Medium Voltage Direct Current Applications

M E D I U M VO LTAG E P R O D U C T S
Technical Application Papers No. 24

Medium voltage direct current applications
2 T E C H N I C A L A P P L I C AT I O N PA P E R S N O . 2 4 - M E D I U M V O LTA G E D I R E C T C U R R E N T A P P L I C AT I O N S
M E D I U M V O LTA G E P R O D U C T S 3

Contents
1. Introduction 4
2. Brief history 5
3. Applications and new scenarios 7
3.1 Evolution of electrical loads and power generation 7
3.2 Power generation from renewable sources 7
3.3 Static converters 8
3.4 Protection of static power converters for AC/DC
interfacing. 14
3.5 DC microgrids 14
3.5.1 Office blocks and shopping centers 14
3.5.2 Rural applications 15
3.5.3 Electric ships 18
3.5.4 Data Centers 22
3.5.5 DC industrial installations 24
3.5.6 HVDC transmission lines 28
3.5.7 Electric traction 31
4. Direct current interruptions 35
4.1 Characteristics and problems 35
4.1.1 Conventional interruption with direct current suppression 35
4.1.2 Interruption with passive oscillating circuit 37
4.1.3 Interruption with active oscillating circuit 39
4.1.4 Interruption with forced oscillation 40
4.1.5 Interruption with parametric oscillation 41
4.1.6 Interruption with semiconductor technology 42
4.1.7 Interruption with hybrid technology 43
4.2 State-of-the-art of medium voltage DC circuit-breakers 44
4.2.1 ABB medium voltage air-insulated circuit-breakers
for railway applications 44
4.2.2 ABB hybrid circuit-breakers 45
4.2.3 Future developments 46
5 Regulatory framework 47
6. Future prospects 50
Bibliography 61
1. Introduction
Although more than a century has passed since The purpose of this guide is to provide an
the heated dispute between Thomas Edison and overview of how direct current can be applied
George Westinghouse known as the War of and an indication as to the state-of-the-art and
Currents, the debate as to whether use of possible future developments.
direct current (DC) is better than the now
widespread use of alternating current (AC)
continues. Development of modern static
converters, growing use of renewable energy
sources (e.g. photovoltaic generation plants)
and new categories of users operating directly
in direct current (e.g. distribution on board
ships, especially military vessels, data centers
and electric vehicles) have re-opened the
possibility for direct current to no longer be
relegated to merely specific applications but to
be used more generally, for electric power
distribution.
2. Brief history
The modern electric power generation and The electrical supply lines were buried under the
distribution industry was only just dawning when, streets of New York and became rapidly popular
in 1882, Thomas Edison (figure 1) founded the with the public, worried about the numerous
first electricity company using direct current as accidents due to electrocution caused by the high
its technology. The initial purpose was to supply voltage AC lines, which had been employed since
110 V DC for lighting people's homes using the the '70's for public lighting purposes using 3500 V
incandescent light bulbs invented by Edison arc lamps.
himself.
Certain limitations to that type of use were,
however, evident: the distance covered by that
voltage was very limited and, to contain the costs
of the copper conductors, the generating plants
had to be sited within residential areas only a few
kilometers from the users.
Invention, in 1884, of the high-efficiency closed-
core transformer by three Hungarian inventors
(Zipernowsky, Blathy, Dri) and its successive
application for supplying users in parallel instead
of in series, played a key role towards the
supremacy of alternating current. Use of
alternating current was immediately successful in
Europe, and Rome was one of the first large cities
to be provided with alternating current
electrification in 1886.
It was during those years that, aware of what was
happening in Europe, George Westinghouse,
Figure 1: Thomas Edison
another entrepreneur and pioneer of the
electricity industry (figure 2), realized how
voltage step-down transformers would help
towards the widespread use of alternating
current while allowing him to develop new
patents instead of getting round those of Edison.
The first demo installation, designed to reduce
voltage to 100 V AC so as to supply incandescent
lamps in homes, was created in 1886 at Great
Barrington, Massachusetts.
It was immediately evident to other electricity
companies that, by cutting down on losses and
the cost of copper lines with successive reduction
for final domestic lighting purposes, high voltage
transmission represented a considerable
economic advantage. So much so, it quickly
became widespread and the very next year, in
1887, Westinghouse and the Thomson-Houston
Electric Company were already supplying more
than 40% of the generating installations.
Figure 2: George
Westinghouse
However, as use of electric power increased, the Edison to take a stand and condemn AC
situation in certain cities as to reliability and distribution as being dangerous This was what
safety became extremely critical. 6 kV AC lines triggered off the so-called War of Currents, which
with the poor quality insulation of those times lasted for several years with no holds barred. The
were routed near all sorts of alarm lines and invention of the electric chair was even used as
telegraph lines. While in the majority of European proof of the dangerous nature of alternating
cities and some American cities, e.g. Chicago, the current. On the other hand, those were the years
authorities decided to route the electricity during which Nikola Telsa invented the induction
distribution lines underground in other cities, motor, thereby endorsing, along with other
such as New York, electricity continued to be inventions, the completeness and convenience of
distributed by overhead lines. The situation using alternating current.
worsened during the Great White Hurricane of The corporate structures of the companies
1888 that also hit New York, causing blackouts involved changed to a considerable extent in just
and the deaths by electrocution of several a few years: Edison had already purchased
workers as they serviced the electricity lines. numerous enterprises by 1890, which became a
The increasing number of deaths by electrocution group called Edison General Electric and which, in
and the continual loss of market shares obliged 1892, merged with the rival company Thomson-
Houston to become General Electric. Thomas
Edision had by that time already lost control of
the company. The two enterprises, General
Electric and Westinghouse, both suppliers of AC
installations, now shared the United States'
electric power generation and distribution
market between them.
At the International Electro-technical Exhibition
held in Frankfurt, Germany, the year before, the
lights and motors of the exhibition had been
energized by a power station 175 km away via a
three-phase transmission line by way of a
demo.
The War of Currents had finally ended.
Even though certain small DC distribution
networks continued to operate until the end of
the '90's, AC electric power generation,
transmission and distribution dominated the
twentieth century and is still the most
widespread system.
However, direct current is now re-awakening the
interest of technicians and electricity authorities
since, thanks to the developments in power
electronics achieved over the last fifty years, the
main problem, i.e. voltage variation difficulties,
has been resolved. Modern static converters are
now so reliable and efficient as to call into
question the possibility of supplying direct
current by harnessing its inherent advantages,
first and foremost, lower transmission losses.
Figure 3: New York

in 1888
3. Applications and new scenarios
3.1 Evolution of electrical loads latter case), certain electrochemical processes,

railways, telecommunication systems and data
and power generation centers, auxiliary services and protection
Since AC distribution has always been systems for power stations and primary
predominant, designers of current-using substations, all UPS (Uninterruptible Power
equipment have been obliged to create such Supply) systems and, generally, all storage
devices or their feeders for use with the AC systems based on batteries.
voltage supply available. Failure to standardize Thus, the chance to save up to 15 - 20% of the
the supply throughout the world, where various total energy losses by supplying these loads
voltage levels are used and two or more directly in DC or improving converter efficiency
frequencies, mainly 50 and 60 Hz, has made it by eliminating the rectifier module, is bound to
even more difficult to develop products since increasingly encourage the use of direct current.
companies have had to differentiate them to suit
the market.
The recent and progressive introduction of digital
devices, the internal power supply of which is
necessarily in DC, has forced designers to 3.2 Power generation from
develop conversion stages, i.e. AC/DC rectifiers, renewable sources
so as to connect them to the grid, initially without
Another important thing to consider is the rapid
worrying too much about their efficiency.
increase in electric power generation from
Nowadays, and even though progress has been
renewable sources, backed by governments and
achieved as to the efficiency and control of these
public opinion worried about the rapid climatic
rectifiers, the massive presence of such loads is
changes (see Technical Guide: Smart grids 1.
forcing a rethink regarding the structure of the
Introduction).
existing power distribution grid.
The main sources of DC power supply are
Loads energized directly in AC are also rapidly
photovoltaic panels (PV) and, to a much lesser
decreasing in sectors such as that of electric
degree, fuel cells. At the present time, these two
motors, which has always been dominated by AC
sources need inverters to connect to the AC grid.
supply systems. Our homes and offices contain a
Electric power storage systems based on
huge number of personal computers, printers,
batteries also need inverters for connection, plus
telephones, televisions and Hi-Fi systems. All
a complex control system for the charging and
these devices are powered in DC via rectifiers and
discharging phases. Wind-powered generators
already represent about 20% of the total electric
can produce in both DC and AC. Nowadays,
power consumption. Considering the increasing
generation typically occurs in AC with long 36 kV
use of LED lights, much more efficient than the
connection lines between the wind farms and
incandescent ones invented by Thomas Edison,
distribution grid.
electric vehicles (charged in DC with a
In view of the widespread installation of
considerable load) and all those loads
photovoltaic panels on buildings for both
traditionally powered in AC (e.g. washing
domestic use and in the services-providing
machines and air conditioners) but which now use
sector, buildings could already be erected with
static frequency converters or inverters for
their own DC network, thereby achieving perfect
improved efficiency and to control the speed of
integration between generation and
their motors, it is easy to forecast that DC loads
consumption, as will be discussed further on.
will reach 50% of the total consumption within
the next few years.
It is worthwhile remembering other DC processes
used in industry, such as electric arc furnaces
(available in AC and DC, but more efficient in this
3.3 Static converters the TRANSISTOR MOSFET (Metal Oxide

Semiconductor Field Effect Transistor) is a
The current AC distribution grid structure needs component that can be switched on and off by
three types of static converter for connecting means of a voltage signal applied to a terminal.
loads and distributed generation: It switches very fast thanks to an extremely
AC-DC converters, also called rectifiers: these simple control circuit (turn-off times are around
devices supply a continuous outgoing quantity 100 ns);
from an alternating incoming quantity; TRANSISTOR IGBT (Insulated Gate Bipolar
DC-AC converters, also called inverters: devices Transistor). A component that combines the low
which supply an alternating outgoing quantity conduction losses of the BJT with the switching
from a continuous incoming quantity; speed of MOSFET;
AC-AC converters, devices with both an lastly, the MCT (MOS-CONTROLLED
incoming and outgoing alternating quantity. THYRISTOR) possesses many of the properties
They are normally used for starting and of the GTO, such as low dissipated power at
adjusting electric motors under the name of medium current and the same switching speed
drivers. as the IGBT. However, owing to its complexity, it
is unable to reach the high current values
Instead of the above, DC grids have DC-DC reached by the GTO.
converters. These devices have both an incoming
and outgoing continuous quantity for the Switching times and the power involved are the
purpose of varying the voltage level, as in the main and most significant differences among the
case of AC transformers. components described above. SCR and GTO are
to be preferred for high powers, since they
Different types of semiconductor devices are withstand higher voltages and currents than
used in static converters: those of a transistor (3000 V/ 3500 A for a diode,
diodes are devices that are only able to conduct 3000 V/1000 A for an SCR, 3600 V/600 A for a
current in one direction and cannot be GTO, 400 V/250 A for a transistor).
controlled;
SCR (Silicon Controlled Rectifier) belongs to the On the other hand, the switching times of a
thyristor family used for controlling the passage transistor are less than 10 s against 40 s for an
of current from anode to cathode; SCR and 25 s for a GTO.
GTO (Gate Turn-Off thyristor) is a device whose
transition from the conducting state to the Thus transistors are better if the power involved
inhibited state is easier to control than that of is relatively low and since they are easier to
the SCR; control, they become the only choice, especially in
TRANSISTOR BJT (Bipolar Junction Transistor) the case of high-frequency converters.
here again, this is a device that allows the Table 1 outlines the main characteristics
current to be controlled more easily; described above in a qualitative manner.
Diode SCR GTO BJT MOSFET IGBT MCT

Power handling capability High High High Medium Low Medium Medium
Switching speed - Slow Slow Medium Fast Medium Medium
Table 1: main characteristics of semiconductors for electric power conversion

The graph in figure 4 below translates these peculiar characteristics into industrial application terms:
voltage High voltage

lines
Thyristors
Heavy
traction
5 kV
4 kV GTO
Large household
appliances
Electric
3 kV
motors
MCT
2 kV
IGBT current
Light vehicles and
small household BJT
appliances 1 kHz
1 kV
MOSFET 10 kHz
100 kHz
1 MHz
500 A 1000 A 1500 A 2000 A 3000 A
frequency
Figure 4: limits and Depending on the type of use required and their DC traction power with alternating feeder lines
applications of
characteristics, some of the components (trolley buses, train and subway locomotives,
electronic components
for AC/DC conversion described previously can be used for designing a cable railways, cable cars, etc.);
static converter. The main criteria are: supply systems for certain electrochemical
voltage and current of the application and processes;
consequent losses; terminal converter substations for HVDC
switching time; transmission lines;
ease with which the component can be the incoming DC supply stages required from
controlled; DC/DC and DC/AC converters.
temperature coefficient of the component in
the case of connection in parallel to obtain DC/DC CONVERTERS
higher currents; In a DC-powered system, a DC-DC converter is
cost of the component. used to vary the DC voltage supplied to the load.
This variation is obtained by varying the fraction
AC-DC CONVERTERS or RECTIFIERS of time in which the load is connected to the
An AC-DC converter converts an alternating power supply. These converters are used as direct
(voltage or current) signal into a continuous current feeders in different sectors:
signal. Rectifiers can be used for numerous electronic equipment such as computers;
purposes: avionic and space applications;
DC drives for industrial use (e.g. rolling mills, laboratory power suppliers;
wire drawing machines, newsprint machines, speed adjustment of motors in DC-powered
conveyor belts, positioning systems, robot electric traction systems and with DC
drives, etc.); propulsion (railways, subways, electric vehicles).
DC-AC CONVERTERS AC-AC CONVERTERS

These devices, also known as inverters, convert An AC-AC converter consists of a rectifier, which
incoming direct voltage to outgoing alternating may or may not be controlled, installed in series
voltage, which can be adjusted both as to with an inverter. Using these converters, it is
amplitude and frequency. These converters are possible to both vary the amplitude of the output
used as AC feeders in different sectors: signal by adjusting the trigger times in the
variable speed drives in DC systems with AC rectifier stage or the trigger sequence in the
motors, where the motor must be supplied with inverter stage, and to change the frequency of
voltage and current with variable frequency and the output signal by adjusting the on and off
amplitude; times of the components in the inverter. Some of
uninterruptible power supply units powered by the uses for these types of converters are listed
DC via storage batteries in which the inverter below:
draws the power required to supply the load in the in the power supplies of aircraft so as to obtain
absence of mains voltage; the 400 Hz frequency typical of avionic
medium frequency furnaces for induction heating; applications and for regulating lighting and
insulation stages based on the use of high- heating systems;
frequency transformers, especially as second in the drives of AC motors, especially for
stage in DC/DC converters when insulation starting and adjusting the speeds of large
between input and output is required. three-phase motors;
in household appliances, for governing the
speed of their small motors.
AC distribution networks normally include at

least three types of converters, such as rectifiers,
inverters and AC/AC converters. The evolution of
electrical loads and distributed generation
involves the increasing use of static converters.
Figure 5 below illustrates the simplified structure
of a modern AC distribution grid.
Non-
Fuel synchronous
Photovoltaic
cells generation
AC power
Existing AC (wind)
transmission generation
infrastructure
HVDC system
MVAC
AC loads
Electric vehicles AC loads
Data Centers
Figure 5: structure
Variable frequency Storage at
of a modern AC Priority loads
drives distribution
distribution grid
level
A similar structure but with DC distribution is

illustrated in the next figure 6:
Fuel Photovoltaic Non-

Existing AC cells power synchronous
AC
infrastructure generation generation
transmission
(wind)
Future
HVDC
connection
HVDC system
MVDC
Loads with DC
distribution
Electric vehicles AC loads
Future DC industrial Future DC Data Motors

site Centers
Storage at
Figure 6: hypothetical Variable frequency distribution Priority loads
DC distribution grid drives level
As can be seen by comparing the two diagrams, 3.4 Protection of static

the DC distribution system requires DC/DC
converters, which actually replace the AC/DC
converters for AC/DC
rectifiers almost completely. interfacing
In addition, a process for standardizing the DC
As explained in the previous section, DC
voltages on the market would reduce the need to
microgrids normally include power generation
use many of the DC/DC converters on individual
from renewable sources (e.g. photovoltaic
loads. DC distribution would therefore be
systems) and will also include DC storage
particularly convenient in limited, independent
systems and local loads to an increasing extent.
areas and with specific functions, generally in
In addition, there will always be a static power
microgrids.
converter to interface with the AC grid.
The most widely used earthing system in medium
voltage AC installations is the isolated neutral
type, while several solutions are available for DC
installations, depending on the safety and
continuity level required. Not only does the
earthing system change, but also the behavior of
the DC system in the case of faults. In some
cases, the converters could be unable to limit the
fault current and could therefore be damaged.
This would require the installation of DC rapid
interruption apparatus.
Now let's look at the circuit illustrated in figure 7 below:
Cdc1 Isc
AC/DC Rdc1
MVAC Input interface VDC
grid filter converter
(FEC) Cdc2 Rdc2
Ppy/VDC
DC Storage
load PV system
system
S1 (ESS)
Figure 7: Medium PE
voltage DC supply RL
and conversion circuit
(AC system with
isolated neutral) a) b) S2
Where the midpoint connections a) connected to the FEC self-controls so as to keep the voltage
earth and b) negative pole connected to earth, are (VDC) at rated value. The value of the current
alternatives. The circuit includes: supplied will be higher, but still close to the value
a bidirectional AC/DC interface converter of the rated current of the actual FEC. If current
(Front-End Converter or FEC) which controls the Isc exceeds the threshold value, the converter
voltage value on the DC side; protection system will limit the current
an Energy Storage System (or ESS) to ensure a consumption of the AC network to a value that
continuous supply to priority loads, normally will not damage the components, thus also
connected by means of a bidirectional DC/DC limiting the active power transferred to the DC
converter; network. In these conditions the voltage (VDC)
a photovoltaic system (PV) connected to the DC decreases until reaching the value corresponding
system by means of a DC/DC step-up converter; to the maximum power absorbable by the AC
a generic DC load represented by a generic side. In addition, if the fault resistance is very
resistive element. small and the VDC drops to less than 2 2 V1
If a fault occurs between the poles immediately (where V1 is the root-mean-square value of the
on the load side of the FEC, the fault current neutral phase voltage on the AC side), the
depends on the resistance of the fault itself. If converter will begin to over-modulate. In this
this is less than the maximum threshold setting condition and for a fraction of time that increases
of the protection of the electronic components, as the fault resistance decreases, the current will
DC
load
Figure 8: currents in
the converter circuit in
the presence of short-
circuit between poles
no longer be controlled and could exceed the The next situation considered is when the
value established for switching the components. midpoint is earthed (connection "a" in figure 7).
This consequently creates a distortion in the The voltage of the pole affected by the fault tends
current on the AC side. Lastly, when the fault towards zero, while the voltage in the unaffected
resistance is near to zero, the converter acts like a pole fluctuates and can reach the full rated
rectifier with short-circuited diodes. In this case, voltage value. This could create insulation
the semiconductors controlled are excluded and problems (if sized for VDC/2). If an earth fault
the current passes through the freewheeling occurs in a generic load, current lg will flow
diodes (figure 8). through the midpoint (figure 9). Since direct fault
current does not reclose through the converter,
In these conditions, the only thing that can limit this latter continues to maintain constant VDC
AC current consumption is the impedance of the voltage on the load regardless of the fault
MV network, with consequent damage to the resistance value.
semiconductors. The current is also strongly Current Ig is therefore excluisvely due to
distorted by the presence of low frequency capacitors Cdc having discharged. The situation
harmonics for as long as the fault lasts. does not change even in the presence of ESS and
Earth faults are the next situations to be PV. Both continue to supply the load normally but
addressed. Even though the power is notmally not the fault, apart from the transient due to the
supplied by means of the FEC, the ESS is also able capacitors having discharged. In the case of PV,
to provide the system with energy and is shown the current is limited to the maximum value of
in figure 7 as a voltage generator E0 . If a fault I PVmax (PPV/VDC).
occurs in a DC system, this converter initially If, on the other hand, the negative point is
supplies a peak current due to the capacitors earthed (connection "b" in figure 7), an earth fault
having discharged and will continue even in the positive pole will cause a short-circuit
afterwards to supply the fault by means of the between the poles. This means that no direct
freewheeling diodes without any further control component passes through the converter, which
by the converter. acts as though the fault were a low impedance
In photovoltaic installations, the task of the load. This case is similar to the first one
converter is to maintain the operating point of considered (fault between poles immediately on
the installation at its maximum power and adapt the load side of the FEC).
the voltage to that of the system. In this case, it is Further details are given, especially about LV
depicted as a current generator equal to P/VDC. If grids, in Technical Application Papers No.14
a fault occurs in a DC system, there will initially be faults in LVDC microgrids with front-end
a peak current after which the converter will converters and in [36].
continue to supply the fault with a short-circuit In view of the above, it is evident that to prevent
current equal to 1.25 times the maximum current the electronic components from being damaged,
in normal conditions. Here again, the converter is the interface converter on the DC side must be
unable to limit the current, which flows by means protected against short-circuit between poles
of the freewheeling diodes. and against earth faults by high-speed circuit-
breakers.
DC
load
Figure 9: currents in the

converter circuit in the
presence of earth fault
3.5 DC microgrids 3.5.1 Office blocks and shopping centers

Many public and private consortia and initiatives
A microgrid can be considered as a group of throughout the world are now actively involved in
electric power sources, loads and storage the subject of Smart Buildings. Use of an internal
systems with well-defined limits from the DC distribution grid is one of the most
electrical aspect, which functions both as an interesting issues being addressed. Regarding
isolated entity disconnected from the network office blocks and shopping centers, it is
and, vice versa, connected to the network itself. worthwhile mentioning the initiatives promoted
Thanks to the development of new types of by Emerge, a consortium of companies (figure 11),
electrical loads and sources, this definition can professionals and institutions to which ABB also
cover many applications especially in future- belongs and which aims to standardize the DC
oriented terms. electric power distribution system in office
buildings and shopping centers. Achieve zero
energy balance in the next generation of
Figure 10: Shopping
Center buildings is the goal.
The basic idea is to promote the use of direct

current and integrate the main services inside
and outside the buildings so as to achieve
enhanced flexibility and reliability, consume less
energy with lower investments and obtain
buildings with zero energy balance where the
power consumed is the same as that produced.
The DC grid would distribute the power
(generated by photovoltaic panels for example),
store it and use it to supply indoor and outdoor
LED lights, the Data Center and all the computers,
communication devices, charging points for
electric vehicles and the building monitoring and
automation system (figure 12).
Figure 11: the companies

belonging to the
Emerge consortium
Solar
panels Controlgear /
Switchgear
Lighting and light

Motion
intensity sensors
detectors
Security
cameras Inverter
Power
distribution Air conditioning
panels control
Information
display Door
surveillance
system
Conditioning, temperature and
humidity metering
Power storage
system
Electric
vehicles
Charging station
Figure 12: DC loads
and supplies in a Building management
residential estate system
Thus the consortium also focuses on creating 3.5.2 Rural applications

new standards, the aim being to reduce the The cost of installing the infrastructures required
number of conversions in the connection to convey electricity to remote rural areas can be
between sources and user devices to the prohibitive. A microgrid could be the ideal
minimum. solution in such cases, since it would allow the
area in question to become independent thanks
to distributed generation based on renewable
sources such as photovoltaic panels or wind-
powered generators. So much so, not only can
this solution make the electrification of remote
areas economically feasible, but also sustainable
from the environmental aspect. The stability of a
microgrid can be increased by means of an
energy storage system based on batteries. The
typical electric loads range from high-efficiency
LED lighting both inside the buildings and
surrounding area and along the roads, to both DC
and AC electric motors and pumps,
communication systems, computers and other
electronic devices, which could even be directly
supplied by DC.
Figure 13 gives an example of how a photovoltaic
source and water pumping and distribution could
be integrated for a rural application.
Photovoltaic system
Well
Power control system
Tank
Electric power supply
Drinking
fountain
Electric pumping unit Drinking trough
Figure 13: Rural electrification,

diagram of installation
As already explained for office blocks and convenient. The remaining AC loads and the
shopping centers, here again DC loads will energy generated by wind-power could be
probably become predominant in the future and connected by means of inverters (figure 14).
this would make DC electric power distribution
Figure 14: example

of rural microgrid RURAL MINIGRID
Hybrid storage system
Power generation from

Hydrogen tank
renewable sources
Electrolyzer Batteries
Super capacitance Fuel cell
DC distribution Connection to
other DC public grid
clients in DC or DC
Computers, Resistive loads:

electronics and
other DC loads
Direct Alternating
current current
motors loads
Compared to an AC microgrid, the DC microgrid The typical voltage to earth ranges from 12.7 to
can lead to significant savings since losses are 19.1 kV, covering tens or even hundreds of
less thanks to the fact that fewer rectifiers are kilometers. When making an economic appraisal,
required. one must first consider the economic viability of
Although they are designed to operate in the laying one single conductor and then the cost of
islanded mode, connection to the public grid can the isolation transformers and other losses due
offer further guarantees as to the stability and to earth return resistance. Regarding this latter,
continuity of the electric power supply. The cost one should bear in mind that in alternating
of this connection would obviously depend on the current systems the return current follows, in the
length of the connection line and this could be ground, the path taken by the electric line above
especially critical in very extensive countries. The even if the route is not linear. In addition, the layer
so-called Single Wire Earth Return (SWER, figure at which it penetrates into the ground is
15) is an already adopted AC solution. This system proportional to , thus the theoretical section
was developed in New Zealand during the '20's. is proportional to 1/f while the resistance per unit
By means of an isolation transformer, it allows the of length is proportional to f. Over 200,000 km of
single-phase circuit to be closed by the earth, such lines have been installed between New
thereby saving on a conductor. Zealand and Australia to date.
A direct current line could also be a practical
method for connecting remote rural areas to the
public grid thanks to the low losses offered by
this solution (figure 16). Another advantage
offered by the DC microgrid would be savings on
the installation of an inverter and transformer
22 kV / 19 kV insulation transformer Other users compared to a simple DC/DC converter. Here
again, use of direct current would allow a single
Earthing switch conductor to be used by returning via ground.
Fuse However, compared to the alternating current
Surge arrester
solution, current distribution in the ground is only
determined by the resistivity of the various layers
19 kV in the subsoil. Current penetrates at great depth
transformer at if the connection points are very far from each
240/0/240 V
other. Thus the current density is very small. In
Single wire addition, the return current does not follow the
distribution system
with earth return Earth return line route but the path with the least resistance.
path Thus the ohmic resistance is small and practically
Separate earthing system for
high and low voltage equal to the resistance of the earth electrodes at
the terminals.
Figure 15: circuit diagram

for a SWER line
DC
microgrid
A.C. grid
Rectifier on Medium voltage Low voltage
grid side DC line DC line
Figure 16: connection

line between AC network
and DC microgrid
A DC distribution line involves considerable initial 3.5.3 Electrical ships

costs due to the static AC/DC converters. Ship installations have undergone considerable
However, the investment improves as the length developments over the years. Safety is obviously
of the line increases and finally becomes of fundamental importance on ships, but other
extremely convenient. Proceed as follows to issues such as reducing consumptions and
calculate this convenience: environmental impact have also acquired
We'll begin with a DC line with a single conductor. increasing priority.
If the section of the conductor is A and its length The growing demand for enhanced safety
I, the volume of that conductor is: imposes stringent requirements as to reliability
and manoeuverability. In addition, reduced
consumptions and lower environmental impact
The losses due to the Joule effect are: require greater efficiency when it comes to the
propulsion system and, more generally, the
technological systems on board, as well as
reduced weights and volumes, and systems for
the electrification of ships moored alongside the
While the initial costs of the conductor are:
quay (Shore-to-ship, figure 18). This last solution
allows ships to connect to the port electricity
grid and, thus, to turn off their engines and stop
having fixed the maximum losses p M and producing harmful emissions, acoustic pollution
consequently calculated the section. and vibrations when berthed, especially as ports
are often located in city areas.
In the case of a three-phase AC system, the
volume equals: The sum of all these requirements has
encouraged the progressive electrification of
ships. Initially built with a steam-powered
And the losses: propulsion system, the Queen Elizabeth II was the
first ship to adopt al electric propulsion system in
1987. Thanks to this conversion, the ship's
efficiency as to fuel consumption improved by
Lastly, losses pM being equal, the cost is: some 35%, with savings amounting to 12 million
pounds a year.
To get an indea of the power involved in a cruise

The ratio between the costs of a three-phase AC ship, we can consider the Costa Fortuna as an
line and a single DC line is therefore: example (figure 19). Built in the Fincantieri
Shipyard of Genoa-Sestri on behalf of Costa
Crociere, the Costa Fortuna is the largest cruise
ship ever built for an Italian company.
Thus the lower cost of the direct current The generation system comprises 6 alternators
conductors compensates, as the length of the line for an overall 90 MVA electric power (enough to
increases, for the higher initial cost until it supply a town of 50,000 inhabitants), connected
actually becomes convenient compared to a to a 6.6 kV AC medium voltage distribution
Figure 17: graph similar alternating current solution. The next switchgear able to supply the hotel services and
illustrating the cost-
graph in figure 17 illustrates this condition in the propell the ship. The actual propulsion system
effectiveness of a
DC connection line qualitative mode. comprises two synchronous electric motors, each
able to deliver 20 MW propulsion power at a
Cost of rotational speed of 140 rpm and drive the vessel
investments at a speed of 23 knots. The rotational speed of
Total AC costs the electric motors is controlled by static drives
at variable frequency allowing the entire speed
Total DC costs range to be accurately regulated. To do this, ABB
line
developed the Azipod propulsion system (figure
of D C
Costs 20) where the electric motor with propeller is
line encased in an pod-shaped enclosure with good
f AC
so hydrodynamic characteristics able to move
C ost
Cost of
DC through 360 degrees under the ship. The system
stations
has evolved into an industrial standard for the
Cost of AC stations
shipbuilding sector.
Distance
Critical distance
This positionable propulsion system can drive

and steer the vessel at the same time and reduces
Complete system on board, fuel consumption by up to 20% while ensuring
including the land-side
connection panel and
accurate manoevrability without the aid of tugs.
cable drum
According to Clarksons Research, one of the

most famous brokers among the principal
shipping companies, the number of ships with
electrical propulsion systems is increasing at the
rhythm of 12 percent per year, three times faster
than the world fleet.
Substation (with 50/60
Hz converter)
If this is what is happening to the merchant fleet,
Transformer on military vessels are proceeding towards
6.6 kV / 11 kV the quay
connection point electrification at an even faster rate owing to the
Underground HV cable (distance 1-5 km) high electric power consumption of new weapons
such as railguns and laser weapons, not to
mention radar systems and increasingly
Figure 18: shore-
to-ship system sophisticated communication systems. The
power required by the facilities on these ships
now amounts to about one third of their
propulsion power but according to forecasts, it
will equal or even exceed this value in the future.
In view of the power involved, the well-defined

limits and the load and generation distribution,
the electrical systems on all-electric ships (AES)
are microgrids to all effects.
However, problems concerning the use of
alternating current were already observed in the
case of the Queen Elizabeth II. For example,
generators must function at constant speed, with
loss of efficiency at different navigation speeds.
Reactive power is generated and problems
concerning quality can arise, such as phase
unbalance and the introduction of harmonics; the
transformers are extremely bulky and heavy;
inability to supply large pulsating loads.
Figure 19: modern ship

with electric propulsion These problems are encourging researchers and
engineers to look into the possibility of adopting
a completely DC distribution network. Beside
automatically resolving the electrical problems
bound to the very nature of AC, DC networks do
not merely use static converters (lighter and more
compact) instead of transformers, but also allow
the size and weight of the electrical distribution
switchgear to be reduced. In addition, the drivers
for feeding and adjusting electric propulsion are
lighter and more compact (DC/AC instead of AC/
AC). Examination of a specific case on a mega-
yacht highlighted a significant convenience with
DC, with 40% less weight and 80% less volume
than the similar AC solution.
Moreover, use of variable speed generators
(possible thanks to the decoupling allowed by the
AC/DC converters) saves up to 15% fuel. DC
distribution in military vessels allows the power
supply to support the most sophisticated
weapons and guarantees vital continuity of
Figure 20: Azipod

propulsion system
service thanks to a more versatile architecture involved are such as to require medium voltage
ensuring control over the aforementioned loads, distribution exceeding 6 kV DC (e.g.: 10 kV DC).
over generation and the ESS (Energy Storage Compared to microgrids for landside
System). installations, a medium voltage ring solution is
Without ESS, the pulsed energy required by the used in AES (a solution usually applied in Data
new weapons would be supplied directly by the Centers), divided into zones (DC Zone Electrical
generators which, in order to do this, would have Distribution System or DC ZEDS) as illustrated in
to be oversized. figure 21. These ZEDS are connected together and
When it comes to designing an electrical system, isolated from each other, and supply all the loads
an AES possesses the same structure as a DC except for particularly important ones like radar
microgrid for landside installations. The powers and the propulsion motors, which are directly
connected.
MVDC
Load
Zone 2 Zone 3 Zone n

Zone 1 Load
Load
Load
Figure 21: circuit

diagram of an AES
Since the power is generated in alternating

current by generators connected to diesel
engines or gas turbines, rectifier converters are
installed immediately on the load side of the
generators. The main advantage of this
configuration is decoupling of the generators,
SFOC (g/kWh)
which can function at optimum speed as to

efficiency in every load configuration without the
problem of having to maintain a certain
frequency. One of the solutions in large ships
with AC distribution is to operate the diesel
generators at their optimum speed and modulate
their number so as to meet the power
requirements of the loads (example in figure 22:
Specific Fuel Oil Consumption (SFOC) on the
Load conditions (pu) basis of the load conditions, using 1 to 4
generators K).
Figure 22: example of
consumption optimization
for four diesel generators However, this adjustment is slow and not very
accurate and the generators may not always
function at maximum efficiency, if only for short
periods. Thus MVDC distribution appears to be
the true solution if maximum fuel savings are to
be achieved.
When MVDC power is distributed, there is no Switchgear equipped with DC circuit-breakers are
need to step up the voltage in order to supply used for MVDC distribution. Circuit-breakers
direct loads. Initial rectification of the power designed for railway applications could already
produced by the generators is all that's required. be used for this sort of distribution even though,
In MVDC distribution, all the ZEDS are supplied in unfortunately, the maximum voltage level (3.6 kV)
DC and converters change the voltage level or is modest for this purpose. On the other hand,
convert it into AC at the frequency required by the there is a whole range of low voltage DC circuit-
loads. In this latter case, transformers can be breakers available. Each zone could be supplied
installed to lower the voltage level still further. by each of the two switchgear in the distribution
UPS with DC/DC converters are installed on ring so as to guarantee maximum continuity of
military ships near the pulsed power loads of service. Figure 24 below shows an example of a
modern weapons for the purpose of tug with MVDC distribution.
compensating the peak power demand and
optimize the size of the generators. As to fault conditions, although, on the one hand,
static converters can limit the outgoing power
thereby preventing serious damage to the loads,
on the other, this situation can lead to critical
conditions in the supply of loads that are not
affected by the fault. To overcome the problem in
microgrids, the tendency is to oversize the
converters. This is not the ideal solution for on-
board installations where space and weight are
subject to limitations. Fault elimination by
isolating the affected area and guaranteeing
continuity in the other zones is of fundamental
importance in large vessels. The availability of
suitable protection devices, circuit-breakers and
load disconnectors is a critical issue currently
being addressed by major research and
development initiatives. Similarly to other
microgrids, reconfiguration after different load
or fault conditions is extremely complex and
needs a centralized hierarchical control system
and an advanced communication system able to
Figure 23: diesel
generator
handle it.
Figure 24: ship with

MVDC distribution
Diesel generation
Energy storage system
Electric propulsion
Thruster
MVDC transmission line
Electronic power substation
DC/DC converter for LVDC grid
LVDC loads
Central control station
Communication system
3.5.4 Data Centers Nowadays, data centers (also known as server

farms) contain most of the information required
for the functioning of our societies. This means
that the availability of data is absolutely the most
important issue to face when the relative
electrical system is designed. However, a second
aspect, i.e. reduced consumption, is peremptorily
emerging as the dimensions and, thus, the power
required increase and the cost of electric power
rises.
According to estimates, the overall cost of the
power consumption of data centers for
energizing and conditioning increased from 10
billion euros in 1996 to about 40 billion euros in
2010.
The typical loads of a data center are digital
devices such as DC-powered computers and
communication systems. The conditioning
system is another important load in a data center,
and the power required is on the increase in view
of the progressive increase in the power of
processors. In the current architecture where the
distribution system is in AC, power suppliers, i.e.
Figure 25: inside a
data center the numerous AC/DC converters with their typical
65 70% efficiency, also consume a large
amount of energy (figure 26).
A typical data center is energized by two or more

Cooling AC sources formed, for example, by two
independent lines connected to the public grid
Supply and an emergency diesel generator. The typical
supply voltage is 400 V AC. Then there are UPS for
the purpose of ensuring maximum continuity of
supply and 400 V AC distribution for powering
Server support
Total cost
the various digital systems, perhaps with further

voltage reduction, and the conditioning system.
Network
As can be seen in figure 27, in this type of
architecture there are at least three conversions,
from AC to DC or vice versa, in order to obtain DC
N servers
supply at the voltage value required by the
electronic components.
Figure 26: types and

distribution of loads
in a data center
AC AC Server rack
400V 400V
System board
System board
Public grid System board
Batteries
Diesel
generator
Public grid
Figure 27: architecture
Non-critical
of a data center with AC
AC loads
distribution system AC loads (cooling)
Starting from the point of supply of the public Thus, by using latest generation UPS and
grid, one can consider an overall 58% efficiency improving the efficiency of the power suppliers
(figure 28). (the weak point in the chain), it is theoretically
possible to achieve approx. 80% efficiency.
70% power supplier Server rack
Figure 28: efficiency

of the components of
an AC supply system
in a data center 400 V AC 400 V AC V AC V DC 12V/5V/3.3V DC
AC-DC losses In the example in figure 29, which illustrates a

typical server board, the losses due to the power
DC-DC losses
supplier account for 36% of the power
Fans consumption and the rectification stage actually
accounts for 29%.
Drive Given the predominance of digital loads, one
PCI boards
might think that use of a direct current
distribution system would optimize consumption
Processors and reduce the number of conversions. Figure 30
shows that compared to the AC distribution
Data storage
Figure 29: division of system, there is only one initial rectifying
losses in a data center Other electronic conversion.
components
DC Server rack
380V
System board
System board
Public grid System board
Non-critical
AC loads
Public grid
AC loads (cooling)
Non-critical
AC loads
Diesel
Batteries
generator
Figure 30: architecture
of a data center with DC
distribution isystem
Calculated from the public grid's point of supply 3.5.5 DC industrial installations
through to the end consumer, the overall The production process in certain industrial
efficiency in this case, is 86%. enterprises requires direct current, this either
because it is required by the process itself or to
reduce consumption as an alternative to a similar
Power supply process in AC.
400 V 12V/5V
V DC 3.3V DC This is the case of certain processes in the
AC
chemical and steel industries.
The electro-winning process illustrated in figure
32 is commonly used for the electrodeposition of
PV metals from ores that have been put in a solution
Batteries generation via a preventive process commonly called
leaching, during which soluble components are
separated from a solid mass by means of a
Efficiency is thus in favour of DC distribution in
solvent. Electrorefining uses a similar process to
Figure 31: efficiency Data Centers, although not to a very great extent.
of the components of remove impurities from a metal. Both processes
DC distribution by the actual Public Utility
a DC supply system use electrolysis. During electrolysis, current is
in a data center Company, first in MVDC and then in LVDC, could
passed from an inert anode through a liquid leach
evidently make the difference, since it would also
solution containing the metal, so that the metal is
do away with losses due to the rectifier and thus
extracted as it is deposited onto the cathode.
raise the efficiency to 93%.
In electrorefining, the anode consists of
Generally speaking, from the reliability aspect,
unrefined impure metal. The current passes
one could say that (for similar components), the
through the electrolyte and the anode is corroded
fewer the components involved the better overall
into the solution so that the electroplating
reliability becomes. Furthermore, rectifiers are
process deposits refined pure metal onto the
generally less reliable than DC/DC converters
cathode. The most common electrowon metals
since they are more complex. Consequently, DC
are lead, copper, gold, silver, zinc, aluminum,
distribution appears to be a better choice as
chromium, cobalt, manganese, rare earths and
regards reliability.
alkali metals. For aluminum, this is the only
The installation of photovoltaic panels for local
production process employed. Many
power generation could provide added impetus
electroextraction systems are used for removing
since DC distribution could save on the inverter.
toxic (and sometimes valuable) metals from
In short, as local generation increases, part of
industrial waste.
which directly in DC, Data Centers would become
During electro-winning or electro-refining, direct
increasingly similar to DC microgrids able to
current is used by means of a static rectifier.
function in the islanded mode, thus separately
Figure 33 illustrates the ABB solution for a copper
from the public grid.
or zinc electro-winning installation.
Figure 32: copper

refining installation
Medium voltage
Medium voltage busbars
MV circuit-breaker
Regulating autotransformer
HMI / DCS
AC current metering
Transformer
Rectifier Rectifier Rectifier transformer

monitoring system monitoring system
Rectifier
DC current metering
DC disconnector
Process
Control system
Human Machine Interface /

Digital Control System
Harmonic filter
As can be seen, DC disconnectors are installed in the relative production facilities are different, but
Figure 33: circuit
diagram of an electro- the direct current part since the task of limiting only as to plant engineering. During the process,
winning system the fault current is left to the static converter. The known as Hall-Hroult, the aluminum is produced
system is protected in AC by circuit-breakers for in an electrolytic cell, where the electrolyte
each riser. consists of a cryolite and alumina bath. Liquid
Similar systems are used for the production of aluminum is deposited at the cathode, which
alkali-chlorides such as chlorine (CI2) and alkalis, consists of a casing made of refractory material
sodium hydroxide (NaOH) and potassium housing the electrodes. These installations are
hydroxide (KOH), again by means of electrolysis enormous and require a vast quantity of AC
of a salt solution. power, which is rectified into DC so as to supply a
When it comes to aluminum (which is only large number of cells.
extracted by means of electrolysis - figure 34),
Figure 34: aluminum

production plant
The typical layout proposed by ABB as as follows:
HVAC
HMI / DCS
Pre-magnetization General Main control

panel protection panel
Rectifier monitoring
system
HV busbars
HV circuit-breaker
Regulating autotransformer
MVAC
Rectifier transformer
Rectifier
DC distribution DC current metering
DC disconnector
Process
Control system
Human Machine Interface /
Digital Control System
Saturable reactances
Pre-magnetization panel
Ring rectifier
Speed-up rectifier
Double-star transformer with
interphase transformer
Figure 35: circuit

diagram of an aluminium A real DC distribution system can be noted in this Electric arc furnaces are in a different category as
production facility case. As in the previous situation, numerous DC to type of finished product and production
disconnectors have been installed since the task process (figure 36).
of limiting the fault current is left to the static When it comes to supply voltage, there are two
converter. Again, the system is protected in AC by types of electric arc furnace:
circuit-breakers for each riser. alternating current (AC) electric arc furnaces.
Figure 36: electric These are normally energized by three-phase

arc furnace
voltage and therefore feature three graphite In addition, the presence of one single electrode
electrodes installed in line (no longer used) or simplifies the design of the crown of the furnace,
which form an equilateral triangle. The arc the tilting mechanism and electric system (with
strikes directly between each electrode and the the exception of the alternating-to-direct current
scrap metal. Once a certain level of molten rectifying system)
metal has been reached, the electrodes bore The typical circuit diagram with the control
into it and continue to produce heat while system proposed by ABB is illustrated in figure 37.
waiting for the entire load of scrap metal to
reach melting point; The advantage of being energized by a direct
direct current (DC) electric arc furnace: unlike current public distribution grid for all these
the alternating current furnace, this type has production systems with processes energized in
one single electrode in the central position. The DC via static rectifiers connected to the medium
arc strikes between the electrode (cathode) and or high voltage AC public distribution grid should
three "anodes" forming a triangle on the be evident. It would eliminate the need for
bottom of the furnace. This configuration transformers and rectifiers, which could be
forces the electric current to pass through the replaced by DC/DC converters so as to use a more
scrap metal, which melts. suitable voltage level, consequently cutting down
on investments and improving the overall
The DC solution provides certain advantages: efficiency. Local generation from renewable
the arc is more stable, thus there is less wear on sources and storage systems able to optimize the
the electrode. The arc strikes at an angle of up supply, all in DC, would complete the conversion
to 30 on the molten metal and tends to turn into DC microgrids. In systems where continuity
continuously along the vertical axis owing to of service is absolutely essential owing to the
the effect of the magnetic field; particular production processes, this would
enhanced efficiency when transferring power to improve reliability and independence from the
the molten pool plus improved localization of public grid.
the central hot spot compared to the three
characteristic spots of the three-phase system;
less network disturbance.
Figure 37: circuit

diagram of an electric
arc furnace
Medium voltage
MV busbars
MV circuit-breaker
Filter
Transformer
Rectifier
Rectifier Power Quality Arc stabilizing reactance
monitoring system control system HF1 ... HFn Disconnector
Adaptive arc control
Optical fiber connection
Power Quality control system
Harmonic filters
3.5.6 HVDC transmission lines Comparison between a high voltage DC

transmission line (HVDC) and a similar
transmission line in AC shows that conduction
losses due to the active power flow are less and
those due to reactive power are actually nil. In just
a few words, this explains the strong interest in
HVDC installations and their rapid increase when
the length of the line leads to a cost of losses able
to balance the higher cost of the static conversion
equipment.
The highest voltage normally used for an AC line
is 800 kV. However, owing to the increased
demand for transmitted power, 1000 kV and even
1200 kV have been considered for some
installations. At over 800 kV, the power dissipated
owing to losses through dielectric hysteresis is
considerable. These losses are due to polarization
phenomena, i.e. rotation and deformation of the
molecules in the presence of a variable electric
field, and depend on the intensity and frequency
of the electric field itself. Their continuous
movement heats the material by the same
principle as a microwave oven.

Figure 38: HVDC
conversion station
Power transmission at 1000 kV AC Figure 39 illustrates the capacitance of a 1000 kV

AC transmission line with 70% compensation and
a 30 degree angle between the terminal points.
Power in MW
Distance in km

Figure 39: capacitance
/ distance graph
for a 1000 kV AC
transmission line
It seems that even for long distances, over 1200

km, a 1000 kV AC line can begin to develop
problems when transmitting high powers. When
Power 12000 MW one assesses the investments required in relation
Length of line 2000 km
800 kV AC 8 lines to the losses (figure 40) for a line able to transmit
Costs in MUSD
1000 kV AC 5 lines 12 GW over 2000 km, one finds that the minimum
500 kV DC 4 lines
800 kV DC 2 lines investment is obtained with 800 kV DC lines.
Generally speaking, the higher the power
transmitted and the longer the line, the more an
HVDC line becomes convenient.
When it comes to rated voltage, what is actually
meant by AC and DC rated voltage must be clearly
specified. In the first case, voltage refers to the
Line losses in % root-mean-square value between two
conductors, which more or less corresponds to
the DC line-line voltage in terms of transmission
Figure 40: graph of
investments in relation
capacity. Thus a 500 kV HVDC line has double
to percentage losses the capacity of a 500 kV AC line (figure 41).
Another advantage of DC lines is that they are not

affected by the so-called skin effect, i.e. the
tendency for alternating current to distribute
itself within a conductor in an uneven way, with a
greater density near the surface than in the
interior.
The depth at which the current penetrates into
the conductor, measured from its surface, is
given by the following formula:
and is about 10 mm at 50 Hz and 8.5 mm at 60 Hz

Figure 41: dimensional (figure 42).
comparison
between AC and DC
transmission lines Figure 42: current This means that there is higher electrical
penetration depth with
resistance in an AC conductor than in a DC one,
respect to frequency
which increases proportionally at the root of the
frequency. For a conductor with diameter D and
D>>:
with: magnetic permeability of the conductor,

specific resistance of the conductor while lastly,
is the angular frequency of the current, equal to
2f.
Thus the result is greater power dissipation,
applied current being equal.
If AC current is flowing through two or more
nearby conductors, the magnetic field generated
affects current distribution in these conductors
and forces it to concentrate on the adjacent
surfaces. This is called the proximity effect.
This means that in actual fact, the R AC/RDC ratio is during transients, e.g. by slowing the voltage rate
higher than that produced by the skin effect of rise at one end of the line from the moment in
alone. However, there are construction solutions which the other end is powered. Figure 43 shows
able to mitigate the skin effect in AC systems, e.g. the world's first HVDC connection made in 1954
by installing several conductors in parallel, each in Gotland, Sweden, with the contribution of ABB.
with R AC/RDC =1 (thus at 50 Hz with a thickness of Lastly, a further advantage of the DC connection
less than 15 20 mm), by plaiting the insulated is decoupling between two connected grids that
wires together, bycoating the conductors with a need not necessarily be synchronized (figure 44).
layer of silver, by using two metals with the more Passage through a conversion into DC makes the
precious one on the outside, or by making hollow two grids independent. In Japan, as many as
conductors. three HVDC connections are used to connect two
Whatever the case, the best solution is to use grids, 60 Hz in the western regions of Okinawa,
direct current, which is free from such effects. Osaka, Kyoto, Kobe, Nagoya, Hiroshima and 50 Hz
If a cable transmission line must be used, e.g. in the eastern regions of Tokyo, Kawasaki,
undersea connections between islands and the Sapporo, Yokohama and Sendai, which are
mainland such as the one between Sardinia and normally incompatible.
mainland Italy, the high capacitance to earth, The cost of the converter substations and circuit-
which is 50 100 times more than that of breakers in DC (figure 45) is thus the only factor
overhead lines, strongly limits the transmissible that limits widespread adoption of HVDC lines. In
power in AC. In DC, the effect of capacitance in actual fact, the same factor also limits the
service conditions is nil and can only be sensed adoption of MVDC distribution grids.
Figure 43: laying of an

HVVDC connection
in Gotland
Network A
HVDC
line
Network B

Figure 44: diagram of
Figure 45: converters
the interconnection
in an HVDC station
between two AC grids
3.5.7 Electric traction The first real electrification of a railway took

place in America in 1895 when 600V DC was used.
Switzerland was the first European country to
follow suit by electrifying the Burgdorf-Thun
railway at 750V AC. Not only was this the first
railway line in Europe, but also the first
experiment with AC traction. The threephase
electric traction motor already offered certain
advantages. It could function as a generator
downhill while acting as an engine brake and
supplying the trains with power uphill. After the
first experiments with DC locomotives powered
by batteries, energizing by a 3 kV DC direct
current catenary was chosen in Italy.
Spain, Poland and Belgium also use this system in
Europe. Other countries like Russia, the Czech
Republic, Slovakia, Slovenia and Ukraine use both
3 kV direct current and 25 kV 50 Hz alternating
Figure 46: Railway The history of railway electric traction is recent current. The 25 kV AC single-phase system (50 Hz
application
yet intense. The first electric train was built in almost everywhere) is becoming the standard
Berlin in 1879, but the potential of this system method for new railway electrification systems
was immediately evident. world-wide and has also been chosen in Italy and
Tramways and suburban lines were the first to be Spain for the new high-speed lines.
electrified, since electric traction was not If the future of railway electric traction tends
considered to be economically or technically valid towards alternating current as a power supply,
for heavily trafficked railway lines. Thus, by the the outlook is more complex for the urban
end of the 19th century, electric trams energized transport systems in medium sized and large
Figure 47: graph of by low voltage direct current (typically 600 and cities. Urban systems comprise buses, trams,
service investments
750 V DC) became widespread in the main cities subways and light rail (figure 47).
for different
transport systems of Europe and overseas.
These means of transport use low voltage direct
Line length investments
current supply systems for safety reasons. To

reduce pollution, buses are also evolving towards
electric traction, with fast battery charging
stations. For example, ABB has supplied the city
of Geneva with flash-charging systems and on-
board electric vehicle technology for 12 TOSA
A
(Trolleybus Optimisation Systme Alimentation)
Subways fully electric buses (e-buses) which will run on
Light rail Line 23, connecting Geneva's airport with
suburban Geneva (figure 48).
B
Rapid transit
Trams Semi-rapid
transit
C
Buses
Road transit
Performance (speed, capacity, reliability)
Figure 48: bus equipped with ABB TOSA

ABB will deliver and deploy 13 flash-charging A recent development introduced for the purpose
stations along an urban transit bus route as well of reducing environmental impact in old town
as three terminal and four depot supply stations. centers is that of discontinous supply with a third
This is the same sort of system as the ones rail buried in the roadway which, for safety
designed for private use, since it requires the reasons, is only energized in the zone underneath
installation of charging points with AC/DC the tram itself. In order to supply trams, electric
rectifiers (figure 49). power from the AC public distribution system is
converted into direct current at between 500 and
750 Volts by means of conversion stations
situated in various parts of the city and is then
conveyed from these via underground cables to
the power boxes of the lines.
Subway trains also use DC power by withdrawing
Connection AC/DC
point for energy voltage from a third rail via contact shoes.
converter
transfer Compared to the overhead lines, this method
Connection to Energy storage does not need pantographs, thus the train can be
the grid unit narrower, ideal for long routes through tunnels.
The construction costs are also lower since pile
driving, installation of overhead lines and relative
systems for stringing the line are not required.
Figure 49: ABB TOSA The third rail system is one of the oldest methods
charging point
since it was already being used in 1890 in the
In tramway systems (figure 50), the electric line is
London subway, the first in the world (figure 46).
much simpler than that of the railway: just one
The maximum voltage used is 1200 V DC.
contact wire is used owing to the lower current
The return current normally runs along the rails
consumption. The suspension system is also very
where the train runs, but with some exceptions.
simple with cross-span feeder cable suspension
For example, the London subway, the M1 line in
on poles or fixed by insulators to the adjacent
Milan and certain lines in the Paris subway have
buildings.
another track for the return current. The
disadvantage of this system is that it is
unsuitable for high speed trains since the
maximum operating voltage is too low. To get
over the problem of low speed, some cities,
especially megacities, have installed a light
railway able to cover larger areas more efficiently.
In this case, the supply system is similar to that of
a railway with DC-powered overhead lines and
maximum voltage up to 3 kV DC.
The structure of a light rail conversion station is
illustrated in figure 51.
Figure 50: example

of a tramway
Control panel
Static rectifiers
Energy
Transformers dissipation
system
AC
switchgear DC
switchgear
Automatic
Energy recovery
Figure 51: railway Grounding Storage system
system
substation for DC System
conversion
Recovery of the braking energy in trains is an meet the individual requirements of different
issue of the highest concern since it saves on transport companies. A diagram of the complete
costs and enables train frequency to be increased braking energy control system is illustrated in
on long distance routes. ABB has developed a figure 52.
range of specific high-efficiency products able to
MV/HV network
Figure 52: diagram of
the complete braking
energy control system TDR
(Traction Diode Rectifier)
ERS energy recuperation

system
ESS energy storage
system
ARU energy dissipation

system
Voltage limiting
device
ABB's ERS and ESS systems from the DC Traction

Power Supply product range are shown in figure 53.
Figure 53: ABB's

ERS (left) and ESS
(right) systems
Use of direct current is not merely limited to the In short, the urban transport system can be
propulsion of vehicles. It is also used as a supply thought of as a direct current microgrid
source for auxiliary circuits in the vehicles operating at different voltage values depending
themselves. In these cases, accumulator batteries on the type of transport involved. Here again,
are installed as a source of backup power in the supply by a DC public distribution grid would
absence of the external source. eliminate the need for transformers and
Subways and light rail transit need hefty rectifiers, which could be replaced by DC/DC
investments but are able to transport large converters so as to use a more suitable voltage
numbers of passengers without polluting, so the level, thereby cutting down on investments and
long-term benefits they provide far exceed the improving the overall efficiency.
investments made.
4. Direct current interruption
4.1 Characteristics and problems 4.1.1 Conventional interruption with direct

current suppression
Direct current is a type of electric current With reference to the circuit diagram illustrated in
characterized by a current flow whose intensity figure 54, the following direct current circulates
and direction remain constant over time. In other when the circuit-breaker is closed:
words, the electrons always flow in the same
direction within the circuit. Consequently, unlike
circuits with alternating current, it can be
important to follow that direction, i.e. comply where U is the direct current applied, L and R are
with the polarity. Unless they are properly the resistance and inductance of the circuit. In
protected, some types of apparatus can break addition, the energy stored in service conditions
down if connected with the wrong polarity. is:
Since the current flow of alternating current
reverses its direction at every half cycle and
consequently crosses through natural zero, the
electric arc which forms during the current The moment in which the circuit-breaker opens, a
interruption process quenches naturally the transient caused by the appearance of arc voltage
moment in which the current returns to zero. Ua is generated. This also introduces a voltage
due to the presence of inductance L (figure 55). If
Crossing does not occur with direct current and the circuit-breaker is replaced by its arc voltage in
this makes the interruption process far more the circuit, the applicable formula becomes:
critical.
Now, we'll suppose that when the circuit-breaker

opens the current remains constant until the
instant the maximum Ua is reached, which we'll
consider as t=0, and that this latter remains
constant. In other words, we're talking about the
application, in t=0, of a generator of step voltage
Ua, which will disappear the moment ta in which
the current tails off.
Resolution of the differential equation gives:
Figure 54: DC electric

circuit with direct with =L/R.
suppression The principle of continuity in inductance L is
applied to calculate constant c, thus the current
in instant t=0- must equal the current in instant
t=0+.
Since:
and
the result is:
lastly:
Figure 55: DC circuit

during transient
The solution of the differential equation for t>0 is therefore:
Since the second term of the equation tends towards zero, it follows that the current only
crosses zero if Ua is higher than U (figure 56).
Figure 56: DC current

trend with direct
interruption
In this case, instant ta represents the arcing time and is thus worth:
The energy dissipated by the arc can now be calculated as:
Proceeding with the calculations, the result becomes:
Multiplying and dividing by U and bearing in mind that U/R=i0 and that W0=1/2 L i02
we can now write:
According to the theory expressed above, This assumption led to the development of high-
interruption will only be successful if the circuit- speed circuit-breakers. The same figure shows
breaker is able to bring arc voltage that exceeds how the interrupted current value also depends
the rated voltage of the circuit into play and if it is on time constant , thus on the parameters of the
able to support energy Wa (function of initial circuit in which the circuit-breaker operates. This
energy W0 and, thus, of the inductance of the is why manufacturers use DC breaking capacity in
circuit, and ratio Ua/U). conjunction with the circuit's time constant limit
values.
Ratio Wa/W0 normally varies from 1 to 1.5 and the
lower it is, the easier interruption becomes. With The limits of this application are evident:
reference to figure 57, which illustrates the real interruption in air is the ideal sort, since it is able
trend of a DC interruption, it appears evident that to support high arcing voltage and can easily
one of the methods for reducing fault current Isc dissipate the energy. However, the voltage cannot
and, thus, energy is to reduce, as far as possible, rise too much as Ua must always be higher than U.
the time between the beginning of the fault and This consequently leads to an increase in the size
the instant in which the contacts start to of the arcing chamber and is therefore not very
separate (called ts). cost-effective.
This type of DC interruption is widely used in low

and medium voltage systems for railway and
industrial applications up to 3600 V.
Other methods have been developed to overcome
the limits described above. For example, one
alternative is to force direct current crossing
through zero by changing the parameters of the
circuit, basically by inducing the appropriate
amplitude oscillation.
There are two categories: self-induced (or

passive) oscillation and forced (or active)
oscillation.
Figure 57: trend of a DC

direct interruption
4.1.2 Interruption with passive oscillating

circuit
An LC circuit is connected in the passive
oscillating circuit in parallel with the contacts of
the circuit-breaker (figure 58).
Where:
I is the circuit-breaker
II is the switching circuit
III is a zinc oxide protection varistor
L and R are the inductance and resistance of the
system
i is the current to be interrupted
is is the current in the circuit-breaker
ic is the current in the switching circuit
Lc and Cc are the inductance and capacitance of
the switching circuit
Rc is the intrinsec resistance of the
Figure 58: passive switching circuit

oscillating
ia is the current of the protection circuit
interruption circuit
As described previously, when the circuit-breaker Term is dimensionally a resistance

is closed, current circulates around the circuit which we'll call
through the circuit-breaker at the value of:
For the sake of simplicity, we'll assume that
i = U
is constant. In addition, to obtain an oscillation,
R the roots of the quadratic differential equation
must be complex and
The circuit-breaker is replaced in the circuit with
its arc voltage Ua upon opening (figure 59): thus
Lastly, as initial conditions, we'll write Ic(0)=0 and

Uc(0)=0 (capacitor discharged).
In the presence of these conditions, the solution
of the differential equation is as follows:
If then ic will oscillate at angular
frequency and with
Figure 59: circuit increasing amplitude, thereby enabling the

of figure 58 with circuit-breaker to interrupt at first zero current.
arc voltage
At this point, the entire current I is transferred to
the switching circuit, thereby charging the
capacitor Cc. Parallelly, the varistor limits the
During the transient, we can write that: maximum voltage at the ends of the capacitor to
the discharge voltage value.
The formulas above show that as the value of Cc
thus: increases, the oscillation frequency decreases
until it disappears but the oscillation amplitude
increases, thus a good capacitance value helps
the interruption process. Vice versa, the
increasing value of Lc not only causes the
frequency to diminish but increases
for the sake of simplicity, Rc will not be
considered, also because it is negligible; thus: the time constant of the circuit and
also decreases the oscillation amplitude .
Thus it is useful for the value of Lc to be low.

Another aspect to consider is that at high
Since the arc voltage is a non-linear function of breaking current values is very low since it
the current, i.e. , we can write: has a negative characteristic. So much so, the
time constant is high. This means that quite some
time can elapse before interruption occurs.
In conclusion, circuit-breakers based on this
By substituting, the result is: principle are unsuitable for interrupting high fault
currents. Figure 60 illustrates the current trend
obtained with Lc=40 H and Cc=9 F
Since, i = is + ic by expressing everything in relation to

ic and bearing in mind that i0 is constant, the result is:
Current [A]
Figure 60: current

trend in a passive
oscillating circuit time [ms]
4.1.3 Interruption with active oscillating circuit Auxiliary contact CS connects the capacitor Cc
This type of interruption overcomes the problem and inductance Lc in series at the same time as
of interrupting the high currents present in the the main circuit-breaker opens. In practice, the
case of passive oscillation circuits. The circuit is difference between active and passive
similar to the previous case, except for the fact interruption lies in the initial condition of the
that the capacitor is precharged. oscillating circuit since the capacitor is now
The principle of this circuit-breaker is illustrated charged, thus Uc(0)0. The relations in the
in figure 61: previous section can all still be considered valid
by adding the presence of a current of a sign
opposite to the current that needs to be
interrupted, the value of which depends on the
charging voltage of the capacitor itself Uc, and
which adds to the arc voltage Ua (figure 62).
Sure enough:
A special capacitor precharging system is

therefore required. One method is to use the
Figure 61: active

actual line voltage.
oscillating circuit
Figure 62: circuit

of figure 61 with
arc voltage
4.1.4 Interruption with forced oscillation

This method of interruption is characterized by the presence of forced oscillation generated by an
external source. As in the previous cases, this oscillation overlaps the DC current that needs to be
interrupted so as to force it to cross zero.
A typical circuit is shown in figure 63:
Figure 63: circuit with

forced oscillation
The typical differential equation of this circuit becomes:
Where Le is the leakage inductance of the transformer, Ue is the amplitude of external angular
frequency source e. Resonance is obtained when
The solution of the previous differential equation, with as a constant, is as follows:
The solution is formed by two terms: the first represents a weak oscillation with angular frequency c
given by Lc and Cc, as in the previous cases, and a second oscillation with angular frequency e imposed
by the external source.
When the current that needs to be interrupted is high and so long as becomes small, one can,
with a certain approximation, write:
This expression shows that for e = c, the denominator of the second term becomes null,
theoretically giving current with an infinite amplitude. Even in real conditions, one can see how
convenient it is to increase the amplitude of the voltage of external source Ue and decrease
frequency e to obtain a greater contribution.
4.1.5 Interruption with parametric oscillation The process is therefore repeated twice per cycle.
This method of interruption is based on the The oscillation is amplified until such time as the
parametric oscillation concept, i.e. an oscillator current crosses zero and the arc quenches. The
whose electrical parameters vary with a inductance can be changed in a similar way by
frequency near to the natural frequency of the varying the position of its magnetic core. Using
circuit. From that viewpoint, this interruption the relation already described for the passive
principle differs from that of forced oscillation oscillator gives:
but is still based on using an oscillating current to
overlap the direct current that needs to be
interrupted. But considering that the voltages on the
The reference circuit is shown in figure 64: components are now a function of the capacitor
charge, the result is:
Ua(t) can now be expressed as product of current

is(t) and resistance Ra(is(t)) which, in turn, is a
function of the current is(t) itself. In addition,
given that is(t)= i(t)-ic(t), is(t) is a function of the
capacitance of the capacitor, the result at the end
is a differential equation with variable
parameters, very difficult to resolve analytically.
Generally speaking, it is not easy to calibrate this
type of circuit, since it only starts oscillating in
the presence of certain circuit parameters.
parametric oscillation
As in the case of the passive oscillating circuit,

oscillation initially begins after the circuit-breaker
has opened, consequently causing an arc to
strike. A pumping effect is obtained if the
capacitance of capacitor C is varied with the
same natural frequency as the oscillating circuit.
In practice, the capacitor is first charged until the
stored energy is at its maximum, then its
capacitance C is reduced. For example, by simply
moving the plates away if the capacitor is the
parallel-plate type, thereby increasing the energy
in the circuit. After this, with the capacitor
discharged, the capacitance is brought to its
initial value by nearing the contacts but this
obviously occurs on an energy-being-equal basis.
4.1.6 Interruption with semiconductor With reference to the schematic circuit diagram,
technology semiconductor T normally conducts and carries
The intrinsic advantage of circuit-breakers based load current i. The current is interrupted the
on this technology (figure 65) is that since they instant that interruption occurs t=0 and
are ultra-fast, they can interrupt the current inductance Ldc limits the di/dt. The overvoltage at
without waiting for it to cross zero. the ends of T is limited by the surge arrester to
These circuit-breakers normally use components the value of discharge voltage Us with Us=U+U.
like GTO, IGBT or MOSFET. These components can Supposing that the fault impedance is negligible,
be connected in series or in parallel in order to one can affirm that at instant t=0 there is:
comply with the rated current and voltage
requirements. Absence of moving mechanical
parts is another advantage of using this
technology. This means that theoretically, the
mechanical life of the circuit-breaker is infinite.
thus:
However, one of the disadvantages is that faults
can occur in the components owing to the high
speed rate of rise of overvoltages dU/dt during
By integrating one obtains
opening operations and of overcurrents di/dt
during closing. This is why a disconnector must where I0=i(0) is the fault current. Interruption
be installed in series so as to ensure galvanic time ta is the time at which the current
isolation. extinguishes, thus:
The energy absorbed by the surge arrester is:
Varistor

semiconductor switch Given that is much smaller than U, it means
that Ws is normally larger than the energy stored
in inductance Ldc alone, which is
Another disadvantage is due to power dissipation and that the difference increases as
owing to internal resistance. Besides the cost of voltage U increases. Thus, the higher the rated
losses, the heat produced can damage the voltage, the larger the surge arrester must be.
component, thus suitable dissipation methods From another viewpoint, once the application has
must be adopted. been chosen and ta and I0 have been entered, the
One must consider that the losses in a only parameter that can be modified is
semiconductor are both those produced during inductance, which changes the time constant of
the conduction stage Won and those produced the current and relates to the discharge voltage
during switching Wc(on) and Wc(off). The of the surge arrester. A shorter interruption
instantaneous losses during switching are higher reduces power dissipation in the surge arrester
than those of normal conduction, but only last for but requires that this latter be sized for a higher
time tc(on) and tc(off) (figure 66). Thus these latter voltage. On the other hand, an increase in the
only become important if the switching level of protection of the surge arrester involves a
frequency is high, as in the case of certain types higher rated voltage for the circuit-breaker,
Figure 66: power
dissipation according of static converters. However, these events are thereby increasing the cost.
to conduction and
switching time
not normal for circuit-breakers.
4.1.7 Interruption with hybrid technology In this case, a mechanical main breaker is
This method is based on the integration of a installed in parallel with the static circuit-
mechanical circuit-breaker with electronic breakers. Arc voltage Ua created when the
devices, as illustrated in figure 67. The main contacts separate must not (as in the direct
components of a circuit-breaker based on this current suppression breaker) return grid voltage
principle are: mechanical circuit-breaker, static U to zero, but merely create a voltage capable of
circuit-breaker in parallel with the former and switching the current in the branch with the
isolating disconnector in series with the previous static breaker switch. Once the arc has quenched
two. The basic solution is illustrated in the figure and the contacts are completely open, the current
below: is returned to zero by the static circuit-breaker.
In a second variant, the mechanical circuit-

breaker is equipped with a static device in series
with the first (figure 68). As in the previous case,
IGBT IGBT
the parallel branch consists of a series of static
circuit-breakers. Typically, the static device in
series in the main branch is an IGBT. The purpose
is just to speed up switching towards the
secondary branch in parallel by rapidly
introducing a high voltage. The number of IGBTs
required for this operation is therefore modest.
Consequently, the losses and voltage drops that
occur in the main branch during the entire
conduction period are low.
Figure 67: interruption

circuit with hybrid
Operation is similar to that of the previous
technology
circuit-breaker. During normal operation with the
disconnector closed, current flows through the
Surge Arrestor Surge Arrestor
main branch towards the low-loss static device
and mechanical disconnector itself. As soon as a
fault occurs in the DC circuit, the auxiliary circuit-
IGBT IGBT breaker with IGBT acts by rapidly switching the
current in the secondary branch and allowing the
Small Size IGBT high-speed disconnector to open practically
Isolator Breaker without current.
At this point, the current is interrupted by the
main static circuit-breaker. This configuration is
Ultra Fast used by ABB for its DC hybrid circuit-breaker.
Disconnector
A variant to the two hybrid circuit-breakers
described above consists in creating a switching
circuit that introduces a current pulse able to
Figure 68: diagram

return the fault current to zero and allow the high-
of hybrid circuit- speed disconnector to open (figure 69).
breaker with static
and mechanical
device in series After this, the current is switched to the limiting
resistor and the capacitor is recharged. Negative
voltage is thus applied to the thyristors, which
are then able to interrupt the current.
The trip times of this second solution are
evidently slightly longer than those of the first
two variants described.
Rlim
Ultra Fast Disconnector
Figure 69: diagram of a

hybrid circuit-breaker
with current pulse
4.2 State-of-the-art of medium voltage DC circuit-breakers

4.2.1 ABB medium voltage air-insulated circuit-breakers for railway applications
ABB's DCBreak DC circuit-breaker is extremely light and compact, thus suitable for use on
trains (or rolling stock), tramways, subways and urban and regional light railways (figure 70).
These circuit-breakers are also based on the direct current-suppression principle described in
section 4.1.1.
Figure 70: ABB DCBreak

circuit-breakers for
applications on trains
In these applications, the weight and volume of the circuit-breaker are obviously fundamental
requirements for their use. Maintenance is also an important issue and particular care has been
taken to design these breakers so that the components most subject to wear, such as the arc
chute and contacts, can be easily replaced without having to disassemble the entire circuit-
breaker, the body of which is securely fixed to the structure of the locomotive. These circuit-
breakers conform to standard IEC 60077-3 Railway applications Electric equipment for rolling
stock Part 3: Electrotechnical components Rules for d.c. circuit-breakers.
Section 4.1.1 describes how the interrupted current value also depends on time constant ,
thus on the parameters of the circuit in which the circuit-breaker operates.
Rated operational voltage Ue [V] 900 1800 3600

Time constant T1 (minimum) (ms) 0 0 0
Time constant T2 (ms) 15 15 15
Table 2: nominal
time constants
The DC making and breaking capacities must be tested at the short-circuit time constant
values given in table 2, in accordance with standard IEC 60850 Railway applications Supply
voltages of traction systems.
Electrification Lowest non- Lowest permanent Nominal voltage Highest permanent Highest non-
system permanent voltage voltage voltage permanent voltage
Umin2 Umin1 Un Umax1 Umax2
V V V V V
500 500 750 900 1000

Table 3: standardized DC (mean values) 1000 1000 1500 1800 1950
voltages for DC 2000 2000 3000 3600 3900
railway applications
The rating for this circuit-breaker ranges from 900 to 1800 V DC (table 3), rated current values up to
1500 A and 30 kA breaking capacity.
4.2.2 ABB hybrid circuit-breakers The ABB hybrid circuit-breaker was designed for
The operating principle of this circuit-breaker is 9.0 kA DC breaking capacity, 320 kV DC voltage
described in section 4.1.7. This particular circuit- and 2,000 A rated current. The breaking capacity
breaker was designed for protecting HVDC lines. depends solely on the size of the main static
Illustrated in figure 71, the circuit-breaker circuit-breaker. Ultra-fast disconnector UFD was
comprises a load commutation switch (or LCS) in designed for voltage exceeding 1.5 p.u.
series with an ultra-fast disconnector (or UFD). considering the voltage transients during
The main circuit-breaker with semi-conductors switching.
consists of several sections equipped with surge Each HVDC compartment (figure 72) contains
arresters, sized for the maximum voltage and for four layers of semiconductors and is equipped
the full breaking capacity envisaged. Vice versa, with surge arresters so as to limit the maximum
the LCS is sized for lower voltages and energy. voltage on the compartment during interruption
to a specific value.
Hybrid DC breaker
Ultra Fast Load Commutation

Current limiting Disconnector Switch
Reactor
Figure 71: schematic Residual Current

diagram of the ABB Disconnecting
hybrid circuit-breaker Circuit Breaker
Main Breaker
Breaking speed is of fundamental importance in Each layer consists of 20 IGBT connected in

these applications since faults must typically be series. Owing to the high di/dt, accurate
extinguished within 5 ms. The first reason for this mechanical design engineering was required in
is to prevent disturbance from being created in order to obtain a very low inductance. This was
the conversion stations, the second is to reduce achieved by using compact IGBT layers with 4.5
stress and, thus, the size of the actual circuit- kV rated voltage. The IGBT gates are optically
breaker so as to also cut costs. Long breaking controlled to make interruption independent of
times require a higher breaking capacity, while disturbance from the grid.
the associated surge arrester would have to be Cooling systems are not required since the rated
oversized to an even greater extent. current does not pass through the semiconductor
compartments of the main circuit-breaker.
It is sufficient to consider the rated voltage
requirement when sizing the LCS. Regarding the
rated current, other modules may have to be
connected in parallel while modules in series
would increase the reliability. A 3x3 matrix for
each current direction was used for this project. A
cooling system was required since the LSC
continuously carries rated current, even though
the losses are modest (tens of kW).
Position of IGBT
breaker Figure 73 illustrates a break at 9 kA DC, which is
the limit with the present generation of semi-
conductors. A 250 s delay time will be noted
(due to opening of the LSC) and 2 ms delay due to
opening of the UFD. As will be seen, the total
break time 5 ms.
HDVC circuit-breaker compartment
Figure 72: HVDC

compartment of ABB
circuit-breaker
Opening
of main
circuit-breaker
Current in main Voltage through

circuit-breaker circuit-breaker
Opening of
load
switch
Time (2 ms/div)
Start of
2 ms delay for
fault current 250 s delay for opening of UFD
Figure 73: oscillogram opening of LCS
of a break at 9 kA DC
4.2.3 Future developments Research into the most suitable semiconductors

Developments are possible in all the technologies is an on-going commitment for both these and
involved in the production of DC circuit-breakers. for purely static circuit-breakers. For example,
Further improvements can be made to the arc ABB is now researching into the use of BiGT (Bi-
chutes and opening speeds of mechanical circuit- mode Insulated Gate Transistors), which include
breakers so as to increase the DC breaking the functionality of back-biased diodes in the
capacities. IGBTs, something that could double the breaking
capacity and raise it to 16 kA DC.
Attempts to improve circuit-breakers with Other SiC-based (Silicon Carbide) or GaN-based
passive and active oscillating circuits will (Gallium Nitride) semiconductors are now being
concentrate on optimizing the components of the produced.
resonant circuit, such as inductances, capacitors Here again, purpose-made dynamic models will
and varistors. The attainable goal is to improve have to be developed for these high voltage and
the dimensions, break time and costs. current circuit-breakers so as to simulate their
Further research into the characteristics of the behavior when breaking.
arc and its shape, both in SF6 and vacuum, can
lead to benefits such as increased oscillation and Lastly, further developments could be obtained
breaking capacity. from the application of technologies able to limit
the fault currents.
When it comes to hybrid circuit-breakers, the
high-speed breaker is certainly one of the critical
issues to be faced in relation to further
developments.
5. Regulatory framework
The absence of specific legislation governing systems, installations on ships and low voltage
direct current is one of the critical factors for the systems.
development of new switchgear and controlgear, Table 4 contains a list of the standards in force at
especially in MV and HV. the present time that deal specifically or at least
To date, the majority of the regulations governing partly with direct current:
direct current concern medium voltage railway
Body Standard Description Purpose
IEC IEC 60850 Railway applications Supply - Specifies the main characteristics of the supply voltages of traction
voltages of traction systems systems, as fixed installations for traction, including the auxiliary devices
supplied by the contact line and the rolling stock, to use in:
railway lines;
guided public transport systems such as trams, light rail, elevated and
underground railways and trolley systems;
systems for transporting materials via rail, e.g. coal or iron ore.
- This standard is also applicable to low speed MagLev trains or transport
systems driven by linear motors.
IEC IEC 60077-3 Railway applications Electric - In addition to the general requirements of IEC 60077-2, this provides
equipment for rolling stock Part regulations for circuit-breakers, the main contacts of which must be
3: Electrotechnical components connected to the DC supply and/or to the auxiliary circuits.
Rules for D.C. circuit-breakers - In accordance with IEC 60850, the rated voltage of these circuits does not
exceed 3000 V DC.
IEC IEC 61992-1 Railway applications - Fixed - The IEC 61992 series specifies the requirements for the DC of electrical
installations - DC switchgear - Part switchgear and controlgear and is intended for use in fixed electrical
1: General systems with rated voltage up to 3000 V DC, which supply power to guided
public transport vehicles, i.e. railway vehicles, vehicles for tramways,
subways and trolleybuses. The general requirements are given in Part 1.
IEC IEC 61992-3 Railway applications - Fixed - Contains the requirements for the DC of disconnectors, switch-
installations - DC switchgear - Part disconnectors and earthing switches used in the fixed indoor installations
3: Indoor D.C. disconnectors, of traction systems.
switch-disconnectors and earthing
switches
IEC IEC 61660-1 Short-circuit currents in D.C. - Describes the method for calculating DC short-circuit currents in the
auxiliary installations in power auxiliary systems of power stations and substations, which can be equipped
plants and substations - Part 1: with the following apparatuses, acting as sources of short-circuit current:
Calculation of short-circuit three-phase AC rectifiers with bridge connection for 50 Hz;
currents fixed lead batteries;
voltage balancing capacitors;
direct current motors with independent energizing
- Provides a generally applicable calculation method, which produces
sufficiently accurate conservative results.
IEC IEC 61975 High-voltage direct current (HVDC) - The tests described in this standard are based on bidirectional and two-pole
installations - System tests high-voltage direct current installations (HVDC) comprising a transmission
terminal and a receiving terminal, each connected to an AC system.
- This standard only serves as a guide for the system tests of high-voltage
direct current installations (HVDC).
- The standard provides potential users with information about how to plan
the putting into service activities.
IEC IEC TS Power installations exceeding 1 kV - Provides, in an appropriate form, common regulations governing the design
61936-2 A.C. and 1.5 kV D.C. - Part 2: D.C. and installation of electrical systems in installations with rated voltage
values over 1.5 kV DC for the purpose of ensuring safety and correct
operation for the required use.
IEC IEC 60204-11 Safety of machinery - Electrical - Applies to the equipment and to the electrical and electronic systems of
equipment of machines - Part 11: machines, including groups of machines that operate together in a
Requirements for HV equipment coordinated way, excluding the aspects of higher-level systems (i.e.
for voltages above 1 000 V A.C. or 1 communication between systems).
500 V D.C. and not exceeding 36 kV
IEC IEC 60364-1 Low-voltage electrical installations - Defines the regulations for designing, assembling and checking electrical
Part 1: Fundamental principles, installations. These regulations intend to protect the safety of persons,
assessment of general animals and property against the dangers and damage that could occur
characteristics, definitions during the proper use of electrical installations and to ensure that such
installations function correctly.
- IEC 60364-1 covers circuits supplied at rated voltage up to 1000 V AC or
1500 V DC.
IEC IEC 60947-2 Low-voltage switchgear and - Applies to circuit-breakers with main contacts designed to be connected to
controlgear Part 2: Circuit- circuits with rated voltage up to 1000 V AC or 1500 V DC. Also contains the
breakers additional requirements for integrally fused circuit-breakers.
IEC SG4 LVDC distribution system up to This deals with:

1500V - Coordinating the standardization of different areas, e.g. data centers, office
blocks and shopping centers, etc.
- Energy efficiency, EMC, reduction of natural resources
- 100% DC installations or with AC and DC hybrid architecture
- life-cycle of protection and earthing equipment
IEEE PC37.20.10/ Approved Draft Standard for - The terms and definitions in the standard are intended to encompass
D6 Definitions for AC (52 kV and products within the scope of AC (38 kV and below for air-insulated
Below) and DC (3.2 kV and Below) equipment, 52 kV and below for gas-insulated equipment) and DC (3.2 kV
Switchgear Assemblies and below) power switchgear assemblies, including components for
switching, interrupting, metering, protection and regulating purposes as
used primarily in connection with generation, transmission, distribution and
conversion of electric power.
IEEE DC@Home DC powered house - Standards and roadmaps for LVDC Microgrid application in residential
houses.
- The aim is to:
Create un business case for DC by determining the effective losses and
their value
Identify the research work required to advance the state-of-the-art
Establish the preliminary recommendations concerning the way in which
DC would be delivered to houses
- Written for the AC system, but some of its contents could be used as a
reference for establishing the standards governing DC systems.
IEEE IEEE 1547 Requirements for interconnecting Contains:

distributed resources with electric - operation in the islanded mode and with connection to the grid
power systems - normal and non-normal operation
- requirements and practices for distributed sources
IEEE IEEE 1709 Recommended Practice for 1 kV to Contains guidelines to specify, procure, design, manufacture and develop
35 kV Medium-Voltage DC Power manuals, safety procedures, practices and procedures for effective
System on Ships maintenance of medium voltage direct current (MVDC) electrical power
systems.
- Recommendations are made for analytical methods, preferred
interconnection interfaces and performance characteristics for reliable
integration of MVDC electrical components into ship MVDC electrical power
systems.
- This guide contains indications about planning and designing DC
connections which terminate at points of connection to AC systems, with
low short-circuit values in the direct current supply.
IEEE IEEE 1204 Guide for Planning DC Links - This guide is limited to the aspects of interactions between AC and DC
Terminating at AC Locations systems that result from the fact that the AC system is "weak" compared to
Having Low Short-Circuit the power of the DC link (i.e. the AC system appears as a high impedance at
Capacities the AC / DC interface bus).
- The guide contains two parts: Part I, AC / DC Interaction Phenomena,
classifies the strength of the AC / DC system, provides information about
interactions between AC and DC systems and their mitigation on economics
and overall system performance, and discusses the studies that need to be
performed.
IEEE IEEE Std Trial-Use Recommended Practice Deals with the earthing of DC equipment enclosures installed in DC traction
1653.6 for Grounding of DC Equipment power distribution facilities as well as the related insulation treatments
Enclosures in Traction Power required for sound and resistant earthing methods.
Distribution Facilities - Guidelines are also given for the material, installation and testing of
insulation used in DC traction facilities and further recommended criteria
for acceptability are provided. Even though related, the earthing system is
not covered in this document.
IEEE IEEE 1227 Guide for the Measurement of DC - The purpose of this document is to provide guidance for the measurement
Electric-Field Strength and Ion of electric field strength, ion-current density, conductivity, monopolar
Related Quantities space-charge density and net-space charge density in the vicinity of high
voltage DC (HVDC) power lines in converter substations and in apparatus
designed to simulate the HV/DC power line environment.
- The document defines the terms used, describes the interrelationship
between electrical parameters, describes operating principles of measuring
instruments, suggests methods of calibration where applicable, describes
measurement procedures and identifies significant sources of measurement
error.
IEEE IEEE 946 Recommended Practice for the Revision of IEEE Std. 946-1992. Guidance for the design of DC auxiliary
Design of DC Auxiliary Power systems for nuclear and non-nuclear power generating stations is provided by
Systems for Generating Systems this recommended practice.
- The components of the DC auxiliary power system addressed by this
recommended practice include lead-acid storage batteries, static battery
chargers and distribution equipment.
- Guidance for selecting the quantity and types of equipment, the equipment
ratings, interconnections, instrumentation, control and protection is also
provided.
IEEE C37.14 Standard for DC (3200 V and - This standard covers enclosed low-voltage DC power circuit-breakers of the
below) Power Circuit Breakers stationary or draw-out type of one- or two-pole construction with one or
Used in Enclosures more rated maximum voltages of 300 V, 325 V, 600 V, 800 V, 1000 V, 1200 V,
1600 V or 3200 V for applications in DC systems having rated voltages of
250 V, 275 V, 500 V, 750 V, 850 V, 1000 V, 1500 V or 3000 V; high-speed
circuit-breakers and for rectifiers; manually or power-operated; with or
without electromechanical or electronic trip devices.
- It also deals with service conditions, ratings, functional components,
temperature limitations and classification of insulating materials, dielectric
withstand voltage requirements, test procedures and applications.
IEEE C37.16 Standard for Preferred Ratings, - This standard defines the preferred ratings for low-voltage AC (635 V and
Related Requirements, and below) power circuit-breakers, general purpose DC (325 V and below) power
Application Recommendations for circuit-breakers, heavy-duty low-voltage DC (3200 V and below) power
Low-Voltage AC (635 V and below) circuit-breakers and fused (integrally or non-integrally) low voltage AC
and DC (3200 V and below) Power (600 V and below) power circuit-breakers.
Circuit Breakers
NEC Article 393, Legal codes including introduction - Presents DC technology

625, 690, 692 of DC technology
MIL STD-1399 Electrical interface characteristics - Includes sections that define the requirements of DC equipment for
for shipboard equipment shipboard supply systems
ETSI EN 300 132- Power supply interface at the input - About data / telecommunications equipment for voltage levels up to 400 V
3-1 to data/telecom equipment - Considers the voltage level during normal operation and the requirements
for various types of non-normal operation, the fault current limits, earthing
and EMC.
Emerge DC Microgrid Standards for occupied spaces and - Describes the architecture and control systems recommended in DC
Alliance data center Microgrids.
REbus Open Open standard for DC electricity - Defines DC distribution for operation parallel to the existing AC system
standard distribution in homes, commercial - Coordinates renewable energy generation on site, including solar modules
buildings, campuses, and other and small wind turbines
settings - Defines a common 380 V DC bus with acceptable variation depending on the
state of power supply, load and storage.
The White Set of definitions and tools to - The Green Grid association is a nonprofit industry consortium of end users,
Green papers, determine and compare policy makers, information and telecommunications technology (ICT)
Grid calculation operational efficiency in data providers, facility architects and utility companies. Its purpose is to improve
tools and centers. the efficiency of IT resources, including use of DC distribution.
industry
glossary
Table 4: list of standards

governing direct current
6. Future prospects
An increase in the use and installation of DC Other possible uses of DC can contribute towards
microgrids with low or medium voltage increasing the potential DC distribution market,
distribution can be forecast for the future. MVDC thereby making dedicated investments profitable
distribution will certainly become a valid option, for the Public Utility Companies. For example, and
especially in industry, where it can be used to as partly described in the sections on
supply processes that already use DC for applications, some of these uses are:
production purposes. This will necessarily lead to Long-distance suburban lines to connect rural
new types of architecture and protection systems areas or remote locations where power is
being researched, for the purpose of generated from renewable sources
guaranteeing flexibility and continuity of service. Urban cable circuits
When it comes to the power distribution To increase the capacity of the lines without
managed by Public Utility Companies at voltage increasing the fault level
below 36 kV, use of direct current must compete Decoupling areas affected by disturbance or
with the tried-and-tested, mature and with different power factors
economical AC system. Decoupling areas with critical voltage variations
Nevertheless, the constant increase in loads and or power flows
power generation that use DC natively plus the Benefits that Public Utility Companies could
attention of public opinion and the authorities obtain from improved use of the existing
towards reducing consumption, will certainly cast structures, allowing them to postpone or even
doubts on the convenience of continuing to avoid making important investments, e.g. for
perform double DC-AC-DC conversions and doubling the capacity in certain areas. Since use
encourage the assessment of new solutions by of static inverters intrinsically allows fault
distributing DC directly to the users. current to be limited, the protection systems on
The DC distribution market is by no means united the AC side could remain unchanged. Benefits
at the present time. This is demonstrated by the deriving from decoupling critical areas could
various consortia established to promote the use also improve flexibility, and lead to economic
of direct current in office blocks and shopping benefits.
centers, in homes, data centers or for If a 20kV DC connection line with two back-to-
telecommunications. For the moment, the Public back converters is added to the example in figure
Utility Companies have mainly invested in HVDC 74, the transmission capacity between two 11 kV
transmission lines for interconnecting microgrids AC substations could be increased by dynamically
or remote parts of the power grid, profiting from managing the active power flow, balancing the
advantages deriving from the use of direct loads and power factors.
current or because they are forced to by the need
to decouple grids at different frequencies (e.g. This solution would prevent the protections of
Japan). the two primary substations from having to be
modified and would keep the disturbances and
power factors confined within their respective
areas.
Development of real MVDC distribution grids,
which is already an effective requirement in
electric ships, poses the problem of protecting
Primary cabin A
Primary Primary cabin B
Primary the grids themselves. A task that cannot be
33kV/11kV entrusted to the fault current limitation provided
at 20kV
12MVA at 20kV DC
DC by static converters since this would not
guarantee continuity of service for users not
affected by the fault. Medium voltage direct
current circuit-breakers must be used to
eliminate the faulty circuits, as is already in use in
AC systems.
In conclusion, the circuit-breaker continues to be
the fundamental factor for future MVDC
distribution grid design.
Figure 74: connection
between two primary
MVDC substations
Bibliography
[1] M. Callavik, A. Blomberg, J. Hfner, B. Jacobson, The [18] G. Bathurst, G. Hwang, L. Tejwani, MVDC The New
Hybrid HVDC Breaker, An innovation breakthrough Technology for Distribution Networks, AC and DC
enabling reliable HVDC grids, ABB Grid Systems, Power Transmission, 11th IET International Conference
Technical Paper Nov2012 on, 10-12 Feb. 2015
[2] Christian M. Franck, HVDC Circuit Breakers: A Review [19] P. Wang, L. Goel, X. Liu, and F. H. Choo, Harmonizing AC
Identifying Future Research Needs, IEEE Transactions and DC, IEEE Power & Energy Magazine, May/June,
on Power Delivery (Volume: 26, Issue: 2, April 2011), 2013 10.1109/MPE.2013.2245587
pages 998 - 1007 [20] R. K. Chauhan, B. S. Rajpurohit, N. M. Pindoriya, DC
[3] M. M. Walter AND C. M. Franck, Flexible Pulsed DC- Power Distribution System for Rural Applications, 8th
Source for Investigations of HVDC Circuit Breaker Arc National Conference on Indian Energy Sector, Sinergy
Resistance, XVIII International Conference on Gas with Energy, October 11-12, 2012
Discharges and Their Applications (GD 2010), [21] Osservatorio europeo LEADER AEIDL Fonti energetiche
Greifswald, Germany, 5 - 10 September 2010 rinnovabili, Fonti di sviluppo sostenibile, Chausse St-
[4] W. Xiang, Y. Hua, J. Wen, M. Yao, N. Li, Research on fast Pierre, 260 B-1040 Bruxelles
solid state DC breaker based on a natural current zero- [22] N. Mohan, T. M. Undeland, W. P. Robbins, Power
crossing point, J. Mod. Power Syst. Clean Energy (2014) Electronics, Chapter 2, Overview of Power
2: 30. Semiconductor Switches, pages: 16-32, Ed. John Wiley &
[5] D. Andersson and A. Henriksson, Passive and Active DC Sons, Inc.
Breakers in the Three Gorges-Changzhou HVDC [23] J. Chalfant, C. Chryssostomidis and M. Angle, Study of
Project, ABB Proc. Int. Conf. Power Systems, 2001 Parallel AC and DC Electrical Distribution in the All-
[6] R.P.P. Smeets, V. Kertesz, Application of a Validated AC Electric Ship, Presented at Grand Challenges in
Black-Box Arc Model to DC Current Interruption, Modeling and Simulation (GCMS10), Ottawa, Canada,
Electric Power Equipment - Switching Technology July 2010
(ICEPE-ST), 2013 2nd International Conference on [24] Angelo DAvico, Master's Degree Thesis, Evoluzione
[7] J. Magnusson, On the design of hybrid DC-breakers Degli Impianti Elettrici Navali Per Le Navi Da Diporto,
consisting of a mechanical switch and semiconductor Politecnico Di Milano, Faculty of Industrial Processes
devices KTH Kungl Tekniska hgskolan Stockholm Engineering, Master's Degree Program in Electrical
SWEDEN, Licenciate Thesis TRITA-EE 2015:011, ISSN Engineering, Department of Electrical Engineering,
1653-5146, ISBN 978-91-7595-481-3 Academic year 2010-2011
[8] B. Pauli, G. Mauthe, E. RUOSS, G. Ecklin, J. Porter, and J. [25] War of Currents, Wikipedia, http://en.wikipedia.org/
Vithayathil, Development of a High Current HVDC wiki/War_of_Currents
Circuit Breaker with Fast Fault Clearing Capability, [26] P. Savage, DC Power Distribution Systems, Next
Transactions on Power Delivery, Vol. 3, No. 4, October Generation Net Zero Energy Commercial Buildings,
1988 CATEE 2012, Galveston Texas
[9] A. Mokhberdoran, A. Carvalho, H. Leite, N. Silva, A [27] E.C.W. de Jong, P.T.M. Vaessen, DC power distribution
Review on HVDC Circuit Breakers, 3rd Renewable for server farms, Leonardo Energy, Briefing Paper,
Power Generation Conference (RPG 2014) Naples, 2014 September 2007
page 5.1.2 [28] National Institute of Health, Division of technical
[10] M. Bonkarev, Concept Analysis for High-Voltage Direct- resources, Office of research facilities, DC Electrical
Current Circuit Breakers for Application in a Network of Power Distribution, Technical News Bulletin, April 2013,
HVDC Transmission, Von der Fakultt fr Issue 15
Maschinenbau, Elektrotechnik und [29] M. Callavik, A. Blomberg, J. Hfner, B. Jacobson, ABBs
Wirtschaftsingenieurwesen der Brandenburgischen hybrid HVDC breaker, an innovation breakthrough
Technischen Universitt Cottbus - Senftenberg zur enabling reliable HVDC grids, ABB Review 2/13
Erlangung des akademischen Grades eines Doktor- [30] J. Tucker, D. Martin, R. Mersenski, A. Barkley, P. Cairoli, U.
Ingenieurs genehmigte Dissertation, 14. April 2015 Ghisla, A. Riccobono, R. Dougal, E. Santi, Fault
[11] E. . Norum, Design and Operation Principles of DC Protection and Ride-Through Scheme for MVDC Power
Circuit Breakers Master of Energy and Environmental Distribution Systems Utilizing a Supervisory Controller,
Engineering, Submission date: January 2016, IEEE Electric Ship Technologies Symposium 2011 (ESTS
Norwegian University of Science and Technology, 2011), pp. 319 325, Alexandria, Virginia, April 10-13,
Department of Electric Power Engineering 2011
[12] J. Mooney, B. K. JohnsonHVDC Transmission and [31] A. Riccobono, E. Santi, Stability Analysis of an All-
Integration into an AC Grid PAC: Protection, Electric Ship MVDC Power Distribution System Using a
Automation and Control magazine, September 2016, Novel Passivity-Based Stability Criterion, Electric Ship
pages 28-33 Technologies Symposium (ESTS), 2013 IEEE
[13] F. Iliceto, Impianti elettrici, Vol.1, Cap. 2, pagg. 19-62, [32] M. J. Korytowski, Comparative Analysis of Medium
Ed. Ptron, 1984 Voltage Dc and Ac Network Infrastructure Models,
[14] ABB, Quaderni di applicazione tecnica N.5, Interruttori Bachelor of Science in Electrical Engineering, University
ABB per applicazioni in corrente continua, of Pittsburgh, 2011
1SDC007104G0902 - 07/2011 - 3.000 [33] S. Kulkarni and S. Santoso, Interrupting Short-Circuit
[15] Jin, Z., Sulligoi, G., Cuzner, R., Meng, L., Quintero, J. C. V., Direct Current Using an AC Circuit Breaker in Series with
& Guerrero, J. M. (2016), Next-Generation Shipboard a Reactor, Hindawi Publishing Corporation, Advances
DC Power System: Introduction Smart Grid and dc in Power Electronics, Volume 2012, Article ID 805958,
Microgrid Technologies into Maritime Electrical 14 pages
Networks, IEEE Electrification Magazine Volume 4, [34] M. N. Ravindra, System Protection for Power Electronic
Issue 2, June 2016, Pages: 4557 Building Block Based DC Distribution Systems, A
[16] Dragicevic, T., Lu, X., Quintero, J. C. V., & Guerrero, J. M. Dissertation submitted to the Graduate Faculty of
(2016). DC MicrogridsPart II: A Review of Power North Carolina State University, Raleigh, November
Architectures, Applications, and Standardization Issues. 2004
IEEE Transactions on Power Electronics, (Volume: 31, [35] M. Kempkes, I. Roth, M. Gaudreau, Solid-State Circuit
Issue: 5, May 2016), Pages: 3528 3549 Breakers for Medium Voltage DC Power, 2011 IEEE
[17] C. Phurailatpam, B. S. Rajpurohit, N.M. Pindoriya, Electric Ship Technologies Symposium, Year 2011,
Embracing Microgrids: Applications for Rural and Pages 254 257
Urban India, 10th National Conference on Indian [36] M. Carminati, E. Ragaini, Considerations on DC side
Energy Sector grounding configurations of LVDC microgrids, 2015
5th International Youth Conference on Energy (IYCE),
Year 2015
ABB S.p.A.
ABB SACE Division
Medium Voltage Products
Via Friuli, 4
I-24044 Dalmine
Tel.: +39 035 6952 111
Fax: +39 035 6952 874
e-mail: info.mv@it.abb.com
www.abb.com
1VCP000681 - Rev - en - 2017.07 - (Technical Guide - Medium voltage direct current) (gs)
The data and illustrations are not binding. We reserve the

right to change the contents of this document without
prior notice following technical and product development.
Copyright 2017 ABB. All rights reserved.

Medium Voltage Direct Current Applications

Uploaded by

Copyright:

Available Formats

Medium Voltage Direct Current Applications

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Medium Voltage Direct Current Applications

Uploaded by

Copyright:

Available Formats

M E D I U M VO LTAG E P R O D U C T S

Technical Application Papers No. 24

Figure 3: New York

3. Applications and new scenarios

3.1 Evolution of electrical loads latter case), certain electrochemical processes,

3.3 Static converters the TRANSISTOR MOSFET (Metal Oxide

Diode SCR GTO BJT MOSFET IGBT MCT

Table 1: main characteristics of semiconductors for electric power conversion

voltage High voltage

DC-AC CONVERTERS AC-AC CONVERTERS

AC distribution networks normally include at

Electric vehicles AC loads

A similar structure but with DC distribution is

Fuel Photovoltaic Non-

Electric vehicles AC loads

Future DC industrial Future DC Data Motors

As can be seen by comparing the two diagrams, 3.4 Protection of static

Now let's look at the circuit illustrated in figure 7 below:

Figure 9: currents in the

3.5 DC microgrids 3.5.1 Office blocks and shopping centers

The basic idea is to promote the use of direct

Figure 11: the companies

Lighting and light

Thus the consortium also focuses on creating 3.5.2 Rural applications

Power control system

Electric power supply

Electric pumping unit Drinking trough

Figure 13: Rural electrification,

Figure 14: example

Power generation from

Computers, Resistive loads:

Figure 15: circuit diagram

Figure 16: connection

A DC distribution line involves considerable initial 3.5.3 Electrical ships

To get an indea of the power involved in a cruise

This positionable propulsion system can drive

According to Clarksons Research, one of the

In view of the power involved, the well-defined

Figure 19: modern ship

Figure 20: Azipod

Zone 2 Zone 3 Zone n

Figure 21: circuit

Since the power is generated in alternating

which can function at optimum speed as to

Figure 24: ship with

3.5.4 Data Centers Nowadays, data centers (also known as server

A typical data center is energized by two or more

the various digital systems, perhaps with further

Figure 26: types and

70% power supplier Server rack

Figure 28: efficiency

AC-DC losses In the example in figure 29, which illustrates a

Public grid System board

Figure 32: copper

Rectifier Rectifier Rectifier transformer

Human Machine Interface /

Figure 34: aluminum

The typical layout proposed by ABB as as follows:

Pre-magnetization General Main control

Figure 35: circuit