Single-phase Full Converters

The circuit arrangement of a single-phase full converter is shown in Figure 10.1a with a highly inductive load so that the load current is
continuous and ripple free.
• During the positive half-cycle,
thyristors T1 and T2 are forward
biased;
• at ωt = α, the load is connected to
the input supply through T1 and
T2.
• Due to the inductive load,
thyristors T1 and T2 continue to
conduct beyond ωt = π, even
though the input voltage is
already negative.
• During the negative half-cycle of
the input voltage, thyristors T3 • T1 and T2 are turned off due
and T4 are forward biased; the to line or natural
turning on of thyristors T3 and T4 commutation and the load
applies the supply voltage across current is transferred from
thyristors T1 and T2 as reverse T1 and T2 to T3 and T4.
blocking voltage.
• During the period from α to π, the input voltage vs and input current is are positive, and the power flows from the
supply to the load. The converter is said to be operated in rectification mode.
• During the period from π to π + α, the input voltage vs is negative, and the input current is is positive, and reverse
power flows from the load to the supply. The converter is said to be operated in inversion mode.
• Depending on the value of α, the average output voltage could be either positive or negative and it provides two-
quadrant operation.
The average output voltage can be found from

and Vdc can be varied from 2Vm/π to -2Vm/π by varying α from 0 to π. The maximum average output voltage is
Vdm=2Vm/π and the normalized average output voltage is

The rms value of the output voltage is given by

Example 10.1: Finding the Input power Factor of a Single-phase Full Converter
The full converter in Figure 10.1a is connected to a 120-V, 60-Hz supply. The load current Ia is continuous, and its ripple
content is negligible. The turns ratio of the transformer is unity.
(a) Express the input current in a Fourier series; determine the HF of the input current, DF, and input PF. (b) If the
delay angle is α=π/3, calculate Vdc, Vn, Vrms, HF, DF, and PF.

Solution
a. The waveform for input current is shown in Figure 10.1c and the instantaneous input current can be expressed in a
Fourier series as
Single-phase Full Converter with RL load
The operation of the converter in Figure 10.1a can be divided into two identical modes: mode 1 when T1 and T2 conduct,
and mode 2 when T3 and T4 conduct.
The output currents during these modes are similar and we need to consider only one mode to find the output current iL.
The critical value of α at which Io becomes zero can be solved for known values of θ, R, L, E, and Vs by an iterative method.
The rms current of a thyristor can be found from Eq. (10.8) as
Discontinuous load current.

The critical value of αc at which ILo becomes zero can be solved. Dividing Eq. (10.9) by 2Vs/Z, and substituting R/Z= cos θ and
ωL/R=tan θ, we get

• where x = E/ 2 Vs is the voltage ratio and θ is the load impedance angle for α ≥ αc, IL0 = 0.
• The load current that is described by Eq. (10.8) flows only during the period α ≤ ωt ≤ β.
• At ωt = β, the load current falls to zero again.
Example 10.2: Finding the Current ratings of Single-phase Full Converter with an RL load
The single-phase full converter of Figure 10.1a has a RL load having L = 6.5 mH, R = 0.5 Ω, and E = 10 V. The input voltage
is Vs = 120 V at (rms) 60 Hz.
Determine (a) the load cur rent ILo at ωt = α = 60°, (b) the average thyristor current IA, (c) the rms thyristor current IR, (d)
the rms output current Irms, (e) the average output current Idc, and (f) the critical delay angle αc.
Single-Phase Dual Converters
We have seen in Section 10.2 that single-phase full converters with inductive loads allow only a two-quadrant operation.
If two of these full converters are connected back-to-back, as shown in Figure 10.2a, both the output voltage and the
load current flow can be reversed.

The system provides a four-quadrant operation and is called a dual converter.

Dual converters are normally used in high-power variable-speed drives. If α1 and α2 are the delay angles of converters
1 and 2, respectively, the corresponding average output voltages are Vdc1 and Vdc2.
The delay angles are controlled such that one converter operates as a rectifier and the other converter operates as an
inverter; but both converters produce the same average output voltage.
Figures 10.2b–f show the output waveforms for two converters, where the two average output voltages are the same.
Figure 10.2b shows the v-i characteristics of a dual converter.
From Eq. (10.1) the average output voltages are

Because one converter is rectifying and the other one is inverting,

Because the instantaneous output voltages of the two converters are out of phase, there can be an instantaneous voltage
difference, and this can result in circulating cur rent between the two converters.
This circulating current cannot flow through the load and is normally limited by a circulating current reactor Lr, as shown in
Figure 10.2a.
If vo1 and vo2 are the instantaneous output voltages of converters 1 and 2, respectively, the circulating current can be
found by integrating the instantaneous voltage difference starting from ωt = π-α1.
Because the two average output voltages during the interval ωt = π+α1 to 2π-α1 are equal and oppositive, their
contributions to the instantaneous circulating current ir is zero.

The instantaneous circulating current depends on the delay angle. For α1, = 0, its magnitude becomes minimum when
ωt = nπ, n = 0, 2, 4, c, and maximum when ωt = nπ, n = 1, 3, 5,c. If the peak load current is Ip, one of the converters that
controls the power flow may carry a peak current of (Ip + 4Vm/ωLr).
The dual converters can be operated with or without a circulating current.
In case of operation without circulating current, only one converter operates at a time and car ries the load current,
and the other converter is completely blocked by inhibiting gate pulses.
However, the operation with circulating current has the following advantages:
1. The circulating current maintains continuous conduction of both converters over the whole control range,
independent of the load.
2. Because one converter always operates as a rectifier and the other converter oper ates as an inverter, the power flow
in either direction at any time is possible.
3. Because both converters are in continuous conduction, the time response for changing from one quadrant operation
to another is faster.
Three-Phase Full Converters
Three-phase converters are extensively used in industrial applications up to the 120-kW level, where a two-quadrant
operation is required.
Figure 10.3a shows a full converter circuit with a highly inductive load.
If the line-to-neutral voltages are defined as the corresponding line-to-line voltages are

The average output voltage is found from

Example 10.4 Finding the performances of a three-phase Full-Wave Converter
A three-phase full-wave converter in Figure 10.3a is operated from a three-phase Y-connected 208-V, 60-Hz supply and the
load resistance is R = 10 Ω.
If it is required to obtain an average output voltage of 50% of the maximum possible output voltage, calculate
(a) the delay angle α, (b) the rms and average output currents, (c) the average and rms thyristor currents,
(d) the rectification efficiency, (e) the TUF, and (f) the input PF.
Example 10.5 Finding the Input power Factor of a three-phase Full Converter
The load current of a three-phase full converter in Figure 10.3a is continuous with a negligible ripple content.
(a) Express the input current in Fourier series, and determine the HF of input current, the DF, and the input PF.
(b) If the delay angle α=π/3, calculate Vn, HF, DF, and PF.
Three-phase Full Converter with RL load

which can be solved for the critical value of α = αc

for known values of x and θ. For α ≥ αc, IL1 = 0. The
load current that is described by Eq. (10.20) flows
only during the period α ≤ ωt ≤ β. At ωt = β, the load
current falls to zero again. The equations derived for
the discontinuous case of diode rectifier in Section
3.8 are applicable to the controlled rectifier.
Example 10.6 Finding the Current ratings of three-phase Full-Converter with an RL load
The three-phase full converter of Figure 10.3a has a load of L = 1.5 mH, R = 2.5 Ω, and E = 10 V.
The line-to-line input voltage is Vab = 208 V (rms), 60 Hz. The delay angle is α = π/3.
Determine
(a) the steady-state load current IL1 at ωt′ = π/3 + α (or ωt = π/6 + α),
(b) the average thyristor current IA,
(c) the rms thyristor current IR,
(d) the rms output current Irms.
PWM RECTIFIERS
The uncontrolled and phase-controlled rectifiers are classified as line-commutated converters, since the turn-off
conditions for their switches are determined by the voltages of the ac supply line.
The non sinusoidal supply currents and dependence of the input power factor on the firing angle are major
disadvantages of those converters.
The availability of fully controlled power switches, such as IGBTs or power MOSFETs, allows to replace the phase
control of output voltage by PWM.
In contrast to line-commutated power electronic converters, those based on fully controlled switches are called force-
commutated.
Principles of PWM
• As an example, consider a simple hypothetical PWM converter described by two switching variables, x1 and x2.
• All possible states are allowed, and each state is designated by the decimal equivalent of the binary number
(x1x2)2
• Thus, the converter in question can assume any of the following states: state 0, when x1x2 = 00; state 1, when x1x2
= 01; state 2, when x1x2 = 10; or state 3, when x1x2 = 11.
• Let us say that the PWM algorithm dictates the following sequence and timing of states within certain switching
interval: 0 (50 μs)–2 (40 μs)–3 (20 μs)–2 (40 μs)–0 (50 μs). State 1 is not used.
• Clearly, the switching period is 200 μs, which corresponds to the switching frequency of 5 kHz.
The switching pattern in question is shown in Figure 4.39.

It can be seen that the pulse width of x1 is 100 μs and that of x2 is 20 μs, which means that the duty ratio of x1 is 0.5
and that of x2 is 0.1.
Both pulses are centered about the middle of the switching interval.
• Other important concepts in control of PWM converters are voltage and current space vectors.
To explain the origin of voltage and current space vectors, a cross-section of stator of a simple three-phase electric ac
machine is shown in Figure 4.40.
Stator currents, iA, iB, and iC are considered positive when they enter the stator winding at front ends of the
corresponding conductors.
Currents in the coils generate magnetomotive forces (MMFs),
which add up to the stator MMF,

It can easily be shown that as time progresses and the currents follow the positive phase sequence,
s rotates counter-clockwise with the angular velocity, 𝜔, equals to the radian frequency of the stator currents.
MMF vector can be represented by a complex number. For example, as illustrated in Figure 4.42,

assuming

An MMF is a product of number of turns in a coil and current in the coil. Therefore, dividing an MMF space vector
by the turn number (which carries no physical units) gives a current space vector,⃗i.
Equation (4.72) can be used to express components of that vector, id and iq, in terms the phase currents, iA, iB, and iC:

This abc → dq conversion, called Park transformation

The concept of current space vectors is somewhat abstract, as it applies to any three-phase currents, not necessarily
those in the stator of an ac machine.
The extension of that concept on the space vector of line-to-neutral voltages given by
 Consider an unspecified converter, whose state X produces current vector ⃗IX, state Y produces vector ⃗IY, and state Z
produces vector ⃗IZ, whose magnitude is zero. These three vectors are to be employed to generate the reference
vector ⃗I ∗ located between ⃗IX and ⃗IY, as shown in Figure 4.43.

𝐼 ∗ = 𝑑𝑥 𝐼𝑥 + 𝑑𝑦 𝐼𝑦
∗
𝜋 ∗
𝜋
𝐼 𝑐𝑜𝑠𝛼 = 𝑑𝑥 𝐼𝑥 + 𝑑𝑦 𝐼𝑦 𝑐𝑜𝑠 , 𝐼 𝑠𝑖𝑛𝛼 = 𝑑𝑦 𝐼𝑦 𝑠𝑖𝑛
3 3
𝜋 𝜋
𝐼 ∗ 𝑠𝑖𝑛𝛼 𝐼 ∗ 𝑐𝑜𝑠𝛼 𝑠𝑖𝑛 − 𝑠𝑖𝑛𝛼 𝑐𝑜𝑠
𝑑𝑦 = → 𝑑𝑥 = 3 3
𝜋 𝐼𝑥 𝑠𝑖𝑛 𝜋/3
𝐼𝑦 𝑠𝑖𝑛 3
𝜋
𝐼∗sin(3 − 𝛼)
=
𝐼𝑥 𝑠𝑖𝑛 𝜋/3
∗ 𝜋 𝜋 𝐼∗
𝐼𝑚𝑎𝑥 = 𝐼𝑦 𝑠𝑖𝑛 3 = 𝐼𝑥 𝑠𝑖𝑛 3 𝑎𝑛𝑑 𝑚 = 𝐼∗
𝑚𝑎𝑥

𝜋
𝑑𝑥 = 𝑚 sin −𝛼 𝑑𝑦 = 𝑚 𝑠𝑖𝑛𝛼
3

As time progresses, the position of reference vector changes. Also, its magnitude can vary. However, switching
intervals are practically so short that within each of them the reference vector can be assumed stationary and
unchanging.
Figure 4.43, the reference vector can be represented as

where dX and dY are duty ratios of state X and state Y, respectively, that is, relative durations of these states with
respect to the length, Tsw, of the switching interval.
The maximum realizable magnitude, I∗max, of the reference vector equals the radius of the arc inscribed in the
triangle formed by the “framing” vectors ⃗IX and ⃗IY.
In such case the sum of dX and dY equals 1.
Otherwise, to fill up the switching interval, a zero vector, IZ, is enforced with such a duty ratio dZ that

In this way, a revolving vector ⃗I∗ is generated as a time average of fractions of stationary vectors, ⃗IX, ⃗IY, and ⃗IZ, that
is, by maintaining state X for a total of dXTsw seconds, state Y for a total of dYTsw seconds, and the zero state Z for a
total of dZTsw seconds. In practice, the switching cycle is usually divided into more than three sub cycles.
Most often six sub-cycles are used, each one holding a half of the total time allocated for a given state.
Current-Type PWM Rectifier
Circuit diagram of the current-type PWM rectifier is shown in Figure 4.45.

• In the technical literature, it is sometimes referred to as a current-source rectifier

• The converter is of the buck type, which in this case means that the output dc voltage, Vo, can only be stepped
down from a specific maximum value, Vo,max.
• PWM rectifiers produce certain notches in the output voltage waveform. As a result, the maximum available dc-
output voltage reaches some 92% of the peak line-to-line input voltage.
Two major advantages of the PWM rectifier are
(1) feasibility of sinusoidal input currents with the unity input factor.
(2) continuous output current, thanks to the narrow pulses and notches of the waveform of output voltage.
Two and only two switches of the rectifier are allowed to conduct at any time, one in the upper row and the other in the
lower row.
Thus, switching variables a, b, c, a′ , b′ , and c′ of Switches SA through SC′ must satisfy the condition

The above condition limits the number of allowable states of the rectifier to 9, namely:
In state 1, currents iA, iB, and iC are equal to Io, –Io, and 0, respectively. Thus, according to Eq. (4.73), the space vector
of input current

The active (nonzero) vectors for states 1 through 6 of the rectifier are shown in Figure 4.46.
States 7, 8, and 9 produce zero vectors of input currents:
Durations of states X and Y framing a sector in which the reference current vector is currently located are given by
An example switching cycle is illustrated in Figure 4.47.

It represents a situation when m = 0.65 and 𝛽 = 70◦, which implies the location of the reference current vector in sector
II. Thus, 𝛼 = 40◦, X =2, Y =3, Z=9, and, according to Eqs. (4.90) through (4.92), T2 = 0.22Tsw, T3 = 0.42Tsw, and T9 =
0.36Tsw.
To illustrate the impact of modulation index, example waveforms of the output voltage, vo, and current, io, in a current-
type PWM rectifier operating in the rectifier mode are shown in Figure 4.49, and waveforms of the input current, ia, and
its fundamental, ia,1, in Figure 4.50.
Inverter operation of the rectifier is illustrated in Figure 4.51.
Example harmonic spectra of the input current are shown in Figure 4.52 to demonstrate the impact of switching
frequency, fsw, on current harmonics.
Example 4.1
A three-pulse and a six-pulse diode rectifiers are supplied from a 460- V ac line. Compare the dc-output voltages
available from these rectifiers.
Next, the conduction modes must be determined for proper selection of formulas for the dc-output voltage.
 which is satisfied.

which is not satisfied, that is, the rectifier operates

in the discontinuous conduction mode.