CHAPTER EIGHT

DC-AC CONVERSION: INVERTERS

8.1 INTRODUCTION
Inverter is an electrical device which converts d.c. voltages
(or currents) to a.c. voltages (or currents). This device has several
important applications in industry such as induction and synchronous
motor drives, uninterruptible power supplies (UPSs’), standby power
supplies, induction heating, traveling, camping, and HVDC systems.
Ideally, the input signal to the inverter is d.c. signal from a battery or an
output of a controlled rectifier and the output signal is a.c. which can be
square wave, quasi-square wave or nearly sinusoidal wave.
Inverters can be classified as voltage source inverters or current source
inverters. Inverter converting voltage is called VOLTAGE SOURCE
INVERTER (VSI), whereas inverter converting current is called
CURRENT SOURCE INVERTER (CSI). The output of the inverter could
be at any desired frequency, voltage or current.
Power transistor such as BJT, MOSFET and IGBT, are widely used in
low and medium power inverters. Thyristors (SCRs) or GTOs inverters
are used for high power inverter. So there are many types of inverters
available in industry today, and each of them is designed to suit particular
application or to meet designed performance requirements.However, For
VSI inverter, the following types are commonly used in industry:
1- Parallel inverter.
2- Single-phase bridge-type inverter.
(a) Single-phase half-Bridge inverter.
(b) Single-phase full-Bridge inverter.
3- Single-phase series inverter.
4- Three-phase inverters.

495
 Power Electronics and Drives

8.2 PARALLEL INVERTER

The simplest type of voltage source inverter used in low power
applications and UPS system for computers is the parallel inverter.
Depending on the switching device used , there are two types of parallel
inverter namely, transistor parallel inverter and thyristor parallel inverter
which is also called parallel capacitor inverter.

8.2.1 Transistor Parallel Inverter

Figure 8.1 shows the diagram of the basic single-phase transistor
parallel inverter circuit which consists of two transistors Tr1 and Tr2 that
are alternatively turned ON and OFF. The transistors employed in the
circuit Tr1 and Tr2 carry the current in the positive and negative half
cycles.

Fig.8.1 Transistor parallel inverter circuit.

Principle of operation

Assuming Tr1 conducts and Tr2 OFF, Tr1 current flows through the
primary winding (a-b) of the output transformer Tx and produces an
induced voltage on the secondary winding. Hence, supplies a load with
positive voltage +Vdc. When Tr1 switched off (ib1→ 0) and Tr2 is switched
on by applying current ib2 in its base, current flows through the primary
winding (c-b) of the output transformer Tx and produce –Vdc at the
secondary of Tx. Hence a square wave of amplitude Vdc is generated. This
square or rectangular wave can be approximated to sine wave by using
additional filter. Waveforms of this inverter are shown in Fig.8.2 for pure
resistive load.
The advantage of the parallel inverter is that, one can use low voltage
d.c. source and obtain any output voltage by using appropriate output
transformer with proper turns ratio.

496
 Chapter 8 : DC to AC Conertion - Inverters

Fig.8.2 Transistor base current and utput voltage waveforms for the
parallel transistor inverter.

8.2.2 Thyristor Parallel Inverter

The basic single-phase parallel inverter circuit consists of two SCRs:
SCR 1 and SCR 2, an inductor L, an output transformer, and commutating
capacitor C as depicted in Fig.8.3. The output voltage and current are vo
and io respectively. The function of L is to make the source current
constant and smooth. During the working of this inverter, capacitor C
comes in parallel with the load via the transformer and helps to
commutate the two thyristors, so it is called a parallel capacitor inverter.
The operation of this inverter can be explained in the following modes.

Fig.8.3 Thyristor parallel capacitor inverter circuit.

497
Power Electronics and Drives

Mode I
In this mode, SCR1 is conducting while SCR2 is OFF. In this case current
flows in the upper half of the primary winding (section a-b) of the output
transformer. As a result an induced voltage on the secondary is produced
and supplied to the load. At this time the voltage between a and c
terminals, Vac , is equal two times the d.c. supply voltage. In other words
total voltage across primary winding is 2Vdc. Now the capacitor C charges
to a voltage of 2Vdc with upper plate as positive.
Mode II
When SCR2 is turned ON, by applying a trigger pulse to its gate, at this
time (t = 0), capacitor voltage 2Vdc appears as a reverse bias across SCR1,
it is therefore turned OFF. A current io begins to flow through SCR2 and
lower half of primary winding. Now the capacitor has charged (upper
plate as negative) from +2Vdc to -2Vdc at time t = t1. Load voltage also
changes from Vdc at t = 0 to –Vdc at t = t1.
Mode III
When capacitor has charged to –Vdc , SCR1 may be tuned ON at any time.
When SCR1 is triggered, capacitor voltage 2Vdc applies a reverse bias
across SCR2, it is therefore turned OFF. After SCR2 is OFF, capacitor
starts discharging, and charged to the opposite direction, the upper plate
as positive. These procedures (Modes) proceed and repeat.
The inductor L in Fig.8.3 is used also to limit the commutating current
during commutation process. In the absence of L, The capacitor will
charged and discharged very rapidly during each of SCR s’ conduction.
The result is that the SCR may have no time to turn off. The waveforms
of the SCR and capacitor voltages and currents are shown in Fig.8.4 (a)
and (b) respectively.

The rms Value of the Output Voltage

The output voltage waveform of the inverter is a square wave as shown
in Fig.8.2, the rms value of the square wave can be evaluated as follows:
Any function vo(ωt) that is periodic in 2π radians has a root mean square
rms or effective value defined by:

√ ∫

The square wave is defined as,

.2)

498
Chapter 8 : DC to AC Conertion - Inverters

(a)

(b)
Fig.8.4 Waveforms of the SCR and capacitor (a) voltage, and (b)
current for the parallel inverter.

In the present case,

√ ∫ √ ∫

Therefore, the rms value of the output voltahe waveform is equal to the
d.c. input voltage source. The main disadvantages of the parallel capacitor
inverter are the heavy transformer required to carry the load current and
the large energy trapped in the commutating capacitor which need to be
removed by additional feedback large ratings diodes.

Harmonic Analysis of the Output Voltage Waveform

The output voltage waveform of the inverter is a square wave (Fig.8.2)
which is non-sinusoidal. Since this wave is periodic, its harmonic content
can be found using Fourier series as follows:
The Fourier coefficients are,

499
Power Electronics and Drives

Referring to the square wave defined in Eq.(8.2),

The d.c. component is:

[∫ ∫ ]

The coefficiants and of the nth order harmonic are,

[∫ ∫ ]

[∫ ∫ ]

* | | +

[ ]

 When n is even (2, 4, 6…) ,

 When n is odd (3, 5, 7…) ,

The Fourier series is given by:

49:
Chapter 8 : DC to AC Conertion - Inverters

Hence the output (load voltage) can be represented by Fourier series as,

From the above equation (8.8), the instantaneous value of the fundam-
ental component (first harmonic, n=1) is

The fundamental component peak value is,

The rms value of the fundamental component is,

√ √ √
Similarly, the instantaneous value of the third harmonic component
(n=3) is,

The peak amplitude of third harmonic , n=3 is

and the rms value of the third harmonic is,

√ √ √
and so on, the rms values for the 5th, 7th, …. odd harmonics can be
evaluate. The harmonic spectrum of the square wave is as shown in
Fig.8.5(a). It is clear from this figure that:

 Harmonic amplitudes decrease as n increases.

 No even harmonics.

4:3
 Power Electronics and Drives

 Nearest harmonic is the third, if the fundamental is 50 Hz, then the

third harmonic is 150Hz and the fifth harmonic 250 Hz as shown in
Fig.8.5 (b).

(a) (b)

Fig.8.5 (a) Harmonic spectrum of the square wave, (b) The square wave
and its odd harmonics up to the fifth order.

8.2.3 Inverter Performance Parameters

It is obvious from the previous analysis that the output voltage
waveform of the inverter is a square wave (or quasi-square wave) which
is rich of harmonics. The harmonic components are harmful for the load
and the system, because they increase the I2R losses, generate heat and
produce mechanical vibrations when applied to a.c. motors. However, the
performance parameters used to assess the amount of distortion produced
by the harmonic components in the output voltage waveform of the
inverter. These parameters are as follow:

1. The distortion actor DF

The distortion factor is defined as

[∑ ( ) ]

where Vn is the rms value of the nth harmonic component.

V1 is the rms value of the fundamental component.
Which is a factor for measuring the harmonic distortion remains in a
particular waveform after filtering. The distortion factor of the n th order
harmonic component may be calculated as

4:3
Chapter 8 : DC to AC Conertion - Inverters

2. The harmonic factor HFn

The harmonic factor of nth order harmonic is defined as the ratio of
the rms value of the nth harmonic component to the rms value of the
fundamental component.

This factor represents the contribution of each harmonic component to the

harmonic distortion and to evaluate the dominant harmonic component in
the output waveform.

3. The total harmonic distortion factor THD

This factor gives the ratio of the rms values of all harmonic
components to the rms value of the fundamental component, which is
calculated as,
∑
√

The total harmonic distortion factor is very important factor that indicates
the amount of distortion in the waveform caused by the harmonic
components. It is also represents the measure of closeness of the
waveform to pure sine wave. The THD can also be written in more
general form as:

where V is the rms value of the output waveform.

V1 is the rms value of the fundamental component.

4. The lowest order harmonic factor LOH

Finally there is another factor which is the lowest order harmonic
(LOH) that indicates the harmonic of frequency nearest to the fundam-
ental and has amplitude within 3% of the fundamental.

Example 8.1

A single-phase MOSFET parallel inverter has a supply d.c. voltage of

100V supplying a resistive load with R =10 Ω via a center-tap transf-
ormer with 1:1 ratio. The output frequency is 50 Hz.
(a) Draw the circuit diagram and the output voltage waveform of the
inverter.
(b) Determine the rms value of the output voltage waveform.

4:3
Power Electronics and Drives

(c) Determine the amplitude of the Fourier series terms for the square
output voltage waveform up to 9th order harmonics.
(d) Calculate the rms value of the output voltage in terms of harmonic
components that obtained in (b).
(e) Determine the power absorbed by the load consider up to 9th order
harmonic.
(f) Draw the frequency spectra of the output voltage waveform.
(g) Calculate the total harmonic distortion factor THD.

Solution

(a) The circuit and the output voltage waveform as shown in Fig.8.6.

Fig. 8.6.

(b) The rms value of the output voltage is found as follows:

Referring to Fig. 8.6,

⁄
√ ∫
⁄

(c) From Eq.(8.5), the Fourier series of the output voltage is

∑ ∑

The amplitude cn of the n th order harmonic is:

4:4
Chapter 8 : DC to AC Conertion - Inverters

Hence the output voltage Fourier representation is,

(d) In terms of the harmonics , the rms value of the output voltage is

√( ) ( ) ( ) ( ) ( )
√ √ √ √ √

This value is less than Vo(rms) since we calculate up to 9th order harmonics
only.

(e) To calculate the power we most calculate the rms value of the current
for each harmonic the amplitude of the nth harmonic current

where √

( )
√

n fn (Hz) cn (V) Zn (Ώ) In (A) Pn (W)

1 50 127.3 10 12.73 810
3 150 42.4 10 4.24 89.8
5 250 25.5 10 2.55 32.5
7 350 18.2 10 1.82 16.5
9 450 14.1 10 1.41 9.99

The total power is ∑ Pn = P1 + P3 + P5 + P7 + P9 = 958.8 W

4:5
 Power Electronics and Drives

(f) The frequency spectrum is given in Fig.8.7 below:

Fig. 8.7 The frequency spectrum of the output voltage waveform.

(g) The total harmonic distortion factor

Vo(rms) = 100V

Vo1(rms) = c1 /√2 =127.3/√2 = 90 V

This is very high THD , the practical value of THD is about (3-10)%
hence we need to use low-pass filter at the output to filter out most of the
undesirable harmonic component and to produce nearly sinusoidal output
waveform.

8.3 SINGLE-PHASE BRIDGE-TYPE INVERTERS

As it has been stated earlier, the parallel inverter discussed previously
has disadvantages in that it uses heavy transformer to carry the load
current. Also in the parallel capacitor inverter large amount of energy is
trapped in the commutating capacitor which is need to be removed with
additional circuit components. These disadvantages may be overcome by
using bridge-type inverter which eliminates the need of the magnetic
components such as the center-tap transformer or large capacitor. Only
semiconductor component are used in this type of inverter.

4:6
 Chapter 8 : DC to AC Conertion - Inverters

8.3.1 Single-Phase Half-Bridge Inverter

This type of inverter is very simple in construction. It does not need
output transformer like parallel inverter. It sometimes called center-tapped
source inverter. The basic configuration of this inverter is shown in
Fig 8.8.

Fig.8.8 Single-phase inverter half-bridge circuit.

The top and bottom switch has to be “complementary” i.e., if the top
switch is closed (ON), the bottom must be OFF, and vice-versa. The
output voltage waveform is a square wave as that indicated in Fig.8.2. In
practice, a dead time between ON and OFF (td) for Q1 & Q2 is required to
avoid short circuit or “shoot-through” faults. This leads to produce quasi-
square wave output voltage (Fig.8.9).

Fig.8.9 Quasi-square output voltage waveform.

Performance of half-bridge inverter with series resistive-inductive

loads

If the load is resistive, the output current waveform will be a copy of the
voltage waveform as shown in Fig.8.10(a). The output voltage is a square (or
quasi-square) wave. However with a series R-L load Fig.8.10 (b), the load
current i is delayed although the output voltage wave is still a square.

4:7
Power Electronics and Drives

(a) (b)

Fig.8.10 Output voltage and current for single-phase half-bridge inverter:

(a) Operation with resistive load, (b) Operation with R-L load.

In Fig. 8.10(b), Q1 is applied the driving current ib1 from 0 to T/2. Since the
current is still negative, this cause diode D1 to conduct from 0 to t1. The
output current io increases from – In toward zero causing Q1 to be reversed
biased and it does not conduct till D1 stop conducting at t1. When Q1
conducts from t1 to T/2, the output current increases from zero to Ip . Hence,
the current io will grow exponentially during the positive half-cycle from – In
to Ip according to the following equation:

- Through D1 [load returning power to the upper half of the source].

- Through Q1 [load absorbing power from the upper half of the
source] until t = T/2 , whereby i = Ip.
At t = T/2, Q1 is turned off and base current drive is applied to Q2.
Negative half cycle starts by conduction of Q2, but at this instant the
current is still positive which cause Q2 to be reversed bias and D2 to be
forward bias. Current starts to change direction from Ip to zero through
D2 while Q2 remains off until the it becomes zero at t2. At this instant Q2

4:8
Chapter 8 : DC to AC Conertion - Inverters

start to conduct from t2 to T until io = -In. The change of current from Ip

through D2 and then to – In through Q2 is governed by the following
equation:

The rms value of the load voltage is

√ ∫ [ ( ) ]

The load voltage can be expressed in terms of harmonics by

Fourier series as:

∑ ( )

Here for n =2,4,6…

where: ω = 2 f0 is the frequency of the output voltage in (rad /sec).

The fundamental component of the load voltage has a peak value of,

and it has rms value of

√
√
For an R-L load, the instantaneous load current can be found by
dividing the instantaneous output voltage given in Eq.(8.21) by
the load harmonic impedance Zn= R + jnωL , or

| | √ , thus

| |

∑
√
where ,

4:9
Power Electronics and Drives

If Io1 is the rms fundamental load current, c1 / √2, then the fundamental
output power (for n = 1) is

[ ]
√ √

Note: In most applications (e.g. electric motor drives) the output power
due to the fundamental current is generally the useful power, and the
power due to harmonic currents is dissipated as heat and increases the
load temperature.

Example 8.2

The single-phase half-bridge transistor inverter shown in Fig.8.8 has a

resistive load of R = 3Ω and the d.c. input voltage Vdc = 60 V. Determine:

(a) The rms value of the output voltage.

(b) The rms value of the load voltage at the fundamental frequency
.
(c) The output power.
(d) The average and peak current of each transistor.
(e) The peak reverse blocking voltage VBR of each transistor.
(f) The total harmonic distortion factor.

Solution

(a) The rms value of the output voltage is

√ ∫ [ ( ) ]

(b) The rms value of the load voltage at the fundamental frequency is
√
√

(c) The output power is

4::
 Chapter 8 : DC to AC Conertion - Inverters

(d) The average and peak current of each transistor are

Because each transistor conducts for a 50% duty cycle, the average
current of each transistor is

(e) The peak reverse blocking voltage VBR of each transistor is

(f) The total harmonic distortion factor is

√ √

8.3.2 Single-Phase Full-Bridge Inverter

The single-phase half-bridge Inverter discussed in the previous sub-
section although it has the advantage that it uses only two semiconductor
switches, but it has the distinct disadvantage in that it needs a center-tap
d.c. supply. However, this problem can be solved using full-bridge
inverter shown in Fig. 8.11.

Fig.8.11 Single-phase full-bridge inverter circuit.

The single-phase full-bridge inverter uses four BJT transistors in its

operation. Practically, this inverter is constructed from two half-bridge
inverter using single d.c. source Vdc and the load is connected between the
centers of the two legs. The four diodes shown in Fig.8.11 are used for
feedback when the load is inductive.

Operation with resistive load

When the load is resistive, the operation of the inverter can be
described as follows: Output voltage is a square wave and the current is
replica of the voltage shape. In this circuit, Q1 and Q2 are triggered
simultaneously and so are Q3 and Q4. Each device is made to conduct for
half-time of the output cycle, the load voltage waveform with the

533
Power Electronics and Drives

transistor base currents are shown in Fig 8.12. The output voltage is a
square-wave and the current is replica of the voltage shape.

Fig.8.12 Operation of the full-bridge inverter with resistive load.

Operation with resistive-inductive load

With resistive-inductive load, the current i0 lags the square-wave

output voltage v0 as shown in Fig.8.13. The operation of the inverter in
this case can be explained as follows :

Fig.8.13 Current through a single-phase full-bridge inverter.

533
Chapter 8 : DC to AC Conertion - Inverters

- Triggering Q1 & Q2 connects the load to Vd . For steady load

condition, i0 grows exponentially through D1& D2 and then through
Q1&Q2 from – In to Ip according to (Vdc = Ri + L ) .
- Triggering Q1&Q2 connects the load to Vd. For steady load condition,
i0 grows exponentially through D1&D2 and then through Q1&Q2 from
– In to Ip according to (Vdc = Ri + L ) .
- When diodes D1&D2 are conducting, the energy fed back to the
source, thus they are known as feedback diodes.
- Negative half-cycle starts by triggering Q3&Q4 at , and
Q1&Q2 goes off when ib1 = ib2 = 0 ( base current blocking).
- Load voltage reverses to –Vdc and io will flow through D3&D4 and
then through Q3&Q4 according to the equation (-Vd = Ri + L ). At
the end of negative-half cycle io = -In.
Note that:

( )

Load rms voltage is

[ ∫ ]

Load voltage may be expressed as:

The peak value of the fundamental component (n=1) of the load voltage
is,

The rms value of the fundamental component is

√
Load rms current and power can be determine from

533
Power Electronics and Drives

[ ∫ ]

where the instantaneous value of the load current io for an R-L load is

∑
√

The angle by which the load current lags the load voltage is

The total harmonic distortion factor is,

Example 8.3

For the single-phase MOSFET bridge inverter circuit shown in

Fig. 8.14, the source Vdc =125 V, load resistance R =10 Ω and output
voltage frequency fo = 50 Hz.

Fig.8.14 Single-phase full-bridge inverter.

534
Chapter 8 : DC to AC Conertion - Inverters

(a) Draw the output voltage and load current waveforms.

(b) Derive the rms value of the output voltage waveform and hence
calculate the output power Po in terms of the output voltage.
(c) Analyse the amplitude of the Fourier series terms of the output
voltage waveform by considering up to the 7th order harmonic.
Determine the value of the rms output voltage in terms of
harmonics rms values.
(d) Calculate the average and peak currents of each transistor.
(e) Estimate the total harmonic distortion factor THD of the circuit.

Solution
(a) The waveforms of the output voltage and current are depicted in
Fig.8.15 as:

Fig.8.15 Output voltage and current waveforms.

(b) The rms value of the output voltage is found from Eq.(8.25) as
follows:

[ ∫ ]

(c) The Fourier series of the output voltage is found from Eq.(7.26) as

The amplitude cn of the nth harmonic is:

535
Power Electronics and Drives

Hence the output voltage Fourier representation is,

In terms of the harmonics:

√( ) ( ) ( ) ( ) ( )
√ √ √ √ √

This value is less than Vo(rms) since we calculate up to 7th order harmonics
only.

(d) Since the duty cycle of each transistor is 0.5, the current waveform is
as shown in Fig.8.16.

Fig.8.16 Transistor current waveform.

∫ [ ]

Peak current

536
Chapter 8 : DC to AC Conertion - Inverters

(e) The total harmonic distortion factor:

Vo1(rms) = c1/√2 =159.12 /√2 =112.6 V

√ √

Example8.4

For the single-phase full-bridge, transistor inverter shown in Fig.8.14 of

the previous example, Vdc =100 V, load is a series resistance-inductance
with R=10 Ω, and L = 25 mH and the output frequency is 50 Hz. It is
required to analyse the circuit by determining: (a)The amplitudes of the
Fourier series terms for the output voltage wave up to the 9 th order
harmonics, (b)The amplitudes of the Fourier series terms for the load
current wave up to the 9th order harmonics, (c) The power absorbed by the
load in terms of harmonics, (d) Compute the total harmonic distortion
factor (THD).

Solution

(a) From Eq.(8.5) ,the Fourier series of the output voltage is

The amplitude of the n th order harmonic cn is:

cn = 4Vdc / 2π = (4x100) / nπ = 127.3 / n

vo(ωt) = c1 sinωt + c2 sin 3ωt + c5 sin 5ωt + c7 sin 7ωt + c9 sin 9ωt
= (127.3/1) sinωt + (127.3/3) sin 3ωt + (127.3/5) sin 5ωt
+ (127.3/7) sin 7ωt + (127.3/9) sin 9ωt

Hence the output voltage Fourier representation is,

vo (ωt) = 127.3 sin ωt + 42.4 sin 3ωt + 25.5 sin 5ωt + 18.2 sin 7ωt
+ 14.1 sin 9ωt.

537
Power Electronics and Drives

(b) From Eq.(8.31) , the Fourier series for the current is ,

∑
√

∑ ∑

where

| | √

Calculating In from the above equation yields,

io (ωt) = 9.29 sin (ωt – 43.25˚)+ 1.42 sin (3ωt –70.52˚)

+ 0.53 sin( 5ωt – 78.02˚) + 0.25 sin (7ωt – 81.37˚)
+ 0.17 (sin 9ωt – 83.27˚).

(c) To calculate the power, the rms value of the current for each harmonic
must be calculated, hence the power Pn of the nth harmonic is

( )
√

n fn (Hz) cn (V) Zn (Ώ) In (A) Pn (W)

1 60 127.3 13.73 9.29 431
3 180 42.4 30.0 1.42 10
5 300 25.5 48.2 0.53 1.4
7 420 18.2 66.7 0.25 0.34
9 540 14.1 85.4 0.17 0.14

The total power is ∑ Pn = P1 + P3 + P5 + P7 + P9 = 443 W

(d) The total harmonic distortion factor

⁄
√ ∫
⁄

Vo1(rms) = c1 /√2 =127.3/√2 = 90 V

538
 Chapter 8 : DC to AC Conertion - Inverters

√ √

or THD = 48.43 %

8.4 THREE-PHASE INVERTER

In high-power applications, three-phase inverters are normally used.
Moreover when three-phase voltage source required from a d.c. source
three-phase inverter is needed. However, a three-phase inverter can be
constructed using three single-phase parallel inverters which are
connected in parallel as shown in Fig.8.17. The gate signals of single-
phase inverters should be advanced or delayed by with respect to
each other in order to obtain three-phase balanced (fundamental) voltages.
The transformer secondary windings may be connected in wye or delta
whereas the primary windings must be isolated from each other.

Fig.8.17 Schematic diagram of a three-phase inverter constructed from

three single-phase parallel inverters.

Three single-phase half-bridge inverters can be connected as shown in

Fig.8.18 to form a configuration of three-phase bridge inverter. In this
circuit, there are six BJTs, Q1, Q2, Q3, Q4, Q5, and Q6. Also, there are six
diodes connected across each transistor as shown in the figure. Observe
that the upper transistors are numbered as Q1, Q3, and Q5. Similarly, the
lower transistors are numbered as Q2, Q4, and Q6. Here, Q1 and Q4 are
connected to phase-a. When Q1 conducts, phase-a is connected to +Vdc
and when Q4 conducts, phase-a is connected to –Vdc. Similarly, Q3 and Q6

539
 Power Electronics and Drives

Fig.8.18 Three-phase bridge inverter circuit diagram with star-connected

R-load.

are connected to phase-b and Q5 and Q2 are connected to phase-c.

In practice, for the three-phase bridge inverter there are two types of
conduction depending on the control signal used to drive the six
transistors with 120˚ or 180˚ conduction. Referring to Fig.8.18 : in 120˚,
each BJT transistor conducts for 120˚ and in 180˚, each BJT transistor
conducts for 180˚ as it will be described in the following subsection.

8.4.1 120 – Degree Conduction

In this type of control, only two transistors conduct at the same time
such that each transistor conducts for 120˚ and remains OFF for 240˚.
This means that only two transistors remain “ON” at any instant of time.
The per-phase and line-to-line voltage waveforms for this type of
operation are shown in Fig.8.19.
The conduction sequence of the transistors is : 1,6-1,2-2,3-3,4-4,5-1,6-
1,2-2,3. For star-connected load, three modes for three-phase bridge
inverter operation exist. In each case, the effective resistance across the
source is 2R as shown in Fig.8.20.

 During mode-1 for 0 ≤ ωt ≤ π/3 transistor 1and 6 conduct:

 During mode-2 for π/3 ≤ ωt ≤ 2π/3 transistor 1and 2 conduct:

53:
Chapter 8 : DC to AC Conertion - Inverters

 During mode-3 for 2π/3 ≤ ωt ≤ π transistor 3 and 2 conduct:

Fig.8.19 Voltage waveforms when two transistor in 120˚conduction.

It should be noted from Fig.8.20 that the phase voltage is a quasi-square

waveform. It has peak value of . They are shifted by 120˚ with
respect to each other. The output line to line voltage is a six-step
waveform, it has peak value of .

533
Power Electronics and Drives

Fig.8.20 Equivalent circuits for the three modes of operation for 120˚
conduction for the three-phase bridge inverter.

Mathematical analysis of 120˚ Mode inverter

The rms value of the output voltage waveform of 120˚ mode inverter
shown in Fig.8.19 for phase-a can be obtain as,

√∫ √∫

√
√

The line-to-line voltages for star-connected load can be found as:

√ √
√ √

The line-to-neutral voltages shown in Fig.8.19 (a) to (c) can be

expressed in Fourier series as:

∑ ( )

∑ ( )

∑ ( )

533
 Chapter 8 : DC to AC Conertion - Inverters

The peak amplitude of the fundamental component (n=1): for phase a, b

or c is:

The rms value of the fundamental component is

√
√
The line-to-line voltage waveform shown in Fig.8.19(d) is vab = √3van
with phase lead of 30˚ can be expressed in Fourier series as:

√
∑ [ ]

Similarly for the other line voltages vbc and vca , the Fourier representation
are

√
∑ [ ( ) ]

√
∑ [ ( ) ]

where for positive sequence harmonic components

(n=1,7,13,19,……) and for negative sequence harmonic
components (n=5,11,17,23,……).

8.4.2 180 –Degree Conduction

In this mode of operation of the three-phase inverter, three transistors
remain on at any instant of time. The base current drive of Q1, in Fig.8.18,
is applied for 180˚ (half-cycle) and is off for the remaining 180˚ (next
half-cycle). Base current drive of Q2 is applied with 60˚ delay with respect
to Q1. Hence for half a cycle, the following modes are presented for star-
connected load:
 During mode-1 for 0 ≤ ωt ≤ π/3 , transistor 1,5 and 6 conduct,
the equivalent circuit is shown in Fig.8.21 (a).

533
Power Electronics and Drives

(a) (b) (c)

Fig. 8.21 Equivalent circuits for the three modes of operation for 180˚
conduction.

 During mode-2 for π/3 ≤ ωt ≤ 2π/3 , transistor 1,2 and 6 conduct:

Equivalent circuit is shown in Fig.8.21 (b).

 During mode-3 for 2π/3 ≤ ωt ≤ π , transistor 1, 3 and 2 conduct:

Equivalent circuit is shown in Fig.8.21 (c).

534
Chapter 8 : DC to AC Conertion - Inverters

Waveforms for the transistors base currents and load phase voltage and
current for R-load are shown in Fig.8.22 for 180-degree conduction. The
line-to-neural voltage and line current waveform, in case of resistive load,
will be replica of the phase voltage waveform as shown in Fig.8.22(e).
However, if the load is inductive, the phase current through each load
branch would be delayed to its phase voltage as shown in Fig.8.22(f).

Fig.8.22 Line-to-neural voltage and line current waveforms for 180˚

conduction, case of R-load and phase current, (f) in case of
R-L load.

535
Power Electronics and Drives

Fourier representation of the phase and line voltages

The line-to-neutral voltage waveform shown in Fig.8.22(d) can be expr-

essed in Fourier series, which is given in many reference and text books
for the three phases a, b and c as,

∑ [ ( ) ]
√

∑ [ ( ) ]
√

∑ [ ( ) ]
√

where : for positive sequence harmonic components (n=1,7,13,19,…)

and for negative sequence harmonic components (n=5,11,17,23,…)
with respect to the line to line voltage vab .
For Y-connected load, the line to line voltage is √ without
delay angle, hence the three line-to-line voltages are given by

∑ ( )

∑ ( )

∑ ( )

The line current iL in phase-a for an R-L load is obtained by dividing the
phase voltage van in by the load impedance, Z=R+jnωL

∑ [ ( ) ]

where

√ √

536
Chapter 8 : DC to AC Conertion - Inverters

and ⁄ , the impedance angle for the nth order harmonic.

The line-to-line rms voltage is
⁄
√ ∫ √

From (8.38a), the rms value of the nth harmonic component of the phase
voltage is

From which, the rms value of the fundamental component (n =1) of the
output phase voltage is

√
√

Example 8.5

The three-phase inverter in Fig.8.18 used to feed a Y-connected resistive

load with R =15 Ω per-phase. The d.c. input to the inverter Vdc = 300 V
and the output frequency is 50 Hz. If the inverter is operating with 120˚
conduction mode, calculate : (a) The peak and rms value of the load
current IL, (b) The output power, and the average and rms values of the
current of each transistor.

Solution

(a) For 120ᵒ conduction mode, at any time the load resistances of two
phases are connected in series, hence, peak value of load current is

From Eq.(8.32), the rms value of the phase voltage is

√ √

Hence the rms value of the load current is

537
 Power Electronics and Drives

(b) The load power is

(c) For 120ᵒ conduction mode, each transistor carries current for (1/3)rd of
a cycle, hence the average transistor current is

The rms value of the thyristor current is

√ √

8.5 INVERTER OUTPUT FREQUENCY AND VOLTAGE CONTROL

In many applications of the inverter require means of controlling the

output voltage as well as its output frequency. In drive systems, the ratio
of voltage to frequency (V/f ) need to be maintained constant in order not
to drive the machine into the saturation region of its magnetic circuit.
Similarly, in UPS the output voltage of the inverter should be regulated.
The output frequency can be easily controlled by controlling the trigg-
ering instants of the power switch electronically. However, the control of
the voltage in most of these applications is usually required in order to
provide stepless adjustment of the inverter output voltage. The methods of
voltage control can be grouped into three broad categories:

1. Control of voltage supplies to the inverter.

2. Control of voltage delivered by the inverter.
3. Control of voltage within the inverter.

Control of voltage supplies to the inverter can be achieved by several

methods including the use of d.c. choppers and phase-controlled rectifiers.
The principal disadvantage of these methods is that the power delivered
by the inverter is handed twice, once by the d.c. or a.c. voltage control and

538
Chapter 8 : DC to AC Conertion - Inverters

once by the inverter. In the second category, the output may be controlled
by means of auto transformer.
In the third category, control of the inverter output voltage may be
achieved by incorporating time-ratio controls within the inverter circuit.
A more common method of controlling the voltage within an inverter
involves the use of pulse width modulation (PWM) techniques. With this
technique the inverter output voltage is controlled by varying the duration
of the output voltage pulses. Moreover, the output frequency can also be
controlled by this technique.

Principle of pulse width modulation (PWM) technique

PWM is simply the variation (modulation) of the duty of a square
pulse to produce a controlled average voltage. In its simple form, PWM is
obtained by comparing a reference signal, Ar (also called the modulating
wave) with a triangular carrier wave, Ac as shown in Fig.8.23.

Fig.8.23 PWM technique: (a) low value of modulation index M,

(b) moderate value of M, (c) high value of M.

539
 Power Electronics and Drives

By varying Ar from 0 to Ac , the pulse width  can be varied from 0o

to 180o. Thus, the voltage and hence, the amount of power sent to the load
is controlled. The modulation index is defined as,

In power electronics, the modulating signal is produced by controlling

the turn-on and turn-off of the power semiconductor devices such as
thyristors, BJTs, MOSFETs, GTOs, etc. The duty of the positive and
negative pulses (rms value) depends upon switching period which is
controlled using PWM in the gate signals in the electronic switches. By
the vision of Power Electronic applications, the quality of the PWM
technique is measured by its powerful to eliminate the output harmonics
or at least to eliminate the low order harmonics (LOH) which are difficult
to filtered out.
There are many types of PWM, however, the commonly used
techniques for controlling output voltage of a single-phase inverter are:

1- Single-pulse width modulation

2- Multiple-pulse-width modulation
3- Sinusoidal pulse-width modulation
4- Modified sinusoidal pulse-width modulation
5- Phase displacement control

Sinusoidal pulse-width modulation (SPWM) is the most commonly

used among all these techniques. However, this type of modulation has
some drawbacks that will be discussed later on in this chapter. Advanced
modulation methods are also used recently such as staircase modulation,
harmonic injection modulation, trapezoidal modulation and delta
modulation. These methods will not be discussed in this book due to their
complex theory. However in all above techniques the aim is to generate a
sinusoidal output voltage, but they differ from each other in the harmonic
content in their respective output voltage.

8.5.1 Single Pulse Width Modulation

The simplest form of pulse width modulation technique is the single
pulse width modulation. This type of modulation gives quasi – square
wave output as shown in Fig.8.24(c). According to this figure one can
observe that, there is a single pulse of output voltage during each half-
cycle and the width of the pulse is varied to control the output voltage.
A carrier signal of frequency fc with amplitude Ac is modulated by another
signal or reference signal of amplitude Ar as shown in Fig.8.24 (a).

53:
Chapter 8 : DC to AC Conertion - Inverters

Fig.8.24 Single-pulse width modulation (SPM): (a) Reference signal,

(b) Gate signals , (c) Output voltage waveform.

The main objective of using modulation process in power electronics

engineering is to generate the gate signal to the power switches and
thereby determine the output frequency of the inverter. As mentioned
before, the output voltage can be varied by varying the pulse width δ by
varying the amplitude Ar from 0 to Ac. i.e by varying the modulation
index M from 0 to 1.

Referring to Fig.8.24, the rms of the output voltage is given by

√ ∫ √

The output voltage waveform has the Fourier series

533
 Power Electronics and Drives

The peak value of the fundamental component is

Pulse width δ has a maximum value of π radians at which the fundamental

term in Eq.(8.47) is a maximum. The nth order harmonic in Eq.(8.46) is
seen to have peak value,

The distortion factor of the single-pulse waveform is therefore

√ √
√

which has a maximum value of 0.9 when δ = π.

8.5.2 Multiple-PulseWidth Modulation

An alternative waveform consisting m symmetrical spaced pulses per
half-cycle can be obtained by control the output voltage of the inverter
such that it can be switched on and off rapidly several times during each
half-cycle to produce a train of constant magnitude pulses. Fig.8.25 shows
the idea of multiple pulse-width modulation.

Fig.8.25 Multiple pulse width modulation.

533
Chapter 8 : DC to AC Conertion - Inverters

The output voltage waveform consists of p pulses for each half-cycle of

the required output voltage. Minimum pulse width is zero and maximum
pulse width will be π/p.
Let fo be the output frequency of the inverter, and T = 1/fo =2π. In this
period there are 2p pulses of equal width. The first pulse is located at π/p,
then its width is to as depicted in Fig.8.25. In this case the
rms value of the output voltage will be:

⁄
√ ∫
⁄

√ [ ] √

The frequency of the pulses is fp = 2 p fo which must be the same

frequency of the carrier signal fc that is always greater than the output
frequency fo. Thus, the number of equal symmetrical pulses p per half-
cycle can be calculated as

where mf = fc / fo is a factor defined as the frequency modulation ratio.

The Fourier series of the output voltage is found to be

where

∑ [ ( )

( )

533
 Power Electronics and Drives

here the angle αm (m = 1,2,3,4,……) is the switching angle of each

individual pulse in the positive half-cycle of the output voltage wave as
depicted in Fig.8.26.

Fig.8.26 The output voltage wave and the meaning of the angle αm.

The normalized amplitude of harmonics and the percentage distortion

factor variation with the modulation index M are shown in Fig.8.27 for
p = 5 of the multiple pulse modulation.

Fig.8.27 Normalized harmonic amplitude and distortion factor variations

with the modulation index for p = 5.

8.5.3 Sinusoidal Pulse-Width Modulation (SPWM)

Using sinusoidal reference signal will produce varied width pulses that
proportional to the amplitude of the sine wave as shown in Fig.8.28. In
this technique, the lower order harmonics of the modulated voltage wave
are greatly reduce. The rms value of the output voltage of the inverter

534
Chapter 8 : DC to AC Conertion - Inverters

depends on the widths of the pulses (δm). These widths depend on the
mdulation index M which controls the output voltage of the inverter.

Fig.8.28 Sinusoidal pulse width modulation.

The magnitude of the fundamental component of the output voltage is

clearly proportional to the modulation index Ar /Ac . But the highest
practical value of M is unity. If Ar ˃ Ac the output voltage waveform
vo(ωt) approaches a rectangular form and undesirable low frequency
harmonics such as the third , fifth and seventh harmonics are introduced
and intensified.

The rms output voltage is

√∑

and the coefficient (Bn) of the Fourier series of the output voltage will be

535
Power Electronics and Drives

∑ [ ( )

( )]

Another type of sinusoidal modulation can obtained by using two anti-

phase sinusoidal reference (modulating) signal as shown in Fig.8.29. This
technique is called double-sided triangular carrier wave modulation. The
reference signal vra produces the resultant modulated wave va , Fig.8.29
(b), whereas the reference signal vrb produces the resultant modulated
wave vb , Fig.8.29(c). The corresponding line voltage vab = va -vb has a
fundamental component as shown in Fig.8.29(d).

Fig.8.29 PWM voltage waveforms obtained by sinusoidal modulation of

double-sided triangular carrier wave.

536
 Chapter 8 : DC to AC Conertion - Inverters

8.5.4 Modified Sinusoidal PulseWidth Modulation

Here the carrier signal is not symmetry triangle wave. The carrier wave
is generated so the widths of the pulses that are near to the peak of the
sine wave not change much when modulation index changed. Such
scheme is shown in Fig.8.30 and known as MSPWM. Note that the
triangular wave is present for the period of first 60˚ of the half cycle of
sine wave. The MSPWM increase the fundamental component and
improve the harmonic characteristic. This technique reduces the number
of switching of power devices and also reduces switching losses.

Fig. 8.30 Modified sinusoidal pulse-width modulation.

Fig.8.30 Modifide PWM voltage waveforms.

8.5.5 Other PWM Methods

Besides SPWM, other PWM methods are proposed for generating
three-phase voltages to deliver more power, including third-harmonic
PWM, sixty-degree PWM and space vector PWM. All these methods are
aimed at making better use of the d.c. bus voltage and thus increasing

537
Power Electronics and Drives

modulation index. These methods are shortly mentioned here without

detailed analysis.

1. Third-harmonic PWM
In this method, the reference signal is not a pure sinusoidal wave, but
the sum of the fundamental and the third harmonic. It is shown with curve
f(ωt) in Fig.8.31. Same as the SPWM method, this signal will be
compared with the triangle signal to generate PWM signals.

Fig.8.31 Reference signal of third harmonic.

Same as the SPWM method, this signal will be compared with the triangle
signal to generate PWM signals. This concept is derived from the nature
of three-phase motors that the third harmonic will be filtered out in the
windings. Thus only the fundamental part will remain and the modulation
index of this method reaches 1.

2. Sixty-degree PWM
The sixty-degree PWM is an extension of third-harmonic PWM. It is
based on the nature of the motor that not only third harmonic, but also all
none-even triple harmonics are filtered out by the windings. Adding all
these harmonics with the fundamental together, a function with flat
segments are obtained as shown in Fig.8.32. The period of the flat part
covers 60° signal phase. The modulation index of this method can
reach 1, too.

Fig.8.32 Concept of sixty-degree PWM.

538
 Chapter 8 : DC to AC Conertion - Inverters

Drawbacks of PWM
Pulse width modulation is a simple, flexible process. It is very
common in inverter that need to provide adjustable performance and in
small battery powered audio amplifiers. However it has a few limitation
that preclude certain specialized applications.

 The highest output amplitude is equal to the input.

 Numerous switching operations occur each period. In real switches,

a small energy loss is incurred every time a switch turns on or off.
In a PMW converter with frequency ratio f switch / f out , the energy
lost in switching increases by the same ratio, relative to the loss in
a VSI inverter.

 Distortion : Although a low-pass filter should be adequate for

recovery modulation function, the inverter output has large
harmonics extending to very high frequencies. Even small residual
components at megahertz frequencies can interfere with commu-
nication equipment, sensors, analog signal processing, and
sometime digital logic.

8.6 VOLTAGE CONTROL OF THREE-PHASE INVERTER

For the three-phase inverters, output voltage control can be achieved
by the same techniques discussed in the previous section. However, the
following methods are mostly used for voltage control in three-phase
inverters.
 Sinusoidal pulse width modulation.
 60˚ pulse width modulation.
 Third harmonic pulse width modulation.
 Space vector modulation.
In sinusoidal pulse width modulation, three reference sinusoidal signals
are used one for each phase. These three reference waves have phase shift
of 120˚ between each other so that they produce three – phase output
voltage waves.

8.7 HARMONIC REDUCTIONS IN THE INVERTER

OUTPUT VOLTAGE
The output voltage waveform of the inverter is rich of harmonics; the
main objective is to obtain a sinusoidal ac output voltage waveform where
the fundamental component can be adjusted within a range and the
intrinsic harmonics selectively eliminated and to element the unwanted
harmonics.

539
 Power Electronics and Drives

There are several industrial applications, which may allow a harmonic

content of 5 % of its fundamental component if, input voltage when
inverter is used. Actually, the inverter output voltage may have harmonic
content much higher than 5 % of its fundamental component. In order to
bring this content to reasonable limit, there are several methods which can
be used:

1- Harmonic reduction by PWM.

2- Harmonic reduction by transformer connections.
3- Harmonic reduction by stepped-wave inverter.
4- Harmonic reduction using phase displacement control.
8.7.1 Harmonic Reduction by PWM
With single pulse width modulation discussed in subsection 8.5.1, the
amplitude cn of the nth order harmonic as given from the Fourier series of
the output voltage in Eq.(8.46) ,

where δ is the pulse width

If it is required to eliminate the third order harmonic (n=3) from the
output voltage waveform , let

To eliminate the fifth order harmonic:

It is obvious that m with this method only single specific order harmonic
can be eliminated. However, since the third harmonic is the next most
predominant one in the output voltage waveform of the inverter, δ must
always be 120˚.

8.7.2 Harmonic Reduction by Transformer Connections

The output of two or more inverters may be connected together using
separate transformers to produce one output voltage waveform that has
other harmonic characteristics. In order to understand the method let us
assume that we want to eliminate the third harmonic as well as the triplen
order harmonics (9, 15, 18,…), two inverter outputs vo1 andvo2 are
connected through two transformers as shown in Fig.8.33(a).

53:
Chapter 8 : DC to AC Conertion - Inverters

(a)
(b)

Fig.8.33 Harmonic elimination using transformer connection: (a) circuit,

(b) waveforms.

Let the two voltages are expressed in Fourier series as

For inverter 1

For inverter 2 , the output voltage is made to lag by π/3, thus

( )

The resultant output voltage is , hence

√ * ( ) ( )

It is obvious from Eq.(8.58) that the third order harmonic as well as all the
triplen harmonics (n=9,15,18,…) are entirely eliminated from the output

543
 Power Electronics and Drives

voltage wave. Waveforms of the system are shown in Fig.8.33(b) for

clarity.
Alternatively the triplen harmonics can also be suppressed using delta-
star transformer at the output of a three-phase inverter, since these
harmonic componenets are all in phase and they will cause circulating
currents that cancel each other through the delta-connection. Therefore,
the output voltage from the star side will not contain any triplen order
harmonic voltage.

8.7.3 Harmonic Reduction by Stepped-Wave Inverter

This method of harmonic reduction is similar to the previous one as it
also requires two inverters with separate transformers that their secondary
windings have different turn ratios and are connected in series as shown
in Fig.8.33(a). Transformer 1 has 1:1 turns ratio whereas transformer 1
may has 1:3 turns ratio such that the output resultant waveform is a
stepped wave of different amplitudes.
Alternatively, one can use parallel inverter with multi taps on its
primary winding as shown in Fig.8.34 (a).When the thyristors on the left
side of the battery conduct, the output voltage is positive. When the
thyristors on the right side conduct the output voltage is negative. The
conduction sequence of the thyristors is 1-2-3-2-1-1’-2’-3’-2’-1’to get 12
steps output voltage waveform.

(a) (b)
Fig.8.34 Stepped wave inverter (a) Circuit, (b) output voltage waveform.

8.7.4 Harmonic Reduction Using Filters

Filters of various types such as LC, LCL or OTT are also used to
reduce the harmonic contents of the output voltage waveform of an
inverter. The reactance (XL =2π f L) of an inductor increases with
frequency, therefore the effect of series inductance is to attenuate high

543
 Chapter 8 : DC to AC Conertion - Inverters

frequencies. On the other hand the reactance of a capacitor (XC = 1/2π f C)

is inversely proportional to frequency, hence a parallel capacitor provides
by-pass to high frequencies. These features are used in combination to
eliminate higher order harmonics to some extent. The types of filters in
common used for inverter circuits are depicted in Fig. 8.35.
Although LC filter shown in Fig.35(a) is simple and gives nearly
sinusoidal output voltage waveform , OTT filter shown in Fig.8.35(b)
gives better performance characteristics due to the generation of high
quality and very low harmonic content output voltage waveform. The
drastic reduction in the higher order harmonics leads to transfer of their
energies to the fundamental component of the desired frequency.
However LC filter does not has resonance problems as it exist in other
types of filters.Hoever , for an inductive load that requires a smooth
current (e.g. an electrical machine), the machine inductance provides the
filtering because it acts as a low pass filter.

(a) (b)

(c) (d)
Fig.8.35 Four types of filters commonly used with inverter circuits:
(a) Simple LC filter, (b) OTT filter, (c) LCL filter, (d) Resonant
filter.

8.8 THREE-PHASE NATURALLY

COMMUTATED INVERTER
The three-phase full-wave, fully-controlled bridge rectifier, discussed
in Chapter Three, can be used as an inverter if the load is replaced by a
d.c. supply with reverse polarity voltage as shown in Fig.8.36. The current
direction, on the d.c. side is unchanged. If this rectifier bridge is used as
inverter, the voltage and frequency as well as the shape of the wave on the
a.c. side are set by the bus and cannot be changed.

543
 Power Electronics and Drives

Fig.8.36 Three-phase naturally commutated bridge inverter.

The instantaneous values vL on the d.c. side of the bridge and the
output waveforms are identical to those of rectifier operation with passive
load shown in Fig.3.37. For α ˂ 60ᵒ , vL always positive and at α = 90ᵒ,
vL = 0 , whereas for α ˃ 90ᵒ, the average value of vL is negative.
As it was given in Chapter Three , the average value of the bridge voltage
is , from Eq.(3.40),

where is the peak phase voltage and is the rms phase voltage.
The average value of the current on the d.c. side is, from Fig.3.36,

[ ]

The power on the d.c. side is

8.9 CURRENT SOURCE INVERTER

The current source inverter (CSI) is a device that converts the input
direct current into an alternating current. It is also called current fed
inverter in which the output current is maintained constant irrespective of
load on the inverter. This means that, the magnitude and nature of the load
current depends on the nature of load impedance. The output voltage of
the inverter is independent of the load. The major advantage of current

544
 Chapter 8 : DC to AC Conertion - Inverters

source inverter is its reliability. In the case of current source inverter a

commutation failure in the same leg does not occur due to the presence of
a large inductance Ld connected in series with the voltage source.
Compared with a VSI system, the output current of a CSI system is not
influenced by the supply voltage, so its output current has low THD and
high PF. Hence, in the 1980s the current source inverters were the main
commonly used electric machine feeding devices. The current source
inverter was constructed of a thyristor bridge with large inductance and
large commutation capacitors. Serious problems in such drive systems
were unavoidable overvoltage cases during the thyristor commutation, as
the current source inverter current is supplied in a cycle from a dc-link
circuit to the machine phase winding. The thyristor CSI has been replaced
recently by the transistor reverse blocking IGBT devices (RBIGBT),
where the diode is series-connected and placed in one casing with
transistor. The power transistors like RBIGBT or Silicon Carbide (SiC)
used in the modern CSIs guarantee superior static and dynamic drive
characteristics. The use of current sources for the electric machine control
ensures better drive properties than in case of voltage sources, where it
may be necessary to use an additional passive filter at the inverter output.
Therefore, due to its advantages, the current source inverters are used in
many industrial applications such as induction heating, static var compen-
sators (SVC) , variable speed a.c. motors etc.
The freewheeling diodes that are used for voltage source inverters
become useless if an inverter is supplied from a d.c. current source, this is
because the current in any half-leg of the inverter cannot change its
polarity, hence it can only flow through the power semiconductor
switches. Therefore, absence of the freewheeling diodes reduces the size
and weight of the inverter circuit.
In practice CSI is mainly used for a.c. motor control, hence it
invariably of three-phase type. However, we shall describe both the
single-phase and three-phase types hereafter.

8.9.1 Single-Phase Current Source Inverter

Fig.8.37 shows the circuit of a single-phase current source inverter.
A constant current source is assumed here, which may be realized by
using a large inductance Ld of suitable value in series with the current
limited d.c. voltage source Vdc. The thyristors of the inverter circuits are,
T1 , T3, T2 and T4 , are alternatively turned ON to obtain a nearly square
wave current waveform. Two commutating capacitors – C1 in the upper
half, and C2 in the lower half, are used. Four diodes, D1, D2, D3 and D4 are
connected in series with each thyristor to isolate the commutating
capacitors from the load and prevent them from discharging through the
load.

545
 Power Electronics and Drives

(a) (b)

Fig. 8.37 Single-phase current source inverter: (a) Circuit, (b) Output
current waveform.

The output frequency of the inverter is controlled by varying the half

time period, (T/2), at which the thyristors in pair are triggered by pulses
being fed to the respective gates by the control circuit, to turn them ON.
The operation of the circuit can be described as :
In the circuit of Fig.8.37(a), two thyristors must trigger simultaneously to
permit current to flow. For example, T1 and T3 must be triggered, while in
reverse, T2 and T4 must trigger at the same time. The output current
waveform is shown in Fig.8.37(b) for the case of inductive load.
When T1 andT3 conduct, capacitors C1 and C2 would be charged with
the polarity as shown. When T2 andT4 are turned on, thyristors T1 and T3
are reversed biased by the capacitors C1 and C2 respectively to commutate
them. At this instant, the load current flows through T2 – C1 – D1 – load –
D3 – C2 – T4 charging capacitors C1 and C2 with opposite polarity and are
ready now to commutate T2 and T4 , while the current in the load changes
its direction and the cycle is repeated.

8.9.2 Three-Phase Current Source Inverter

The circuit of a three-phase current source inverter is shown in
Fig.8.38. As in the circuit of a single-phase CSI, the input is also
a constant current source. In this circuit, six thyristors, two in each of
three arms, are used, as in a three-phase VSI. Also, six diodes D1, D2, D3,
D4, D5 and D6, each one in series with the respective thyristor, are needed
here, as used for single-phase CSI.

546
 Chapter 8 : DC to AC Conertion - Inverters

Fig. 8.38 Three-phase current source inverter.

Six capacitors, three each in two (top and bottom) halves, are used for
commutation. It may be noted that six capacitors are equal, i.e. C1= C2 =
C3.....= C6. The six diodes are needed in CSI, so as to prevent the
capacitors from discharging through the load. The numbering scheme for
the thyristors and diodes are same, as used in a three-phase VSI, with the
thyristors being triggered in sequence as per number assigned in 120˚
conduction mode or 180˚ conduction mode discussed in three-phase
voltage source inverter in subsections 8.4.1 and 8.4.2 respectively.

PROBLEMS

8. 1 (a) Draw the circuit diagram of a single-phase half-bridge inverter.

(b) Sketch the output voltage waveform and show conducting devices.
(c) Give an expression for the rms values of the inverter output (load)
voltage.
(d) Sketch one transistor current and give its average value for a purely
resistive load.
(e) What is the peak inverse voltage of the transistors?

547
 Power Electronics and Drives

8.2 A single-phase half-bridge inverter has a resistive load of 2.5 Ω and input
voltage of 50 V. Calculate the following:

(a) The rms voltage of the fundamental frequency component,

(b) The output power,
(c) The average current and peak current,
(d) Harmonic rms voltage, and
(e) The total harmonic distortion factor.
[Ans : (a) 45 V, (b) 1000 W, (c ) 10 A , 20 A , (d ) 21.8 V , (e) 48 %]

8.3 For a single-phase transistor bridge inverter, the d.c. supply voltage
Vdc = 300 V, the load consists of resistance R = 20 Ω in series with
L = 40 mH and the output voltage frequency f o = 50 Hz.

(a) Draw the output voltage and load current waveforms for highly
inductive load.
(b) Determine the amplitude of the Fourier series terms of the output
voltagewaveform up to the 9th order harmonics.
(c) What is the maximum value of the load current (iomax)?
(d) Calculate the power absorbed by the load (consider up to 9 th order
harmonics).

[Ans : (b) c1=382 V, c3 = 127.3 V, c5 = 76.3 V, c7 = 54.5 V, c9 = 42.4 V

, (c) Iomax = 37.5 A , (d) 4316 W]

8.4 For the single-phase bridge inverter circuit shown in Fig.8.39, Vdc =125V,
load resistance R =10 Ω and output voltage frequency fo = 50 Hz.

Fig.8.39.

Fg.8.39.

(a) Write the Fourier series terms of the output voltage waveform
up to the 7th order harmonic.
(b) Draw the output voltage and load current waveforms.
(c) Determine the value of the rms output voltage in terms of

548
 Chapter 8 : DC to AC Conertion - Inverters

harmonics rms values and the output power Po (consider up to

7th order harmonics).
(d) Calculate the average and peak currents of each transistor.
(e) Estimate the total harmonic distortion factor THD of the circuit.

[ Ans : (a) vo(ωt) = 159.12 sinωt + 53.04 sin3ωt +31.82 sin 5ωt +
22.73 sin7ωt , (c) 121.78 V, 1483 W (d) 6.25 A, 12.5 A ,
(e) 48.36 %]

8.5 In a single-phase full-bridge PWM inverter, the control logic for power
electronics devices is so adjusted that the output voltage waveform
produced is a quasi-square as shown in Fig.8.40. The inverter has a
resistive load of 5 Ω and input voltage of 100V. Find an expression for
the Fourier coefficient of the output voltage for the first sixth harmonics
and their corresponding harmonic currents.

Fig.8.40.

[Ans: The Fourier series is

( )
√
{ Peak harmonics voltage and currents are : V1 = 77.97 V, V3 = 0 ,
V3 = - 15.95 V, Idc = 0A , I1 = 15.58 A , I3 = 0A , I5 = 3.19 A}

8.6 A three-phase transistor voltage-source inverter supplies a three-phase load,

as shown in Fig.8.41. The load consists of star connected resistance of
10 Ω in each phase. The inverter supply voltage is 200V d.c. and each
inverter switch conducts for 120˚.
(a) Sketch the switching signals for the six transistors.
(b) Sketch the line-to-neutral voltage for one complete cycle of the
output voltage.
(c) Sketch the line-to-line voltage for one complete cycle of the
output voltage.

549
 Power Electronics and Drives

(d) Calculate the rms values of the first five harmonics in the line-
to-line output voltage, including the fundamental.
(e) Calculate the rms values of the first five harmonics in the line-
to-neutral output voltage, including the fundamental.

[ Ans: (d) VL-L = 244.95 V, (e) Vph = 141.42 V]

Fig.8.41.

8.7 A single-phase full-bridge voltage source inverter, fed from 200 V d.c.
source. The output voltage is controlled by sinusoidal pulse width
modulation technique. The carrier and modulating signals are so adjusted
that the modulation index produces three pulses per half a cycle of widths,
20˚, 60˚, and 20˚. Assuming the load is purely resistive, find the rms value
of the output voltage.

[Ans: 149 V]

8.8 For the inverter of problem 8.7, the strategy of output voltage control is
changed such that the technique used is multiple pulse width modulation.
The width of each pulse is 20˚ and each half cycle has 6 pulses,
(a) Determine the rms value of the output voltage,
(b) Find the pulse width to maintain the output rms value in (a)
constant if the input voltage increases by 20%,
(c) What is the maximum input voltage if the maximum possible pulse
width is δ max = 25˚ ?

[Ans: (a) 163.3 V, (b) δ =13.8˚, (c) 179.45 V]

54:
 Chapter 8 : DC to AC Conertion - Inverters

8.9 Repeat problem 8.7 above if single-pulse width modulation technique with
δ = 120 ˚ is used to control the output voltage.

[Ans: 163.3V]

8.10 Calculate the rms value of the output voltage waveform shown in
Fig.8.19. The d.c. input voltage is 200V and the fundamental frequency
component of the output voltage waveform is 100Hz.

[Ans: 94V]

8.11 (a) Explain with the aid of neat sketches of waveforms the voltage control
using phase-displacement thchnique.

(b) The d.c. supply voltage of a single pulse width modulated inverter is
120V. Calculate the displacement angle to produce rms output
voltages of 50V and 100V.

[Ans : 31.25ᵒ , 125ᵒ ]

8.12 A three-phase natural commutated inverter is used to transfer power from

a 300 V battery to a three-phase 230 V, 50 Hz a.c. bus. A large filter
inductor with resistance 10Ω is included in the d.c. side. Estimate the
power transferred for (a) α = 120ᵒ, (b) α = 150ᵒ.

[ Hint : Use equations (8.59) – (8.61)]

[Ans: (a) 2248 W, (b) 834 W]

553