Chapter 9. Controller Design: 9.2. Effect of Negative Feedback On The Network Transfer Functions

Chapter 9.
Controller Design
9.1. Introduction
9.2. Effect of negative feedback on the network transfer
functions
9.2.1. Feedback reduces the transfer function from disturbances
to the output
9.2.2. Feedback causes the transfer function from the reference
input to the output to be insensitive to variations in the gains
in the forward path of the loop
9.3. Construction of the important quantities 1/(1+T) and
T/(1+T) and the closed-loop transfer functions
Fundamentals of Power Electronics 1 Chapter 9: Controller design

Controller design
9.4. Stability
9.4.1. The phase margin test
9.4.2. The relation between phase margin and closed-loop
damping factor
9.4.3. Transient response vs. damping factor
9.5. Regulator design
9.5.1. Lead (PD) compensator
9.5.2. Lag (PI) compensator
9.5.3. Combined (PID) compensator
9.5.4. Design example

Controller design
9.6. Measurement of loop gains

9.6.1. Voltage injection
9.6.2. Current injection
9.6.3. Measurement of unstable systems
9.7. Summary of key points

9.1. Introduction
Switching converter Load

Output voltage of a
+ iload(t) switching converter
depends on duty cycle
vg(t) + v(t) d, input voltage vg, and
–
load current iload.
–
Transistor
gate driver Switching converter
δ(t) δ(t) Pulse-width vc(t) v(t) = f(vg , iload , d)
modulator vg(t)
dTs Ts t
iload (t) } Disturbances
v(t)
d(t)
} Control input
The dc regulator application
Switching converter
Objective: maintain constant v(t) = f(vg , iload , d)
output voltage v(t) = V, in spite vg(t)
of disturbances in vg(t) and
iload(t).
Typical variation in vg(t): 100Hz
iload (t) } Disturbances
v(t)
or 120Hz ripple, produced by d(t)

rectifier circuit.
} Control input
Load current variations: a significant step-change in load current, such
as from 50% to 100% of rated value, may be applied.
A typical output voltage regulation specification: 5V ± 0.1V.
Circuit elements are constructed to some specified tolerance. In high
volume manufacturing of converters, all output voltages must meet
specifications.

The dc regulator application
So we cannot expect to set the duty cycle to a single value, and obtain
a given constant output voltage under all conditions.
Negative feedback: build a circuit that automatically adjusts the duty
cycle as necessary, to obtain the specified output voltage with high
accuracy, regardless of disturbances or component tolerances.

Negative feedback:
a switching regulator system
Power Switching converter Load

input
+
iload
vg + v
–
Sensor
– H(s) gain
Transistor Error
gate driver signal
δ Pulse-width vc G (s) ve –+ Hv
modulator c
Compensator
Reference
input vref
Negative feedback
Switching converter
v(t) = f(vg , iload , d)
vg(t)
Error
signal
ve(t)
iload (t)
vc Pulse-width d(t)
} Disturbances
v(t)
vref +– Compensator
modulator
} Control input
Reference
input
Sensor
gain

9.2. Effect of negative feedback on the
network transfer functions
Small signal model: open-loop converter

e(s) d(s) Le
buck converter
1 : M(D)
+
–
+
vg(s) + j(s) d(s) C v(s) iload (s)

– R
–
Output voltage can be expressed as
v(s) = Gvd(s) d(s) + Gvg(s) vg(s) – Z out(s) i load(s)
where
v(s) v(s) v(s)
Gvd(s) = Gvg(s) = Z out(s) = –
d(s) vg = 0 vg(s) d=0 i load(s) d=0
i load = 0 i load = 0 vg = 0

Voltage regulator system small-signal model
e(s)d(s) Le
1 : M(D)
+
–
• Use small-signal +
converter model
vg(s) + j(s)d(s) C v(s) iload (s)
– R
• Perturb and
linearize remainder –
of feedback loop:
d(s)
Error
vref (t) = Vref + vref (t) signal
vref (s) +– ve (s) vc (s) 1
Gc (s)
ve(t) = Ve + ve(t) Reference
VM
input Compensator Pulse-width
etc. modulator
H(s)v(s)
H(s)
Sensor
gain

Regulator system small-signal block diagram
iload (s) Load current

variation
vg (s) Zout (s)

Gvg(s)
ac line
variation
Pulse-width
Compensator modulator –
vref (s) + ve (s) vc (s) 1 d(s) + v(s)
– Gc(s) VM Gvd (s)
Reference Error Duty cycle + Output voltage
input signal variation variation
Converter power stage
H(s) v(s)
H(s)
Sensor
gain

Solution of block diagram
Manipulate block diagram to solve for v(s) . Result is
GcGvd / VM Gvg Z out

v = vref + vg – i load
1 + HGcGvd / VM 1 + HGcGvd / VM 1 + HGcGvd / VM
which is of the form

Gvg Z
v = vref 1 T + vg – i load out
H 1+T 1+T 1+T
with T(s) = H(s) Gc(s) Gvd(s) / VM = "loop gain"
Loop gain T(s) = products of the gains around the negative

feedback loop.

9.2.1. Feedback reduces the transfer functions
from disturbances to the output
Original (open-loop) line-to-output transfer function:

v(s)
Gvg(s) =
vg(s) d=0
i load = 0
With addition of negative feedback, the line-to-output transfer function

becomes:
v(s) Gvg(s)
=
vg(s) vref = 0 1 + T(s)
i load = 0
Feedback reduces the line-to-output transfer function by a factor of

1
1 + T(s)
If T(s) is large in magnitude, then the line-to-output transfer function

becomes small.

Closed-loop output impedance
Original (open-loop) output impedance:

v(s)
Z out(s) = –
i load(s) d=0
vg = 0
With addition of negative feedback, the output impedance becomes:
v(s) Z out(s)
=
– i load(s) vref = 0 1 + T(s)
vg = 0
Feedback reduces the output impedance by a factor of

1
1 + T(s)
If T(s) is large in magnitude, then the output impedance is greatly

reduced in magnitude.

9.2.2. Feedback causes the transfer function from the
reference input to the output to be insensitive to
variations in the gains in the forward path of the loop
Closed-loop transfer function from vref to v(s) is:

v(s) 1 T(s)
=
vref (s) vg = 0 H(s) 1 + T(s)
i load = 0
If the loop gain is large in magnitude, i.e., || T || >> 1, then (1+T) ≈ T and
T/(1+T) ≈ T/T = 1. The transfer function then becomes
v(s)
≈ 1
vref (s) H(s)
which is independent of the gains in the forward path of the loop.

This result applies equally well to dc values:
V = 1 T(0)
≈ 1
Vref H(0) 1 + T(0) H(0)

9.3. Construction of the important quantities
1/(1+T) and T/(1+T)
Example
80 dB
1 + ωs
z
|| T || T(s) = T0
QdB 1+ s + s 2
1 + ωs
60 dB | T0 |dB Qω p1 ω p1 p2
fp1
40 dB
– 40 dB/decade
20 dB
fz
– 20 dB/decade
0 dB
fc fp2
–20 dB Crossover – 40 dB/decade
frequency
–40 dB
1 Hz 10 Hz 100 Hz 1 kHz 10 kHz 100 kHz
At the crossover frequency fc, || T || = 1 f

Approximating 1/(1+T) and T/(1+T)
T ≈ 1 for || T || >> 1
1+T T for || T || << 1
1 for || T || >> 1
1 ≈ T(s)
1+T(s) 1 for || T || << 1

Example: construction of T/(1+T)
80 dB
T ≈ 1 for || T || >> 1
1+T T for || T || << 1
60 dB
fp1 || T ||
40 dB
Crossover
20 dB frequency
fz
– 20 dB/decade fc
0 dB
T fp2
–20 dB 1+T – 40 dB/decade
–40 dB
f
Example: analytical expressions for approximate
reference to output transfer function
At frequencies sufficiently less that the crossover frequency, the loop

gain T(s) has large magnitude. The transfer function from the reference
to the output becomes
v(s) T(s)
= 1 ≈ 1
vref (s) H(s) 1 + T(s) H(s)
This is the desired behavior: the output follows the reference
according to the ideal gain 1/H(s). The feedback loop works well at
frequencies where the loop gain T(s) has large magnitude.
At frequencies above the crossover frequency, || T || < 1. The quantity
T/(1+T) then has magnitude approximately equal to 1, and we obtain
v(s) T(s) T(s) Gc(s)Gvd (s)
= 1 ≈ =
vref (s) H(s) 1 + T(s) H(s) VM
This coincides with the open-loop transfer function from the reference
to the output. At frequencies where || T || < 1, the loop has essentially
no effect on the transfer function from the reference to the output.
Same example: construction of 1/(1+T)
80 dB
1 for || T || >> 1
60 dB | T0 |dB QdB 1 ≈ T(s)
1+T(s) 1 for || T || << 1
fp1
40 dB
|| T ||
– 40 dB/decade
20 dB
fz
– 20 dB/decade
0 dB
+ 20 dB/decade fc fp2
–20 dB
fz Crossover – 40 dB/decade
frequency
+ 40 dB/decade
–40 dB
– | T0 |dB fp1 1
1+T
–60 dB QdB
–80 dB
f
Interpretation: how the loop rejects disturbances
Below the crossover frequency: f < fc

and || T || > 1 1 for || T || >> 1
1 ≈ T(s)
Then 1/(1+T) ≈ 1/T, and 1+T(s) 1 for || T || << 1
disturbances are reduced in
magnitude by 1/ || T ||
Above the crossover frequency: f > fc
and || T || < 1
Then 1/(1+T) ≈ 1, and the
feedback loop has essentially
no effect on disturbances

Terminology: open-loop vs. closed-loop
Original transfer functions, before introduction of feedback (“open-loop

transfer functions”):
Gvd(s) Gvg(s) Z out(s)
Upon introduction of feedback, these transfer functions become

(“closed-loop transfer functions”):
1 T(s) Gvg(s) Z out(s)

H(s) 1 + T(s) 1 + T(s) 1 + T(s)
The loop gain:

T(s)

9.4. Stability
Even though the original open-loop system is stable, the closed-loop

transfer functions can be unstable and contain right half-plane poles. Even
when the closed-loop system is stable, the transient response can exhibit
undesirable ringing and overshoot, due to the high Q -factor of the closed-
loop poles in the vicinity of the crossover frequency.
When feedback destabilizes the system, the denominator (1+T(s)) terms in
the closed-loop transfer functions contain roots in the right half-plane (i.e.,
with positive real parts). If T(s) is a rational fraction of the form N(s) / D(s),
where N(s) and D(s) are polynomials, then we can write
N(s)
T(s) D(s) N(s) • Could evaluate stability by
= = evaluating N(s) + D(s), then
1 + T(s) N(s) N(s) + D(s)
1+
D(s) factoring to evaluate roots.
1 1 D(s) This is a lot of work, and is
= =
1 + T(s) N(s) N(s) + D(s) not very illuminating.
1+
D(s)

Determination of stability directly from T(s)
• Nyquist stability theorem: general result.

• A special case of the Nyquist stability theorem: the phase margin test
Allows determination of closed-loop stability (i.e., whether 1/(1+T(s))
contains RHP poles) directly from the magnitude and phase of T(s).
A good design tool: yields insight into how T(s) should be shaped, to
obtain good performance in transfer functions containing 1/(1+T(s))
terms.

9.4.1. The phase margin test
A test on T(s), to determine whether 1/(1+T(s)) contains RHP poles.

The crossover frequency fc is defined as the frequency where
|| T(j2πfc) || = 1 ⇒ 0dB
The phase margin ϕm is determined from the phase of T(s) at fc , as

follows:
ϕm = 180˚ + ∠T(j2πfc)
If there is exactly one crossover frequency, and if T(s) contains no

RHP poles, then
the quantities T(s)/(1+T(s)) and 1/(1+T(s)) contain no RHP poles
whenever the phase margin ϕm is positive.

Example: a loop gain leading to
a stable closed-loop system
60 dB
|| T || ∠T
|| T ||
40 dB
fp1
Crossover
20 dB fz frequency
fc
∠T
0 dB 0˚
–20 dB –90˚
ϕm
–40 dB –180˚
–270˚

f
∠T(j2πfc) = – 112˚
ϕm = 180˚ – 112˚ = + 68˚

Example: a loop gain leading to
an unstable closed-loop system
60 dB
|| T || ∠T
|| T ||
40 dB
fp1
Crossover
20 dB frequency
∠T fp2 fc
0 dB 0˚
–20 dB –90˚
–40 dB –180˚
ϕm (< 0)
–270˚

f
∠T(j2πfc) = – 230˚
ϕm = 180˚ – 230˚ = – 50˚

9.4.2. The relation between phase margin
and closed-loop damping factor
How much phase margin is required?

A small positive phase margin leads to a stable closed-loop system
having complex poles near the crossover frequency with high Q. The
transient response exhibits overshoot and ringing.
Increasing the phase margin reduces the Q. Obtaining real poles, with
no overshoot and ringing, requires a large phase margin.
The relation between phase margin and closed-loop Q is quantified in
this section.

A simple second-order system
40 dB || T || f0
|| T || ∠T
f
20 dB
– 20 dB/decade
f0
Consider the 0 dB
case where T(s) f2 f0 f2
–20 dB f2
can be well-
approximated in –40 dB 0˚
– 40 dB/decade
the vicinity of the ∠T – 90˚
f2 /10
–90˚
crossover f2
ϕm
frequency as –180˚
10f2
T(s) = 1
s 1 + ωs
–270˚
ω0 2 f

Closed-loop response
If
T(s) = 1
s 1 + ωs
ω0 2
Then
T(s) 1 1
= =
1 + T(s) 1+ 1 1 + s + s2
T(s) ω 0 ω 0ω 2
or,
T(s) 1
=
1 + T(s) 1 + s + ωs
2
Qωc c
where
ω ω0
ωc = ω0ω2 = 2π fc Q = ω0 = ω2
c

Low-Q case
ω ω0
Q = ω0 =
c ω2 low-Q approximation: Q ωc = ω0 ωc
= ω2
Q
40 dB || T || f0
f
20 dB
– 20 dB/decade
fc = f0 f2
0 dB
T f0 Q = f0 / fc
1+T
–20 dB
f2
–40 dB f0 f2 – 40 dB/decade
f2
f
High-Q case
ω ω0
ωc = ω0ω2 = 2π fc Q = ω0 = ω2
c
60 dB f0
|| T || f
40 dB
– 20 dB/decade
20 dB f2
Q = f0 /fc
0 dB
T fc = f0 f2 f0
1+T
–20 dB
f0 f2 – 40 dB/decade
f2
–40 dB
f
Q vs. ϕm
Solve for exact crossover frequency, evaluate phase margin, express

as function of ϕm. Result is:
cos ϕ m
Q=
sin ϕ m
1+ 1 + 4Q 4
ϕ m = tan -1
2Q 4

Q vs. ϕm
20 dB
Q
15 dB
10 dB
5 dB
0 dB Q = 1 ⇒ 0 dB
ϕm = 52˚
–5 dB
Q = 0.5 ⇒ –6 dB
–10 dB ϕm = 76˚
–15 dB
–20 dB
0° 10° 20° 30° 40° 50° 60° 70° 80° 90°
ϕm
9.4.3. Transient response vs. damping factor
Unit-step response of second-order system T(s)/(1+T(s))
2Q e -ωct/2Q 4Q 2 – 1
v(t) = 1 + sin ωc t + tan -1 4Q 2 – 1 Q > 0.5
4Q 2 – 1 2Q
ω ω Q < 0.5
v(t) = 1 – ω –2ω e –ω1t – ω –1ω e –ω2t
2 1 1 2
ωc
ω 1, ω 2 = 1± 1 – 4Q 2
2Q
For Q > 0.5 , the peak value is

4Q 2 – 1
peak v(t) = 1 + e – π /

Transient response vs. damping factor
2
Q = 50
v(t) Q = 10
Q=4
1.5
Q=2
Q=1
1
Q = 0.75
Q = 0.5
Q = 0.3
Q = 0.2
0.5
Q = 0.1
Q = 0.05
Q = 0.01
0
0 5 10 15
ωc t, radians

9.5. Regulator design
Typical specifications:
• Effect of load current variations on output voltage regulation
This is a limit on the maximum allowable output impedance
• Effect of input voltage variations on the output voltage
regulation
This limits the maximum allowable line-to-output transfer
function
• Transient response time
This requires a sufficiently high crossover frequency
• Overshoot and ringing
An adequate phase margin must be obtained
The regulator design problem: add compensator network Gc(s) to
modify T(s) such that all specifications are met.

9.5.1. Lead (PD) compensator
1 + ωs fp
z
Gc(s) = Gc0 Gc0
1 + ωs fz fp
p || Gc || Gc0
fϕmax
fz
= fz fp
Improves phase
margin
fp /10 10fz
+ 45˚/decade
fz /10
0˚ – 45˚/decade
∠ Gc

Lead compensator: maximum phase lead
Maximum phase occurs at

Maximum 90˚
phase lead fϕmax = fz fp
75˚
60˚
fp fz
–
45˚
fz fp
∠ Gc( fϕmax) = tan -1
2
30˚
15˚
fp 1 + sin θ
=
fz 1 – sin θ
0˚
1 10 100 1000 θ =-Gc( fϕmax)
fp / fz
To obtain the maximum improvement in phase margin, we should design our compensator so that the frequency fϕmax coincides with the loop gain crossover frequency fc

Lead compensator design
To optimally obtain a compensator phase lead of θ at frequency fc, the

pole and zero frequencies should be chosen as follows:
1 – sin θ fp
fz = fc Gc0
fz fp
1 + sin θ || Gc || Gc0
fϕmax
1 + sin θ fz
= fz fp
fp = fc
1 – sin θ
fp /10 10fz
If it is desired that the magnitude + 45˚/decade
of the compensator gain at fc be fz /10
0˚ – 45˚/decade
unity, then Gc0 should be chosen ∠ Gc
as
fz f
Gc0 =
fp

Example: lead compensation
60 dB
T0
40 dB || T || T0 Gc0 Original gain
|| T || ∠T
20 dB
Compensated gain f0
0 dB fz
fc
–20 dB fp
Compensated phase asymptotes
0˚
–40 dB 0˚
∠T
Original phase asymptotes –90˚
ϕm
–180˚
–270˚
f

9.5.2. Lag (PI) compensation
|| Gc ||
ω
Gc(s) = Gc∞ 1 + sL
– 20 dB /decade Gc∞
Improves low- fL
frequency loop gain
and regulation
10fL 0˚
∠ Gc
+ 45˚/decade
– 90˚
fL /10

Example: lag compensation
original
|| T ||
(uncompensated) 40 dB
loop gain is Gc∞Tu0
Tu0 20 dB fL f0
Tu(s) =
1 + ωs || Tu || Tu0 fc
0 0 dB
f0
compensator: –20 dB 90˚
ω
Gc(s) = Gc∞ 1 + sL ∠ Tu
–40 dB 0˚
10fL
Design strategy:
–90˚
choose ∠T 10f0 ϕm
Gc∞ to obtain desired –180˚
crossover frequency
ωL sufficiently low to 1 Hz 10 Hz 100 Hz 1 kHz 10 kHz 100 kHz
maintain adequate f
phase margin
Example, continued
Construction of 1/(1+T), lag compensator example:
|| T ||
40 dB
Gc∞Tu0
20 dB fL f0
fc
0 dB
–20 dB fL f0
1 1
1+T G c∞ T u0
–40 dB

f

9.5.3. Combined (PID) compensator
ω
1 + sL 1 + ωs
z
Gc(s) = Gcm
1 + ωs 1 + ωs
40 dB p1 p2
|| Gc || ∠ Gc
|| Gc || fp2
20 dB fp1
Gcm fc
0 dB
fL fz
–20 dB 10 fz f /10 90˚

10 fL
45˚/decade p2
–40 dB 0˚
fp1 /10
– 90˚/decade
– 90˚ 90˚/decade
fz /10 10 fp1 –90˚
∠ Gc fL /10
–180˚
f
9.5.4. Design example
L
50 µH
+
iload
vg(t) + C R
v(t)
28 V – 500 µF 3Ω
Sensor
– H(s) gain
fs = 100 kHz
Transistor Error
gate driver signal
δ Pulse-width vc G (s) ve –+ Hv
modulator c
VM = 4 V Compensator vref
5V

Quiescent operating point
Input voltage Vg = 28V

Output V = 15V, Iload = 5A, R = 3Ω
Quiescent duty cycle D = 15/28 = 0.536
Reference voltage Vref = 5V
Quiescent value of control voltage Vc = DVM = 2.14V
Gain H(s) H = Vref/V = 5/15 = 1/3

Small-signal model
V d
D2 1:D
L
+
–
+
vg(s) + V d C v(s) iload (s)

– R
R
Error d(s)
signal
vref (= 0) + ve (s) vc (s) 1
– Gc(s) VM
T(s)
Compensator VM = 4 V
H(s) v(s)
H(s)
H=1
3

Open-loop control-to-output transfer function Gvd(s)
60 dBV
|| Gvd || ∠ Gvd
40 dBV || Gvd || Q0 = 9.5 ⇒ 19.5 dB
Gd0 = 28 V ⇒ 29 dBV
Gvd(s) = V 1
f0
D 1 + s L + s 2LC 20 dBV
R ∠ Gvd 10 –1/2Q 0 f0 = 900 Hz
0 dBV 0˚
standard form:
–20 dBV –90˚
Gvd(s) = Gd0 1
1+ s + s 2
–40 dBV –180˚
Q 0 ω0 ω0 10 1/2Q 0
f0 = 1.1 kHz
–270˚
salient features:
Gd0 = V = 28V 1 Hz 10 Hz 100 Hz 1 kHz 10 kHz 100 kHz

D f
ω0 1
f0 = = = 1kHz
2π 2π LC
Q0 = R C = 9.5 ⇒ 19.5dB
L

Open-loop line-to-output transfer function
and output impedance
Gvg(s) = D 1
1 + s L + s 2LC
R
—same poles as control-to-output transfer function

standard form:
Gvg(s) = Gg0 1
1+ s + s 2
Q 0 ω0 ω0
Output impedance:
Z out(s) = R || 1 || sL = sL
sC 1 + s L + s 2LC
R

System block diagram
T(s) = Gc(s) 1 G (s) H(s)

vd
VM
iload (s) Load current
variation
Gc(s) H(s) V 1
T(s) =
VM D 1+ s + s 2
Zout (s)
Q 0 ω0 ω0 vg(s)
Gvg (s)
ac line
variation
VM = 4 V –
vref ( = 0 ) + ve(s) vc(s) +
1 d(s) v(s)
– Gc(s) VM Gvd (s)
Duty cycle +
variation
Converter power stage
T(s)
H=1
3
H(s)

Uncompensated loop gain (with Gc = 1)
40 dB
|| Tu || ∠ Tu
20 dB Q0 = 9.5 ⇒ 19.5 dB
|| Tu || Tu0 2.33 ⇒ 7.4 dB
0 dB f0
1 kHz
–20 dB
With Gc = 1, the – 1
10 2Q f0 = 900 Hz
– 40 dB/decade
0˚
loop gain is –40 dB 0˚
∠ Tu
Tu(s) = Tu0 1 –90˚
1+ s + s 2
Q 0 ω0 ω0 –180˚
1
10 2Q f0 = 1.1 kHz
Tu0 = H V = 2.33 ⇒ 7.4dB –270˚
D VM 1 Hz 10 Hz 100 Hz 1 kHz 10 kHz 100 kHz
f
fc = 1.8 kHz, ϕm = 5˚

Lead compensator design
• Obtain a crossover frequency of 5 kHz, with phase margin of 52˚

• Tu has phase of approximately – 180˚ at 5 kHz, hence lead (PD)
compensator is needed to increase phase margin.
• Lead compensator should have phase of + 52˚ at 5 kHz
• Tu has magnitude of – 20.6 dB at 5 kHz
• Lead compensator gain should have magnitude of + 20.6 dB at 5 kHz
• Lead compensator pole and zero frequencies should be
1 – sin (52°)
fz = (5kHz) = 1.7kHz
1 + sin (52°)
1 + sin (52°)
fp = (5kHz) = 14.5kHz
1 – sin (52°)
2
fc 1 fz
• Compensator dc gain should be G c0 = = 3.7 ⇒ 11.3dB
f0 Tu0 fp

Lead compensator Bode plot
40 dB
|| Gc || fp ∠ Gc
Gc0
fz fp
20 dB || Gc || Gc0
fc
0 dB
fz = fz fp
–20 dB fp /10 10fz 90˚

fz /10
0˚
–40 dB 0˚
∠ Gc
–90˚
–180˚

f
Loop gain, with lead compensator
1 + ωs
z
T(s) = Tu0 Gc0
1 + ωs s + s 2
1+ ω0
p Q 0 ω0
40 dB
|| T || Q0 = 9.5 ⇒ 19.5 dB
∠T
|| T || T0 = 8.6 ⇒ 18.7 dB
20 dB
f0
0 dB 1 kHz fz
1.7 kHz fc
–20 dB 5 kHz fp
900 Hz 14 kHz
0˚
–40 dB 0˚
∠T 170 Hz
–90˚
1.4 kHz 17 kHz
ϕm=52˚
1.1 kHz –180˚
–270˚
f
1/(1+T), with lead compensator
40 dB
Q0 = 9.5 ⇒ 19.5 dB
• need more
20 dB
|| T || T0 = 8.6 ⇒ 18.7 dB low-frequency
f0 loop gain
0 dB fz
1 fc • hence, add
1+T 1/T0 = 0.12 ⇒ – 18.7 dB inverted zero
–20 dB fp
Q0 (PID controller)
–40 dB

f

Improved compensator (PID)
ω
1 + ωs 1 + sL
z
Gc(s) = Gcm
1 + ωs
40 dB p
|| Gc || || Gc || ∠ Gc
20 dB fp
fc • add inverted
Gcm
0 dB
fL fz zero to PD
compensator,
–20 dB
10fL 10fz 90˚ without
45˚/decade – 45˚/dec changing dc
–40 dB 0˚ gain or corner
fp /10
∠ Gc – 90˚ 90˚/decade frequencies
fz /10 –90˚
fL /10
• choose fL to be
–180˚
fc/10, so that
phase margin
1 Hz 10 Hz 100 Hz 1 kHz 10 kHz 100 kHz is unchanged
f
T(s) and 1/(1+T(s)), with PID compensator
60 dB
40 dB || T ||
Q0
20 dB
fL f0
0 dB fz
fc
–20 dB fp
1 Q0
–40 dB 1+T
–60 dB
–80 dB
f
Line-to-output transfer function
v 20 dB
vg Q0
0 dB Gvg(0) = D
Open-loop || Gvg || f0
–20 dB D
T u0G cm
fL fz
–40 dB
fc
20 dB/decade
–60 dB
G vg – 40 dB/decade
–80 dB Closed-loop
1+T
–100 dB

f

9.6. Measurement of loop gains
Block 1 Block 2
A
Z1(s)
+
vref (s) +– ve (s) G2 (s) vx (s) = v(s)

G1 (s) ve (s) + vx (s) Z2(s)
–
T(s)
H(s)
Objective: experimentally determine loop gain T(s), by making

measurements at point A
Correct result is
Z 2(s)
T(s) = G1(s) G2(s) H(s)
Z 1(s) + Z 2(s)

Conventional approach: break loop,
measure T(s) as conventional transfer function
VCC
Block 1 Block 2
dc bias
Z1(s)
0 – +
vref (s) +– ve (s) G2 (s)vx (s) = v(s)

G1 (s)ve (s) + vy (s) vx (s) Z2(s)
–
vz
+ –
Tm (s)
H(s)
measured gain is
vy(s)
Tm(s) = Tm(s) = G1(s) G2(s) H(s)
vx(s) vref = 0
vg = 0

Measured vs. actual loop gain
Actual loop gain:

Z 2(s)
T(s) = G1(s) G2(s) H(s)
Z 1(s) + Z 2(s)
Measured loop gain:
Tm(s) = G1(s) G2(s) H(s)
Express Tm as function of T:
Z 1(s)
Tm(s) = T(s) 1 +
Z 2(s)
Tm(s) ≈ T(s) provided that Z 2 >> Z 1

Discussion
• Breaking the loop disrupts the loading of block 2 on block 1.

A suitable injection point must be found, where loading is not
significant.
• Breaking the loop disrupts the dc biasing and quiescent operating
point.
A potentiometer must be used, to correctly bias the input to block 2.
In the common case where the dc loop gain is large, it is very
difficult to correctly set the dc bias.
• It would be desirable to avoid breaking the loop, such that the biasing
circuits of the system itself set the quiescent operating point.

9.6.1. Voltage injection
Block 1 – vz + Block 2
i(s)
Z1(s) Zs(s)
0 – +

G1 (s) ve (s) + vy (s) vx (s) Z2(s)
–
+ –
Tv (s)
H(s)
• Ac injection source vz is connected between blocks 1 and 2

• Dc bias is determined by biasing circuits of the system itself
• Injection source does modify loading of block 2 on block 1

Voltage injection: measured transfer function Tv(s)
Block 1 – vz + Block 2
i(s)
Network analyzer 0
Z1(s)
–
Zs(s)
+
measures vref (s) +– ve (s) G2 (s) vx (s) = v(s)
G1 (s) ve (s) + vy (s) vx (s) Z2(s)
–
vy(s)
Tv(s) = + –
vx(s) vref = 0
vg = 0 Tv (s)
H(s)
Solve block diagram:
ve(s) = – H(s) G2(s) vx(s)
– vy(s) = G1(s) ve(s) – i(s) Z 1(s) Substitute:

Z 1(s)
Hence vy(s) = vx(s) G1(s) G2(s) H(s) +
Z 2(s)
– vy(s) = – vx(s) G2(s) H(s) G1(s) – i(s) Z 1(s)
which leads to the measured gain
with Z (s)
vx(s) Tv(s) = G1(s) G2(s) H(s) + 1
i(s) = Z 2(s)
Z 2(s)

Comparison of Tv(s) with T(s)
Actual loop gain is Gain measured via voltage

injection:
Z 2(s) Z (s)
T(s) = G1(s) G2(s) H(s) Tv(s) = G1(s) G2(s) H(s) + 1
Z 1(s) + Z 2(s) Z 2(s)
Express Tv(s) in terms of T(s):

Z 1(s) Z (s)
Tv(s) = T(s) 1 + + 1
Z 2(s) Z 2(s)
Condition for accurate measurement:

Tv(s) ≈ T(s) provided (i) Z 1(s) << Z 2(s) , and
Z 1(s)
(ii) T(s) >>
Z 2(s)

Example: voltage injection
Block 1 Block 2
vz
50 Ω Z 1(s) = 50Ω
+
–
– + Z 2(s) = 500Ω
Z 1(s)
+ = 0.1 ⇒ – 20dB
– vy (s) vx (s) 500 Ω Z 2(s)
Z 1(s)
+ – 1+ = 1.1 ⇒ 0.83dB
Z 2(s)
suppose actual T(s) = 10 4

1+ s 1+ s
2π 10Hz 2π 100kHz

Example: measured Tv(s) and actual T(s)
100 dB
Z 1(s) Z (s)
Tv(s) = T(s) 1 + + 1
80 dB Z 2(s) Z 2(s)
60 dB || Tv ||
|| T ||
40 dB
20 dB
0 dB Z1
⇒ – 20 dB || Tv ||
Z2
–20 dB
|| T ||
–40 dB
10 Hz 100 Hz 1 kHz 10 kHz 100 kHz 1 MHz
f

9.6.2. Current injection
i y(s)
Ti(s) =
i x(s) vref = 0
vg = 0
Block 1 Block 2
iy (s) ix (s)
Z1(s)
0 iz (s)
G1 (s) ve (s) + Z2(s)
– Zs(s)
Ti (s)
H(s)

Current injection
It can be shown that Injection source impedance Zs

is irrelevant. We could inject
Z 2(s) Z (s) using a Thevenin-equivalent
Ti(s) = T(s) 1 + + 2 voltage source:
Z 1(s) Z 1(s)
iy (s) iz (s) ix (s)

Conditions for obtaining accurate Cb
measurement:
Rs
(i) Z 2(s) << Z 1(s) , and
(ii) T(s) >>

Z 2(s) vz (s)
Z 1(s)

9.6.3. Measurement of unstable systems
• Injection source impedance Zs • Original (unstable) loop gain is

does not affect measurement measured (not including Zs ),
while circuit operates stabily
• Increasing Zs reduces loop
gain of circuit, tending to
stabilize system
– vz +
Block 1 Block 2
Rext
Z1(s) Zs(s)
0 – +
vref (s) Lext

+– ve (s) G2 (s) vx (s) = v(s)
G1 (s) ve (s) + vy (s) vx (s) Z2(s)
–
+ –
Tv (s)
H(s)

9.7. Summary of key points
1. Negative feedback causes the system output to closely follow the

reference input, according to the gain 1/H(s). The influence on the
output of disturbances and variation of gains in the forward path is
reduced.
2. The loop gain T(s) is equal to the products of the gains in the
forward and feedback paths. The loop gain is a measure of how well
the feedback system works: a large loop gain leads to better
regulation of the output. The crossover frequency fc is the frequency
at which the loop gain T has unity magnitude, and is a measure of
the bandwidth of the control system.

Summary of key points
3. The introduction of feedback causes the transfer functions from

disturbances to the output to be multiplied by the factor 1/(1+T(s)). At
frequencies where T is large in magnitude (i.e., below the crossover
frequency), this factor is approximately equal to 1/T(s). Hence, the
influence of low-frequency disturbances on the output is reduced by a
factor of 1/T(s). At frequencies where T is small in magnitude (i.e.,
above the crossover frequency), the factor is approximately equal to 1.
The feedback loop then has no effect. Closed-loop disturbance-to-
output transfer functions, such as the line-to-output transfer function or
the output impedance, can easily be constructed using the algebra-on-
the-graph method.
4. Stability can be assessed using the phase margin test. The phase of T
is evaluated at the crossover frequency, and the stability of the
important closed-loop quantities T/(1+T) and 1/(1+T) is then deduced.
Inadequate phase margin leads to ringing and overshoot in the system
transient response, and peaking in the closed-loop transfer functions.

Summary of key points
5. Compensators are added in the forward paths of feedback loops to

shape the loop gain, such that desired performance is obtained.
Lead compensators, or PD controllers, are added to improve the
phase margin and extend the control system bandwidth. PI
controllers are used to increase the low-frequency loop gain, to
improve the rejection of low-frequency disturbances and reduce the
steady-state error.
6. Loop gains can be experimentally measured by use of voltage or
current injection. This approach avoids the problem of establishing
the correct quiescent operating conditions in the system, a common
difficulty in systems having a large dc loop gain. An injection point
must be found where interstage loading is not significant. Unstable
loop gains can also be measured.

Chapter 9. Controller Design: 9.2. Effect of Negative Feedback On The Network Transfer Functions

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter 9. Controller Design: 9.2. Effect of Negative Feedback On The Network Transfer Functions

Uploaded by

Copyright:

Available Formats

Chapter 9.

Fundamentals of Power Electronics 1 Chapter 9: Controller design

Fundamentals of Power Electronics 2 Chapter 9: Controller design

9.6. Measurement of loop gains

Fundamentals of Power Electronics 3 Chapter 9: Controller design

Switching converter Load

or 120Hz ripple, produced by d(t)

Fundamentals of Power Electronics 5 Chapter 9: Controller design

Fundamentals of Power Electronics 6 Chapter 9: Controller design

Power Switching converter Load

Fundamentals of Power Electronics 8 Chapter 9: Controller design

Small signal model: open-loop converter

vg(s) + j(s) d(s) C v(s) iload (s)

Fundamentals of Power Electronics 9 Chapter 9: Controller design

Fundamentals of Power Electronics 10 Chapter 9: Controller design

iload (s) Load current

vg (s) Zout (s)

Fundamentals of Power Electronics 11 Chapter 9: Controller design

Manipulate block diagram to solve for v(s) . Result is

GcGvd / VM Gvg Z out

which is of the form

with T(s) = H(s) Gc(s) Gvd(s) / VM = "loop gain"

Loop gain T(s) = products of the gains around the negative

Fundamentals of Power Electronics 12 Chapter 9: Controller design

Original (open-loop) line-to-output transfer function:

With addition of negative feedback, the line-to-output transfer function

Feedback reduces the line-to-output transfer function by a factor of

If T(s) is large in magnitude, then the line-to-output transfer function

Fundamentals of Power Electronics 13 Chapter 9: Controller design

Original (open-loop) output impedance:

With addition of negative feedback, the output impedance becomes:

Feedback reduces the output impedance by a factor of

If T(s) is large in magnitude, then the output impedance is greatly

Fundamentals of Power Electronics 14 Chapter 9: Controller design

Closed-loop transfer function from vref to v(s) is:

which is independent of the gains in the forward path of the loop.

Fundamentals of Power Electronics 15 Chapter 9: Controller design

At the crossover frequency fc, || T || = 1 f

Fundamentals of Power Electronics 16 Chapter 9: Controller design

Fundamentals of Power Electronics 17 Chapter 9: Controller design

At frequencies sufficiently less that the crossover frequency, the loop

Below the crossover frequency: f < fc

Fundamentals of Power Electronics 21 Chapter 9: Controller design

Original transfer functions, before introduction of feedback (“open-loop

Gvd(s) Gvg(s) Z out(s)

Upon introduction of feedback, these transfer functions become

1 T(s) Gvg(s) Z out(s)

The loop gain:

Fundamentals of Power Electronics 22 Chapter 9: Controller design

Even though the original open-loop system is stable, the closed-loop

Fundamentals of Power Electronics 23 Chapter 9: Controller design

• Nyquist stability theorem: general result.

Fundamentals of Power Electronics 24 Chapter 9: Controller design

A test on T(s), to determine whether 1/(1+T(s)) contains RHP poles.

The phase margin ϕm is determined from the phase of T(s) at fc , as

If there is exactly one crossover frequency, and if T(s) contains no

Fundamentals of Power Electronics 25 Chapter 9: Controller design

1 Hz 10 Hz 100 Hz 1 kHz 10 kHz 100 kHz

Fundamentals of Power Electronics 26 Chapter 9: Controller design

1 Hz 10 Hz 100 Hz 1 kHz 10 kHz 100 kHz

Fundamentals of Power Electronics 27 Chapter 9: Controller design

How much phase margin is required?

Fundamentals of Power Electronics 28 Chapter 9: Controller design