Digital Control Systems

Lecture-12
Stability Analysis Techniques

Dr. Mountasser Mohamed Ramadan

email: [email protected]
 Lecture Outline
• Example of Routh Hurwitz Stability Criterion for DT
Systems.
• Example of Jury’s Stability Test.
• Root Locus.
• Frequency Domain Analysis:
The Nyquist Criterion.
The Bode Diagram.
• Interpretation of the Frequency Response.
• Closed-Loop Frequency Response.
2
 Example of Routh-Hurwitz Criterion
• Find the range of a gain factor 𝐾 to be the system of
the following figure (with 𝑇 = 1 𝑠 ) stable, using
Routh-Hurwitz approach.

• Solution:
0.368z + 0.264
G z = 2
z − 1.368z + 0.368
• The characteristic equation is given by
1 + 𝐾𝐺 𝑤 = 1 + 𝐾 𝐺 𝑧 𝑤+1 =0
3
𝑧=
 Example of Routh-Hurwitz Criterion
0.368z + 0.264
1 + KG w = 1 + 2
z − 1.368z + 0.368 z=(w+1)/(w−1)

𝑤+1
𝐾 0.368 + 0.264
=1+ 𝑤 − 1
𝑤+1 2 𝑤+1
− 1.368 + 0.368
𝑤−1 𝑤−1
𝐾 0.368 𝑤 + 1 𝑤 − 1 + 0.264 𝑤 − 1 2
=1+
𝑤 + 1 2 − 1.368 𝑤 + 1 𝑤 − 1 + 0.368 𝑤 − 1 2

1.264𝑤 + 2.736 + 0.632𝐾𝑤 2 − 0.528𝐾𝑤 − 0.104𝐾

= 2 2
=0
4
𝑤 + 1 − 1.368 𝑤 + 1 𝑤 − 1 + 0.368 𝑤 − 1
 Example of Routh-Hurwitz Criterion
• Thus the characteristic equation may by expressed as
0.614𝐾𝑤 2 + (1.264 − 0.0528𝐾)𝑤 + (2.736 − 0.104𝐾) = 0
• The Routh array is then
𝑤 2 0.614K 2.736-0.104K ⟹𝐾>0

𝑤 1 1.264-0.528K 0 ⇒ 𝐾 < 2.39

𝑤 0 2.736-0.104K ⇒ 𝐾 < 26.3

• Hence the system is stable for

0 < 𝐾 < 2.39.
5
 Example of Routh-Hurwitz Criterion
𝑤 2 0.614K 2.736-0.104K ⟹𝐾>0

𝑤 1 1.264-0.528K 0 ⇒ 𝐾 < 2.39

𝑤 0 2.736-0.104K ⇒ 𝐾 < 26.3

• The system is marginally stable for 𝐾 = 2.39.

• The auxiliary equation is
• 0.614𝐾𝑤 2 + 2.736 − .104𝐾 𝐾=2.39 = 0
2.48744
𝑤 = ±𝑗 =± 𝑗1.32 𝑟𝑎𝑑/𝑠
6
1.46746
 Example of Jury’s Stability Test.
• Find the range of a gain factor 𝐾 to be the system of
the following figure (with 𝑇 = 1 𝑠 ) stable, using
Jury’s Stability Test.

• Solution:
0.368z + 0.264
G z = 2
z − 1.368z + 0.368
• The characteristic equation is given by
1 + 𝐾𝐺(𝑧) = 0
7
 Example of Jury’s Stability Test.
K(0.368z + 0.264)
1 + KG z = 1 + 2 =0
z − 1.368z + 0.368
• Thus the characteristic equation may by expressed as
z 2 + 0368K − 1.368 z + 0.368 + 0.264K = 0
• The Jury’s Table is
Row 𝒛𝟎 𝒛𝟏 𝒛𝟐
1 0.368+0.264K 0.368K-1.368 1

• The constraint 𝐹 1 > 0 yields

1 + 0368K − 1.368 + 0.368 + 0.264K > 0 ⇒
8 0.632𝐾 > 0 ⇒ 𝐾 > 0
 Example of Jury’s Stability Test.
• The constraint (−1)2 𝐹 −1 > 0 yeilds
1 − 0368K − 1.368 + 0.368 + 0.264K > 0 ⇒
2.736
2.736 − 0.104𝐾 > 0 ⇒ 𝐾 < = 26.3
0.104
• The constraint 𝑎𝑜 < 𝑎2 yeilds
0.632
0.368 + 0.264K < 1 ⇒ 𝐾 < = 2.39
0264
• Hence the system is stable for
0 < 𝐾 < 2.39.
9
 Example of Jury’s Stability Test.
• The system is marginally stable for 𝐾 = 2.39.
• For this value of 𝐾, the characteristic equation is
𝑧 2 + 0.368𝐾 − 1.368 𝑧 + (0.368 + 0.264𝐾)
𝐾=2.39

= 𝑧 2 − 0.488𝑧 + 1 = 0
• The roots of this equation are
𝑧 = 0.244 ± 𝑗0.970 = 1∠ ± 75.9° = 1∠ ± 1.32 𝑟𝑎𝑑
= 1∠ ± 𝜔𝑇
• Since 𝑇 = 1 𝑠, the system will oscillate at frequency of
1.32 𝑟𝑎𝑑/𝑠.
 Root Locus
• The root locus method provides a quick means of
predicting the closed-loop behavior of a system
based on its open-loop poles and zeros.
• For the LTI sampled-data system of the following
Figure,

𝐶(𝑧) 𝐾𝐺(𝑧)
=
𝑅(𝑧) 1 + 𝐾𝐺𝐻(𝑧)
 Root Locus
•The method is based on the properties of the closed-
loop characteristic equation
1 + 𝐾𝐺𝐻 𝑧 = 0
•The root locus for this system is a plot of the locus in
the previous equation in the z-plane as a function of K.
•Thus the rules of root locus constructed for discrete
time systems are identical to those for continuous
time systems.
 Root Locus
• Rules for root locus construction

• For the characteristic equation

1 + 𝐾𝐺𝐻 𝑧 = 0

1. Loci originate on poles of 𝐺𝐻 𝑧 and terminate on

the zeros of 𝐺𝐻 𝑧 .

2. The root locus on the real axis always lies in a section

of the real axis to the left of an odd number of poles
and zeros on the real axis.
 Root Locus

• Rules for root locus construction

3. The root locus is symmetrical with respect to the

real axis.

4. The number of asymptotes is equal to the number

of poles of 𝐺𝐻 𝑧 , 𝑛𝑝, minus the zeros of 𝐺𝐻 𝑧 , 𝑛𝑧,
with angles given by 2𝑘 + 1 𝜋 𝑛𝑝 − 𝑛𝑧 .
 Root Locus
• Rules for root locus construction
5. The asymptotes intersect the real axis at 𝜎𝐴 , where

𝑝𝑜𝑙𝑒𝑠 𝑜𝑓𝐺𝐻 𝑧 − 𝑧𝑒𝑟𝑜𝑠 𝑜𝑓𝐺𝐻 𝑧

𝜎𝐴 =
𝑛𝑝 − 𝑛𝑧
6. The breakaway points are given by the roots of
𝑑 𝐺𝐻(𝑧)
=0
𝑑𝑧
Or, equivalently,
𝑑𝑁 𝑧 𝑑𝐷 𝑧 𝑁(𝑧)
𝐷 𝑧 −𝑁 𝑧 = 0, 𝐺𝐻 𝑧 =
𝑑𝑧 𝑑𝑧 𝐷(𝑧)
 Example
• Consider the unity negative feedback system with
0368𝐾(𝑧+0.717)
𝐾𝐺 𝑧 =
(𝑧−1)(𝑧−0.368)
• Draw the root locus
• Solution
1. The loci originate at 𝑧 = 1, 𝑧 = 0.368, and terminate
at 𝑧 = −0.717 and 𝑧 = ∞. (𝑛𝑝 = 2, 𝑛𝑧 = 1)
2. The root locus on the real axis
3. The root locus is symmetrical w.r.t the real axis.
4. 𝑛𝑝 − 𝑛𝑧 = 1 then 𝜃𝐴 = 180°
1+.368 −(−.717)
5. 𝜎𝐴 = 2−1
= 2.085
 Example
𝑑 𝐺𝐻(𝑧) 𝑑𝑁 𝑧 𝑑𝐷 𝑧
6. The breakaway points = 0 Or, 𝐷 𝑧 −𝑁 𝑧 = 0,
𝑑𝑧 𝑑𝑧 𝑑𝑧

occur at 𝑧 = 0.65 and 𝑧 = −2.08

K=15.0 K=0.196
 Example
• The points of intersection of root loci with the unit
circle may be found by
1. Graphical construction, with 𝐾𝐺𝐻(𝑧) = 1; i.e;

0.368𝐾(𝑍1 ) 0.368𝐾(1.364)
= =1
(𝑃1 )(𝑃2 ) (1.229)(0.978)

• Then
𝐾 = 2.39
2. The Jury stability test,
3. The Routh-Hurwitz criterion.
 Example

K=2.39

0.97
𝑃1
𝑍1
K=15.0 𝑃2 K=0.196

0.244
 Example (using matlab)
• Using MATLAB as
• ex1
>> n=[0 0.368 0.264];
>> d=[1 -1.368 0.368];
>> r=rlocus(n,d);
>> r
>> pause
>> plot(real(r),imag(r))
>> ex2
>> g = tf([0 0.368 0.264],[1 -1.368 0.368])
>> rlocus(g);
 Frequency Domain Analysis
• When a sinusoidal input is given to a stable LTI
system it produces a sinusoidal output of same
frequency but with different magnitude and
phase.
• The variation of output magnitude and phase with
input frequency is known as frequency response
of the system.
• Frequency domain analysis provides a good design
in presence of uncertainty in plant model.
 Frequency Domain Analysis
• Experimental results can be used to construct
frequency response even if the plant model is
unknown.
• Analysis of digital control systems in frequency
domain depends on the extension of the existing
techniques in continuous time case.
• Two most popular graphical representations in
frequency domain are Nyquist plot and Bode
diagram.
 Nyquist plot (Polar Plot)
• The Nyquist plot of a transfer function, usually the
loop transfer function GH(z), is a mapping of
Nyquist contour in z-plane onto GH(z) plane which
is in polar coordinates.
• Thus it is sometimes known as polar plot.
Absolute and relative stabilities can be determined
from the Nyquist plot using Nyquist stability
criterion.
• Given the loop transfer function 𝐺𝐻(𝑧) of a digital
control system, the polar plot of 𝐺𝐻(𝑧) is obtained
by setting 𝑧 = 𝑒𝑗𝜔𝑇 and varying 𝜔 from 0 to ∞.
 Nyquist Plot (Polar Plot)
• The polar plot of a sinusoidal
transfer function G(jω) is a
plot of the magnitude of
G(jω) versus the phase angle
of G(jω) on polar
coordinates as ω is varied
from zero to infinity.
• Thus, the polar plot is the
locus of vectors
as ω is varied from zero to
infinity.
 Nyquist Plot (Polar Plot)
• Each point on the polar
plot of G(jω) represents
the terminal point of a
vector at a particular
value of ω.

• The projections of G(jω)

on the real and
imaginary axes are its
real and imaginary
components.
 Nyquist Plot (Polar Plot)
• An advantage in using a
polar plot is that it depicts
the frequency response
characteristics of a system
over the entire frequency
range in a single plot.
• One disadvantage is that
the plot does not clearly
indicate the contributions
of each individual factor of
the open-loop transfer
function.
 Nyquist Plot of Integral and Derivative Factors

• The polar plot of G(jω)=1/jω is the negative

imaginary axis, since
Im
1
G( j ) 
j

1  j 1
G( j )    j Re
j  j  ω=∞ -90o

1
In polarform G( j )    90
 ω=0
Nyquist Plot of Integral and Derivative Factors

• The polar plot of G(jω)=jω is the positive

imaginary axis, since
Im
ω=∞
G( j )  j

In polarform G( j )  90 
ω=0
90o
Re
 Nyquist Plot of First Order Factors

• The polar plot of first order factor in numerator

is
G( j )  j  1

𝜔 𝑅𝑒 𝐼𝑚 Im ω= ∞

0 1 0 2 ω=2

1 ω=1
1 1 1 ω=0
Re
1

2 1 2
∞ 1 ∞
 Nyquist Plot of First Order Factors
• The polar plot of first order factor in denominator is
1
G( j )  𝜔 𝑅𝑒 𝐼𝑚
j  1

1 1  j 0 1 0
G( j )  
j  1 1  j 0.5 0.8 0.4
1  j 1 1/2 −1/2
G( j ) 
1 2 2 1/5 −2/5
1  ∞ 0 0
G( j )  2
j
1  1 2
 Nyquist Plot of First Order Factors
• The polar plot of first order factor in denominator is

𝜔 𝑅𝑒 𝐼𝑚 Im

0 1 0
0.5 0.8 −0.4 ω= ∞ 0.2 0.5 0.8 1
Re
ω=0
1 0.5 −0.5 -0.4 ω=2
ω=0.5
-0.5
ω=1

2 0.2 −0.4
∞ 0 0
 Nyquist Plot of First Order Factors
• The polar plot of first order factor in denominator is

𝜔 𝑅𝑒 𝐼𝑚 G( j ) G( j ) Im

0 1 0 1 0𝑜
0.5 0.8 −0.4 0.9 −26𝑜 ω= ∞
ω=0
Re

1 0.5 −0.5 0.7 −45𝑜 ω=2

ω=0.5
ω=1

2 0.2 −0.4 0.4 −63𝑜

∞ 0 0 0 −90𝑜
 Example
• Draw the polar plot of following open loop transfer
function. 1
G( s ) 
s( s  1)
Solution Put s  j
1
G( j ) 
j( j  1)
1
G( j ) 
  2  j
1   2  j
G( j )  
2
   j   2  j
  2  j
G( j ) 
4  2
 Example
𝜔 𝑅𝑒 𝐼𝑚
  2  j
G( j ) 
4  2 0 ∞ ∞
0.1 −1 −10
 2 
G( j )  4 2
j 4 2
0.5 −0.8 −1.6
   
1 −0.5 −0.5
1 1 2 −0.2 −0.1
G( j )  2
j
 1 ( 2  1)
3 −0.1 −0.03
∞ 0 0
 Example
𝜔 𝑅𝑒 𝐼𝑚 Im

0 ∞ ∞
0.1 −1 −10
0.5 −0.8 −1.6 -1 ω=∞ Re
ω=2 ω=3
1 −0.5 −0.5 ω=1

2 −0.2 −0.1
ω=0.5

3 −0.1 −0.03
ω=0.1 -10
∞ 0 0 ω=0
 The Nyquist Criterion.

• The Nyquist criterion allows us to answer two

questions:
1. Does the system have closed-loop poles
outside the unit circle?
2. If the answer to the first question is yes, how
many closed-loop poles are outside the unit
circle?
 The Nyquist Criterion.
• To develop the Nyquist criterion for discrete- time
systems, we consider the systems shown in the
following figures
 The Nyquist Criterion.
• The transfer function for the continuous-time
system of Figure a is

𝐶(𝑠) 𝐺𝑝 (𝑠)
=
𝑅(𝑠) 1 + 𝐺𝑝 𝑠 𝐻(𝑠)
• And for the sampled-data system of Figure b the
transfer function is

𝐶 ∗ (𝑠) 𝐺 ∗ (𝑠)
∗
=
𝑅 (𝑠) 1 + 𝐺𝐻∗ (𝑠)
 The Nyquist Criterion.
• The characteristic equation for the continuous-time
system is
1 + 𝐺𝑝 𝑠 𝐻 𝑠 = 0
• And for the sampled-data system the characteristic
equation is
1 + 𝐺𝐻∗ 𝑠 = 0
• The characteristic equation of the sampled-data
system can also be written as
1 + 𝐺𝐻 𝑧 = 0
 The Nyquist Criterion.
• Recall that the continuous-time system is stable if
the roots of the equation 1 + 𝐺𝑝 𝑠 𝐻 𝑠 = 0 are
all contained in the left half-plan.
• Similarly, the sampled-data system is stable if the
roots of the equation 1 + 𝐺𝐻∗ 𝑠 = 0 all lie in the
left half-plan.
• Or if the roots of the equation 1 + 𝐺𝐻 𝑧 = 0 all lie
Within the unit circle.
 The Nyquist Criterion.
• Before discussing Nyquist stability criterion for the
digital system, following steps are necessary.
1. Defining the Nyquist path in the 𝑧-plane that encloses
the exterior of the unit circle. Here the region to the
left of a closed path is considered to be enclosed by
that path when the direction of the path is taken
anticlockwise.
2. Mapping the Nyquist path in z-plane onto the 𝐺𝐻(𝑧)
plane which results in Nyquist plot of 𝐺𝐻(𝑧).
3. Stability of the closed loop system is investigated by
studying the behavior of Nyquist plot with respect to
the critical point (−1, 𝑗0) in the 𝐺𝐻(𝑧) plane.
 The Nyquist Criterion.
• Theorem. Let 𝑓(𝑧) be the ratio of two polynomials in
𝑧. Let the closed curve 𝐶 in the 𝑧-plane be the
mapped into the complex plane through the
mapping 𝑓(𝑧). If 𝑓(𝑧) is analytic and on 𝐶, except at
a finite number of poles, and if 𝑓(𝑧) has neither
poles nor zeros on 𝐶, then
𝑁 =𝑍−𝑃
• where Z is the number of zeros of 𝑓(𝑧) in 𝐶, 𝑃 is
the number of poles of 𝑓(𝑧) in 𝐶, and 𝑁 is the
number of the encirclements of the origin, taken in
the same sense as 𝐶.
 Nyquist Stability Criterion
• The Nyquist stability Im

criterion determines the

stability of a closed-loop
system from its open-loop
frequency response and (-1, j0)
Re
open-loop poles.

• A minimum phase closed

loop system will be stable
if the Nyquist plot of open
loop transfer function does
not encircle (-1, j0) point.
 Gain Margin

Phase Margin
Gain cross-over point
Phase cross-over point

45 4/1/2017
 Example
Examine the stability of the given system using Nyquist
Criterion
0.368𝑧 + 0.264
𝐺 𝑧 = 2
𝑧 − 1.368𝑧 + 0.368
Solution:
𝑇
1+ 2 𝑤
Pot 𝑧 = 𝑇 , with 𝑇 = 1 𝑠
1− 2 𝑤
Then 𝐺(𝑤) is given by,
1 + 0.5𝑤
0.368 + 0.264
1 − 0.5𝑤
𝐺 𝑤 =
1 + 0.5𝑤 2 1 + 0.5𝑤
− 1.368 + 0.368
1 − 0.5𝑤 1 − 0.5𝑤
−0.381 𝑤 − 2 𝑤 + 12.14
=
𝑤 𝑤 + 0.924
 Example
• The Nyquist diagram can be obtained from this
equation by allowing 𝑤 to assume values from
𝑗0 to 𝑗∞.
• Since 𝐺(𝑤) has a pole at the origin, the Nyquist
path must detour around this point.
• The Nyquist diagram generated by 𝐺(𝑤) is
identical to that generated using 𝐺(𝑧) as shown
in the following
 Example

• the system is stable,

1
• The Gain Margin is
𝑎
• The Phase Margin is ∅
 The Bode Diagram
• Bode plot is the graphical tool for drawing the
frequency response of a system.
• It is represented by two separate plots, one is the
magnitude 𝑣𝑠 frequency and the other one is
phase 𝑣𝑠 frequency.
• The magnitude is expressed in 𝑑𝐵 and the
frequency is generally plotted in log scale.
 The Bode Diagram
• One of the advantages of the Bode plot in s-
domain is that the magnitude curve can be
approximated by straight lines which allows the
sketching of the magnitude plot without exact
computation.
• This feature is lost when we plot Bode diagram in
z-domain .
 The Bode Diagram

• To incorporate this feature we use bi-linear

transformation to transform unit circle of the z-
plane into the imaginary axis of another complex
plane, w plane, where

1
𝑤 = ln(𝑧)
𝑇
 The Bode Diagram
• From the power series expansion

2 𝑧−1
𝑤=
𝑇 𝑧+1

2 𝑤𝑇
+𝑤 1+
⇒𝑧= 𝑇 = 2
2 𝑤𝑇
−𝑤 1− 2
𝑇
• For frequency domain analysis the above bi-linear
transformation may be used to convert
GH(z) to GH(w) and then construct the Bode plot.
Decade
 Basic Factors of a Transfer Function
• The basic factors that very frequently occur in an
arbitrary transfer function are

1. Gain 𝐾
±1
2. Integral and Derivative Factors (𝑗𝜔)
±1
3. First Order Factors (𝑗𝜔𝑇 + 1)

2 ±1
𝑗𝜔 𝑗𝜔
4. Quadratic Factors 1 + 2𝜁 +
𝜔𝑛 𝜔𝑛
 Basic Factors of a Transfer Function
1. Gain 𝐾
• The log-magnitude curve for a constant gain 𝐾 is a
horizontal straight line at the magnitude of
20 𝑙𝑜𝑔(𝐾) decibels.

• The phase angle of the gain 𝐾 is zero.

• The effect of varying the gain 𝐾 in the transfer

function is that it raises or lowers the log-magnitude
curve of the transfer function by the corresponding
constant amount, but it has no effect on the phase
curve.
 If K  5 Then20 log (K)  20 log ( 5 )  14db
15
Magnitude (decibels)

-5

-15
0.1 1 10 100 103 104 105 106 107 108 109
Frequency (rad/sec)
 Im 0
If K  5 Then  tan-1( )  tan-1( )  0
90o
Re 5

30o
Phase (degrees)

0o

-300

-90o
0.1 1 10 100 103 104 105 106 107 108 109
Frequency (rad/sec)
 Basic Factors of a Transfer Function
2. Integral and Derivative Factors (jω)±1
Derivative Factor G ( s )  s, where s  j
G( j )  20 log( )
Magnitude
𝝎 0.1 0.2 0.4 0.5 0.7 0.8 0.9 1
𝒅𝒃 −20 −14 −8 −6 −3 −2 −1 0

Slope=6d`b/octave
Slope=20db/decade

Phase 1 
G( j )  tan ( )  90
0
 20db
decade
30
Magnitude (decibels)

10

-10

-20

-30
0.1 1 10 100 103 104 105 106 107 108 109
Frequency (rad/sec)
 
  tan-1( )  90
180o
0

900

60o
Phase (degrees)

0o

-600

-180o
0.1 1 10 100 103 104 105 106 107 108 109
Frequency (rad/sec)
 Basic Factors of a Transfer Function
±1
2. Integral and Derivative Factors (𝑗𝜔)
• When expressed in decibels, the reciprocal of a number
differs from its value only in sign; that is, for the number N,
1
20 log( N )  20 log( )
N
• Therefore, for Integral Factor the slope of the magnitude line would
be same but with opposite sign (i.e -6db/octave or -20db/decade).

1
Magnitude G( j )   20 log( )
j

1 
Phase G( j )   tan ( )  90
0
 20db
decade
30

20
Magnitude (decibels)

10

-10

-30
0.1 1 10 100 103 104 105 106 107 108 109
Frequency (rad/sec)
 
  tan-1( )  90
180o
0

60o
Phase (degrees)

0o

-600

-900

-180o
0.1 1 10 100 103 104 105 106 107 108 109
Frequency (rad/sec)
Basic Factors of a Transfer Function
3. First Order Factors (jωT+1)
M ( )  20 log( 1  jT )
M ( )  20 log( 1   2T 2 )
1
– For Low frequencies ω<<1/T G( s )  ( s  3)  ( s  1)
3

M ( )  20 log(1)  0 1 T
T
– For high frequencies ω>>1/T

M ( )  20 log(T )
 Basic Factors of a Transfer Function
3. First Order Factors (𝑗𝜔𝑇 + 1)

 ( )  tan-1(T )

when   0, ( )  tan-1(0)  0

1 -1
when   ,  ( )  tan (1)  45 
T
when   , ( )  tan-1()  90
 1
G( s )  ( s  3)  ( s  1)
3
30

20

6 db/octave
Magnitude (decibels)

10
20 db/decade

0
ω=3

-10

-30
0.1 1 10 100 103 104 105 106 107 108 109
Frequency (rad/sec)
 90o

45o
30o
Phase (degrees)

0o

-300

-90o
0.1 1 10 100 103 104 105 106 107 108 109
Frequency (rad/sec)
Basic Factors of a Transfer Function
−1
3. First Order Factors (𝑗𝜔𝑇 + 1)
M ( )  20 log( 1  jT )
M ( )  20 log( 1   2T 2 )
1
– For Low frequencies ω<<1/T G( s ) 
( s  3)

M ( )  20 log(1)  0
– For high frequencies ω>>1/T

M ( )  20 log(T )
 Basic Factors of a Transfer Function
−1
3. First Order Factors (𝑗𝜔𝑇 + 1)

 ( )  tan-1(T )

when   0, ( )  tan-1(0)  0

1 -1
when   ,  ( )  tan (1)  45 
T
when   , ( )  tan-1()  90
 1
G( s ) 
( s  3)
30
Magnitude (decibels)

10

0
ω=3

-10
-6 db/octave
-20 db/decade
-20

-30
0.1 1 10 100 103 104 105 106 107 108 109
Frequency (rad/sec)
 90o

30o
Phase (degrees)

0o

-300

-45o

-90o
0.1 1 10 100 103 104 105 106 107 108 109
Frequency (rad/sec)
 Relative Stability
 Phase crossover frequency (ωp) is the frequency at which
the phase angle of the open-loop transfer function equals –
180°.
 The gain crossover frequency (ωg) is the frequency at which
the magnitude of the open loop transfer function, is unity.
 The gain margin (Kg) is the reciprocal of the magnitude of
G(jω) at the phase cross over frequency.

 The phase margin (γ) is that amount of additional phase lag

at the gain crossover frequency required to bring the system
to the verge of instability.
Relative Stability
 Gain cross-over point

Phase cross-over point

ωg ωp
75 4/1/2017
 Unstable Stable

Stable Gain Margin

Stable
Phase Margin

Unstable Stable

ωg ωp
76 4/1/2017
 Example
• Examine the stability of the given system using
Bode Plot

0.381(𝑤 − 2)(𝑤 + 12.104)

𝐺 𝑤 =
𝑤(𝑤 + 0.924)
• Solution:

0.381(𝑗𝜔𝑤 − 2)(𝑗𝜔𝑤 + 12.104)

𝐺 𝑗𝜔𝑤 =
𝑗𝜔𝑤 (𝑗𝜔𝑤 + 0.924)

𝜔𝑤 𝜔𝑤
− 𝑗 −1 𝑗 +1
= 2 12.14
𝜔𝑤
𝑗𝜔𝑤 𝑗 +1
0.924
 Example
0.381(𝑗𝜔𝑤 − 2)(𝑗𝜔𝑤 + 12.104)
𝑗𝜔𝑤 =
𝑗𝜔𝑤 (𝑗𝜔𝑤 + 0.924)

𝜔𝑤 𝜔𝑤
− 𝑗 −1 𝑗 +1
= 2 12.14
𝜔𝑤
𝑗𝜔𝑤 𝑗 +1
0.924
Note that the numerator break frequencies are 𝜔𝑤 = 2
and 𝜔𝑤 = 12.14 and the denominator break
frequencies are 𝜔𝑤 = 0 and 𝜔𝑤 = 0.924.
The Bode diagram for this system, using straight-line
approximations, is shown in the following Figure.
Example
 Example (using matlab)
• Using MATLAB as
• ex3
>> ww=logspace(-1,1,10);
>>n=[-0.0381 -0.386 0.924];
>> d=[1 0.924 0];
>> [mag,phase,ww]=bode(n,d,ww);
>> db=20*log10(mag);
>>disp(‘ omegaw mag dB phase’);
>>[ww’,mag,db,phase]
>> pause
>> subplot(211), semilogx(ww,db)
>> title(‘Bode diagram’);xlabel(‘w-plane frequency’);ylabel(‘db’);grid
>>subplot(212), semilogx(ww,phase)
>> xlabel(‘w-plane frequency’);ylabel(‘phase’);grid
>>pause,subplot(111)