Chapter 4 Multiple Degree of

Freedom Systems
The Millennium bridge required
many degrees of freedom to model
and design with.

Extending the first 3 chapters to

more then one degree of freedom
D.J. Inman

1/58

Mechanical Engineering at Virginia Tech

The first step in analyzing multiple

degrees of freedom (DOF) is to look at 2
DOF
DOF: Minimum number of coordinates to specify the
position of a system
Many systems have more than 1 DOF
Examples of 2 DOF systems
car with sprung and unsprung mass (both heave)
elastic pendulum (radial and angular)
motions of a ship (roll and pitch)

D.J. Inman

2/58

Fig 4.1

Mechanical Engineering at Virginia Tech

4.1 Two-Degree-of-Freedom
Model (Undamped)

A 2 degree of freedom system used to base

much of the analysis and conceptual
development of MDOF systems on.
D.J. Inman

3/58

Mechanical Engineering at Virginia Tech

Free-Body Diagram of each mass

Figure 4.2
k2(x2 -x1)
k1 x1

m2

m1
x1

D.J. Inman

4/58

Mechanical Engineering at Virginia Tech

x2

Summing forces yields the

equations of motion:
m1 !!
x1 (t) = k1 x1 (t) + k2 ( x2 (t) x1 (t))
m2 !!
x2 (t) = k2 ( x2 (t) x1 (t))
Rearranging terms:
m1 !!
x1 (t) + (k1 + k2 )x1 (t) k2 x2 (t) = 0
m2 !!
x2 (t) k2 x1 (t) + k2 x2 (t) = 0
D.J. Inman

5/58

Mechanical Engineering at Virginia Tech

(4.1)

(4.2)

Note that it is always the case

that
A 2 Degree-of-Freedom system has
Two equations of motion!
Two natural frequencies (as we shall see)!

D.J. Inman

6/58

Mechanical Engineering at Virginia Tech

The dynamics of a 2 DOF system consists

of 2 homogeneous and coupled equations
Free vibrations, so homogeneous eqs.
Equations are coupled:
Both have x1 and x2.
If only one mass moves, the other follows
Example: pitch and heave of a car model

In this case the coupling is due to k2.

Mathematically and Physically
If k2 = 0, no coupling occurs and can be solved
as two independent SDOF systems
D.J. Inman

7/58

Mechanical Engineering at Virginia Tech

Initial Conditions
Two coupled, second -order, ordinary
differential equations with constant
coefficients
Needs 4 constants of integration to solve
Thus 4 initial conditions on positions and
velocities

x1 (0) = x10 , x!1 (0) = x!10 , x2 (0) = x20 , x!2 (0) = x!20
D.J. Inman

8/58

Mechanical Engineering at Virginia Tech

Solution by Matrix Methods

The two equations can be written in the form of a
single matrix equation (see pages 272-275 if matrices and
vectors are a struggle for you) :
x1 (t)
x1 (t)
x!1 (t)
!!
x(t) =
, x! (t) =
, !!
x(t) =

!
!!
x
(t)
x
(t)
x
(t)
2
2
2
m1
M =
0

0
,

m2

M!!
x + Kx = 0
D.J. Inman

9/58

k1 + k2
K=
k2

(4.4), (4.5)

k2
k2

(4.6), (4.9)

m1 x1 (t) + (k1 + k 2 )x1 (t) k2 x 2 (t) = 0

m2 x2 (t) k 2 x1 (t) + k 2 x 2 (t) = 0
Mechanical Engineering at Virginia Tech

Initial Conditions
ICs can also be written in vector form

x10
x!10
x(0) = , and x! (0) =
x20
x!20
D.J. Inman

10/58

Mechanical Engineering at Virginia Tech

The approach to a Solution:

For 1DOF we assumed the scalar solution aet
Similarly, now we assume the vector form:

Let x(t) = ue

j t

(4.15)

j = 1, u =/ 0, , u unknown

- M + K ue
2

j t

=0

- M + K u = 0
D.J. Inman

11/58

Mechanical Engineering at Virginia Tech

(4.16)
(4.17)

This changes the differential

equation of motion into algebraic
vector equation:

- 2 M + K u = 0

(4.17)

This is two algebraic equation in 3 uknowns

( 1 vector of two elements and 1 scalar):
u1
u= , and
u2
D.J. Inman

12/58

Mechanical Engineering at Virginia Tech

The condition for solution of this matrix

equation requires that the the matrix
inverse does not exist:
If the inv ( - M + K ) exists u = 0: which is the
2

static equilibrium position. For motion to occur

u =/ 0 - M + K

or det - 2 M + K = 0

does not exist

(4.19)

The determinant results in 1 equation

in one unknown (called the characteristic equation)

D.J. Inman

13/58

Mechanical Engineering at Virginia Tech

Back to our specific system: the

characteristic equation is defined as
det(- M + K ) = 0
2

2 m1 + k1 + k2
det
k2

(4.20)

k2
=0
2
m2 + k2

m1m2 4 (m1k2 + m2 k1 + m2 k2 ) 2 + k1 k2 = 0
Eq. (4.21) is quadratic in 2 so four solutions result:

D.J. Inman

14/58

2
1

and 1 and 2
2
2

Mechanical Engineering at Virginia Tech

(4.21)

Once is known, use equation (4.17) again to

calculate the corresponding vectors u1 and u2
This yields vector equation for each squared frequency:"
( 12 M + K )u1 = 0
(4.22)
and
( 22 M + K )u 2 = 0

(4.23)

Each of these matrix equations represents 2

equations in the 2 unknowns components of the
vector, but the coefficient matrix is singular so each
matrix equation results in only 1 independent
equation. The following examples clarify this."
D.J. Inman

15/58

Mechanical Engineering at Virginia Tech

Examples 4.1.5 & 4.1.6:calculating u

and
m1=9 kg,m2=1kg, k1=24 N/m and k2=3 N/m
The characteristic equation becomes
4-62+8=(2-2)(2-4)=0
2 = 2 and 2 =4 or

1,3 = 2 rad/s, 2,4 = 2 rad/s

Each value of 2 yields an expression for u:"
D.J. Inman

16/58

Mechanical Engineering at Virginia Tech

Computing the vectors u

u11
For =2, denote u1 = then we have
u12
(- 12 M + K )u1 = 0
2
1

3 u11 0
27 9(2)
=

3 (2) u12 0

9u11 3u12 = 0 and 3u11 + u12 = 0

2 equations, 2 unknowns but DEPENDENT!"

(the 2nd equation is -3 times the first) "
D.J. Inman

17/58

Mechanical Engineering at Virginia Tech

Only the direction of vectors u can be

determined, not the magnitude as it
remains arbitrary
u11 1
1
= u11 = u12 results from both equations:
u12 3
3
only the direction, not the magnitude can be determined!
This is because: det( 12 M + K ) = 0.
The magnitude of the vector is arbitrary. To see this suppose
that u1 satisfies
( 12 M + K )u1 = 0, so does au1 , a arbitrary. So
( 12 M + K )au1 = 0 ( 12 M + K )u1 = 0
D.J. Inman

18/58

Mechanical Engineering at Virginia Tech

Likewise for the second value of 2

u21
For = 4, let u 2 = then we have
u22
(- 12 M + K )u = 0
2
2

3 u21 0
27 9(4)
=

3 (4) u22 0

1
9u21 3u22 = 0 or u21 = u22
3
D.J. Inman

19/58

Note that the other equation is the same"

Mechanical Engineering at Virginia Tech

What to do about the magnitude!

Several possibilities, here we just fix one element:
Choose:"

Choose:"

3
u12 = 1 u1 =
1
3
u22 = 1 u 2 =
1

D.J. Inman

20/58

Mechanical Engineering at Virginia Tech

Thus the solution to the algebraic

matrix equation is:
1, 3

13
= 2, has mode shape u1 =
1

1 3
2, 4 = 2, has mode shape u 2 =
1
Here we have introduce the name
mode shape to describe the vectors
u1 and u2. The origin of this name comes later
D.J. Inman

21/58

Mechanical Engineering at Virginia Tech

Return now to the time response:

We have computed four solutions:"

x(t) = u1e j1t , u1e j1t , u 2 e j 2 t , u 2 e j 2 t

(4.24)

Since linear, we can combine as:"

x(t) = au1e j1t + bu1e j1t + cu 2 e j 2 t + du 2 e j 2 t

x(t) = ae j1t + be j1t u1 + ce j 2 t + de j 2 t u 2

= A1 sin( 1t + 1 )u1 + A2 sin( 2t + 2 )u 2

(4.26)

where A1 , A2 , 1 , and 2 are constants of integration

determined by initial conditions. "
D.J. Inman

22/58

Note that to go from the exponential

form to to sine requires Eulers formula
for trig functions and uses up the
+/- sign on omega

Mechanical Engineering at Virginia Tech

Physical interpretation of all that

math!
Each of the TWO masses is oscillating at TWO
natural frequencies 1and 2
The relative magnitude of each sine term, and
hence of the magnitude of oscillation of m1 and m2
is determined by the value of A1u1 and A2u2
The vectors u1 and u2 are called mode shapes
because the describe the relative magnitude of
oscillation between the two masses
D.J. Inman

23/58

Mechanical Engineering at Virginia Tech

What is a mode shape?

First note that A1, A2, 1 and 2 are determined by
the initial conditions
Choose them so that A2 = 1 = 2 =0
Then:
x (t)
u
1
11
x(t) =
= A1 sin 1t = A1u1 sin 1t

x2 (t)
u12

Thus each mass oscillates at (one) frequency w1

with magnitudes proportional to u1 the 1st mode
shape
D.J. Inman

24/58

Mechanical Engineering at Virginia Tech

A graphic look at mode shapes:

If ICs correspond to mode 1 or 2, then the response is purely
in mode 1 or mode 2.
x1
k1

m1

Mode 1:

k2

Mode 2:

x1

m1
x1=-A/3

D.J. Inman

25/58

m2

13
u1 =
1

m2

1 3
u2 =
1

x2=A

x1=A/3
k1

x2

k2

x2

x2=A

Mechanical Engineering at Virginia Tech

Example 4.1.7 given the initial

conditions compute the time response
1
0
consider x(0) = mm, x! (0) =
0
0
A1
A2

x
(
t
)
1
sin ( 2t + 1 ) sin (2t + 2 )

3
x (t ) = 3
2 A sin ( 2t + )+ A sin (2t + )
1
1
2
2
A
A

1
2
x!1 (t )
2 cos( 2t + 1 ) 2 cos(2t + 2 )

3
x! (t ) = 3
2 A 2 cos( 2t + )+ A 2 cos(2t + )
1
1
2
2
D.J. Inman

26/58

Mechanical Engineering at Virginia Tech

At t = 0 we have"

A
A
1
2
1 mm
sin(1 ) sin( 2 )

= 3
0 A sin( ) + A sin( )
1
1
2
2
A1

A
2
0
2 cos(1 ) 2 cos( 2 )
3
= 3

0
A1 2 cos(1 ) + 2A2 cos( 2 )
D.J. Inman

27/58

Mechanical Engineering at Virginia Tech

4 equations in 4 unknowns: "

3 = A1 sin ( 1 ) A2 sin (2 )
0 = A1 sin (1 ) + A2 sin ( 2 )
0 = A1 2 cos( 1 ) A2 2 cos( 2 )
0 = A1 2 cos( 1 ) + A2 2 cos(2 )
Yields:"

A1 = 1.5 mm, A2 = 1.5 mm, 1 = 2 = rad

2
D.J. Inman

28/58

Mechanical Engineering at Virginia Tech

The final solution is:

x1 (t) = 0.5 cos 2t + 0.5 cos 2t
x2 (t) = 1.5 cos 2t 1.5 cos 2t

(4.34)

These initial conditions gives a response that is a combination

of modes. Both harmonic, but their summation is not.

Figure 4.3
D.J. Inman

29/58

Mechanical Engineering at Virginia Tech

Solution as a sum of modes

x(t) = a1u1 cos 1t + a2u2 cos 2 t

Determines how the first "

frequency contributes to the"
response"

D.J. Inman

30/58

Determines how the second "

frequency contributes to the"
response"

Mechanical Engineering at Virginia Tech

Things to note
Two degrees of freedom implies two natural
frequencies
Each mass oscillates at with these two frequencies
present in the response and beats could result
Frequencies are not those of two component
systems
1 =

k1
= 1.63, 2 = 2
m1

k2
= 1.732
m2

The above is not the most efficient way to

calculate frequencies as the following describes
D.J. Inman

31/58

Mechanical Engineering at Virginia Tech

Some matrix and vector reminders

a b
1 d b
1
A=
A =

ad cb c a
c c
T
2
2
x x = x1 + x2
m1 0
T
2
2
M =
x Mx = m1 x1 + m2 x2

0 m2
T
M > 0 x Mx > 0 for every value of x except 0

Then M is said to be positive definite

D.J. Inman

32/58

Mechanical Engineering at Virginia Tech

4.2 Eigenvalues and Natural

Frequencies
Can connect the vibration problem with the
algebraic eigenvalue problem developed in
math
This will give us some powerful
computational skills
And some powerful theory
All the codes have eigen-solvers so these
painful calculations can be automated
D.J. Inman

33/58

Mechanical Engineering at Virginia Tech

Some matrix results to help us use

available computational tools:
A matrix M is defined to be symmetric if

M=M

A symmetric matrix M is positive definite if

x Mx > 0 for all nonzero vectors x

T

A symmetric positive definite matrix M can

T
be factored
M = LL
Here L is upper triangular, called a Cholesky matrix
D.J. Inman

34/58

Mechanical Engineering at Virginia Tech

If the matrix L is diagonal, it defines

the matrix square root
The matrix square root is the matrix M 1/2 such that
M 1/2 M 1/2 = M
If M is diagonal, then the matrix square root is just the root
of the diagonal elements:
L = M 1/2

D.J. Inman

35/58

m1
=
0

m2

Mechanical Engineering at Virginia Tech

(4.35)

A change of coordinates is introduced to

capitalize on existing mathematics
For a diagonal, positive definite matrix M:"
1

0
m1
1/2
,
M
=

1
m2
0
Let x(t) = M 1/2 q(t) and multiply by M 1/2 :

m1
M =
0

0
1 m1
1
, M =

m2
0

1
m2

1/2
1/2
1/2
1/2
%%
M
MM
q(t)
+
M
KM
!#
#"##
$
!#
#"##
$ q(t) = 0
I identity

or

(4.38)

K& symmetric

&
%% + Kq(t)
q(t)
= 0 where K& = M 1/2 KM 1/2

k
&
K is called the mass normalized stiffness and is similar to the scalar
m
used extensively in single degree of freedom analysis. The key here is that
K& is a SYMMETRIC matrix allowing the use of many nice properties and
computational tools
D.J. Inman

36/58

Mechanical Engineering at Virginia Tech

How the vibration problem relates to the

real symmetric eigenvalue problem
j t
"
!! + Kq(t)
Assume q(t) = ve in q(t)
=0
2
j t
j t
"
ve + Kve = 0,
v 0 or
2
"
Kv
= %
#%$
&v

vibration problem
(4.40)

" = v
Kv
#
%$%
&

v0

real symmetric
eigenvalue problem
(4.41)

Note that the martrix K! contains the same type of information

as does n2 in the single degree of freedom case.
D.J. Inman

37/58

Mechanical Engineering at Virginia Tech

Important Properties of the n x n Real

Symmetric Eigenvalue Problem
There are n eigenvalues and they are all real
valued
There are n eigenvectors and they are all real
valued
The set of eigenvectors are orthogonal
The set of eigenvectors are linearly
independent
The matrix is similar to a diagonal matrix
Window 4.1 page 285
D.J. Inman

38/58

Mechanical Engineering at Virginia Tech

Square Matrix Review

Let aik be the ikth element of A then A is symmetric
if aik = aki denoted AT=A
A is positive definite if xTAx > 0 for all nonzero x
(also implies each i > 0)
The stiffness matrix is usually symmetric and
positive semi definite (could have a zero
eigenvalue)
The mass matrix is positive definite and
symmetric (and so far, its diagonal)

D.J. Inman

39/58

Mechanical Engineering at Virginia Tech

Normal and orthogonal vectors

x1
y1
n

T
x = ! , y = ! , inner product is x y = xi yi
i =1
xn
yn
x orthogonal to y if xT y = 0

T
x is normal if x x = 1

if a the set of vectores is is both orthogonal and normal it

is called an orthonormal set
The norm of x is x = xT x
D.J. Inman

40/58

(4.43)

Mechanical Engineering at Virginia Tech

Normalizing any vector can be

done by dividing it by its norm:
x

has norm of 1

x x

To see this compute

x
xT x

D.J. Inman

41/58

xT

xT x

xT x

xT x
=1
T
x x

Mechanical Engineering at Virginia Tech

(4.44)

Examples 4.2.2 through 4.2.4

1
1
0
27
3
0

3
3
1/2
1/2
K! = M KM
=
3 3 0 1
0
1

3 1

so K! =
which is symmetric.

1 3
3- -1

2
det(K! I ) = det
=

6 + 8 = 0

-1 3-

which has roots: 1 = 2 = and 2 = 4 =

2
1

D.J. Inman

42/58

Mechanical Engineering at Virginia Tech

2
2

( K! 1 I )v1 = 0
3 2 1 v11 0
1 3 2 v = 0

12
1
v11 v12 = 0 v1 =
1
v1 = (1 + 1) = 1 =
2

1
v1 =
2
D.J. Inman

43/58

1
1

The first normalized eigenvector"

Mechanical Engineering at Virginia Tech

Likewise the second normalized eigenvector is

computed and shown to be orthogonal to the first,
so that the set is orthonormal

1 1
1
T
v2 =
1 , v1 v 2 = 2 (1 1) = 0
2
1
T
v1 v1 = (1 + 1) = 1
2
1
T
v 2 v 2 = (1 + (1)(1)) = 1
2
v i are orthonormal

D.J. Inman

44/58

Mechanical Engineering at Virginia Tech

Modes u and Eigenvectors v are

different but related:
u1 v1 and u 2 v 2
x=M
Note

1/2

qu= M

1/2

(4.37)

1 1

3
0

1/2
3 =
M u1 =
= v1

0 1 1 1
D.J. Inman

45/58

Mechanical Engineering at Virginia Tech

This orthonormal set of vectors is

used to form an Orthogonal Matrix
P = [ v1 v 2 ]

called a matrix of eigenvectors (normalized)"

T
T

v
v
v
1 0
1 1
1 v2
T
P P= T
= 0 1 = I
T

v 2 v1 v 2 v 2
P is called an orthogonal matrix"
! = P T Kv
! 1 Kv
! 2 = P T [ 1v1 2 v 2 ]
P T KP

1v1T v1 2 v1T v 2 1 0
2
2
= T
=
=
diag(

0
1
2 ) = (4.47)
T
2
1v 2 v1 2 v 2 v 2

P is also called a modal matrix"

D.J. Inman

46/58

Mechanical Engineering at Virginia Tech

Example 4.2.3 compute P and show

that it is an orthogonal matrix
From the previous example:
P = [ v1

1 1
P P=
2 2
T

1 1 + 1
=
2 1 1
D.J. Inman

47/58

1 1
1 1

1 1 1 1
1 1 1 1

1 1 1 2 0
=
=I

1 + 1 2 0 2

1
v1 ] =
2

Mechanical Engineering at Virginia Tech

Example 4.2.4 Compute the square of

the frequencies by matrix manipulation
1 1 1 3 1 1 1 1
!
P KP =
1 1 1 3
1 1
2

1 1 1 2 4
=
2 1 1 2 4
T

12 0
1 4 0 2 0
=
=
==

2
2 0 8 0 4
0

1 = 2 rad/s and 2 = 2 rad/s

In general:
D.J. Inman

48/58

2
!
= P KP = diag ( i ) = diag( i )
T

Mechanical Engineering at Virginia Tech

(4.48)

Example 4.2.5
The equations of motion:

Figure 4.4

m1 !!
x1 + (k1 + k2 )x1 k2 x2 = 0
m2 !!
x2 k2 x1 + (k2 + k3 )x2 = 0

(4.49)

In matrix form these become:

m1
0

D.J. Inman

49/58

0
k1 + k2
!!
x+

m2
k2

k2
x=0

k2 + k 3

Mechanical Engineering at Virginia Tech

(4.50)

Next substitute numerical values

and compute P and
m1 = 1 kg, m2 = 4 kg, k1 = k3 = 10 N/m and k2 =2 N/m
1 0
12
M =
, K=

0 4
2
12
1/2
1/2
!
K = M KM
=
1

2
12
1
12

1
12
2
!
det K I = det
=

15 + 35 = 0

12
1
1 = 2.8902 and 2 = 12.1098
1 = 1.7 rad/s and 2 = 12.1098 rad/s

D.J. Inman

50/58

Mechanical Engineering at Virginia Tech

Next compute the eigenvectors

For 1 equation (4.41 ) becomes:
1 v11
12 - 2.8902
=0

1
3 - 2.8902 v21

9.1089v11 = v21
Normalizing v1 yields
2
1 = v1 = v112 + v21
= v112 + (9.1089)2 v112

v11 = 0.1091, and v21 = 0.9940

0.1091
v1 =
,

0.9940
D.J. Inman

51/58

0.9940
likewise v 2 =

0.1091

Mechanical Engineering at Virginia Tech

Next check the value of P to see

if it behaves as its suppose to:
P = [ v1

0.1091 0.9940
v2 ] =

0.9940 0.1091

0.1091 0.9940 12 1 0.1091 0.9940 2.8402

0
! =
P T KP
=
0.9940 0.1091 1 3 0.9940 0.1091 0
12.1098

0.1091 0.9940 0.1091 0.9940 1 0
PT P =
=

0.9940 0.1091 0.9940 0.1091 0 1

Yes!
D.J. Inman

52/58

Mechanical Engineering at Virginia Tech

A note on eigenvectors
In the previous section, we could have chosed v 2 to be
0.9940
-0.9940
v2 =
instead of v 2 =

0.1091
0.1091

because one can always multiple an eigenvector by a constant

and if the constant is -1 the result is still a normalized vector.

Does this make any difference?

No! Try it in the previous example
D.J. Inman

53/58

Mechanical Engineering at Virginia Tech

All of the previous examples can and should

be solved by hand to learn the methods
However, they can also be solved on
calculators with matrix functions and
with the codes listed in the last section
In fact, for more then two DOF one must
use a code to solve for the natural
frequencies and mode shapes.
Next we examine 3 other formulations for
solving for modal data
D.J. Inman

54/58

Mechanical Engineering at Virginia Tech

Matlab commands
To compute the inverse of the square matrix
A: inv(A) or use A\eye(n) where n is
the size of the matrix
[P,D]=eig(A) computes the eigenvalues
and normalized eigenvectors (watch the
order). Stores them in the eigenvector
matrix P and the diagonal matrix D (D=L)

D.J. Inman

55/58

Mechanical Engineering at Virginia Tech

More commands
To compute the matrix square root use
sqrtm(A)
To compute the Cholesky factor: L= chol(M)!
To compute the norm: norm(x)!
To compute the determinant det(A)!
To enter a matrix:
K=[27 -3;-3 3]; M=[9 0;0 1];!
To multiply: K*inv(chol(M))!

D.J. Inman

56/58

Mechanical Engineering at Virginia Tech

An alternate approach to normalizing

mode shapes
From equation (4.17) ( M 2 + K ) u = 0,

u0

Now scale the mode shapes by computing such that

1
T
( i ui ) M ( i ui ) = 1 i = T
ui ui

w i = i ui is called mass normalized and it satisfies:

i2 Mw i + Kw i = 0 i2 = w Ti Kw i , i = 1, 2
(4.53)
D.J. Inman

57/58

Mechanical Engineering at Virginia Tech

There are 3 approaches to computing

mode shapes and frequencies
(i) Mu = Ku (ii) u = M Ku
2

(iii) v = M KM 2 v
2

(i) Is the Generalized Symmetric Eigenvalue Problem

easy for hand computations, inefficient for computers
(ii) Is the Asymmetric Eigenvalue Problem
very expensive computationally
(iii) Is the Symmetric Eigenvalue Problem
the cheapest computationally
D.J. Inman

58/58

Mechanical Engineering at Virginia Tech