied
Superconductivity, vol. 11, no. 1, pp. 2090-2093, March 2001.
[26] Z. Lu, D. Jiang and Z. Wu. “A New Topology of Fault-Current Limiter and Its Parameters
Optimization,” 34th Annual IEEE Power Electronics Specialists Conference (PESC03), pp. 462-
465, 2003.
[27] W. Fei and Y. Zhang, “A Novel IGCT-based Half-controlled Bridge Type Fault Current
Limiter,” CES/IEEE 5th International Power Electronics and Motion Control Conference
(IPEMC 2006), pp. 1-5, 2006.
132

[28] W. Fei, Y. Zhang and Z. Meng. “A Novel Solid-State Bridge Type FCL for Three-Phase
Three-Wire Power Systems,” Twenty-Second Annual IEEE Applied Electronics Conference and
Exposition (APEC07), vol. 2, pp. 1369-1372, 2007.
[29] W. Fei, Y. Zhang and Q. Wang, “A Novel Bridge Type FCL Based on Single Controllable
Switch,” Seventh International Conference on Power Electronics and Drive Systems (PEDS ‘07),
pp. 113-116, 2007.
[30] W. Fei, Y. Zhang, Z. Lu. “Novel Bridge-Type FCL Based on Self-Turnoff Devices for
Three-Phase Power Systems,” IEEE Transactions on Power Delivery, vol. 23, issue 4, pp.
2068-2078, October 2008.
[31] V. K. Sood, S. Alam. “Three-Phase Fault Current Limiter for Distribution Systems,” Sixth
International Conference on Power Electronics and Drive Systems (PEDS ’06), pp. 1-6, 2006.
[32] T. Hoshino, K. M. Salim, A. Kawasaki, I. Muta, T. Nakamura, M. Yamada. “Design of 6.6
kV, 100 A Saturated DC Reactor Type Superconducting Fault Current Limiter,” IEEE
Transactions on Applied Superconductivity, vol. 13, no. 2, pp. 2012-2015, 2003.
[33] E. Yokoyama, T. Sato, T. Nomura, S. Fukui and M. Yamaguchi. “Application of Single DC
Reactor Type Fault Current Limiter As a Power Source,” IEEE Transactions on Applied
Superconductivity, vol. 11, no. 1, March 2001.
[34] Z. Caihong, W. Zikai, Z. Dong, Z. Jingye, D. Xiaoji, G. Wengyong, X. Liye, L. Liangzhen.
“Development and Test of a Superconducting Fault Current Limiter-Magnetic Energy
Storage (SFCL-MES) System,” IEEE Transactions on Applied Superconductivity, vol. 17, issue
2, pp. 2014-2017, 2007.
[35] T. Hoshino, K. M. Salim, M. Nishikawa, I. Nuta, and T. Nakamura, “DC Reactor Effect on
Bridge Type Superconducting Fault Current Limiter During Load Increasing,” IEEE
Transactions on Applied Superconductivity, vol. 11, no. 1, pp. 1944-1947, March 2001.
[36] T. Nomura, M. Yamaguchi, S. Fukui, K. Yokoyama. “Study of a Single DC Device Type
FCL for Three Phase Power System,” Cryogenics, vol. 41, pp. 125-130, 2001.
[37] M. T. Hagh, M. Abapour. ”Nonsuperconducting Fault Current Limiter with Controlling
the Magnitudes of Fault Currents,” IEEE Transactions on Power Electronics, vol. 24, no. 3,
pp. 613-619, March 2009.
[38] M. M. Lanes, H. A. C. Braga, P. G. Barbosa. “Limitador de Corrente de Curto-Circuito
Baseado em Circuito Ressonante Controlado por Dispositivos Semicondutores de
Potência,” IEEE Latin America Transactions, vol. 5, no. 5, pp. 311-320, September 2007.
[39] G. G. Karady. “Principles of Fault Current Limitation by a Resonant LC Circuit,” IEE
Proceedings on Generation, Transmission and Distribution, vol. 139, pp. 1-6, 1992.
[40] S. Sugimoto, J. Kida, H. Arita, C. Fukui, T. Yamagiwa. “Principle and Characteristics of a
Fault Current Limiter with Series Compensation,” IEEE Transactions on Power Delivery,
vol.11, no. 2, pp. 842-847, 1996.
133

[41] C. S. Chang, P. C. Loh. “Designs Synthesis of Resonant Fault Current Limiter for Voltage
Sag Mitigation and Current Limitation,” Proceedings of IEEE Power Engineering Society
Winter Meeting, vol. 4, pp. 2482 – 2487, January 2000.
[42] M. Hojo, Y. Fujimura, T. Ohnishi, T. Funabashi. “Experimental Studies on Fault Current
Limiter by Voltage Source Inverter with Line Voltage Harmonics Compensation,”
European Conference on Power Electronics and Applications 2007, pp. 1-8.
[43] S. S. Choi, T. X. Wang, D. M. Vilathgamuwa. “A Series Compensator with Fault Current
Limiting Function,” IEEE Transactions on Power Delivery, vol. 20, pp. 2248-2256, 2005.
[44] M. Hojo, N. Kuroe, T. Ohnishi. “Fault Current Limiter by Series Connected Voltage Source
Inverter,” IEEE Transactions on Industry Applications, Special Issue on IPEC-Niigata, vol.
126, no. 4, pp. 438-443, 2006.
[45] B. P. Raju, K. C. Parton, T. C. Bartram. “A Current Limiting Device Using
Superconducting DC Bias, Applications and Prospects,” IEEE Transactions on Power
Apparatus and Systems, vol. 101, pp. 3173–3177, 1982.
[46] G. A. Oberbeck, W. E. Stanton, A. W. Stewart. “Saturable Reactor Limiter for Current,” US
Patent 4152637, May 1979.
[47] Y. Pan, J. Jiang. “Experimental Study on the Magnetic-controlled Switcher Type Fault
Current Limiter,” Third International Conference on Electric Utility Deregulation Restructuring
and Power Technologies 2008, DRPT 2008, pp. 1987-1991.
[48] D. Cvoric, S. W.H. de Haan, J.A. Ferreira. “Improved Configuration of the Inductive Core-
Saturation Fault Current Limiter with the Magnetic Decoupling,” IEEE Industry
Applications Society Annual Meeting 2008 (IAS ’08), pp. 1-7, 2008.
[49] D. Cvoric, S. W.H. de Haan, J.A. Ferreira. “New Saturable-Core Fault Current Limiter
Topology with Reduced Core Size,” IEEE 6th International Power Electronics and Motion
Control Conference (IPEMC ’09), pp. 920-926, 2009.
[50] M. Iwdiara, S. C. Mukl, S. Yaniada, et al. “Development of Passive Fault Current Limiter
in Parallel Biasing Mode,” IEEE Transactions on Magnetics, vol. 35, no. 3, pp. 3523-3525,
1999.
[51] H. Liu, Q. Li, L. Zou, Y. Ma, W. H. Siew. “An Equivalent Magnetic Circuit Coupled Model
of PMFCL for Transient Simulation,” Proceedings of the 4th International Universities Power
Engineering Conference (UPEC), pp. 1-5, 2009.
[52] S. Young, F. P. Dawson, A. Konrad. “An Extended Magnet in a Passive dI/dt Limiter,”
Journal of Applied Physics, vol. 76, no. 10, pp. 6874-6876, November 1994.
[53] J-L. Rasolonjanahary, J. P. Sturgess, E. F. H. Chong, A. E. Baker, C. L. Sasse. “Design and
Construction of a Magnetic Fault Current Limiter,” The 3rd International Conference on
Power Electronics, Machines and Drives, pp. 681-685, 2006.
[54] S. Young, F. P. Dawson, A. Konrad. “A Three-material Passive dI/dt Limiter,” Journal of
Applied Physics, vol. 76, no. 10, pp. 6871-6873, November 1994.
134

[55] S. Young, F. P. Dawson, M. Iwahara, S. Yamada. “A Comparison Between a Two-material
and Three-material Magnetic Current Limiter,” Journal of Applied Physics, vol. 83, no. 11,
pp. 7103-7105, June 1998.
[56] V. Rozenshtein, A. Friedman, Y. Wolfus, F. Kopansky, E. Perel, Y. Yeshurun, Z. Bar-Haim,
Z. Ron, E. Harel, N. Pundak. “Saturated Cores FCL—A New Approach,” IEEE Transactions
on Applied Superconductivity, vol. 17, no. 2, pp. 1756-1759, June 2007.
[57] H. Hong, Z. Cao, J. Zhang, X. Hu, J. Wang, X. Niu, B. Tian, Y. Wang, W. Gong, Y. Xin. “DC
Magnetization System for a 35 kV/90 MVA Superconducting Saturated Iron-Core Fault
Current Limiter,” IEEE Transactions on Applied Superconductivity, vol. 19, no. 3, pp. 1851-
1854, June 2009.
[58] Y. F. He, J. H. Li, X. H. Zong, J. Sun, Y. N. Wang, C. L. Wu, J. X. Wang. “The High Voltage
Problem in the Saturated Core HTS Fault Current Limiter,” Physica C
(www.sciencedirect.com), vol. 386, pp. 527-530, 2003. Available at:
http://www.sciencedirect.com/science/article/pii/S0921453402021536
[59] Y. Xin, J. Zhang, W. Gong. “Voltage Surge Protection Circuit for Superconducting Bias
Coil,” IEEE Transactions on Applied Superconductivity, vol. 20, no. 3, pp. 1118 - 1121, 2010.
[60] V. Temple. “Super GTO’s Push the Limits of Thyristor Physics,” 35th Annual IEEE Power
Electronics Specialists Conference (PESC04), vol. 1, pp. 604-610, 2004.
[61] V. Temple. “GTO’s Fight Back,” Power Electronic Europe, Issue 6, pp. 32-34, 2004.
[62] V. Keilin, I. Kovalev, S. Kruglov, V. Stepanov, I. Shugaev, V. Shcherbakov. “Model of HTS
Three-Phase Saturated Core Fault Current Limiter,” IEEE Transactions on Applied
Superconductivity, vol. 10, No. 1, pp. 836-839, 2000.
[63] Y. Xin. “Development of Saturated Iron Core HTS Fault Current Limiters,” IEEE
Transactions on Applied Superconductivity, vol. 17, no. 2, pp. 1760-1763, 2007.
[64] F. Moriconi, N. Koshnick, F. De La Rosa, A. Singh, “Modeling and Test Validation of a
15kV 24MVA Superconducting Fault Current Limiter,” 2010 IEEE PES Transmission and
Distribution Conference and Exposition, pp. 1-6, April 19-22, 2010.
[65] S. B. Abbott, D. A. Robinson, S. Perera, F. A. Darmann, C. J. Hawley, T. P. Beales.
“Simulation of HTS Saturable Core-Type FCLs for MV Distribution Systems,” IEEE
Transactions on Power Delivery, vol. 21, no. 2, April 2006.
[66] A. Nelson, L. Masur, F. Moriconi, F. De La Rosa, D. Kirsten. “Saturated-Core Fault Current
Limiter Experience at a Distribution Substation,” 21st International Conference on Electricity
Distribution, June 6-9, 2011.
[67] F. Moriconi, F. De La Rosa, F. Darmann, A. Nelson, L. Masur. “Development and
Deployment of Saturated-Core Fault Current Limiters in Distribution and Transmission
Substations,” IEEE Transactions on Applied Superconductivity, vol. 21, no. 3, pp. 1288-1293,
June 2011.
135

[68] C. R. Clarke, F. Moriconi, A. Singh, A. Kamiab, R. Neal, A. Rodriguez, F. De La Rosa, N.
Koshnick. “Resonance of a Distribution Feeder with a Saturable Core Fault Current
Limiter,” 2010 IEEE PES Transmission and Distribution Conference and Exposition, pp. 1-8,
April 19-22, 2010.
[69] A. Abramovitz, K. Smedley. “Review of the Saturable Core Fault Current Limiter
Technology,” submitted to IEEE Transactions on Power Delivery, Nov. 2010.
[70] A. Abramovitz, K. Smedley, et al. “Prototyping and Testing of a 15kV/1.2kA Saturable
Core Reactor High Temperature Superconductive Fault Current Limiter,” submitted to
IEEE Transactions on Power Delivery, April, 2011.
[71] A. Abramovitz, K. Smedley. “Survey of Solid State Fault Current Limiters,” accepted for
publication in the IEEE Transactions on Power Electronics.
[72] “Fault Current Limiters: Report on the Activities of CIGRE WG A3.10,” ELECTRA, no.
194, pp. 23-29, February 2001.
[73] Available at: http://www.siliconpower.com/documents/comp_data/spt315.pdf
[74] Available at: http://www.siliconpower.com/_documents/sscl-datasheet.pdf
[75] IEEE Standard C57.16-1996: Requirements, Terminology, and Test Code for Dry-Type Air-
Core Series-Connected Reactors, IEEE 1996.
[76] IEEE Standard C57.12.00-2010: IEEE Standard for General Requirements for Liquid-
Immersed Distribution, Power, and Regulating Transformers, IEEE 2010.
[77] IEEE Standard C57.12.01-2005: General Requirements for Dry-Type Distribution and
Power Transformers Including Those with Solid-Cast and/or Resin-Encapsulated
Windings, Rev. 2005, IEEE 2005.
[78] IEEE Standard 693-2005: IEEE Recommended Practice for Seismic Design of Substations,
Rev. 2005, IEEE 2005.
[79] IEEE Standard 519-1992: IEEE Recommended Practices and Requirements for Harmonic
Control in Power Systems, Rev. 1992, IEEE 1992.
[80] S. Weckmann, “Dynamic Electrothermal Model of a Sputtered Thermopile Thermal
Radiation Detector for Earth Radiation Budget Applications,” Master’s Thesis, Mechanical
Engineering, Virginia Polytechnic University, August 2008.
[81] L. Sanford. “First HTS FCL Goes into a US Grid,” Modern Power Systems, May 2009.
136

APPENDIX A:
List of Attachments
Attachment I: Zenergy Power HTS FCL Test Plan
Attachment II: Zenergy Power HTS FCL Laboratory Test Report
Attachment III: Zenergy Power HTS FCL Dielectric and HV Test
Attachment IV: Zenergy Power HTS FCL Normal State Temperature Rise Test
Attachment V: Zenergy Power HTS FCL Short Circuit Test
Attachment VI: Zenergy Power HTS FCL High Voltage Field Test
Attachment VII: Zenergy Power HTS FCL Operation Manual
Attachment VIII: Zenergy Power HTS FCL Cryostat Evacuation and Moisture Removal
Procedure
Attachment IX: Zenergy Power HTS FCL Liquid Nitrogen Fill Procedure
Attachment X: Silicon Power SSCL Test Plan
A-1

California Energy Comission 500-2013-134 DEVELOPMENT OF FAULT CURRENT CONTROLLER TECHNOLOGY Prototyping, Laboratory Testing, and Field Demonstration

More Related Content

Similar to California Energy Comission 500-2013-134 DEVELOPMENT OF FAULT CURRENT CONTROLLER TECHNOLOGY Prototyping, Laboratory Testing, and Field Demonstration (20)

More from Franco Moriconi (16)

Recently uploaded (20)

California Energy Comission 500-2013-134 DEVELOPMENT OF FAULT CURRENT CONTROLLER TECHNOLOGY Prototyping, Laboratory Testing, and Field Demonstration