Lecture 8: AC-DC

single-phase
conversion
Monday, 10 May, 2021 8:27 PM

Right from the outset, it is crucial, crucial to recognize

your approach to this lecture should be
understanding the problem-solving techniques that
are grounded on
• the physics of devices at the operational level;
• the concepts/definitions of average values, RMS
values and power factor;
• the mathematical theory of differential equations,
Fourier series;
• symbolic computing and simulation.

You should not approach this lecture by treating the

voltage/current equations as “formulas” that you will
just look up when you are given a circuit.

So far in the course, we covered devices (diode, power MOSFET, IGBT) and DC-DC converters (buck, boost,
buck-boost), circuit modelling and controller design.

Time to cover other converters, starting with AC-DC converters, i.e., rectifiers.

Back in Lecture 1:

Rectifiers can be classified as:

Lecture 8 Page 1
Rectifiers can be classified as:
Uncontrolled Controlled
Example

Examples Half-wave Full-wave

Uncontrolled

Controlled

Example Single-phase Three-phase

Controlled
half-wave None

Lecture 8 Page 2
half-wave None

If it is controlled, it had better be full-wave

Controlled
full-wave

Lecture 8 Page 3
1. Half-wave
rectification
Monday, 10 May, 2021 9:19 PM

This section discusses uncontrolled and controlled half-wave rectification.

1.1 Uncontrolled half-wave rectification

Our circuit of interest is the half-wave rectifier with a resistive-inductive load.

Why resistive-inductive load? Industrial loads typically contain inductance as well as resistance.

The source is AC, and the objective is to create a load voltage that has a nonzero DC component.

To analyze this circuit, consider two cases separately:

1. When the diode is forward biased and conducting,
what is ?
2. When the diode is reverse biased and blocking, .

When the diode is conducting, neglect the diode’s

forward voltage drop, and apply KVL, we get a first-
order differential equation:

Standard time-domain solution (see knowledge base

entry) to the first-order differential equation:

is

Therefore, if ,

Define time constant . Substituting

sin into the above, we get
sin
Lecture 8 Page 4
Define time constant . Substituting
sin into the above, we get
sin

sin sin ,
(1.1)
sin sin ,
where , t n , .

The point when the current reaches zero occurs

when the diode turns off. The first positive value of
that results in zero current is called the
extinction angle, denoted . In other words, Eq.
(1.1) is defined for .

By definition,
sin sin

(1.2)
sin sin

Example 1.1
This example was adapted from [Har11, EXAMPLE 3-2]. For the uncontrolled half-wave rectifier, ,
H, rad/s, and V. Determine (a) an expression for the current in this circuit, (b)
the average current, (c) the RMS current, (d) the average power absorbed by the RL load, and (e) the power
factor.

Solution: Due to the computational nature of this problem, detailed solution exists in the accompanying
Live Script. The following is an outline of the solution. Based on the given data, we can straightaway write
down , , , t n , .
, o
(a)
, o

where the extinction angle is the solution to sin sin , i.e., rad, which can be
found using MATLAB function fzero.

(b) Average current , where is from (a), and . Using the MATLAB function int,
with integration limits from 0 to , we can work out to be 0.3081 A.

Lecture 8 Page 5
with integration limits from 0 to , we can work out to be 0.3081 A.

(c) RMS current A.

(d) Output power sin W.

(e) Power factor , where . So, .

1.2 Controlled half-wave rectification

The fundamental limitation of the uncontrolled rectifier is once the source and load parameters are
established, the DC level of the output and the power transferred to the load are fixed quantities.

A way to control the output of a half-wave rectifier is to use an SCR instead of a diode.

Two conditions must be met before the SCR can conduct:

1. The SCR must be forward-biased.
2. A current must be applied to the gate of the SCR.

Unlike the diode, the SCR will not begin to conduct as soon as the source becomes positive. Conduction is
delayed until a gate current is applied, which is the basis for using the SCR as a means of control.

Once the SCR is conducting, the gate current can be removed and the SCR remains on until the (anode)
current goes below the holding current level, which is by conventional treated as zero in circuit analyses.

Three types of phase-controlled rectifiers [Ras14, Sect. 10.1]:

• Semiconverter: One-quadrant converter with unipolar ( in figure below) and output current.
Not of interest.
• Full converter: Two-quadrant converter with bipolar but unipolar output current. Covered in this
lecture and the next.
⚠ can be bipolar, but its DC component remains positive.
• Dual converter: Four-quadrant converter with bipolar and output current. Covered in the next lecture.
⚠ Bipolar AC. Output current flows in one direction for motoring, another direction for regeneration.

Consider the full-converting phase-controlled rectifier below:

Lecture 8 Page 6
Suppose the SCR is fired at time , where is called the delay angle [Har11] or firing angle [Trz16].

Rectifiers like this where the control variable is the (firing) phase angle are called phase-controlled
rectifiers.

Using the technique as before, except instead of using the initial condition , which no longer applies,
we use the boundary condition , it is straightforward (see accompanying Live Script) to get
(1.3)
sin sin , ,
Compared to (1.1), only the exponent has changed.

The extinction angle in (1.3) satisfies

(1.4)
sin sin
Compared to (1.2), only the exponent has changed.

How long the conduction lasts is measured by the conduction angle, defined as .

For an example, see the full-wave case in Example 3.1.

Lecture 8 Page 7
2. Full-wave
uncontrolled
rectification
Monday, 10 May, 2021 9:23 PM

The objective of a full-wave rectifier is to produce a voltage or current that is purely DC or has some specified
DC component.

Full-wave rectifiers have some fundamental advantages

over half-wave rectifiers:
• The average current in the AC source is zero in the
full-wave rectifier, thus avoiding problems associated
with nonzero average source currents, particularly in
transformers.
• The output of the full-wave rectifier has inherently
less ripple than the half-wave rectifier.

Discussion of full-wave rectification is divided into four sections, i.e., Sections 2-3, starting with uncontrolled,
single-phase, full-wave rectification.

Let us start with the basic circuit:

Lecture 8 Page 8
Characteristics:
• Diodes and conduct together, and and
conduct together.

• By KVL, we can tell and cannot conduct at

the same time, and and cannot conduct at
the same time.

• When and are on, , . When

and are on, , .

• can be positive or zero but can never be

negative.

• The RMS source current (denoted , ) is the

same as the RMS load current (denoted , ).
The squares of the load and source currents are
the same, so the RMS currents are equal.

• Since two periods of the output occur for every

period of the input, the fundamental frequency of
the output voltage is , where is the frequency
of the AC input .

• The Fourier series of the output consists of a DC

term, and the even harmonics of the source
frequency.

For our analysis, let us consider resistive-inductive load.

The full-wave rectified sinusoidal voltage across the load

can be expressed as a Fourier series consisting
of a DC term and the even harmonics (see Tutorial 8 for
derivation):
(2.1)
cos
, ,

The current through the RL load is the

superposition of the currents associated with
each frequency.

Applying phasor analysis,

Lecture 8 Page 9
 ,

Note:
• As increases, decreases, but the impedance
increases, so decreases rapidly with .
• This means the DC term and only a few, if any,
of the AC terms are usually necessary to
describe current in an RL load.

The RMS current through the load is thus

, ,

Example 2.1
This example was adapted from [Har11, EXAMPLE 4-1]. The bridge rectifier circuit of has an AC source with
V at 60 Hz and a series RL load with and mH.
(a) Determine the average current in the load.
(b) Estimate the peak-to-peak variation in load current based on the first AC term in the Fourier series.
(c) Determine the average power absorbed by the load and the power factor of the circuit.
(d) Determine the average and RMS currents in the diodes.

Solution: Detailed solution is provided in the accompanying Live Script.

(a) The DC load voltage is and the DC load current is = 6.3662 A.

(b) The first AC term has magnitude

So, peak-to-peak variation in is 6.7776 A.

(c) A. (Quiz: Based on the Live Script, why did the computation stop
at and not extend to higher-order terms?) The average power W, and thus the
power factor is

(d) Since each diode only carries the current half of the time, average diode current is = 3.1831 A. For
the same reason, RMS diode current is

A widely used configuration involves the use of an output capacitor; see the output stage of the current-fed
resonant full-bridge boost DC/AC/DC Converter below [JC08].

Lecture 8 Page 10
Main advantages of indirect and isolated DC/AC/DC converters [JC08]:
• Output filter capacitors are less bulky and less expensive than output filter inductors for high-voltage
applications [CLL05].
• Output rectifier diodes can be operated in a discontinuous conduction mode with zero-current
switching, minimizing switching losses.

Applications: battery chargers and dischargers, uninterruptible power systems, alternative energy systems,
hybrid electric vehicles, and medical X-ray imaging [JC08].

Thus, let us consider the full-wave rectifier with a capacitive output filter below:

The capacitor smoothens out the output voltage.

Take as the angle where a diode pair becomes reverse-

biased and turns off, then for the first cycle,

sin ,
sin e ,

• The expression containing an exponential above can be obtained by solving the first-order differential
equation associated with a parallel RC circuit. Derivation omitted here.

• The angle is when another diode pair becomes forward-biased and turns on.

Let us determine the voltage ripple , but first we need to determine , , .

Having an expression relating to circuit parameters , allows us to determine what , should be,
given .

Lecture 8 Page 11
Let us start by determining . Define

sin , sin e

At , clearly we have , but continuity in rate of change also tells us

Since
cos , sin e ,

we have

cos sin t n
We must be careful when taking arctan, because a negative tangent can be in the second or fourth quadrant,
but we know this tangent is in the second quadrant. Therefore,
(2.2) t n ,

Now, let us tackle by applying the same technique as before. Clearly, we must have
, i.e.,
sin sin sin sin e
In the absence of a closed-form solution to the equation above, we can only say is the solution to
(2.3)
sin sin e

Once we have and , the voltage ripple is just

(2.4) sin sin
In practical circuits with large output capacitance, i.e., , , , in which case,
e ,
and
(2.5)
e

Above, we used the well-known approximation that when is small e

.

Example 2.2
This example was adapted from [Har11, EXAMPLE 4-4]. (a) From (2.2), t n ,
A full-wave rectifier with a capacitive output filter has a rad. By solving (2.3):
120 Vrms source at 60 Hz. Additionally, and
. sin sin e ,
(a) Determine the output voltage ripple. we get rad. Now, applying (2.4) gives
(b) Determine the value of for reducing the output us
voltage ripple to 1% of the DC value. sin

Solution: Given , , , (b) Given , the low ripple implies a

, the plan is to find , , and then based on (2.4). high and hence small , making (2.5)
When is given and is small, can be estimated from
Lecture 8 Page 12
 high and hence small , making (2.5)
When is given and is small, can be estimated from applicable. So,
(2.5). Detailed solution is provided in the accompanying
Live Script.

A related rectifier design to the rectifier just discussed is the voltage doubler.

The voltage doubler [Har11, Figure 4-7(a)]: Dual-voltage rectifier [Har11, Figure 4-1(b)]:

• charges to through when . • The full-wave rectifier with a capacitive output

• charges to through when . filter can be combined with the voltage doubler,
as shown above.
• is the sum of the capacitor voltages.
• When switch is open, the circuit is similar to the
• Useful when the output voltage of a rectifier
full-wave rectifier, with output at approximately
must be larger than the peak input voltage.
if the capacitors are large.
• Removes the need for a transformer to step up
• When switch is closed, the circuit acts as the
voltage, saving volume, weight and cost.
voltage doubler:
○ Capacitor charges to through when
.
○ Capacitor charges to through when
.
○ Diodes and remain reverse-biased
in this mode.
○

Applications of the voltage doubler:

• Circuits for catering to two voltage standards, e.g., 240 V for Australia and 120 V for the US.
• Output stage of a more advanced converter, e.g., the power-factor-corrected boost rectifier below [JJ07].

Topological stages for :

Lecture 8 Page 13
Without derivation [JJ07], , , which is double that of the classical boost converter; in
this sense, this circuit acts as a voltage doubler.

Lecture 8 Page 14
3. Full-wave
controlled
rectification
Monday, 10 May, 2021 9:23 PM

Clearly, being able to control the output of a full-wave rectifier has many advantages, including tracking the
desired voltage level.

Substituting the diodes with SCRs gives us a control mechanism based on the delay angle of each SCR.

Most common topology: bridge rectifier [Ras14,

p. 107]

• SCRs and will become forward-biased when

, but will not conduct until gate signals are
applied.

• Similarly, and will become forward-biased

when , but will not conduct until they receive
gate signals.

• The delay angle, , is the angle interval between the

forward biasing of the SCR and the gate signal
application.

• If the delay angle is zero, the rectifiers

behave exactly as uncontrolled rectifiers with
diodes.

Alternative topology: center-tapped transformer

rectifier

• is forward-biased when , and is

forward-biased when , but each will not
conduct until it receives a gate signal.

Pros and cons of bridge rectifier and center-tapped transformer rectifier [Ras14, TABLE 3.1]:

Lecture 8 Page 15
Pros and cons of bridge rectifier and center-tapped transformer rectifier [Ras14, TABLE 3.1]:

Quiz
mong “semiconve te ”, “ ull conve te ” nd “du l conve te ”, wh t ty e o conve te e the b idge
converter and center-tapped transformer converter?

Load current for a controlled full-wave rectifier with an

RL load can be either continuous or discontinuous, and a
different analysis is applicable to each.

Let us consider discontinuous current, then continuous

current.

Discontinuous conduction mode (DCM): Analysis of

the controlled full-wave rectifier operating in the DCM
is identical to that of the controlled half-wave rectifier
(see Sec. 1.2) except that the period for the output
current is rather than rad.

When , SCRs and in the bridge rectifier are

forward-biased, while and are reverse-biased.

At , and are fired and become conducting,

causing .

At , where , , and turn off

and become blocking.

Between angles and , (1.3) is applicable: Between and , remains at

sin sin zero.

At , and are fired and become

can be determined from (1.4). conducting.
Lecture 8 Page 16
 sin sin zero.

At , and are fired and become

can be determined from (1.4). conducting.

Example 3.1
This example was adapted from [Har11, EXAMPLE (c) The average load current is
4-7]. A controlled full-wave bridge rectifier has a
source of 120 Vrms at 60 Hz. Additionally, ,
mH, .
(a) Determine if the rectifier is operating in DCM. ⚠ is half of the period of the input!
(b) Determine an expression for the load current.
(c) Determine the average load current. (d) The power absorbed by the load is
(d) Determine the power absorbed by the load.
(e) Make a plot of power factor against , . W

Solution: Due to the computational nature of this

problem, detailed solution exists in the accompanying (e) Making a plot of power factors for a range of
Live Script. The following is an outline of the solution. del y ngles equi es us to w ite “ o ” loo to
Based on the given data, we can straightaway write repeat the calculations above for the range of angles.
Done correctly, the plot should look like
down , , ,
t n , .

(a) To determine if the rectifier is operating in DCM,

we need to check if the extinction angle is less than
Solving (1.4), we get rad or
216.4169 , which is less than , so the
rectifier is operating in DCM.

(b) We have all the ingredients we need to assemble

an expression for the current:
sin sin
sin , The plot shows increasing the delay angle worsens
where . the power factor.

What we can learn from the previous example: while being controllable, phase-controlled rectifiers have
poor power factors. The next lecture will introduce better rectifiers.

Lecture 8 Page 17
Continuous conduction mode (CCM): Load current is
still positive at , when and are turned
on and and are forced off.

Since the initial condition for current in the second

half-cycle is not zero, the current function does not
repeat, i.e., (1.3) does not apply.

So that CCM occurs, , and , or

equivalently, from (1.4),
sin sin
sin sin
sin
sin
We know and t n ,
are in the first
quadrant, so . The inequality
(a negative number) is not useful, so we use
the inequality instead, i.e.,
(3.1) t n ,
Either or (3.1) can be used as a criterion for
the CCM.

Without derivation (see [Har11, pp. 136-137]), the

output voltage can be expressed as Fourier series:
(3.2)
cos cos ,
, ,
where An expression for is not given, because it is not
needed for our purpose.
,
For calculation of the RMS current, power and
cos cos
, power factor, the technique of Sec. 2 is applicable.

sin sin For an example, refer to Tutorial 8.

Lecture 8 Page 18
References
Friday, 14 May 2021 4:01 PM

[CLL05] R. Y. Chen, R. L. Lin, T. J. Liang, J. F. Chen and K. C. Tseng, Current-fed full-bridge boost converter
with zero current switching for high voltage applications, Fourtieth IAS Annual Meeting.
Conference Record of the 2005 Industry Applications Conference, 2005, pp. 2000-2006 Vol. 3,
doi: 10.1109/IAS.2005.1518722. Available at [Link]
[Link]/abstract/document/1518722
[CM10] F. Chen and D. Maksimović, Digital Control for Improved Efficiency and Reduced Harmonic
Distortion Over Wide Load Range in Boost PFC Rectifiers, IEEE Transactions on Power
Electronics, 25(10):2683-2692, Oct. 2010, doi: 10.1109/TPEL.2010.2050702. Available at
[Link]
[Har11] D. W. Hart, Power Electronics, international ed., McGraw-Hill Education, 2011.
[JC08] S. Jalbrzykowski and T. Citko, Current-Fed Resonant Full-Bridge Boost DC/AC/DC Converter,
55(3):1198-1205, Mar 2008. More than 100 Google Scholar citations. Available at
[Link]
[JJ07] Y. Jang and M. M. Jovanovic, Interleaved Boost Converter With Intrinsic Voltage-Doubler
Characteristic for Universal-Line PFC Front End, IEEE Transactions on Power Electronics,
22(4):1394-1401, July 2007, doi: 10.1109/TPEL.2007.900502. Available at [Link]
[Link]/document/4267757
[Kre15] P. T. Krein, Elements of Power Electronics, 2nd ed., Oxford University Press, 2015.
[LY18] F.L. Luo and H. Ye, Power Electronics: Advanced Conversion Technologies, 2nd edition, CRC Press,
2018. Available at [Link]
ye/10.1201/9781315186276
[Ons20] ON Semiconductor, 6.6 kW On Board EV Charger Reference Design, reference design
TND6320/D, July 2020. Available at [Link]
[Ras14] M. H. Rashid, Power Electronics: Devices, Circuits, and Applications, Pearson Education Limited,
2014.
[Trz16] A. M. Trzynadlowski, Introduction to modern power electronics, 3rd ed., John Wi-
ley & Sons, 2016

Lecture 8 Page 19