LECTURE 3

VIBRATION MITIGATION TECHNIQUES:

THEORY AND APPLICATION

1
TOPICS TO BE COVERED……

MDoF system, eigen values, eigen vectors, modal analysis, orthogonal properties.

MDoF Model of Three Storey Building

2
MDoF Model of a Heavy Machine Tool

3
A Simplified Analysis of the Vibration of an Airplane
4
Multi‐degree of freedom systems
• Modeling of continuous systems as multi degree of freedom systems
• Eigen value problem
• As stated before, most engineering systems are continuous and have an
infinite number of degrees of freedom. The vibration analysis of continuous
systems requires the solution of partial differential equations, which is quite
difficult.
• In fact, analytical solutions do not exist for many partial differential
equations. The analysis of a multi degree of freedom system on the other
hand, requires the solution of a set of ordinary differential equations, which
is relatively simple. Hence, for simplicity of analysis, continuous systems
are often approximated as multi degree of freedom systems.
• For a system having n degrees of freedom, there are n associated natural
frequencies, each associated with its own mode shape.
• Different methods can be used to approximate a continuous system as a
multi degree of freedom system. A simple method involves replacing the
distributed mass or inertia of the system by a finite number of lumped
masses or rigid bodies.
• The lumped masses are assumed to be connected by mass-less elastic and
damping members.
5
• Linear coordinates are used to describe the motion of the lumped masses. Such
models are called lumped parameter of lumped mass or discrete mass systems.
• The minimum number of coordinates necessary to describe the motion of the
lumped masses and rigid bodies defines the number of degrees of freedom of
the system. Naturally, the larger the number of lumped masses used in the
model, the higher the accuracy of the resulting analysis.
• Some problems automatically indicate the type of lumped parameter model to
be used.
• For example, the three storey building shown in the figure automatically
suggests using a three lumped mass model as indicated in the figure.
• In this model, the inertia of the system is assumed to be concentrated as three
point masses located at the floor levels, and the elasticity's of the columns are
replaced by the springs.
• Another popular method of approximating a continuous system as a multi
degree of freedom system involves replacing the geometry of the system by a
large number of small elements.
• By assuming a simple solution within each element, the principles of
compatibility and equilibrium are used to find an approximate solution to the
original system. This method is known as the finite element method.
6
Using Newton’s second law to derive equations of motion

The following procedure can be adopted to derive the equations of motion of

a multi-degree of freedom system using Newton’s second law of motion.

• Set up suitable coordinates to describe the positions of the various point

masses and rigid bodies in the system. Assume suitable positive
directions for the displacements, velocities and accelerations of the
masses and rigid bodies.
• Determine the static equilibrium configuration of the system and
measure the displacements of the masses and rigid bodies from their
respective static equilibrium positions.
• Draw the free body diagram of each mass or rigid body in the system.
Indicate the spring, damping and external forces acting on each mass or
rigid body when positive displacement or velocity are given to that mass
or rigid body.
• Apply Newton’s second law of motion to each mass or rigid body shown
by the free body diagram as:
or

7
8
 The equations of motion of complete system:

Mass
matrix

Damping
matrix

Stiffness
matrix
9
 Displacement Velocity Acceleration Force Vector

For an undamped system the equations of motion reduce to:

• The differential equations of the spring‐mass system considered in the

example, can be seen to be coupled. Each equation involves more than one
coordinate. This means that the equations can not be solved individually one
at a time; they can only be solved simultaneously.
• In addition, the system can be seen to be statically coupled since stiffness are
coupled‐ that is the stiffness matrix has at least one nonzero off‐diagonal
term. On the other hand, if the mass matrix has at least one off‐diagonal term
nonzero, the system is said to be dynamically coupled. Further, if both the
stiffness and the mass matrices have nonzero off‐diagonal terms, the system
is said to be coupled both statically and dynamically. 10
Influence Coefficients

The influence coefficients corresponding to the inverse stiffness matrix are

called the flexibility influence coefficients, and those corresponding to the
inverse mass matrix are known as the inverse inertia coefficients.
Stiffness Influence Coefficients

11
First, we set the displacement of mass 1 equal to one and the displacements
of mass 2 and mass 3 equal to zero The equilibrium of forces for the different
masses and in the horizontal direction yields:

The solution

Next the displacements of the masses are assumed as x1=0, x2=1, x3=0

12
The force equilibrium equations of the masses are:

The solution yields:

Next the displacements of the masses are assumed as x1=0, x2=0, x3=1
The force equilibrium equations lead to:

The solution yields:

Thus the stiffness matrix of the system is given by

The stiffness and flexibility matrices are the inverse of one another
13
Eigenvalue Problem
if the system is conservative, there are no non conservative forces so the equations
of motion become

Note also that if the generalized coordinates x are same as the actual (physical)
displacements, the mass matrix [m] is a diagonal matrix.

The solution of Eq. above corresponds to the undamped free vibration of the
system. In this case, if the system is given some energy in the form of initial
displacements or initial velocities or both, it vibrates indefinitely, because there
is no dissipation of energy. We can find the solution of above Eq. by assuming a
solution of the form

The configuration of the system, given by the vector

is known as the mode shape of the system.

Substituting in EOM, we get

Equation of motion can be written in scalar form as n separate equations

14
 From which we can obtain the relations

Since the left side of Eq. above is independent of the index i, and the right side is
independent of t, both sides must be equal to a constant. By assuming this constant
as ω² we can write Eq. above as

OR

Eigen value or characteristic value problem:

For a nontrivial solution of Eq. above, the determinant of the coefficient matrix must
be zero. Characteristic equation:
ω² is known as the eigen value or the characteristic value, and ω is called the natural
frequency of the system. 15
The expansion of characteristic equation leads to an nth-order polynomial
equation in ω². The solution (roots) of this polynomial or characteristic
equation gives n values of ω² . It can be shown that all the n roots are real and
positive when the matrices [k] and [m] are symmetric and positive definite.

Solution of the Characteristic (Polynomial) Equation

where

By pre-multiplying Eq. above by we obtain

OR

where [I] is the identity matrix and is called dynamic matrix.

16
Problem for Eigen Value, Eigen Vector, Natural Frequencies, Mode Shapes
A three-degree-of-freedom spring-mass system

Where,

The free-vibration equations can be obtained as:

The dynamical matrix is given by:

17
For and
Flexibility matrix [a] can be calculated by taking inverse of [k] matrix.

and

Thus,

By setting the characteristic determinant equal to zero, we obtain the

frequency equation.

where

By dividing throughout by λ, gives

18
 where

The roots of the cubic equation are given by:

Once the natural frequencies are known, the mode shapes or eigenvectors can be
calculated using equation:

where

denotes the ith mode shape. The procedure is outlined next slide.

19
First Mode: By substituting the value of ω1 (i.e. ), we obtain:

That is,

The solution of Eqs. can be obtained and first mode shape is given by:

Second Mode: By substituting the value of ω2. Solution of Eqs.:

Third Mode: By substituting the value of ω3. Solution of Eqs.:

20
The values of is chosen arbitrarily. Values are usually taken as 1.
The mode shapes are shown in Figure below:

21
Orthogonality of Normal Modes

Eigen value or characteristic value problem for i th mode

Eigen value or characteristic value problem for j th mode

By pre multiplying above Eqs. By and respectively, we obtain, by

considering the symmetry of the matrices [k] and [m],

By subtracting Eqs. above, we obtain

In general, leads to

we also obtain,

22
Equations indicate that the modal vectors and are orthogonal with
respect to both mass and stiffness matrices.
When i=j the left sides of Eqs. are not equal to zero, but they do yield the
generalized mass and stiffness coefficients of the i th mode:

Equations can be written in matrix form as

23
where [X] is called the modal matrix, in which the i th column corresponds to
the i th modal vector:
In many cases, we normalize the modal vectors such that that is

In this case the matrix reduces to

24
Modal Analysis
Modal analysis is a powerful tool to determine the free and forced vibrations of
MDOF systems. We can Consider Physical MDOF system to be replaced by
several SDOF Systems, each SDOF system representing one Specific Natural
Mode. This process of determining the modal masses and stiffness in each mode
of Vibration of a MDOF and determine the response in each of the modes to
determine the total behavior is modal analysis.
Undamped Free Vibrations:
We saw that there exist a set of coordinates called the principal coordinates, which
express the equations of motion in uncoupled form. Such coordinates enabled each
equation to be solved independently of others.
A modal matrix [U] is referred to a square matrix where in each column represents
an eigenvector.
Thus for an n degree of freedom system

25
Let the differential equation of motion for an n-degree of freedom undamped system be
M   
x   K x   0 
 
To decuple these equations, let us use the linear transformation
x  Uy
where yare the principal coordinate s and can be found out by pre
- multiplyin g the above equation by U to
1

U1x  U1 Uy; y  U1x

or substituting in EOM

MUy  K Uy  0
 
Pre-multiplying the above equation by transpose modal matrix U   '

 
U  M U  y  U ' K U y  0
'

 
The term U ' M U and U '
K U  in above equation are each a diagonal matrix.
Since the off diagonal terms which involve rth row and sth column express the
orthogonality relationships which are zero:
26
 that is Xr M Xs  0, r  s
'

and Xr K Xs  0, r  s

'

The diagonal terms inU  M U  and U K U  which involves rth row and rth
' '

column give us the generalized mass and the generalized stiffness terms,
that is Xr M Xs  M r , r  s
'

and Xr K Xs  K r , r  s

'

Thus equation of motion becomes

Further, it can easily be seen that K r  λ r Mr

27
 Thus EOM becomes:

Where r  r  the eigen valu e for the r th mode

2

Thus the above equation as detailed below are n-uncoupled differential equations of
motion for an n-degree of freedom system in terms of the principal coordinates y.

yr   2 r yr  0 (r  1,2,......n)
The solution of the above equation is
yr  {A r cosr t  Br sin r t] (r  1,2,......n)
The above equation can be rewritten as
y  {Acos t  Bsin t]
so

Which give the vibratory response of undamped free vibrations Ar and Br (r  1,2,......n)
can be obtained from initial conditions.
Previous equation can be seen to be expanded and put in the following form which is
more convenient at times.

or
x  X1 y1  X2 y 2  ....Xr y r  ....Xn y n

29
Damped Free Vibrations

Differential equation of motion for a n-degree of freedom damped system be

  
M x   [c] 

   K x  0 
x
   
Where [c] is the damping matrix

Apart from static and/or dynamic coupling amongst the generalized coordinates
x1 , x 2 ......x n , there exist now damping coupling also. The equations can get uncoupled
with regard to damping if the damping matrix contains only the diagonal terms i.e. if
Cij  0, i  j
Let us use a linear transformation x  Uy as before in undamped case, {y} being
the column of principal coordinates.
Substituting above and pre-multiplying transpose of [U], we get:
  
U  M U  y  U CU  y  U ' K U y  0
' '

   
30
First and last term of previous equation reduce to following diagonal matrices:

Term containing [C] does not reduce to a diagonal matrix unless we use a concept of
proportional damping i.e. [C] being proportional to [M] or to [K] or a linear
combination of both. Let C  α[M]  β[K]
Where α and β are constants.
Then U CU  αU [M]U  βU [K]U
' ' '

OR

 
Equating term α  βω2 r  2ζ r ωr 31
So final equation give us

Above equations are n-uncoupled differential equations in terms of principal

coordinates as follows:  
yr  (2 rr ) y r   r yr  0 (r  1,2,......n)
2

 r  modal damping ratio in r th mode

r  undamped natural frequency in r th mode
Solution of above equation
yr  e rr t [A r cos 1 -  r r t  Br sin 1 -  r r t] (r  1,2,......n)
2 2

Now solution can be written as: OR

32
Forced vibrations
The differential equation of motion for forced vibration of a n-degree of freedom
system can be written as: M  


 x   [c] x   K x  F
   

To decouple use linear transformation as before x  Uy

Substituting and pre-multiplying by transpose of [U], we get
  
U  M U  y  U CU  y  U ' K U y  U ' F
' '

   
Using the concept of proportional damping equation becomes:

Left hand side is identical to damped vibration case and right hand side of equation can
be written as follows:

where 33
So final equation is as follows:

It gives n uncoupled equations of vibrations.

E r is referred to as excitation function for the r th mode.
Now solution can be written as:

34
 Multi-rotor system example These system are quite common in internal
combustion engines besides many other machinery

Consider a n - rotor system having (n - 1) connecting shafts . The moments

of inertia of rotors are represente d by J1 , J 2 ........J n . Torsional stiffness of
shafts by k t1 , k t 2 .....k t n-1 . Let, at an instant θ1 , θ 2 .....θ n be the displaceme nts
of various rotors from their equillibri um positions when system is having
torsional vibration s. The twists of respective shafts at this instants are
1   2 ,  2  3 ... n -1   n .
From Newton’s second law of motion the differential equations of motion for different
rotors are:

35
 Adding all the above equations, we have
The left hand side of the above equations represents the sum of the inertia torques on all
discs and it is equal to zero since there is no external torque on the system. However, if
there are external torques acting on the system at different points, then we must have

where Text is the sum of all external torques on the system.

For free vibration, let us assume for principal mode of vibration, the solution to be

Substituting in equation of motion:

Eliminatin g β1 , β 2 ...β n from above n homogenous set of equations

in β1 , β 2 ...β n , and the resulting n th degree equation in ω2 would then
give n natural frequencie s of the system. 36
In terms of matrix equations, the frequency equation for the system as given in
determinant form is:

After substituting the matrix values, the frequency equation can be obtained directly.

37
Special Case Three Rotor System
For a case of a three rotor system having two connecting shafts as shown in figure
equations becomes:

On rearranging we have

This is a homogeneou s set of equations in

β1 , β 2 and β 3 , and can have a solution only
if the determinen t formed with thei r co -
efficients vanishes; or
Three rotor system and its
mode shapes

38
In the determinant form, the matrix expressions for the system of Fig. are:

Frequency equation is:

Expanding the determinant of equation gives

The above equation is cubic in ω2 with one of the roots of ω2  0.

This should have been expected as we are dealing with what is known as a semi-
39
definite system
The two definite frequencies in this three rotor system can be obtained form equation as

The mode shapes can be obtained from the first and third equation and these are

When ω=0, both the above ratios are unity, indicating that the whole system rotates
rigidly, and there is no vibration as such. This is typical in a semi-definite system. For
value of   1 (the smaller of the two natural frequencies), one of the ratios in
equation is positive while the other is negative. And for   2 both ratios are negative.
40
The mode shapes are of the form shown in figure.