Design, Test and Demonstration of Saturable Reactor High-Temperature Superconductor Fault Current Limiters

ZP/Error! Reference source not found.
Page 1 of 71
Final Engineering Report
Design, Test and Demonstration of Saturable
Reactor High-Temperature Superconductor
Fault Current Limiters
Final DOE Technical Report, DE-FC26-07NT43242
By: Frank Darmann, Robert Lombaerde, Franco Moriconi, Albert Nelson
Author: Robert Lombaerde
For: U. S. Department of Energy, National Energy Technology Laboratory
Client: Donald Geiling
March 2012
1616 Rollins Road, Burlingame, CA 94010 USA
www.zenergypower.com

ZP-ER2011-009
Page 2 of 71
Final Engineering Report
Design, Test and Demonstration of Saturable
Reactor High-Temperature Superconductor
Fault Current Limiters
Final DOE Technical Report, DE-FC26-07NT43242
By: Frank Darmann, Robert Lombaerde, Franco Moriconi, Albert Nelson
Author: Robert Lombaerde
For: U. S. Department of Energy, National Energy Technology Laboratory
Client: Donald Geiling
March 2012
www.zenergypower.com

ZP-ER2011-009
Page 3 of 71
Engineering Report
Status Final
Reporting Entity:
Zenergy Power Inc.
Responsible Person: Robert Lombaerde
Project Name:
Design, Test and Demonstration of Saturable Reactor
High-Temperature Superconductor Fault Current Limiters
Document Title: Final DOE Technical Report, DE-FC26-07NT43242
Date of issue: 02/24/2012
Client(s):
U. S. Department of Energy
National Energy Technology Laboratory
Don Geiling
donald.geiling@netl.doe.gov
Author(s): Frank Darmann, Robert Lombaerde, Franco Moriconi, Albert Nelson
Distribution Limitations: None Approved: A. Nelson
DISCLAIMER
This report was prepared as an account of work sponsored by an agency of the United States
Government. Neither the United States Government nor any agency thereof, nor any of their employees,
makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy,
completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents
that its use would not infringe privately-owned rights. Reference herein to any specific commercial
product, process or service by trade name, trademark, manufacturer, or otherwise does not necessarily
constitute or imply its endorsement, recommendation, or favoring by the United States Government or any
agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect
those of the United States Government or any agency thereof.

ZP-ER2011-009
Page 4 of 71
Contents
DISCLAIMER 3
1 Executive Summary 6
2 Introduction and Background 7
3 Technical Approach and Results 8
3.1 How the Saturable Reactor HTS Fault Current Limiter Works 8
3.2 Budget Period 1 9
3.2.1 Testing First Prototype at PG&E San Ramon, CA 10
3.2.2 Testing First Prototype at Powertech, Surry, British Columbia, Canada 11
3.2.3 Southern California Edison FCL, installed in Avanti Grid, San Bernardino, CA 12
3.2.4 Utility Partner Implementation 18
3.2.5 The Compact Design 19
3.2.6 Powertech Compact FCL Test Results 25
3.2.7 Powertech Testing Conclusions 27
3.3 Special 2G Wire Test 28
3.3.1 2G Wire Report Summary 28
3.3.2 2G Wire Coil 29
3.3.3 2G Test Results 31
3.3.4 Conclusions – Improvements – Next Steps 32
3.4 Budget Period 2 32
3.4.1 138 kV Transmission Class FCL Design 32
3.4.2 138 kV Design Challenges 35
3.4.3 HTS Magnet Design 47
3.4.4 FCL Auxiliary Enclosure 59
3.4.5 Zenergy FCL Software 59
3.4.6 FCL Structure 61
3.4.7 Factors Affecting the FCL Timeline 64
3.4.8 Zenergy Power Change of Focus 65
4 Summary and Conclusions 66
5 Acknowledgements 67

ZP-ER2011-009
Page 5 of 71
6 Index of Figures 68
7 Index of Tables 69
8 List of Acronyms and Definitions 69
8.1 Acronyms 69
8.2 Definitions None 70
9 References 70
10 Appendix. Public Documents 70
10.1 Patents Granted or Applied For, Software Developed 70
10.2 Publications 70
10.3 Technical Presentations 71

ZP-ER2011-009
Page 6 of 71
1 Executive Summary
Fault current limiters (FCLs) when utilized in the electric grid will limit the current in the system when a fault,
commonly thought of as a short circuit, occurs. (Nominal current in utility electrical lines may easily be as much as
2,000 Amps or more, and during a fault when the lines are shorted together by wind or shorted to ground from a
downed utility pole or equipment internal failure, the current can typically be 25,000 Amps, 40,000 Amps, 60,000
Amps or more.) While the usual utility implementation purely for fault limitation is an air-core reactor (utilities often
use other means such as splitting electric buses or relocating generator or electrical feeder circuit tie-ins to limit total
fault current potential), the air-core reactor will typically have a relatively high nominal insertion impedance and
subsequent voltage drop that will increase linearly with current. The Zenergy Power FCL is a saturable reactor high-
temperature superconductor fault current limiter design and is rather unique in that it presents low impedance to
electrical currents under nominal conditions; however, under fault conditions the impedance is large and will limit the
flow of current. A key factor in the Zenergy Power design is that the fault current itself initiates the change to high
impedance with no other outside triggering and can, therefore, be considered a “Smart Grid” component. The
Zenergy Power FCL reacts quickly, for the impedance change can occur within the first quarter cycle of the 50-Hertz
or 60-Hertz current. The Zenergy Power design has no AC load current or voltage through the superconducting
components, and once the fault is cleared, the FCL returns instantly to low impedance and is ready for the next fault.
The fast reaction time allows the FCL to handle multiple faults in rapid succession. The application of FCLs to an
electric grid can provide protection to upstream circuitry already at its maximum rating, lower the maximum fault
current in the system, improve the lifetime and reliability of downstream electrical circuits and components, enable the
“ride-through” of short duration electrical transients without power interruption, and facilitate interconnection of
distributed energy sources (wind, solar, hydro) without increasing the fault current design rating.
Zenergy Power has successfully designed, built, tested, and installed in the U.S. electrical grid a saturable reactor
FCL. Beginning in 2007, as SC Power Systems and from 2008 as Zenergy Power, Inc., Zenergy Power employed a
DOE matching funding opportunity cost-share grant and American Recovery and Reinvestment Act funding to design,
build and test a saturable reactor fault current limiter suitable for commercialization. The demonstration effort began
with a very basic FCL design utilizing 1G HTS conductor for a DC bias magnet housed in a LN2 cryostat. Substantial
progress was made in two areas. First, the cryogenics progressed from containers of liquid nitrogen housing the HTS
coils to cryostats utilizing dry conduction cooling reaching temperatures down to sub 20 degrees K. Large round
cryostats with “warm bore” diameters of 1.7 meters enabled the design of large tanks to hold the AC components.
Second, the design of the AC part of the FCL was refined from a six-legged, picture-frame magnetic core “spider”
design to a more compact and lighter design with better limiting capability. Refinement of the flux path and core
shape led to an efficient saturable reactor design requiring less Ampere-Turns to saturate the core. In conclusion, the
refinement of the saturable reactor FCL led to a more efficient design not requiring the HTS wire and associated
peripheral equipment for a more economical product in line with the electric utility industry expectations.
The work funded by this project assisted Zenergy Power’s efforts to design, build and prove the commercial potential
of the saturable core FCL concept and to demonstrate its fault current limiting potential with full-scale prototypes.
Computer models enabled Zenergy Power to first predict the fault current limiting of a design and then to verify the
design’s performance from actual measurements taken at test laboratories. The original plan was to build three sets
of FCLs with voltage classes of 15 kV, 69 kV, and 138 kV over five Budget Periods (BPs), however, positive results of
the initial tests on the low-voltage FCLs initiated a revised plan of only two budget periods leading directly to a 138 kV
FCL. The key factor in this decision was that the superconductor components and the AC electrostatics were
completely separate, and for a HV design the electrostatic considerations within the AC components “tank” alone
were the major risk factor for the design to work.

ZP-ER2011-009
Page 7 of 71
Although the 138 kV paper design had been finished and major components were to be procured under issued
purchase orders, the project was terminated due to the realization that with proper flux manipulation and containment,
the FCL could be designed with a copper coil replacing the HTS wire and its associated cryostat, cold heads,
cryogenic compressors, chillers, HVAC units, auxiliary power supplies and distribution systems, controls and notably,
the associated poor reliability and high maintenance requirements of these cryogenic components and auxiliary
support systems. The Zenergy Power management team decided to steer the direction of the company away from
HTS wire magnets and focus on copper magnet FCLs to take advantage of their inherent lower cost and higher
reliability. Crucial to this decision was the realization that the projections for the increase in performance and cost-
down of HTS conductor and cryogenic systems projected over the next 5 years at the start of the project in 2007 had
not materialized and were not likely to do so for the foreseeable future.
The original goal for the DOE funding of the Zenergy Power project “Design, Test and Demonstration of Saturable
Reactor High-Temperature Superconductor Fault Current Limiters” was to stimulate the HTS wire industry with first
1G then 2G HTS wire applications. Over the approximately 5 years of Zenergy Power’s use of HTS wire, the cost
over time had not dropped nor had the availability increased (today the availability of 2G wire is essentially non-
existent outside of specific programs and partnerships of the few performing 2G HTS conductor manufacturers).
Zenergy Power had foreseen a large potential for FCLs at a reduced price from the initial models built with 1G wire.
The attraction of 2G wires at greatly reduced prices from 1G made the commercial production of FCLs feasible for the
utility industry. However, the price of HTS wire has not dropped as expected, and today one cannot buy quantities of
2G wire and the price is higher than the limited available amounts of 1G wire. With the recent reductions in the DC
magnet bias flux required to operate the Zenergy Power FCL, the commercial pathway for Zenergy Power to provide
a reliable and reasonably priced FCL to the utility industry is to use conventional resistive magnets rather than HTS
magnets. Since the premise of the original funding was to stimulate the HTS wire industry and Zenergy Power has
determined methods to use copper based magnets, the DOE and Zenergy Power decided to mutually terminate the
project.
2 Introduction and Background
To promote the HTS wire industry the United States Department of Energy provided matching funds for several HTS
projects. In September 2007, SC Power Systems, Inc. was granted DE-FC26-07NT43242 to Design, Test and
Demonstrate Saturable Reactor High Temperature Superconductor Fault Current Limiters. The original proposal was
split into five budget periods which encompassed design, building and testing of three classes, 15 kV, 69, kV, and
138 kV, FCLs. Subsequent to the award, SC Power Systems joined with Australian Superconductor, and Trithor
GmbH to form Zenergy Power, Plc headquartered in the UK. The merger in 2008 produced a synergistic company of
design and modeling in Australia, FCL manufacturing in the US, and a turnkey superconducting magnet source in
Germany. The Saturable Reactor HTS FCL, which was first researched in Australia as far back as the late 1990s,
was refined and analyzed from models tested under fault conditions. Design equations were verified from the
models, and Zenergy Power produced an FCL for the California Energy Commission in conjunction with Southern
California Edison and placed it in the SCE grid in San Bernardino, CA for eighteen months. The saturable reactor
FCL design relies very heavily on a large mass of iron which must be saturated with a DC magnetic bias flux. The
original presumption was that this DC magnetic flux biasing could be done most efficiently with an HTS magnet. The
improvements over time of the Zenergy Power FCL have centered on the constant reduction of iron mass and size to
develop a commercially viable product for the utility industry. The progressive development of the saturated reactor
FCL led to the elimination of the HTS wire as more efficient methods of core saturation were developed which could
be accomplished without the cryogenics and their associated poor reliability and maintenance requirements. Without
the need for HTS wire in the FCL, the original purpose of the matching grant was not met and DOE and Zenergy
Power mutually terminated the FCL project.

ZP-ER2011-009
Page 8 of 71
3 Technical Approach and Results
3.1 How the Saturable Reactor HTS Fault Current Limiter Works
The saturable reactor fault current limiter works on the principle that an AC coil with an air core will have relatively low
impedance, but if a (magnetically unsaturated) iron core is placed in the air core of the AC coil, the coil will have much
higher impedance. If the iron core in the AC coil is magnetically saturated, then the impedance will again be low and
will approach that of an air core coil.
 AC coil with air core – low impedance
 AC coil with iron core (magnetically unsaturated iron) - high impedance
 AC coil with iron core (magnetically saturated iron) – low impedance (close to air core value)
Figure 1: AC Coils with Iron Cores Saturated by DC-Powered HTS Coil

ZP-ER2011-009
Page 9 of 71
Figure 1 depicts two AC coils with iron cores and an HTS magnet powered by a DC power supply able to saturate
both cores with magnetic flux. Note: the AC current into the coil on the left side will generate a magnetic flux that
boosts the flux generated by the DC magnet in the associated iron core. The same AC current passing through the
coil on the right side bucks the flux generated by the DC magnet in the right-hand core. If this were a saturated
reactor fault current limiter, the nominal AC load current (which typically needs to be significantly less than the limited
value of any fault current) would flow through the AC coils with little impedance due to the saturated iron cores. (The
magnetic field generated by the HTS magnet is much greater than the magnetic field generated by the AC current in
the AC coil.) When a fault occurs, the fault current is sufficiently large that the flux generated in one iron core from
the fault current through the AC coil during one-half cycle will be enough to desaturate the iron – resulting in the AC
coil having much greater impedance and limiting the fault current. Upon reversal of the AC cycle, the other iron core
will desaturate during the second half-cycle and again limit the AC current flow as the first iron core resaturates. The
key to the operation of the saturated reactor FCL is sizing the iron core and the DC magnetic flux bias magnet to have
low impedance under normal operating conditions when the iron cores are magnetically saturated by the DC magnet,
but designing the AC coils to generate enough magnetic flux to desaturate the iron cores based on the predicted
amount of prospective fault current and the acceptable value of limited fault current.
3.2 Budget Period 1
Zenergy Power’s first FCL was built to prove the saturable reactor concept. It was built with narrow picture frame
style cores and used a nitrogen filled cryostat to cool the HTS 1G wire. Figure 2 is the FCL before it was enclosed.

ZP-ER2011-009
Page 10 of 71
3.2.1 Testing First Prototype at PG&E San Ramon, CA
In general, the first prototype HTS FCL device succeeded in accomplishing about 90% of the overall objectives during
the first phase of testing at PG&E San Ramon, California. Specifically, Zenergy Power personnel were able to:
 Configure the test yard equipment to provide the desired levels of voltage and current,
 Reliably and accurately measure the steady-state and transient voltages and currents,
 Repeatedly measure the test yard voltages and currents with a high degree of repeatability,
 Repeatedly and reliably trigger a bolted three-phase fault at a zero-crossing voltage of one phase of the
circuit,
 Measure the steady-state impedance of the HTS FCL,
 Measure the steady-state voltage drop of the HTS FCL,
 Record the steady-state voltage and current waveforms (both AC and DC) of the HTS FCL,
Figure 2: First FCL Produced by Zenergy Power

ZP-ER2011-009
Page 11 of 71
 Measure the steady-state and transient heat-loads and cooling requirements of the HTS FCL,
 Measure the fault transient response (both AC and DC) of the HTS FCL, and
 Obtain measurements of several other parameters (such as temperature, mechanical loads and flux levels)
for the HTS FCL.
Testing was less than 100% successful, because of two observed issues with the HTS FCL device. The first issue, a
known leak in the cryostat vacuum space, resulted in much higher liquid nitrogen consumption than expected. The
second issue, a higher than desired steady-state insertion impedance, appeared to result from lower than expected
magnetic flux saturation levels in the outer limbs of the HTS FCL iron cores. This was ultimately attributed to
measured lower than target value flux levels in the iron cores, which were determined to result from the existence of
parallel reluctance paths or other flux “short circuits” stemming from certain mechanical construction characteristics of
the HTS FCL.
3.2.2 Testing First Prototype at Powertech, Surry, British Columbia, Canada
High-power testing performed at Powertech Labs followed a test plan summarized in SCP test report SCPS-TR
07_03-03. High voltage fault current tests were conducted at 2.5, 5, 10 and 12.5 kA, under a variety of reactive and
resistive source impedances. X/R ratios were in the range of 10-11. Fault current reduction levels achieved were
14%, 14%, 20% and 22% respectively for tests conducted at 2.5, 5, 10 and 12.5 kA RMS. Peak measured fault
current for the test conducted at 12.5 kA was measured at 16.5 kA. As expected, at full-rated power conditions
(12.47 kV and 10,000 amps RMS) at Powertech, fault current reduction levels achieved were much lower than the
low voltage testing performed at PG&E.
Summarizing other test results includes:
 Maximum Symmetrical Current: 12.5 kA RMS
 Maximum Asymmetrical Current: 39 kA
 No appreciable variations or excursions in cryostat temperature
 No significant AC coil to DC coil coupling was observed.
 Up to the lab’s thermal limits, the HTS FCL was subjected to the labs full-power, steady-state voltage and
current, faulted and returned to nominal steady-state operations within 30 cycles.
 Insertion impedance was reduced or improved by more than 50% as compared to the earlier testing
performed at PG&E. Additional modeling and analysis, performed to determine how to lower the insertion
impedance to meet SCP standards, revealed the need to improve the low-reluctance flux return path.
With the exception of insertion impedance, finite element analysis (FEA) modeling results concurred with the test
results. The higher measured insertion impedance was attributed to insufficient iron core saturation in the vicinity of
the AC coils. Thermal and electrical FCL stressing proved the basic design was sufficiently robust for electrical
distribution system service.
The HTS FCL results exceeded and accomplished 100% of the overall objectives during this second phase of testing.
Specifically, SCP personnel were able to:
 Configure the test yard equipment to provide the desired levels of voltage and current,
 Reliably and accurately measure the steady-state and transient voltages and currents,
 Repeatedly measure the test yard voltages and currents with a high degree of repeatability,
 Repeatedly and reliably trigger a bolted three-phase fault at a zero-crossing voltage of one phase of the
circuit,

ZP-ER2011-009
Page 12 of 71
 Measure the steady-state impedance of the HTS FCL,
 Measure the steady-state voltage drop of the HTS FCL,
 Record the steady-state voltage and current waveforms (both AC and DC) of the HTS FCL,
 Measure the apparent steady-state and transient heat-loads and cooling requirements of the HTS FCL,
 Measure the fault transient response (both AC and DC) of the HTS FCL, and
 Obtain measurements of several other parameters (such as temperature, mechanical loads and flux levels)
for the HTS FCL.
The performance of the HTS DC bias coil was measured several times using several methods throughout the testing
program, and its performance was nominal at all times with no variation or deterioration detected.
The results of the first and second phase testing programs were used to modify Zenergy’s modeling and simulation
algorithms, engineering and design techniques, and baseline FCL manufacturing processes for the DOE three-phase,
15 kV-class, 2G+1G HTS FCL, which was designed, built and tested in fiscal year 2008.
3.2.3 Southern California Edison FCL, installed in Avanti Grid, San Bernardino, CA
This distribution class (15 kV) saturated reactor FCL was the first Saturated Reactor FCL installed in the United
States electrical grid. The demonstration lasted for eighteen months and proved to be a learning experience for all
parties. The unit was the object of several tours – specifically CIGRE, and DOE and the source of several papers on
FCLs. The FCL also experienced a real fault in the Avanti grid. Figure 3 shows the 15 kV FCL with several panels
removed during the HV testing at Powertech in Surry, British Columbia, CN.

ZP-ER2011-009
Page 13 of 71
Figure 4 shows the Zenergy Power 15 kV FCL installed at the Avanti Substation of Southern California Edison. The
FCL was installed on the low-voltage side of a 115 kV to 12 kV transformer in line with a feeder circuit. This particular
feeder circuit was a live, commercial circuit with residential, commercial and light industrial customers designated the
“Circuit of the Future,” which had been specifically configured to demonstrate new technologies for the electric grid.
The bypass switch was installed to facilitate isolation of the FCL for maintenance and repair and for quick, remote
removal from the circuit in the event of any problems. The FCL installation was metered for voltage and current on
both sides of the FCL, and the station was equipped with digital fault recorders in order to collect data on events at
the substation and the performance of the FCL under normal and transient conditions.
Figure 3: Zenergy Power’s 15 kV FCL at HV Test

ZP-ER2011-009
Page 14 of 71
The goals of the SCE demonstration program, which was also supported by the California Energy Commission
(CEC), were to demonstrate that ZP could design a FCL to a specific set of performance requirements, build the
device, test it to verify that it met the specified requirements, and qualify it for installation in the SCE grid. A key
aspect was to improve the models used in the software design programs that could accurately predict performance.
Through several refinements ZP has implemented a PSCAD model of the FCL and has been able to predict the
performance of the device. Figures 5 and 6 show the comparison of the predicted fault limitation and the actual
measured fault limiting.
Figure 4: Zenergy Power’s 15 kV FCL Installed at SCE Avanti Substation with Bypass Switch

ZP-ER2011-009
Page 15 of 71
Figure 5: Calculated FCL Limiting of 17% of 21 kA Fault
Figure 6: Powertech Result Showing 19% Fault Reduction of 23 kA Fault
300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000ms
-40
-20
0
20
40
60
80
kA
23kA Prospective
FCL IN

ZP-ER2011-009
Page 16 of 71
The design of the Avanti FCL followed the ZP original picture frame model with twelve limbs (four limbs per phase,
each joined top and bottom with yokes to make six rectangular closed cores for a total of 24 laminated magnetic steel
structures per FCL) and a cryostat in the middle. The cryostat used super-cooled liquid Nitrogen created by two cold
heads in the cryostat which employed a drip condensation method for super-cooling. The two air-cooled cryogenic
compressors provided redundancy (only one was needed to maintain normal operations), and the FCL was monitored
by a microcontroller with remote monitoring capability. Two major problems were encountered and corrected during
the course of the demonstration program. First, the microcontroller had a memory leak and memory overflow causing
the unit to shutdown. This was resolved by implementing an industry standard Siemens PLC for the microcontroller
and SCE installing an automated bypass switch (the bypass switch was not related specifically to the memory
overflow problem, but the overflow problem underscored the need for SCE to be able to remotely connect and
disconnect the FCL). Second, the installation was in the Southern California desert area with very hot temperatures
in the summer. When the outside temperature reached over 108 degrees F the HVAC system could no longer cool
the cryogenic compressors and the system shut down. The solution was to install a larger capacity HVAC system
and a sunshade over the cryogenic and instrumentation area.
Summary of the operational experience at the Avanti Substation includes:
 18 months of operation
 Operated in a harsh, hot, dry, dusty environment
 Maximum ambient temperature reached was 108°F in Summer ‘09
 Heavy winds and considerable wind-borne dust and debris
 Experienced one loss of DC with consequent “resonance” condition (IEEE paper)
 Successful integration with automatic bypass switch
 Experienced one fault event with multiple faults in quick succession (14 January ‘10)
 Experienced three “auxiliary power failures”
 Effective bypass of FCL and shut-down of the HTS coil (as expected)
 “Auxiliary power failures” caused by grid disturbances
 Instantaneous bypassing achieved by SCE (2 out of three)
 Successful recovery after 2-minute black out of auxiliary power
 Performed routine maintenance on cryogenics compressors
 Replaced cryogenic compressor after 8000 hours
 Replenished LN2 after HVAC loss (hot Summer 2009)
The fault event was recorded by SCE instrumentation on 14 January 2010 and is presented here to show the
capability of the FCL.

ZP-ER2011-009
Page 17 of 71
A summary of the experience of ZP FCL for SCE includes:
 First successful integration superconducting FCL in US grid
 Operational experience at SCE has been a big success for both ZP and SCE
 Resonance study revealed minimal effects and defined protocol for FCL grid application analysis
 Both parties gained invaluable experience by operating through all four seasons
 ZP learned to address unplanned events such as loss of station power
 The host utility learned about preventive maintenance and how the device responded to real fault event
Figure 7: Three-Second Fault at SCE Avanti Sub-station, 14 January 2010

ZP-ER2011-009
Page 18 of 71
3.2.4 Utility Partner Implementation
Upon the success of the SCE FCL, ZP implemented a utility partner investigation. Zenergy Power and Seattle City
Light agreed to pursue an FCL for their grid. The design parameters were:
 Line Voltage 26 K Volts
 Steady State Current 1200 Amps
 Prospective Fault Current 30 K Amps
 Fault Current Clipping (Desired) 40 – 50 %
 Allowable Voltage Drop 1 %
These parameters led to a very large and expensive FCL in the picture frame or spider type design. Zenergy
Australia determined that a more compact design would be feasible and suggested a new implementation. Table 1
lists the comparison of the two designs, and Figure 8 contrasts models of the two concepts.
Parameter Spider Compact
Iron Core Weight (lbs) 252K 67K
Cost of Iron @ $3.00/lb $ 756K $ 201K
FCL Size (core iron + AC coils) 19' x 19' 10' x 7'
Table 1: Comparison of Spider versus Compact Design
Figure 8: Spider and Compact FCL Design Examples

ZP-ER2011-009
Page 19 of 71
While the Seattle City Light partnership did not come to fruition due to budget constraints and loss of personnel in
Seattle, ZP was able to determine realistic utility wants and needs. The spider (picture frame) design was deemed
not feasible for production due to the excessive size, weight, and cost.
3.2.5 The Compact Design
The premise of a compact FCL design to reduce the size and weight led ZP to investigate several configurations.
The compact design still works on the same saturated reactor principle as the spider except instead of an interior core
saturated by the HTS magnet and the AC coils over the outside limbs of the picture frame cores, the iron core with
AC coils are all encased by the HTS magnet. The size of the core/AC coils is now limited by the interior bore of the
HTS magnet.
Before committing to spend the monies to manufacture full-scale prototypes, ZP through its Australian sister company
built several small feasibility models to verify the design. While the mini prototypes would receive scaled-down
stresses, they would nonetheless prove the design feasible. Figure 9 shows a typical compact mini prototype.

ZP-ER2011-009
Page 20 of 71
With the agreement of the DOE and the RRT, ZP proceeded to design and build three prototype compact FCLs to
test for their fault current limiting capability. The prototype compact FCLs are commonly referred to by their numeric
code, “Y x Z” where Y is the number of cores per layer and Z is the number of layers. The 6x1 is composed of 6
cores arranged in one layer and is a three-phase device. The 3x2 has three cores in each of two layers and is also a
three-phase device; the 2x1 is composed of two cores arranged in one layer and is a single-phase device.
The HTS magnets were of a rectangular or “race-track” design and produced by ZP’s sister company in Germany.
The design was similar to the magnets employed in a HTS billet heater product for the metals extrusion industry. The
magnets were able to be fabricated from existing designs saving design time and allowing the project to be fast-
tracked. T&R Electric in Colman, South Dakota was the manufacturing partner for the compact FCL build. Figure 10
provides a comparison of the different models built for the compact design and shows the interior construction.
Figures 11-13 are of the actual full size FCLs. The FCL designs were built taking advantage of the ZP design
Figure 9: Mini 6x1 Feasibility Prototype

ZP-ER2011-009
Page 21 of 71
concept of the AC and HTS portions being completely separate. The design consisted of a stack of two HTS
magnets. The three different FCLs (the AC portion of the FCL) in their stainless steel tanks were lowered into the
magnet stack for testing. The ZP trailer was setup with the DC power supplies, cryogenic compressors and chiller,
making it a rolling test bed (Figure 14).
Figure 10: Comparison of the Compact Prototype Models (L to R) 6x1, 3x2, 2x1

ZP-ER2011-009
Page 22 of 71
Figure 11: Full-Scale 6x1 Prototype under Test at Powertech Labs

ZP-ER2011-009
Page 23 of 71

ZP-ER2011-009
Page 24 of 71

ZP-ER2011-009
Page 25 of 71
3.2.6 Powertech Compact FCL Test Results
The test results at Powertech conclusively proved the viability of the compact design FCL. While the testing was
shortened due to equipment problems, enough data was taken to correlate the test data to the predicted
performance. As an example, the fault current limiting results of the 6x1 are shown in Figures 15 and 16. Note the
predicted and measured currents are almost identical and that the graph traces nearly overlap. These tests proved
conclusively the efficacy of the compact FCL design and the accuracy of the design algorithms that had been
developed to engineer the FCL to meet specific performance targets.
Figure 14: Zenergy Trailer Setup as Rolling Test Bed at Powertech Labs

ZP-ER2011-009
Page 26 of 71
Figure 15: Measured versus Calculated 15 kA Prospective and Limited Fault Current – 6x1 FCL
0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75
-20
-15
-10
-5
0
5
10
15
20
25
30
Test 97 - ZENERGYPOWER Compact FCL - 15kA Prospective Fault - 120A DC Bias
Time [sec]
LineCurrent[kA]
MEASURED LIMITED
MEASURED PROSPECTIVE
MODEL LIMITED

ZP-ER2011-009
Page 27 of 71
3.2.7 Powertech Testing Conclusions
The goal of the experimental work was to validate ZP’s design tools and to conduct sufficient empirical work to
optimize the fault-limiting behavior, the steady-state voltage drop, and the size of the Compact FCL prototypes. The
units performed extremely well during testing at steady-state and under the application of sequences involving load
and fault current periods. In several of the configurations, they showed exceptionally low insertion impedance at
maximum load currents and significant fault current limiting capability. For example, the 6x1 design showed less than
70 Volts (1%) insertion voltage drop at 1.2 kA load current and a fault limiting capability of 46% at 25 kA prospective
fault current. For comparison, the 3x2 design showed 40% clipping for the same 25 kA prospective fault current, but
slightly higher insertion impedance at the same load currents. The back emf during faults showed peak values of the
order of 4,000-6,000 Volts with no signs of reverse saturation. This indicates that the magnetic cores and the number
of AC turns for the DC bias magnetic flux available were properly sized. Most of the tests were run with the HTS coils
at 120 A DC bias current, and very minimum coupling was observed during 20-cycle faults at various fault current
levels. There were no mechanical/electrical/thermal issues with the HTS coils during the short-circuit tests. The
power supply and protection circuit performed well, as expected.
While the extensive planned test program was curtailed due to the onset of ground currents in the AC tanks before
the complete test plan was finished, enough data was collected to satisfy the program objectives and to validate the
Figure 16: Measured versus Calculated 15 kA Prospective and Limited Fault Current – 6x1 FCL
0.62 0.63 0.64 0.65 0.66 0.67 0.68 0.69 0.7 0.71 0.72
-20
-15
-10
-5
0
5
10
15
20
Test 97 - ZENERGYPOWER Compact FCL - 15kA Prospective Fault - 120A DC Bias
Time [sec]
LineCurrent[kA]
MEASURED LIMITED
MEASURED PROSPECTIVE
MODEL LIMITED

ZP-ER2011-009
Page 28 of 71
FCL design methodologies. Results pointed clearly to the importance of the AC ampere-turns during the fault
condition adequately de-saturating the core and suggested that the FCL design protocols may over emphasize the
core area parameter. Also, a fourth prototype not included in the set of DOE units was “yoked” (meaning that the two
core half-phases were joined magnetically in a closed-loop by transformer steel). This was done in an effort to
improve the ability of the HTS coils to saturate the iron cores in order to help reduce the steady-state voltage drop.
Analysis of the data and comparison of the performance of the yoked design with the DOE unyoked designs suggest
that “yoking” does improve the DC magnetic flux saturation of the core, but it also reduces the ability of the AC coils
and the fault current to de-saturate the core. These observations must be taken into account under particular
requirements involving a large impedance change with a relatively low change in current from a high steady-state
current and a low fault current.
ZP believes that the experimental work conclusively validated the FCL design protocols and that it is possible to
optimize the trade-offs between fault clipping, steady-state voltage drop, and FCL size. The main parameter
governing such trade-offs would be the maximum steady-state load current voltage drop that can be tolerated and the
rate at which the steady-state voltage drop could be allowed to increase at given load currents.
In conclusion, ZP proved that the Compact FCL design is a sound concept and that ZP should pursue the design of a
Transmission Class FCL to fulfill the intent of the DOE project.
3.3 Special 2G Wire Test
The limited availability and high cost of 2G wire made the construction of a 2G magnet system for an FCL impossible.
ZP had an open order for a 500-meter length of 2G wire with a manufacturer that went unfilled for almost two years
during the FCL program, and which was canceled when the FCL project was terminated. A special report testing the
applicability of 2G wire in an HTS coil was approved by DOE to fulfill the 2G coil requirement. (10.2.1)
3.3.1 2G Wire Report Summary
A test coil was made from a 12 mm wide 2G HTS wire. The manufacturer of the 2G wire was SuperPower, Inc. (US).
The test coil was wound onto a 150 mm diameter GFRP former as a solenoid with 8 layers and 35 turns. The winding
was done without problems, even though the HTS wire is quite wide for a solenoid winding with a small diameter. The
coil was placed in a cryostat and cooled by a single-stage cryogenic cooler. During cool-down, the transition to
superconductivity was found at about 90 K, as expected.
Up to 330 A pulse current was applied to the coil at low (28 K) and medium (45 K) temperatures without significant
voltage drops over the coil being measured. This showed there was no major degradation of the superconductor due
to manufacturing and testing. The coil operation was stable at these currents. At high temperatures (78 K coil, 87.5 K
current lead), a voltage drop in the HTS wire was detected at about 100A, and the voltage did rise with time indicating
an unstable operating point. No fast quenching was detected during the tests done, and no burn out of the wire was
found, which was promising for larger coils, and which suggested that 2G wire might be appropriate for FCL magnet
applications in the future.
The critical current at different temperatures could not be determined exactly due to limits of the test setup. An
improvement of the thermal anchoring of the current leads would be required before the critical current of the coil can
be determined exactly. Because of the high critical current of the wire and the small self-field of the test coil, the
expected critical currents at low temperatures are very high.

ZP-ER2011-009
Page 29 of 71
3.3.2 2G Wire Coil
Using SCS12050 wire, a test coil was designed and manufactured. The coil was designed for 8 layers, which is the
same number of layers employed in a former FCL magnet (the magnet for the CEC/SCE device). The arrangement of
the test coil is shown in Figure 17 below.
A layer winding was chosen for the test coil, as the previous magnets were also made as layer windings. As the
former outer diameter is quite small and the 2G wire is quite wide, it was more difficult to do layer winding.
Test Coil Data:
HTS Wire: SCS12050 (SuperPower)
ID Winding: 150 mm
Former: GFRP Tube
Winding: Layer
No. of Layers: 8
Length: About 50 mm (winding height)
No. of Turns: 35 (in total)
HTS Wire Length: Approximately 16.5 m
Inductance: Approximately 0.236 mH
Center Field: 0.27 mT/A
Perpendicular Field: Approximately 0.62 mT/A Perpendicular Field at Ends of Solenoid

ZP-ER2011-009
Page 30 of 71
Figure 17: Test Coil before Impregnation with Resin
Figure 18: 2G Test Coil Mounted in Cryostat

ZP-ER2011-009
Page 31 of 71
3.3.3 2G Test Results
A series of measurements were taken at the lowest recorded operating temperature of 28 K.
Without heating the cold-head, the coil reached a lowest recorder temperature of about 28 K without application of
current to the test coil. The temperature of the current connection near to the coil was slightly above 35 K. Pulsed
current, up to about 300A, was applied to the coil at a base temperature of about 28K as shown in Figure 19 below.
During current ramp-up and ramp-down a lot of noise was seen in the voltage signals, as the wires picked up the
inductive components. The pick-up depended on the number of turns enclosed in the measuring loop and it was
lowest for the sensors measuring the contacts (UK-1, 8-OK). Despite the noise, clear trends can be seen of the
voltage and current changes over the test time.
With constant current, the approximate voltage-drop over the coil could be determined. This was compared to a
voltage criterion, e.g. 0.1 μV/cm for 16.5 m of HTS wire (-> Uc=165 μV).
The measured current lead contact voltages were about 175μV at 300 A each (UK-1 and 8-OK). The voltage drop
over the HTS layers (1-8) was smaller.
Figure 19: Measured Voltage Drops for Coil Sp215 at about 28 K vs. Time, Maximum Current 300 A

ZP-ER2011-009
Page 32 of 71
No fast quenching was detected during the tests, and no burn-out of the wire was observed, which suggests a stable
transition voltage regime, considerable margin in the critical current operating reserve, homogenous wire quality, and
the potential for a predictable and measureable transition to quench, all of which is favorable for larger coils.
The critical current at different temperatures could not be determined exactly due to limits of the test setup. An
improvement of the thermal anchoring of the current leads would have been required before the critical current of the
coil can be determined exactly. Because of the high critical current of the wire and the small self-field of the test coil
the expected critical currents at low temperatures are very high.
3.3.4 Conclusions – Improvements – Next Steps
 Improve the thermal anchoring of the current leads to the HTS Coil
 Install compensation coil(s) to reduce inductive voltages during current ramps
 Install a second feed-through for signal wires
 Perform additional measurements at different temperatures and greater accuracy
 Increase the magnet field strength for higher-field measurements
 Perform measurements during quench heater operation
3.4 Budget Period 2
The success of the distribution-class FCL in the SCE Avanti grid and the successful testing of the distribution-class
compact designs at Powertech led ZP to propose accelerating the project and using the next budget period, Budget
Period 2, for building the transmission-class FCL. With the design equations for iron mass, and copper coils well
understood and the benefits of the ZP design approach of separating the high voltage AC from the HTS wire
component of the FCL thoroughly established, ZP embarked on the transmission-class FCL design where the
greatest risk was the electrostatic design of the FCL to meet the BIL requirements. To mitigate the BIL risk, ZP
contracted with several experienced transformer designers to solidify the design. A contract with American Electric
Power (AEP) was signed to install the transmission class FCL in their 138 kV grid at the Tidd substation in Brilliant,
Ohio. Simultaneously Pennsylvania Transformer Technology Inc. in Canonsburg, Pennsylvania was selected as the
contract manufacturer to build the transmission class FCL due to their prior HV experience and manufacturing
capability.
3.4.1 138 kV Transmission Class FCL Design
AEP presented ZP with these design parameters for the 138 kV FCL
Requirements Summary
• 138 kV
• 1300 Arms
• ~ 20 kArms prospective fault
• Reduce fault by 43%
• Recovery under load required
These parameters were then transcribed into a ZP specification (Table 2) with a few modifications:

ZP-ER2011-009
Page 33 of 71
Parameter Data
Name of project for communication purposes AEP Tidd
Location of project: city/state/country Brilliant, Ohio USA
Rated Voltage 138 kV
Line Frequency 60 Hz
Lightning Impulse Voltage Withstand Level 650 kV
Power Frequency Voltage Withstand Level 310 kV (1 minute dry), 275 kV
(10 seconds wet). See footnote
Continuous Normal Current 1300 A-rms
Maximum Normal Current (Magnitude and Duration) 1300 A-rms
Symmetrical Current through FCL for 3-Phase Fault at 138 kV &
X/R for Total Fault
13,775 A-rms & 31.65 ANSI
(without the series 2% reactor)
Symmetrical Current through FCL for L-G Fault at 138 kV & X/R
for Total Fault
18,805 A-rms & 119.26 ANSI
(without the series 2% reactor)
Symmetrical Current through FCL for 3-Phase Fault at 138 kV &
X/R for Total Fault
8,090 A-rms & 33.60 ANSI
(with the series 2% reactor)
Symmetrical Current through FCL for L-G Fault at 138 kV & X/R
for Total Fault
9,831 A-rms & 21 ANSI
(with the series 2% reactor)
Thevenin System Equivalent at 138 kV Side
Z1, Z2, Z0 (Ohms)
0.31343+j2.56709 0.31345+j2.56713
1.68751+j7.46171 (Ohms)
Series Reactor at 138 kV Side = 2% on 100 MVA 3.81 (Ohms)
Thevenin System Equivalent at 345kV Side
Z1, Z2, Z0 (Ohms)
0.34019+j5.46865 0.34044+j5.46871
0.57858+j6.00828 (Ohms)
Transformer Test Report Data:
Rated H/M/T: 339/137.5/13.8 kV
450/450/100 MVA OFAF
Z H-M = 10.84 % @ 450 MVA
ZM-L = 16.58 % @ 100 MVA
Z H-L = 29.08 % @ 100 MVA
Transformer B Is kept disconnected
Table 2: AEP Tidd Substation FCL Parameters

ZP-ER2011-009
Page 34 of 71
An important concept learned from testing the compact FCL full-scale prototypes was that round construction was
better than square construction. Round HTS coils have less of a problem with delamination from the induced
stresses during a fault, and round cryostats are easier to manufacturer. Typically, transformer manufacturers build
stepped round iron cores that have helical copper coils carrying the AC current. In an attempt to follow industry
standards with round construction, the initial design analyses for the transmission-class FCL led to a design having a
round horizontal FCL with round cryostats over the AC tank. One “AC tank” per phase would be used due to the
transportation weight and height restrictions, making for a total of three AC tanks for a three-phase system. The
configuration chosen was a 1x2 (two layers each with a single core) with the stepped round cores axially aligned.
Figure 20: Proposed Location for the Tidd HTS FCL in Brilliant, Ohio

ZP-ER2011-009
Page 35 of 71
3.4.2 138 kV Design Challenges
Figure 21 shows an original concept for the 1x2 single phase FCL for the Tidd location. To reduce the technical risk,
several prototypes of varying scale were built in Australia by Zenergy Power Pty Ltd.
Figure 21: Initial 1x2 FCL Concept for AEP Tidd 138 kV FCL

ZP-ER2011-009
Page 36 of 71
The first prototype of the initial design, as tested at the Australian national Lane Cove test facility, is shown in Figure
23.
Figure 22: Initial Design Model of 1x2 FCL for AEP Tidd

ZP-ER2011-009
Page 37 of 71
As it turned out, during a design review, the DOE Readiness Review Team (RRT) correctly questioned the
performance of the 1x2 design. During the earlier Powertech compact FCL developmental testing, the 3x2 compact
prototype (which has similar construction) did not perform as well as expected. There were also questions raised
regarding whether or not the large voltages induced by the flux into the HTS coils from the fault current would cancel
to acceptable values due to the non-symmetric nature of the 1x2 design. To answer the questions, a half-scale
model was designed, built, and tested by ZP Pty Ltd. Figures 23-25 show the results of the 1x2 prototype testing at
the Lane Cove test facility for the 138 kV prototype test.
Figure 23: Initial Approximately Half-Scale Prototype of 1x2 Design at Lane Cove Test Facility

ZP-ER2011-009
Page 38 of 71
Figure 24: Excellent 1x2 Lane Cove Testing Results for Limiting the Fault Current
Figure 25: Arcing of 1x2 Compact FCL Prototype during Test

ZP-ER2011-009
Page 39 of 71
As predicted by the DOE RRT, the asymmetric design of the 1x2 device experienced imbalances in the
electromechanical and electromagnetic forces induced in the device under fault conditions, which lead to particularly
high induced voltages in the DC coils. This led ZP to rethink its design approach. A summary of the Lane Cove testing
results is presented below; the test results from Lane Cove clearly demonstrated that:
 The proposed 1x2 device was not optimal
 While feasible, it could not be implemented easily
 The design was inherently unbalanced electromagnetically
 It displayed large mechanical forces, high variable AC flux outside device envelope, pronounced shielding
effect, and high induced voltages under maximum fault conditions
The axially aligned cores and AC coils in series did not have the cancellation of flux and induced voltages as
demonstrated in the 2xX designs. Flashovers occurred on the DC windings during the testing and made the 1x2 design
unsuitable for the FCL. Consequently, ZP concluded from its Lane Cove testing of the 1x2 design that, while the 1x2
design was chosen originally for perceived manufacturing ease due to the incorporation of round cores with round coils
on round formers in round tanks, a radical shift in the design approach was needed. This shift led to the adoption of a
“new” 2x1 design that would place two electrical half-phases alongside each other in one AC tank to make a HV single-
phase device. This arrangement, referred to as a 2x1, was previously tested at Powertech in July 2009 during the
compact FCL prototype development testing. Test results suggested the 2x1 design should be more balanced, and
should yield reduced shielding and minimize induced voltage effects. Extensive FEA modeling of the design predicted
significantly improved performance, with the electrical half-phases, which are located side-by-side, significantly
canceling each other’s effects, thereby reducing the induced DC coil voltage and current transients and significantly
reducing the AC flux variations outside of device envelope. The “cost” for this improved performance was an AC tank
structure that would be more complicated to build, with “D-shaped” AC cores and coils inside a round dielectric tank.
Subsequent formal design evaluation meetings led to a change in the project baseline FCL configuration in agreement
with the DOE RRT.
A new 2x1 FCL design was chosen as new project baseline, and a large-scale prototype (Figure 26) was slated for
testing at Lane Cove in July 2010. Plans called for a high-voltage (138 kV L-G) single-phase HTS prototype to be
tested. An AC “test tank” was to be built using production device techniques, and it was to be configured for testing
using HTS magnets from another project (a commercial project that was scheduled to deliver a 1250-amperes,11 kV
three-phase FCL to Applied Superconductors Limited for installation at CE Electric in the UK). If completed, this test
would have resulted in an approximately full-scale (ASL is 3-θ at 11 kV versus 1-θ at 138 kV) testing of a near-
commercial device. Testing was scheduled for the KEMA facility in Chalfont, PA during November 2010, and a
preliminary 2x1 design was completed. The electrostatic analyses were compatible with largest Zenergy production
magnet, and the project device performance was consistent with the RRT’s recommendations.

ZP-ER2011-009
Page 40 of 71
Figure 26: Model of AEP 2x1 Prototype FCL

ZP-ER2011-009
Page 41 of 71
While the original plan was to quickly build an AC HV “test tank” and test at KEMA, Chalfont, PA, using a “borrowed”
set of round magnets from the ASL CE Electric project, rapidly evolving schedule divergences affected the timing of
the two projects and did not allow this to happen. ZP subsequently elected to build another large-scale prototype
(Figure 27) and to test it at the Lane Cove test facility to validate the design approach. The AEP project continued
independently into the design stage of an FCL specific for AEP’s Tidd substation, which was to be modeled and
verified based upon the Lane Cove test results. Figure 28 shows some of the test results from the Lane Cove
experiments, which were very encouraging. Taken with other results, which demonstrated greatly reduced current
and voltage transients on the DC coils, the Lane Cove experiments validated the 2x1 design concept for the AEP
project.
Figure 27: 2x1 Prototype at Lane Cove Testing

ZP-ER2011-009
Page 42 of 71
The Lane Cove large-scale prototype testing led to a significant set of accomplishments including:
 Down-selection of the 2x1 format as the design of choice for all future single-phase HV FCL
applications
 Demonstration of the potential for fault-limiting capabilities in excess of 50% of the peak prospective
fault current values
 Demonstration of a balanced design with low mechanical stresses and low induced voltages on the
DC circuit
 Exhibition of repeatable fault current limiting performance that tracked well with FEA predictions
 Support from the RRT for a decision to proceed with full-scale 138 kV single-phase design for TIDD
substation in which the first test phase would also be the first phase of the commercial installation
Figure 28: Lane Cove Test Results of 2x1Prototype FCL

ZP-ER2011-009
Page 43 of 71
The 2x1 design concept was then applied to the specifications of the AEP Tidd station and the resultant fault current
reduction was calculated as 43.5% (Figure 29). The required iron core mass and AC coils were then designed to fit
within the constraints of the AC tank diameter which was designed to fit within the “warm bore” of the largest HTS
magnets that ZP was then able to produce. The HTS magnet “warm bore” size (open center space of the cryostat)
was limited by the manufacturing VPI oven used to cure the HTS wire on the GFRP former.
As previously discussed, the Lane Cove tests of the 2x1 prototype demonstrated the expected voltage cancelling in
the DC coils, and the FCL transmission-class FCL design was changed to two AC cores and coils, side-by-side, in
one layer oriented vertically. The problem was now constructing two cores and coils in a rounded enclosure to
operate with the electrostatic gradients required for 138 kV. The solution was to design a “D-shaped” core with “D-
shaped” coils as in the prototype. The concept was presented to PTTI and a small prototype “D-shaped” AC coil was
built (Figure. 30) to verify the manufacturability of the concept and the ability to wind the “D-shaped” coils with an
appropriately-rated copper CTC conductor.
Figure 29: 2x1 FCL Predicted 43.5% Fault Current Reduction at the AEP Tidd Substation

ZP-ER2011-009
Page 44 of 71
The next major step was to design for the electrostatic clearances required for 138 kV. With a fixed “warm bore” of
the large cryostats, the challenge was to design as much iron core into the AC tank as can be accommodated with
the helical copper coils and still maintain the required 138 kV electrostatic clearances. Through many weeks of
iteration and the collegial design of three experts in the field, the electrostatic design was finished. Figure 31 shows
the electrostatic design of the “D-shaped” core design in the tank. Figure 32 shows the electrostatic analysis of the
final design. However, to obtain the clearances for 138 kV and maximize the iron core mass available to “toggle”
through the magnetically-biased saturated and unsaturated conditions, the design grew very tall. Several design
proposals were evaluated to incorporate a removable top section housing the bushings that would be removed during
shipping. Top section removal still made the single-phase device too tall for easy transport vertically (the preferred
orientation for the strength and stability of the FCL). A check with the Ohio Department of Transportation revealed
that the transport limiting height was 13.5 ft due to the bridge overhead clearances into Brilliant, Ohio. The vertical
operating position was very desirable for the magnets for the cold heads were designed to function more efficiently
vertically oriented. The solution was to design the interior of the tank such that the frame interior would support the
core and coils for horizontal shipment; however, the final operating position was to be vertical. Calculations showed
Figure 30: Short Test AC Coil Wound with CTC Wire on “D-Shaped” Former at PTTI

ZP-ER2011-009
Page 45 of 71
the iron core and AC coil compression and construction would provide the necessary support for horizontal shipment.
The key factor would be to design the manufacturing fixtures to rotate the cores from horizontal to vertical during
assembly of the FCL.
Figure 31: Final Design of the Core, Coil, Tank Spacing for Electrostatic Compliance

ZP-ER2011-009
Page 46 of 71
Figure 32: Electrostatic Design Analysis for 138 kV AEP Tidd FCL

ZP-ER2011-009
Page 47 of 71
3.4.3 HTS Magnet Design
Zenergy Power GmbH was to be the turn key supplier for the HTS magnets. To mitigate the risk of building a sub-par
performance magnet, ZP GmbH designed a half scale magnet based on copper wire rather than HTS. The prototype
DC coil was built on a G10 former with copper cold conduction plates to emulate the final HTS design. The copper
coil was placed in a laboratory cryostat to evaluate the cold bus design. Figure 33 shows the copper coil used to
verify the thermal design and performance of the magnet. The copper magnet was tested in a laboratory cryostat
with a production cold-head to verify the performance of the coil and the thermal bus design (Figure 34).
Figure 33: Copper Coil on G10 Former used as Test Coil for Heat Transfer Verification

ZP-ER2011-009
Page 48 of 71
Extensive testing was conducted over several weeks to evaluate the test coil. The test routines included several
cool-down and heat-up cycles and evaluations of the temperature distributions around the coil and thermal bus in
relation to the cold-head connection. The copper magnet experiment yielded significant test results that validated the
thermal design, including:
 A half-scale (0.7 m) copper coil constructed and tested
 The “Dry” (conductive) cooling thermal design was optimized (cold-finger location, spacing, dimensions and
shape) and validated for the production magnet
 Performance evaluation included the use of resistive heaters installed at coil edge to simulate AC losses, and
ambient heat input from the DC bias current leads
 A novel, shock-resistant copper cold-bus implemented
 Testing was completed in a cryostat with production cold-heads
 Heat transfer models developed and validated for correlation with the full-scale FCL
Figure 34: Super-insulated Copper Coil in Laboratory Cryostat

ZP-ER2011-009
Page 49 of 71
Results of the copper magnet test are shown in Figure 35. The correlation of the predicted to actual temperature is
very close for both temperature locations on the coil.
3.4.3.1 HTS Coil and Cryostat
The HTS magnet coil was wound with the specification that the OD have a maximum of 1700 mm and the inner
“warm bore” of the cryostat to be a minimum of 1500 mm. The HTS magnet was designed with conductive cooling
utilizing a copper bus connected between the HTS wire and the redundant cold heads that cool the HTS wire to less
than 20 degrees K. Figure 36 shows the magnet covered with super insulation ready to be installed in the cryostat.
The magnets were designed with a thick top plate to which all components (cold heads, lifting hooks, mounting pads,
and electrical connections) are mounted, see Figure 37. With the mounting pads on the top surface of the cryostat,
the magnets are then suspended from the mounting structure keeping the cold heads upright. The structure holding
the magnets is to be built with substantial integrity to withstand the 40,000 pound attractive force between the two
magnets when in use.
Figure 35: Copper Magnet Cool Down to 17 degrees K, Showing Correlation of Predicted to Actual

ZP-ER2011-009
Page 50 of 71
Figure 36: AEP Magnet with Super-Insulation Ready to be Mounted in Cryostat

ZP-ER2011-009
Page 51 of 71
Figure 37: AEP Assembled Cryostat with Components Mounted to Top Plate.

ZP-ER2011-009
Page 52 of 71
The cryostat assembly has a weather resistant enclosure protecting the cold-heads, see Figure 38. All
interconnecting wires are cabled with plug-in connectors for interfacing to the auxiliary enclosure, which houses the
monitor and control instrumentation, and the auxiliary power supplies. The interconnecting wires and cables were
designed on a “plug and play” basis in which the cryostat side of the cables and the auxiliary enclosure side of the
cables were marked and matched to corresponding weatherproof multi-pin connectors on cryostat and the exterior of
the auxiliary enclosure. This reduced the number of connections that needed to be made, improved their integrity
and reliability, and reduced the potential for installation and connection errors.
Figure 38: Assembled Cryostat with Weather-Resistant Housing over Cold-Heads.

ZP-ER2011-009
Page 53 of 71
As shown in Figure 39, the two cold heads in the weather resistant housing are connected to their respective
cryogenic compressors with helium pressure lines and electrical control cables. A closed loop water chiller is
required to provide cooling for the cryogenic compressors. In the final field installation, two units are provided for
redundancy; for shop testing, a portable unit (the large dark grey metal box in the right of the picture) was employed.
Figure 39: Cold-Heads Connected to Cryogenic Compressors through Conduit with Plug Connectors

ZP-ER2011-009
Page 54 of 71
3.4.3.2 Cryostat Cold Performance
Figure 40 documents the excellent cool-down performance of the magnets. The first magnet was assembled and
cooled down to 25 degree K. The process took 3.5 days as only one cold-head could be turned on at a time and
turning on the cryogenic compressor of the second cold-head exceeded the capacity of the portable chiller, and the
over-temperature conditions would shut down the compressors). Two cold-heads were able to be operated in parallel
for short periods of time, and an additional incremental drop in temperature less than 20 K was noted. The jagged
line on the right side of Figure 40 at ~23 K shows the multiple events in which both cold-heads were cycled and the
resulting temperature decreases and increases.
3.4.3.3 AC Losses in HTS Wire
The unknown factor when calculating the performance of the magnet and the amount of cooling required was the AC
losses of the HTS coil during normal operating condition with maximum normal AC load current. AC currents in the
HTS coil result from weak coupling from the flux of the AC current through the AC coils and cause heating of the HTS
coil. In addition to ambient heat leakage from convection and radiation, and conductive inputs from current leads and
direct inputs from DC bias current, the induced AC heat load must also be removed by the cold-heads or the magnet
Figure 40: Cryostat Cool-Down to 25 deg K with One Cold-Head

ZP-ER2011-009
Page 55 of 71
temperature would increase. The extent of these AC losses was unknown for 2G wire. Robert Duckworth of ONL
was asked to perform several experiments to determine the AC loss for future 2G wire designs.
3.4.3.4 Importance of Knowing AC Loss
Due to the efficiencies that result from operating HTS wire at temperatures significantly below transition temperatures
and due to the high cost of HTS, ZP chose to design and operate its HTS magnets at extremely low temperatures.
The objectives of this strategy were to reduce the total cost of HTS conductor used in a FCL system and to increase
the margin to quench by taking advantage of the significantly higher magnetic field performance and IC (critical
current) available at reduced temperatures. But, operating at reduced temperatures significantly increased the
demand on the cryogenic systems and the thermal design of the HTS magnets – among other things, the thermal lift
or cooling capacity of the Cryomech AL-325 cryogenic compressors were reduced by more than two-thirds when
operating at 20K versus 77K. These severe refrigeration requirements made understanding the AC losses of the FCL
magnets extremely important. Among other things:
 Fringing ac fields generate losses in HTS dc coil
 Losses must be characterized and accommodated in thermal design
 Estimates range from 10’s to 100’s of watts per coil – this spread unacceptably complicated the thermal design
 Measurements were needed instead of calculations, because established analytic methods were known to be
inaccurate
 Losses concentrated on coil edge; not uniformly distributed (see Figure 41)
Figure 41: AC Field Strength in Typical HTS Coil

ZP-ER2011-009
Page 56 of 71
3.4.3.5 Excerpts from the Robert Duckworth Report - Executive Summary
As part of the ZP FCL project design, it was determined that the AC losses generated in the DC bias coil from the
fields that were produced by the room temperature AC components of the three-phase FCL circuit were significant.
These AC fields, which were calculated as having values up to 25 mT, would impart a dynamic resistance on different
portions of the DC bias coil, which would in turn generate heat during normal operation. Since the DC bias coil has
regions which operate at different percentages of DC transport current in relation to the local critical current and DC
fields up to 2 T, knowledge of the dependence of the dynamic resistance on these factors would help to determine
the appropriate refrigeration load for the system.
While dynamic resistance has been examined in 1G BSCCO conductors at these operating conditions, 2G YBCO
coated conductors have not been characterized as extensively. Expressions for dynamic resistance that have been
previously used may require some revision due to the difference in aspect ratio between 1G and 2G conductors and
the composition of 2G YBCO coated conductors compared to 1G BSCCO conductors. Dynamic resistance
measurements were carried out at 77 K on an as-manufactured AMSC 2G YBCO coated conductor as a function of
DC transport current with applied DC background fields up to 2 T and applied AC magnetic fields up to 20 mT. These
results were compared to existing dynamic resistance theories to determine best expressions to represent dynamic
resistance for 2G YBCO coated conductors.
With respect to tape orientation, there was a clear difference in the dynamic resistance dependence. The dynamic
resistance due to an applied AC field perpendicular to the tape face is nearly two orders of magnitude higher than the
dynamic resistance with the AC field parallel to the tape face. For the case of AC field parallel to the tape face, no
measurable voltage was observed at applied AC fields up to 25 mT below 85% of the sample critical current. For a
resistance greater than 1 x 10-5 Ω/m with the AC field perpendicular to the tape face, the dynamic resistance had a
linear dependence on the applied AC field. As the fraction of the DC transport current to the sample critical current
was increased, the resistance increased and threshold field where the dynamic resistance appeared decreased.
These functional behaviors of the dynamic resistance were consistent with existing theories; however there was
sufficient difference in the absolute values to suggest refinement of the theories to better represent YBCO coated
conductors.
When DC background fields perpendicular to the tape face were applied, the functional dependence of the dynamic
resistance with respect to applied AC fields and fraction of DC transport current to sample critical current did not
change. The applied DC field did change the dependence of the threshold field with respect to the fraction of DC
transport current to sample critical current. This dependence was consistent with the change in critical current of the
sample due to applied DC field. It was concluded that additional work with AC and DC fields parallel to the tape face
would be beneficial.
3.4.3.6 ORNL Experiment
The extensive characterization experiments of 1G and 2G wires carried out ORNL were performed under the
following conditions:
 AC field (0-20 mT)
 DC field (0 – 2 T)
 Parallel / Perpendicular field orientations
 DC current (0-100% Ic)
 Data collection at 77 K initially, then repeated at 30 K

ZP-ER2011-009
Page 57 of 71
Additional experimentation at other temperatures (20K, 25K, 35K, 40K, 50K and 60K) was suggested for possible
future work. Figure 42 shows one of the test rigs and sample arrangements that were used for collecting data. A
direct refrigeration system was used for data collection at 30K, but the magnetic flux orientations remained consistent
with those shown in Figure 42.
Figure 42: Voltage Arrangement used to Measure Dynamic Resistance in HTS Samples
Liquid
nitrogen
dewar
DC
background
LTS magnet
AMI ac
magnet
Sample
location
S-shaped voltage
taps
I
dc
Bdc Bac
HTS sample

ZP-ER2011-009
Page 58 of 71
Figure 43 shows a typical data set that was collected. With a perpendicular AC magnetic field from 0-25 mT
losses were measured at transport currents ranging from 50% to 90% of critical current. As expected, losses
increased as transport current increased as a percentage of critical current and AC perpendicular magnetic field
increased, but a consistent range of increase of about 300% from 50% to 90% of critical current was identified.
As a result of the ORNL experiments, ZP concluded that:
 While AC losses increased as Iop / Ic and background perpendicular magnetic fields increased, they were
tolerable for the proposed project
 ZP would have been able to optimize HTS magnet design for wire cost versus cryogenics cost
 Using 77 K data, loss from dynamic resistance is between 0.1 mW/m to 15 mW/m for the regimes
contemplated
 Operation of HTS coil at 30 to 50 K would significantly increase the critical current, which would in turn
increase the threshold current
 Measurements on a 1G wire done at 77 K on a cryogenic system matched reasonably well with the
preliminary 2G data and confirmed the first-order approximations of magnet system AC losses
 The custom equipment developed by ORNL was uniquely suited for the purpose and provided 2G wire
performance data that had never before been collected.
Figure 43: Dynamic Resistance as a Function of Peak Perpendicular Field at Different Percentages of DC
Current to the Sample DC Critical Current (No External DC Field)

ZP-ER2011-009
Page 59 of 71
3.4.4 FCL Auxiliary Enclosure
An auxiliary enclosure (Figure 46) was designed to house the cryogenic compressors, DC power supplies, and PLC
monitoring equipment and computer/SCADA interface. The HTS FCL components required monitoring to ensure that
they were operating efficiently and that all components were operating properly. The FCL system was designed to be
built with redundant components to ensure consistently high operational duty cycles and to improve the reliability of
the FCL system.
To ensure proper operation of the HTS FCL, key parameters on the FCL were to be monitored. Several of the
monitoring points included: cryostat temperatures, chiller water temperatures, HTS magnet system DC currents, and
ambient temperatures. Because of the system redundancy, almost one hundred channels of data were to be
collected and monitored. With the goal of easy access in mind, the auxiliary enclosure was designed with several
doors on the front and back. The heavy (500 pound) cryogenic compressors were to be located on the first level, and
the electronics were to be located on the second level.
3.4.5 Zenergy FCL Software
Software used in the development of the HTS FCLs was standard off-the-shelf applications of ANSYS, FEA, and
PSCAD where the variables were the parameters of the FCL. Software developed specifically for the FCL with cost-
share funds was for the HMI monitoring of the FCL. The PLC control software was standard Siemens code, the
Wonderware interface was developed with help from Serra an E&M Company. The process software was specifically
designed for the Siemens PLC to operate the controls for the FCL. A Wonderware - Human Machine Interface (HMI)
program running on a “Toughbook” communicated with the user to provide a graphical depiction of the levels present
and switch positions and operating conditions of the FCL. While automatic control of the FCL was normally the mode
of operation, manual control could be activated for servicing or testing the operation under various parameter
changes. A SCADA data line was also output from the FCL which enabled an external monitor system to record and
display the data at the utility operations center. Figure 44 shows the display of the HMI for configuring the set points,
scan rates, and refresh rates. Figure 45 is an example of the actual computer screen with a graphical representation
showing levels, switch positions, and numeric readouts.

ZP-ER2011-009
Page 60 of 71
Figure 44: Wonderware Parameter Configuration Page

ZP-ER2011-009
Page 61 of 71
3.4.6 FCL Structure
The FCL structure was designed to support the magnets at the correct vertical distance around the AC tank. Figure
47 shows the general external arrangement of the FCL, which was considerably different than the earlier 11 kV FCL
with a similar architecture that was manufactured for ASL and CE Electric in the UK. The 11 kV, two-magnet FCL
used an external support structure to position the magnets. For the AEP device, the AC tank was to be a structural
component in order to reduce the overall external dimensions of the FCL. In the AEP design, the bottom magnet was
to be placed in its holder mounted to the foundation, and the tank inserted into the bottom magnet by a crane. The
top magnet would have been attached to supports cantilevered off of the “top hat” portion of the AC tank prior to the
entire assembly having been lowered into the bottom tank. The radiators would have been attached last. The AC
Figure 45: Wonderware Human Machine Interface Display

ZP-ER2011-009
Page 62 of 71
tank and cryostats were design for continuous outside exposure without an additional enclosure. The cold-heads
were to be protected by a weatherproof housing, the interconnecting cables were to be sheathed in a conduit, and the
electrical interfaces were to have weatherproof plug-in connections for ease of assembly.
Figure 46: Model of Auxiliary Enclosure for AEP FCL

ZP-ER2011-009
Page 63 of 71
Figure 47: AEP Single-Phase FCL Model

ZP-ER2011-009
Page 64 of 71
Figure 48 shows the overall dimensions and weights of a single-phase of the AEP FCL. Due to the finite and
somewhat limited warm-bore radius of the cryostat, the active core area of the FCL was limited, and the FCL AC tank
had to grow in length to meet the iron mass required for the fault reduction. The FCL height exceeded allowable road
clearances, so the AC tank design was modified in order to allow the FCL to be shipped horizontally. While the AC
tank design was proceeding, other vendors were contracted to design and build the auxiliary enclosure to house the
cryogenic compressors, DC power supplies, and control electronics. ZP in Germany was building the two complete
HTS magnet subsystems, and the first magnet had only been cryogenic tested when the project was terminated.
Purchase orders were issued for the iron cores, CTC copper for the coils, and the insulation material for the coil
winding.
3.4.7 Factors Affecting the FCL Timeline
Several factors influenced the timetable for the Transmission Class FCL. ZP was informed by PTTI, the
manufacturing partner that had been selected for the project, that the manufacturing timetable would slip due to:
1. The loss of manufacturing engineers on the shop floor needed for the manufacturing improvements to handle the
long horizontal cores and modifications to the coil winding lathe required to upgrade it to handle the long coils.
Figure 48: AEP FCL Overall Schematic

ZP-ER2011-009
Page 65 of 71
2. The incompatibility of the software between ZP and PTT, which necessitated that printed manufacturing drawings
would have to be produced in order to check tolerances and make judgments regarding the final manufacturability of
the design.
ZP also met with AEP to finalize the planned interconnections with the grid and to review testing and installation
requirements. The two-day meeting with AEP generated many questions and answers and much activity to close the
gaps. However, the question of installation costs which had been determined at the beginning of the contract to be
$850,000 and were to be split by ZP and AEP had been revised by AEP to a figure of at least $1.6 M, with additional
costs possible and expected, and neither AEP nor ZP had sufficient budget to absorb the additional costs.
ZP had also determined that no power lab in North America could test the 138 kV FCL to its full limits at full voltage
and full current. The only test facility able to test at full voltage and current was at KEMA in The Netherlands. KEMA
in Chalfont, PA could perform limited current testing at reduced voltage, but the FCL would not receive the full stress
under test conditions. AEP was concerned with placing the FCL into their grid, if it had not been tested to the full
stress it could experience under maximum fault conditions. AEP asked that ZP discuss this matter with NEETRAC,
DOE, and the Readiness Review Team. The RRT recommended that the FCL be installed with the air core reactors
(ACRs) (which were already in place at the Tidd station) always in series with the FCL. The ACRs would insure that
the FCL would not receive any greater stress than was experienced during the laboratory testing available at KEMA
Chalfont, PA.
3.4.8 Zenergy Power Change of Focus
A major change occurred in early September 2011 with this project. The ZP Australian team determined that with
proper iron core and magnetic circuit design techniques, it would be possible to use only one superconductor magnet
to saturate the iron cores of the FCL. Further investigation led to determining that further enhancements to the
magnetic circuit design with judiciously placed iron would further reduce the number of Ampere-turns (ATs) required
to saturate the iron core. The extension of this research concluded that the ATs required for saturation was within the
capabilities of DC power supplies and conventional resistive copper DC coils. This profound realization was an
epiphany for the inductive FCL technology, since suitable FCL performance could be achieved without the cost of the
HTS wire, associated cryostats, cold-heads, cryogenic compressors, chillers, and the maintenance, training, cost to
operate, and the poor reliability of these elements. While these conclusions were the result of an analytical study,
there was no indication from a physics point of view that it would not work. A fast-paced test of the concept
performed in mid October 2011 in Australia, verified the concept and proved the efficacy of the conventional DC
magnet inductive FCL.
The ZP FCLs have been built with magnets made from 1G HTS wire since 2007. The plan was to utilize 2G HTS
wire when the availability and cost made the switch feasible. Unfortunately 2G HTS conductor was not generally
available in quantity during the performance of this project, nor did the price come down substantially from the initial
short-length and test and qualification sample offerings from the 2008 and 2009 timeframe. ZP’s FCL program was
predicated on the availability 2G HTS conductor with greatly increased performance and reduced price that would
bring the cost of saturated reactor FCLs more in line with the expectations of the utility companies.
Simon Cleaver, the CEO of Zenergy Power, PLC (Zenergy Power, Incorporated’s parent company) recognized the
business opportunity in a less expensive, HTS-free FCL and after careful consideration, choose to focus the
company’s Australian and United States teams toward making the non-HTS FCL a commercial success. ZP has built
several FCLs, and through extensive testing, has proven the Saturable Reactor FCL concept. Having a FCL with no
cryogenics, requiring far less maintenance, consuming less auxiliary power and with the appearance, reliability,
functionality and maintenance profile of standard utility equipment will greatly ease the adoption of the saturated

ZP-ER2011-009
Page 66 of 71
reactor FCL by the utility industry. The FCLs will now fit seamlessly into utility substation configurations without any
adjustment in utility operations and maintenance practices, or field maintenance and repair capabilities.
The ramifications of Mr. Cleaver’s decision were several. First, all POs for the AEP project were put on hold and then
subsequently cancelled, once DOE concurred with project termination. Fortunately the POs of the major components
of iron, copper, and insulation materials were stopped in time to not incur any penalties. Second, DOE concurred
with ZP’s assessment, and through DOE Program Manager Mr. Don Geiling, ZP was given permission to terminate
the project. Third, AEP was informed of the Zenergy Power decision and given several options for future FCLs.
The original goal for the DOE funding of the Zenergy Power project “Design, Test and Demonstration of Saturable
Reactor High-Temperature Superconductor Fault Current Limiters” was to stimulate the HTS wire industry by
developing commercial products which would consume massive quantities of, first 1G, then 2G, HTS conductor.
Since the premise of the original funding was to stimulate the HTS wire industry, and since ZP had concluded that for
the foreseeable future its saturated reactor FCL was not economically viable with HTS wire, the DOE and Zenergy
Power mutually terminated the project.
4 Summary and Conclusions
Zenergy Power has successfully designed, built, tested, and installed in the US electrical grid a saturable reactor
Fault Current Limiter. Beginning in 2007, first as SC Power Systems and from 2008 as Zenergy Power, Inc., ZP used
DOE matching grant and ARRA funds to help refine the design of the saturated reactor fault current limiter. ZP
ultimately perfected the design of the saturated reactor FCL to the point that ZP could reliably design a suitable FCL
for most utility applications. Beginning with a very basic FCL design using 1G HTS for a coil housed in a LN2 cryostat
for the DC bias magnet, the technology progressed to a commercial system that was offered for sale internationally.
Substantial progress was made in two areas. First, the cryogenics cooling system progressed from a sub-cooled
liquid nitrogen container housing the HTS coils to cryostats utilizing dry conduction cooling and reaching
temperatures down to less than 20 degrees K. Large, round cryostats with “warm bore” diameters of 1.7 meters
enabled the design of large tanks to hold the AC components. Second, the design of the AC part of the FCL was
refined from a six legged “spider” design to a more compact and lighter design with better fault current limiting
capability. Further refinement of the flux path and core shape led to an efficient saturated reactor design requiring
less Ampere-turns to saturate the core. In conclusion, the development of the saturable reactor FCL led to a more
efficient design not requiring HTS magnets and their associated peripheral equipment, which yielded a more
economical product in line with the electric utility industry expectations.
The original goal for the DOE funding of the ZP project “Design, Test and Demonstration of Saturable Reactor High-
Temperature Superconductor Fault Current Limiters” was to stimulate the HTS wire industry with, first 1G, then 2G,
HTS wire applications. Over the approximately 5 years of ZP’s product development program, the amount of HTS
wire employed per FCL and its cost as a percentage of the total FCL product content had not dropped substantially
from an unsustainable level of more than 50% of the total cost of the FCL, nor had the availability increased (today
the availability of 2G wire for commercial applications outside of specific partnerships with the leading 2G wire
manufacturers is extremely limited). ZP had projected a very significant commercial potential for FCLs with higher
performance and lower costs compared to the initial models built with 1G wire, which would come about from the
widespread availability of low-cost, high-performance 2G HTS wire. The potential for 2G wires at greatly reduced
performance-based prices compared to 1G HTS conductor held out the potential for the commercial production of
FCLs at price and performance levels attractive to the utility industry. However, the price of HTS wire did not drop as
expected and today the available quantities of 2G wire are limited, and the price is higher than the currently available
supplies of 1G wire. The commercial option for ZP to provide a reliable and reasonably priced FCL to the utility
industry is to employ conventional resistive conductor DC electromagnets to bias the FCL. Since the premise of the

ZP-ER2011-009
Page 67 of 71
original funding was to stimulate the HTS wire industry and ZP concluded that copper-based magnets were more
economical for the foreseeable future, DOE and ZP decided to mutually terminate the project.
5 Acknowledgements
ZP would like to recognize and thank the DOE Readiness Review Team for their support and contributions to the
development of the saturable reactor FCL technology and Dr. Robert Duckworth of ORNL for his groundbreaking
research in measuring and modeling the AC losses of 2G HTS conductor under a variety of operational
environments.

ZP-ER2011-009
Page 68 of 71
6 Index of Figures
Figure 1: AC Coils with Iron Cores Saturated by DC-Powered HTS Coil 8
Figure 2: First FCL Produced by Zenergy Power 10
Figure 3: Zenergy Power’s 15 kV FCL at HV Test 13
Figure 4: Zenergy Power’s 15 kV FCL Installed at SCE Avanti Substation with Bypass Switch 14
Figure 5: Calculated FCL Limiting of 17% of 21 kA Fault 15
Figure 6: Powertech Result Showing 19% Fault Reduction of 23 kA Fault 15
Figure 7: Three-Second Fault at SCE Avanti Sub-station, 14 January 2010 17
Figure 8: Spider and Compact FCL Design Examples 18
Figure 9: Mini 6x1 Feasibility Prototype 20
Figure 10: Comparison of the Compact Prototype Models (L to R) 6x1, 3x2, 2x1 21
Figure 11: Full-Scale 6x1 Prototype under Test at Powertech Labs 22
Figure 14: Zenergy Trailer Setup as Rolling Test Bed at Powertech Labs 25
Figure 15: Measured versus Calculated 15 kA Prospective and Limited Fault Current – 6x1 FCL 26
Figure 16: Measured versus Calculated 15 kA Prospective and Limited Fault Current – 6x1 FCL 27
Figure 17: Test Coil before Impregnation with Resin 30
Figure 18: 2G Test Coil Mounted in Cryostat 30
Figure 19: Measured Voltage Drops for Coil Sp215 at about 28 K vs. Time, Maximum Current 300 A 31
Figure 20: Proposed Location for the Tidd HTS FCL in Brilliant, Ohio 34
Figure 21: Initial 1x2 FCL Concept for AEP Tidd 138 kV FCL 35
Figure 22: Initial Design Model of 1x2 FCL for AEP Tidd 36
Figure 23: Initial Approximately Half-Scale Prototype of 1x2 Design at Lane Cove Test Facility 37
Figure 24: Excellent 1x2 Lane Cove Testing Results for Limiting the Fault Current 38
Figure 25: Arcing of 1x2 Compact FCL Prototype during Test 38
Figure 26: Model of AEP 2x1 Prototype FCL 40
Figure 27: 2x1 Prototype at Lane Cove Testing 41
Figure 28: Lane Cove Test Results of 2x1Prototype FCL 42
Figure 29: 2x1 FCL Predicted 43.5% Fault Current Reduction at the AEP Tidd Substation 43

ZP-ER2011-009
Page 69 of 71
Figure 30: Short Test AC Coil Wound with CTC Wire on “D-Shaped” Former at PTTI 44
Figure 31: Final Design of the Core, Coil, Tank Spacing for Electrostatic Compliance 45
Figure 32: Electrostatic Design Analysis for 138 kV AEP Tidd FCL 46
Figure 33: Copper Coil on G10 Former used as Test Coil for Heat Transfer Verification 47
Figure 34: Super-insulated Copper Coil in Laboratory Cryostat 48
Figure 35: Copper Magnet Cool Down to 17 degrees K, Showing Correlation of Predicted to Actual 49
Figure 36: AEP Magnet with Super-Insulation Ready to be Mounted in Cryostat 50
Figure 37: AEP Assembled Cryostat with Components Mounted to Top Plate. 51
Figure 38: Assembled Cryostat with Weather-Resistant Housing over Cold-Heads. 52
Figure 39: Cold-Heads Connected to Cryogenic Compressors through Conduit with Plug Connectors 53
Figure 40: Cryostat Cool-Down to 25 deg K with One Cold-Head 54
Figure 41: AC Field Strength in Typical HTS Coil 55
Figure 42: Voltage Arrangement used to Measure Dynamic Resistance in HTS Samples 57
Figure 43: Dynamic Resistance as a Function of Peak Perpendicular Field at Different Percentages of DC Current to
the Sample DC Critical Current (No External DC Field) 58
Figure 44: Wonderware Parameter Configuration Page 60
Figure 45: Wonderware Human Machine Interface Display 61
Figure 46: Model of Auxiliary Enclosure for AEP FCL 62
Figure 47: AEP Single-Phase FCL Model 63
Figure 48: AEP FCL Overall Schematic 64
7 Index of Tables
Table 1: Comparison of Spider versus Compact Design 18
Table 2: AEP Tidd Substation FCL Parameters 33
8 List of Acronyms and Definitions
8.1 Acronyms
ACR Air Core Reactor
ARRA American Recovery and Reinvestment Act
AT Ampere Turn
BIL Basic Insulation Level
BP Budget Period
DOE Department of Energy
FCL Fault Current Limiter

ZP-ER2011-009
Page 70 of 71
HTS High Temperature Superconductor
HVAC Heating Ventilating and Cooling
LN2 Liquid Nitrogen
ONL Oak Ridge National Laboratory
PLC Programmable Logic Controller
PTTI Pennsylvania Transformer Technology Inc.
RRT Readiness Review Team
SCADA Supervisory, Control and Data Acquisition
SCE Southern California Edison
ZP Zenergy Power, Inc
8.2 Definitions
None
9 References
1. Robert Duckworth, “Dynamic Resistance Measurement at 77 K of Zenergy FCL Project Update”, January 2010
2. Robert Duckworth, “ORNL contribution to Zenergy FCL presentation” (2010 Peer Review), June 2010
3. Franco Moriconi, “High Voltage Testing”, Powertech Labs, Surry, BC Canada, Dec 2007 (SCP/TR-07/03)
10 Appendix. Public Documents
10.1 Patents Granted or Applied For, Software Developed
1. Patent documentation submitted under Intellectual Property Provisions (CSB-1003)
Cooperative Agreement Research, Development, or Demonstration Domestic Small Businesses
2. Software for Wonderware HMI developed under this grant submitted under DOE Form 241.4
10.2 Publications
1. “2G HTS Wire Testing”, Dr Jan Wiezoreck, for „Design Test and Demonstration of Saturable Reactor High-
Temperature Superconducting Fault Current Limiters“ DE-FC26-07NT43242, 21 September 2009
2. “Analysis of the Effect of a Saturable-Core HTS Fault Current Limiter on the Circuit Breaker Transient Recovery
Voltage”, J. Lopez-Roldan, A. C. Price, F. De La Rosa, and F. Moriconi, 2011 IEEE PES General Meeting,
Detroit, Michigan, July 24-29, 2011
3. “Modeling and Test Validation of a 15kV 24MVA Superconducting Fault Current Limiter”, Franco Moriconi, Nick
Koshnick, Francisco De La Rosa, and Amandeep Singh, 2010 IEEE T&D Conference, N. Orleans April 19-22,
2010.
4. “An Innovative Compact Saturable-Core HTS Fault Current Limiter - Development, Testing and Application to
Transmission Class Networks”, Franco Moriconi, Francisco De La Rosa, Amandeep Singh, Bill Chen, Marina
Levitskaya, Albert Nelson, IEEE Transmission and Distribution, November 24, 2010, 2010 IEEE PES
General Meeting, July 26-29, 2010, Minneapolis, Minnesota.
5. “Resonance of a Distribution Feeder with a Saturable Core Fault Current Limiter”, Christopher R. Clarke, Franco
Moriconi, Amandeep Singh, Ardalan Kamiab, Russ Neal, Alonso Rodriguez, Francisco De La Rosa, and Nick

ZP-ER2011-009
Page 71 of 71
Koshnick, IEEE Transmission and Distribution Conference and Exposition, November 24, 2010, N. Orleans April
19-22, 2010.
6. “Development and Deployment of Saturated-Core Fault Current Limiters in Distribution and Transmission
Substations”, Franco Moriconi, Francisco De La Rosa, Frank Darmann, Albert Nelson, and Larry Masur, Applied
Superconductivity Conference August 2-6, 2010, Washington, DC, USA. Also published in Trans., on Applied
Superconductivity, Vol. 21, Issue 3, pp 1288-1293, June 2011
7. “Saturated-Core Fault Current Limiter Field Experience at a Distribution Substation”, Albert Nelson, Larry Masur,
Franco Moriconi, Francisco De La Rosa, Detlev Kirsten, C I R E D, 21st International Conference on Electricity
Distribution, Frankfurt, Germany,6-9 June 2011
8. “Wind Power Plant Enhancement with a Fault-Current Limiter”, E. Muljadi and V. Gevorgian, F. DeLaRosa,
IEEE PES Annual Meeting, Detroit, Michigan, July 24-29, 2011
10.3 Technical Presentations
1. “Design, Test and Demonstration of Saturable-Core Reactor HTS Fault Current Limiter”, Larry Masur, Frank
Darmann, Robert Lombaerde, Presented at the U.S. Department of Energy Annual Peer Review, Crystal City,
VA, July 31, 2008
2. “Design, Test and Demonstration of Saturable-Core Reactor HTS Fault Current Limiter”, Franco Moriconi,
Frank Darmann, Robert Lombaerde, U.S. Department of Energy Annual Peer Review, Alexandria, VA, Aug 4-
6, 2009
3. “Design, Test and Demonstration of Saturable-Core Reactor HTS Fault Current Limiter”, Albert Nelson, Franco
Moriconi, Robert Lombaerde, U.S. Department of Energy Annual Peer Review, Alexandria, VA, June 29 – July
1, 2010
4. ASL 11 kV FCL for the CE Electric Grid, Newcastle, UK, Albert Nelson, Franco Moriconi, EPRI
Superconductivity Conference, Tallahassee, Florida, October 12, 2011

Design, Test and Demonstration of Saturable Reactor High-Temperature Superconductor Fault Current Limiters

More Related Content

What's hot (17)

Viewers also liked (10)

Similar to Design, Test and Demonstration of Saturable Reactor High-Temperature Superconductor Fault Current Limiters (20)

More from Franco Moriconi (9)

Recently uploaded (20)

Design, Test and Demonstration of Saturable Reactor High-Temperature Superconductor Fault Current Limiters