‘

‘
‘
‘
‘
‘

‘‘ ‘

 
‘
‘‘‘‘‘‘‘‘‘‘‘‘‘‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
c

 ‘
‘
‘
‘‘‘‘ ‘ ‘
‘
Power line carrier communication, also known as PLCC, refers to the concept of
transmitting information using the electrical power distribution network as a
communication channel. This technology allows a flow of information through the same
cabling that supplies electrical power. This novel idea of communication helps in
bridging the gap existing between the electrical and communication network. It offers the
prospect of being able to construct intelligent buildings, which contains many devices in
a Local Area Network.

Power line carrier communication uses existing power distribution wires to communicate
data. This, however, is not a new idea. In 1838 the first remote electricity supply
metering appeared and in 1897 the first patent on power line signalling was issued in the
United Kingdom. In the 1920's two patents were issued to the American Telephone and
Telegraph Company in the field of "Carrier Transmission over Power Circuits". One
would think that the long-ago conceived idea of power line communications would be
well developed by now. However, this is not the case because the power line is not well
suited for data communication.

There are two main applications for power line communication - one for broadband
Internet access to the home and the other for home and office networking. This work
focuses on using power lines for home and office networking.

Power line communication technology has been slow to evolve because the lines were
designed solely for the purpose of 50 Hz main power distribution. Unfortunately, power
lines are a rather hostile medium for data transmission, it is difficult to communicate data
effectively because the medium was not designed for data transmission. Attenuation,
variable impedance, and noise are three factors which make this a harsh medium,
making it difficult to achieve optimum signal transfer, low distortion, and high signal-to-
noise ratios (SNR).


 ‘ !‘"‘#‘‘
‘
The project aims towards thoroughly exploring the theoretical and practical aspects of
power line carrier communication (PLCC) techniques. To this end a number of specific
goals were proposed at the start of the project.

1. To gain a detailed knowledge of the challenges faced by PLCC techniques and to

understand why they are not a widespread communication methods.

2. To research and design a working PLCC system, which could be employed to couple
speech signals with AC signals to transfer voice from one end of a building to the other
end using home line AC wiring.


$‘ %!‘ ‘& ‘"‘'(‘‘

The design of a network should consider several factors, of which the two most important
are predicted network traffic and installation cost. The nature of traffic generated by
applications such as email, streaming audio or video, file transfer, control systems,
application or resource sharing, etc. predicts the type of service needed. The various
types of traffic can have different throughput, data integrity, latency, and other
requirements. A simple control system network that performs functions such as turning
lights on and off, opening and closing the garage door, and controlling the air conditioner
does not require high speeds. A high speed network would be much better utilized by a
multiple computer network where there is a large amount of file and application sharing
or video.

The cost factor refers to the installation cost of a network. High speed networks often
require more expensive equipment than low speed networks, so for low speed networks it
is not economically smart to install high speed equipment. Installation cost is also
affected by the actual setup of the network. Wireless equipment is becoming popular
because it is simple to set up and provides high speed and high mobility (computers can
access the network as long as they are within a certain distance of the access point).
However, the wireless equipment may be too costly for low to medium speed
applications. Another solution is to use dedicated network cabling but this is also a high
cost solution because retrofitting a home with the required cabling becomes a time
consuming and expensive job. Also, once network cabling is installed in a home or
office, it does not lend itself easily to reconfiguration ± resulting in down time when
location of network entities changes.

What is missing is a medium speed technology that is low cost and allows for easy and
ubiquitous network access. This project addresses a possible solution to the problem of
mobility, ease of installation, and cost of networks by using the in-building power
distribution system.

Current power line communication technology that can support applications such as
control of devices, network gaming, low resolution image sequences from cameras,
security applications and several other applications.

‘
‘

[
)‘! %!‘ ‘!(‘‘‘

‘
It is difficult to communicate data effectively because the medium was not designed for
data transmission. Attenuation, variable impedance, and noise are three factors which
make this a harsh medium, making it difficult to achieve optimum signal transfer, low
distortion, and high signal-to-noise ratios (SNR).

The most important technical problem in power line PLC is to device methods and
equipment to couple the low frequency and high voltage signal to the high frequency and
low voltage signal. Following problems had to be dealt with

½‘ ÷isturbance created in nearby cast receivers.

½‘ Protection against high voltage.
½‘ rore than one carrier circuit may cause interference.
½‘ Switching may cause undesirable disturbances.

The modern practices are achieved by connecting a capacitor of proper rating between
the carrier terminal and high voltage line

º
 ‘
‘
‘
‘
‘

‘'‘‘

* +‘‘ ‘
‘‘‘‘‘‘‘‘‘‘‘‘‘‘
‘

ü

 ‘
‘
‘‘ ‘‘ ‘,(%‘‘
‘
In order to better understand PLC, the following section provides an overview of a
general communications system. This includes a discussion of the elements of a
communications system, the methods for transmitting data, and performance measures.

‘‘‘ !‘+-!‘&.‘
Figure 2.1 shows a simplified model of a digital communications system. The overall
objective of a communications system is to communicate information (the
transmission of digital information this thesis considers) from a source to a destination
over some channel.

Figure 2.1 Communication system model.

+‘‘! ‘‘The source can be any digital source of information. If the
source is analog such as speech, then an analog to digital converter must precede the
transmitter. At the receiving end, the decoded information is delivered to the
destination.

The source may also compress redundant data, which minimizes the number of bits
transmitted over the channel, but can also create a loss of source information. The data
is unpacked at the destination to either an exact replica of the source information
(lossless data compression) or a distorted version (lossy data compression).

G
 #.‘ ‘ ‘ #.‘  ‘ ‘ Channel coding reduces the bit error
probability by adding redundancy (extra check bits) to the bit sequence. The check bits
are computed over a k-symbol input sequence to create an n-symbol output code
sequence. This determines the code rate Rc where Rc = k/n and Rc ” 1. This is the
ratio of the number of actual data bits to the total number of bits transmitted. The
channel decoder uses the extra bits to detect and possibly correct errors which occurred
during transmission. The number of extra bits added depends on how much error
detection and correction is needed. Channel coding (also known as error control
coding) is a heavily studied area. It is used to improve performance over noisy channels
(such as the power line). Two major classes of codes exist: block codes and
convolutional codes. Block codes are implemented by combinational logic circuits.
Reed-Solomon (RS) codes are a popular block code. Convolutional codes (also known
as tree codes or trellis codes) are implemented by sequential logic circuits.

#.‘&.‘‘ #.‘. ‘ The purpose of the modulator is to

take the encoded data and produce an analog signal suitable to propagate over the
channel. The data is converted from a stream of bits into an analog signal. An r-ary
modulator takes a block of Y binary digits from the channel encoder to select and
transmit one of r analog waveforms at its disposal where r = 2Y and Y • 1. At the
receiver, the demodulator tries to detect which waveform was transmitted, and convert
the analog information back to the sequence of bits. rodulation is typically performed
by varying the amplitude, the phase, or the frequency of a high-frequency carrier signal.
For example, if the input signal of the modulator is used to vary the amplitude of the
carrier signal, the modulation is called Amplitude Shift Keying (ASK). There are
several other modulation techniques including FSK (Frequency Shift Keying), PSK
(Phase Shift Keying) and QAr (Quadrature Amplitude rodulation).

#. ‘ The channel can be any physical transmission medium including coaxial
cable, twisted pair, optical fibre, air, water, or for this work - the power line. It is
important to know the characteristics of the channel, such as the attenuation and noise
level because these parameters directly affect the performance of the communication
system.


‘‘‘!/‘"‘
/‘"‘!

+-.‘‘- This is the transmission rate or number of symbols per second from
the modulator. If the signal duration is T seconds then the symbol rate is 1/T symbols
per second. The symbol rate is also known as the baud rate.

,!‘/‘!‘0/!1‘± Also known as bit rate, bits per second is directly related to
the symbol rate. If each symbol represents Y bits and the symbol rate is 1/T baud,
then the bit rate is Y*(1/T) bps. On high quality channels it is easier to send more bits
with one symbol, resulting in higher bps.

,‘‘.-‘01‘± Pb is the probability that a bit is incorrectly received at

the destination. This is an important performance measure for any digital
communication system that is affected by noise and the disturbances in the channel.

,#‘0,'1‘± The range of frequencies used by the communication system. For a

specific communication method, the bandwidth needed is proportional to the symbol
rate. Bandwidth is a limited resource and is often constrained to a certain small range.

,#‘""-‘2‘This is the ratio between the bit rate and the bandwidth of a
communication system (bps/BW). Today a telephone system can achieve a bit rate
of 56.6 kbps using a bandwidth of 4 kHz, so the bandwidth efficiency is 56.6/4 =
14.15 bps/Hz.

!‘± This is an unwanted signal on the channel that interferes with the desired signal.
Noise on the power line is a sum of many different disturbances originating from
devices such as television receivers, computers, and vacuum cleaners. The amount of
noise can drastically affect the quality of communication.

‘± When the signal is propagating from the transmitter to the receiver the
signal gets attenuated (loses power). If the attenuation is high, the received signal
power can become low and might not be detected. Attenuation is shown to be high on
a power line, and this puts a restriction on the distance from the transmitter to the
receiver.

0
+%.‘‘!‘‘0+1‘± This is the ratio of received power to noise power. A
higher SNR makes for easier communication because noise has a smaller effect on the
signal. SNR is also affected by attenuation, which reduces signal power and thus SNR.
SNR can be increased by using filters to reduce noise outside of the bandwidth
occupied by the signal.

 !-‘ ± Used to reduce the error probability of harsh channels. Examples of

diversity are time diversity and frequency diversity. In time diversity the same
information is transmitted at different time instants with the idea that if the channel is
bad at some time instance it might not be at another. Frequency diversity transmits the
same information
in different frequency bands. It can be compared to having two antennas transmitting at
different frequencies; if one of them fails the other might work. Several variations of
time and frequency diversity exist. This thesis explores a form of time diversity,
although not exactly as described above.

$‘‘&.‘

/. !#"‘ (-% (+3) is a form of modulation that represents digital data as
variations in the amplitude of a carrier wave.The amplitude of an analog
carrier signal varies in accordance with the bit stream (modulating signal),
keeping frequency and phase constant. The level of amplitude can be used to
represent binary logic 0s and 1s. We can think of a carrier signal as an ON or OFF
switch. In the modulated signal, logic 0 is represented by the absence of a carrier, thus
giving OFF/ON keying operation and hence the name given.

Like Ar, ASK is also linear and sensitive to atmospheric noise, distortions, propagation
conditions on different routes in PSTN, etc. Both ASK modulation and demodulation
processes are relatively inexpensive. The ASK technique is also commonly used to
transmit digital data over optical fiber. For LE÷ transmitters, binary 1 is represented by a
short pulse of light and binary 0 by the absence of light. Laser transmitters normally have
a fixed "bias" current that causes the device to emit a low light level. This low level
represents binary 0, while a higher-amplitude light wave represents binary 1.

*4- !#"‘ (-%‘ 0*+31 is a frequency modulation scheme in which digital

information is transmitted through discrete frequency changes of a carrier wave. The

c

simplest FSK is binary FSK(BFSK). BFSK uses a pair of discrete frequencies to transmit
binary (0s and 1s) information. With this scheme, the "1" is called the mark frequency
and the "0" is called the space frequency.

‘ *+3‘ Audio frequency-shift keying (AFSK) is a modulation technique by

which digital data is represented by changes in the frequency (pitch) of an audio tone,
yielding an encoded signal suitable for transmission via radio or telephone. Normally, the
transmitted audio alternates between two tones: one, the "mark", represents a binary one;
the other, the "space", represents a binary zero.

AFSK differs from regular frequency-shift keying in performing the modulation

at baseband frequencies. In radio applications, the AFSK-modulated signal normally is
being used to modulate an RF carrier for transmission.

#! !#"‘(-%‘0+31 is a digital modulation scheme that conveys data by changing,

or modulating, the phase of a reference signal (the carrier wave).

Any digital modulation scheme uses a finite number of distinct signals to represent
digital data. PSK uses a finite number of phases, each assigned a unique pattern of binary
digits. Usually, each phase encodes an equal number of bits. Each pattern of bits forms
the symbol that is represented by the particular phase. The demodulator, which is
designed specifically for the symbol-set used by the modulator, determines the phase of
the received signal and maps it back to the symbol it represents, thus recovering the
original data. This requires the receiver to be able to compare the phase of the received
signal to a reference signal ² such a system is termed coherent (and referred to as
CPSK).

Alternatively, instead of using the bit patterns to set the phase of the wave, it can instead
be used to change it by a specified amount. The demodulator then determines
the changes in the phase of the received signal rather than the phase itself. Since this
scheme depends on the difference between successive phases, it is termed ÷"".‘
/#! !#"‘ (-%‘ 0+31. ÷PSK can be significantly simpler to implement than
ordinary PSK since there is no need for the demodulator to have a copy of the reference
signal to determine the exact phase of the received signal (it is a non-coherent scheme).
In exchange, it produces more erroneous demodulations. The exact requirements of the
particular scenario under consideration determine which scheme is used.

5‘/#! !#"‘ (-%‘05+31‘ Sometimes this is known as quaternary PSK

or 4-PSK. QPSK uses four points on the constellation diagram, equi-spaced around a
circle. With four phases, QPSK can encode two bits per symbol with gray coding to
minimize the bit error rate (BER) ² sometimes misperceived as twice the BER of
BPSK.

cc
 The mathematical analysis shows that QPSK can be used either to double the data rate
compared with a BPSK system while maintaining the same bandwidth of the signal, or
to maintain the data-rate of BPSK but halving the bandwidth needed. In this latter case,
the BER of QPSK is exactly the same as the BER of BPSK - and deciding differently is a
common confusion when considering or describing QPSK.

5‘ /.‘ . (5&) is both an analog and a

digital modulation scheme. It conveys two analog message signals, or two digital bit
streams, by modulating the amplitudes of two carrier waves, using the amplitude-shift
keying(ASK) in digital modulation scheme or amplitude modulation (Ar) in analog
modulation scheme. These two waves, usually sinusoids, are out of phase with each other
by 90° and are thus called quadrature carriers or quadrature components ² hence the
name of the scheme. The modulated waves are summed, and the resulting waveform is a
combination of both phase-shift keying (PSK) and amplitude-shift keying(ASK),. In the
digital QAr case, a finite number of at least two phases, and at least two amplitudes are
used. PSK modulators are often designed using the QAr principle, but are not
considered as QAr since the amplitude of the modulated carrier signal is constant. QAr
is used extensively as a modulation scheme for digital telecommunication systems.
Bandwidth speeds of 6 bytes/hz can be achieved with QAr.

%.‘ 5& : Like all modulation schemes, QAr conveys data by changing some
aspect of a carrier signal, or the carrier wave, (usually a sinusoid) in response to a data
signal. In the case of QAr, the amplitude of two waves, 90 degrees out-of-phase with
each other (in quadrature) are changed (modulated or keyed) to represent the data signal.
Amplitude modulating two carriers in quadrature can be equivalently viewed as both
amplitude modulating and phase modulating a single carrier.

)‘‘‘.‘#!‘
Error control methods for a communication system can be divided into two categories:

½‘ Error detection
½‘ Error correction

‘ ‘

Error detection is the process of monitoring received data and determining when a
transmission error has occurred. The most common error detection techniques are:

½‘ ‘‘- :
Redundancy involves transmitting each character multiple times. If the same character is
not received a fixed number of times in succession, a transmission error is deduced.

c
 ½‘ ‘#/.6 :
Echoplex is used in data communication systems where human operators enter data
manually from the keyboard. Each character received at the receiver is resent to the
transmitter for the operator to confirm that character was actually transmitted.

½‘ 6‘‘%:
With exact count encoding, the number of ones in each character is the same. The
number of ones in a character, and if this total does not equal the present value, then an
error has occurred.

½‘ -‘#(!:
In parity checking, a single bit(the bit parity) is added to each data unit to force the octal
number of binary µ1¶s in a data unit to be either odd(odd parity) or even (even parity).
Parity techniques fail when an even number of bits are in error, making it possible to
miss a large number of errors. However parity methods are simple and easy to
implement.

½‘ #(!:
Checksum is defined as the least significant byte of the arithmetic sum of the data set
transmitted. Transmitter and receiver both perform summing operation on the data
transmitted, with the checksum appended to the end of a data message at transmission. If
the receiver checksum does not equal to the transmitter checksum, an error has occurred.
Checksum techniques detect 95% of errors, but are more computation intensive than
parity methods.

½‘ -.‘-‘#(!:
Cyclic redundancy checking uses a division operation on the transmitted sequence,
appending the remainder of the division operation to the message transmitted. At the
receiver, this same division process is non-zero, an error has occurred. CRC error
checking methods detect approximately 99.95% of errors, but are very computation-
intensive.
‘
‘‘#! ‘
Error correction is the process of deducing the correct data that is supposed to be
received, and modifying the input data set so that the error ceases to exist. The most
common methods of error correction are:
½‘ !!! ‘
Here, an error detected and the receiver automatically requests for the message to be
retransmitted. This method is often referred to as ARQ or automatic request for

c
retransmission. ARQ techniques are simple, but can be hindered by the overhead
involved with acknowledgement and repeat requests.
‘
½‘ *‘‘0* 1 ‘
FEC techniques detect and correct errors at the receiving end without calling for
retransmission. With FEC, a number of bits are added to the message. These extra bits
are coded in a way that allows for a certain number of errors per message to be detected
and corrected. FEC techniques increases message overhead by the addition of these bits,
and are relatively computation intensive.

‘6!%‘ ‘+!‘‘
Various protocols have been developed for the purpose of carrier communication over
power-line. They differ in their modulation Techniques, channel access mechanisms and
the frequency bands they use. A brief overview of most popular protocols is presented
here.

‘‘‘78
The X10 specification was designed for low-bandwidth signalling over power lines
within the home. The product was developed by a company in Scotland - Pico
Electronics -with the first shipped product to market in 1978. The patent on the standard
has since expired and prices have fallen sharply. X10 applications include controlling
lights and thermostats as well as devices like the stereo amplifier, garage door opener,
television receiver and more.

The X10 system is simple and easy to use. It transmits over the electrical wiring using on
off keying (OOK). rore precisely, it uses 120 kHz signal bursts, each one millisecond
long. These signal bursts are synchronized to the zero crossings (both positive and
negative) of the ac power line signal. The specification allows for a signal burst to be
within a maximum of 200 s of the zero crossing point. Each bit transmitted occupies
two zero crossings; a binary 1 is represented by a burst followed by a no-burst, while a
binary 0 is a no-burst followed by a burst. Also, each one millisecond burst is equally
transmitted 3 times to coincide with the zero crossing point of all three phases in a three
phase distribution system. Figure 2.2 shows the timing relationship of these bursts
relative to the zero crossing.

c[
 Figure 2.2 X10 timing on 60 Hz waveform

An X10 packet, shown in Figure 2.3, encompasses eleven cycles of the power line. It
begins with a start of packet identifier consisting of the sequence 'burst, burst, burst, no-
burst', which occupies the first two cycles (four zero crossings). The next four cycles
represent the House Code, and the last five cycles represent a number code (1 through
16) or a Function Code (On, Off, etc.) This complete block is always transmitted twice,
with 3 power line cycles between each group of 2 codes. Hence the total number of
power line cycles required to complete a transmission is 2*11 + 3 = 25 power line cycles.

Figure 2.3 X10 packet format.

With this simplicity also comes a low bit rate. 25 power line cycles are required to
transmit a frame - which consists of 11 bits. The resulting bit rate is then 60*(11/25) =
26.4 bits/second. This bit rate is only useful for trivial applications, and is much too slow
for transmission of audio, video, network gaming traffic, and other higher bandwidth
network traffic. However, X10 is inexpensive and easy to use; hence it is a popular
choice for basic home automation.

‘‘‘‘(!‘
Lon works is a network protocol created by Echelon Corporation and is intended to
support communication between control devices or nodes. Each node in the network ± a
switch, sensor, motor, motion detector, etc. - performs a simple task. The overall
network performs a complex control application such as automating a building. Early
standards for the Lonworks protocol used spread spectrum modulation. Spread spectrum
communication techniques can be used to improve performance in the presence of tonal

cº
noise (noise that is present at specific frequencies only). Spread spectrum improves
performance by using a wider bandwidth for communication than what is required. The
amount of improvement depends on the available bandwidth, or in other words, the
degree of spreading. It was first used with a bandwidth of 100 kHz - 400 kHz, but this
band was found to be too narrow to provide acceptable performance given the type of
noise present on the power line. In addition, European regulations prohibit power line
signalling above 150 kHz due to potential interference with low frequency licensed radio
services.

Instead, Echelon's latest product, the PLT-22 transceiver, operates using a novel ÷ual
Carrier Frequency mode along with ÷igital Signal Processing (÷SP). The purpose of the
÷SP is to provide adaptive carrier and data correlation, impulse noise cancellation, tone
rejection, and low overhead error correction. The PLT-22 communicates using BPSK
with frequency ranges 125 kHz - 140 kHz (primary) and 110 kHz - 125 kHz (secondary).
The primary frequency range is used unless impairments prevent successful
communication in this range. When this occurs, the PLT-22 automatically switches to
the secondary frequency range. The PLT-22 communicates at a raw bit rate of 5 kbps.
This is much faster than X10, and hence more useful for more complex control of
electrical devices.

$‘‘‘ ,!‘
In 1984, the Electronic Industries Alliance (EIA) Consumer Electronics Group began an
effort whose goal was the formulation of a standard for a communication network for
consumer products in the home. The standard came to be called the Consumer Electronic
Bus (CEBus). The suite of specifications includes communication on many different
types of medium including power line, twisted pair cable, coaxial cable, infrared, radio
frequency, and fibre optic. The suite of specifications was labelled EIA-600. The full
specification was released in 1992.

This work is concerned with the physical layer coding employed by CEBus. CEBus uses
non return to zero (NRZ), pulse width encoding. There are four symbols: µ1¶, µ0¶, EOF,
EOP. These symbols are encoded using chirp spread spectrum in the bandwidth 100 kHz
to 400 kHz. In spread spectrum, the carrier signal frequency is swept over a range of
frequencies. CEBus employs a sequence of up and down frequency sweeps of the carrier
that in total for one symbol occupies a period of 100 µ s. This symbol interval is the
shortest symbol time ³1´, or unit symbol time. A 0.1 % margin of error is also defined
(100 ns for 100 µ s). Also, the time to transmit binary ³1´ is a unit symbol time (100 µ s),
while to transmit a binary 0, two unit symbol times are used (200 µ s). For random binary
data, the average symbol time is then 150 µ s, for a bit rate of 7.5 kbps. A unit symbol
time is shown in 2.4.

cü
Figure 2.4 CEBus spread spectrum chirp

One other interesting note in the CEBus standard is coupling between power phases
within the home. There are two 60 Hz phases, L1 and L2, in a home that are 180 degrees
apart in phase. Home 120V electrical devices - appliances, lights, motors, etc. - normally
connect to either L1 or L2. Only 240V devices that connect to L1 and L2 simultaneously
provide a signal path between these two branches other than the minimal coupling
provided by the distribution transformer. Therefore, a CEBus 120V device on L1 may
not communicate with a CEBus 120V device on L2 due to inadequate signal coupling
between L1 and L2. To help solve this problem, the CEBus standard says that a signal
coupler should be placed between L1 and L2 when needed to improve signal propagation
within the power line network.

)‘‘‘
.%
HomePlug is a non-profit consortium founded in rarch 2000 by thirteen leading IT
companies who have a mutual interest in high-speed networking technologies over power
lines. Its membership of now more than 80 companies includes companies specializing
in semi-conductor manufacturing, hardware/software supply, and service. The goal of
the consortium is to create an open specification for high speed power line networking
technology and to promote new products to accelerate its adoption. In June 2001, the
HomePlug v1.0 specifications were published.

The HomePlug specification is the most complex of all power line technologies. To
achieve higher bit rates, higher frequencies and bandwidth must be used than that for
X10, Lonworks and CEBus as discussed above (whose frequencies are less than 500
kHz). HomePlug communicates using Orthogonal Frequency ÷ivision rultiplexing
(OF÷r) in the 4.49 to 20.7 rHz frequency band. This method of multiplexing divides
up the available bandwidth into sub-bands. These sub-bands are mathematically
orthogonal, meaning that for the specific symbol rate they are placed at specific intervals
in the frequency domain that minimizes interference between them. In the bandwidth 0 -
25 rHz, there are 128 evenly spaced sub-carriers of which HomePlug uses 84, from the
band 4.49 to 20.7 rHz (carriers 23-106 inclusive).

cG
Before forming a symbol to be transmitted, data bits are processed using several error
control coding schemes. ÷ata bits are modulated onto the sub-carriers using differential
quadrature phase shift keying (÷QPSK), or differential binary phase shift keying
(÷BPSK). The Inverse Fast Fourier Transform (IFFT) is used at the transmitter to create
individual channel waveforms. The whole process is reversed at the receiver .

In addition to this, HomePlug adapts to channel conditions. Special frames are sent and
analyzed by receivers to determine which of the 84 sub-carriers are available for
communication. Tone raps (Tr) are then created and used by sender-receiver pairs to
adapt to varying channel conditions. Only good sub-carriers are used for
communication. Also, the modulation scheme can be changed (÷BPSK or ÷QPSK), and
the error-control coding can be modified. Altogether, 139 distinct physical data rates are
available from 1rbps to 14.1 rbps .

Several manufacturers have demonstrated HomePlug technology and it looks promising.

Field tests with HomePlug V1.0 devices in 500 homes show that 80% of outlet pairs
were able to communicate with each other at about 5 rbps or higher, and 98% could
support data rates greater than 1 rbps [13]. The HomePlug alliance has announced
plans for the development of next generation specifications. Named HomePlug AV, the
new specification will be designed to support distribution of data and multimedia-
streaming entertainment including High ÷efinition television (H÷TV) and data rates of
100 rbps throughout the home.

The ability to adapt is the real strength of HomePlug. Obviously if the power line
channel becomes harsh for communication, data rates will be slow, but reliability will be
maintained. Note that HomePlug employs a method of frequency diversity. It also uses
complex error control coding and modulation techniques that are good for reliability, but
are computationally demanding, power consuming and expensive. HomePlug provides
high enough data rates for medium speed, but its complexity and cost is more than
necessary.

9‘‘‘#‘‘"‘‘&‘+/‘#.%-
In the coming years, people are likely to use PLC to network anything that is electrically
powered such as heating, ventilation, and air conditioning (HVAC). ÷ifferent devices
and applications will have different throughput requirements ranging from 10 bps to 100
rbps or more. Simple control of devices (turning lights on and off, controlling
thermostat, etc.) is achievable with low bit rates (< 5 kbps). High quality video and high-
speed computer networking is at the opposite end of the spectrum requiring speeds up to
100 rbps.

From the different technologies given above one can see that there exists low cost,
reliable systems (X-10, Lonworks, and CEBus) and that high speed systems exist using

c
 the HomePlug protocol. These systems work well for what they were designed, but have
several drawbacks. There really isn¶t a device designed for ³medium speed´ (100 kbps
to 1 rbps). Therefore this project focuses on a low complexity, low cost medium speed
technology for power line communication.

‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘‘‘‘‘‘‘‘‘
‘
‘
‘
‘
‘

c0
‘‘‘‘‘‘‘‘‘‘‘‘$‘ ‘
/.‘ #..%!‘
#‘ #.
The characteristics of the channel must first be explored in order to design a timed
transmission protocol. Varying impedance, considerable noise, and high attenuation are
the main issues. Channel characteristics depend on the location of the transmitter and
receiver in the specific power line infrastructure and are both time and frequency
dependent. Hence, the channel can be described as random time varying with a
frequency-dependent signal-to-noise ratio (SNR) over the communication bandwidth.

‘
$‘‘‘
Attenuation is the loss of signals strength as the signal travels over distances for a
transmission line the input impedance depends on the type of line. Its length and the
termination at the far end. The characteristics impedances of a transmission line (Z) is the
impedances measured at the input of this line when its length is infinite. Under these
conditions the type of termination at the far end has no effect. A standard distributed
parameter model can be obtain the characteristics impedances of an unloaded power
cable, and it is given by the equations that follows

At the frequencies of interest for PLC communication (the high frequency range), this
approximates to

, where L and C are the line impedance and capacitance per length.
High frequency signals can be injected on the power line by using an approximately
designed high pass filter. raximum signal power line will be received when the
impedances of the transmitter, power line and the receiver are matched. Power line
networks are usually made of a variety of conductor types and cross sections joined
almost at random. Therefore a wide variety of characteristics impedances are
encountered in the networks.

Unfortunately, a uniform distributed line is not a suitable model for PLC

communications, since the power line has a number of load (appliances) of differing
impedances connected to it for variable amounts of time. Channel impedance is a
strongly fluctuating variable that is difficult to predict. The overall impedances of the low



voltage network¶s loads, so the small impedances will play a dominant role in
determining overall impedances.

Overall network impedances are not easy to predict either. The most typical coaxial cable
impedances used are 50 and75 ohm coaxial cables. A twisted pair of guage-22wire with
reasonable insulation on the wires measures at about 120ohms. Clearly, channel
impedances are low. This presents significant challenges when designing a coupling
network for PLC communications. raximum power transfer theory states that the
transmitter and channel impedances must be matched for maximum power transfer. With
strongly varying channel impedance, this is tough. We need to design the transmitter and
receiver with sufficiently low output/input impedance respectively to approximately
match channel impedances in the majority of expected situation.

‘
$‘‘‘
/
Power line impedance is important because a transmitter must match this impedance over
the desired frequency range to avoid frequency dependent distortion of a broadband
signal. Nicholson and ralack measured line impedance in the frequency range 20 kHz
to 30 rHz at 36 different commercial locations in the United States. They found that the
characteristic impedance increased with frequency. The impedance, averaged over all
sites, was approximately 1 Ÿ at 20 kHz increasing to 100 Ÿ at 30 rHz. Similar results
were obtained in European countries and in Japan. Nicholson and ralack explained that
any variation from site to site was attributed to variations in load connected to the line.
Figure 2.6 is extracted from their work to show the frequency range 50 to 500 kHz.

$$‘‘‘!‘
The major sources of noise on power line are from electrical appliances, which utilizes
the 50Hz electric supplies and generate noise component, which extend well into the
high frequency spectrum. Apart from these induced radio frequency signals from
broadcast, commercial , military, citizen band and amateur stations severly impair certain
frequency bands on power line. The primary sources of noise in residential environments
are universal motors, light dimmers and televisions. This noise can be classified as:

98‘
:‘‘/‘!‘‘
Noise synchronous to the sinusoidal power line carrier can be found on the line. The
sources of this noise tend to be silicon-controlled rectifiers (SCRs) that switch when the
power crosses a certain value , placing a voltage spike on the line. This category of niose
has line spectra at multiples of 50Hz.

c
‘
+%.  ‘/.!‘!‘
This category includes spike placed on the line by single events, such as a lighting strike
or a light switch turn on or off. Capacitor banks switched in and out create impulse
noise.‘

‘/.! ‘!‘‘
The most common impulse noise sources are triac controlled light dimmers. These
devices introduce noise a they connect the lamp to the AC line part way through each AC
cycle. These impulse occur at twice the AC line frequency as this process is repeated
every ½ AC cycle.

!‘/.! ‘!‘
This kind of noise is produced by a variety of series wound Ac motors. This type of
motor is found in devices such as found in vacuum cleaner, drillers, electric shavers and
many common kitchen appliances. Commutator arcing from these motors produces
impulses at repetition rates in the several kilohertz range. Continuous impulsive noise is
the most severe of all the noise sources.

 !-#!‘/‘!‘
This type of noise has line spectra uncorrelated with 650Hz sinusoidal carriers.
Television sets generates noise synchronous to their 15734 Hz horizontal scanning
frequency. rultiples of this frequency must be avoided when designing a
communications transceiver. It was found that noise levels in a closed residential
environment fluctuate greatly as measured from different locations in the building. Noise
levels tend to decreases in power level as the frequency increases in other words,
spectrum density of power line noise tends to concentrate at lower frequencies. This
implies that a communication carrier frequency would compete with less noise if its
frequency were higher.

,(%‘!‘‘
This is what every subscriber sees as already present on the line, and not caused by
subscriber appliances. Typically, this originates from the distribution transformer, public
lighting systems etc.

‘‘


 ‘
‘
‘
‘
‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘
‘
‘‘

‘
‘;&‘*‘ ‘;
 ‘
+
<‘
+++‘
‘‘‘‘‘‘‘‘‘‘‘‘‘‘
‘
‘
‘
‘
‘


$‘!%‘!!!‘‘
For the design of any communication system, we have to address a number of important
design issues. rodulation technique and transmission methods need to be selected to
give suitable performance in the communication environment of choice. Our
communication environment, i.e., the power line network possesses some unique design
issues of its own.

$ #‘"‘#‘.‘#!‘
Transmission of data across a noisy communication channel requires some manner of
separating the valid data from the background noise. The most common way to
accomplish this is to modulate the data at the data at the transmission end and to
demodulate the data on the reception endpoint, to make sure that the data coming from
the receiver is the same as the data being presented to the transmitter. The efficiency of
the modulation/demodulation process determines the accuracy of the data coming from
the receiver. Therefore, careful consideration must be given to the selection of an
appropriate modulation-demodulation scheme.

The modulation band selected for power line communications must meet the required
data rate while maximizing resistance to noise and interference with the signal because in
any power line, there are several sources of noise and interference, each with its own
individual characteristics.

ASK is the simplest scheme but is very rarely used, because of its relatively poor noise
performance. The amplitude variations in an ASK signal becomes a source of difficulty.
Such signals when amplified by nonlinear amplifiers generate spurious out-of-band
spectral components, which are filtered out only with difficulty.

Unlike ASK, a carrier is always present with FSK modulation. This affords the designer
several benefits. First, the carrier will load the receiver at all times providing greatly
increased noise immunity. Secondly, the strength (or amplitude) of the carrier can be
used to determine the quality of the incoming signal. FSK is a µnon-return to zero¶
modulation method. This means that the non-modulated condition is between the ³off´
and ³on ´ condition. In other words, the carrier should never be at the centre frequency
when modulation is present. The benefit here is noise immunity. Since FSK relies on
frequency change, and not amplitude change, to indicate data states, an FSK receiver is
inherently immune to amplitude noise. This increased noise immunity suggests a
potential for higher data rates.

In fact, FSK systems can achieve significantly higher data rates than the ASK
counterparts, albeit at the sacrifice of cost and power consumption.

[
Considering now the PSK techniques, BPSK and QPSK generate discontinuities in the
carrier phase, which are further sources of difficulty. When it is necessary to avoid such
amplitude and phase discontinuities, frequency modulation is the feasible solution. The
FSK waveform has constant amplitude and no matter how discontinuities the modulating
waveform maybe, its phase is continuous. Phase delay in the PLC channel is expected
and is also unpredictable. The reliable performance of FSK with any reasonable amount
of phase delay makes it the modulation scheme of choice for PLCC techniques.

$ #‘"‘‘"4-‘
Generally, the power frequencies are very low such as 50 or 60Hz. Hence if we use low
carrier frequency over power frequency then it would be very difficult to separate out the
carrier signal from the power signal. Again, the power currents have higher harmonics
and due to surge currents and corona effects higher frequency currents are generated.
These generally lie between 100 to 50,000Hz. So if carrier frequency is chosen within
this range then noise introduced in this path will be large and signal to noise ratio would
become too low. However, if the carrier frequency is chosen as high as 500 KHz then the
open long line behaves as Arial and radiates out such high frequency signal. For this
reason frequency between 50 to 500 KHz is chosen to be used as carrier frequency over
the power lines.
The type of channel equipment and bandwidth being used will dictate the minimum
frequency separation requirements. Table 3.1 has typical values for FSK transmitters to
transmitters (uni-directional) and transmitters to receivers (bi-directional). These tables
reflect the minimum requirements, assuming a 15 dB isolation is provided with an
external device between the equipment.

º
Table 3.1 ± Frequency Spacing Requirements in kHz for FSK Equipment

The typical frequencies used in Power Line Carrier range from 30 to 500 kHz. In
considering which frequency to use for the specific application, several things must first
be considered.

1. Application requirements
‡ What are the bandwidth/frequency spacing requirements?
‡ Is there interference from other sources?
2. Surrounding frequencies in use
3. Frequency Planning

4. Coupling rethod
5. Line configuration for noise and attenuation considerations
6. Overhead and/or power cable

$$‘+.‘"‘‘.‘#‘‘#‘ ‘ ‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘‘

In general, the complexity of the PLCC system influences the selection of the error
control method. The optimal solution is to use FEC error detection and correction
methods to cope with the majority of errors, and then CRC and ARQ detection and
correction methods to cope with errors missed by FEC. However, such a system would

ü
be complex to implement. A less complex system with moderate performance would be
straight ARQ techniques, with cyclic redundancy checking to detect errors or for simpler
system. Straight parity checking and ARQ. Accordingly, we implement parity checking
and ARQ.

‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘

G
$‘ /.%‘+%-‘‘
Once the data signal has been generated, it needs to be placed on the power line by some
kind of coupling network. The idea is to superimpose the data signal onto the 240V,
50Hz power waveform, and extract it afterwards at the receiving end.
There are three possible combinations of lines on which to couple a signal.
½‘ Live to ground,
½‘ Neutral to line
½‘ Neutral to ground
÷ifferential mode coupling is a scheme where the live wire is used as one terminal and
the neutral as the other. In case where a neutral line is not present the ground line acts as
the second terminal.

Common mode coupling involves the live and neutral being treated as one of terminal
and the ground as the other. Such a kind of coupling is not potentially safe and hence is
not used.

For our coupler implementation we use differential mode coupling. The basic component
used for the coupling may be capacitive or inductive. Inductive coupling provides a
physical separation between the power network and communication network, making it
safer to install. For safety reasons we have decided upon using an inductive coupling
method.

At the receiver side the coupling device should posses a band pass characteristics,
blocking the 50Hz mains voltage and passing signal at the carrier frequency . At
transmitter side, the coupler should posses high pass properties, passing the
communication signal un-attenuated. The coupler should also be impedance matched to
the power line for maximum power line for maximum power transfer.

$‘&!‘"‘/.%‘‘/‘.‘
As with most systems, there is more than one way to couple the carrier to the power line.
The deciding factor may be economic, performance or a compromise of the two. That is,
the best performance may be expensive to justify for the line being protected so the next
best one may be the preference. rost protective relay channels use single-phase-to-
ground coupling, requiring only one set of coupling equipment (line tuner, coupling
capacitor and line trap). rulti-phase coupling may be used to improve dependability, but
requires multi-sets of coupling equipment. As stated before, the coupling schemes with
least losses (ranked in order of least losses) are shown below:

‡ rode 1 Coupling (Out on two outer phases, in on the center phase)

‡ Center phase to outer phase (push-pull)


‡ Center phase to ground
‡ Outer phase to outer phase with ground return (push-push)
‡ Outer phase to ground (only on short lines)

The major methods of coupling are discussed as below :

<‘‘/#!‘/.%‘
The best single-phase-to-ground scheme uses the centre phase for coupling. The centre
phase provides the most mode 1 coupling. Using one of the outside phases will introduce
more mode 2 and mode 3 coupling than desired. Figure 3.1 shows an example of phase-
to-ground coupling.
Here the carrier terminals are connected between one phase conductor and the ground.
The wave trap prevents the carrier currents from flowing to the power station side. The
ground is used as the return path. Advantage of using this method is that only half of the
coupling devices are required and the cost is lowered.

Figure 3.1. Single Phase-to-Ground (Center Phase) Coupling

#!‘‘#!‘ /.%‘
Some applications will require more dependability. When the protected line is of
significant importance and the type of protection requires receipt of the signal during an
internal fault, multiphase coupling improves ependability of the signal being transmitted
through the fault. Since the most frequent type of power system fault is a phase to

0
ground, you can improve your chances of receiving the signal through the fault if more
that one phase is used. Figure 3.2 shows how to couple using the push, pull type coupling
It provides metallic go and run paths to the carrier currents. In this case the uncoupled
conductor does not have any appreciable influence on the transmission. It has such some
advantage as It has less attenuation & transmission characteristics are very constant.
Hence, radiation loss is smaller & signal/noise ratio is high. Hence it is used although
costlier.

Figure 3.2. Phase-to-Phase Coupling

$‘ /.%‘4/:
/.%‘ /! ‘‘
The coupling capacitor is used as part of the tuning circuit. The coupling capacitor is the
device which provides a low impedance path for the carrier energy to the high voltage
line, and at the same time blocks the power frequency current by being a high impedance
path at those frequencies. It can only perform its function of dropping line voltage across
its capacitance if the low voltage end is at ground potential. Since it is desirable to
connect the line tuner output to this low voltage point a device must be used to provide a
high impedance path to ground for the carrier signal and a low impedance path for the
power frequency current. ÷epending on line voltage and capacitor type, the capacitance
values in use range from 0.001 to .05 microfarads.

‘
‘



‘!‘
‘
In conjunction with the coupling capacitor, the line tuner provides a low loss path to the
power line for the carrier signal. There are two basic types of line tuners, resonant and
broad-band. The type used depends on the transmission line and the number of carrier
channels to be placed on the line.

‘/!‘
To block the energy from going back into the bus and direct it toward the remote line
terminal. This device is called a line trap. The general design of a line trap is that of a
parallel LC circuit. This type of a circuit presents high impedance to the carrier signal at
its resonant frequency. Thus if the parallel LC circuit were placed in series with the
transmission line, between the bus and the coupling capacitor, then the carrier signal
would propagate toward the remote terminal. The line trap must be capable of providing
a very low impedance path to the power frequency current. The inductor in the trap
provides this path, and it is designed to carry the large currents required. Another
important function of the line trap is to isolate the carrier signal from changes in the bus
impedance, thus making the carrier circuit more independent of switching conditions.

-!‘=‘*.!‘
The purpose of the hybrid circuits is to enable the connection of two or more
transmitters together on one coaxial cable without causing inter modulation distortion
due to the signal from one transmitter affecting the output stages of the other transmitter.

Hybrids may also be required between transmitters and receivers, depending on the
application. High/low-pass and band-pass networks may also be used, in some
applications, to isolate carrier equipment from each other.

c
$$‘‘‘‘ #!!‘‘*‘
‘
Carrier frequencies exceed power frequencies by a factor of 500 or more. As a result, a
transmission line¶s response to carrier frequencies will be different from its response to
power frequencies. At the power frequency, all power lines are electrically short in terms
of wavelength. At carrier frequencies, however, most lines are many wavelengths long
because of the much shorter wavelength. The (ƒC) frequency to wavelength (Ȝ)
relationship is approximated by:

Remember that c= 3 x108 meters/seconds (speed of light) or 186,000 miles/second. From

this relationship it is clear that a 250 kHz signal will have a wavelength of 1,176 meters
(0.73 miles). This means that a 100 Kilometer (62 mile) line will be 85 wavelengths long.
At 60 Hz, this line will be only 0.02 of a wavelength long.

$$‘‘!"‘ #!!‘‘*‘‘
There is no up-to-date reference available on the impedances of power transformers at
the carrier frequencies. The discussion below is a general discussion based on past
experience, and it must be remembered that the results may be entirely different.
Generally power transformers are accepted as being a high shunt impedance at the carrier
frequencies. ÷epending on their location in the carrier channel, their effect may or may
not affect carrier channel performance. It is also commonly accepted that a power
transformer connecting two transmission lines of different voltages constitutes a broad
band high-frequency blocking device, preventing carrier on one line from reaching the
other. Thus when a power transformer is at the terminal location of a carrier channel it
will probably appear to the carrier signal as a trap.
If the transformer is terminating a load tap, what effect does it have on the carrier signal?
That depends largely on the effective RF impedance to ground of the transformer and
how far the transformer is from the tap point. Testing has shown that delta connected
windings are more capacitive than wye-connected. This high capacitance produces a
lower impedance to ground than might be expected. As discussed in the previous section,
if the tap is an odd quarter wavelength long, then the impedance presented to the carrier
channel is the opposite value of the terminating transformer impedance. That is, if the
transformer impedance is low, then the impedance at the tap point will be high and the
tap will have little effect. On the other hand, if the transformer impedance is high, then
the impedance at the tap point will be low and the tap will have a significant effect on the
carrier channel. In the case of taps at even quarter wavelengths, the high terminating


impedance will be reflected as a high impedance with little effect on the channel and the
low terminating impedance is reflected as a low impedance with a large effect.

‘
$$‘""!‘"‘&!#!‘
Any time there is a change of impedance along the carrier signal path, there will be some
reflection of the signal. This reflection is caused by the mismatching of the impedances.
An example of this is when an overhead line is combined with a power cable circuit. This
reflection results in a loss of the carrier signal in the transmitted direction. This loss can
be calculated by the following equation:

where rL is mismatch loss.

$$$‘‘ #!‘
/‘‘
The characteristic impedance of a transmission line is defined as the ratio of the voltage
to the current of a traveling wave on a line of infinite length. This ratio of voltage to its
corresponding current at any point the line is a constant impedance, Z0. Carrier terminals
and line coupling equipment must match the characteristic impedance for best power
transfer.

In practice, the jȦC and jȦL are so large in relationship to R and G, this equation can be
reduced to :

By applying appropriate formulas for L and C, this equation can be expressed in terms of
the distance between conductors and the radius of the conductor as follows:

‘
‘

‘
‘
‘
‘
‘
‘
‘
‘
‘
‘
‘

‘*‘

‘
;&‘*‘ ‘;
 ‘
+
<‘
&&
‘
‘
‘
‘
‘
‘
‘

[
)‘‘ ‘+-!‘  ‘
‘
Our device consists of two parts TRANSrITTER and RECEIVER. Here both ends are
Transceiver modules; however the calling end is referred to as Transmitter while the
called end is sighted as the Receiver. Each room may have the equipment connected to
each other through power line. Transmitted signals shall reach the intended receivers, and
the reply must also be obtained to the calling end. We have demonstrated the working by
only two transceiver modules.

‘
Figure 4.1 The PLC System Block ÷iagram
‘
‘
)‘‘!%‘ !‘
rany factors will affect the reliability of a power line carrier (PLC) channel. The goal is
to get a signal level to the remote terminal that is above the sensitivity of the receiver,
and with a signal-to-noise ratio (SNR) well above the minimum, so that the receiver can
make a correct decision based on the information transmitted. If both of these
requirements are met then the PLC channel will be reliable. The factors affecting
reliability are:

‡ The amount of power out of the transmitter.

‡ The type and number of hybrids required to parallel transmitters and receivers.

‡ The type of line tuner applied.

‡ The size of the coupling capacitor in terms of capacitance.

º
‡ The type and size, in terms of inductance, of the line trap used.

‡ The power line voltage and the physical configuration of the power line.

‡ The phase(s) to which the PLC signal is coupled.

‡ The length of the circuit and transpositions in the circuit.

‡ The decoupling equipment at the receiving terminal (usually the same as the
transmitting end).

‡ The type of modulation used to transmit the information, and the type of demodulation
circuits in the receiver.

‡ The received signal-to-noise ratio (SNR).

The above list may not be all inclusive, but these are the major factors involved in the
success or failure of a PLC channel. The paper will deal with each one of the above items
in detail, and then use this information to design a reliable power-line carrier channel
using an example.

‘
Figure 4.2. Basic Power Line Carrier Terminal

‘
‘
‘

ü
)‘‘‘ /.%‘ /!‘‘ !‘
‘

)‘‘ /.%‘ /!‘‘

The coupling capacitors will be discussed before the line tuners since they play a large
part in the response of the line tuner. In fact the coupling capacitor is used as part of the
tuning circuit. The coupling capacitor is the device which provides a low impedance path
for the carrier energy to the high voltage line, and at the same time blocks the power
frequency current by being a high impedance path at those frequencies. It can only
perform its function of dropping line voltage across its capacitance if the low voltage end
is at ground potential. Since it is desirable to connect the line tuner output to this low
voltage point a device must be used to provide a high impedance path to ground for the
carrier signal and a low impedance path for the power frequency current. This device is
an inductor and is called a drain coil. The coupling capacitor and drain coil circuit are
shown in Figure 4.3.‘

‘‘ ‘ ‘
Figure 4.3 Coupling Capacitor & ÷rain Coil Combination

It is desirable to have the coupling capacitor value as large as possible in order to lower
the loss of carrier energy and keep the bandwidth of the coupling system as wide as
possible. However, due to the high voltage that must be handled and financial budget
limitations, the coupling capacitor values are not as high as one might desire. Technology
has enabled suppliers to continually increase the capacitance of the coupling capacitor for
the same price thus improving performance. ÷epending on line voltage and capacitor
type, the capacitance values in use range from 0.001 to .05 microfarads.

G
[



‘
In conjunction with the coupling capacitor, the line tuner provides a low loss path to the
power line for the carrier signal. There are two basic types of line tuners, resonant and
broad-band. The type used depends on the transmission line and the number of carrier
channels to be placed on the line.

The line tuner should be mounted either in the base of the coupling capacitor, if space is
available, or on the structure that supports the coupling capacitor. The reason is that the
lead between the coupling capacitor and tuner should be as short as possible. Since the
coupling capacitor is part of the filter circuit, the point of connection between it and the
line tuner is generally a high impedance point. Any capacitance to ground in the
connecting cable will cause losses and change the tuning circuit characteristics. This
cable is typically a single conductor that is insulated for high voltage and has a very low
shunt capacitance to ground. As mentioned before, coaxial cable should not be used for
this connection.

All line tuners will have a protector unit which is connected from the output lead to
ground. This protector unit must consist of a grounding switch and a protective gap. The
gap is present to protect the tuner from failure during large transients on the power line.
These transients have large amounts of high frequency energy which is passed by the
coupling capacitor and are present at the tuner because the drain coil is high impedance
to these frequencies. The grounding switch is for personnel protection during
maintenance. Sometimes the line tuners are supplied with a drain coil in addition to the
one supplied in the coupling capacitor. This drain coil should not be considered as the
primary drainage path. The coupling capacitor must always have a drain coil and it is
considered the primary drainage path for power frequency currents.
! +%.‘*4-‘
The single-frequency tuner, shown in Figure 4.4, has a single inductor and a matching
transformer. The inductor is arranged so that it and the coupling capacitor form a series
resonant circuit. When this circuit is tuned to the carrier frequency it will provide a low
impedance path for the carrier signal to the power line. The matching transformer
provides the impedance match between the 50 or 75 ohm coaxial cable and the
characteristic impedance of the power line (150 to 500 ohms). This tuner will tune at one
frequency,thus the name single frequency tuner.


 Figure 4.4. Single Frequency Line Tuner
! .‘*4-

The double-frequency tuner, on the other hand, has two sets of resonant circuits so it may
be tuned to pass two frequencies to the power line. The two-frequency tuner shown in
Figure 4.5 not only provides a low loss path for two frequencies, but it also isolates the
two sets of carrier equipment from each other. As seen in Figure 4.5 there are two paths,
each with its own matching transformer and series inductor, but each path also has a
parallel LC circuit used for blocking the carrier signal from the other path. Each path is
tuned to series resonance with the coupling capacitor at its given frequency, and the
parallel LC circuits are tuned to resonate at the frequency passed by the other path. For
the two-frequency tuners, the minimum frequency separation is generally 25 per cent of
the lower frequency or 25 kHz, whichever is smaller.

Figure 4.5. ÷ouble Frequency Line Tuner

0
‘

,‘,‘!‘

If it is desired to place more than two narrow band frequency groups on the line then one
must use broad-band coupling. There are two forms of broad-band coupling used: high-
pass and band-pass tuners.

%# /!!‘

The high-pass tuner is the simpler of the two and in most cases is the preferred type. It is
usually small enough to fit in the base of the coupling capacitor and as a result does not
need an extra outdoor cabinet. Another advantage of the high-pass tuner is that the high
impedance lead to the coupling capacitor is very short and not exposed to the elements.
The high-pass tuner is shown in Figure 4.6. The equivalent circuit for the high-pass tuner
is shown in Figure 4.7. Note that the coupling capacitor is used as one of the series
branches of the high-pass circuit. The low-frequency cutoff of the circuit is determined
by the size of the coupling capacitor and the terminating impedance of the power line.
One should not apply any carrier frequencies close to the cutoff frequency of the circuit
since it does not have a stable characteristic impedance in that area. The high-pass tuner
has one coaxial cable input. Therefore, all of the carrier sets must be paralleled using the
principles described in the section on ³Paralleling Transmitters & Receivers.´

Figure 4.6. High Pass Tuner

[

Figure 4.7. Equivalent Circuit of High Pass Tuner

, /!!‘‘ ‘

This tuner provides a large bandwidth with a constant coupling impedance over a band of
carrier frequencies. The band-band tuner is as shown in Figure. The bandwidth of the
band-pass tuner depends on coupling capacitance, the terminating impedance, and the
square of the geometric mean frequency (GrF) to which the filter is tuned. One should
be careful in applying frequencies too close to the band edges of a band-pass tuner since
this area can change with varying temperature and changes in standing waves which may
be produced on the power line due to changes in line termination.

ner

Figure 4.8. Band-pass Tuner

)$‘‘‘/!‘
When the carrier signal is coupled to the power line it can propagate in two directions,
either to the remote line terminal or into the station bus and onto other lines. If the signal
goes into the station bus much of its energy will be shunted to ground by the bus
capacitance. Also some of this energy would propagate out on other lines thus
transmitting the signal to a large portion of the system. This is undesirable since the same
frequency may be used on another line. Because of these problems, a device is needed to
block the energy from going back into the bus and direct it toward the remote line

[c
terminal. This device is called a line trap. The general design of a line trap is that of a
parallel LC circuit. This type of a circuit presents high impedance to the carrier signal at
its resonant frequency. Thus if the parallel LC circuit were placed in series with the
transmission line, between the bus and the coupling capacitor, then the carrier signal
would propagate toward the remote terminal. The line trap must be capable of providing
a very low impedance path to the power frequency current. The inductor in the trap
provides this path, and it is designed to carry the large currents required. Another
important function of the line trap is to isolate the carrier signal from changes in the bus
impedance, thus making the carrier circuit more independent of switching conditions.

Figure
4.9. Characteristic of Single Frequency Trap

Figure 4.10. Characteristic of ÷ouble Frequency Trap

Note that both the single- and double-frequency traps have a rather sharp resonance peak
which provides a 7,000 to 10,000 ohm blocking impedance at one given frequency, as
shown in following figures. On the other hand, the wide-band trap will block a large

[
bandwidth of frequencies but its blocking impedance is low, on the order of 500 ohms.
Therefore, the resonant traps will have less losses than the wide-band type.

Figure 4.11. Characteristics of a Band-pass Trap

Line traps come in several versions just as the tuners do, and these types are single-
frequency, double-frequency, and band-pass. Usually the trap used is the same type as
the line tuner, that is, if the tuner is a single-frequency type, the trap will also be a single-
frequency type. However, it is not absolutely necessary that the line trap be of the same
type as the tuner. As an example wide-band traps could be used at all times. The question
of economics and blocking impedance will dictate the type of trap to be applied.

+%.‘*4-‘‘.‘*4-‘‘
Figure 4.9 shows the typical characteristic for the single-frequency trap and Figure 4.10
shows double-frequency traps. The trap can have both a low-Q and a high-Q setting. The
low-Q setting of the trap provides a lower blocking impedance, but has a wider
bandwidth. This setting can thus be used to couple two or more very close frequencies to
the line. The high-Q setting of the trap provides the normal high blocking impedance, but
it has a very narrow bandwidth which may be very susceptible to variations in the bus
impedance. The bus is capacitive at carrier frequencies and it can form a series resonant
circuit with the inductance of the trap, and this then can create a low impedance path to
ground. Power transformers on the line behind the trap have been known to affect the

[
trap and change the tuning characteristic. These types of effects can be detected by
comparing the received signal at the other end for two conditions. The first level is
measured with the disconnect switch between the trap and the bus open. The second level
is with the line normal (disconnect closed). If the signal level changes by a large amount
between these two conditions and you are certain the trap is tuned properly, then the low-
Q setting should be selected since the station impedance will have less effect on the trap
tuning. The channel losses will be a little higher, but the channel will be less affected by
switching conditions. Note, that not all traps have the low-Q option, and you should
check with the manufacturer of the trap.

' ‘‘
When applying a wide band trap, two things must be decided, that is, bandwidth
requirements and how much blocking impedance is needed. Both these factors will
greatly affect the cost of the trap. The blocking impedance and bandwidth are directly
related to the required inductance which is a large part of the cost. Also it is suggested
that frequencies not be used that are near the band edge of the trap because the tuning in
that area may change with system conditions. Figure 16 shows the typical characteristic
for the wide band trap.

))‘‘
-!‘‘
The purpose of the hybrid circuits is to enable the connection of two or more
transmitters together on one coaxial cable without causing inter modulation distortion
due to the signal from one transmitter affecting the output stages of the other transmitter.

Hybrids may also be required between transmitters and receivers, depending on the
application. The hybrid circuits can, of course, cause large losses in the carrier path and
must be used appropriately. High/low-pass and band-pass networks may also be used, in
some applications, to isolate carrier equipment from each other.

There are many forms of hybrids, such as resistive hybrids, reactance hybrids, and
skewed hybrids to name the most popular types. Simply stated a hybrid is a bridge
network. The complete bridge is made up of components internal to the hybrid and the
external circuits connected to the hybrid.

[[
/‘"‘!! ‘
-‘‘

Figure 4.12. Resistive Hybrid

Refer to Figure 4.12. The hybrid, in this case, is made of a resistor of 25 ohms, and a
transformer with a center tap on the primary. The transformer turns ratio is ¥2/1 with the
¥2 turns on the center tapped primary. Let¶s assume the secondary of the transformer is
terminated with a 50 ohm resistor and a voltage (V) is applied to input port #1. The 50
ohm load will be reflected in the primary of the transformer as a 25 ohm quantity from
point (a) to the center tap (ct). This is because there is 1 turn on the primary, (a) to (ct),
for every ¥ ¥2 turns on the secondary. The impedance will be transferred as the square of
the turns ratio which in this case is 2 to 1. The voltage V will divide equally between the
25 ohm resistor and the 25 ohm reflected load into the top half of the primary. Thus each
voltage has a value of V/2, and in the direction as shown. Since the center tapped primary
of the transformer will act as an autotransformer, a voltage V/2 will also appear on the
other half of the primary between point (ct) and (b). The voltage appearing across input
port #2 due to the voltage V at input port #1 is the sum of the voltages around the loop
from (g) to (y). As shown in Figure 4.12, this resultant voltage is 0 volts, and the hybrid
isolates the voltage at one input port from the other input port. This isolation expressed in
decibels is called trans-hybrid loss and is the same as return loss. Return loss is the ratio
in decibels of the power into a discontinuity to the power reflected from the
discontinuity. In terms of impedance this would be the ratio of sum of the impedances to
the difference of the impedance. The reciprocal of this impedance ratio is call the
reflection coefficient.

The mathematical expressions are:

where RT is its terminating resistance and R is the designed impedance.

[º
A price must be paid for this isolation, and that is attenuation of the carrier signal from
either input port to the output port. This loss is the ratio of the input voltage V and the
output voltage V/¥2, expressed in dB. The result of this calculation will be 3 dB.
However, the transformer will have some losses and the loss from input to output will be
on the order of 3.5 dB for most hybrids of the type shown in Figure 4.12. The difference
in decibels between the input power to a device and the output power of the device is the
insertion loss. This can be expressed as follows:

If we analysis of the hybrid shown in Figure 2 using a termination of 45 ohms, the results
would be different than discussed above. That is, the voltage will not divide equally
between (a) to (ct) and (ct) to (g) and a resultant voltage will appear across input port #2.
Thus the hybrid can only provide the best isolation when it is properly terminated, in this
case, with a 50 ohm resistor. It is then appropriate to only apply a non-adjustable hybrid
in an area of known termination. In cases where there is a termination of 45 ohms, the
trans-hybrid loss (return loss) will be:

Adequate return loss is 30 dB or greater.

‘
-‘

Figure 4.13. Reactance Hybrid

When a hybrid is connected to the power line through a line tuner and coupling capacitor
the termination impedance may not always be a 50 Ÿ resistive. Therefore, the hybrid
which is connected to the tuner should be an adjustable type and should be designed to
handle non-resistive terminations in order to obtain the best performance. This type of
hybrid is called a reactance hybrid and is shown in Figure 4.13. Note that the transformer

[ü
has impedance matching taps to adjust to different magnitudes of termination. The
balance network is no longer a simple resistor, as in the resistance hybrid, but a resistor,
inductor, and capacitor. This is to enable the hybrid to adjust to non-resistive loads. The
reactance hybrid will also use an impedance matching transformer similar to the one used
by resistive hybrids. Balanced hybrids have equal losses from each input port to the
output port. The success of a PLC channel will depend on the received SNR, and this can
be obtained by maximizing the amount of transmitter signal that is coupled to the phase
wire.

.‘0+(1‘

It is desirable to use balanced hybrids in most applications, but there may be other factors
to consider on long lines where losses may be high. Another type of hybrid can be used
in an application of this type. It is called a skewed hybrid. Its name comes from the fact
that the losses from input port #1 to the output are not the same as the losses from port #2
to the output. The skewed hybrid may be designed with different magnitudes of
unbalance, but the most common is 0.5/12 dB. That is, the loss from input port #1
(transmit port) and the output is 0.5 dB and the loss from output port to the input port #2
(receive port) is 12 dB.

The skewed hybrid then allows the transmitter to be isolated from the receiver with only
a 0.5 dB loss instead of the 3.5 dB loss of the balanced hybrid. Thus twice as much
transmitter power (3 dB) is applied to the line, and the SNR will be improved by 3 dB.
The high losses in the receive path do not affect the SNR since the noise is attenuated by
the same amount as the signal. The skewed hybrid will generally have an impedance
matching network with a fixed balance network and would be considered a resistive type
hybrid. When using a skewed hybrid, the receiver port must be terminated in 50 ohms.

A summary of some of the more important application rules are given below:

‡ All hybrids in a chain should be resistive type hybrids except the last hybrid, that is, the
one connected to the line tuner.

‡ The last hybrid in the chain should be a reactance type hybrid or a skewed type hybrid.

‡ When applying transmitters to reactance type hybrids the frequency spacing between
the widest spaced transmitters is about 4% for frequencies below 50 kHz and 6% for
frequencies above 50 kHz.If this rule is not followed then the hybrid cannot be adjusted
to provide the best possible isolation between all transmitters.

‡ When applying transmitters and receivers to a reactance type hybrid the frequency
spacing between the transmitter group and receiver group is of no concern; however, all
the transmitter frequencies must meet the frequency spacing rule above. This rule is
based on receivers with a high input impedance.

[G
‡ When the last hybrid is a skewed type then the receiver port should be terminated with
a 50 ohm resistor to obtain proper isolation.

Figure 4.14 Hybrid Connections ± ÷ual Bi-÷irectional Channel

Figure 4.15. Hybrid Connections ± Four Transmitters (Unequal Losses)

‘
)9‘‘*.!‘
‘
L/C Filters: While not providing the isolation of a hybrid, L/C filters may be used to
combine two or more transmitters. The bandwidth response of the series resonant L/C
filter is a function of the L:C ratio and the frequency to which it is tuned. The insertion
loss of the L/C filter is typically around 2 dB, while the return loss is only around 10 to
15 dB, depending on application. Another disadvantage of the L/C filter is the tuning
required during installation dictates accurate tuning to maintain the needed isolation.
rinimum frequency separation of the transmitters should be 25 kHz or 10% of the
highest frequency. These would typically be used where hybrids could not be applied.
However, one should calculate the isolation resulting from use of a resistive hybrid as
compared to the LC unit. A miss termination of a resistive hybrid of anywhere from 25 to
100 ohms will produce a 10 dB or greater return loss. The advantage here would not have
to tune a LC unit.
‘
[
)$‘+-‘"‘ ‘"‘‘‘. ‘&.‘.-!!‘

Prediction of carrier performance can be accomplished through the use of rodal

Analysis. rodal Analysis is a mathematical tool similar to symmetrical components used
for analyzing unbalanced faults on three phase power systems. Like symmetrical
components, modal analysis is a practical means whose modes can be electrically
generated and measured separately. rodal theory is based on the premise that there are
as many independent modes of propagation on a multiconductor line as there are
conductors involved in the propagation of energy. What follows is a simplified
explanation of rodal Analysis. There are five characteristics of natural modes:

1. The phase-conductor currents or voltages can be resolved into three sets of natural-
mode components at any point on a lossy, reflection-free three-phase line.

2. At any point on a line, the mode components will add to the actual phase quantities, as
well the total power derived be equal to the sum of the mode powers.

3. The mode characteristic impedance, which is the ratio of mode voltage to mode
current, is constant on each phase conductor.

4. Each mode propagates with a specific attenuation, wave length and velocity.

5. One set of mode components can not be resolved into other mode components. There
is no inter-mode coupling on a uniform line since the modes are independent.

Figure 4.16. rode ÷istribution for a Three-Phase Line

Each mode has it own characteristics. rode 1 is the least attenuated and least frequency
dependent of the three and makes carrier channels possible on long EHV lines. The
energy is propagated on the two outer phases and returns on the centre phase. rode 2 is
propagated on one outside phase and returns on the other outside phase. It is more
frequency dependent and has more attenuation than mode 1. rode 3 is the highest

[0
attenuated mode and is propagated on all three phases and returns via the ground. The
attenuation is so high that beyond 10 miles, mode 3 is negligible. Figure 4.16 shows the
mode propagation characteristics.

This explanation of rodal Analysis applies to a horizontally spaced, single-circuit three-

phase EHV line with two overhead static wires, grounded at each tower. The static wires
do not generate any transmission modes if grounded at each tower.

Figure 4.17. rode Components for Various Types of Coupling

º

Figure 4.18. Simplified Presentation of Basic rodes
‘
Calculations from modal analysis can become very complex but for explanation purposes
a few assumptions can be made to simplify the process. Assume the following:
‡ All phases and modes have the same surge impedances.
‡ Frequency will not be considered.
‡ Instantaneous currents (phase or modal) will be either in phase or 180 out of phase.
‘
‘
‘
‘
‘
‘
‘
ºc
‘
‘
‘
‘
‘
‘
‘
‘
‘

‘*
;‘

‘
;&‘*‘ ‘;
 ‘

'‘
&&
‘
‘
‘
‘
‘
‘
‘
‘
‘

º

 9‘
‘
9‘‘!%‘"‘ ,‘"‘ ‘+-!‘
/.‘

9‘ % ‘&(%‘+/!

(a) The printout of the design made on

$‘ software is taken on a transparent sheet.
And its dimension should approximate with the dimension of the design on software.

(b) Transparent sheet should be cut in proper dimension, then along with AgBr sheet,
they kept in a negative making unit in which X-rays is passed, and its print is taken on
AgBr sheet.

(c) Then AgBr sheet is dipped into a mixture of NaOH and . Then sheet is
dipped into the fresh water. After that it is dipped into AgI solution. The sheet is kept
for 10-15 minutes and proper shaking is done all the above processes.

Thus the negative has been developed.

9‘ 
 
‘+
*



÷IELECTRIC THICKNESS 1.6 mm

÷IELECTRIC CONSTANT 4.4

VALUE

.‘9 ‘÷ielectric specification

º
‘
‘
‘