ECE4510/5510: Feedback Control Systems.

3–1

DYNAMIC RESPONSE

3.1: System response in the time domain

■ We can now model dynamic systems with differential equations.
What do these equations mean?
■ We’ll proceed by looking at a system’s response to certain inputs in
the time domain.
■ Then, we’ll see how the Laplace transform can make our lives a lot
easier by simplifying the math.
■ This will give insights into how we might specify the way the system
should respond.
■ Finally, we’ll preview how adding dynamics (e.g., a controller) can
change how the system responds.

Some important input signals

■ Several signals recur throughout this course.
■ The unit step function: 1(t)
!
1, t ≥ 0;
1(t) =
0, otherwise.
t
■ The unit ramp function: r (t)
!
t, t ≥ 0;
r(t) =
0, otherwise.
t
c 1998–2013, Gregory L. Plett and M. Scott Trimboli
Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–2
■ The unit parabola function:
p(t)
⎧ 2
⎨t
, t ≥ 0;
p(t) = 2
⎩
0, otherwise.
t
■ The cosine/sine functions:
cos(t) sin(t)

t t

■ The (ideal) impulse function, δ(t):

• Very strange “generalized” function, defined only under an integral.

δ(t) = 0, t ̸= 0 zero duration

% ∞
δ(t) dt = 1. unit area.
−∞
δ(t)

t
Symbol
• Sifting property:1
% ∞
x(τ )δ(t − τ ) dτ = x(t).
−∞

1
Assumes that x(t) is continuous at t = τ . Interpretation: no value of x(t) matters
except that over the short range where δ(t) occurs.

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–3
Time response of a linear time invariant system

■ Let y(t) be the output of an LTI system with input x(t).

y(t) = T[x(t)]
&% ∞ '
=T x(τ )δ(t − τ ) dτ (sifting)
−∞
% ∞
= x(τ )T[δ(t − τ )] dτ . (linear)
−∞
Let h(t, τ ) = T[δ(t − τ )]
% ∞
= x(τ )h(t, τ ) dτ
−∞
If the system is time invariant, h(t, τ ) = h(t − τ )
% ∞
= x(τ )h(t − τ ) dτ (time invariant)
−∞
△
= x(t) ∗ h(t).
■ The output of an LTI system is equal to the convolution of its impulse
response with the input.
■ This makes life EASY (TRUST me!)

EXAMPLE : Finding an impulse response:

■ Consider a first-order system, ẏ(t) + ky(t) = u(t).
■ Let y(0−) = 0, u(t) = δ(t).
■ For positive time we have ẏ(t) + ky(t) = 0. Recall from your
differential-equation math course: y(t) = Aest , solve for A, s.

ẏ(t) = Asest

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–4

Asest + k Aest = 0

s+k = 0

s = −k.
■ We have solved for s; now, solve for A.
% 0+ % 0+ % 0+
ẏ(t) dt + k y(t) dt = δ(t) dt
− − −
( 0 )* + ( 0 )* + ( 0 )* +
+ 0 1
y(t)|00−

y(0+) − y(0−) = 1
+
Ae−k0 − 0 = 1

A = 1.

■ Response to impulse: h(t) = e−kt , t > 0.

■ h(t) = e−kt 1(t).
■ Response of this system to general input:
% ∞
y(t) = h(τ )u(t − τ ) dτ
−∞
% ∞
= e−kτ 1(τ )u(t − τ ) dτ
−∞
% ∞
= e−kτ u(t − τ ) dτ .
0

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–5
3.2: Transfer functions
■ Response to impulse = “impulse response”: h(t).
■ Response to general input = messy convolution: h(t) ∗ u(t).
■ To choose a simpler example, what is the response to a cosine?
A , j ωt -
A cos(ωt) = e + e− j ωt
2
Break it down: What is the response to an exponential?
■ Let u(t) = est , where s is complex.
% ∞ % ∞
y(t) = h(τ )u(t − τ ) dτ = h(τ )es(t−τ ) dτ
−∞ −∞
% ∞
= h(τ )est e−sτ dτ
−∞
% ∞
st
=e h(τ )e−sτ dτ
( −∞ )* +
Transfer function, H (s)

= est H (s).
■ An est input decouples the convolution into two independent parts: a
part depending on est and a part depending on h(t).
EXAMPLE : ẏ(t) + ky(t) = u(t) = est :
but , y(t) = H (s)est , ẏ(t) = s H (s)est ,

s H (s)est + k H (s)est = est

1
H (s) = (I never integrated!)
s +k
est
y(t) = .
s +k
c 1998–2013, Gregory L. Plett and M. Scott Trimboli
Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–6
Response to a cosinusoid (revisited)

Let s= jω u(t)=e j ωt y(t)=H ( jω)e j ωt

s=− jω u(t)=e− j ωt y(t)=H (− jω)e− j ωt
A. /
u(t)=A cos(ωt) y(t)= H ( jω)e j ωt + H (− jω)e− j ωt
2

△
Now, H ( jω) = Me j φ

H (− jω) = Me− j φ (can be shown for h(t) real)

AM . j (ωt+φ) /
y(t) = e + e− j (ωt+φ)
2
= AM cos(ωt + φ).
■ The response of an LTI system to a sinusoid is a sinusoid! (of the
same frequency).

EXAMPLE : Frequency response of our first order system:

1
H (s) =
s+k
1
H ( jω) =
jω + k
1
M = |H ( jω)| = √
ω2 + k 2
0ω 1
−1
φ = ̸ H ( jω) = − tan
k
A 0 0 11
−1 ω
y(t) = √ cos ωt − tan .
ω2 + k 2 k
■ Can we use these results to simplify convolution and get an easier
way to understand dynamic response?

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–7
Defining the Laplace L− transform
■ We have seen that if a system has an impulse response h(t), we can
compute a transfer function H (s),
% ∞
H (s) = h(t)e−st dt.
−∞
■ Since we deal with causal systems (possibly with an impulse at
t = 0), we can integrate from 0− instead of negative infinity.
% ∞
H (s) = h(t)e−st dt.
0−
■ This is called the one-sided (uni-lateral) Laplace transform of h(t).

Laplace Transforms of Common Signals

Name Time function, f (t) Laplace tx., F (s)
Unit impulse δ(t) 1
1
Unit step 1(t)
s
1
Unit ramp t · 1(t)
s2
n!
nth order ramp t n · 1(t)
s n+1
1
Exponential exp(−at)1(t)
s+a
1
Ramped exponential t exp(−at)1(t)
(s + a)2
b
Sine sin(bt)1(t)
s 2 + b2
s
Cosine cos(bt)1(t)
s 2 + b2
b
Damped sine e−at sin(bt)1(t)
(s + a)2 + b2
s+a
Damped cosine e−at cos(bt)1(t)
(s + a)2 + b2
2bs
Diverging sine t sin(bt)1(t)
(s 2 + b2 )2
s 2 − b2
Diverging cosine t cos(bt)1(t)
(s 2 + b2 )2
c 1998–2013, Gregory L. Plett and M. Scott Trimboli
Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–8
Properties of the Laplace transform
■ Superposition: L {a f 1(t) + b f 2(t)} = a F1(s) + bF2(s).
■ Time delay: L { f (t − τ )} = e−sτ F(s).
1 0s 1
■ Time scaling: L { f (at)} = F .
|a| a
(useful if original equations are expressed poorly in time scale. e.g.,
measuring disk-drive seek speed in hours).
■ Differentiation:
2 3
L f˙(t) = s F(s) − f (0−)
2 3
L f (t) = s 2 F(s) − s f (0−) − f˙(0−)
¨
2 (m) 3
L f (t) = s m F(s) − s m−1 f (0−) − . . . − f (m−1)(0−).
4% t 5
1
■ Integration: L f (τ ) dτ = F(s).
0− s
■ Convolution: Recall that y(t) = h(t) ∗ u(t)

Y (s) = L {y(t)} = L {h(t) ∗ u(t)}

4% t 5 t
=L h(τ )u(t − τ ) dτ τ =t
τ =0−
% ∞ % t
= h(τ )u(t − τ ) dτ e−st dt
t=0− τ =0−
% ∞ % ∞
= h(τ )u(t − τ ) e−st dt dτ . τ
τ =0− t=τ −
Region of
integration
■ Multiply by e−sτ esτ
% ∞ % ∞
−sτ
Y (s) = h(τ )e u(t − τ )e−s(t−τ ) dt dτ .
τ =0− t=τ −
c 1998–2013, Gregory L. Plett and M. Scott Trimboli
Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–9

Let t ′ = t − τ :
% ∞ % ∞
−sτ ′
Y (s) = h(τ )e dτ u(t ′)e−st dt ′
τ =0− t ′ =0−
Y (s) = H (s)U (s).

■ The Laplace transform “unwraps” convolution for general input

signals. Makes system easy to analyze.
■ This is the most important property of the Laplace transform. This is
why we use it. It converts differential equations into algebraic
equations that we can solve quite readily.

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–10
3.3: The inverse Laplace transform

■ The inverse Laplace transform converts F(s) → f (t).

■ Once we get an intuitive feel for F(s), we won’t need to do this often.
■ The main tool for ILT is partial-fraction-expansion.
b0s m + b1s m−1 + · · · + bm
Assume : F(s) = n
s + a1s n−1 + · · · + an
6m
(s − zi ) (zeros)
= k 6ni =1
➠
i =1 (s − pi ) (poles)
c1 c2 cn
= + +··· + if { pi } distinct.
s − p1 s − p2 s − pn
c2(s − p1) cn (s − p1 )
so, (s − p1 )F(s) = c1 + +··· +
s − p2 s − pn
let s = p1 : c1 = (s − p1)F(s)|s= p1

ci = (s − pi )F(s)|s= pi
n
7 . / 1
f (t) = ci e pi t 1(t) since L ekt 1(t) = .
i =1
s−k

5 5
EXAMPLE : F(s) = = .
s 2 + 3s + 2 (s + 1)(s + 2)
8 5 88
8
c1 = (s + 1)F(s)8 = 8 =5
s=−1 s + 2 s=−1
8 5 88
8
c2 = (s + 2)F(s)8 = 8 = −5
s=−2 s+1 s=−2

f (t) = (5e−t − 5e−2t )1(t).

■ If F(s) has repeated roots, we must modify the procedure. e.g.,
repeated three times:

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–11
k
F(s) =
(s − p1)3(s − p2) · · ·
c1,1 c1,2 c1,3 c2
= + + + + ···
s − p1 (s − p1)2 (s − p1)3 s − p2
3
8
c1,3 = (s − p1) F(s)8s= p1
& 8
d , 3
-8
c1,2 = (s − p1) F(s) 88
ds s= p1
& 2 8
1 d , 3
- 8
c1,1 = 2
(s − p1 ) F(s) 88
2 ds s= p1
& i 8
1 d , k
-8
8
cx,k−i = i
(s − p i ) F(s) 8 .
i! ds s= pi

EXAMPLE : Find the ILT of

s+2 A B C
H (s) = 2
= + 2
+ .
(s + 1) (s + 3) s + 1 (s + 1) s+3
■ We start with B, 8
s + 3 88 1
B= = .
s + 3 8s=−1 2
■ Next, we find A,
& 9 :8
d s + 2 88
A=
ds s + 3 8s=−1
& 8
d 8
−1 8
= (s + 2)(s + 3) 8
ds s=−1
. −2
8
−1 8
= (s + 2)(−1)(s + 3) + (s + 3) s=−1
& 8
s+2 1 88
= − +
(s + 3)2 s + 3 8s=−1
1 1 1
=− + = .
4 2 4
c 1998–2013, Gregory L. Plett and M. Scott Trimboli
Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–12
■ Lastly, we find C, 8
s + 2 88 1
C= = − .
(s + 1)2 8s=−3 4
■ Therefore, the inverse Laplace transform we are looking for is
& '
1 1 1
h(t) = te−t + e−t − e−3t 1(t).
2 4 4
s+3
EXAMPLE : Find ILT of .
(s + 1)(s + 2)2
■ ans: f (t) = (2e−t − 2e−2t − (te)*
−2t
+)1(t).
from repeated root.
■ Note that this is quite tedious, but MATLAB can help.
5
■ Try MATLAB with two examples; first, F(s) = .
s 2 + 3s + 2

Example 1. Example 2.

>> Fnum = [0 0 5]; >> Fnum = [0 0 1 3];

>> Fden = [1 3 2]; >> Fden = conv([1 1],conv([1 2],[1 2]));

[r,p,k] = residue(Fnum,Fden); [r,p,k] = residue(Fnum,Fden);

r = -5 r = -2

5 -1

p = -2 2

-1 p = -2

k = [] -2

-1

k = []

■ When you use “residue” and get repeated roots, BE SURE to type
“help residue” to correctly interpret the result.

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–13
Using the Laplace transform to solve problems

■ We can use the Laplace transform to solve both homogeneous and

forced differential equations.

EXAMPLE : ÿ(t) + y(t) = 0, y(0−) = α, ẏ(0−) = β.

s 2Y (s) − αs − β + Y (s) = 0

Y (s)(s 2 + 1) = αs + β
αs + β
Y (s) = 2
s +1
αs β
= 2 + 2 .
s +1 s +1
From tables, y(t) = [α cos(t) + β sin(t)]1(t).
■ If initial conditions are zero, things are very simple.

EXAMPLE :
ÿ(t)+5 ẏ(t)+4y(t) = u(t), y(0−) = 0, ẏ(0−) = 0, u(t) = 2e−2t 1(t).
2
s 2Y (s) + 5sY (s) + 4Y (s) =
s+2
2
Y (s) =
(s + 2)(s + 1)(s + 4)
−1 2/3 1/3
= + + .
s+2 s +1 s+4
& '
2 1
From tables, y(t) = −e−2t + e−t + e−4t 1(t).
3 3

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–14
3.4: Time response versus pole locations

■ If we wish to know how a system responds to some input (for

example, an impulse response, or a step response), it seems like we
need to do the following:

1. Find the Laplace transform U (s) of the input u(t),

2. Find the Laplace transform of the output Y (s) = H (s)U (s),
3. Find the time response by taking the inverse Laplace transform of
Y (s). That is, y(t) = L−1(Y (s)).
■ This is true if we want a precise, quantitative answer.
■ But, if we’re interested only in a qualitative answer, we can learn a lot
simply by looking at the pole locations of the transfer function.
■ If we can represent H (s) = num H (s)/den H (s) and
U (s) = numU (s)/denU (s), then we have
num H (s)numU (s)
Y (s) =
den H (s)denU (s)
7 rk
= ,
k
s + p k

where “pole” s = − pk is a root of either den H (s) or denU (s).

■ So, some of the system’s response is due to the poles of the input
signal, and some is due to the poles of the plant.
■ Here, we’re interested in the contribution due to the poles of the plant.

• Neglecting the residues rk , which simply scale the output by some

fixed amount, we’re interested in “what does an output of the type
1
look like?”
s + pk
c 1998–2013, Gregory L. Plett and M. Scott Trimboli
Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–15

• That is, the poles qualitatively determine the behavior of the

system; zeros (equivalently, residues) quantify this relationship.
■ Note that the poles pk may be real, or they may occur in
complex-conjugate pairs.
■ So, in the next sections, we look at the time responses of real poles
and of complex-conjugate poles.

Time response due to a real pole

■ Consider a transfer function having a single real pole:
1
H (s) = ➠ h(t) = e−σ t 1(t).
s+σ
■ If σ > 0, pole is at s < 0, STABLE i.e., impulse response decays, and
any bounded input produces bounded output.
■ If σ < 0, pole is at s > 0, UNSTABLE.
■ σ is “time constant” factor: τ = 1/σ .
impulse([0 1],[1 1]); step([0 1],[1 1]);
1 1

0.8 0.8
K (1 − e−t/τ )
−σ t
e System response. K = dc gain
y(t) × K

0.6 0.6
h(t)

0.4 1 0.4
←−
e
Response to initial condition
0.2 0.2
−→ 0.
0 0
0 1 2 3 4 5 0 1 2 3 4 5
t =τ t =τ
Time (sec × τ ) Time (sec × τ )
Time response due to complex-conjugate poles
■ Now, consider a second-order transfer function having
complex-conjugate poles
c 1998–2013, Gregory L. Plett and M. Scott Trimboli
Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–16

b0 ωn2
H (s) = 2 =K 2 .
s + a1 s + a2 s + 2ζ ωn s + ωn2
( )* +
standard form
ζ = damping ratio.

ωn = natural frequency or undamped frequency.

ωn
h(t) = ; e−σ t (sin(ωd t)) 1(t),
1 − ζ2
where, σ = ζ ωn ,
;
ωd = ωn 1 − ζ 2 = damped frequency.

I(s) I(s) I(s) I(s)

−1 ◦ ◦ ◦
θ = sin (ζ ) 45 30 17.5

ωn
R(s) R(s) R(s) R(s)
ωd
σ

ζ = 0.707 ζ = 0.5 ζ = 0.3

Impulse Response
1

0.5
e−σ t
h(t)

0
Envelope of sinusoid decays as e−σ t
−0.5
−e−σ t

−1
0 5 10 15 20 25 30
Time (sec)

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–17
Impulse Responses of 2nd-Order Systems Step Responses of 2nd-Order Systems
1 2
ζ =0 ζ =0
0.2 0.2
0.5 0.4 1.5
0.4
0.6
0.6
y(t)

y(t)
0 1
ζ =1 0.8

−0.5 0.5 0.8

1.0

−1 0
0 2 4 6 8 10 12 0 2 4 6 8 10 12
ωn t ωn t

■ Low damping, ζ ≈ 0, oscillatory; High damping, ζ ≈ 1, no oscillations.

I(s) I(s)

R(s) R(s)

Impulse responses vs. pole locations Step responses vs. pole locations

■ 0<ζ <1 underdamped.

■ ζ =1 critically damped, ζ > 1 over-damped.

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–18
3.5: Time-domain specifications
■ We have seen impulse and step responses for first- and second-order
systems.
■ Our control problem may be to specify exactly what the response
SHOULD be.
■ Usually expressed in terms of the step response.

tp Mp

1
0.9

0.1
tr t
ts

■ tr = Rise time = time to reach vicinity of new set point.

■ ts = Settling time = time for transients to decay (to 5 %, 2 %, 1 %).
■ M p = Percent overshoot.
■ t p = Time to peak.

Rise Time
■ All step responses rise in roughly the same amount of time (see
pg. 3–17.) Take ζ = 0.5 to be average.
➠ time from 0.1 to 0.9 is approx ωn tr = 1.8:
1.8
tr ≈ .
ωn
c 1998–2013, Gregory L. Plett and M. Scott Trimboli
Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–19
■ We could make this more accurate, but note:

• Only valid for 2nd-order systems with no zeros.

• Use this as approximate design “rule of thumb” and iterate design

until spec. is met.

Peak Time and Overshoot

■ Step response can be found from ILT of H (s)/s.

9 :
σ
y(t) = 1 − e−σ t cos(ωd t) + sin(ωd t) ,
ωd

;
ωd = ωn 1 − ζ 2 , σ = ζ ωn .

■ Peak occurs when ẏ(t) = 0

9 :
σ
ẏ(t) = σ e−σ t cos(ωd t) + sin(ωd t) − e−σ t (−ωd sin(ωd t) + σ cos(ωd t))
ωd
9 2 :
σ
= e−σ t sin(ωd t) + ωd sin(ωd t) = 0.
ωd
■ So, 100
90

ωd t p = π, 80
70
π π
tp = = ; Text . 60
M p, %

ωd ωn 1 − ζ 2 50

√ 40
−ζ π/ 1−ζ 2
■ Mp = e × 100. 30
20

■ (common values: M p = 16% for 10

0
ζ = 0.5; M p = 5% for ζ = 0.7). 0 0.2 0.4 0.6 0.8 1.0
ζ

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–20
Settling Time

■ Determined mostly by decaying exponential

e−ωn ζ ts = ϵ ... ϵ = 0.01, 0.02, or 0.05

EXAMPLE :

ϵ = 0.01 ϵ ts
e−ωn ζ ts = 0.01 0.01 ts = 4.6/σ
0.02 ts = 3.9/σ
ωn ζ ts = 4.6
0.05 ts = 3.0/σ
4.6 4.6
ts = =
ζ ωn σ
Design synthesis

■ Specifications on tr , ts , M p determine pole locations.

■ ωn ≥ 1.8/tr .
■ ζ ≥ fn(M p ). (read off of ζ versus M p graph on page 3–19)
■ σ ≥ 4.6/ts . (for example—settling to 1%)

I(s) I(s) I(s) I(s)

σ
ωn sin−1 ζ
R(s) R(s) R(s) R(s)

EXAMPLE : Converting specs. to s-plane

■ Specs: tr ≤ 0.6, M p ≤ 10%, ts ≤ 3 sec. at 1%
c 1998–2013, Gregory L. Plett and M. Scott Trimboli
Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–21
■ ωn ≥ 1.8/tr = 3.0 rad/sec.
■ From graph of M p versus ζ, ζ ≥ 0.6.
■ σ ≥ 4.6/3 = 1.5 sec.
3

1
I(s)
0

−1

−2

−3
−5 −4 −3 −2 −1 0 1
R(s)
EXAMPLE : Designing motor compensator
■ Suppose a servo-motor system for a pen-plotter has transfer function
0.5K a ωn2
= 2 .
s 2 + 2s + 0.5K a s + 2ζ ωn s + ωn2
■ Only one adjustable parameter K a , so can choose only one spec: tr ,
ts or M p ➠ Allow NO overshoot.
■ M p = 0, ζ = 1.
■ From transfer fn: 2 = 2ζ ωn ➠ ωn = 1.
■ ωn2 = 12 = 0.5K a , K a = 2.0
■ Note: ts = 4.6 seconds. We will need a better controller than this for a
pen plotter!

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–22
3.6: Time response vs. pole locations: Higher order systems

■ We have looked at first-order and second-order systems without

zeros, and with unity gain.

Non-unity gain
■ If we multiply by K , the dc gain is K . tr , ts , M p , t p are not affected.

Add a zero to a second-order system

2 2(s + 1.1)
H1(s) = H2(s) =
(s + 1)(s + 2) 1.1(s
9 + 1)(s + 2) :
2 2 2 0.1 0.9
= − = +
s+1 s +2 1.1 s + 1 s + 2
0.18 1.64
= +
s+1 s+2

■ Same dc gain (at s = 0).

■ Coefficient of (s + 1) pole GREATLY reduced.
■ General conclusion: a zero “near” a pole tends to cancel the effect of
that pole.
■ How about transient response?
s
αζ ωn
+1
H (s) = .
(s/ωn )2 + 2ζ s/ω n +1
• Zero at s = −ασ .

• Poles at R(s) = −σ .

■ Large α, zero far from poles ➠ no effect.

■ α ≈ 1, large effect.
c 1998–2013, Gregory L. Plett and M. Scott Trimboli
Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–23
■ Notice that the overshoot goes up as α → 0.
2nd-order system with zero Overshoot versus normalized zero loc.
2 2
Step Response

1.5 1.5
ζ = 0.3
0.5

Mp
1 1
α=1 0.7
2
0.5 4 0.5

100
0 0
0 2 4 6 8 10 0 2 4 6 8 10
ωn t α

■ A little more analysis; set ωn = 1

s
αζ
+1
H (s) =
s 2 + 2ζ s + 1
9 :
1 1 s
= 2 +
s + 2ζ s + 1 αζ s 2 + 2ζ s + 1
= Ho (s) + Hd (s).

■ Ho (s) is the original response, without the zero.

■ Hd (s) is the added term due to the zero. Notice that
1
Hd (s) = s Ho(s).
αζ
The time response is a scaled version of the derivative of the time
response of Ho (s).
■ If any of the zeros in RHP, system is nonminimum phase.

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–24
2nd-order min-phase step resp. 2nd-order nonmin-phase step resp.
2 1.5

H (s) 1
1.5 Ho (s)
0.5
1
Ho (s) H (s)
y(t)

y(t)
0
0.5
−0.5

0
Hd (s)
Hd (s) −1

−0.5 −1.5
0 2 4 6 8 10 0 2 4 6 8 10
Time (sec) Time (sec)

Add a pole to a second order system

1
H (s) = 0 1 .
s
αζ ωn
+ 1 [(s/ωn )2 + 2ζ s/ωn + 1]
■ Original poles at R(s) = −σ = −ζ ωn .
■ New pole at s = −αζ ωn .
■ Major effect is an increase in rise time.
2nd-order system with pole Norm. rise time vs. norm. pole loc.
1.4 9

1.2 8
Step Response

7
1
ζ = 1.0
6
0.8 0.7
ωn tr

100 5
0.6 0.5
0.4
5 4

3
0.2
2
2

0
α=1 1
0 1 2 3 4 5 6 7 8 0 2 4 6 8 10
ωn t α

Summary of higher-order approximations

■ Extra zero in LHP will increase overshoot if the zero is within a factor
of ≈ 4 from the real part of complex poles.
c 1998–2013, Gregory L. Plett and M. Scott Trimboli
Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–25
■ Extra zero in RHP depresses overshoot, and may cause step
response to start in wrong direction. DELAY .
■ Extra pole in LHP increases rise-time if extra pole is within a factor of
≈ 4 from the real part of complex poles.

I(s)
Unstable
Insignificant Dominant region
R(s)

■ MATLAB ‘step’ and ‘impulse’ commands can plot higher order system
responses.
■ Since a model is an approximation of a true system, it may be all right
to reduce the order of the system to a first or second order system. If
higher order poles and zeros are a factor of 5 or 10 time farther from
the imaginary axis.

• Analysis and design much easier.

• Numerical accuracy of simulations better for low-order models.

• 1st- and 2nd-order models provide us with great intuition into how
the system works.
• May be just as accurate as high-order model, since high-order
model itself may be inaccurate.

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–26
3.7: Changing dynamic response
■ Topic of the rest of the course.
■ Important tool: block diagram manipulation.

Block-diagram manipulation
■ We have already seen block diagrams (see pg. 1–4).
■ Shows information/energy flow in a system, and when used with
Laplace transforms, can simplify complex system dynamics.
■ Four BASIC configurations:

U (s) H (s) Y (s) Y (s) = H (s)U (s)

U (s) H1 (s) H2 (s) Y (s) Y (s) = [H2 (s)H1(s)] U (s)

H1 (s)

U (s) Y (s) Y (s) = [H1 (s) + H2(s)] U (s)

H2 (s)

U1 (s) = R(s) − Y2 (s)

U1 (s) Y2 (s) = H2 (s)H1(s)U1 (s)

R(s) H1(s) Y (s)
so, U1 (s) = R(s) − H2 (s)H1(s)U1 (s)
R(s)
=
H2(s) 1 + H2 (s)H1(s)
Y2 (s) U2 (s)
Y (s) = H1 (s)U1(s)
H1 (s)
= R(s)
1 + H2 (s)H1(s)
c 1998–2013, Gregory L. Plett and M. Scott Trimboli
Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–27
■ Alternate representation
H (s)
U (s) Y (s)

H1(s) H2(s)
U (s) Y (s)

H1(s)
U (s) Y (s)
H2(s)
H1(s)
R(s) Y (s)
−H2(s)
EXAMPLE : Recall dc generator dynamics from page 2–19
i f (t) R f Ra L a i a (t)

e f (t) Lf eg (t) ea (t) Zl

( )* + ( )* +( )* +
Field Rotor Load
circuit circuit circuit
i f (t) eg (t) i a (t)
e f (t) Field Kg Rotor Zl ea (t)
circuit circuit

■ Compute the transfer functions of the four blocks.

d
e f (t) = R f i f (t) + L f i f (t) eg (t) = K g i f (t)
dt
E f (s) = R f I f (s) + L f s I f (s) E g (s) = K g I f (s)
I f (s) 1 E g (s)
= . = Kg.
E f (s) Rf + L fs I f (s)

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–28

d
eg (t) = Ra i a (t) + L a i a (t) + ea (t)
ea (t) = i a (t)Z l dt
E g (s) = Ra Ia (s) + L a s Ia (s) + E a (s)
E a (s) = Z l Ia (s)
E a (s) = (Ra + L a s + Z l ) Ia (s)
= Zl . Ia (s) 1
Ia (s) = .
E g (s) L a s + Ra + Z l
■ Put everything together.
E a (s) E a (s) Ia (s) E g (s) I f (s)
=
E f (s) Ia (s) E g (s) I f (s) E f (s)
K g Zl
=, - .
L f s + R f (L a s + Ra + Z l )
Block-diagram algebra

U (s) H (s) Y1 (s) U (s) H (s) Y1 (s)

⇐⇒
1
Y2 (s) Y2 (s)
H (s)

U1 (s) H (s) Y (s) U1 (s) H (s) Y (s)

⇐⇒
U2 (s)
U2 (s) H (s)

1
R(s) H1(s) Y (s) R(s) H2(s) H1(s) Y (s)
H2 (s)
⇐⇒
H2(s)
“Unity Feedback”

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝
ECE4510/ECE5510, DYNAMIC RESPONSE 3–29

EXAMPLE : Simplify:
H6 (s)

R(s) H1 (s) H2 (s) H5 (s) Y (s)

H3 (s)

H4 (s)

H6 (s)

H1(s)
R(s) H2 (s) H5 (s) Y (s)
1 − H1 (s)H3(s)

H4 (s)

H6(s)
H2(s)

H1(s)
R(s) H2 (s) H5 (s) Y (s)
1 − H1 (s)H3(s)

H4 (s)

( 9
)* :
+( )* +
H1 (s)H2(s) H (s)
1−H1(s)H3 (s) H5(s)+ H6(s)
9 : 2
H1(s)H2 (s)H4 (s)
1+
1−H1(s)H3 (s)

H1 (s)H2(s)H5(s) + H1 (s)H6(s)
R(s) Y (s)
1 − H1 (s)H3(s) + H1(s)H2(s)H4(s)

c 1998–2013, Gregory L. Plett and M. Scott Trimboli

Lecture notes prepared by and copyright ⃝