Isolated Switch-Mode DC/DC Converters DC Power Supply Control
Isolated Switch-Mode DC/DC Converters DC Power Supply Control
17Jun13
The dots used on the primary and secondary windings define the
relationship between the voltage polarities and current directions on
two sides of the transformer.
The model of an ideal 2-winding transformer and the relationship
between input and output voltages and currents are:
v1 i2 N1
v2 i1 N 2
17Jun13
17Jun13
Flyback Converter-1
[1]
Flyback Converter
Buck-Boost Converter
17Jun13
Flyback Converter-2
Lets assume the components are ideal, the transformer has a
magnetizing inductance of Lm, the circuit is at steady-state, and C is
large enough for the output voltage to be considered constant.
Flyback Converter
[1]
Flyback Converter-3
When the switch is turned ON,
v1 Vi Lm
I Lm
diLm
dt
on state
diLm
I Lm
dt
on state
I Lm
dTs
Vi
Lm
dTsVi
Lm
[1]
17Jun13
Flyback Converter-4
v1 Vi v2
N2
Vi
N1
i2 0 i1
N2
i2 0 ii iLm
N1
[1]
Flyback Converter-5
Flyback Converter
[1]
10
Flyback Converter-6
[1]
11
Flyback Converter-7
When the switch is turned OFF,
v2 Vo v1
N1
Vo
N2
diLm
diLm
N1
Vo Lm
N2
dt
dt
I Lm
off state
I Lm
off state
I Lm
(1 d )Ts
N1 Vo
N 2 Lm
(1 d )TsVo N1
Lm
N2
[1]
12
Flyback Converter-8
When the switch is turned OFF,
iLm i1
iD i2
I D Io
N1
N
i1 1 iLm
N2
N2
Vo
R
vS Vi v1 Vi
iC iD I o
N1
Vo
N2
V
N1
iLm o
N2
R
[1]
Topological Mode II: Switch OFF
17Jun13
13
Flyback Converter-9
Output-to-Input Voltage Ratio
on state
I Lm
off state
I Lm
dTsVi (1 d )TsVo N1
V
d N2
o
Lm
Lm
N2
Vi 1 d N1
[1]
Flyback Converter
17Jun13
14
Flyback Converter-10
Other Relations for Flyback Converter
I i I S dI Lm
I Lm
Po V 2
o
2
2
I i Vi
V
V
dV
N
N
o
i
2
2
R o
2
d
d dVi dVi R (1 d ) R N1 (1 d ) R N1
I Lm ,max I Lm
I Lm ,min I Lm
I Lm
2
I Lm
2
I C I D I 2
N1
N
I1 1 I Lm ,max
N2
N2
[1]
Flyback Converter
17Jun13
15
Flyback Converter-11
Output Voltage Ripple and Capacitor Design
Due to the fact that the output stage of flyback converter is the same
as that of buck-boost converter, the formula for output voltage ripple
will be the same as that for buck-boost converter.
Vo
d
Vo
RCf s
if I Lm ,min
N1
d
ID. C
Vo
N2
Rf s
Vo
N1
I D , a different formula has to be derived for
N2
N1
I o (1 d )
I Lm ,max
N
Vo
N
2
, if I Lm ,min 1 I D .
N
Vo
N2
2I Lm 1 f s CVo
N2
17Jun13
16
Example
The flyback converter shown below has the following parameters:
Vi = 36V
N1:N2 =5:2
Lm = 400H
C = 150F
fs = 30kHz
R = 10
Vo= 12V
Flyback Converter
17Jun13
[1]
17
Solution
The duty ratio can be found as follows:
Vo
V N
d N2
d
o 1
Vi 1 d N1
Vi N 2 1 d
Vo N1
Vi N 2
1
1
d
0.455
36 2
Vo N1
Vi N 2
1
1
1
12 5
Vi N 2
Vo N1
The magnetizing inductance current average, minimum and
maximum values can be found in the following way:
I Lm
Vo2
122
0.879 A
dVi R 0.455 36 10
I Lm
dTsVi
dVi
0.455 36
1.365 A
3
6
Lm
Lm f s 400 10 30 10
1.365
1.56 A
2
2
I Lm
1.365
0.879
0.197 A 0 continuous iLm
2
2
I Lm ,max I Lm
I Lm ,min I Lm
17Jun13
I Lm
0.879
18
Solution (Cont.)
The output voltage ripple calculation starts with a check on the
validity of the conventional formula.
N1
5
I
0.197
0.493 A
Lm ,min N
2
2
V 12
I D I o o 1.2 A
R 10
N1
I Lm ,min
I D Conventional formula is not valid.
N2
2
N1
5
I o (1 d )
I Lm ,max
1.56 1.2 (1 0.455)
N2
Vo
2
5
N
Vo
2 1.365 30 103 150 106 12
2I Lm 1 f s CVo
2
N2
0.011 1.1%
17Jun13
19
Solution (Cont.)
The setup in PSIM will look like the following:
17Jun13
20
Solution (Cont.)
The simulation results are as follows:
17Jun13
21
Solution (Cont.)
The simulation results are as follows:
I D Io
1.2 A
Vo
0.0109
Vo
1.1%
17Jun13
22
Solution (Cont.)
The simulation results are as follows:
17Jun13
23
Flyback Converter-12
PROS
Single-switch, no output inductor
Single-ended topology (Emitter or Source of switch at the same
potential as the input voltage negative terminal.)
Simplest circuit, low-cost
Low turn-on loss (Switch current rises gradually at a rate controlled
by Lm.)
CONS
High output capacitor ripple current
High turn-off loss (Switch voltage rises quickly to a level dictated
by input and output voltages and transformer turns-ratio.)
Unidirectional core excitation, poor transformer utilization
High switch and diode voltage stresses
[2]
Flyback Converter
17Jun13
24
Flyback Converter-13
Discontinuous Mode of Operation
1st-order system (easy to stabilize)
Better utilization of magnetics
High output capacitor ripple current
Continuous Mode of Operation
2nd-order system
Right-Half-Plane (RHP) zero
Difficult to stabilize
Flyback Converter
17Jun13
[2]
25
[2]
1-Switch Forward Converter
(derived from buck converter)
17Jun13
26
[2]
17Jun13
27
[2]
PROS
Low switch voltage stress
Simple circuit
Low output ripple
No reset winding needed
Higher power capability than 1-switch forward converter
CONS
Large output inductor
Unidirectional core excitation, poor transformer utilization
Switching losses
Double-ended topology; driving S1 and S2 needs two isolated
power supplies, since the emitters (sources) of the two switches
are not at the same potential. Also, isolation between control
circuit and power circuit is necessary.
17Jun13
28
[2]
2-Switch Forward Converter
17Jun13
29
PROS
Push-Pull Converter
CONS
[2]
Push-Pull Converter
17Jun13
30
[2]
Half-Bridge DC-DC Converter
PROS
CONS
31
[2]
Full-Bridge DC-DC Converter
PROS
CONS
32
17Jun13
33
[1]
A 4-output DC power supply based on flyback converter topology
17Jun13
34
[1]
17Jun13
35
110V
AC
EMI
Filter
Fuse
Diode
Rectifier
Thermistor
LPF
Thermistor
4-output
Flyback
Converter
+12V
Regulator
+12V
+5V
Regulator
+5V
-5V
Regulator
-5V
-12V
Regulator
-12V
[3]
17Jun13
36
110V
AC
EMI
Filter
Fuse
Diode
Rectifier
Thermistor
Thermistor
LPF
4-output
Flyback
Converter
+12V
Regulator
+12V
+5V
Regulator
+5V
-5V
Regulator
-5V
-12V
Regulator
-12V
Bleeder
Resistors
Coupled
Inductors
[3]
17Jun13
37
110V
AC
EMI
Filter
Fuse
Diode
Rectifier
Thermistor
LPF
Thermistor
4-output
Flyback
Converter
+12V
Regulator
+12V
+5V
Regulator
+5V
-5V
Regulator
-5V
-12V
Regulator
-12V
[3]
17Jun13
38
110V
AC
EMI
Filter
Fuse
Diode
Rectifier
Thermistor
Thermistor
LPF
4-output
Flyback
Converter
+12V
Regulator
+12V
+5V
Regulator
+5V
-5V
Regulator
-5V
-12V
Regulator
-12V
[3]
17Jun13
39
110V
AC
EMI
Filter
Fuse
Diode
Rectifier
Thermistor
4-output
Flyback
Converter
LPF
Thermistor
+12V
Regulator
+12V
+5V
Regulator
+5V
-5V
Regulator
-5V
-12V
Regulator
-12V
[3]
17Jun13
40
110V
AC
EMI
Filter
Fuse
Diode
Rectifier
Thermistor
Thermistor
LPF
4-output
Flyback
Converter
+12V
Regulator
+12V
+5V
Regulator
+5V
-5V
Regulator
-5V
-12V
Regulator
-12V
[3]
17Jun13
41
[3]
17Jun13
42
[1],[4],[5]
A closed-loop control is required to compensate for components nonidealities and variations in the source voltage and load power, and
deliver regulated output voltages at the desired levels.
Feedback control loops are designed to satisfy conflicting
requirements of speed of response, accuracy of regulation and
stability.
It is desired to start from a set of desired specifications and reach at
exact circuit parameters and component values that satisfy those
specifications.
In the discussions to follow, a buck converter operated in continuousconduction mode is used to illustrate the basic concepts and common
techniques involved in the systematic design of closed-loop controls
for DC power supplies.
17Jun13
43
17Jun13
44
Concept of Stability-1
Amplifier vcontrol
Vo ,ref
GA ( s)
Vo , fbk
PWM
GPWM ( s )
Sensor
Gsensor ( s )
Control Block Diagram
17Jun13
45
Concept of Stability-2
Based on the block diagram of the system, the following closed-loop
transfer function can be derived.
GF ( s )
1 GLoop ( s )
GF ( s ) GA ( s ) GPWM ( s ) GC ( s )
GLoop GF ( s ) Gsensor ( s )
Amplifier vcontrol
Vo ,ref
GA ( s)
Vo , fbk
PWM
GPWM ( s )
Sensor
Gsensor ( s )
Control Block Diagram
17Jun13
46
Stability Indices-1
Gloop ( s )
@ f cross over
(180 )
Bode Diagram
20
0db
0
-20
-40
-60
-80
0
Phase (deg)
-45
Bode Plots
-90
-135
Phase
Margin
-180
-2
10
17Jun13
-1
10
10
Frequency (rad/sec)
10
10
47
Stability Indices-2
Gain Margin is defined as the difference between unity (or 0db)
and the overall loop gain at the frequency where the total phase shift
of the blocks around the loop is -180.
Magnitude (dB)
0.001s 2 20
Gloop ( s ) 2
s s 0.25 0
@ Loop 180
Bode Diagram
-20
Gain
Margin
-40
-60
-80
0
Phase (deg)
-45
Bode Plots
-90
-135
-180
-2
10
17Jun13
-1
10
10
Frequency (rad/sec)
10
-180
10
48
Concept of Stability-3
To stabilize a loop, one can reduce the amplifier gain so that the
loop gain cross-over occurs at a frequency well below where phase
shifts from reactive elements and time delays become significant.
Problem: unacceptable speed of transient response!
Desired: a cross-over frequency of as high as possible while
keeping a good phase margin.
Solution: Tailoring the frequency response of the Amplifier to
compensate for some of the Modulator phase shift in the region of
gain cross-over.
Small-Signal Linear Model is used to design the compensator
based on the given performance specifications in time and
frequency domains.
17Jun13
49
vo
Vi
1 sRC C
GC ( s )
( s)
1
1
L C 2 RC
d
s
s
L
RC
LC
Note that this is the same as the linear model derived for buck
converter, except for the fact that the inductors series resistance has
been ignored and the load resistance has been considered.
Buck Converter
17Jun13
LPF
50
d control GPWM ( s )
vcontrol Vtri
Vtri
Note that in PWM, using a sawtooth waveform as the carrier signal,
if the peak value of the triangular signal is equal to 1, control signal
will represent duty ratio.
Vtri
0
vcontrol
t
ton AC BC vcontrol
d
Ts AE DE
Vtri Vi
vtri
ON
vo ,unfiltered
OFF
t
ON
Vo
0
17Jun13
toff
ton
Ts 1/ f s
OFF
51
Bode Diagram
60
3 0 0 0 0 s + 4 .3 0 4 1 0 8
(s)
s 2 1, 3 5 9 s 1 .4 3 5 1 0
Magnitude (dB)
40
20
0
-20
Vi
1 sRC C
GC ( s )
R
1
1
LC 2 C
s
s
LC
L RC
-40
0
Phase (deg)
-45
-90
-135
-180
2
10
10
10
10
10
Frequency (rad/sec)
52
Type 1 Amplifier
GA ( s)
10
Magnitude (dB)
5
0
-5
-10
-15
-89
Phase (deg)
-89.5
-90
-90.5
-91
0
10
10
Frequency (rad/sec)
53
1
R1C1 s
2
s
5
Magnitude (dB)
GA ( s)
GA ( s)
0
-5
-10
-15
-89
Phase (deg)
-89.5
-90
-90.5
-91
0
10
10
Frequency (rad/sec)
54
30
Magnitude (dB)
GA ( s )
Bode Diagram
40
20
10
0
-10
Phase (deg)
-20
0
-45
-90
2
10
17Jun13
10
10
10
Frequency (rad/sec)
10
10
55
GA ( s)
Bode Diagram
40
30
Magnitude (dB)
1
s
R2 C1
GA ( s)
C C2
R1C2 s s 1
R
C
C
2 1 2
20
10
0
-10
Phase (deg)
-20
0
-45
17Jun13
10
10
10
10
Frequency (rad/sec)
10
10
56
Magnitude (dB)
20
0
-20
-40
Phase (deg)
-60
45
-45
-90
0
10
10
10
10
10
10
Frequency (rad/sec)
17Jun13
57
Magnitude (dB)
1
1
s
s
( R1 R3 )C3
R2 C1
R1 R3
GA ( s )
R1 R3C2
C C2
1
ss 1
s
R2 C1C2
R3C3
40
20
0
-20
-40
Phase (deg)
-60
45
-45
-90
0
10
10
10
10
10
10
Frequency (rad/sec)
17Jun13
58
40
Magnitude (dB)
20
-20
-40
Phase (deg)
-60
45
-45
-90
0
10
17Jun13
10
10
10
Frequency (rad/sec)
10
10
59
Idea of K Factor-1
fp
fz
(Type 2 Amplifier), K
fp
fz
(Type 3 Amplifier)
Type 1 Amplifier
Type 1 Amplifier is a special case of Type 2 Amplifier, where fp and fz
coincide and cancel each other resulting in a K equal to 1.
Bode Diagram
10
Magnitude (dB)
5
0
-5
-10
-15
-89
Phase (deg)
-89.5
-90
-90.5
-91
0
10
17Jun13
10
Frequency (rad/sec)
60
Idea of K Factor-2
Type 2 Amplifier
In Type 2 Amplifier, the loop cross-over frequency is chosen at the geometrical
mean of the pole-zero pair frequencies for maximum phase boost. Therefore,
fp
K
fz
f c f p f z
f p Kf c
fc
f
z K
The amplifier is designed such that loop cross-over frequency occurs at fc.
The larger the K, the larger the distance between fp and fz and the larger the
phase boost.
Bode Diagram
40
Magnitude (dB)
30
Type 1: Reference
20
10
0
K fc
fc
fc / K
-10
Phase (deg)
-20
0
-45
10
17Jun13
10
10
10
Frequency (rad/sec)
10
10
61
Type 3 Amplifier
Idea of K Factor-3
fp fp
fp
K
fz fz
fz
f c f p f z
f p K fc
fc
f
z
K
The amplifier is designed such that loop cross-over frequency occurs at fc.
The larger the K, the larger the distance between fp and fz and the larger the phase boost.
Bode Diagram
40
Magnitude (dB)
20
Type 1: Reference
0
-20
-40
fc / K
fc
K fc
-60
45
Phase (deg)
-45
-90
0
10
10
10
10
10
10
Frequency (rad/sec)
17Jun13
62
Derivation of K Factor-1
Type 2 Amplifier
z tan 1
f
f
, p tan 1
fz
fp
Boost z p tan 1
fc
f
f
f
1
tan 1 c tan 1 c tan 1 c tan 1 K tan 1
fc
fz
fp
Kf c
K
K
From trigonometry,
63
Derivation of K Factor-2
Type 3 Amplifier
Type 3 Amplifier has two coincident zero-pole pairs. The phase boost due to each zeropole pair is given by:
1
K
For two coincident zero-pole pairs, the phase boost is two times that for one zero-pole
pair, i.e.,
1
Boost 2 tan 1 K tan 1
17Jun13
64
Using K Factor-1
Step 1: Make Bode Plots of Modulator
This can be done in two ways:
By Analysis: The small-signal model of modulator can be used to plot the Bode
diagrams. This method is not accurate, as some parasitic effects are not included
in the model.
By Measurement: This method is preferred. The Bode plots are created by a
special type of equipment called Frequency Response Analyzer. A small
sinusoidal signal superimposed on a dc signal is used as the duty ratio for the
converter switch. As the frequency of the ac signal is varied over a wide range,
the magnitude and phase angle of the ratio of the ac component of the output
voltage to the ac component of duty ratio is measured at each frequency. The
data thus collected will be used to create the Bode plots of the converter.
65
Using K Factor-2
Step 3: Choose the Desired Phase Margin
A phase margin of 90 means that your system is stable as a rock.
Phase margin of 60 is a compromise between fast transient response
and stability.
Phase margins of 30 or less cause the system to have substantial
ringing when subjected to transients and little tolerance for
component or environmental variations.
In terms of dB,
Amplifier Gain ( Modulator Gain Voltage sensor Gain PWM Controller Gain)
17Jun13
66
Using K Factor-3
Step 5: Calculate the Required Phase Boost
From the definition of Phase Margin,
PHASE MARGIN M ph (Modulator Amplifier ) (180 )
Therefore,
M ph (Modulator Amplifier ) (180 ) [Modulator (90 Boost )] (180 )
Modulator Boost 90
Boost M ph Modulator 90
17Jun13
67
Using K Factor-4
Step 6: Choose an Amplifier Type
Type 1 Amplifier: This amplifier is the simplest in structure and
requires the lowest number of parts, but offers no phase boost. It is
proper to use this amplifier for the cases where the loop is crossed
over before the resonant frequency of the LC filter.
Type 2 Amplifier: This amplifier is used where the required phase
boost is below 90. It is most practical where the required phase
boost is less than about 70. Note that a large boost calls for a large
K, and for a large K one has to pay a penalty on the low-frequency
gain. Type 2 Amplifier is used where the Modulator gain is falling
off at about a -20db/decade and the phase shift is about -90. This is
the case in current regulators, or in voltage regulators above the
frequency of the ESR zero of the main filter capacitor.
Type 3 Amplifier: This amplifier is used where the required phase
boost is below 180. It offers the largest boost for a given K factor.
The drawback is using the highest number of parts. Type 3 Amplifier
outperforms Type 1 and Type 2 Amplifiers in the sense that it offers
the highest low-frequency gain and lowest high-frequency gain for a
given cross-over frequency and phase margin.
17Jun13
68
Using K Factor-5
Step 7: Choose a Value for R1
17Jun13
69
Using K Factor-6
Step 7: Choose a Value for R1 (Cont.)
17Jun13
70
Using K Factor-7
Step 8: Calculating the Amplifier Component Values
Assumptions:
Type 1 Amplifier
K 1
C1 1/(2 f c GA R1 )
Type 1 Amplifier provides no phase boost. The phase margin can be found as follows:
Boost M ph Modulator 90
Boost 0 M ph Modulator 90
17Jun13
71
Using K Factor-8
Step 8: Calculating the Amplifier Component Values
(Cont.)
Type 2 Amplifier
K tan[( Boost / 2) 45 ]
GA ( s )
k A (1 s / z )
s (1 s / p )
C2 1/(k A K 2 R1 )
C1 C2 ( K 2 1)
R2 K /(2 f c C1 )
17Jun13
72
Using K Factor-9
Step 8: Calculating the Amplifier Component Values
(Cont.)
Type 3 Amplifier
K tan[( Boost / 4) 45 ]
C2 1/(k A K R1 )
k A (1 s / z ) 2
GA ( s)
s (1 s / p ) 2
C1 C2 ( K 1)
R2 1/( z C1 )
R3 R1 /( K 1)
C3 1/( p R3 )
73
Using K Factor-10
Optimization of Amplifiers
Type 1 Amplifier
Type 2 Amplifier
There is nothing that can be done to optimize a Type 2 Amplifier. With the K
factor, the gain and the locations of the zero and pole are determined for a
particular operating point. There is nothing that can be done about the
performance at other operating points.
Type 3 Amplifier
In Type 3 Amplifier, it was assumed that the two zero-pole pairs were
coincident. This results in a pointy phase bump at the cross-over frequency
leading to a large enough phase margin.
Spreading zeros and poles tends to broaden and flatten the phase bump,
reducing the phase boost at cross-over frequency and the phase margin.
Optimum performance, i.e., largest phase margin for smallest K factor, is
obtained when zeros are poles are coincident.
In cases, where a wide range of variations for input voltage, load and
temperature is expected, Modulator transfer function experiences wide
variations. In such cases, it is wise to sacrifice optimum performance at a
certain operating point to gain satisfactory performance over a wide
operating range.
17Jun13
74
Example
17Jun13
75
Solution
vo
Vi
1 sRC C
GC ( s)
(s)
1
1
L C 2 RC
d
s
s
LC
L RC
G
3 0 0 0 0 s + 4 .3 0 4 1 0 8
(s)
s 2 1, 3 5 9 s 1 .4 3 5 1 0
Bode Diagram
24.66db
40
Magnitude (dB)
20
-20
-40
0
Phase (deg)
-45
-90
-135
-180
2
10
17Jun13
10
10
1kHz
Frequency (rad/sec)
10
138
10
76
Solution (Cont.)
Step 2: Cross-Over Frequency
Choose fc=1kHz which is larger than the resonant frequency
of the LC filter (fr,LC=602.84 Hz). In this way, the loop phase
angle will be larger than -180 at all frequencies.
Step 3: Phase Margin
Mph=60
With this phase margin, the closed-loop system settles
quickly without oscillations.
Step 4: Required Amplifier Gain at fc
GC @ fc = 24.66 dB = 17.1
Therefore, at fc,
GAmplifier
17Jun13
GPWM
1
1
0.526
Gsensor GC 0.556 0.2 17.1
77
Solution (Cont.)
Step 5: Required Phase Boost
Type 1: Reference
0
-20
-40
fc / K
fc
K fc
-60
45
Phase (deg)
k A (1 s / z ) 2
GA ( s )
s (1 s / p ) 2
Magnitude (dB)
20
-45
-90
0
10
17Jun13
10
10
10
Frequency (rad/sec)
10
10
78
Solution (Cont.)
Step 7: Choose R1
Choose R1=100k.
Step 8: Find Amplifier Component Values
tan[(108 / 4) 45 ]
K tan[( Boost / 4) 45 ]
f p K fc
fc
z
K
k A (1 s / z ) 2
one can find kA in GA ( s )
s (1 s / p ) 2
9.47
From
to be
2
2
2
2
1 c / p
1 jc / p
k A GA ( s )
jc
G
s
(
)
A
c
2
2
@ fc
@ fc
1 jc / z
12 / 2
c z
1
2
1
1 c / p
9.47 348.99
GA ( s)
c
0.526
2
1000
2
@ fc
1 9.47
1 c / z
17Jun13
79
Solution (Cont.)
Step 8: Find Amplifier Component Values (Cont.)
From
C2 1/(k A K R1 )
C1 C2 ( K 1)
R2 1/( z C1 )
R3 R1 /( K 1)
C3 1/( p R3 )
This completes the loop design. Now, lets verify the design
through simulation of the circuit using PSIM.
17Jun13
80
Solution (Cont.)
PSIM Simulation
From Po=36W and Vo=12V, one can find the load resistance
as follows:
Vo2
Vo2 (12) 2
R
4
Po
R
Po
36
Lets try the transient response following steady-state
conditions by changing the load by 25%, i.e., from 4 to 3
and back to 4 .
17Jun13
81
Solution (Cont.)
PSIM Simulation (Type 3 Amplifier)
17Jun13
82
Solution (Cont.)
PSIM Simulation Results (Type 3 Amplifier)
17Jun13
83
Solution (Cont.)
PSIM Simulation (PI Controller)
17Jun13
84
Solution (Cont.)
PSIM Simulation Results (PI Controller)
17Jun13
85
References:
[1] D.W. Hart, Introduction to Power Electronics, Prentice
Hall, 1997.
or
D.W.Hart, Power Electronics, McGraw-Hill, 2011.
[2] Prof. P. Jains Course Notes (Concordia University, 1996)
[3] Horowitz and Hill, The Art of Electronics, Cambridge University
Press, 2nd Edition, 1994.
[4] H. Dean Venable (President of Venable Industries, Inc.*),
THE K FACTOR: A NEW MATHEMATICAL TOOL
FOR STABILITY ANALYSIS AND SYNTHESIS,
Proceedings of the tenth National Power Conversion Conference,
Powercon 10, 1983, pp. H1-1 to H1-12.
* Venable Industries offers Frequency Response Analysis Systems
based on Fourier Integral Analysis machines.
[5] N. Mohan, Internet-Based Short Course: Power Electronics
Principles for Practicing Engineers, Module 3: Designing Feedback
Control of DC-DC Converters and Soft-Switching, Aug. 20, 2004,
10:00-11:30 a.m., Central Time, University of Minnesota.
17Jun13
86