199

SUPERCONDUCTING CABLES
IMPACT ON NETWORK STRUCTURE
AND CONTROL

Task Force
38.01.11

February 2002
 SUPERCONDUCTING CABLES
IMPACT ON NETWORKS STRUCTURE
AND CONTROL

TASK FORCE
38.01.11

Copyright © 2005
“Ownership of a CIGRE publication, whether in paper form or on electronic support only infers right of
use for personal purposes. Are prohibited, except if explicitly agreed by CIGRE, total or partial
reproduction of the publication for use other than personal and transfer to a third party; hence
circulation on any intranet or other company network is forbidden”.
Disclaimer notice
“CIGRE gives no warranty or assurance about the contents of this publication, nor does it accept any
responsibility, as to the accuracy or exhaustiveness of the information. All implied warranties and
conditions are excluded to the maximum extent permitted by law”.
 Contents

- Members of the task force

1. Introduction .....................................................................................................p1

2. Power transmission and technology of superconductivity .................................p2

2.1 Features of superconducting cables
2.2 Impacts of introducing superconducting cables into power networks
2.3 Technical issues for the application of superconducting cables

3. Present Status of superconducting cable technologies and development...........p5

3.1 Present status of superconducting cable technologies
3.2 R & D projects in the world

4. Electro-magnetic properties of superconducting power transmission cables.....p22

4.1 Superconducting cable with low temperature dielectric
4.2 Superconducting cable with room temperature dielectric
(Reference: Conventional cables)
4.3 Comparison of characteristics of cold dielectric and room temperature dielectric
superconducting cables

5. Superconducting cable transmission systems and their ratings..........................p32

5.1 Basic superconducting cable system design and performance
5.2 Ratings of superconducting cables
5.3 System performance
5.4 Technical issues concerning system reliability
5.5 Operations and problems when a fault occurs in superconducting cable
systems

6. Conclusion ......................................................................................................p38

App. I. Synthesis of the answers to questionnaire on conventional and

superconducting cables in power grids
App. II. Effects of thermal insulation on constants of RTD SC cables
 Member List of the SCTF 38.01.11

Convenor Seiji Sekine Japan

Secretary Noboru Higuchi Japan
Members Wolfgang Blendl UK
Christophe Boisseau France
Donald W. Von Dollen USA
Klaus Fröhlich Switzerland
Alan Jenner Australia
Pierluigi Ladie Italy
Evanise Neves de Mesquita Brazil
Yasuhiro Noro Japan
Ole Tønnesen Denmark
Observers Aldo Bolza Italy
Bjørløw-Larsen Kjell Norway
Haruhito Taniguchi Japan

Contributed
persons Laura Gherardi Italy
Michael Gibbard Australia
Nelson Martins Brazil
Steve Norman UK

Shirabe Akita Japan

Masanao Egawa Japan
Nobuo Fukushima Japan
Shoichi Honjoh Japan
Yoshihiro Iwata Japan
Noriyuki Kimura Japan
Shin-ichi Mukoyama Japan
1. Introduction
This report is a distillation of many discussions on the possible impacts of
superconducting (SC) power cables in power transmission systems in the future. In this
report, only oxide superconductors, so-called “high temperature superconductors” are
discussed, since metallic superconductors will never be used for power cables due to
economic reasons and superconducting characteristics.
In order to clarify the possibility of using SC cables, the Task Force collected and
discussed worldwide background data concerning underground (or undersea) cables,
and analyzed the trend of SC cable applications in future power grids.
Two types of SC cable, which have different technical and economic
characteristics as described below, have been studied to obtain a high degree of
flexibility for use in power grids with high power density. The use of SC cables is
expected to reduce the right-of-way, and so SC power cables are a possible candidate
for replacing conventional XLPE cables or GILs.
SC cables for use in power systems are now under development around the world, but
there are many difficulties to estimate the manufacturing and installation costs. Several
kinds of conceptual design have been proposed, and field testing in commercial systems
is now planned.
Under these circumstances, the Task Force was organized in CIGRE SC-38.
The mission of the Task Force is to clarify the basic network characteristics of SC
cables and their potential use in networks; to compile key information on the impacts of
SC cables on the construction and operation of networks; and to analyze the
fundamental characteristics of SC cables in power systems.
Legal restrictions and public awareness of the need to preserve the environment are
becoming increasingly strict in most parts of the world, making it difficult to construct
overhead transmission lines in the suburbs of big cities and even in rural areas. Only
underground transmission systems will be accepted, even though these are much more
expensive than overhead line systems, and the cost of acquiring land such as for tunnels
and trenches, including right-of-way, will account for most of the total cost.
Laying SC cables in existing trenches of conventional cables is an attractive
option, since an SC cable may carry several times the power of a conventional cable.
New SC transmission cable systems are expected to reduce right-of-way and to make it
easy and economical to plan underground transmission systems. Especially, the
replacement of existing conventional cables with SC cables is one of the most attractive
applications.

-1-
2. Power transmission and superconductor technology
2.1 Features of superconducting cables
The key feature of SC power transmission cables is that large power
transmission with extremely small losses is possible. With SC cables, the operating
temperature of the conductor is controlled by a coolant within a suitable range to
maintain the superconducting state. Due to extremely small Joule losses and appropriate
temperature control, the cable is not subject to transmitting current restrictions
dependent on thermal tolerance. However, regarding transmitting power efficiency, SC
cables have not only Joule losses but also hysteresis losses and dielectric losses, and the
efficiency of the refrigerator liquefier must also be taken into account when evaluating
the total loss of the cable system. Cable losses at liquid nitrogen temperature have to be
estimated by considering the efficiency of the Carnot cycle and that of the refrigerator.
That is, a loss of 1W at the temperature of LN2 will be equivalent to 14.5W, assuming
the efficiency of the refrigerator is about 20%. In order to express the performance, the
‘penalty’ of the refrigeration is defined as the operating power ratio of refrigerators to
remove losses at the operating temperature. In this case, it is 14.5. Equivalent loss will
decrease with improvement of the refrigerator. Even so, the current restriction due to
thermal capacity (critical temperature of superconducting materials) of cable conductors
will be far less severe compared with conventional cables, and so superconducting
cables will allow low voltage and large current transmission systems to be constructed.
Conventional cables are generally operated at a high voltage and small current
to reduce the Joule losses and to suppress the temperature rise of the conductor.
However, this increases the no-load charging current and restricts the possible cable
length to transmit a given amount of power.
Since the development of long-length SC tapes (BSCCO), a number of SC
power cable projects have emerged around the world. Due to the remarkable
development of SC tape processing technologies, the use of SC cables operating at
liquid nitrogen temperature for power networks has become a much more attractive
prospect. However, SC cable materials are still being developed to improve the critical
current and amount of current that can be carried, and researches on geometrical cable
design, manufacturing processes and reduction of AC losses are also still underway.
Compared with conventional cable conductors, superconductors have
intrinsically different characteristics, that is, there exists a transition between the
superconducting state and the normal state, which depends on the current density,
operating temperature and magnetic field. The development of SC materials having high
Tc and high Jc (high Je: engineering current density) with low AC losses is essential for
SC cables, and the manufacturing process must also be improved to achieve
cost-effective, reliable cables.

2.2 Impacts of introducing SC cables into power networks

There have been many discussions on the merits of using SC cables around the
world. The key benefits of SC cables over conventional cables (XLPE) or even
-2-
overhead lines are as follows.
a. Capacity
SC cables can carry large currents at high current density, so large amounts of
power can be conveyed at a relatively low line voltage.
b. Efficiency
Because of the very small AC losses, total transmission power losses are
reduced, even taking into consideration the penalty of refrigerators (operating power
ratio of refrigerators to remove losses at liquid nitrogen) in estimating the total losses.
Reduction of losses also reduces the impact on the environment (lower emissions of
CO2).
c. Compactness
Because of the high current density and low voltage operation, cables (and
auxiliary equipment) can be made compact and lightweight. Therefore, the cost of
constructing transmission lines is much lower than that of conventional lines, and it is
easier to obtain right-of-way or to use existing tunnels and trenches effectively. Also,
operation of the cable system at low voltage decreases the cable charging current, and
so less VAR compensating equipment (shunt reactor) is needed. This also decreases the
losses of shunt reactors, and so reduces the total losses of the cable system.

The impacts of introducing SC power cables into the existing power grid are as
follows.
a. Environmental issues
In every country, environmental protection requirements are becoming
increasingly strict. Installation of overhead lines is becoming very difficult due to issues
such as legal regulations, scenic beauty, right-of-way and so on, especially in urban
areas, and so demand for underground cables is growing at the expense of high voltage
overhead transmission systems. SC cable systems that can transmit large power at low
voltage and high power density are thus expected to have a major impact on power
system configurations.
b. Power systems in populated areas
In urban areas, there is little space available for sub-stations above ground. In
many countries, the power grid is becoming crowded and public demand for measures
to prevent disfiguration of the scenery by high voltage overhead lines or open-air
substations means that transmission facilities must be placed underground. Underground
substations and cable systems will inevitably become necessary, and compactness and
lightweight are important factors when designing an underground power grid. Moreover,
thermal radiation due to the transmission power losses must be minimized to prevent
thermal pollution of the soil.
SC transmission systems meet these requirements and there is no risk of fire as
liquid nitrogen coolant is used.

-3-
c. Necessity of close cooperation
SC cable systems offer many benefits for the supply of electric power with
minimal environmental impact. However, the industry cannot afford to pay exorbitant
prices to construct the power transmission system, and the cost of producing SC cables
is a major concern. To develop economically viable SC cable systems and low life-cycle
cost networks, close partnership between power engineers and SC engineers is crucial.

2.3 Technical issues for the application of superconducting cables

Generally, the application of SC technology in the power industry is expected
to make a major contribution to solving technical and social problems. SC cable
development is promising, and work is underway around the world. The acceptance of
this new technology primarily depends on the development of reliable superconductors
itself. Notable progress has been made in this decade on investigations and
manufacturing of HTS conductors, but cable designing is still at the development stage.
Technical issues concerning cable system configurations and operations still exist, and
they closely depend on the cable concept and design.

Typical technical issues for the application of SC cables are as follows:

a. Evaluation of economics and environmental benefits
· Installation cost including corridor
· Cost of cooling system
· Life-cycle cost evaluation
· Maintenance cost (cost evaluation including transformers: SC cables are capable of
transmitting large power at low voltage and, in some systems, transformers can be
eliminated)
b. How to ensure power feeding security
· Reliability of SC cables and refrigeration systems
· Investigation and construction of suitable cable network with adequate redundancy,
where operational reliability of the refrigerator, maintenance procedures and so on
have to be taken into account
· Capability to withstand short-circuit current (adequate SC cable design
corresponding to S/N transition, development of fault current limiters)
· Development of reliable SC cables in case of cable line faults (close investigation
of S/N transition phenomena of SC cables)
c. Clarification of SC cable system performance
· System behavior in case of overload and fault
· Analysis of grounding system
· Balance of surge impedance of the SC cable system
· Damping characteristics in SC cable systems

-4-
3. Present status of superconducting cable technologies and development

The first field-testing of an SC cable assembled using BSCCO tapes began last
spring in 2000 as a power supply to a factory. However, there still exist many technical
issues to be solved, and further R&D is required for full-scale utilization. Typical
technical issues and major R&D projects are described in this chapter.

3.1 Present status of SC cable technologies

3.1.1 Superconductors
There are two candidate superconducting materials for the superconductors of
SC cables: BSCCO and YBCO. Intensive development of both materials is underway
around the world to make wires of high current density. At present, BSCCO wires are
used to assemble SC cables, because longer wires with high current density such as
1000-meter ones with a practical current density of ten kilo-amperes per square
centimeter can be made.
Some of the important characteristics of BSCCO wires are as follows.
• AC losses: Reduction of AC losses of short wires has been greatly improved, but
for practical long wires, AC losses must be reduced.
• Mechanical strength: By using a silver alloy for the sheaths of the wires,
mechanical characteristics were greatly improved. As a result, less deterioration
was observed in assembled conductors, and the allowable bending radii became
much smaller, making SC wires easier to handle.
• Shape of wires: Most of the wires are in the shape of tapes, because of the
processing to obtain higher current densities, but round wires have advantages
for assembling conductors. Further research is required.
Even though YBCO has great potential to carry higher current at lower losses, wires
long enough to assemble test cables are still not available; further R&D is required.

3.1.2 Conductor assembling

The current capacity of each superconducting wire is not so large, typically tens of
amperes, thus many wires must be assembled to attain the designed current capacity of
several thousand amperes.
The method of assembling conductors is one of the most important issues, because
sophisticated designs are required to obtain a uniform current distribution in conductor
assemblies. Uneven current distributions would result in insufficient current capacity
and excessive AC losses of conductor assemblies.
The achievable current capacity of an assembled conductor depends on the
structure of the model, so respective cases should be considered as shown in section 3.2.
Investigation of the thermal and mechanical properties of conductor assemblies is
also important. Since high-temperature superconductors are brittle, and the stress caused
by thermal contraction due to the temperature difference between cable assembling and
operation may damage them. Recently, remarkable progress has been made.
-5-
3.1.3 Electric insulation
Two types of SC cable are under development, and the major difference is in the
electric insulation. The first type of SC cable adopts conventional electric insulation and
so it is not necessary to develop SC cables of higher line voltage such as 275 kV or
higher. These are called room temperature dielectric (RTD) type SC cables. Electric
insulation in the second type of SC cable is at cryogenic temperature, which requires
extensive studies; these are called cold dielectric (CD) type SC cables.
The difference affects the cable characteristics considerably as described later.

3.1.4 Long distance cooling

Superconductors of SC cables must be cooled below the critical temperature
along the whole length during operation. Nitrogen liquefiers are in use around the world
at present, but refrigerators still require improvements in terms of both efficiency and
reliability for use in power systems. Other than the development of reliable refrigerators,
technical issues concerning cooling remain to be solved, such as the initial cooling, the
design of cooling channels, the design of return paths of the coolant and so on.
The coolant will be sub-cooled liquid nitrogen, since saturated liquid nitrogen
generates bubbles that cause unacceptable pressure drops along the cooling channels.
The cooling distance, which is the distance between cooling stations, depends on the
dimensions of cooling channels available and losses, and optimization is required to
minimize the cost. In the case of RTD SC cables, the cooling channels of phase
conductors are independent of each other, and the features are quite simple. However,
interactions between the coolants in CD SC cables must be considered when the
conductors are cooled in the same channel. Heat exchange inside the channel may make
cooling more complex, so much work is underway in this field.

3.1.5 Others
There are important hardware technical issues other than those described above,
such as those concerning terminations, joints, installation techniques, diagnostic skill
and so on. Further effort is required for the practical use of SC cables.

-6-
3.2 R & D projects in the world
Many efforts have been made worldwide, and some of them are based on
international collaborations. Typical ones are presented here. Each one has its own
distinctive target, so the achieved stages should be evaluated respectively.

3.2.1 R&D in Denmark

1) Introduction
Superconducting power cables have been of considerable interest since the
discovery of the high Tc superconducting materials. They represent the most drastic
change to the power cable industry since 1880. The Danish superconducting power
cable project consists of: three NKT companies, namely NKT Research, NKT Cables,
and Nordic Superconducting Technologies (NST), the Institutes of Electric Power
Engineering (ELTEK) and Process Technique (IPT) of the Technical University of
Denmark, the national laboratory Risø, the research centre of the Danish power utilities
DEFU, the Danish utilities Eltra, Elkraft and Københavns Energi, and the Ministry of
Energy. The Danish project use conductors made of the (Bi,Pb)2Sr2Ca2Cu3O10+x
(Bi2223) material (made by NST). Attractive is the larger amount of current and energy
that can be transferred using superconductors compared to copper cables and the energy
savings that can be obtained with the superconductor operating with liquid nitrogen
cooling. Here the advancements and status of a project aimed at demonstrating a 36 kV,
2 kArms AC cable system by installing a 30 m long full-scale functional model in a
power utility are summarised. The expected benefits of such a system includes reduced
energy loss and increased power rating in a small cross-section. Electrical loss below 1
W/m/phase at 2 kArms has been obtained in these conductors.

2) The Cable
Each of the 30 m long cables (total number is 3, for a 3 phase transmission line)
consists of several different components: the former (containing the coolant), the
superconductor itself, the thermal insulation, the electrical insulation and a screen and
protection layer. The cable is of the room temperature dielectric (RTD) type as shown in
figures 1 and 2. In this case a stainless steel corrugated tube was used as a cooling pipe
(also called former). The eight superconducting layers are then wound on top of the
former using an adapted industrial paper winding machine. In this process the conductor
is spooled up on the uptake onto a spool with 1.5m bending radius. The longest length
wound up to date is 60m (to become 2x30m, cut after winding). Bending tests show that
the strain in the cable during bending does not exceed the maximum strain allowed. The
critical current of the superconducting cable was found to decrease somewhat (2.5%
after 19 bends and 14 thermal cycles). However, this decrease is not considered
seriously affecting the performance. No difference in ac loss before and after the
bending was found. It should be noted that the cable model also experienced many
thermal cycles during this bending test. Both thermal cycling and mechanical strain are
known to have a possible effect on the performance of the superconductor, which in our
-7-
Figure 3-1. Schematic of the NKT superconducting
power cable.

Figure 3-2. Cut of actual cable.

case turned out to be virtually negligible.
The thermal insulation (cryostat) of the cable consists of two concentric corrugated
stainless steel tubes that between them contain multilayer insulation and spacers. In
order to avoid thermal leakage through gas convection the space between the two
corrugated tubes is evacuated. The thermal leak rate was measured to be about 1.5
W/m/phase with liquid nitrogen on the inside, and room temperature on the outside. In
our case it can be seen that the thermal loss is significantly higher than the ac loss - thus
at this point it is of some importance to improve the thermal loss.
The electrical insulation consists of an industry standard polyethylene layer that is
extruded onto the cryostat. On top of that layer follows a standard copper screen with a
protection layer as outermost surface. The electrical insulation passed tests with
standard routine high voltage partial discharge tests (10s at 42kV dc, 300s at 63kV ac).

3) Accessories
Typically cable splices, terminations, stress cones, etc. are referred to as cable
accessories. In the framework of this article the most important accessory to the cable is
the termination where a thermal and electrical termination have been integrated into a
compact entity. The termination has several tasks:
• to transfer the current from room temperature to the superconducting cable at low
temperature (typically 70-80K) and vice versa. In this case this is done by a copper
rod that carries the current through the temperature gradient to copper end pieces
onto which the superconductor is soldered.
• to transfer the coolant - in this case liquid nitrogen - in and out of the cable.
• to accept all mechanical forces that rise from the cooling of the cable from room to
low temperature.
The whole body of the termination is situated at high electrical potential.
However, the thermal out-coupling is at ground potential. The potential in this case falls
to ground over the whole length of the insulator shown, which on the inside only
consists of electrically non-conducting materials. The terminations passed the standard
routine high voltage tests. The heat leak through the terminations is of order 85 to 150W,
depending on current load. For a 3 phase cable system 6 of the terminations shown are
needed.

4) Utility installation
It is planned to install the superconducting cable in a substation in the Danish
power grid close to Copenhagen. This will be at the 36kV level. The superconducting
cable will connect two high current (2kA) bus bars. Thus this installation will allow an
evaluation of the performance of the system as a whole as well as allow the utilities
involved to gain experience with the operation of a superconducting power system.

5) Conclusion
• A factory made 30 m long 36kV 2kArms superconducting three phase cable has been
-8-
 produced.
• It will be installed and integrated in a large power substation close to Copenhagen.
• The superconducting cable will operate under the same conditions as conventional
power cables.
• The superconducting cable will fulfil all functions of a demonstration cable.
Records of the cable performance will give the necessary data for further
development.

-9-
3.2.2 R&D in Italy and USA
1) General
In response to the combined effects of growing energy demand and the impact
of de-regulation of the electrical energy industry, utilities are ensuring flexibility and
robustness of their networks, upgrading, where necessary, installed capacity in both in
transmission and distribution. In this regard, high-temperature superconducting cable
systems offer advantages where space, thermal capability and environmental constraints
dominate.
Two basic HTS cable designs are recognised for electric power transmission;
Warm Dielectric design and the Cold Dielectric Coaxial design.
For the Warm Dielectric design (WD), only the HTS conductor assembly is
enclosed in the cryogenic environment; the WD design is best suited for retrofitting
existing duct or pipe systems and has the potential to carry more than twice the power
of an existing cable with approximately the same losses.
In the Cold Dielectric Coaxial design the two concentric HTS operate in the
cryogenic environment. Cold Dielectric Coaxial cables promise the greatest level of
transmission capacity and efficiency thanks to the superconducting coaxial “return”
conductor. The potential power capacity for the CDC design exceeds a conventional
circuit by 500%, while having specific losses (kW lost per MVA delivered) that are over
30% lower.

2) HTS cable system development programmes

Pirelli is involved in several HTS R&D projects (for both WD and CD design)
each with distinct, specific goals ranging from system design studies to experimental
development and demonstration of complete HTS cable systems. The approach adopted

Effect of cabling and jointing

Cable ; Ic = 3100 A
100
Conductor ; Ic = 3320 A
System ; Ic = 2860 A
y = 9E-40x11.185
y = 5E-35x9.9322
10
µV/cm

y = 2E-44x12.428

0.1
1000 10000
DC current (Amp)

Fig. 3-3 Effect of cabling and jointing

- 10 -
for all these programmes has involved initial feasibility studies in order to analyse the
suitability and effectiveness of various HTS system options within the context of a
specific transmission

Table 3-1 R&D projects

Developer Main Partners Cable type Present milestone

Pirelli DOE, EPRI, ASC 115 kV, 50 m complete system (cable,
WD joint, termination, cooling),HV
laboratory test completed (2
kA rms) in Jan 1999
Pirelli Detroit. Edison, DOE, 24 kV, WD 120 m, 3 single phases field
EPRI, ASC, Lotepro test in a utility network (2.4
kArms) in ‘01/’02, retrofit in
existing ducts of 4“
Pirelli EdF 225 kV, CD 30÷50 m complete system
(cable, joint, termination,
cooling), 2,6 kArms in ‘01
/’02
Pirelli Edison, ENEL, CESI 132 kV, CD 30 m 1- phase, 3 kArms field
trials in ‘02

3) Pirelli_EPRI 115 kV cable system

Thus as early as 1993, Pirelli and EPRI commenced their first feasibility
evaluation of an HTS pipe-type retrofit cable that was subsequently finalised at the end
of 94, under the US D CABLE DESIGN
epartment & KEYSuperco
of Energy DATA nductivity Partnership Initiative
(SPI), to the development of a complete prototype cable system.
PIPE
Liquid N2
FLUID
SKID WIRES
BSSCO HTS Tape (ASC)

Tape Specification (1994):

LAPPED PPL Ic DC > 28 A
DIELECTRIC
AC Loss < 10-4 W/m @ 7 A rms & 55 Hz
Strain tolerance: 0.3%@77 K, Ic/Ico > 95%
CRYOSTAT
Thermal Tolerance: Ic/Ico > 90%

HTS
CONDUCTOR Conductor Design:
3-Core Cable in Pipe HTSTapes arranged in 8 layers

Fig.3-4 Cable design and key data

- 11 -
 The conductor design was based on BSCCO2223 HTS tapes supplied by
American Superconductor Corporation arranged in 8-layer.
Ampacity tests were undertaken on each tape before cabling, on the completed
«bare» conductor, on the cable after completion of all manufacturing stages, and on the
finished cable system assembly comprising cable, terminations and one in-line joint.
Before installing the joint and the terminations d.c. measurements were performed
varying the operating temperature. The Ic measurement confirmed that almost no
degradation of the BSCCO tape performance took place through all the steps of the
cable manufacturing. Fig. 1 shows results for the ampacity at 77 K for the conductor,
cable, and system respectively. It can be seen that the performance of the conductor in
the complete system exceeded the targeted minimum performance of Ic > 2800A d.c..
The electrical performances of the insulation system, joint, and terminations were
checked with a complete range of Type Tests, generally in accordance with AEIC
CS2/90, including a.c. withstand tests to 165kV a.c. for 24 hrs, 205 kV for 1 minute,
and lightning impulse tests at +/-550kV (10 pulses each).

4) Detroit Edison 24 kV HTS cable project

The demonstration circuit will be installed in Detroit Edison’s Frisbie substation.
It connects the secondary of a 120 kV-24 kV transformer, with a circuit rating of 100
MVA, to 24 kV bus in the building approximately 120 m away. At full power, the circuit
will be responsible to carry 2400 A
The cable employs a Warm Dielectric design, taking advantage of the compact
construction to fit into the nominal 100mm diameter ducts.

Fig.3-5 Conceptual view of the conductor assembly

The conductor assembly will have four layers of Ag-sheathed Bi2Ca2Sr2Cu3Ox

tapes wound around a flexible former. The tapes will have a minimum critical current
Ic,DC ∼ 110A, measured over the total length of the tape at 77K and in self-field.
The winding angles of the different layers are chosen in such a way to:
- limit the strain experienced by the HTS tapes due to thermal contraction and cable
bending
- achieve the current distribution among the HTS layers as uniform as possible
- minimise the magnetic field acting on a single tape in the cable configuration to

- 12 -
 increase the cable critical current and to reduce the AC losses
The demonstration cable uses Ethylene Propylene Rubber (EPR) insulation. The
design of the insulation system was derived from techniques and standards.
The demonstration cable will include six cable terminations and a single joint.
The terminations designed for the demonstration system follows the same
philosophy as for the cable design; to separate the dielectric and thermal insulation
systems. To perform this, a conventional outdoor insulator is combined with a novel
cryogenic portion.
A joint will be included to complete the accessories which would be necessary for
system installation. The techniques for completing the joint are based on the joint
developed during the EPRI 115 kV program, and will employ pre-moulded, slip-on joint
bodies.

5) Pirelli-ENEL-Edison HTS Cable System Development

In May 1999 Edison, ENEL and Pirelli announced a joint program to develop a
cold dielectric cable system; the system is designed for a service voltage of 132 kV and
a nominal current of 3 kArms. The scope of the cable system development includes the
HTS cable itself, accessories, cryostats, cooling devices, and measuring and control
systems. The 30 m single phase cable prototype together with the accessories and the
cryogenic system will be manufactured, installed and tested by 2003.

3.2.3 R&D in France and Italy

1) Pirelli-EDF HTS Cable System Development
In October 1997 Electricité de France (EDF) and Pirelli announced a joint project on high-power
HTS cable link. The system is designed to carry a total power of 3000 MVA at 225 kV with 4
circuits in parallel with a total length of 10 km.
Several main areas of investigations were identified to characterise all the key
elements of the superconducting coaxial cable system starting from the HTS and the
dielectric materials to the cable, accessories and auxiliary systems. Based on the
positive results of the experimental phase, the construction of full size prototype of the
system components is under way. The first set of full-size prototype test will start in
2000.

HTS Tapes
Extensive research work was undertaken in collaboration with American
Superconductor Corporation (ASC) to develop HTS tapes offering superior tensile
strain performance, and to define the design limits for the maximum allowable stresses
both at room and LN temperature. Fig. 2 shows the typical ampacity retention and
stress-strain characteristic versus tensile stress (Ic/Ico) at 77K in LN for BSCCO tape :

- 13 -
 Ic/Ico Ic/Ico Strain Strain
1 0.008

0.007
0.8
0.006

0.6 0.005

0.004
0.4 0.003

0.002
0.2
0.001

0 0
0 100 200 300 400 500 600

Stress at 77K (MPa)

Fig. 3-6 Effects of stress on the tapes

HTS phase and return conductors

The Bean model was assumed to predict the losses both for the phase and the
return superconductor. A target of 0.6 W/m for the phase conductor was fixed. This led
to the design of two layers phase and return superconductors, each layer being formed
by 24 HTS tapes. The layers were applied in a counter-helical way with an angle of 30°
to minimise the axial magnetic field.
The specified level of short-circuit current was 31.5 kA for a duration of 0.5 s.

Electrical insulation
A number of dielectric materials were characterised in liquid nitrogen at 77 K
and 1 atmosphere as flat or cylindrical models.
Polypropylene Paper Laminate (PPL) tapes were selected because they exhibit
the best combination of performance properties between the different needs: easiness of
application, dry mechanical performances, easiness of impregnation and importantly the
best dielectric strength.

Outdoor termination design

The design of the termination represents a real challenge: it must control the
electrical field in LN environment, realise the transition between LN temperature and
ambient temperature and sustain the mechanical forces due to the LN pressure and
thermo-mechanical effects.
Among a few possibilities, three solutions were developed up to a concept
design study; all three solutions foresaw a pass-through bushing from the cable
dielectric to LN dielectric contained inside a pressurised vessel. For the first alternative,
a simple thermally isolated current lead was considered to effect the transition between
the LN environment inside the casing and the external ambient.
For the second alternative, an epoxy-resin bushing, embodying the current lead and also
acting as a stress-relief cone was considered. These two solutions were rejected due to
electrical stress considerations.

- 14 -
 For the third solution, an epoxy-resin bushing containing the current lead and
embodying a stress-grading capacitor for the control of the electric field was adopted.
This design solution is very promising from the mechanical, electrical and thermal
points of view. The schematic construction of this termination is shown in Fig. 3-5

Fig. 3-7 Schematic drawing of the outdoor termination

1 – Conventional outdoor insulator
2 – Epoxy-resin bushing
3 – Thermally isolated vessel
4 – Pressurised LN
5 – Corona shields
6 – Epoxy-resin insulator

Straight joint design

The joint design was derived from conventional fluid-filled cable one. For the
extra-insulation the same PPL material used for the cable were adopted up to an outer
insulation diameter of approx. 215 mm.

3.2.4 R&D in Japan

Tokyo Electric Power Company, TEPCO, is developing 66kV, 350-700MVA
superconducting cables with low temperature dielectric collaborating with Sumitomo
Electric Industries (SEI) and Furukawa Electric Co. (Furukawa), respectively. Their
target is the development of a compact and three-phase type superconducting cable
small enough to install existing cable ducts 150mm in inner-diameter. Their R&D
activities are listed below.

- 15 -
 Table 3-2 R&D activities

Year Milestone
1993 5m/3000A flexible conductor was developed, and bending test to a
radius of 1.3m was conducted without any degradation.
1994 7m three-phase/1000A cable, having a dielectric layer and a
superconducting shielding layer on each phase conductor, was
developed. Outer diameter was set to 130mm, which is the development
target.
1995 50m/2000A-class conductors were developed using proto-machines for
combining wires.
1997 30m single-phase/1000A cable system including outdoor terminations
and closed cycle cooling system using sub-cooled LN2 was developed.
100hours of 40kV/1000A loading test was successfully conducted.

As described above, the element technologies needed for realizing a high-Tc

superconducting power cable have proceeded smoothly, clearing development
objectives at each stage. In addition to these activities, it is important to certify the
manufacturing ability and the practicability of the high-Tc superconducting cable system
as actual power system equipment. For these purposes, TEPCO has begun with SEI a
new project in which they are developing a 100m long three-phase 66kV/1kA/100MVA
high-Tc superconducting cable system prototype by integrating element technologies,
and conduct long-term loading tests in collaboration with the Central Research Institute
of Electric Power Industry (CRIEPI). The specifications of this model system were set
as described above in consideration of the current technical level of each element and
the content to be verified.
There are, mainly, three objectives. The first is the establishment of newly
required manufacturing technologies, including a conductor structure to reduce AC loss,
or the so called spiral pitch adjustment technique, measures to handle heat contraction
force, and a splitter box to separate three conductors and lead them to each termination.
The second is to establish technology for transporting parts made at the factory to the
testing site and assembling them into a system on site. The third is to conduct long-term
current-voltage loading tests while continuously operating a cooling system in which
sub-cooled liquid nitrogen is re-circulated, and to evaluate the state of superconductivity,
dielectric property, and thermal insulation performance of the terminations and thermal
insulation pipe.
This project began in October 1999 with cable design researches. They have
tested 30m preliminary cable system to check the feasibility of cable design in October
2000, and then they have started the test of 100m cable system on June 2001 at the test
site of CRIEPI for a one-year period.
A 30m long preliminary cable system has been under testing in order to check
- 16 -
the propriety of cable design, and to verify the manufacturing and integration
technology of the cable system. This preliminary cable consists of three conductors
having different roles; one is used for current loading test, one is for voltage test, and
one is for mechanical property test. The cable is bent into an L-shape to check the
effects of bending on cable characteristics. Theoretical Ic value of these parts is 2378A
and 2738A, respectively. This theoretical estimation takes into account the effects of
both the mechanical force and the magnetic field on the Ic of wires. Test results were
2416A and 2718A, respectively. Both are in good agreement with the theoretical
estimation, indicating there is no additional reduction factor in wire property during
cable manufacturing, integration and cooling down. According to the theoretical
estimation, if the current is completely uniform among layers, here supposing the
number of superconducting layers as k, AC loss is reduced to 1/k of that of the
mono-block model when the peak of sinusoidal current reaches the Ic value. They
confirmed that AC loss of conductor consisting of 4 layers of superconducting layer was
reduced to 1/4 at around Ic, indicating that the current is uniformly distributed among
layers by so-called pitch adjustment technique, in which each layer’s pitch angle is
adjusted respectively so as to equalize each layer’s inductance. Heat cycle test and other
tests are now under way.
100m-long cable is bent into a U-shape in the middle and is bent back 180
degrees. Three core conductors are twisted together so that the bending distortion will
be uniformly applied to the three cores, with a space inserted between the three cores
for the absorption of heat contraction force during cooling. Splitter boxes are used to
separate three conductors and lead them to each termination. They are also used for
short connection of three magnetic shielding layers.
Through the integration tests, they intend to extract problems that might be
difficult to foresee in individual parts tests. The next step is to undertake new
developments to solve these problems. Beside this task, there are other aspects that have
been inadequate and have not been handled in the research and development conducted
thus far. One is the scaling up of conductor capacity by further improving characteristics
of bismuth-based wires. Feasibility study of manufacturing a conductor using
next-generation wires is also important since it might be another solution for scaling up
the capacity. The next challenge is to reduce AC loss by adjusting the structure of the
conductor, and also the wire itself.
It must also be taken into account the long-term characteristics of dielectric layer,
short circuit withstanding characteristics, cooling characteristics of cable in the order of
kilometer, and the development of cable jointing technology for the use of
superconducting cables in an actual power network, as well as the development of
connecting technology with conventional power equipment such as GIS (gas insulated
switch).
Through these research and development activities, they expect to start feasible
study with installing a long-length superconducting test cable to an actual power
network around 2010.
- 17 -
 Superconducting conductor
PVC jacket
Liquid nitrogen PPLP insulation layer
Former

Supeconducting
layer Superconducting
shielding layer
Inner corrugated pipe
Vacuum insulation layer
Superconducting Outer corrugated pipe
cable

130mm Superconducting wire

Fig.3-8 Outline of the 100m cable structure.

(TEPCO/SEI)
Fig. 3-9 Cross section of 100m cable.

- 18 -
 XLPE Cable
Transformer

Duct
Splitter Boxes φ150mm

Current
Transformer

Cooling System 100m Three-Core HTSC cable

for Terminations
Termination
Cooling System
for Cable

(TEPCO/SEI)
Fig.3-10 Outline of 100m cable for verification tests.

Chubu Electric Power Company, CEPCO, is also making efforts to develop

superconducting cables. CRIEPI has made a network analysis proposing a model
network including superconducting cables as shown in Fig.3-11. The results showed
improved voltage stability in the system.

3.2.5 R&D in Switzerland

1) The PSEL project.

A 5m cable based on HTSC were developed and tested by EPFL-CRPP- Brugg
Cable AG and Metallwerke Dornach. under the PSEL contract.
The SC-strip based on Bi 2223 and Ag was manufactured by Vacuumschmelze Hanau in
Germany.
Novel about the task was to cool the conductor with gaseous Ne to 50-60K
The current density of the SC went up from 1570 A at 77K to 4900 A at 52 K.
Losses from the leads to the SC cable were 8.5 W, through the super insulation 0.6 W.
Electrical losses: 0.3W/m.
Taking the efficiency of the refrigerator as 20 %, considering Carnot efficiency
the electrical losses were 0.16W/MW/m which is a factor of two (2) lower than a
conventional cable with 200 mm Cu cross section.
Cable data: Operational current 2.5 kA, Voltage 110 kV 50 Hz.

2) The ETHZ- EPFL-ABB studies

Extensive theoretical studies were carried out to compare the current carrying
capacity of HTSC and conventional cables, economical aspects, electrical losses,
- 19 -
thermal losses, at LN2 and converted to room temperature, refrigeration, connection of
SC to the main grid.
Four case studies were looked upon:
a. Transmission of three-phase cable across the Alps (389 kV, 1.5 kA, 100 km long).
b. Energy input from rural area into a city.(220 kV; 24 kV ac and 24 kV dc)
Maximum transmission capacity 2x300 MW, 800 A at 220 kV
Maximum transmission capacity 600 MW, 52 kA at 24 kV.
In the later case, dc cables were studied and the efficiency of present inverter,
converter taken into account.

3) HTSL - Material Research

Activities on Bi2223 strips are supported by the university of Geneva, where the
substrates (still Ag and Ag alloys) are studied in combination with the superconductor.
The conductor has not reached commercial level yet.
Activities on Bi2212 are mainly performed by research groups of ABB. Power
switching using superconducting elements, current limiters on inductive basis have been
advanced to a high degree and even used in combination with three phase commercial
cables.
An inductive current limiter was installed in a power plant and operated for about 6
months. Also a three phase transformer with superconducting primary and secondary
windings was built and tested. To our knowledge both items are the only devices used in
combination with conventional energy systems.

3.2.6 USA
DOE has been sponsoring three activities of R&D on superconducting cables. The
first one is cosponsored with EPRI, and developed by Pirelli and American
Superconductor Corporation, ASC. A 50-meter cable was assembled and tested to carry
3300A dc in 1996.
Oak Ridge National Laboratory, ORNL, Southwire Co. and others organized the
second one, and Technical University of Denmark joined in. They successfully
demonstrated the cable for the factory power supply of Southwire. This is of CD type
the rating is 1250A, 12.4kV, and it is the first SC cable assembled with high temperature
superconductors (BSCCO) installed and operated in the power system of practical use.

- 20 -
 500kV-100km 500kV-200km
Overhead-line Systems

500kV-100km
40GW 4GW

66kV-SCC 66kV-SCC
30km 30km
Superconducting Cable Systems

- 21 -
66kV-SCC 66kV-SCC
20km 66kV-SCC 66kV-SCC 20km
10km 10km

P=1.5 P=0.4 P=0.4 P=1.5

1GW
Fig. 3-11 A model network with SC cable systems
4. Electro-magnetic properties of superconducting power transmission cables
4.1 Superconducting cable with low temperature dielectric
A conceptual view of an SC cable with low temperature dielectric is shown in Fig.4-1. Two
superconducting layers are laid for each phase, one as a phase conductor and the other as a shield.
Thermal insulation Superconductor

Electric insulation

Protection layer

d
D
Superconductor
Normal conductor
Electric insulation
Thermal insulation
Protection layer

D: inner diameter of magnetic shielding, d: diameter of the superconductor for transport current
Fig. 4-1 Conceptual view of an SC cable with low temperature dielectric.

4.1.1 Fundamental items to determine the properties of SC cables

a. Superconductor for transport current (phase conductor)
In most of the designs, the conductors of SC cables consist of helically wound
superconducting wires or tapes, and the magnetic field is generated inside and outside
of this conductor layer while the electric field is generated at the surface of the
conductor. The outer diameter, d, is used for approximately calculating the inductance,
L, and the capacitance, C, of the cable.
b. Shielding superconductor
The magnetic field generated by the transport current will be shielded by the
induced current in this conductor. When the three phase currents are balanced, no fields
would exist outside. The electric field is also confined inside of this conductor, so the
inner diameter, D, can be used for the estimation of L and C.
c. Insulation materials

- 22 -
 Dielectric constant of insulation materials, ε, should be considered when
determining C.

4.1.2 The equivalent circuit and the values of the constants

Lp
C M
Ls

Fig. 4-2. Equivalent circuit of an SC cable with low temperature dielectric

The equivalent circuit is shown in Fig. 4-2. It is quite simple and consisted of
the inductance of phase conductor, Lp and the capacitance, C derived by the two
superconducting layers. The self-inductance of the shielding conductor is equivalent to
the mutual inductance between the phase conductor and shielding conductor. When
the shielding conductors of the three phases are connected to each other as shown in Fig.
4-3, both of the positive and negative phase sequence currents cancel each other and
only the zero phase sequence current generates magnetic field outside of the cables. .

Fig.4-3. Conceptual view of a connection of the shielding layers.

The positive phase sequence inductance, which represents the characteristics of

cables, L, and the capacitance, C, are shown below,
L=µ0/2π·ln(D/d)
C=2πε0 εs /ln(D/d)
That is;
L=0.4605 log (D/d) [mH/km]
C=0.02406ε s /log(D/d) [µF/km]

- 23 -
4.2 Superconducting cable with room temperature dielectric
A conceptual view of an SC cable with room temperature dielectric is shown in
Fig.4-4.
Superconductor
Thermal insulation
Electric insulation
Normal conductor
Protection layer

d
Di
S Do

S: distance between centers of phase conductors, Do: outer diameter of the electric insulation layer,
Di: inner diameter of electric insulation layer, d: outer diameter of the superconductor.

Fig.4-4 Conceptual view of a superconducting cable with room temperature dielectric.

4.2.1 Fundamental items to determine the properties

a. Superconductor for transport current (phase conductor)
The situation concerning inductance in RTD SC cables is similar to that in CD
SC cables, but no electric field exists around the conductor.
b. Thermal insulation
This layer itself does not affect the electric and magnetic characteristics. But
the thickness changes the diameters of the electric insulation.
c. Electric insulation with normal conductors
This layer is laid outside of the thermal insulation. Normal and
semi-conductive layers are laid both of outside and the inside of this layer to fix the
potential for the reliable electric insulation. The inner normal conductor should be
connected to the phase conductor electrically at the termination, and the outer normal
conductor is connected to the ground potential. As a result, these layers become paths
of the charging current.

- 24 -
d. Insulation materials
Relative dielectric constant of insulation materials, εs, should be taken into
account.
e. Installation profile of cable
This type of SC cable does not have a shielding superconductor, so the features
of magnetic field out side of the cables are similar to the conventional ones.
Consequently, the inductance of the cable depends on the profile of cable installation,
whether they are laid in parallel or in trefoil, like conventional cables.

4.2.2 Equivalent circuit and values of the constants

1) Approximate values
Precise considerations are required to estimate the constants, but approximate
values can be obtained as follows. The inductance, L, is roughly determined using
the outer diameter of the superconductor and the distance of the center of the phase
conductors, like conventional cables. The capacitance, C, depends on the specification
of the electric insulation as follows,
L=µ0 /2π·ln(2S/d)
C=2πε0 εs /ln(Do/Di)
That is;
L=0.4605 log (2S/d) [mH/km]
C=0.02406ε s /log(Do/Di) [µF/km]
Where, the notations are the same to those shown in Fig.4-4.
As a result, the inductances and capacities of RTD SC cables are similar to those of conventional
cables.

2) Effects of thermal insulation on constants

The equivalent circuit is shown in Fig. 4-5 for the estimation of the effects of
thermal insulation below the electric insulation on the electric constants of RTD SC
cables.
．
V1
． Phase superconductor
Vs ． ． ．
Z1 ． Z1 Vr
Y１ Thermal insulation
．
V2 Normal conductor
at high potential
． ． ．
Z2 Y2 Z2 Electric insulation

Normal conductor
at ground potential
D
Fig. 4-5. Equivalent circuit of an SC cable with room temperature dielectric
- 25 -
 The impedances should be chosen based on not only line voltage drops but also
voltages across the thermal insulation, which are generated by the charging current.
．
Z1 is the impedance of the superconducting phase conductor, which mainly
consists of inductance. The inductance is determined by the outer diameter and the
．
distance between the centers of phase conductors. Z2 is the impedance of the inner
normal conductor, which is at high potential, and consists of the resistance and the
．
mutual inductance of the two conductors. Y1 is the admittance between the
．
superconductor and the inner normal conductor, and Y2 is the admittance of the electric
insulation.
Usually, the thermal insulation is not designed to withstand high voltage, so the
．
voltage across Y1 caused by the charging current must be kept sufficiently low, typically
several hundred volts. For that purpose, the resistance of the normal conductor R2
should be controlled in the cases of long distance and high line voltage cables. Because
too high R2 may cause break down of the thermal insulation whereas too low R2 may
increase the current in the normal conductor such that losses become intolerable.
Details are given in Appendix II.

- 26 -
 (Reference: Conventional cables)
R-1. General features of conventional cables
Fig. 4-R-1 shows a conceptual view of a conventional cable for the comparison
with SC cables. RTD SC cables and these cables are the same other than the conductors.
R-2. Cable constants

Normal conductor
Electric insulation
Protection layer

S D

S: distance between centers of cable conductors, D: outer diameter of

electric insulation layer, d: diameter of cable conductors.

Fig. 4-R-1. Conceptual view of a conventional cable.

Cable constants of conventional cables are as follows.

L=µ0/2π·ln(2S/d)
C=2πε0 εs /ln(D/d)
That is;
L=0.4605 log (2S/d) [mH/km]
C=0.02406ε s /log(D/d) [µF/km]

In RTD SC cables, the conductors of the conventional cables are replaced with
thermally insulated superconductors, and so their cable constants are similar.

- 27 -
 4.3 Comparison of characteristics of cold dielectric and room temperature dielectric
superconducting cables
In case of conventional cables, higher voltage classes are preferable to reduce
Joule losses and the amount of the conductors. So, the voltage ratings are chosen as high
as possible. On the contrary, lower voltage classes are to be selected for SC cable
systems because they can carry large currents without increasing losses and the costs of
auxiliary equipments are cheaper in the lower voltage class systems. However, there
exist lower limits determined by several factors, such as the critical currents of
superconductors, the system stability and so on, and the limits are different in the cases
of RTD and CD SC cables.
This chapter compares conventional cables, RTD and CD SC cables.

4.3.1 Comparison of constants

In the comparison with CD SC cables, the inductances of RTD SC cables are
much larger than those of CD SC cables. Since the magnetic field generated by the
transport currents of RTD SC cables is not confined inside of the cables as in
conventional ones. As a result, the voltage ratings of RTD SC cables are to be chosen
higher than those of CD SC cables from the point of the view of voltage stability. Hence
the capacitances become smaller with the increase of the insulation thickness. Typical
cases are shown in Table 4-1.

Table 4-1. Comparison of constants of cables for 1 circuit

%base=1GVA
CD SC cable RTD SC cable Conventional cable
Number of circuits 1 2 2
Transmission voltage/current 66kV/9kA 115kV/2.5kA 275kV/1kA
-1
Reactance X(Ωkm ) 0.0165 0.086 0.114
base X(Ω) 4.36 13.2 75.6
-1
%X(%km ) 0.378 0.652 0.151
Y(10 Ω km )
-4 -1 -1
Admittance 1.53 1.84 1.19
-1
base Y(Ω ) 0.23 0.075 0.0132
-1
%Y(%km ) 0.0665 0.245 0.902
Surge impedance R0 (Ω) 10.39 21.62 30.95
Propagation constant β (10-3) 1.59 3.98 3.69
Surge impedance loading(SIL) P(MW/cct) 419 611 2444
P(MW/route) 419 1222 4888
Positive and Negative
-1
Sequence Reactance X1 (=X2) (Ωkm ) 0.0165 0.086 0.114
-1
Zero Sequence Reactance X0(Ωkm ) 0.0165 0.26 0.26

- 28 -
Note that this example does not mean SC cables cannot be used in the high voltage
systems. The table shows CD SC cables have the possibility to transport the huge bulk
power, such as 1GVA at the system voltage of 66kV. If the stability problems in the
SC cable system occur due to the low system voltage, a higher voltage class, such as
154 kV, 275 kV or even 500 kV may be used without serious technical problems. In this
sense, SC cables offer a wider choice of electric parameters compared with
conventional cables. In the comparison between RTD and CD SC cables, the latter ones
are more flexible concerning parameter selection.

4.3.2 Critical cable length

Power transmission cables have three kinds of critical lengths due to charging
currents, voltage rises and phase shifts, denoted as Lcc, Lcv and Lcs, respectively. In the
case of conventional cables, the first one is dominant and practically the only item to be
considered. However, these lengths must be reconsidered for SC cables.
The critical cable length Lcc due to the maximum allowable charging current Icm is
given by:
Icm
Lcc = (4-1)
(V / 3) ⋅ω ⋅ C

where C is the cable capacitance per unit length, ω the angular frequency and V the line
voltage of the cable. Icm of the conventional cable can be determined by the heat
generation of the cable and the thermal resistance for the cooling. On the contrary, Icm
of the superconducting cable will primarily depend on the critical current of the
superconductor, which is an extremely large value.
From the point of views of the cable system operation, cable length will be
restricted to maintain the voltage rise at the receiving end under the no-load operation
(Feranti effect). The critical cable length, Lcv determined by the allowable no-load
voltage at the receiving terminal, Vr, is given by:

1 Vr
L cv = cosh −1 (4-2)
β Vs
where β is propagation velocity and Vs sending end voltage.
Concerning the critical cable length, Lcs due to power transmission stability, the
transmission power P between two points is generally given by the following equation:
Vs ⋅Vr
P = sin δ s (4-3)
Xs + X c

Where,
Vs, Vr ; sending and receiving end line voltages,
Xs ; system reactance except the cable,
Xc ; cable reactance = R0sinh(β·Lcs),
- 29 -
 R0 ; characteristic impedance = L / C ,

β ; propagation velocity = ω LC ,

δs ; phase shift of the voltages at the both ends.

Transmission power at δs =900 means the maximum capable transmission power of
the system.
To maintain stable operation of the system under transient operating conditions, δs
must be less than 900 (e.g., δs=400). It is now simply assumed that the phase shift
angle δc between the sending and receiving end voltages of the cable is 300. If the
voltages at the sending and receiving ends of the cable are assumed to be kept at the
constant voltage V, then the transmission power on the cable is given by:
V2
P= sin δ c (4-4)
R0 sinh β L cs

The critical cable length Lcs to keep the phase sift angle δc at 300 is given by.
V 2 sin 30 0 1 V 2
Lcs ≈ = (4-5)
PX c 2 PX c

The impedance of the superconducting cables is generally smaller than those of

existing cables, but not necessarily so small in the per unit values because of the low
voltage/large current power transmission. Thus when designing a low voltage, very
high power cable transmission system, it is necessary to consider the critical cable
length restricted by the phase shift angle between the voltages of the sending and
receiving ends of the cable to maintain system stability.
Examples of critical line lengths in the case of using SC cables for the trunk lines
of huge power transmission systems are shown in Table 4-2, with those of overhead
lines and conventional cables also given for comparison. As shown here, appropriate
choices of voltage/current are required even though superconducting cables have the
capability to transport bulk power at lower voltages.

Table 4-2. Examples of critical lengths

Line Critical Length Critical Length Critical Length

Power,
Type Voltage, (charging curr.), (voltage rise), (phase shift-30deg),
P [MW]
U [kV] Lcc [km] Lcv [km] Lcs [km]
RTD SC cable 50 200 N/A 106 57
CD SC cable 50 600 N/A 212-149 38
Overhead line 400 1000 ~3,000 ~300
Overhead line 132 225 ~4,000 ~100
Conv. cable 400 900 ~100 ~1,000
Conv. cable 132 175 ~100 ~300
N/A: not applicable
- 30 -
 4.3.3 Possibility of the surge impedance loading
Power transmission under SIL conditions will not suffer the problems of critical
length of cables discussed above. In case of conventional cables, their operating
voltages are usually high and the current capacities are not large enough for the SIL. But
in the case of SC cables, they generally have small characteristic impedance and their
current capacities are usually very large, so there exists a possibility of designing SC
cables applicable to surge impedance loading (SIL).
Some examples are shown in Table 4-1.

- 31 -
5. Superconducting cable transmission systems and their ratings
5.1 Basic superconducting cable system design and performance
Symmetrical impedance (zero, positive and negative phase sequence) of a cable
depends on the grounding schemes of the magnetic shielding layer. An example of this
impedance is listed in Table 4-1.
To discuss the power feeding security, it is important to understand the current
phenomena under fault conditions (symmetrical and asymmetrical faults) on the cable
system. The fault current path in the cable depends on the grounding scheme of the
shield layer, while the fault current in case of an asymmetrical fault depends on the
grounding scheme of the neutral transformer.
Concerning the reliability and security of feeding in superconducting cable systems,
the important factors can be classified as follows:
a. Reliability of cable hardware, cable design, manufacturing, etc.
b. Auxiliary equipment of cable system, refrigerator, etc.
c. Compatibility of cable capacity and transient disturbances in the AC network.

To deal with these factors, fault current analysis is important to design a tolerable
cable capacity for emergencies and also to evaluate the system security in case of
possible fault conditions. These issues are discussed below.
In case of an RTD SC cable, the current path is complicated compared with that of
a conventional cable due to the thermal insulation layer, but the return path of the fault
current is the same. Therefore, the purpose of grounding the shielding layers is to
minimize the Joule losses in the shielding conductor (normal conductor).
In contrast, the shielding layers of a CD SC cable are also superconductors, and
they will be grounded at both ends of the cable to shield magnetic fluxes, so the
symmetrical components of cable impedances have the same values.
The magnitude of fault current depends on the symmetrical components of the
impedances and SC cables generally have small impedance. According to the cable
characteristics and how the cable is installed, careful estimation of the fault current and
investigation of cable protection and system protection are important technical issues.

5.2 Ratings of superconducting cables

One of the specific features of SC cables is the transition phenomenon from
superconducting state to normal state; that is, beyond the critical values of current
density, temperature and magnetic field, the conductors cannot be superconducting. The
ratings of SC cables should be determined in a quite different way from those of
conventional ones.

Generally, the allowable current determines the transmission capacity of

conventional cables, where the allowable currents are classified into the following three,
according to the duration for which tolerable current flows:
a. Allowable current for continuous load Continuous rating current
- 32 -
b. Allowable current for short time duration (e.g. less than 10 hours)
Short time rating current
c. Allowable current for short circuit conditions Transient current rating

These current ratings are primarily determined by the thermal characteristics of

cable components such as cable conductor, dielectric, sheath conductor and
surroundings.
As mentioned above, SC cables have intrinsic current density and operating
temperature characteristics, and the allowable currents of SC cables are restricted by
these critical values. Regarding cable design and application for power systems, it is
important to consider the cable current rating.
It is common to decide the operational point (cable current rating) based on the
load factor to the critical current. In the design of SC cables, the selection of load factor
is decisive in determining the cable size, AC losses, and cable cost. Manufacturers take
into account these parameters to decide the continuous current rating and to develop SC
cables.
Regarding system operation, cables must be able to withstand short-duration
overload or transient overcurrent similar to conventional ones. However, a general
consensus on the overcurrent capability of SC cables (short duration or transient) in
close conjunction with system operation has not been achieved yet. As SC cables do not
have the concept of “tolerable overcurrent”, the allowable overcurrent depends on the
load factor, which affects the cable concept, cable cost and system operation security.
The two types of system configuration are proposed to secure the reliability of
SC cable systems under overload operation for a short time due to possible variations of
demand, including putting the transmission line off-line for maintenance and so on, as
shown in Fig.5-1a and 5-1b.
These system configurations are proposed based on the idea of transmitting the
rated power under the off-service condition of one-parallel line. The first one consists of
three parallel circuits each having a capacity of half of the route capacity. In this route,
two circuits will carry the full route power in case one circuit is off. The second one
consists of two circuits, each of which has the ability to carry the full power of the route.
The concept of the former one is generally adopted in conventional systems because of
the cost. But the latter one can be applied to SC cable systems, since the scale merit of
SC cables is much larger than that of conventional cables. As a result, the increase in
cost of cable fabrication may be offset by the decrease of installation cost.

Fig. 5-1a Three parallel circuits Fig. 5-1b Two parallel circuits

- 33 -
 These configurations make it possible to transmit a certain amount of
overcurrent (e.g. 1.5 or 2.0 times), and the characteristics are comparable to the
short-duration current rating of conventional cables.
Concerning the transient overcurrent, superconductors must return to the
superconducting state as soon as the abnormal current decreases to the rated value. It is
possible, theoretically, to design the cable to withstand the fault current by setting the
load factor small enough. However, This produces an extremely large current margin
and jeopardizes the merits of SC cables. In case of BSCCO SC cables, silver sheathes
are generally expected to stabilize the superconductors and these silver sheathes may be
designed to have a capacity to carry the transient abnormal current, when
superconductors temporarily enter the normal state. Naturally, the capability of the SC
cable to recover from abnormal currents depends on the current magnitude in excess of
the critical current/time duration, I2 t, and also the cooling capability of the system.
The general concept of system fault currents on SC cables has not yet been
discussed sufficiently. It is true that the stabilizing conductor or silver sheath should
have a fundamental ability to protect the superconductors from system fault currents.
However, an overall investigation of the stabilizing scheme is important in terms of
cost-effective cable design and fault current protection. For example, a stabilizing
scheme by using a porous type copper metal has been proposed to withstand transient
fault currents. The common concept for the transient current rating of SC cables will be
established as progress is made in stabilizing the technology and information of
dynamic system performance. In terms of power system management, three ratings have
been set up: constant rating, short time rating, and tolerable transient rating as shown in
Fig. 5-2. The capability and ratings of conventional cables are designed to meet the
requirement as shown in Fig. 5-3. As SC cables do not have the concept of “tolerable
overcurrent” as shown in Fig. 5-4, selection of an adequate operating current and
coordination with protective devices (protection relay system, circuit breaker’s action
and so on) has a strong influence on the system security.

- 34 -
 Transient rating
Itran

Short time duration rating

Current, 　I
Ish

Icon
Constant rating

ttran tsh
Duration, t
Fig. 5-2: System requirement

Itran
Current, 　I

Ish

Icon

ttran tsh
Duration, t
Fig. 5-3: Capability and ratings of conventional cables.

Transient rating
Itran Stabilization limit

Short time duration rating

Current, 　I

Ish

Icon
Constant rating

ttran tsh
Duration, t

Fig. 5-4: Capability and ratings of SC cables.

- 35 -
5.3 System performance
As regards system security, investigations on system performance including
overload capacities and tolerable transient current and estimations of the costs of SC
cables are the principal technical issues to introduce SC cables into the present networks.
Even so, few investigations on dynamic system simulation have been reported to date.
An example of dynamic system simulation is shown below. These investigations will
contribute to the progress of SC cable development.

5.4 Technical issues concerning system reliability

The technology concerning SC cables is still at the development stage, but
great progress has been made in recent years.

5.4.1 Stability of superconductors

The stability of superconductors of SC cables is one of the main concerns to
maintain system reliability. This issue strongly depends on both the cable design and the
cooling system, and many R&D studies have been conducted on this issue. It is
particularly important to develop technology for stabilizing SC cables against transient
fault currents.

5.4.2 System planning

(1) One of the most important issues is close communication between cable
manufacturing engineers and power system engineers to establish a common concept
for SC cable systems. From this point of view, one of the key issues is to establish a
common concept for cable ratings. Decisions on the current ratings corresponding to
system behavior have major impact on the cable concept and cable cost. On the other
hand, the capability of the SC cable to withstand tolerable currents affects the reliability
of the cable system.
(2) The characteristics of intrinsic tolerable current capacity versus time duration
are quite different between normal conductors and superconductors.

5.4.3 Auxiliary equipment

Regarding application development of superconducting technology, R&D of fault
current limiters (FCL) has progressed significantly. When FCLs become available for
SC cable systems, system operation will become much easier and system security will
be improved.

5.5 Operations and problems when a fault occurs in superconducting cable systems
The ratings and reliability of superconducting cables are studied in another
chapter; the entire system or cable peripheral systems are considered here.

- 36 -
5.5.1 Advantages of introducing superconducting cable systems
By introducing the superconducting cable system, low voltage/large current
power transmission is made possible and existing power transmission voltages can be
reduced. For example, 275 kV is used at present for 1000 MW power transmission, but
it can be reduced to 66 kV by using superconducting cable. As the result, 275 kV
transformers will no longer be necessary. Furthermore, the cable charge capacity
decreases, and so shunt reactors for compensation of charge current will also not be
necessary or only very small capacity reactors will be sufficient. In addition to these
merits, the equipment in substations can be made smaller and space can be saved thanks
to a relatively low voltage system compared with the existing system.
(Example: For 1 GVA power transmission, a system of 275 kV/1 kA/2 cct with
conventional cables is required, but a 66 kV/9 kA/1 cct is enough with superconducting
cables.)

5.5.2 Characteristics of system

In case of power transmission of the nominal capacity by underground cables, the
impedance of the superconducting cable becomes much smaller than that of existing
cables because low voltage/large current (high current density) cable design is possible.
Nevertheless, the impedance is not necessarily so small in the power system in terms of
per unit value because of low voltage/large current operation. Consequently, the voltage
stability at the ends of the load and the transient stability of generators connected via the
superconducting cable are likely to worsen in the case of large power transmission.
However, many transformers can be omitted as stated in section 1 above and so
the impedance of the total system will be lower. According to one analysis example
reported, the nose point of the P-V curve is higher and the phase differential angle
variation of the generators at a transient is smaller in the superconducting cable.

5.5.3 System problems

Generally in power systems, in the case of a three-phase short circuit or
three-phase grounding, a fault current will exceed several times the rated current. And
when the effect of transient DC current is taken into consideration, the peak value may
be doubled. Existing transmission lines and cables are designed with a large short-time
overload capacity so that they can tolerate the fault current and resume power
transmission shortly after the fault has cleared. On the other hand, overcurrent
durability has not yet been defined for superconducting cable systems, and as
mentioned regarding the definition of cable ratings, careful cable design will be
required under the following conditions.

- When a fault occurs in a power transmission circuit in a multiple-circuit power

transmission network, an adjacent circuit is affected by the fault and a large induced
current flows through the sound circuit
- 37 -
- When a short circuit has occurred in the lower level system beyond a
superconducting cable in a network, and a large fault current also flows through the
superconducting cable

As stated above, the superconducting cable has the possibility of shifting from the
superconducting state to the normal state due to the extremely large fault current. It is
therefore very important to manage the fault current successfully in terms of system
security, and a fault current limiter or adequate circuit breakers must be introduced
appropriately. A fault current limiter would be the best solution to achieve system
security, but such limiters are still under development at present.
Another consideration is a protective relay for the superconducting cable system. If
a fault occurs in the superconducting cable network in which fault current limiters are
installed at both ends of the cable, the impedance of the protective relay will be affected
greatly due to the S/N transition and operation of fault current limiters. In order to solve
this problem, protective relays and relaying systems have to be developed.
Furthermore, when the existing high voltage conventional power network such as
the 275 kV system, is replaced by a low voltage superconducting cable network such as
a 66 kV system for a large power consuming populated city, a large fault current will
flow through the cable system connected to the large capacity power grids.
If the fault current exceeds 31.5 kA to 40 kA, it is required to use circuit breakers
of larger capacity than the existing ones, and the ratings of the power circuit equipment
must also be updated.

5.5.4 For use at power plant terminal

Similar and intrinsic problems will occur when superconducting cables are
introduced at power plant terminals, as follows.
A power swing will occur in the power generator after a fault has occurred. The
magnitude of the power swing sometimes reaches 50% to 100% more than the rated
power. Consequently, the ratings of the superconducting cable must be decided
accordingly to secure the reliability of the system.

- 38 -
6. Conclusions
Due to the remarkable development of high Tc superconducting materials,
research and development on the application of superconducting materials for power
apparatus is attracting much interest around the world. Among these power applications
of HTc superconducting materials, power transmission cables and fault current limiters
are expected to have a great impact on the system configurations and operation.
The Task Force for “Superconducting Cables, Impact on Network Structure
and Control” (CIGRE TF 38.01.11) was organized at the SC 38 of CIGRE. SC cables,
especially superconductors, are still at the development stage including test operation in
the field, and the characteristics of the cables are expected to be improved in the near
future. As the cable technology progresses, SC cable systems are expected to provide a
very attractive power transmission scheme and to enable compact, high efficiency, small
environmental impact power grids to be developed.
This report described the present status of cable development around the world
and discussed the key characteristics of SC cables. Particularly, the differences in
concepts of operational ratings to the conventional ones are very important in estimating
the effects of SC cable networks and basic information of SC cable systems.
Many technical issues remain to be solved for the construction of SC cable
networks as pointed out here. However, it is also true that present technology is
sufficient to make pre-commercial SC cables under restricted conditions.
In modern R&D, close communication between SC cable engineers and power
system engineers is critically important. The report noted the importance of close
exchange of information on SC cable technology and operational requirements of power
systems in order to promote the development and application of SC cable networks.
Possibilities of SC cable systems are clarified, however, precise research needs
to be conducted on SC cable development and compatibility with power systems for
estimating the impact of SC cables on the power network along with the rapid progress
in superconducting technology.

- 39 -
 Appendix I

SYNTHESYS OF THE ANSERS TO

THE QUESTIONNAIRE
ON CONVENTIONAL AND SUPERCONDUCTING CABLES
IN POWER GRIDS

Survey made in 1999

App.I - 1
Questionnaire and Answers
(Questionnaire and answers on conventional and superconducting cables in power grids)

1.Objective
The collected data will be used to design SC cable systems to evaluate the
characteristics.
2. Items
(1) Conventional high voltage (>60 kV) insulated cables in the present grid
(1.1) The reasons and the cases of the installation of insulated cables
(Q1.1) What is the extension of cable lines (compared to total) in the power grid in
your country (or district)?
(A1.1) See the attached Answer Table.

(Q1.2) Why the cables are chosen in those cases? Since their costs are usually much higher
than those of over head power transmission lines, so there must be particular reasons of the
choice. (Please list them up)
(A1.2)
ix) Mostly in housing and industrial area near S/S NOK
x) The installation of overhead lines is impossible or very difficult due to the following
restriction:
・ Legal restriction,
・ The restriction on land space acquisition TEPCO
xi) From the standpoint of harmony with local environment and economy, underground
cable lines are totally more advantageous than overhead lines
TEPCO
xii) Cables are mainly chosen where transmission lines have to cross waterways, pass
through urban area or through environmentally sensitive area where overhead line cannot
be accepted. ELTRA
xiii) An agreement between Danish government and the electricity supply companies.
・ 50 kV lines shall be installed as cables.
・ 132 kV lines shall be installed as cables in urban areas.*
・ 400 kV lines shall be installed as cables in urban areas.*
* It will not be possible to install OH line in urban areas.
* In rural areas OH lines can be installed but permission
has to be obtained by the authorities. NESA
xiv) Restriction on right of way in urban areas
• Submarine cables (380 kV)
• Connection of power stations with overhead lines
• HVDC cables for long distance ENEL
vii) No right of way available for overhead transmission lines. SCECO
viii) In the over crowded cities, overhead transmission lines are not possible.
• To avoid overhead obstructions.
• Where bearing capacity of soil is very low. TABOUK
ix) Restriction on right-of-way. Environmental concerns REN
x) Used only in the city centers and for short distance sea connections.
SYDKRAFT
xi) For environmental reasons (the view, density of inhabitants next to big town,
legal requirements ) EdF

(1.2) Cable and system concept

App.I - 2
 (Q1.3) Voltage levels:
(Q1.4) Capacities:
(Q1.5) Typical length:
(Q1.6) Types of the cables and their installation features
(Q1.7) Where are they installed
(Q1.8) Methods and spans of the forced cooling, if any:
(Q1.9) Typical number of the circuits in a route:
(A1.3)~(A1.9) See the attached Answer Table.
(1.3) Future requirements on power transmission lines:
(Q1.10) What types of power transmission lines are going to be required in your regional area,
country or in the world?
(A1.10)
i) The tendency is towards building loop-circuits for upgraded transmission
Line (present 220kV lines into 380kV). ETH
ii) 110kV-lines for replace 50kV systems NOK
iii) High capacity ( high voltage and/or large current)
• Loss reduction
• Low cost (construction and maintenance)
• High reliability TEPCO
iv) 400kV grid in Jutland / Funen has not been completed yet. About 250km of line still
remains to be completed. However, it is very difficult to get the right of way today.
ELTRA
v) Transmission lines; (a) between substation in urban area, (b) between power
generating units and substation, (c) from power generating units in urban areas to station
in rural areas where cable is connected to over-head lines (without breakers o.a)
NESA
vi) Same type of transmission lines as nowadays.
• For special cases (e.g. urban areas) and short routes (some km) XLPE and GIL cables
are presently evaluated. ENEL
vii) Conventional over head and underground systems will continue to be built.
• High power density circuit will find application where bulk power transmission is
required in congested row corridors. SCECO
viii) (a) High efficiency. (b) High power density. REN
ix) High efficiency and high power density.
• High reliability
• Low electromagnetic environmental impact. SYDKRAFT
x) HV: over-head lines and underground cables for environmental purposes
• EHV: mainly over-head lines, and underground cables in urban areas
EdF
(2) Sc cables
(2.1) Exploitation of advantages of sc cables:
(Q2.1) What kinds of application for sc cables do you think are feasible in the future
grid( in your networks or others)?
(A2.1)
i) Application of sc cables not probable (length, cool core ) NOK
ii) Using compact type sc cable in ducts or underground (urban corridors)
TEPCO
iii) Massive power transmission into urban area and only for relatively short line
ELTRA
iv) It will take several years before SC cable have been proven to be reliable both from an

App.I - 3
 electrical point of view. This includes reliability and maintenance for the cooling systems.
• From maintenance point of view SC cables are a step back. (In NESA area) All new
cables are XLPE cables, which means the maintenance cost is greatly reduced.
• Fault current limiters and SC transformers are interesting new features, although they
must be proven to be reliable.
NESA
v) Penetration in congested urban area, congestion in the row corridors
• Upgrading of existing conventional underground cable
• High power cables in environmentally sensitive area
ENEL
vi) SC cable will be applied in high-density urban areas due to increase
In future load demand and congestion in row corridors. SCECO
vii) Urban corridors using compact type sc cables.
(short circuit current limiters). SYDKRAFT
viii) In France : interconnections with scheduled interchange and environmental
Constraints.
• Worldwide: evacuation of large power units with environmental constraints.
EdF

(Q2.2) And what are the reasons of the choice?

(A2.2)
i) Efficient use of the underground.
• Lower installation cost. TEPCO
ii) Cost NESA
iii) High power cable with very low EMF and reduced transmission losses.
• Saving of a transformation step in urban area. ENEL
iv) Load demand, restriction of row availability, lower cost per MVA
SCECO
v) To avoid hurdles such as overhead crossing
• Lay in the streets as well as in busy road. TABOUK
vi) Cost. Lcc (low cost corridors ?) SYDKRAFT
vii) To surmount the over-cost of SC cable vis-a-vis “classic” ones,
you need large power flow connection, used as close as
possible from their maximum rating and as long as
possible over the year. EdF

(2.2) Easing restrictions:

(Q2.3) Which are the main constraints that you think could be overcome by SC cables?
(please list in order of importance )
(A2.3)
i)Space limitations of underground in metropolitan area.
• Environmental issues (EMF, Joule heating, etc ). TEPCO
ii) It is probably only feasible for high power transmission lines
(The cooling arrangements and the power consumption necessary
to cool the cable make the installation very expensive) ELTRA
iii)Cost.
• Impedance (reactance) compared with ordinary cables. Current dividing
with parallel cables.
• Ability to handle overloading and short circuit currents. It is vital for the
system that the grid remains intact during and after a fault.

App.I - 4
 • Physical handling during transportation and installation. Is it possible
do directly bury the SC cables, or must they be installed in duct/tunnel?
NESA
iv) Very high transmitted power through limited corridors.
• Technology with a low environmental impact.
• Matching capability of overhead line. ENEL
v) Space limitation for the row.
• Power transfer capability. SCECO
vi) Space limitations- very costly installation.
• Environmental issues (EMF, Joule heating, etc.).
• Very high transmitted power.
• Short circuit current limiter. SYDKRAFT
vii) Mainly the width necessary for underground large power link in cities
EdF
(Q2.4) Which new technical constraints do you expect SC cables to bring about from
the point of views of performances (please list in order of importance)?
(A2.4)
i) Cooling systems
• Compatibility with other apparatus and maintenance. TEPCO
ii) Reduced damping in the systems.
• Ability to handle overloading and short circuit currents. NESA
iii)Cooling system issues (refilling substations, maintenance, length)
• Compatibility of SC cables with other existing apparatus
• Reliability issues
• Analysis of the ground system; possible increase of short circuit power
• Balance of surge impedance of the various line sections ENEL
iv) Careful drsign, installation,. Splicing and termination procedures will be
essential to obtain long time trouble free service SCECO
v) Difficulties in location of faults
• Difficulties in extension in crowded cities.
• Maintenance. TABOUK
vi) Maintenance and compatibility. SYDCRAFT
vii) New technical constraints SC cable could bring about :
• More maintenance due to refrigerating unit
• Less feeding security due to less cables (with large capacity)
EdF
(Q2.5) Typical specifications of the sc cables listed above
Please describe the capacities, the voltage level, the length,
the dimensions and/or the special features.

(A2.5)

i) 66kV, 1000MW, km order length, expected dia. : approximately 130mmφ

TEPCO
ii) Relatively short lines (<500 m, urban areas ). ELTRA
iii) Penetration in urban areas; 132kV, 1GVA, typical length ≤10km
ENEL
iv) Cannot estimate at this stage. SCECO
v) 132kV; interconnection of S/S inside big cities. TABOUK
vi) 130kV, 400kV, short circuit current limiter. SYDKRAFT

App.I - 5
 vii) Typical specifications of SC cable ;
• Voltage ; 225kV, • Capacity ; more than 1000MVA,
• Length ; few ten kilometers.
• Capability to auto-limit the short-circuit current would be an interesting
special feature. EdF

(3) Others
When sc cables are applied in dc systems, the advantages over the ac cases may be
enlarged because of no conductor losses, higher current densities, reduction in cable
diameters, and so on. Recent proposal of a small scale dc system, “HVDC Light” (please refer
to ‘ABB Review’, No.1/98), seems to be suggestive. Then it will be worth to study the
possible dc cases.
(3.1) Dc power transmission cables (either conventional or superconducting)
(Q3.1) The cases of installation of dc cables in the grid:
What are the expected needs (in the next 10 years) for DC cables?
(A3.1)
i) No need. NOK
ii) Non-synchronous linkages.
• Power transmission from remote generation. TEPCO
iii) A 600MW HVDC installation is planned between the islands Funen
and Zealand for commissioning in the year 2003 ( cable ≈40 km). It is
planned to use conventional oil-filled HVDC cable for line.
• HVDC Light has been discussed for the transmission of power from large
scale offshore wind farms (≈140 MW) situated 40-60 km from the existing
grid. It is planned to use XLPE insulated HVDC cable.
The alternative is 150 kV XLPE or oil-filled cables.
• SC cables will no be considered for any of the above installations ad no
references for commercial use of SC cable systems are expected to exist
within the next 5 years. ELTRA
iv) Great belt link ; non synchronous linkage.
• Wind farms ; power transmission from remote generators. NESA
v) Oversea links ( Italy-Greece interconnection). ENEL
vi) No plans for dc cables so far. SCECO
vii) Not feasible. TABOUK
viii) In Portugal ; do not foreseen the needs for DC cables installation
In the next 10 years. REN
ix) Strengthen different areas the weakly connected ac-systems.
SKTDKRAFT
x) World wide; some high power interconnections (underwater or in particular
areas) [in France; probably none] EdF

(Q3.2) And what are the reasons of the choice?

(A3.2)
i) Lower cost TEPCO
ii) Great Belt link: Non synchronous linkage of two major AC systems.
• Wind farms : Power transmission from remote generators. HVDC light may be an
alternative to expensive reinforcement of the ac system. NESA
iii) Need for a stronger European grid interconnection. ENEL
iv) Not applicable. TABOUK
v) Not applicable. REN

App.I - 6
 vi) Lower cost. SKYDKRAFT
vii) Cannot use overhead lines EdF

(Q3.3) How would you consider an HTS solution for such cases?
(A3.3)
i) Primary reason to select a sc cable is not necessarily to reduce Joule losses.
・ To expose the dielectrics at room temperature: cable reactance is substantially
reduced. Voltage drop is small, highest power level at low voltage.
・ Nitrogen as a coolant has merits.
・ Old conventional cable must be upgraded and replaced by new one with higher
capacities; Using existing tunnels and trenches and place sc cable. ETH
ii) An HTS solution is a possibility ; it will take several years before it will become
reality because of the relatively small transmission capabilities in the NESA area.
And the technology must be proven to be reliable. NESA
iii) At present there is interest in :
(a) Research in HTSC cables for urban routes.
(b) Monitoring of international tech. Activity in HTS materials,
Cables and devices. ENEL
iv) With a.c. transmission. TABOUK
v) Not applicable. REN
vi) For very specific applications, it can be interesting to study HTS
SYDKRAFT
vii) No electrical losses.
• Limited number of cables (due to high power capability).
• Maintenance difficulties. EdF

(Q3.4) Please note whatever you think or feel concerning the task force works.
(A3.4)
i) Task Force : Necessary to study the problems relating to the network and provide
the possible solution for giving guidance to the electric utilities.
TABOUK

App.I - 7
Ref. Comments on conventional cables and superconducting cables in electric power
grids from EZH, Zurich.

Worldwide there are no SC-Cables incorporated in grids. If we could build a SC-Cable, it

must be with regard to voltage levels in that particular country.
In addition to our study we would like to ask the following questions:
a. Why not a sc-d.c. cable ?
b. Why 66 kV in Japan?
We think 110 kV would be more suitable.
c. Why cables for 1 GW?
What happens when a quench occurs and the cable cannot be operable for some time?
d. Problems of redundancy of sc.cables should be discussed.

a) Conventional insulated cables in the present grid

Reasons and cases for of the installation of insulated cables

The central European grid is crowded. People’s sentiments and the laws in various
European countries to preserve the environment are very strict. In Switzerland it is difficult to
build 380 kV overhead transmission-lines even in rural areas: e.g. To an existing 220kV
transmission line from the atomic power plant Mühleberg to the substation of Verbois a 380
kV three phase overhead line must be added to complete a loop circuit, which will guarantee
the power delivery in the southwest of Switzerland. For more than 6 years there have been
discussions to prevent the installation of such an overhead line. The environmentalists prefer a
380kV Cable system or they indicate to jeopardize the 380 kV transmission line all together.
A similar example can be quoted in Austria, where from Wien- Südost to Tauern a 380 kV line
should be installed to create a loop circuit. Again in this case a 380 kV cable is preferred by
the environmentalists and by the majority of the people.
The cable is much more costly than overhead transmission lines: (15 - 20 times per km
higher in Switzerland). In Switzerland in large cities no overhead transmission lines are
allowed. In these cases only cables are permitted in spite of their high cost.
How are the economical aspects for selecting cables?
The lifetime of a cable is about 40 years. A transmission line has a lifetime of 60 to 70
years. The design of a cable transmission circuit is a matter of sound engineering judgement.
Regards for circuit security, standby and overload capacities and its intrinsic commercial
features are more pre-dominant than academic formulae which attempt to convert the future
into the present. From the engineering point of view several possibilities are feasible for the
engineers: Three phase circuits, (single or double circuits; redundant circuits; ring systems;
feeding from two or more points etc.) dc systems using the FACTS- electronic circuits).
Superconducting cable-systems can be incorporated in a grid using FACTS* and current
limiters.
Any cost estimate is based on previous experiences with conventional systems and on old
and new assumptions as well as present component costs. However the cost estimate between
transmission lines and conventional cable for power transmission systems is based on
experience of the last 30 to 40 years. Within limits new estimates based on experience and
sound judgement are certainly feasible.
Selection of superconducting cables instead of conventional isolated cables is new and
needs more careful economical evaluation, as well as consideration of safety, flexibility in
operation, important advantages by using low voltages and high current and operating costs,
assuming low Joule losses.
As seen below, the transmission lines dominate over the cables in Switzerland for voltage

App.I - 8
levels higher than 24 kV. The continuous trend to concentrate industry and the population in
cities necessitates clustering the power. Open-Air substations are not permitted in suburbs,
because of today’s sentimental reasons and also because there is no adequate space for
open-air circuit-breakers, disconnectors, transformers, safety and diagnostic devices etc.
conductor lineout structures in open air. Today’s trend lead towards underground substations.

________________________________________________
* FACTS : flexible ac transmission systems.

b) Future requirements on power transmission lines

Future requirements on power transmission lines are the upgrading present 220 kV three
phase lines into 380 kV lines for safety reasons and also to guarantee immediate power
delivery in cases of failures in a circuit. The tendency is towards building loop circuits for
upgraded transmission lines. The implementation of overhead lines will face great resistance
from the environmentalists.

c) Superconducting Cables

c1) Reasons to select superconducting cables:

The primary reason to select a superconducting cable is not necessary to reduce Joule
losses. An ac three phase superconducting cable has hysteretic losses, heat transmission losses
to the coolant and dielectric losses, if the dielectric is at cryogenic temperatures. The total
losses must be carried away by the coolant. In new studies the dielectric is kept at room
temperature. To compare these losses to those of a conventional cable at 80oC the Carnot
efficiency and the efficiency of the refrigerator liquefier must be considered. It is certainly
possible to reduce losses of conventional cables to levels, so that they become comparable to
HTSC cables.
To expose the dielectrics (cable electrical insulation) at room temperature has also the
advantage, that the grounded thermal shield acts as an electric shield. In this case the cable
reactance is substantially reduced, resulting in a smaller cable impedance, compared to a
room temperature cable. Since the voltage drop along the cable is small, even at highest
power levels low voltage can be considered.
One reason why high voltages are used in conventional transmission lines is to send
smaller currents through the cable and thus the total voltage drop along the line and the
Joules-losses can be kept at reasonable limits. In superconducting cables the highest voltage
levels (> 115 kV) can be eliminated. By reducing voltage levels in step-up step-down voltage
transformation not only the transformer can be eliminated, but the number of switches,
control and safety circuits and voltage interruptors can be reduced. Less real estate is
necessary. This would lead to substantial savings, especially if in the grid-system
superconducting current limiters and sc-transformers are implemented. In this case the whole
system could be installed underground.
Conventional underground cables normally incorporate an insulant-coolant such as SF6 or
oil. In HTSC the utilization of liquid nitrogen (LN2 ) as a coolant is not an substantial
departure from already established methods. Nitrogen has merits, which oil and SF6 have not.
The populations of cities are growing. Required electrical power density in congested
urban areas and in clusters of industrial complexes are increasing. Old conventional cables
must be upgraded and replaced by new ones with higher capacities. Overhead transmission
lines and open air substations in urban areas must be replaced by cables and underground

App.I - 9
substations to comply with the popular demand.
Why not using existing tunnels and trenches and place superconductors in them, since real
estate is expensive and the sc cable dimensions are even smaller than those of room
temperature cables. Open-air substations, which provide energy to cities can be replaced by
underground substations located not far from cities. The whole system will not be noticeable
above ground..

d) DC-superconducting cables:

Using dc superconducting cables require inverter-converters. For short lengths cables such
as in power corridors into big cities (< 10 km) the cost of high efficiency converting ac to dc
and back is at present higher than the cost of the superconducting cable. This maybe the
reason why no superconducting dc transmission lines have been studied in detail.

App.I - 10
Answer Table Answers to the Questionnaire(Q1.1, Q1.3-9)
Region Voltage OHTL Cable
Level (kV) Total Total Typical Typical Installation Forced cooling Type of Length
Length Length Capacity & Length (if any) Cables (km)
(km) (km) No. of cct/route (km)
Switzerland 245-420 1 422 0
(ETH) 170-245 5 122 35(0.7%) 300(MVA/cct), Tunnel Oil-Paper 32,52
2(1)(cct/route) XLPE,VPE 1,8
EPR 0,24
72.5-170 1 844 298(14%) 200 Duct Oil-Paper 91,29
GIL 82
XLPE,VPE 79,5
EPR 45,5
24-72.5 3 086 427(12%) 40 Oil-Paper 175,85
XLPE,VPE 170,77
EPR 34
PE 9
GIL 3
12-24 6 781 5,669(46%) 20 Directly　buried XLPE,VPE 1445,57
EPR 476,31
PE 495,43
1-12 315 626(67%) 5 XLPE,VPE 124
EPR 87,2
PE 11,8
<1 1 537 6,504(81%) 1 XLPE,VPE 2903
EPR 600
PE 100
NOK 220 650 13(2%)
supply area 110 920 184(20%) 140(MVA/cct), 0.2-0.3 Duct none XLPE
(~1/3 of Swiss 2(cct/route) (a few km at max.)
power consum.) 50 1 000 80(8%)
16 90 18(20%)
TEPCO 500 4 016 (under construction)
supply area (80-120km) (850-1150) (40) (Tunnel) (Indirect water c.) (XLPE)
(Japan) (MVA/cct),
(2-3) (cct/route)
275 2 484 952(28%) 500 15 Directly　buried(12%) Indirect water c. Oil-Paper (75%)
(0.2-32.5) Duct(6%) Direct water c. XLPE (25%)
Tunnel(82%)
154 6 286 782(11%) 200 7 Directly　buried(3%) Direct air c. Oil-Paper (80%)
(0.1-24.1) Duct(47%) XLPE (20%)
Tunnel(50%) GIL (0.1%)
66 14 961 5,544(27%) 50 5 Directly　buried(1%) Direct air c. XLPE (80%)
(0.1-24.1) Duct(69%) Oil-Paper (20%)
Tunnel(30%)
Eltra 400 (Cable installation planned for commissioning in 2002)
supply area 800(MVA/cct), (2) (Directly buried) (none) (XLPE)
(western 2(cct/route)
Denmark)
150 2 356 35.34(1.5%) 200 5 (0.65-12) Directly buried none XLPE
NESA 400 165 35(17%) 1000-800(MVA/cct), 8-12 Directly buried none XLPE
supply area 1(cct/route)
(Denmark) 132 584 92(14%) 150-250 1-10
50 254 247(49%) 60-70 1-15
ENEL 380 9 412 16(0.17%) 1000(MVA/cct), 9 (0.6-9) Buried in cement mortar, none Oil-Paper
(Most of Italy) 1(cct/route) Duct,
220 10 645 99(0.93%) 380 2 (0.4-11) Tunnel Oil-Paper
120-150 3 409 375(1.1%) 260 2 Oil-Paper,
(0.2-9) EPR
200(DC) 857 240(28%)
SCECO 230 9 000 36(0.4%) 500-600, 3-9 Directly buried none XLPE,
supply area 2(usually) or 3 LPOF
(Saudi Arabia) 115 9 273 102(1.1%) 300 2-10
69 9 444 680(7.2%) 50-90 1-14
Tabouk 132 30.6(0%) 150(MVA/cct), 9 Directly buried none LPOF
supply area 1(cct/route) (3-12)
(Saudi Arabia) (132kV XLPE cables will be utilized in future)
13,8 5 000 400(8%) 10 17 Directly buried none XLPE
REN 400 0
supply area 220 2 500 7.5(0.3%) 380(MVA/cct), 6 Directly buried, none LDPE
(Portugal) 2-4(cct/route) (10 at max) Tunnel
150 2 500 7.5(0.3%) 260(MVA/cct),
( EDP?) 2-4(cct/route)
63 5 960 298(5%) 40(MVA/cct), 5 Directly buried LDPE,
4-6(cct/route) (10 at max) XLPE
SYDKRAFT 130 20 000 200(1%) 100-200(MVA/cct), 2 Directly buried, none EPR
supply area
1(cct/route) (3 at max) Duct
(Sweden)
EdF 400 2 600 2.6(0.1%) 1200mm2 0,25 all of the available none XLPE
(France) 1(mainly) or (0.1-0.4) techniques (new cables)
2(sometimes)
225 26 400 870(3.3%) 400-1600mm2 2,5
(0.1-19)
2
90 15 600 452(2.9%) 240-1200mm 2
63 36 000 1,442(4.0%) (0.1-12)
 Appendix II

EFFECTS OF THERMAL INSULATION ON CONSTANTS OF

ROOM TEMPERATURE DIELECTRIC SUPERCONDUCTING

CABLES

App.II - 1 -
 Effects of thermal insulation on constants of RTD SC cables

The equivalent circuit is shown in App.II-1 for estimating the effects of thermal insulation
below the electric insulation. The impedances should be chosen in terms of not only line voltage
drops but also voltages across the thermal insulation, which is generated by the charging current.

．
V1
． ． ． ． ． ．
Vs I1s ． 1Z ． I1r ． Z1 Vr
Zm Y１ Zm
． ． ． ．
Is V2 I1c Iｒ
． ． ． ．
I2s Z2 ． I2r Z2
Y
．2
I2c

D
Fig. App.II-1. Equivalent circuit of SC cable with warm temperature dielectric

．
Z1 is the impedance of the superconducting phase conductor, which mainly consists of
inductance. The inductance per unit length L1 [H/m] is determined by the outer diameter, d, and
．
the distance between the centers of phase conductors, S. Z2 is the impedance of the inner normal
conductor on the electric insulation. It is consisted of the resistance and the inductance. The
． ．
resistance is R2 [Ω] and Zm is the mutual impedance of the two conductors. Y1 is the admittance
between the superconductor and the inner normal conductor. The capacitance per unit length is C1
．
[F/m] and depends on d and the inner diameter of the electric insulation, Di. Y2 is the admittance
of the electric insulation. The capacitance per unit length C2 [F/m] is given by the outer diameter of
electric insulation, Do, Di and εs.
Usually, the thermal insulation is not designed to withstand high voltage, so the voltage
． ．． ．
across Y1 caused by the charging current, V1-V2 (=∆V), must be kept low enough, typically several
hundred volts.
． ． ． ． ． ．
When Z1 = Z12 + Zm and Z2 = R2+ Zm, ∆V can be expressed as follows.

Z& 2 − Z& m
∆V& = Y&2V&2
2 + Y&1 (Z& 1 + Z& 2 − 2 Z& m )

In order to keep this value low enough, the resistance of the normal conductor R2 should be
controlled in the cases of long distance and high line voltage cables. Too high R2 may cause break
down in the thermal insulation whereas too low R2 increases the current in the normal conductor
and losses may become intolerable. In order to show typical features, some results of calculations
App.II - 2 -
 ． ．
are described below. The current at the sending end, Is, the current in the phase conductor, I1s, and
．
the current in the inner normal conductor, I2s, are given as follows.

I&s = I&1s + I&2 s

R2 1
I&1s = I&s − (V&1 − V&2 )
R 2 + Z&12 R 2 + Z&12
Z&12 1
I&2 s = I&s + (V&1 − V&2 )
R + Z&
2 12 R 2 + Z&
12

I&1s R ⎧ 1 I&c 2 ⎫
= 2 ⎨1 − ⎬
I&2 s Z&12 ⎩ 4 + Y&1 (R 2 + Z&12 ) I&2 r ⎭

． ． ．
The ratio of the currents I2s / I1s approaches R2 / Z12 when the charging current becomes small.
Supposing the outer diameter of the superconductor to 30mm and the thicknesses of the thermal
insulation and the electric insulation are 10mm each, the upper limit of R2, R2max, to keep ∆V less
than 1kV and 500V can be calculated using the parameters of the line voltage and the distance. The
results are shown in Fig.App.II-2, where r2[Ω/m] and r2max [Ω/m] are unit length values of R2 and
R2max. In this figure, the required conductor areas to attain the resistance are also calculated by using
．
the specific resistance of copper, 2 x 10-8 [Ωm]. The absolute value of R2 / Z12 is plotted in
Fig.App.II-3.
Following is an example when 3 V2 =100kV, Is=2500A, Ps=433MVA, ∆Vmax=1kV, and
． ． ．
the cable length is 10km. r2max is 4.45 x 10-3 [Ω/m], and the second term of I1s and I2s, |∆V| / | R2
． ． ． ． ．
+ Z12 |, is 22.47A. | I1s| becomes the smallest when Is and ∆V are in phase. Thus, the results | I1s |
． ． ． ．
=2477A and | I2s | =37.4A are obtained. In this case, | I2s / I1s | is about 66. In this case, R2 / Z12 is
83.5. The current in the normal conductor depend on the cable load. The current ratio becomes
between those values.

App.II - 3 -
 1.0E-01

∆V=1,000V
∆V= 500V

1.0E-02 S=2mm
2
r2,max [ Ω /m]

D=10km
2
1.0E-03 S=20mm
D=20km

D=40km
2
1.0E-04 S=200mm

ρ=2x10 Ωm
-8

1.0E-05
0 100 200 300 400 500
Line Voltage [kV]
Fig.App.II-2 The upper limit of r2 to keep the ∆V less than 1kV or 500V

App.II - 4 -
 1.0E+03

∆V=1,000V
∆V= 500V

1.0E+02

D=10km
R 2,max/Z 12

1.0E+01
D=20km

D=40km

1.0E+00

1.0E-01
0 50 100 150 200 250 300 350 400 450 500

Line Voltage [kV]

． ． ．
Fig.App.II-3 |R2 / Z12|, the upper limit of the ratio, | I2s / I1s |.

App.II - 5 -