Software Engineering 3DX3

Slides 2: Modelling in the Frequency

Domain

Dr. Ryan Leduc

Department of Computing and Software

McMaster University

Material based on lecture notes by P. Taylor and M. Lawford, Control Systems Engineering (4th ed.) by N.
Nise, and Calculus with Analytic Geometry (3rd ed.) by Larson et al.
c
°2006 R.J. Leduc 1
Block Diagram Representation of a System
I Differential equations can be used to represent relationship
between input and output of a system.
I Problem: system parameters, and input (r(t)) and output
(c(t)) appear throughout equation.
I Prefer to represent system as in Fig. 2.1(a) below where input
and output are separate.
Figure 2.1

Want to be able to represent

system as series of cascading
subsystems, which can easily
be combined together.

This can not be achieved with

differential equations.
c
°2006 R.J. Leduc 2
Laplace Transform Review

I Helps us understand the dynamic behaviour of processes.

I Essential for: stability analysis, block diagrams, and controller

design.

I Converts differential equations (time domain) into algebraic

equations (s-domain).

time domain Laplace domain time domain

inverse
differential Laplace algebraic Laplace time domain
equations transform equations transform solution

c
°2006 R.J. Leduc 3
Laplace Transform Definition
Z ∞
L{f (t)} = f (t)e−st dt = F (s)
0

I L is the Laplace transform operator.

I s = σ + jω is the Laplace transform variable and is a complex

number.

I A sufficient condition for existence of Laplace transform is

function is piecewise continuous in time over interval of
integration.

I F (s) contains no information about f (t) for time t < 0.

I When doing the inverse Laplace transform, we can represent

this by multiplying f (t) by u(t), the unit step function.
c
°2006 R.J. Leduc 4
Common Laplace Transforms

Table 2.1

I Table shows the Laplace

transform for several
common signals.

I Rather than evaluating

Laplace transform using
its definition, common to
use table and apply
various theorems.

c
°2006 R.J. Leduc 5
Important Properties of Laplace Transform

Linearity: L{k1 f1 (t) ± k2 f2 (t)} = k1 F1 (s) ± k2 F2 (s)

Differentiation:

df (t)
L{ } = sF (s) − f (0−)
dt

d2 f (t)
L{ } = s2 F (s) − sf (0−) − f 0 (0−)
dt2
Frequency Shifting: L{e−at f (t)} = F (s + a)

c
°2006 R.J. Leduc 6
Laplace Transform Theorems

Table 2.2

Commonly
used theorems.

c
°2006 R.J. Leduc 7
The Transfer Function
I We are now ready to represent system in the form of the
diagram below.

I We will separate the system into the following distinct parts:

system input, output, and transfer function.

I In general, an nth order linear, time-invariant differential

equation is of the form:
dn c(t) dn−1 c(t)
an + an−1 + · · · + a0 c(t)
dtn dtn−1
dm r(t) dm−1 r(t)
= bm + b m−1 + · · · + b0 r(t)
dtm dtm−1

I c(t) is output, r(t) is input, and differential

equation represents system’s behavior.
c
°2006 R.J. Leduc 8
The Transfer Function - II
I Taking Laplace transform of both sides, gives algebraic
equation.
an sn C(s) + an−1 sn−1 C(s) + · · · + a0 C(s) + init terms for c(t)
= bm sm R(s) + bm−1 sm−1 R(s) + · · · + b0 R(s) + init terms for r(t)
I If we assume initial conditions are zero, we can reduce this to:
(an sn + an−1 sn−1 + · · · + a0 )C(s)
= (bm sm + bm−1 sm−1 + · · · + b0 )R(s)
I
C(s) bm sm + bm−1 sm−1 + · · · + b0
= G(s) =
R(s) an sn + an−1 sn−1 + · · · + a0
I G(s) is called the transfer function.

I Can find ouput by

C(s) = R(s)G(s).
c
°2006 R.J. Leduc 9
Inverse Laplace Transform

I Once we have solved for the output as a function of the input

in the s-domain, we typically want to convert this back into
the time domain.

I We use the inverse Laplace transform:

Z σ+jω
−1 1
L {F (s)} = F (s)est ds
2πj σ−jω

I Normally, people look up up F (s) in tables, and make use of

the theorems.

c
°2006 R.J. Leduc 10
Partial Fraction Expansion
I The Laplace transform model of a system will typically be of
form:
N (s)
F (s) =
D(s)
I where N (s) is an nth order polynomial in s and D(s) is an
mth order polynomial in s.

I For example:
s2 + 2s − 3
F (s) =
s5 + s4 − s − 1
I We know from algebra that all polynomials that have real
coefficients can be factored into linear and irreducible
quadratic factors.

I For example: s5 + s4 − s − 1 = (s − 1)(s + 1)2 (s2 + 1)

c
°2006 R.J. Leduc 11
Partial Fraction Expansion - II

I Using partial fraction decomposition and this factorization, we

can write the transfer function as follows:
s2 + 2s − 3
F (s) =
(s − 1)(s + 1)2 (s2 + 1)
A B C Ds + E
= + + 2
+ 2
s − 1 s + 1 (s + 1) s +1
I To complete the decomposition for this example, we would
then determine values for A, B, C, D, and E that satisfy the
equation.

I We can then apply the linearity and the frequency shifting

properties to easily find the inverse Laplace transform.

c
°2006 R.J. Leduc 12
 N (s)
Steps for Decomposing D(s) into Partial Fractions

1. Divide if Improper: If N (s)

D(s) improper fraction (ie. degree of
N (s) is ≥ degree of D(s)) then divide N (s) by D(s) to get

N (s) N1 (s)
= (a polynomial) +
D(s) D(s)
N1 (s)
where D(s) is not an improper fraction. We would then apply
N1 (s)
the following steps to D(s) .

2. Factor Denominator: Factor denominator into factors of

form:

(ps + q)m and (as2 + bs + c)n

where as2 + bs + c can not be further reduced.

c
°2006 R.J. Leduc 13
 N (s)
Steps for Decomposing D(s) - II

3. Linear Factors: For factor (ps + q)m , include the sum below:

A1 A2 Am
+ 2
+ ··· +
(ps + q) (ps + q) (ps + q)m

4. Quadratic Factors: For factor (as2 + bs + c)n , include the

sum below:
B1 s + C1 B2 s + C2 Bn s + Cn
2
+ 2 2
+ ··· +
(as + bs + c) (as + bs + c) (as2 + bs + c)n

5. Determine Unkowns: Equate original fraction to the sums

found in steps 3 and 4, and solve for unknowns.

c
°2006 R.J. Leduc 14
Quadratic Factors e.g.
3
Find: L−1 { s(s2 +2s+5) }

For a quadratic of the form as2 + bs + c, we can use the quadratic

equation to determine its factors.
√
−b ± b2 − 4ac
s=
2a
For us, this gives complex factors, thus the term is irreducible.
p
−2 ± 22 − 4(1)(5)
s= = −1 ± 2j
2(1)

3 A Bs + C
= + 2
s(s2 + 2s + 5) s s + 2s + 5

c
°2006 R.J. Leduc 15
Quadratic Factors e.g. - II
Basic Equation:

3 = A(s2 + 2s + 5) + s(Bs + C) = (A + B)s2 + (2A + C)s + 5A

3
For s = 0, we get: 3 = 5A thus A = 5

Substituting A in gives: 3 = ( 53 + B)s2 + ( 65 + C)s + 3

Equating coefficients of powers of s gives:

3 6
( + B) = 0 and ( + C) = 0
5 5
We thus have B = − 53 and C = − 65 .

We thus have:
3
3 5 − 35 s + − 65 3
5 3 s+2
2
= + 2
= + (− ) 2
s(s + 2s + 5) s s + 2s + 5 s 5 s + 2s + 5
c
°2006 R.J. Leduc 16
Quadratic Factors e.g. - III
The quadratic term is the sum of the Laplace transforms of
frequency shifted sine and cosine.

This gives:
K1 (s+a) K2 ω
L{K1 e−at cosωt} = (s+a)2 +ω 2
L{K2 e−at sinωt} = (s+a)2 +ω 2
.

Adding the transforms together, we get:

K1 (s + a) + K2 ω
L{K1 e−at cosωt + K2 e−at sinωt} =
(s + a)2 + ω 2

We next note that (s + a)2 + ω 2 = s2 + 2as + (a2 + ω 2 )

Equating to our denominator (s2 + 2s + 5), gives: 2a = 2 and

(a2 + ω 2 ) = 5

We thus have a = 1 and ω = 2.

c
°2006 R.J. Leduc 17
Quadratic Factors e.g. - IV
We now equate our numerators and get:

s + 2 = K1 (s + 1) + 2K2 = K1 s + (K1 + 2K2 )

1
This gives us K1 = 1 and K2 = 2

We thus have:

s+2 (s + 1) + ( 12 )2
=
s2 + 2s + 5 (s + 1)2 + 22

Putting it all together:

3
3 s+2 3 3 1
L−1 { 5 + (− ) 2 } = − e−t (cos2t + sin2t)
s 5 s + 2s + 5 5 5 2

We note that the factors of s2 + 2s + 5 were s = −1 ± 2j

= −a ± ωj.
c
°2006 R.J. Leduc 18
Stability Analysis Using Roots of D(s)
I The roots of D(s) correspond to the exponential terms in the
time domain response for G(s), the transfer function.

N (s) N (s)
G(s) = =
D(s) (s + p1 )(s + p2 ) · · · (s + pn )
A1 A2 An
= + + ··· +
(s + p1 ) (s + p2 ) (s + pn )

We refer to the roots of D(s) as poles.

Solving for g(t) gives:

A1 A2 An
g(t) = L−1 { } + L−1 { } + · · · + L−1 { }
(s + p1 ) (s + p2 ) (s + pn )
= A1 e−p1 t + A2 e−p2 t + · · · + An e−pn t

c
°2006 R.J. Leduc 19
Stability Analysis: Real Roots
Considering: g(t) = A1 e−p1 t + A2 e−p2 t + · · · + An e−pn t

If we have pole(s) at s = −pi = −σi ± jωi :

I If ωi = 0, then pole is strictly real and corresponds to Ai e−σi t

If σi > 0, then pole in left side of imaginary plane, and

response decreases to zero over time and system is stable.

i.e. σi = 2, then we get

s = −2, pi = 2 and
response
Ai e−σi t = Ai e−2t .

c
°2006 R.J. Leduc 20
Stability Analysis: Real and Imaginary Roots
I If ωi = 0 and σi < 0, then pole in right side of imaginary
plane, and response increases over time and system is
unstable. i.e. σi = −1, then we get s = 1, pi = −1 and
response Ai e−σi t = Ai et .
I If σi = 0 and ωi 6= 0, then we have two pure imaginary roots.
This corresponds to a sinusoidal response with no damping,
technically considered stable.

i.e. ωi = 2, then we get

s = ±2j, pi = ±2j and
response
Ai sinωi t = Ai sin2t.

c
°2006 R.J. Leduc 21
Stability Analysis: Complex Roots
If we have σi 6= 0 and ωi 6= 0, we have complex roots.
I For poles at s = −σi ± jωi we get the partial fraction
expansion below:
α + jβ α − jβ
+
s + σi + jωi s + σi − jωi
I This results in the time domain response of the form:

e−σi t [2α cosωi t + 2β sinωi t]

If σi > 0, then response is stable.

i.e. σi = −2 and ωi = 3,
then we get s = 2 ± 3j,
pi = −2 ± 3j and response
e−σi t [2α cosωi t + 2β sinωi t]
= e2t [2α cos3t + 2β sin3t].

c
°2006 R.J. Leduc 22
Time Functions Associated with s-plane

Figure 2.5 from Dorf and Bishop, Modern Control Systems (10th Edition), Prentice-Hall, 2004.

c
°2006 R.J. Leduc 23
PID Controller Intro
I Transfer function for a proportional-plus-integral-plus-
derivative (PID) controller is:

K2 K1 s + K2 + K3 s2 K3 (s2 + K
K3 s +
1 K2
K3 )
Gc (s) = K1 + +K3 s = =
s s s

Figure 9.30: PID Controller.

c
°2006 R.J. Leduc 24
Electric Network Transfer Function
I We will generalize notion of resistance to impedance.

I This will allow us to treat capacitors and inductors in similar

manner as resistors in analyzing circuits.

I Recall that Ohms law says: v(t) = Ri(t)

I Taking Laplace Transform of both sides gives: V (s) = RI(s)

V (s)
I We define the impedance to be Z(s) = I(s) =R

I(s) 1
I We also define the admittance to be: Y (s) = V (s) = R =G

c
°2006 R.J. Leduc 25
Impedance of Inductor

I The voltage-current relation for an inductor is: v(t) = L di(t)

dt .

I To find transfer function, take Laplace transform of each side

and assume zero initial conditions:

V (s) = LsI(s)
V (s)
I The impedance is thus Z(s) = I(s) = Ls

c
°2006 R.J. Leduc 26
Impedance of Capacitor
I The voltage-current relation for a capacitor is:
Z
1 t
v(t) = i(τ )dτ
C 0
I Taking derivative of both sides gives:
dv(t) 1
= i(t)
dt C
I Taking Laplace transform (assuming zero initial conditions)
gives:
1
sV (s) = I(s)
C
V (s) 1
I The impedance is thus Z(s) = I(s) = Cs

c
°2006 R.J. Leduc 27
Summary of Circuit Elements.
I Table 2.3 below shows for each element type the
voltage-current, current-voltage, and voltage-charge
relationship, as well as the elements impedance and
admittance.

I All elements here are passive as they do not contain an

internal energy source.

c
°2006 R.J. Leduc 28
Equivalent Resistance and Impedance.
I It can be shown that resistance in serial can be replaced by an
equivalent resistor with resistance Rs calculated as follows:
Rs = R1 + R2 + · · · + RN
I This can be generalized to impedance as follows:
Zs = Z1 + Z2 + · · · + ZN
I The equivalent resistance (Rp ) and impedance (Zp ) when
elements are connected in parallel is:
1 1 1 1 1 1 1 1
Rp = R1 + R2 + · · · + RN Zp = Z1 + Z2 + · · · + ZN

Figures 2.16 and 2.18 from Electric Circuit Analysis, D.E. Jounson et al., 1989.
c
°2006 R.J. Leduc 29
Equivalent Resistance Example
Figure (a) to (b): Rs1 = 1 + 5 = 6Ω
1 1 1
Figure (b) to (c): Rp = 12 + 6 ⇒ Rp = 4Ω

Figure (c) to (d): Rs2 = 7 + 4 = 11Ω

Figure 2.19 from Electric Circuit Analysis, D.E. Jounson et al., 1989.

c
°2006 R.J. Leduc 30
Kirchhoff’s Current Law
I A point of connection between two or more circuit elements is
referred to as a node.
I Circuit show in Fig. (a) contains three nodes and is electrically
equivalent to that of Figure (b).
I Kirchoff’s current law (KCL) says that the algebraic sum of
the currents entering any node is zero.
I Currents flowing into the node are considered positive, and
those leaving negative.
I For node on right, this would give: 5 + i − (−3) − 2 = 0

Figures 2.8 and 2.9 from Electric Circuit Analysis, D.E. Jounson et al., 1989.
c
°2006 R.J. Leduc 31
Kirchhoff’s Voltage Law
I Kirchoff’s voltage law (KVL) says that the algebraic sum of
voltages around any closed path is zero.

I In the direction we traverse the path, voltages that go from −

to + (lower to higher potential) are considered to be positive,
and those going from + to − to be negative.

I Traversing the circuit below in clockwise direction gives:

5 − v − 10 + 2 = 0
I Solving for v gives v = −3V .

Figure 2.11 from Electric Circuit Analysis, D.E. Jounson et al., 1989.

c
°2006 R.J. Leduc 32
Kirchhoff’s Voltage Law eg.

VC (s)
I Find transfer function relation V (s) .

Figures 2.4 and 2.5.

c
°2006 R.J. Leduc 33
Mesh Analysis
I Replace passive elements with their impedances, and all
sources and time variables with their Laplace transform.

I Assume a transform current and direction in each mesh.

I Assume for each element a voltage polarity

I Assume a current I3 (s) and

direction through the shared
segment (ie. inductor)

I Apply Kirchoff’s voltage law to

each mesh going in direction of
the mesh current.

I Use Kirchoff’s current law to

relate the currents. Figure 2.6.
c
°2006 R.J. Leduc 34
Mesh Analysis eg.
I2 (s)
I Find transfer function relation V (s) .

c
°2006 R.J. Leduc Figures 2.6. 35
Cramer’s Rule

Cramer’s Rule: If Ax = B is a system of n linear equations in n

unknowns such that det(A) 6= 0, then the system has a unique
solution. The solution is:
det(A1 ) det(A2 ) det(An )
x1 = , x2 = , · · · , xn =
det(A) det(A) det(A)

where Aj is the matrix obtained by replacing the entries of the j th

column of A by the entries in the matrix B.

Definition taken from Elementary Linear Algebra with Applications by H. Anton et al. , 1987.

c
°2006 R.J. Leduc 36
Matrix Determinant
The formula of a determinant for a 2 × 2 and a 3 × 3 matrix:

Definition taken from Elementary Linear Algebra with Applications by H. Anton et al. , 1987.

Mnemonic device for remembering determinant formula.

Figure 2.2 from Elementary Linear Algebra with Applications by H. Anton et al. , 1987.

c
°2006 R.J. Leduc 37
Nodal Analysis
I Transform circuit into Laplace Domain.
I Assume for each element a voltage polarity.
I Determine nodes for circuit. Choose one (treat like ground) -
others relative to it.
I Assume transform currents and directions entering and leaving
each node.

I Use Kirchoff’s current law to

create equations for each node
with an unknown voltage.

I Use relation I(s) = VZ(s)

(s)
to get
equation for currents in terms of
voltages.
c
°2006 R.J. Leduc 38
Nodal Analysis eg.
VC (s)
I Find transfer function relation V (s) .

Figures 2.6 and 2.7.

c
°2006 R.J. Leduc 39
Operational Amplifiers
I So far we have only used passive elements for transfer
functions.

I we can use operational amplifiers (op amps) to construct

active circuits that can be used as transfer functions.

I Op amps have the following characteristics:

1. Output: vo (t) = A(v2 (t) − v1 (t))

2. High input impedance: Zi = ∞ (ideal)
3. Low output impedance: Zo = 0 (ideal)
4. High constant gain: A = ∞ (ideal)

Figure 2.10.
c
°2006 R.J. Leduc 40
Inverting Op Amps
I If we tie v2 (t) to ground, we get an inverting op amp, with
output vo (t) = −Av1 (t).
I Usually use with feedback in form of figure on the right.
I As input impedance very large, Ia (s) ≈ 0.
I Using nodal analysis at v1 , we find I1 (s) = −I2 (s).
I If gain A very large, feedback forces v1 (t) ≈ 0.

Figure2.10.

c
°2006 R.J. Leduc 41
Inverting Op Amps - II
Vi (s) Vo (s)
I With v1 (t) ≈ 0, we thus have I1 (s) = Z1 (s) and I2 (s) = Z2 (s) .

I Substituting into eqn I1 (s) = −I2 (s) gives:

Vi (s) Vo (s)
=−
Z1 (s) Z2 (s)
I This gives the transfer function:
Vo (s) Z2 (s)
=−
Vi (s) Z1 (s)

Figure2.10.

c
°2006 R.J. Leduc 42
Inverting Op Amp eg.
Vo (s)
I Find transfer function Vi (s) .

Figure 2.11.

Transfer function of a PID controller:

K1 K2
K3 (s2 + K3 s + K3 )
Gc (s) =
s
c
°2006 R.J. Leduc 43
Noninverting Op Amp
I We know that Vo (s) = A(Vi (s) − V1 (s))
I Using voltage division, we can derive:
Z1 (s)
V1 (s) = Vo (s)
Z1 (s) + Z2 (s)
I Substituting into top equation for V1 (s), gives:
Vo (s) A
= AZ1 (s)
Vi (s) 1+ Z1 (s)+Z2 (s)

I As A is large, we can disregard the 1 term

in denominator, which gives:

Vo (s) Z1 (s) + Z2 (s)

=
Vi (s) Z1 (s)
c
°2006 R.J. Leduc 44
Mechanical Systems

I We now examine how to relate the Laplace transform of the

input to that of the output using a transfer function.

I We will consider two types of mechanical systems:

1. translational mechanical systems.
2. rotational mechanical systems.

I As then end result is a transfer function that is mathematically

indistinguishable from that of an electrical network, we can
thus connect the two by cascading their transfer functions.

c
°2006 R.J. Leduc 45
Mass Component
I Newton’s Second Law of Motion: Σf = M a

I We are interested in the following properties of the object:

1. Σf = f (t): the summation of the forces acting on the object.
2. M : the object’s mass.
3. x(t): the position of the object.
4. v(t): the velocity of the object which is dx
dt .
2
5. a(t): the acceleration of the object which is ddtx2 .

F (s)
I We wish to find the transfer function Zm (s) = X(s) .
2
I We have: f (t) = M a(t) = M ddtx2 .
I Taking the Laplace transform of both
sides gives: F (s) = M s2 X(s).
F (s)
I Thus Zm (s) = X(s) = M s2 .
0 (s) = F (s)
I Similarly, Zm V (s) = Ms
Table 2.4
c
°2006 R.J. Leduc 46
Viscous Damper
I Middle part is typically moving through some sort of fluid.
I Frictional force generated is proportional to object’s velocity:
f (t) = fv v(t)
where fv is called coefficient of viscous friction.
I Substituting v(t) = dx
dt gives
dx
f (t) = fv
dt
I Taking the Laplace transform of both sides gives:
F (s) = fv sX(s).

F (s)
I Thus Zm (s) = X(s) = fv s.
0 (s) = F (s)
I Similarly, Zm V (s) = fv

c
°2006 R.J. Leduc Table 2.4. 47
Spring Component
I The spring constant is denoted K.

I Spring’s force is proportional to distance x which gives us:

f (t) = Kx(t)
I Taking the Laplace transform of both sides gives:
F (s) = KX(s)
F (s)
I We thus have: Zm (s) = X(s) = K.

I With zero
R t initial conditions, we have:
x(t) = 0 v(τ )dτ
K
I Thus F (s) = sV (s)
0 (s) = F (s) K
I This gives: Zm V (s) = s
Table 2.4.
c
°2006 R.J. Leduc 48
Summary of Translational Elements
I Table gives Force-velocity, force-displacement, and impedance
translational relationships for springs, viscous dampers, and
mass.

Table 2.4.
c
°2006 R.J. Leduc 49
Translational eg. - one equation of motion
X(s)
I Find transfer function F (s)

I f (t) is the force we are applying in the direction of x(t).

I To solve for transfer function, do the following:
1. Draw free body drawing showing all forces acting on object,
and their directions.
2. Replace mechanical components with mechanical impedances.
3. Newton’s law says that M a(t) equals the sum of the forces
acting on the object. Use this to create your force equation.

Figure 2.15.

c
°2006 R.J. Leduc 50
Translational eg. - one equation of motion - II
I We assume the mass is travelling to the right.
I This means only the applied force (f (t)) points to the right.
I The spring and viscous damping forces impede the motion,
thus act to oppose it.
2
I Newton’s second law: M ddtx2 = f (t) − fv dx
dt − Kx(t)
d2 x
I Rearranging this gives: f (t) = M dt2 + fv dx
dt + Kx(t)

I (a) Proper application

of Newton’s law.

I (b) Text book’s

“method.”
Free body diagram..
c
°2006 R.J. Leduc 51
Translational eg. - one equation of motion - III

I Can think of the M a(t) term as force beyond spring and

viscous damping forces, needed to attain acceleration a(t).

I Using textbook’s method, forces acting in assumed direction

of motion are positive, those acting in opposite are negative,
and they should sum to zero.

I (a) Free-body diagram

of mass, spring, and
damper system.

I (b) Transformed
free-body diagram.
Figure 2.16.

c
°2006 R.J. Leduc 52
Translational eg. - two degrees of freedom
I Number of equations of motion needed = number of linearly
independent motions.

I A linearly independence means that one object can still move

when all other objects are held still.

I We also refer to number of linearly independent motions as

the number of degrees of freedom of the system.

Figure 2.17 (a).

c
°2006 R.J. Leduc 53
Translational eg. - two degrees of freedom - II
X2 (s)
I Find transfer function F (s)

I To solve a problem like this, we will use the principal of

superposition. To find the forces acting on a given object (say
M1 ), we do the following:
1. Draw the free body diagram for object (ie. M1 ) by holding all
other objects (ie. M2 ) still, and finding the forces acting on
the object (ie. M1 ) due only to its motion.
2. We then hold the object (ie. M1 ) still, and activate each other
object (ie. M2 ) one at a time, and determine the forces acting
on original object (ie. M1 ) due to the other object’s motion.
3. The total force acting on the original object is the
superposition of the forces found in steps 1 and 2.

c
°2006 R.J. Leduc Figure 2.17 (a). 54
Translational eg. - two degrees of freedom - III

I Use Newton’s law to find equation of motion for M1 .

I (a) Forces on M1 due

to only motion of M1 .

I (b) Forces on M1 due

to only motion of M2 .

I (c) All forces on M1 .

Figure 2.18.

c
°2006 R.J. Leduc 55
Translational eg. - two degrees of freedom - IV

I Use Newton’s law to find equation of motion for M2 .

I (a) Forces on M2 due

to only motion of M2 .

I (b) Forces on M2 due

to only motion of M1 .

I (c) All forces on M2 .

Figure 2.19.

c
°2006 R.J. Leduc 56
Rotational Mechanical Systems

I With rotational mechanical systems, we are dealing with

objects that rotate about a fixed axis.

I The main differences from a translational system are as

follows:
I x(t) displacement becomes θ(t) angular displacement.
I v(t) = dx(t) dθ(t)
dt velocity becomes ω(t) = dt angular velocity.
I f (t) force becomes T (t) torque.
I M mass becomes J moment of inertia.
I fv coefficient of viscous damping becomes D coefficient of
viscous angular damping.
I K spring constant becomes K angular spring constant.

c
°2006 R.J. Leduc 57
Summary of Rotational Components
I Same symbols for components, but undergoing rotation.
I We are interested in the relation: T (s) = Zm (s)θ(s)

Table 2.5.

0 (s) T (s) K
Would have for Zm = ω(s) the values s, D, and Js.
c
°2006 R.J. Leduc 58
Degrees of Freedom

I Similar idea as for translation systems.

I The degrees of freedom is the number of objects (points of

motions) that can still rotate when all other objects are held
still.

I The degrees of freedom equals the number of equations of

motion we need to describe a system.

Figure 2.22 (b).

c
°2006 R.J. Leduc 59
Writing Equations of Motion - Rotational
I Create free body diagram, but show torques and angular
displacement instead.
I We will use superposition. To find the torques action on J1 :
1. Draw the free body diagram for J1 by holding J2 still, and find
the torques acting on J1 due only to its own motion.
2. We then hold J1 still, and rotate J2 , and determine the
torques acting on J1 due to the motion of J2 .
3. The total torque acting on J1 is the superposition of the
torques found in steps 1 and 2.
I Repeat above, but for J2 .
I The sum of torques acting in direction of assumed rotation
equals the sum of the torques opposing.

Figure 2.22 (b).

c
°2006 R.J. Leduc 60
Rotational eg.
θ2 (S)
I Find the transfer function T (S) .

I We have a flexible rod being supported at both ends by

bearings, and is undergoing torsion.

I We assume system can be approximated by a spring at one

point in the rod, with an inertia of J1 on the left, and J2 on
the right.

I (a) Physical
system.
I (b) Schematic.
Figure 2.22.

c
°2006 R.J. Leduc 61
Free body Diagrams for J1

I (a) Torques on J1
due only to motion
of J1 .

I (b) Torques on J1
due only to motion
of J2 .

I (c) All Torques on

Figure 2.23.
J1 .

c
°2006 R.J. Leduc 62
Free body Diagrams for J2

I (a) Torques on J2
due only to motion
of J2 .

I (b) Torques on J2
due only to motion
of J1 .

I (c) All torques on

Figure 2.24.
J2 .

c
°2006 R.J. Leduc 63
Rotational Mechanical System with Gears
I Usually when rotational systems are driven by motors, we find
associated gear trains driving the load.

I In diagram below, we see that if we apply torque T1 (t) to our

input gear which has N1 teeth and radius r1 , we will get a
corresponding torque T2 (t) out of gear 2, which has N2 teeth,
and radius r2 .
θ2 T2
I We want to determine the ratios θ1 and T1 .

Figure 2.27.

c
°2006 R.J. Leduc 64
Transfer Functions for Gears
I When gears turn, distance travelled around each gear’s
circumference is equal.
θ2 r1
I We thus have: r1 θ1 = r2 θ2 ⇒ θ1 = r2

r1 N1
I It can be shown that r2 = N2

θ2 N1
I We thus have: θ1 = N2

Figure 2.27 and Figure 2.28.

c
°2006 R.J. Leduc 65
Transfer Functions for Gears - lossless
I If we assume the gears have negligible inertia and damping,
then the energy put into gear 1 equals energy out of gear 2.
T2 θ1
I We thus have: T1 θ1 = T2 θ2 ⇒ T1 = θ2

θ2 N1
I Substituting in transfer function: θ1 = N2

T2 N2
I We get: T1 = N1

Figure 2.27 and Figure 2.28.

c
°2006 R.J. Leduc 66
Using Transfer Function for Lossless Gears
I Our first step is to reflect T1 to the other side of output gear
giving: T1 (s) N
N1 .
2

I This gives equation of motion:

N2
(Js2 + Ds + K)θ2 (s) = T1 (s)
N1
θ2 (s)
I This would allows us to find: T1 (s)

Figure 2.29.

c
°2006 R.J. Leduc 67
Lossless Gears: reflecting impedance
I Substituting in for θ2 (s) gives
N1 N2
(Js2 + Ds + K) θ1 (s) = T1 (s)
N2 N1
I Simplifying: [J( N1 2 2 N1 2 N1 2
N2 ) s + D( N2 ) s + K( N2 ) ]θ1 (s) = T1 (s)

Figure 2.29.
c
°2006 R.J. Leduc 68
Generalizing Reflecting Impedance

I Rule: “Rotational mechanical impedances can be reflected

through gear trains by multiplying the mechanical impedance
by the ratio:“
µ ¶2
Number of teeth of gear on destination shaft
Number of teeth of gear on source shaft
I The impedance to be reflected is attached to the source shaft
and is being reflected to the destination shaft.

I In previous example gear 2 was the source shaft, and gear 1

was the destination shaft.

c
°2006 R.J. Leduc 69
Lossless Gears eg.

θ2 (s)
I Find transfer function: T1 (s)

Figure 2.30.

c
°2006 R.J. Leduc 70
Gear Trains
100
I What if you wanted a high ratio, say 1 ?

I To avoid a gear with 100× as big radius, a gear train is used.

I The total equivalent gear ratio for the gear train, is the
product of the gear ratios for each step.

Figure 2.31.

T4 N2 N4 N6
For torques, we would have: T1 = N1 N3 N5
c
°2006 R.J. Leduc 71
Gears With Loss eg.
I Find transfer function Tθ11(s)
(s) .
I Each gear below has its own inertia, and some shafts have
nonnegligible damping.
I Need to reflect all moments of inertial and viscous dampers to
the left of gear 1.

Figure 2.32.

c
°2006 R.J. Leduc 72
Gears With Loss eg. - II
I Equation of motion:
(Je s2 + De s)θ1 (s) = T1 (s)
I The transfer function is thus:
θ1 (s) 1
G(s) = =
T1 (s) Je s2 + De s

Figure 2.32.
c
°2006 R.J. Leduc 73
Electromechanical System Transfer System

I Electromechanical systems are systems with a mixture of

electrical and mechanical variables.

Figure 2.34: NASA flight simulator robot arm with

electromechanical control system components.

c
°2006 R.J. Leduc 74
DC Motors

I A motor takes a voltage as input and produces a physical

displacement as output.

I We will derive transfer function for the armature-controlled dc

servomotor.

Figure 2.35.

c
°2006 R.J. Leduc 75
DC Motor Principles

From: http://hyperphysics.phy-astr.gsu.edu/hbase/magnetic/motdc.html.

c
°2006 R.J. Leduc 76
DC Motors: basics
I DC motor contains stationary magnetic field provided by
permanent or electromagnet with field strength B.
I Motor contains a rotating circuit called the armature (the
”coils“) through which a current ia (t) flows.
I Current flows perpendicular through the magnetic field and
feels a force F = Blia (t) acting on it.
I Parameter l is the length of the conductor the current is
flowing through within the field.

I Associated with the armature is a

resistance Ra , and an inductance La .

Figure 2.35.
c
°2006 R.J. Leduc 77
Back EMF
I When a conductor moves at right angles to a magnetic field,
it generates a voltage at terminals of the conductor (think
generator).
I Since armature is rotating in a magnetic field, it produces a
voltage proportional to its velocity.
dθm (t)
vb (t) = Kb
dt
I We refer to vb (t) as the back electromotive force (back emf),
and Kb as the back emf constant.

I Taking Laplace transform we get:

Vb (s) = Kb sθm (s)

c
°2006 R.J. Leduc 78
Figure 2.35.
Analyzing Armature Circuit
θm (s)
I We wish to find the transfer function: Ea (s)

I Applying Kirchoff’s voltage law to armature circuit gives:

Ra Ia (s) + La sIa (s) + Vb (s) = Ea (s)
I The torque developed by motor is:
1
Tm (s) = Kt Ia (s) ⇒ Ia (s) = Tm (s)
Kt
I Substituting in for Ia (s) and Vb (s) gives:

(Ra + La s)Tm (s)

+ Kb sθm (s) = Ea (s)
Kt

c
°2006 R.J. Leduc Figure 2.35. 79
Analyzing Armature Circuit - II
I Figure shows a typical mechanical loading of a motor.

I Variable Jm is the moment of inertia of armature plus that of

the load reflected to the armature.

I Variable Dm is the viscous damping of the armature plus that

of the load reflected to the armature.

I This gives the equation of motion:

Tm (s) = (Jm s2 + Dm s)θm (s)

Figure 2.36.

c
°2006 R.J. Leduc 80
DC Motor Transfer function

I We can now substitute into our previously derived equation

for Ea (s), and noting that typically (Ra + La s) ≈ Ra gives:

Ra
[ (Jm s + Dm ) + Kb ]sθm (s) = Ea (s)
Kt
I This gives the transfer function:
Kt
θm (s) K
= h ³Ra Jm ´i =
Ea (s) s s+ 1
Dm + Kt Kb s(s + α)
Jm Ra

c
°2006 R.J. Leduc 81
Deriving Jm and Dm
I Figure shows typical usage of motor.
I Variable Ja is the moment of inertia of armature, and Da is
the viscous damping of the armature.
I Variable JL is the load moment of inertia, and DL is the
viscous damping on the load.
I Reflecting onto the armature gives:
N1 N1
Jm = Ja + ( )2 JL and Dm = Da + ( )2 DL
N2 N2

Figure 2.37.

c
°2006 R.J. Leduc 82
Deriving Electrical Constants
I A dynamometer measues the torque and speed of a motor
under a constant applied voltage.

I Previously, we derived:
Ra
Tm (s) + Kb sθm (s) = Ea (s)
Kt
I Inverse Laplace transform gives:
Ra
Tm (t) + Kb ωm (t) = Ea (t)
Kt
I As voltage is constant, so is torque and velocity. Solving for
Tm gives
Kb Kt Kt
Tm = − ωm + ea
Ra Ra
c
°2006 R.J. Leduc 83
Deriving Electrical Constants - II
Kb Kt Kt
Tm = − ωm + ea
Ra Ra
I When ωm = 0 we get the stall torque given by:
Kt
Tstall = ea
Ra
I When Tm = 0 we get the angular velocity called no-load
ea
speed given by: ωno−load = K b

I We thus have:
Kt Tstall ea
= and Kb =
Ra ea ωno−load

c
°2006 R.J. Leduc Figure 2.38. 84
Linear Systems
I So far, we have developed models for systems that can be
approximately represented by linear, time-invariant differential
equations.

I In developing these models, we assumed they were linear.

I A linear system must have two properties:

Superposition: means that the output response to a sum of
inputs is equal to the sum of the output
response of each individual input that makes up
the ”sum of inputs.“

ie. if input r1 (t) generates output response c1 (t), and input

r2 (t) generates output response c2 (t), the the output response
to input r1 (t) + r2 (t) will be c1 (t) + c2 (t).
c
°2006 R.J. Leduc 85
Linear Systems -II
Homogeneity: means that when the input is multipled by a scalar,
the result is a response multiplied by same scalar

ie. if input r1 (t) generates output response c1 (t), then input

Ar1 (t) will result in output Ac1 (t).

Figure 2.45.

c
°2006 R.J. Leduc 86
Nonlinear Systems
I Nonlinear systems are systems that are not linear.
I Some examples are:
I Op amps are linear over a given range, but exhibits saturation
at high input voltages.
I A motor exhibits a deadzone where it does not respond to low
input voltages because of friction.
I Backlash occurs when a pair of gears do not fit tightly. This is
when the input gear moves through a small angle before the
output gear starts to move.

c
°2006 R.J. Leduc Figure 2.46. 87
Nonlinear Systems - II

I Sometimes we can make a linear approximation to a nonlinear

system.

I For example, if system is linear for a portion, the range of

input values about the point is small, and we can shift this
portion to the origin, we can establish a linear relationship.

I Op amps are an example of this. We make sure our input

values stay within the linear range.

c
°2006 R.J. Leduc 88
Linearization
I If a system contains any nonlinear components, we must
linearize the system before we can apply our modelling
methods.

I We need to find linear approximations in order to be able to

find transfer functions.
I To linearize a system, we do the following:
1. Identify the nonlinear components and write nonlinear
differential equations for them.
2. We then choose a small range of input values over which the
system behaves approximately linear. Refer to this range as a
small-signal input.

This range is centered around a steady state solution, called

equilibrium, where the small-signal input is equal to zero.
3. We then write a linear differential expression for this range,
and then apply Laplace transforms and form our transfer
function as normal.
c
°2006 R.J. Leduc 89
Linearizing a Function
I Assume we have a nonlinear system that is operating about
point A: [xo , f (xo )].
I For small changes around A, we can calculate the approximate
value of f (x) using the slope of the curve at xo .
I If slope at point A is ma we can determine the change in
f (x), called δf (x), for very small changes in x, called δx.
δf (x) = [f (x) − f (xo )] ≈ ma (x − xo ) = ma δx

I This gives us:

f (x) ≈ f (xo ) + ma (x − xo )
≈ f (xo ) + ma δx

Figure 2.47.

c
°2006 R.J. Leduc 90
Linearizing a Function - eg.
I Linearize f (x) = 5 cos x about π/2.

Figure 2.48.
c
°2006 R.J. Leduc 91
Formalizing the Linearization Method.

I Using the Taylor series expansion, we can express the value of

a function about point xo using the equation:
¯ ¯
df ¯¯ (x − xo ) d2 f ¯¯ (x − xo )2
f (x) = f (xo ) + + +...
dx ¯x=xo 1! dx2 ¯x=xo 2!

I For small range around xo , we can neglect higher order terms

and we get:
¯
df ¯¯
f (x) − f (xo ) ≈ (x − xo ) = m|x=xo (x − xo )
dx ¯x=xo

c
°2006 R.J. Leduc 92
Linearizing a Differential Eqn eg.

I Linearize the differential equation below about x = π/4:

d2 x dx
+2 + cosx = 0
dt2 dt

c
°2006 R.J. Leduc 93
Transfer Function - Nonlinear Electrical System eg.

I Find transfer function VVL(s)

(s)
, where voltage current
relationship for resistor is ir = 2e0.1vr , and v(t) is a
small-signal voltage source.

Figure 2.49.

c
°2006 R.J. Leduc 94