Chapter 8

Industrial Networks of PLCs

PLCs, which are the leading, stand-alone control tools of industrial processes today, can be inter-
connected with each other to exchange data, information, and control signals. The current form
of automation of most modernized industries is based on integrated control systems implemented
by the communicative interconnection of computers and PLCs. The interconnected systems of
PLCs offer to an extensive industrial process, flexibility, transparency, and effective coordination
of all its parts. The utilization of PLC networks has a direct huge economic benefit, as has been
proven in practice in terms of quality improvement, reduction of production costs, flexibility, and
reliability of industrial production.
It is an undisputable fact that it is impossible for an industrial process to take place in a limited
space or in a single building. Typically, an entire industrial process is implemented in a number
of buildings (production departments) and therefore a need is created for effective coordination
of these various sections of production. Initially, during the first decades of the industrial age,
a lot of time was wasted and labor costs were high mainly due to the need to send “messages”
along all sub-processes in order to achieve better process operation and exploitation. The first step
towards automation is the introduction of central control stations for the production process.
However, the management and control of an industrial process is effective only if information
can be received from every point in the network. This means that for every “bit” of information
sent to the central control station, a corresponding wire is needed. Therefore, for an increased
volume of transmitted information, an increased amount of wires or, equivalently, a significant
cabling task is needed. Hence, it is understandable that in order to install a central station for
monitoring and controlling an industrial process, many, large bundles of wires are needed (peer-
to-peer connections), which leads to a significant increase in the cost of the automation task. At
this time, the idea of using communication networks in the industrial world has matured as a
physical step after the parallel developments in the area of computer networks. With the help of
communication networks, all the bundles of cables are replaced by a single cable, which contains
only two or four conductors instead of the thousands that we had before. In this approach, digital

355
356 ◾ Introduction to Industrial Automation

devices of any kind, participating in the operation of an industrial process, are connected via a
network, and can communicate and exchange data with each other and with a central monitor-
ing station. The data received by a digital device can relate the information required for its own
control decisions, or the automation commands of a coordinator PLC for specific control actions
directly enforceable to the actuators, etc. In conclusion, it can be said that one of the purposes of
using communication networks in an industrial environment is to reduce the enormous task of
wiring. The essential result is a significant reduction in installation, pipeline, and cable channel
costs, as well as maintenance costs of wiring. But the financial benefit is not the only profit that
can be achieved from the application of computer networks in industrial automations. With the
use of communication networks in an industry, we have the ability to receive and process any
data desired in a central control station. This, in addition to many others, leads to faster and more
accurate production planning, as well as faster optimization and tuning of controlled processes
based on reference values received from the central control station. With the use of communica-
tion networks, increased operational reliability is also achieved. In the networked automation
approach, errors are detected at an early stage, while the source of errors can be easily identified
using the existing data of a control process. With the adoption of communication networks,
individual processes can be controled independently and take place in parallel. This has, as a
consequence, the whole industrial process to be controlled in a decentralized manner and at
high speed from a central control station providing coordination, monitoring, and recording of
data. The communication networks developed for the use in an industrial environment are called
“industrial networks”. They interconnect primarily programmable logic controllers, as shown
in Figure 8.1, and secondly personal computers or any other form of digital communication­-
enabled devices, such as a robot controller.

Figure 8.1 Interconnection of various PLCs and remote I/O units through the Profibus network
(Siemens).
 Industrial Networks of PLCs ◾ 357

In the field of industrial networks, a lot of related products have been developed over the years
by European and US companies, and organizations of automation. For only indicating the range
and the variety of the current industrial networks, the names of the most famous industrial net-
works existing in the market today are listed below:

AS-I Hart Interbus Modbus

Profibus CC-Link LonWorks CANbus

Industrial Ethernet Device Net SERCOS EtherCAT

DIN MESSbus P-NET SDS ControlNet

Foundation Fieldbus DH+/DH-485 Profinet EthernetIP

Ethernet Power Link

In addition to those, the networks LIN-bus and FlexRay are used mainly in motorcars, and the
network Instabus is used in building automation. More details on the specific functionalities and
characteristics of these networks can be found on the Internet in the form of online manuals, as well
as from the corresponding manufacturers, thus these networks will not be further analyzed in detail.

8.1 Topology of a Network
The term “topology” refers to the way in which the PLCs of a network are joined together or, in
other words, to the shape of the graph formed by interconnected lines. The topologies of networks
are derived in two basic ways from their node interconnection, point-to-point connection and multi-
point interconnection. The first consists of a circuit that connects two PLCs of the network without
the interference of any intermediate node, while the second consists of a communication line that
is shared between more than two PLCs in the network. Generally, industrial network topologies
have various forms regarding the natural method of their node interconnection. The most prevalent
topologies are the star, the ring, and the bus and these are analyzed in the following subsections.

8.1.1 Star Topology
In networks with a star topology, the connection between several PLCs is made through a central
PLC, as shown in Figure 8.2. The central PLC accepts messages from the PLCs’ senders and for-
wards them to the destination PLCs, except if the messages are intended for the receiving PLC.
The star topology of an industrial network has the advantage that the linking of any two PLCs
requires two communication lines at most. Thus, the time needed from sending until receiving a
message is relatively small. On the other hand, the fact that the host PLC participates in the pro-
cedure of dispatching any sent message, can lead to situations of overcrowding and, consequently,
to delays of the messages’ distribution. The extension of a star network is fairly easy, because the
only nodes involved in this implementation are the new and central PLCs. To achieve the net-
work’s extension, (a) a new communication line should be installed, (b) the two PLCs should be
connected in the line and (c) the network routing tables that exist on the remaining PLCs should
358 ◾ Introduction to Industrial Automation

PLC

PLC PLC

PLC

PLC
PLC

Figure 8.2 Star topology of a PLC network.

be updated for the addition of the new node. The networks with a star topology have relatively
low reliability. Any damage to the central PLC will result in the shutdown of the entire network,
while damage to any other PLC does not have a serious impact on the functioning to the network.

8.1.2 Ring Topology
In networks with a ring topology, all PLCs are connected in a point-to-point configuration, thus
forming a continuous ring, as shown in Figure 8.3. Every message transmitted from a PLC sender
contains the address of the destination PLC and is circulated from node to node in the direction

PLC PLC

PLC PLC

PLC PLC

Figure 8.3 Ring topology of a PLC network.

 Industrial Networks of PLCs ◾ 359

PLC PLC PLC

PLC PLC

Figure 8.4 Bus topology of a PLC network.

of a ring, until it is received from the destination PLC that recognizes its address and is relayed to
the PLC sender. The PLC receiver, before relaying the message, can notice that it has been read,
and therefore when the message is returned to the PLC sender, it understands that the message
has been received correctly. All the PLCs of this network usually have the same rights to send and
receive messages, and therefore the control of the information flow in the network is distributed.
Ring topology networks have a high reliability. In case of a failure of one PLC, the messages are
driven to the adjacent PLC, and thus the network ring remains unbroken. Obviously, if a PLC
is disabled, all other PLCs continue to communicate normally. The extension of a ring topology
network is performed relatively easily, because only the PLCs that are adjacent to the new node
are affected.

8.1.3 Bus Topology
In networks with a bus topology, all PLCs are connected to a common communication line that
is a multi-point or multi-drop physical medium with open ends for transmitting data-messages, as
shown in Figure 8.4. This common line is called a “bus” and is the only channel of communica-
tion between the PLCs of the network. Only one PLC can send a message through the common
communication channel at any time. The message with the address of the destination node is
transmitted along the bus until received by the PLC that recognizes its address.
The reliability of bus-type networks is quite satisfactory. The only case that may cease the
operation of the entire network is when the common communication channel becomes totally
disabled, something that is very rare. Any possible damage to a PLC has no effect on the function-
ing of the other PLCs, except that they cannot communicate with it. In general, the extension of a
bus-topology network is excellent and straightforward by the addition of new PLCs. This addition
is performed very easily at any point of the network, without requiring multiple connections and
without hindering the operation of the network. Due to the above advantages, bus topology is
prevalent and widespread in the area of industrial networks.

8.2 Communication Protocols
The term “protocol” in the field of communication networks generally refers to the rules according
which a node can have access to the network bus. Access rules essentially constitute the control
methods for the information flow in the network, which ensure that there will not be collisions
360 ◾ Introduction to Industrial Automation

in the channel of communication that could lead the network to instability, and that they will
not interrupt or corrupt small or large amounts of transmitted information. The control of nodes’
access may be central or distributed.

◾◾ In the case of centralized control, the access of PLCs to the network is determined by a cen-
tral PLC called a “master PLC”.
◾◾ In the case of distributed control, all PLCs of the network have the same rights to send and
receive messages. Therefore, they don’t have to wait in order to receive rights from a master
PLC.

Access methods of the first category are deterministic, because each PLC communicates
with another in a given time period. Access methods of the second category may be determinis-
tic or stochastic. In the case of stochastic methods, the time a PLC communicates with another
PLC is not definite, and is generally dependent on different access methods that also define the
access times.

8.2.1 Master/Slave Method of Access

According to the Master/Slave protocol, a PLC in a communication network is defined as the
master PLC, and thus all the remaining PLCs become slaves, as shown in Figure 8.5. The PLC
defined as the master station controls the bus traffic entirely. The master PLC with a predefined
order gives the control of the bus to the rest of the PLCs, and thus the ability to communicate to
each slave PLC. The slave PLC acquires the exclusive use of the network for some definite time and
distributes the information that is required by the corresponding application to the other PLC/
PLCs in the network.

8.2.2 Carrier Sense with Collision Detection Method of Access

The full name of this access method is “carrier sense multiple access with collision detection”
(CSMA/CD protocol). The term “carrier sense” derives from the fact that each PLC can sense

Master
PLC

Active
station

PLC PLC PLC

Passive
stations

Slave 1 Slave 2 Slave n

Figure 8.5 A network with one master and multiple slave PLCs.
 Industrial Networks of PLCs ◾ 361

PLC PLC PLC

Collision

PLC PLC PLC PLC

Figure 8.6 Communication collision in a network with carrier sense.

the bus (check its availability for transmitting information), while the term “multiple access” is
defined from the characteristic that all the PLCs in the network have the privilege to access the
bus. Furthermore, the term “collision detection” is defined from the protocol’s ability to detect
communication collisions that may occur, application of methods for avoiding them, and guaran-
tee of the proper operation of the network. According to this access method, when a PLC needs to
transmit a data packet, first it “hears” the bus, which means that it senses whether the line is idle
and therefore available to be used. If the bus is busy, the PLC waits until it is released, otherwise
it immediately transmits the intended information. In the case that two or more PLCs start to
transmit at the same time in the free bus, then a collision will occur, as shown in Figure 8.6. In
this situation, after the detection of the collision from the PLCs in the same network, all the col-
lided PLCs stop their transmission, wait for the elapse of a random period of time (standby), and
after that, repeat the whole process again from the beginning until they successfully achieve the
transmission of their data packets. In this approach, the main disadvantage of this method is that
theoretically there is no upper limit to the time that a PLC should wait in order to send a message,
and thus it is characterized as a stochastic method with the characteristic that this back-off time
can be increased widely until the collapse of the network. This drawback is the main reason that
the method is not used when real-time control is needed, which normally happens in industrial
processes.

8.2.3 Token Passing Method of Access

According to this method, a PLC gets the token, which is equivalent to the authorization of a
data transmission, and transmits the data to the desired PLC. After the end of the transmission,
the token passes to the next PLC, and the previous procedure is followed. The token is a certain
number of bits, also called a “coded packet”, which circulates throughout the network. Depending
on the network’s topology, the token circulation could pass in a ring, a bus approach, or a com-
bination of these access methods. Figure 8.7 shows the application of an access method where
the token passes initially in a bus topology. In this configuration, the token passes only by some
selected PLCs in the network with a predetermined a priori sequence, creating a logical ring in this
way. Each PLC of the logical ring knows which PLC is before and after it. When the logical ring
is created for the first time, the PLC that initially gets the token transmits the data and, after
362 ◾ Introduction to Industrial Automation

Logical ring (token)

PLC PLC PLC

Active
stations

PLC PLC PLC PLC

Passive
stations

Figure 8.7 Token bus network with a logical ring.

the end of the transmission, passes the token to the next PLC in the sequence. Subsequently, the
process is repeated cyclically along the logical ring, while the station that has the token is the only
one that can transmit data on the bus. Finally, it should be noted that in industrial networks, the
hybrid method of multiple master PLCs is commonly encountered, which combines the features
of the token passing method on a bus and master/slave method.

8.3 Implementation of Industrial Networks

8.3.1 Data Transmission Media
In general, all the physical media for the communication of data transfers for classical public
communication networks can also be utilized in the case of the industrial networks. The most
common data transmission media (hardware) are presented next.
Twisted-pair cable. The simplest medium for data transmission in a communication network
is a twisted-pair cable. It consists of two insulated copper conductors twisted around one another.
The twisting of wires is made in order to avoid the electromagnetic interactions between two or
more pairs when they are near. When several wire-pairs go parallel and at long distances, then these
are grouped in bundles with an external protective casing. The wire pairs at these bundles would
create an electromagnetic interference between them if they were not twisted. The signal trans-
mission with twisted-pair cables is quite satisfactory (amplification is not needed) for distances
of a few kilometers. For very large distances between two network nodes, repeaters are required
in order to communicate. Analog and digital signals can both be transmitted via a twisted-pair
cable. In this case, the bandwidth depends on the thickness of the wires and the transmission
distance, while generally some Mbits/sec for distances of a few kilometers can be achieved. In the
twisted-pair case, the main advantages are the low cost and the ease of installation, while the main
disadvantages are their sensitivity to noise and the relatively slow speed of transmission.
Coaxial cable. The coaxial cable consists of a stiff copper wire (core) that is surrounded by
an insulating material. The insulating material is covered by a woven copper braid or a metal-
lic foil that acts both as the second wire in the circuit and as a shield for the inner conductor.
The external conductor (grid) is covered by a protective plastic cover. There are two types of
 Industrial Networks of PLCs ◾ 363

coaxial cables, the 50 Ω cable used for digital transmission (base band) and the 75 Ω cable
used for analog transmission (wide band). The coaxial cable is characterized by a high band-
width and a satisfactory tolerance to noise. The bandwidth varies depending on the length of
the cable. For coaxial cables with a length of 1 km, the data rate can be as high as 10 Mbps.
For smaller or larger distances the data rate is correspondingly higher or lower. Coaxial cables
are used by many companies in commercial industrial networks for the interconnection of
PLCs.
The 75 Ω coaxial cable through which analog signal transmission is performed, is called a
broadband cable, and is used between others in cable television systems for transmitting video,
voice, and data. It is characterized by a high bandwidth and a long transmission distance. In
wideband transmission systems, the total available bandwidth is divided into a series of non-
overlapping frequency channels, usually in 6 MHz channels, used for television broadcast.
Each channel could transmit analog TV signals, audible signals, and digital data indepen-
dently from other channels. Video channels, after being digitized, could be compressed to
better fit the channel width. For transmitting digital signals in an analog network, every
digital connected device should be equipped with electronic circuits able to convert the out-
bound train of bits into an analog signal and the incoming analog signal into a train of bits.
In conclusion, the 75 Ω coaxial cable is suitable for all the types of digital audio, digital video,
and data signals.
Fiber-optic cable. Digital data transmission via an optical fiber can now be considered as a
classic broadband trend. Many manufacturers of PLCs have adopted optical fibers as the transmis-
sion medium in the respective networks on the market. It is known that a light pulse can be used
to represent the binary digit “1”, while the absence of a pulse can represent the binary digit “0”.
To perform an optical transmission, a light source, a transmission medium, and a light detector
are required. The transmission medium is a very thin filament of glass or fused silica. The light
source is either an LED or a laser diode that both emit pulses of light when voltage is applied. The
detector is a photodiode that produces an electrical signal in response to an incident of light. By
connecting an LED or a laser diode at one end of a fiber optic cable and a photodiode at the other
end, a one-way data transmission system can be created, which is able to accept an electrical signal
to convert and transmit light pulses subsequently, and then, at the other end or at the signal exit,
reconvert it into a similar electrical signal as the input one.

8.3.2 The ISO/OSI Model

In the field of communication networks, there has been a complete incompatibility in the
various hardware and software manufacturers, as well as between the communication proto-
cols used in these networks. Thus, at the beginning of the communication network era, it was
impossible to connect two computers or PLCs from different manufacturers on a common net-
work. The International Organization for Standardization (ISO) developed the Open Systems
Interconnection model (OSI) in order to describe the communication process in the field of net-
works. The model has a key feature, that of a hierarchical seven-layer stratification. Each layer is
directly linked with the previous one and has a definite relation with the rest. Each layer takes
care of a very specific job, and then offers its services onto the next higher layer. The interface is
so flexible that each manufacturer has the possibility to create its own communication protocol
consistent with the OSI model. The ISO/OSI model is a reference tool for data communications,
364 ◾ Introduction to Industrial Automation

and its seven layers are ranked from the lowest physical layer, data link, network, transport, and
presentation to the highest application layer. In these layers, all the specifications of the commu-
nication networks are defined, from electrical details of bit transmission, the voltage that would
represent the binary digit “1” and bit “0”, the time length of each bit-pulse, up to the applications
concerning the user. Finally, the ISO/OSI model defines terms that networking professionals can
use to compare basic functional relationships of different networks.

8.3.3 Network Devices
Due to various factors, industrial networks (as well as networks in general) have a limited capa-
bility in terms of both the number of interconnected stations as well as the maximum length
of interconnecting network cables. Therefore, problems are often created when it is necessary
to extend a network of PLCs or to interconnect an existing network with another one. To
overcome these limitations, four network devices are used, namely the repeater, the bridge, the
router, and the gateway. With the help of these devices, it becomes possible to extend a PLC
network, either in terms of the number of connected stations, or in terms of geographic area
and range. Also, with the help of these devices, the communication between heterogeneous
networks of PLCs is possible.
Repeater. The repeater is used only in homogeneous networks, i.e., networks that follow the
same exact communication protocol. The repeater just regenerates and forwards bits from one
sub-network to another, making two sub-networks look like one. Great distances are covered by
concatenating two or more network segments via a repeater. Repeaters do not need software to
operate, since they just copy and regenerate bits blindly, without any further processing on the
repeated data packets.
Bridge. The bridge performs the connection of two networks which have different data link
layers (data frames with different media access control addresses) but the same network layer as
in the OSI model. The data frames from one network segment, reaching the bridge, are analyzed.
After checking and filtering the destination address, they are forwarded to the second network
segment. Bridges, unlike repeaters, use software for their operation. They are programmed for
exchanging certain data between subnets and generally for carrying out the necessary modifi-
cations during their operation. As an industrial example, by using two CAN@net II/Generic
bridges by the HMS/IXXAT Company, a CAN-Ethernet-CAN* bridging can be implemented.
This bridge allows the exchange of CAN messages between two separate CAN networks via the
Internet, where filter tables can be defined.
Router. A router can connect two dissimilar networks. In terms of the OSI model layers, a
router connects two networks that differ in the three lower levels. This means that the router
is used when two networks have the same transfer layer but different network layers of the OSI
model. The router, unlike the bridge, may have its own address in the network. Some routers are
also used as industrial secure devices protecting control networks from critical assets.
Gateway. The gateway is the most general type of device for interconnecting heterogeneous
networks. It has the ability to interlink completely different networks in terms of communication
protocols, even networks that are different in all the seven layers of the OSI model. The gateway
is also used for the interconnection of networks that do not follow the OSI model. Gateways are
generally more complex than bridges or routers. Figure 8.8 shows the use of the above devices

* Controller area network (CAN) is a well-known industrial network.

 Industrial Networks of PLCs ◾ 365

C C
P P P
S U

Industrial ethernet
Gate Gate
C C
P P
P U
S

Repeater AS-I
PROFIBUS
C C
P P
C C C P U
C C C S
P P P P P P
P U P U P U
S S S

Figure 8.8 Three different networks (corresponding to machine, production control, and
supervisory levels) are interconnected through network devices.

to create an integrated interconnection of different or similar networks, called a “multi-network

communication structure”.

8.3.4 The Communication Task of PLCs

The need for data exchange between PLCs and other digital devices or controllers in an automated
industrial process has led all manufacturers to supply their controllers with additional commu-
nication capabilities. The hardware for the communication capability of a PLC appears in two
forms, as is described subsequently. In compact type PLCs, the communication hardware and
software are embedded in the module of their CPU, with the communication port as the only
visible part. In modular type PLCs, the communication hardware is available as separate modules
of communication of various types and features, which have already been mentioned in Section
6.7. Regardless of the form of the hardware, the communication process of a PLC fits together
with the classical automation program inside the scanning cycle of the PLC, as shown in Figure
8.9, where two PLCs communicate over a network. It is obvious that this communication process
refers to the read/write operations of the variables to be transferred over the network and not to the
communication task of the network operation that is performed by the communication processor-
module. The transmitted data over the network are stored in the buffers of the communication
modules according to rules defined by the communication protocol of each network. As shown
in Figure 8.9, the PLC in each scan cycle updates its memory with data that have arrived over
the network, executes the corresponding instructions based on the updated values in the PLC’s
memory, and updates the buffer of the communication module with variable values obtained
from the execution of the automation program that needs to be transmitted through the network.
These steps are combined suitably with the rest of the classical steps of the automation program
execution, including the reading of inputs and writing of outputs in the PLC, which have been
described in detail in Section 6.1.
The addition of the communication task on the rest of the computing work for the scanning
cycle increases the duration of the latter. On the other hand, the communication network has its
own response times, which are not related to the duration of the scanning cycle. For the connected
system of two PLCs of Figure 8.9, it is defined as a response time RT; the time that elapses from the
moment that an input is activated in the PLC1 until the moment that an output will be activated
in the PLC2. Obviously, in the automation logic program of the PLC2, the output status depends
366 ◾ Introduction to Industrial Automation

PLC 1 PLC 2

CPU Communication Communication CPU

module module

Transfer memory Transfer memory

buffer A from buffer A from
communication communication
module to PLC module to PLC
central memory central memory

Read Read
Inputs Inputs

User program User program

A A
Scan cycle

Scan cycle
I/O I/O
Instructions for bus bus Instructions for
variables that variables that
will be send to will be send to
other PLC B B other PLC

Update Update
outputs outputs

Transfer specific Transfer specific

memory words memory words
of PLC to the of PLC to the
memory buffer memory buffer
B of the B of the
communication communication
module module

Physical medium Network line

Figure 8.9 The scan cycle in relation to the communication task of two PLCs.

on the status of the PLC1 input. The response time or the total communication time is calculated
by the equation,

R T = IT + 2ST1 + PT1 + A T + TT + PT2 + 2ST2 + OT

where
IT = Input delay time, which is the time that elapses from the moment that an input contact
closes until the moment that this 1-bit information is available for reading in the PLC1.
ST1 = The scan cycle time of PLC1.
ST2 = The scan cycle time of PLC2.
PT1 = The processing time in the PLC for the preparation of data to be transmitted.
PT2 = The processing time in the PLC2 for receiving arrived data and preparation in order to be
available for the automation program executed in the PLC2.
 Industrial Networks of PLCs ◾ 367

AT = Waiting time of a PLC to access the physical network.

T T = Data transmission time for the given data transfer rate (baud rate) of the network.
OT = Output delay time, the time that elapses from the moment that 1-bit information is avail-
able in the output module until the moment that the corresponding output is energized
in the PLC2.

The numeric values for all the above parameters make sense only in the case that these refer
to a specific network and with a certain PLC. To give a sense of the time correlation between a
PLC and a corresponding network, it is reported that, in many cases, the access cycle time of the
network may be faster than the scanning cycle of a PLC that is connected to the network. For
example, in a Profibus network with the smallest possible speed (selected and programmed based
on the length of the network) and a usual number of nodes, it is possible to have a 2–3 network
operation cycles per scan cycle of a PLC.

8.3.5 The Actuator-Sensor Interface (AS-I) Network

In an industrial process, all actuators and sensors should be connected to the programmable auto-
mation system. The industrial network AS-I that first appeared in 1993 and was reissued in 2000
supports communication at this level, known as machine level. Figure 8.10 shows a typical con-
figuration of the industrial network AS-I. At the machine level, where there is a large number of
sensors and actuators, the installation and use of the AS-I network allows us to avoid the numer-
ous wiring that is replaced with a two-conductor cable for power and data transfer simultaneously,
as shown in Figure 8.11a. The special cable of the AS-I network is depicted in Figure 8.11b, and
is known as the yellow cable, and the applied technique of connection without removing the
insulation make the wiring work simple, while providing greater security in the data transmis-
sion. In the same figure, it can also be observed that nowadays it is possible to interconnect even
simple push buttons, something that was impossible as a non-supported feature a few years ago.
The topologies that can be configured to implement the AS-I network are the bus, star, ring, and

Figure 8.10 Interconnection of PLCs, sensors, and actuators through the AS-I network (Siemens).
368 ◾ Introduction to Industrial Automation

(a) (b)

Figure 8.11 The AS-I network can interconnect sensors and actuators (a), as also simple devices
(buttons, switches, etc.) (b) at the machine level.

tree. Furthermore, the AS-I network has the ability to communicate easily with other industrial
networks like DeviceNet, Profibus, etc.

8.3.6 The Profibus Network

Except for interconnection of sensors and actuators, there is a big need for interconnecting control-
lers of all kinds that exist at the level of control of a production process, such as robot controllers,
industrial PCs, PLCs, CNC machine controllers, programming devices of PLCs, etc. At this level
of a computer integrated industrial environment, strong communication abilities are required,
such as a large volume of data, real-time high speed, a large number of nodes, and a large range.
One of the networks with such characteristics is the Profibus network for a wide range of indus-
trial applications in manufacturing, process, and building automation. A PLC can communicate
with machines from different manufacturers via a Profibus network that is a vendor-independent
fieldbus standard without the need for special adaptive settings. A typical Profibus network con-
figuration is shown in Figure 8.12. In the Profibus network, there is separation of PLCs on master/
slave units. The token passing procedure is used for the communication of master PLCs. The mas-
ter PLCs designate the data communication on the bus, and can transmit messages without the
need for external requests when the PLC has the access rights (token) to the bus. Slave units may
be peripheral devices such as PLCs, I/O devices, servo drives, inverters, and temperature control-
lers. These devices do not have direct access to the bus, except for the cases where it is requested by
the master PLC. The master PLC sends messages to the slaves, who send acknowledgment or send
response messages to the master upon corresponding requests.

8.3.7 SCADA Systems
The establishment and broad utilization of communication networks in the field of industrial
automation brought, along with the economic and operational benefits, the development of the
supervisory control and data acquisition (SCADA) systems. Nowadays, a SCADA system is used
 Industrial Networks of PLCs ◾ 369

Siemens S7-300
Class-1 DP-Master PC(W/Class-2
DP-Master function)

PROFIBUS
i-7550 PROFI-8455

RS-232
RS-485

RS-232

RS-232
temperature
sensor i-8K i-87K
module module
Remote I/O solution

Figure 8.12 The Profibus network.

to monitor and control an entire production system or plant and consists of a series of operations,
the collection of information, the transfer of it to a central station through the network, the car-
rying out of the necessary data analysis, the indication of data in a number of virtual “screens”,
and finally the possible control actions. It is essentially a large software package-tool, which col-
lects all the required information through the network from the controlled system, so that it can
implement a number of operational functions while visualizing it to the user in a friendly and
realistic approach. The most significant feature of SCADA systems is the ability for downloading
and storing data, graph production, the graphical display of the process with simulated motion of
the machines, the announcing of alarm situations, the presentation of real-time values in active
fields, the recording and printing of normal or exceptional events, the statistical processing of
data, the status monitoring of the overall communication network, the communication with the
external databases, and many other features that provide the end user with a direct overview of the
controlled industrial process. It is obvious that the control actions applied by a SCADA system
can be automatic or manual. This means that an operator can start or stop machines through the
SCADA system while Figure 8.13 shows an example of an industrial process depicted in a SCADA
software environment.
370 ◾ Introduction to Industrial Automation

Pump station “Mesopotamos” Full screen WinCC bar Help Screen with motion 11:44:20 AM 12/9/2016
Air compressor tank 1 Air compressor tank 2
Suction tank Home screen
Level Level
Storage tank 0.0 0.0
2 Low voltage
power station
Level
Level Monitor variables
###
0.0
Print daily report

Minimum pressure Maximum flow rate

Pumps operation
Air compressor
Hand Auto Hand Auto Hand Auto Hand Auto Hand Auto Hand Auto Hand Auto Hand Auto
Hand Auto

On On On On On On On On
8 On

Off Off Off Off Off Off Off Off Off

1 2 3 4 5 6 7 8 Open 0.0 A
0.0

Open Open Open Open Open Open Open

Exit
Motorized valve
Hand Auto
0.0 A 0.0 A 0.0 A 0.0 A 0.0 A 0.0 A 0.0 A
On

1 2 3 4 5 6 7 Off

Open
Evacuation pump
Stop
Close

Suction
tank
1 Level
###

Figure 8.13 Virtual screen of a pump station SCADA system.

Review Questions
8.1. What is the purpose and what are the advantages of using communication networks in
industry? Comment upon the existence of many industrial networks in the international
market.
8.2. In addition to PLCs and cable, what others devices can be used in an industrial network and
for what purpose?
8.3. At the lowest level of communication of production machines, the so-called “field level”,
what is the most significant characteristic that the industrial network must present and for
what reason?
8.4. Why do we need industrial networks in an industry? Write down as many technical impacts
of the operation of an industrial manufacturing or process line as possible.
8.5. A new PLC has to be connected on a preexisting industrial network. What hardware and
software are required in order to perform the connection? What kind of programming and
which units are required in order to achieve the operative connection of the new PLC with
the other interconnected PLCs to exchange information and control commands?
8.6. Describe the access method of a PLC in an industrial network called “token passing in bus
topology”.
8.7. In your opinion, which are the two most basic network topologies that are used in industrial
networks? Please describe their advantages and disadvantages.
8.8. What is a “SCADA system”? What does it include, how does it operate, and for what pur-
pose? Which operational functions does it offer the user?
 Industrial Networks of PLCs ◾ 371

8.9. If a system is geographically distributed (like a town water system, a gas or oil pipeline, etc.),
how do you think that its communication integration is achieved?
8.10. In an industrial process, there are 100 smart sensors with embedded microprocessors and
communication ports, and 150 pure analog sensors all of which must be interconnected.
In which manner will you interconnect the 250 sensors through a suitable industrial
network?
8.11. Using Internet databases, find what is called a “computer-integrated manufacturing”
(CIM) model. What kind of networks and technical characteristics are suitable at each
one of the three levels of the CIM model?
8.12. The figure shows a possible configuration of an industrial Ethernet network consisting of
a number of the same kind of subnetworks. Consider any kind of information that can be
derived from the figure and write comments for any topic such as communication archi-
tecture, topology, devices, physical medium, etc.

PLC PLC PLC

PLC Wireless
Copper
cable PLC PLC PLC
PLC

Fiber optic PLC

C C PLC

Star coupler CON

PLC
PLC
PLC PLC PLC CON Concentrator unit