KON 313

Feedback Control Systems

Frequency Response Techniques - Bode

Dr. Aysa Jafari Farmand

[email protected]
Contents
• Introduction
• Introduction - The concept of frequency response
• Introduction - Plotting the frequency response
• Bode Plots
• Stability, Gain Margin, and Phase Margin via Bode Plots

Aysa Jafari Farmand

Introduction
Frequency Response. The term frequency response means the steady-state
response of a system to a sinusoidal input.
• In frequency-response methods, we vary the frequency of the input signal over a
certain range and study the resulting response.
Frequency response compensation presents the same design of feedback control
systems through gain adjustment and compensation networks from the perspective
of frequency response.
• The results of frequency response compensation techniques are not new or
different from the results of root locus techniques. Frequency response methods,
developed by Nyquist and Bode in the 1930s, are older than the root locus
method, which was discovered by Evans in 1948.

Aysa Jafari Farmand

Introduction
• This technique provides a new point of view and has distinct advantages:
1. When modelling transfer functions from physical data (we can use the data
obtained from measurements on the physical system without deriving its
mathematical model).
2. When designing lead compensators to meet a steady-state error requirement
and a transient response requirement.
3. When finding the stability of nonlinear systems.
4. In settling ambiguities when sketching a root locus.
In fact, the frequency response and root-locus approaches complement each other.
In many practical designs of control systems both approaches are employed.

Aysa Jafari Farmand

Introduction - The concept of frequency response
Recall. Sinusoids can be represented as complex numbers called phasors.
• The magnitude of the complex number is the amplitude of the sinusoid.
• The angle of the complex number is the phase angle of the sinusoid.
Thus,
𝑀 cos(𝜔𝑡 + 𝜙)
can be represented as
𝑀∠𝜙
where 𝜔 is the frequency.

Note. The product of the input phasor and the system function yields the
phasor representation of the output.

Aysa Jafari Farmand

Introduction - The concept of frequency response
Recall. In the steady state, sinusoidal inputs to a linear system generate
sinusoidal responses of the same frequency, but different amplitude and phase
from the input.
Input: 𝑥 𝑡 = 𝐴𝑐𝑜𝑠(𝜔𝑡 + 𝜑)
Output: 𝑦 𝑡 = 𝐴 𝐺 𝑗𝜔 𝑐𝑜𝑠(𝜔𝑡 + 𝜑 + ∠𝐺(𝑗𝜔))
These differences ( 𝐺 𝑗𝜔 and ∠𝐺(𝑗𝜔)) are functions of frequency.
Frequency Response Function. 𝐺 𝑗𝜔 is called the frequency response of the
system.
𝐺 𝑗𝜔 = 𝐺 𝑗𝜔 ∠𝐺 𝑗𝜔
= 𝐺 𝑗𝜔 𝑒 ∠𝑗𝐺 𝑗𝜔
Note that 𝑔(𝑡) is the impulse response of the system.

Aysa Jafari Farmand

Introduction - The concept of frequency response
Waveform Representation

Aysa Jafari Farmand

Introduction - Plotting the frequency response
• The system frequency response is the combination of the magnitude frequency
response and the phase frequency response.

• The frequency response can be plotted in several ways; two of them are
(1) Bode Plot. Separate magnitude and phase plots as a function of frequency
(2) Nyquist. As a polar plot, where the phasor length is the magnitude, and
the phasor angle is the phase.

Aysa Jafari Farmand

 Introduction - Plotting the frequency response
• Note that each point on the frequency response is equivalent to substituting that
point, 𝑠 = 𝑗𝜔, into 𝐺(𝑠) and evaluating its value.
• Data for the plots can also be obtained using vectors on the 𝑠-plane drawn from
the poles and zeros of 𝐺(𝑠) to the imaginary axis.
oThe magnitude response at a particular frequency is the product of the vector
lengths from the zeros of 𝐺(𝑠) divided by the product of the vector lengths
from the poles of 𝐺(𝑠), drawn to points on the imaginary axis.
oThe phase response is the sum of the angles from the zeros of 𝐺(𝑠) minus the
sum of the angles from the poles of 𝐺(𝑠) drawn to points on the imaginary axis.
oPerforming this operation for successive points along the imaginary axis yields
the data for the frequency response.

Aysa Jafari Farmand

Introduction - Plotting the frequency response
Ex. Find the analytical expression for the magnitude frequency response and the
phase frequency response for a system 𝐺 𝑠 = 1/(𝑠 + 2). Also, plot both the
separate magnitude and phase diagrams and the polar plot.
Soln. First substitute 𝑠 = 𝑗𝜔,

• The magnitude frequency response:

• The phase frequency response:

Aysa Jafari Farmand

 Introduction - Plotting the frequency response
Soln. Cont. The separate magnitude and phase diagrams.
• The magnitude diagram is

• The phase diagram is

Aysa Jafari Farmand

 Introduction - Plotting the frequency response
Soln. Cont. The polar plot.
• Plot of
for different 𝜔.

Note. Either of these frequency response presentations can also be obtained from Aysa
theJafari
other.
Farmand
Bode Plots
Bode plots. The log-magnitude and phase frequency response curves as functions
of log 𝜔 are called Bode plots or Bode diagrams.
• Magnitude Plot. logarithmic vertical axis (in 𝑑𝐵), vs. logarithmic frequency axis
(log 𝜔)
• Phase Plot. linear vertical axis (in degree or radian), logarithmic frequency axis
(log 𝜔)

Asymptotic Approximations. An approximation of the magnitude and phase plots

in order to sketch them rapidly.
• Sketching Bode plots can be simplified because they can be approximated as a
sequence of straight lines (asymptotes).

Aysa Jafari Farmand

Bode Plots
Consider the following general system

• The magnitude frequency response is

Converting to dB

Thus, the total response in 𝑑𝐵 is the algebraic sum of the response of each term.
• The phase frequency response is also the sum of the phase frequency response
curves of the zero terms minus that sum of the pole terms.
Aysa Jafari Farmand
Bode Plots - Asymptotic Approximations
Let us now show how to approximate the frequency response of all possible pole
and zero terms by straight-line approximations. Consider the basic factors:

1. Gain 𝐾

2. Integral and derivative factors 𝑠 ±1 ⟹ (𝑗𝜔)±1

3. First order factors (𝑠 + 𝑎)±1 ⟹ (1 + 𝑗𝜔𝑇)±1

2
𝑗𝜔 𝑗𝜔
4. Quadratic factors 𝐺 𝑠 = (𝑠 2 + 2𝜁𝑠 + 𝜔𝑛 2 )±1 ⟹ (1 + 2𝜁 + )±1
𝜔𝑛 𝜔𝑛

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations
1. Gain 𝐾
• As a number increases by a factor of 10, the corresponding decibel value
increases by a factor of 20.

• The reciprocal of a number, expressed in decibels, differs from its value only in
sign; that is, for the number 𝐾.

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations
2. Integral and derivative factors 𝑠 ±1 ⟹ (𝑗𝜔)±1

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations
3. First order factors (𝑠 + 𝑎)±1 ⟹ (𝑗𝜔 + 𝑎)±1
𝑗𝜔
𝐺 𝑠 = 𝑠 + 𝑎 ⟹ 𝐺 𝑗𝜔 = 𝑗𝜔 + 𝑎 = 𝑎( + 1)
𝑎
• The low-frequency asymptote, 𝜔 ≪ 𝑎 ∶ 𝐺 𝑗𝜔 ≈ 𝑎
• The high-frequency asymptote,. 𝜔 ≫ 𝑎 ∶

Note. The high-

freq. approximation
is equal to the low-
freq. approximation
at break frequency,
𝜔 = 𝑎.
Aysa Jafari Farmand
Bode Plots - Asymptotic Approximations
3. First order factors (𝑠 + 𝑎)±1 ⟹ (𝑗𝜔 + 𝑎)±1
𝑗𝜔
𝐺 𝑠 = 𝑠 + 𝑎 ⟹ 𝐺 𝑗𝜔 = 𝑗𝜔 + 𝑎 = 𝑎( + 1)
𝑎
• The low-frequency asymptote, 𝜔 ≪ 𝑎 ∶ 𝐺 𝑗𝜔 ≈ 𝑎
• The high-frequency asymptote,. 𝜔 ≫ 𝑎 ∶

Note. The high-

freq. approximation
is equal to the low-
freq. approximation
at break frequency,
𝜔 = 𝑎.
Aysa Jafari Farmand
Bode Plots - Asymptotic Approximations
3. First order factors (𝑠 + 𝑎)±1 ⟹ 𝑗𝜔 + 𝑎 ±1
Normalizing and Scaling
• It is often convenient to normalize the magnitude and scale the frequency so that
the log-magnitude plot will be 0 dB at a break frequency of unity. It helps:
1. When comparing different first- or second-order frequency response plots.
2. When sketching the frequency response of a function makes it easier to add
components to obtain the Bode plot.
𝑠
• To normalize (𝑠 + 𝑎), we factor out the quantity 𝑎 → 𝑎[ + 1]
𝑎
The frequency is scaled by defining a new frequency variable,𝑠1 = 𝑠/𝑎.
The magnitude is divided by 𝑎 to yield 0 dB at the break frequency.
Hence, the normalized and scaled function is (𝑠1 + 1).
To obtain the original frequency response, the magnitude and frequency are
multiplied by 𝑎. Aysa Jafari Farmand
Bode Plots - Asymptotic Approximations
3. First order factors (𝑠 + 𝑎)±1 ⟹ (𝑗𝜔 + 𝑎)±1

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations
3. First order factors (𝑠 + 𝑎)±1 ⟹ (𝑗𝜔 + 𝑎)±1

This max. difference of

the actual magnitude
curve from the
asymptotes is 3.01 dB,
and occurs at the
break frequency.

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations
3. First order factors (𝑠 + 𝑎)±1 ⟹ (𝑗𝜔 + 𝑎)±1

The maximum difference

for the actual phase curve
form the asymptotes is
5.71°, which occurs at the
decades above and
below the break
frequency.

Aysa Jafari Farmand

 Bode Plots - Asymptotic Approximations 2
𝑗𝜔 𝑗𝜔
4. Quadratic factors 𝐺 𝑠 = (𝑠 2 + 2𝜁𝑠 + 𝜔𝑛 2 )±1 ⟹ (𝜔𝑛 2 (1 + 2𝜁 + ))±1
𝜔𝑛 𝜔𝑛

• The low-frequency asymptote, 𝜔 ≪ 𝜔𝑛 :

• The high-frequency asymptote,. 𝜔 ≫ 𝜔𝑛 :
Note. The high-
freq. asymptote is
equal to the low-
freq. asymptote at
break frequency,
𝜔 = 𝜔𝑛 .

Here, we normalize the magnitude, dividing by 𝜔𝑛 2 ,

and scale the freq., dividing by 𝜔𝑛 . Aysa Jafari Farmand
 Bode Plots - Asymptotic Approximations 2
𝑗𝜔 𝑗𝜔
4. Quadratic factors 𝐺 𝑠 = (𝑠 2 + 2𝜁𝑠 + 𝜔𝑛 2 )±1 ⟹ (𝜔𝑛 2 (1 + 2𝜁 + ))±1
𝜔𝑛 𝜔𝑛

• The low-frequency asymptote, 𝜔 ≪ 𝜔𝑛 :

• The high-frequency asymptote,. 𝜔 ≫ 𝜔𝑛 :

Note. The phase at

the natural freq.,
𝜔𝑛 , is +90°.

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations 2
𝑗𝜔 𝑗𝜔
4. Quadratic factors 𝐺 𝑠 = (𝑠 2 + 2𝜁𝑠 + 𝜔𝑛 2 )±1 ⟹ (𝜔𝑛 2 (1 + 2𝜁 + ))±1
𝜔𝑛 𝜔𝑛
• The actual magnitude
and phase of the
second-order function
has a greater disparity,
which depends upon
the value of 𝜻.
• The normalized log-
magnitude at the
scaled natural freq. is
+ 20 log 2𝜁.

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations 2
𝑗𝜔 𝑗𝜔
4. Quadratic factors 𝐺 𝑠 = (𝑠 2 + 2𝜁𝑠 + 𝜔𝑛 2 )±1 ⟹ (𝜔𝑛 2 (1 + 2𝜁 + ))±1
𝜔𝑛 𝜔𝑛
The actual magnitude
and phase of the
second-order function
has a greater disparity,
which depends upon
the value of 𝜻.

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations 2
𝑗𝜔 𝑗𝜔
4. Quadratic factors 𝐺 𝑠 = (𝑠 2 + 2𝜁𝑠 + 𝜔𝑛 2 )±1 ⟹ (𝜔𝑛 2 (1 + 2𝜁 + ))±1
𝜔𝑛 𝜔𝑛

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations 2
𝑗𝜔 𝑗𝜔
4. Quadratic factors 𝐺 𝑠 = (𝑠 2 + 2𝜁𝑠 + 𝜔𝑛 2 )±1 ⟹ (𝜔𝑛 2 (1 + 2𝜁 + ))±1
𝜔𝑛 𝜔𝑛

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations
Ex. Bode Plots for Ratio of First-Order Factors
Draw the Bode plots for the following system, where
𝐺 𝑠 = 𝐾(𝑠 + 3)/[𝑠 𝑠 + 1 𝑠 + 2 ].
Soln. Bode plot is made for the open-loop function.
It is convenient to use the normalized plot for each term so that the low-frequency
asymptote of each term, except the pole at the origin, is at 0 dB, making it easier to
add the components of the Bode plot.

• The effect of 𝐾 is to move the magnitude curve up (increasing 𝐾) or down

(decreasing 𝐾) by the amount of 20 log𝐾. Choosing 𝐾 = 1 at first, the magnitude
plot can be denormalized later for any value of 𝐾.
• K has no effect upon the phase curve. Aysa Jafari Farmand
Bode Plots - Asymptotic Approximations
Ex. Bode Plots for Ratio of First-Order Factors
Soln. Cont.

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations
Ex. Bode Plots for Ratio of First-Order Factors
Soln. Cont.

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations
Ex. Bode Plots for Ratio of First-Order Factors
Soln. Cont.

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations
Ex. Bode Plots for Ratio of First-Order Factors
Soln. Cont.

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations
Ex. Bode Plots for Ratio of First- and Second-Order Factors
Draw the Bode plots for the following system, where
𝐺 𝑠 = 𝐾(𝑠 + 3)/[ 𝑠 + 2 (𝑠 2 + 2𝑠 + 25)].
Soln. Bode plot is made for the open-loop function.

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations
Ex. Bode Plots for Ratio of First- and Second-Order Factors
Soln. Cont.

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations
Ex. Bode Plots for Ratio of First- and Second-Order Factors
Soln. Cont.

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations
Ex. Bode Plots for Ratio of First- and Second-Order Factors
Soln. Cont.

Aysa Jafari Farmand

Bode Plots - Asymptotic Approximations
Ex. Bode Plots for Ratio of First- and Second-Order Factors
Soln. Cont.

Aysa Jafari Farmand