PCEE-612

Control Systems
GEE-2020
Lecture-07: Time –response Analysis
Note: The lectures are prepared referring recommended books that are listed in Introductory Lecture

06-Oct-22 Prof. J.S. Dhillon 1

Time –response Analysis
Time-domain analysis:
• In the analysis of a system, if the input, output and
Step Input
other variables in the system are represented as
functions of the time.
• The output of a system for the given input is the

Response
response of the system.
• Three characteristics are know to study the system
response of feedback system
• The transient response
Time, t
• The steady-state response
• Stability Steady state
Transient Response
The transient and steady state response of system are
studied with respect to response of the system to certain
standard signals.
• Unit Step signal
• Unit ramp signal
• Unit parabolic signal
• Unit impulse signal
• 06-Oct-22
Sinusoidal signal. Prof. J.S. Dhillon 2
Test Signals
Unit Step Signal: Unit Ramp Signal: u(t)
Displacement input signal u(t) Velocity input signal
1 ;𝑡 ≥ 0 𝑡 ;𝑡 ≥ 0
𝑢 𝑡 =ቊ 1 𝑢 𝑡 =ቊ
0 ;𝑡 < 0 0 ;𝑡 < 0
1 1
𝑈 𝑠 = 𝑈 𝑠 = 2
𝑠 t 𝑠
Derivative of unit step 0 Derivative of unit ramp
signal is unit impulse signal is unit step
signal. signal.

u(t) Unit Parabolic Signal:

Acceleration input signal
1 2
𝑢 𝑡 = ቐ2 𝑡 ;𝑡 ≥ 0
0 ;𝑡 < 0
1
𝑈 𝑠 = 3
𝑠
Derivative of unit parabolic
06-Oct-22 signal isProf.
unit ramp signal.
J.S. Dhillon 3
Test Signals
Unit Impulse Signal: u(t)
1
𝑢 𝑡 = ቐ𝑎 ; 0 < 𝑡 < 𝑎
0 ; 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
1
𝐴𝑟𝑒𝑎 = × 𝑎 = 1
𝑎
As the width a tends to 0, with the area remaining the
same , u(t) approaches to impulse signal.
𝑡
u(t)
න 𝛿 𝑡 𝑑𝑡 = 1
−∝ 1
∆ 𝑠 =1

Derivative of unit step signal is unit impulse signal.

t
0

06-Oct-22 Prof. J.S. Dhillon 4

Transient Response
𝐶 𝑠 𝐺 𝑠
=
𝑅 𝑠 1+𝐺 𝑠 𝐻 𝑠
−1
𝐺 𝑠
𝑐 𝑡 =ℒ 𝑅 𝑠
1+𝐺 𝑠 𝐻 𝑠
𝐺 𝑠
𝐶 𝑠 = 𝑅 𝑠
1+𝐺 𝑠 𝐻 𝑠

Characteristic equation 1 + 𝐺 𝑠 𝐻 𝑠 = 0
The roots of characteristic equation are called poles of the transfer function. The poles are called closed loop
poles.
𝐶(𝑠) 𝑏0 𝑠 𝑚 + 𝑏1 𝑠 𝑚−1 + 𝑏2 𝑠 𝑚−2 + ⋯ + 𝑏𝑚−1 𝑠 + 𝑎𝑚 𝑠 = 𝜎 + 𝑗𝜔
𝑇 𝑠 = = ;𝑛 ≥ 𝑚 𝜎 is called Neper frequency
𝑅(𝑠) 𝑎0 𝑠 𝑛 + 𝑎1 𝑠 𝑛−1 + 𝑎2 𝑠 𝑛−2 + ⋯ + 𝑎𝑛−1 𝑠 + 𝑎𝑛
𝜔 is called the radian frequency.
𝐾 𝑠 + 𝑧1 𝑠 + 𝑧2 … 𝑠 + 𝑧𝑚 𝐾 𝑠 + 𝑧1 𝑠 + 𝑧2 … 𝑠 + 𝑧𝑚 1
𝐶 𝑠 = 𝑅 𝑠 = × Poles and zeros are critical
𝑠 + 𝑝1 𝑠 + 𝑝2 … 𝑠 + 𝑝𝑛 𝑠 + 𝑝1 𝑠 + 𝑝2 … 𝑠 + 𝑝𝑛 𝑠 frequencies.
𝐾0 𝐾1 𝐾2 𝐾𝑛
𝐶(𝑠) = + ++ + ⋯+
𝑠 𝑠 − 𝑝1 𝑠 − 𝑝2 𝑠 − 𝑝𝑛
Poles decide the nature of transient
𝐾0 , 𝐾1 , 𝐾2 , … , 𝐾𝑛 are called residues response
𝑐 𝑡 = 𝐾0 + 𝐾1 𝑒 −𝑝1𝑡 + +𝐾2 𝑒 −𝑝2𝑡 + ⋯ + 𝐾𝑛 𝑒 −𝑝𝑛𝑡
06-Oct-22 Prof. J.S. Dhillon 5
Influence of closed loop poles on the Transient Response
Simple Pole at the origin
𝐾0
𝐶(𝑠) =
𝑠
Time domain response is 𝑐 𝑡 = 𝐾0

Simple Pole on negative real axis

𝐾1
𝐶(𝑠) =
𝑠 + 𝜎1
Time domain response is 𝑐 𝑡 = 𝐾1 𝑒 −𝜎1𝑡

Simple Pole on positive real axis

𝐾3
𝐶(𝑠) = K3
𝑠 − 𝜎3
Time domain response is 𝑐 𝑡 = 𝐾3 𝑒 −𝜎3𝑡
06-Oct-22 Prof. J.S. Dhillon 6
Influence of closed loop poles on the Transient Response
A pair of conjugate poles on the imaginary axis A pair of conjugate poles on the left half of s-plane

𝐾1 𝐾2
𝐶 𝑠 = + 𝐾1 𝐾2
𝑠 − 𝑗𝜔1 𝑠 + 𝑗𝜔1 𝐶 𝑠 = + ; 𝑝 = −𝛼 + 𝑗𝛽, 𝑝 ∗
= −𝛼 − 𝑗𝛽
∗ 1 1
𝑀 −𝑗𝜃 𝑀 +𝑗𝜃 𝑠 − 𝑝1 𝑠 − 𝑝1
𝐾1 = 𝑒 , 𝐾2 = 𝑒 𝑀 𝑀
2 2 𝐾1 = 𝑒 −𝑗𝜃 , 𝐾2 = 𝑒 +𝑗𝜃
𝑀 −𝑗𝜃 +𝑗𝜔 𝑡 𝑀 +𝑗𝜃 −𝑗𝜔 𝑡 2 2
𝑐 𝑡 = 𝑒 𝑒 1 + 𝑒 𝑒 1
𝑀 −𝑗𝜃 −𝛼+𝑗𝛽 𝑡 𝑀 +𝑗𝜃 −𝛼−𝑗𝛽 𝑡
2 2 𝑐 𝑡 = 𝑒 𝑒 + 𝑒 𝑒
𝑀 𝑗(𝜔 𝑡−𝜃) 𝑀 −𝑗(𝜔 𝑡−𝜃) 2 2
𝑐 𝑡 = 𝑒 1 + 𝑒 1
𝑀 −𝛼𝑡 𝑗(𝛽𝑡−𝜃)
2 2 𝑐 𝑡 = 𝑒 𝑒 + 𝑒 −𝑗(𝛽𝑡−𝜃)
= 𝑀 cos(𝜔1 𝑡 − 𝜃) 2
06-Oct-22 Dhillon = 𝑀𝑒 −𝛼𝑡 cos(𝛽𝑡 − 𝜃)
Prof. J.S.𝑐(𝑡) 7
Influence of closed loop poles on the Transient Response
A pair of conjugate poles on the right half of s-plane
𝐾1 𝐾2 ∗
𝐶 𝑠 = + ∗ ; 𝑝1 = 𝛼 + 𝑗𝛽, 𝑝1 = 𝛼 − 𝑗𝛽
𝑠 − 𝑝1 𝑠 − 𝑝1
𝑀 𝑀
𝐾1 = 𝑒 −𝑗𝜃 , 𝐾2 = 𝑒 +𝑗𝜃
2 2
𝑀 −𝑗𝜃 𝛼+𝑗𝛽 𝑡 𝑀 +𝑗𝜃 𝛼−𝑗𝛽 𝑡
𝑐 𝑡 = 𝑒 𝑒 + 𝑒 𝑒
2 2
𝑀 𝛼𝑡 𝑗(𝛽𝑡−𝜃)
𝑐 𝑡 = 𝑒 𝑒 + 𝑒 −𝑗(𝛽𝑡−𝜃)
2
𝑐(𝑡) = 𝑀𝑒 𝛼𝑡 cos(𝛽𝑡 − 𝜃)

The closed loop poles located on left half of s-plane → The response decays as time increases. The system
response is controllable and is said to be stable.
The closed loop poles located on imaginary axis→ The response is sinusoidal. The response is oscillatory and
marginally stable.
The closed loop poles located on right half of s-plane → The response increases as time increases. The system
response is not controllable and is said to be unstable.
06-Oct-22 Prof. J.S. Dhillon 8
Unit Step Response Of First Order System
𝐶(𝑠) 1 R(s) 1
= + C(s)
𝑅(𝑠) 𝜏𝑠 + 1 𝜏𝑠
Let unit step input
1 ;𝑡 ≥ 0
r t =𝑢 𝑡 =ቊ
0 ;𝑡 < 0 1
1
𝐶(𝑠) 𝜏𝑠
=
𝑅 𝑠 =𝑢 𝑠 = 𝑅(𝑠) 1 + 1 × 1
𝑠 𝜏𝑠
1 1 1 1
𝐶(𝑠) = × = 𝐶(𝑠)
𝜏𝑠 + 1 𝑠 𝑠(𝜏𝑠 + 1)
1 = 𝜏𝑠
1 𝑅(𝑠) 𝜏𝑠 + 1
𝐶(𝑠) = 𝐴= ቤ =1 𝜏𝑠
𝑠(𝜏𝑠 + 1) 𝜏𝑠 + 1 𝑠=0
𝐴 𝐵 𝐶(𝑠) 1
𝐶 𝑠 = + =
1 𝑅(𝑠) 𝜏𝑠 + 1
𝑠 𝜏𝑠 + 1 𝐵= ቤ = −𝜏
1 𝜏 𝑠 𝑠=−1
𝐶 𝑠 = − 𝜏
06-Oct-22 𝑠 𝜏𝑠 + 1 Prof. J.S. Dhillon 9
Unit Step Response Of First Order System
1 𝜏
𝐶 𝑠 = − R(s)
+ 1 C(s)
𝑠 𝜏𝑠 + 1
𝜏𝑠
Take Laplace Inverse
𝑡
−
c 𝑡 =1−𝑒 𝜏
𝐶(𝑠) 1
=
1 𝑡 𝑛−1 𝑅(𝑠) 𝜏𝑠 + 1
ℒ −1 𝑛
=
𝑠 (𝑛 − 1)!
1 𝑡 1−1 𝑡0
ℒ −1 1
= = =1 𝑡
𝑠 (1 − 1)! 0! −𝜏
e 𝑡 =𝑟 𝑡 −𝑐 𝑡 =1−𝑒
𝜏 𝜏 −
𝑡
ℒ −1 = ℒ −1 e 𝑡 =1− 1−𝑒 𝜏
𝜏𝑠 + 1 1
𝜏 𝑠+ 𝑡
𝜏 −
1 𝑡 e 𝑡 =𝑒 𝜏
−
ℒ −1 = 𝑒 𝜏
1
𝑠 +
06-Oct-22 𝜏 Prof. J.S. Dhillon 10
 𝑡
−𝜏
c 𝑡 =1−𝑒
(i) t = 0
𝑐 0 =1−1=0
(ii) t = 𝜏
𝑐 1 = 1 − 𝑒 −1 = 0.632
(iii) 𝑡 = ∞
𝑐 ∞ = 1 − 𝑒 −∝ = 1

𝑡
−𝜏
c 𝑡 =1−𝑒
Take Derivative w.r.t. ‘t’
𝑑 −𝜏
𝑡
𝑐ሶ 𝑡 = 1−𝑒
𝑑𝑡
−
𝑡 1
𝑐ሶ 𝑡 = −𝑒 𝜏 −
𝜏
1 −𝑡
𝑐ሶ 𝑡 = 𝑒 𝜏
06-Oct-22
𝜏 Prof. J.S. Dhillon 11
Unit Ramp Response Of First Order System
𝐶(𝑠) 1 R(s) 1
+ C(s)
=
𝑅(𝑠) 𝜏𝑠 + 1 𝜏𝑠
Let unit Ramp input
𝑡 ;𝑡 ≥ 0 1
r t =𝑢 𝑡 =ቊ 𝐶(𝑠) 1
0 ;𝑡 < 0 = 𝜏𝑠 =
1 𝑅(𝑠) 𝜏𝑠 + 1 𝜏𝑠 + 1
𝑅 𝑠 =𝑢 𝑠 = 2 𝜏𝑠
𝑠
1 1 1
𝐶(𝑠) = × 2= 2 1 = 𝐴 𝜏𝑠 + 1 + 𝐵𝑠 𝜏𝑠 + 1 + 𝐶𝑠 2
𝜏𝑠 + 1 𝑠 𝑠 (𝜏𝑠 + 1) Compare Constant terms
1 1=A
𝐶(𝑠) = 2
𝑠 (𝜏𝑠 + 1) Compare Coefficients of “s”
𝐴 𝐵 𝐶 0 = 𝐴𝜏 + 𝐵 ⇒ 𝐵 = −𝐴𝜏 ⇒ 𝐵 = −𝜏
𝐶 𝑠 = 2+ +
𝑠 𝑠 𝜏𝑠 + 1 Compare Coefficients of “s2”
1 𝜏 𝜏2 0 = 𝐵𝜏 + 𝐶 ⇒ 𝐶 = −𝐵𝜏 ⇒ 𝐶 = 𝜏 2
𝐶 𝑠 = 2− +
06-Oct-22 𝑠 𝑠 𝜏𝑠 + 1 Prof. J.S. Dhillon 12
Unit Ramp Response Of First Order System
1 𝜏 𝜏2 R(s)
𝐶 𝑠 = 2− + + 1 C(s)
𝑠 𝑠 𝜏𝑠 + 1 𝜏𝑠
Take Laplace Inverse
𝑡
−𝜏
c 𝑡 = 𝑡 − 𝜏 + 𝜏𝑒 𝐶(𝑠) 1
=
1 𝑡 𝑛−1 1 𝑡 𝑛−1 𝑅(𝑠) 𝜏𝑠 + 1
ℒ −1 𝑛
= ℒ −1 𝑛
=
𝑠 (𝑛 − 1)! 𝑠 (𝑛 − 1)! e 𝑡 =𝑟 𝑡 −𝑐 𝑡
1 𝑡 1−1 1 𝑡 2−1 𝑡
ℒ −1 = =1 ℒ −1 = =𝑡 −𝜏
𝑠 1 (1 − 1)! 𝑠 2 (2 − 1)! e 𝑡 = 𝑡 − 𝑡 − 𝜏 + 𝜏𝑒
𝑡
−𝜏
𝜏2 𝜏2 e 𝑡 =𝜏 1−𝑒
ℒ −1 = ℒ −1 0
𝜏𝑠 + 1 1 −𝜏
𝜏 𝑠+ e 0 =𝜏 1−𝑒 =0
𝜏
1 𝑡 ∝
−𝜏 −𝜏
𝜏ℒ −1 = 𝜏𝑒 e ∝ =𝜏 1−𝑒 =𝜏
1
𝑠+
06-Oct-22
𝜏 Prof. J.S. Dhillon 13
Unit Ramp Response Of First Order System
1 𝜏 𝜏2 e 𝑡 =𝑟 𝑡 −𝑐 𝑡 𝐶(𝑠)
=
1
𝐶 𝑠 = 2− + 𝑡 𝑅(𝑠) 𝜏𝑠 + 1
𝑠 𝑠 𝜏𝑠 + 1 e 𝑡 = 𝑡 − 𝑡 − 𝜏 + 𝜏𝑒 −
𝜏
Take Laplace Inverse 𝑡
−𝜏
𝑡
−𝜏 e 𝑡 =𝜏 1+𝑒 𝑒 −∝ = 0
c 𝑡 = 𝑡 − 𝜏 + 𝜏𝑒
Take the derivative w.r.t. “t” Initial Value
𝑑 𝑡
−𝜏
𝑡
−𝜏
𝑐ሶ 𝑡 = 𝑡 − 𝜏 + 𝜏𝑒 c 𝑡 = 𝑡 − 𝜏 + 𝜏𝑒
𝑑𝑡 (i) t = 0
−
𝑡 1 𝑐 0 = 0 − 𝜏 + 𝜏𝑒 0 = 0
𝑐ሶ 𝑡 = 1 + 𝜏𝑒 𝜏 −
𝜏 Final Value
𝑡 (ii) t =∝
−𝜏
𝑐ሶ 𝑡 = 1 − 𝑒 𝑐 ∝ =∝ −𝜏 − 𝜏𝑒 −∝ =∝
𝑐ሶ 0 = 1 − 𝑒 −0 = 0 (iii) t = 𝜏
𝑐ሶ 0 = 1 − 𝑒 −∝ = 1 𝑐 ∝ = 𝜏 − 𝜏 − 𝜏𝑒 −1 = −0.368𝜏
06-Oct-22 Prof. J.S. Dhillon 14
Unit Ramp Response Of First Order System
𝐶(𝑠) 1
1 =
𝐶(𝑠) = 2 𝑅(𝑠) 𝜏𝑠 + 1
𝑠 (𝜏𝑠 + 1)
𝑡
−
c 𝑡 = 𝑡 − 𝜏 + 𝜏𝑒 𝜏

Initial Value Final Value

(i) t = 0 (ii) t =∝
𝑐 0 = lim 𝑠𝐶(𝑠) 𝑐 ∝ = lim 𝑠𝐶(𝑠)
𝑠→∝ 𝑠→0
1 1
𝑐 0 = lim 𝑠 × 2 𝑐 ∝ = lim 𝑠 × 2
𝑠→∝ 𝑠 (𝜏𝑠 + 1) 𝑠→0 𝑠 (𝜏𝑠 + 1)
1 1
𝑐 0 = lim 𝑐 ∝ = lim
𝑠→∝ 𝑠(𝜏𝑠 + 1) 𝑠→0 𝑠(𝜏𝑠 + 1)
1 1 1
𝑐 0 = lim =0 𝑐 ∝ = lim =
𝑠→∝ 1 𝑠→0 0(0 + 1) 0
𝑠2 𝜏 +
𝑠 𝑐 ∝ =∝
06-Oct-22 Prof. J.S. Dhillon 15
 Unit Impulse Response
Unit Ramp Response

06-Oct-22 Prof. J.S. Dhillon 16

Unit impulse Response Of First Order System
𝐶(𝑠) 1 R(s) 1
+ C(s)
=
𝑅(𝑠) 𝜏𝑠 + 1 𝜏𝑠
Let unit impuse input
𝑡
1
න 𝑟 𝑡 𝑑𝑡 = 1 𝐶(𝑠) 𝜏𝑠 1
−∝ = =
𝑅 𝑠 =1 𝑅(𝑠) 𝜏𝑠 + 1 𝜏𝑠 + 1
1 1 𝜏𝑠
𝐶(𝑠) = ×1=
𝜏𝑠 + 1 (𝜏𝑠 + 1) 1 1
1 ℒ −1 = ℒ −1
𝜏𝑠 + 1 1
𝐶(𝑠) = 𝜏 𝑠+
(𝜏𝑠 + 1) 𝜏
Taking Laplace Inverse 1 −1 1 1 −𝑡
ℒ = 𝑒 𝜏
1 −𝑡 𝜏 1 𝜏
c 𝑡 = 𝑒 𝜏 𝑠+
𝜏
𝜏

06-Oct-22 Prof. J.S. Dhillon 17

Unit impulse Response Of First Order System
R(s)
1 −𝑡 + 1 C(s)
c 𝑡 = 𝑒 𝜏
𝜏 𝜏𝑠
Initial Value
1 −𝑡 𝐶(𝑠) 1
c 𝑡 =
𝜏
𝑒 𝜏 =
𝑅(𝑠) 𝜏𝑠 + 1 𝑒 −∝ = 0
(i) t = 0
1 0 1
𝑐 0 = 𝑒 =
𝜏 𝜏
Final Value
(ii) t =∝
1 −∝ 1 −∝
𝑐 ∝ = 𝑒 𝜏 = 𝑒 =0
𝜏 𝜏
(ii) t = 𝜏
1 −𝜏 1 −1 0.368
𝑐 𝜏 = 𝑒 𝜏= 𝑒 =
06-Oct-22
𝜏 𝜏 𝜏 Prof. J.S. Dhillon 18
First Order System
Unit Ramp Input: R(s) 1
𝑡 + C(s)
−𝜏 𝜏𝑠
𝑐1 𝑡 = 𝑡 − 𝜏 + 𝜏𝑒
Unit Step Input:
𝑡
−𝜏 𝐶(𝑠) 1
𝑐2 𝑡 = 1 − 𝑒 =
𝑅(𝑠) 𝜏𝑠 + 1
Unit Impulse Input:
1 −𝑡
𝑐3 𝑡 = 𝑒 𝜏
𝜏 𝑑 𝑑 𝑡
𝑑 𝑑 𝑡
−𝜏 𝑐3 𝑡 = 𝑐2 (𝑡) = 1−𝑒 −𝜏
𝑐2 𝑡 = 𝑐1 (𝑡) = 𝑡 − 𝜏 + 𝜏𝑒 𝑑𝑡 𝑑𝑡
𝑑𝑡 𝑑𝑡
𝑡 1 −
𝑡 1
𝑐2 𝑡 = 1 − 0 + 𝜏𝑒 −𝜏
− 𝑐3 𝑡 = 0 − 𝑒 𝜏 −
𝜏 𝜏
𝑡 1 −𝑡
𝑐2 𝑡 = 1 − 𝑒 −𝜏 𝑐3 𝑡 = 𝑒 𝜏
06-Oct-22 Prof. J.S. Dhillon
𝜏 19
Example: The impulse response of an initially relaxed system is
5𝑒 −10𝑡 . Compute its step response.
Solution: Impulse input = Derivative of step unit Input
Step Input = Integral of Impulse input
Response to Step Input = Integral of response to Impulse input
𝑡

𝑆𝑡𝑒𝑝 𝑅𝑒𝑠𝑝𝑜𝑛𝑠𝑒 = න 5𝑒 −10𝑡 𝑑𝑡

0
−10𝑡 𝑡
𝑒 𝑡
𝑆𝑡𝑒𝑝 𝑅𝑒𝑠𝑝𝑜𝑛𝑠𝑒 = 5 = −0.5 𝑒 −10𝑡 0
−10 0
𝑆𝑡𝑒𝑝 𝑅𝑒𝑠𝑝𝑜𝑛𝑠𝑒 = −0.5 𝑒 −10𝑡 − 𝑒 0 = −0.5 𝑒 −10𝑡 − 1
𝑆𝑡𝑒𝑝 𝑅𝑒𝑠𝑝𝑜𝑛𝑠𝑒 = 0.5 1 − 𝑒 −10𝑡

06-Oct-22 Prof. J.S. Dhillon 20

Example: When the input to a system was with drawn at t=0, its output was
found to decrease exponentially from 1000 units to 500 units in 1.386
seconds. What is time constant of system?
Solution: 𝑡
−𝜏
c t = 1000𝑒 1.386
1.386 τ=−
c 1.386 = 1000𝑒 − 𝜏 −0.693
1.386 τ=2
−
500 = 1000𝑒 𝜏
1.386
0.5 = 𝑒− 𝜏
1.386
−
ln 0.5 = ln 𝑒 𝜏
1.386
ln 0.5 = −
𝜏
1.386
τ=−
ln 0.5
06-Oct-22 Prof. J.S. Dhillon 21
Example: A first order system and its response is shown in Figure,
compute system parameters 𝛼 and k.
𝐶(𝑠) 𝑘 2
Solution: = 𝑅 𝑠 = R(s) C(s)
𝑅(𝑠) 𝑠 + 𝛼 𝑠 𝑘
𝐶(𝑠) 𝑘 𝑠+𝛼
=
𝑅(𝑠) 𝑠 + 𝛼
𝑘 2 2𝑘 𝐴 𝐵
𝐶(𝑠) = × = +
𝑠+𝛼 𝑠 𝑠 𝑠+𝛼 𝑠 𝑠+𝛼
2𝑘 2𝑘 = 𝐴(𝑠 + 𝛼) + 𝐵𝑠
𝐶(𝑠) = Put s=0
𝑠 𝑠+𝛼
𝐴 𝐵 2𝑘
𝐶 𝑠 = + 2𝑘 = 𝐴𝛼 ⇒ 𝐴 =
𝑠 𝑠+𝛼 𝛼
2𝑘 1 2𝑘 1 Put 𝑠 = −𝛼
𝐶 𝑠 = − 2𝑘
𝛼 𝑠 𝛼 𝑠+𝛼 2𝑘 = 𝐵 −𝛼 ⇒ 𝐵 = −
2𝑘 𝛼
𝑐 𝑡 = 1 − 𝑒 −𝛼𝑡
𝛼
06-Oct-22 Prof. J.S. Dhillon 22
Example: A first order system and its response is shown in Figure,
compute system parameters 𝛼 and k.
𝐶(𝑠) 𝑘 2
Solution: = 𝑅 𝑠 = R(s) C(s)
𝑅(𝑠) 𝑠 + 𝛼 𝑠 𝑘
2𝑘 𝑠+𝛼
𝑐 𝑡 = 1 − 𝑒 −𝛼𝑡
𝛼
2𝑘
𝑐ሶ 𝑡 = −𝑒 −𝛼𝑡 (−𝛼 )
𝛼
𝑐ሶ 𝑡 = 2𝑘𝑒 −𝛼𝑡

2𝑘
𝑐 ∝ = 1 − 𝑒 −𝛼×∝
𝛼
2𝑘
=2⇒𝛼=𝑘 𝛼=𝑘=5
𝛼
𝑐ሶ 0 = 2𝑘𝑒 −𝛼×0
2
= 2𝑘 ⇒ 𝑘 = 5
0.2 06-Oct-22 Prof. J.S. Dhillon 23
Unit Step Response of Second Order System
𝐶(𝑠) 𝐺(𝑠) r(t)=u(t) e(t)
= 𝜔𝑛2 c(t)
𝑅(𝑠) 1 + 𝐺 𝑠 𝐻(𝑠) +
1 R(s) E(s) 𝑠 𝑠 + 2𝜉𝜔𝑛 C(s)
𝑅 𝑠 = , 𝐻 𝑠 =1
𝑠
𝜔𝑛2
𝐶(𝑠) 𝑠 𝑠 + 2𝜉𝜔𝑛
=
𝑅(𝑠) 𝜔𝑛2
1+ ×1
𝑠 𝑠 + 2𝜉𝜔𝑛
𝜔𝑛2 1
𝜔𝑛2 𝐶 𝑠 = 2 2×𝑠
𝐶(𝑠) 𝑠 𝑠 + 2𝜉𝜔𝑛 𝑠 + 2𝜉𝜔𝑛 𝑠 + 𝜔𝑛
= 2 𝜔𝑛2 𝐴 𝐵𝑠 + 𝐶
𝑅(𝑠) 𝑠 + 2𝜉𝜔𝑛 𝑠 + 𝜔𝑛2
2 = 𝑠 +
𝑠 𝑠 + 2𝜉𝜔𝑛 2
𝑠 𝑠 + 2𝜉𝜔𝑛 𝑠 + 𝜔𝑛 𝑠 2 + 2𝜉𝜔𝑛 𝑠 + 𝜔𝑛2
𝐶(𝑠) 𝜔𝑛2
= 2
𝑅(𝑠) 𝑠 + 2𝜉𝜔𝑛 𝑠 + 𝜔𝑛2

06-Oct-22 Prof. J.S. Dhillon 24

Unit Step Response of Second Order System
𝜔𝑛2 𝐴 𝐵𝑠 + 𝐶 𝜔𝑛2
𝐶 𝑠 =
2 = 𝑠 + 𝑠 𝑠 2 + 2𝜉𝜔𝑛 𝑠 + 𝜔𝑛2
2
𝑠 𝑠 + 2𝜉𝜔𝑛 𝑠 + 𝜔𝑛 𝑠 2 + 2𝜉𝜔𝑛 𝑠 + 𝜔𝑛2
𝜔𝑛2 = 𝐴 𝑠 2 + 2𝜉𝜔𝑛 𝑠 + 𝜔𝑛2 + 𝐵𝑠 + 𝐶 𝑠 𝑠 2 + 2𝜉𝜔𝑛 𝑠 + 𝜔𝑛2
Put s = 0 = 𝑠 2 + 2 𝜉𝜔𝑛 𝑠 + 𝜉𝜔𝑛 2 − 𝜉𝜔𝑛 2
𝜔𝑛2 = 𝐴 𝜔𝑛2 ⇒ 𝐴 = 1 +𝜔𝑛2
Compare Coefficients of ‘ s2 ’ = 𝑠 + 𝜉𝜔𝑛 2 + 𝜔𝑛2 1 − 𝜉 2
0 = 𝐴 + 𝐵 ⇒ 𝐵 = −𝐴 ⇒ 𝐵 = −1
Compare Coefficients of ‘ s ’ 1 𝑠 + 𝜉𝜔𝑛 + 𝜉𝜔𝑛
0 = 2𝜉𝜔𝑛 𝐴 + 𝐶 ⇒ C = −2𝜉𝜔𝑛 𝐴 𝐶 𝑠 = −
𝑠 𝑠 + 𝜉𝜔𝑛 2 + 𝜔𝑛2 1 − 𝜉 2
C = −2𝜉𝜔𝑛
1 −𝑠 − 2𝜉𝜔𝑛
𝐶 𝑠 = + 2
𝑠 𝑠 + 2𝜉𝜔𝑛 𝑠 + 𝜔𝑛2
1 𝑠 + 2𝜉𝜔𝑛
𝐶 𝑠 = − 2
𝑠 𝑠 + 2𝜉𝜔𝑛 𝑠 + 𝜔𝑛2

06-Oct-22 Prof. J.S. Dhillon 25

Unit Step Response of Second Order System
1 𝑠 + 𝜉𝜔𝑛 𝜉𝜔𝑛
𝐶 𝑠 = − 2 +
𝑠 2
𝑠 + 𝜉𝜔𝑛 + 𝜔𝑛 1 − 𝜉 2 𝑠 + 𝜉𝜔𝑛 2 + 𝜔𝑛2 1 − 𝜉 2
𝜔𝑑
Let σ = 𝜉𝜔𝑛 and 𝜔𝑑2 = 𝜔𝑛2 1 − 𝜉 2 or 𝜔𝑛 =
1 − 𝜉2
σ is damping factor or damping constant σ = 𝜉𝜔𝑛 Time constant of
𝜔𝑑 is damping frequency or conditional 𝜎 𝐴𝑐𝑡𝑢𝑎𝑙 𝐷𝑎𝑚𝑝𝑖𝑛𝑔 under damped
frequency 𝜉= =
𝑤𝑛 𝐶𝑟𝑖𝑡𝑖𝑐𝑎𝑙 𝐷𝑎𝑚𝑝𝑖𝑛𝑔 Response,
𝜔𝑛 is natural un-damped frequency 𝐴𝑐𝑡𝑢𝑎𝑙 𝐷𝑎𝑚𝑝𝑖𝑛𝑔 1 1
𝜉 is damping ratio
𝜉=
𝐶𝑟𝑖𝑡𝑖𝑐𝑎𝑙 𝐷𝑎𝑚𝑝𝑖𝑛𝑔
τ= =
𝜎 𝜉𝜔𝑛

1 𝑠+𝜎 𝜉 𝜔𝑑
𝐶 𝑠 = − 2 2 + × 2 + 𝜔2
𝑠 𝑠 + 𝜎 + 𝜔𝑑 1−𝜉 2 𝑠 + 𝜎 𝑑

−𝜎𝑡
𝜉
𝑐 𝑡 =1− 𝑒 cos 𝜔𝑑 𝑡 + 𝑒 −𝜎𝑡 sin 𝜔𝑑 𝑡
1 − 𝜉2
06-Oct-22 Prof. J.S. Dhillon 26
Unit Step Response of Second Order System σ = 𝜉𝜔𝑛

𝜉 𝜔𝑑
𝜔𝑛 =
𝑐 𝑡 = 1 − 𝑒 −𝜎𝑡 cos 𝜔𝑑 𝑡 + 𝑒 −𝜎𝑡 sin 𝜔𝑑 𝑡 1 − 𝜉2
1 − 𝜉2
𝑒 −𝜎𝑡 1
𝑐 𝑡 =1− 1 − 𝜉 2 cos 𝜔𝑑 𝑡 + 𝜉 sin 𝜔𝑑 𝑡
1 − 𝜉2 1 − 𝜉2
𝜙
𝑒 −𝜎𝑡
𝑐 𝑡 =1− sin 𝜙 cos 𝜔𝑑 𝑡 + cos 𝜙 sin 𝜔𝑑 𝑡 𝜉
1− 𝜉2
𝑒 −𝜎𝑡 sin 𝜙 = 1 − 𝜉2
𝑐 𝑡 =1− sin 𝜔𝑑 𝑡 + 𝜙
1−𝜉 2 cos 𝜙 = 𝜉
1 − 𝜉2 1 − 𝜉2
Where 𝜙= tan−1 tan 𝜙 =
𝜉 𝜉

06-Oct-22 Prof. J.S. Dhillon 27

Transient Response Specifications

Maximum
Overshoot

Delay Rise Peak Settling

time time time time

06-Oct-22 Prof. J.S. Dhillon 28

Peak Time (𝑡𝑝) σ = 𝜉𝜔𝑛
𝜔𝑑
𝑐ሶ 𝑡 = 0 tan 𝜙 =
1 − 𝜉2 𝜔𝑛 =
𝜉 1 − 𝜉2
𝑑 𝑒 −𝜎𝑡
1− sin 𝜔𝑑 𝑡 + 𝜙 =0
𝑑𝑡 1 − 𝜉2
𝑒 −𝜎𝑡 1
0− 𝑐𝑜𝑠 𝜔𝑑 𝑡 + 𝜙 𝜔𝑑 − 𝑒 −𝜎𝑡 −𝜎 sin 𝜔𝑑 𝑡 + 𝜙 =0
1 − 𝜉2 1 − 𝜉2
𝑒 −𝜎𝑡
− 𝑐𝑜𝑠 𝜔𝑑 𝑡 + 𝜙 𝜔𝑑 + 𝜎 sin 𝜔𝑑 𝑡 + 𝜙 = 0
1 − 𝜉2
𝜎 sin 𝜔𝑑 𝑡 + 𝜙 = 𝑐𝑜𝑠 𝜔𝑑 𝑡 + 𝜙 𝜔𝑑
sin 𝜔𝑑 𝑡 + 𝜙 𝜔𝑑
=
cos 𝜔𝑑 𝑡 + 𝜙 𝜎

06-Oct-22 Prof. J.S. Dhillon 29

 𝜔𝑑
Peak Time (𝑡𝑝) 𝜔𝑛 =
1 − 𝜉2
𝑐ሶ 𝑡 = 0
1 − 𝜉2
𝜔𝑛 1 − 𝜉 2 tan 𝜙 =
1 𝜉
tan 𝜔𝑑 𝑡 + 𝜙 =
𝜉𝜔𝑛 1 − 𝜉2 σ = 𝜉𝜔𝑛
𝜙
tan 𝜔𝑑 𝑡 + 𝜙 = tan 𝜙
𝜉
tan 𝜔𝑑 𝑡 + 𝜙 = tan 𝑛𝜋 + 𝜙
𝜔𝑑 𝑡 + 𝜙 = 𝑛𝜋 + 𝜙
𝜔𝑑 𝑡 = 𝑛𝜋
𝑛𝜋 𝜋 Maximum overshoot occurs at n = 1 and 0<𝜉<1
𝑡= = n = 1, 3, 5, 7, …. Overshoot
𝜔𝑑 𝜔𝑛 1 − 𝜉 2 n = 2, 4, 6, 8, …. Undershoot
𝜋 𝜋
𝑡𝑝 = =
𝜔𝑑 𝜔𝑛 1 − 𝜉 2
06-Oct-22 Prof. J.S. Dhillon 30
Periodic overshoots and undershoots

06-Oct-22 Prof. J.S. Dhillon 31

 𝜔𝑑
Rise Time (𝑡𝑟) 𝜔𝑛 =
1 − 𝜉2
• The rise time, 𝑡𝑟 is defined as time required for the step response to
reach from 0% to 90% of its final value for overdamped systems 1 − 𝜉2
tan 𝜙 =
and 0% to 100% for underdamped systems. 𝜉

• The rise time, 𝑡𝑟 is equal to reciprocal of slope response at the σ = 𝜉𝜔𝑛

instant, the response is half the final value
1
𝑐(𝑡)ቚ =1 𝑡𝑟 =
𝑡=𝑡𝑟
ሶ ȁ𝑡=𝑡𝑑
𝑐(𝑡)
𝑒 −𝜎𝑡𝑟
1− sin 𝜔𝑑 𝑡𝑟 + 𝜙 =1
1 − 𝜉2
𝑒 −𝜎𝑡𝑟
sin 𝜔𝑑 𝑡𝑟 + 𝜙 =0
1 − 𝜉2
sin 𝜔𝑑 𝑡𝑟 + 𝜙 = 0
sin 𝜔𝑑 𝑡𝑟 + 𝜙 = sin 𝜋
06-Oct-22 Prof. J.S. Dhillon 32
 𝜔𝑑
Rise Time (𝑡𝑟) 𝜔𝑛 =
1 − 𝜉2
𝜔𝑑 𝑡𝑟 + 𝜙 = 𝜋 1 − 𝜉2
tan 𝜙 =
𝜔𝑑 𝑡𝑟 = 𝜋 − 𝜙 𝜉
𝜋−𝜙 𝜋−𝜙 1 − 𝜉2
𝑡𝑟 = = where 𝜙 = tan−1 σ = 𝜉𝜔𝑛
𝜉
𝜔𝑑 𝜔𝑛 1 − 𝜉 2 and 0<𝜉<1

06-Oct-22 Prof. J.S. Dhillon 33

 𝜔𝑑
Maximum Overshoot (MP) 𝜔𝑛 =
1 − 𝜉2
𝑐𝑚𝑎𝑥 = Maximum output value
𝑀𝑃 = 𝑐𝑚𝑎𝑥 − 𝑐𝑆𝑆 𝑐𝑆𝑆 = Steady state value σ = 𝜉𝜔𝑛
𝑐𝑚𝑎𝑥 − 𝑐𝑆𝑆
%𝑀𝑃 = × 100 𝑛𝜋
𝑐𝑆𝑆 𝑡=
𝜔𝑑
Measure of relative stability of system
Maxima occurs at first overshoot sin 𝜙 = 1 − 𝜉2

𝑒 −𝜎𝑡
𝑐 𝑡 =1− sin 𝜔𝑑 𝑡 + 𝜙
1 − 𝜉2
𝑡 → 𝑡𝑝
𝑒 −𝜎𝑡𝑝
𝑐𝑚𝑎𝑥 = 𝐶 𝑡𝑝 = 1 − sin 𝑛𝜋 + 𝜙
1 − 𝜉2

06-Oct-22 Prof. J.S. Dhillon 34

 𝑛𝜋 𝜔𝑑
Maximum Overshoot (MP) 𝑡= 𝜔𝑛 =
𝜔𝑑 1 − 𝜉2
𝑐𝑚𝑎𝑥 = Maximum output value
𝑀𝑃 = 𝑐𝑚𝑎𝑥 − 𝑐𝑆𝑆 𝑐𝑆𝑆 = Steady state value 𝑛𝜋
𝜔𝑑 𝑡𝑝 = 𝜔𝑑 σ = 𝜉𝜔𝑛
𝑒 −𝜎𝑡𝑝 𝜔𝑑
𝑛𝜋
𝑐𝑚𝑎𝑥 = 𝐶 𝑡𝑝 = 1 − sin 𝑛𝜋 + 𝜙 𝑡𝑝 =
1−𝜉 2 𝜔𝑑
𝑛𝜋𝜉 𝑛𝜋
− 𝜎𝑡𝑝 = 𝜉𝜔𝑛
1−𝜉 2 𝜔 1 − 𝜉 2
𝑒 𝑛
𝑛
𝑐𝑚𝑎𝑥 = 1 − −1 1 − 𝜉2 𝑛𝜋𝜉
1 − 𝜉2 𝜎𝑡𝑝 =
1 − 𝜉2
𝑛𝜋𝜉
−
𝑐𝑚𝑎𝑥 = 1 − −1 𝑒 𝑛 1−𝜉 2 sin 𝑛𝜋 + 𝜙 = −1 𝑛 sin 𝜙
Peak overshoot occurs, n = 1 sin 𝜙 = 1 − 𝜉 2
−
𝑛𝜋𝜉 sin 𝑛𝜋 + 𝜙 = −1 𝑛 1 − 𝜉 2
𝑛𝜋𝜉
𝑐𝑚𝑎𝑥 = 1 + 𝑒 1−𝜉 2 −
𝑛𝜋𝜉 1−𝜉2
− %𝑀𝑝 = 𝑒
𝑀𝑃 06-Oct-22
= 𝑐𝑚𝑎𝑥 − 1 = 𝑒 1−𝜉 2
Prof. J.S. Dhillon 35
Settling Time (𝑡𝑠 )
• The settling time,𝑡𝑠 is defined as time required for the step response to decrease
and stay within the specified tolerance band (2% or 5%) of final value.
𝑒 −𝜎𝑡
For ±2% Tolerance For ±5% Tolerance 𝑐 𝑡 =1− sin 𝜔𝑑 𝑡 + 𝜙
1 − 𝜉 2
𝑒−𝜎𝑡𝑠 = 0.02 −𝜎𝑡
𝑒 𝑠 = 0.05
−𝜎𝑡𝑠 = ln 0.02 −𝜎𝑡𝑠 = ln 0.05 𝜔𝑑 = 2𝜋𝑓𝑑
𝜔𝑑
ln 0.02 ln 0.05 𝑓𝑑 =
𝑡𝑠 = − 𝑡𝑠 = − 2𝜋
𝜎 𝜎
−3.912 −2.99 For ±2% Tolerance
𝑡𝑠 = − 𝑡𝑠 = −
𝜎 𝜎 4𝑓𝑑
3.91 4 2.99 3 𝑡𝑠 𝑓𝑑 =
𝑡𝑠 = ≅ 𝑡𝑠 = ≅ 𝜔𝑛 𝜉
𝜎 𝜎 𝜎 𝜎
4 3 For ±5% Tolerance
𝑡𝑠 = = 4𝜏 𝑡𝑠 = = 3𝜏 3𝑓𝑑
𝜔𝑛 𝜉 𝜔𝑛 𝜉 𝑡𝑠 𝑓𝑑 =
06-Oct-22 Prof. J.S. Dhillon
𝜔𝑛 𝜉 36
Delay Time (𝑡𝑑 )
• The delay time,𝑡𝑑 is defined as time required for the step response to reach 50% of
its final value in its first attempt.
−𝜎𝑡
1 + 0.7𝜉 𝑒
𝑡𝑑 = 𝑐 𝑡 =1− sin 𝜔𝑑 𝑡 + 𝜙
𝜔 1 − 𝜉2
𝑛

Final value 𝑡𝑑 = −ln 0.5

lim 𝑐(𝑡) = 1
𝑡→∝
𝑡𝑑 = ln 0.5 −1
50% 𝑜𝑓 𝑐 𝑡 = 1 − 𝑒 −𝑡𝑑 𝑡𝑑 = ln 2
0.5 = 1 − 𝑒 −𝑡𝑑 𝑡𝑑 = 0.693 𝑠
𝑒 −𝑡𝑑 = 0.5
−𝑡𝑑 = ln 0.5
𝑡𝑑 = −ln 0.5
𝑡𝑑 = − −0.693 = 0.693
06-Oct-22 Prof. J.S. Dhillon 37
 Continued….
Note: The lectures are prepared referring recommended books that are listed in Introductory Lecture

06-Oct-22 Prof. J.S. Dhillon 38