Coordinate Coupling

 The term coupling is used in vibration analysis to indicate a

connection between equations of motion.
 The general differential equation of motion for a 2-DOF
undamped vibration system has the form
 m11 m12   x1 
  k11 k12   x1 
 
0

          
m21 m22  
 x2   k21 k22    
 x2  0


 Coordinate coupling can be defined as a vibration in one part of

the system induces vibration in another part of the same system
due to the force transmitted through the coupling spring or
dashpot.
 OR, the displacement of one mass will be felt by another mass in
the same system since they are coupled together.
Dr Asif Israr - Institute of Space Technology
 Coordinate Coupling
2

 Two types:
 Static/Stiffness coupling: If the stiffness matrix is nondiagonal
 Dynamic/Mass coupling: If the mass matrix is nondiagonal

 In general, it is possible to find a coordinate system that has

neither mass/dynamic coupling nor stiffness/static coupling.
These equation are then decoupled into two independent
equations and can be solved independently. Such coordinates are
called principal or normal coordinates

Dr Asif Israr - Institute of Space Technology

 Coordinate Coupling
3

 Hence the matrix equation for a 2-DOF undamped system with no

dynamic and static coupling has the form
m11 0   x1 
 k11 0   x1 
  0

          
 0 m22  
 x2    0 k22    
 x2  0

 For simplicity, a rigid slab AB of total mass m represents the
automobile body with its mass centre (C.G) at distances l1 and l2
from the springs k1 and k2, respectively.

Dr Asif Israr - Institute of Space Technology

 Coordinate Coupling
4

 For this system, any of the following sets of coordinates may be

used to describe the motion:
 Deflections x1(t) and x2(t) of the two ends of the slab AB
 Deflection x(t) of C.G and rotation θ(t)
 Deflection x1(t) of the end A and rotation θ(t)
 Deflection y(t) of any point P located at a distance ε to the left of the
C.G and rotation θ(t)
 Thus any set of these coordinates (x1,x2),(x,θ),(x1,θ), and (y,θ)
represents the generalized coordinates for this system

Dr Asif Israr - Institute of Space Technology

 Coordinate Coupling
5

 Figure shows a FBD corresponding to the body in displaced

position. The displacements consist of the vertical translation x(t)
of the C.G and rotation θ(t) about the C.G.

 Static coupling:

mx  k1 ( x  l1 )  k2 ( x  l2 )

J  k ( x  l  )l  k ( x  l  )l
1 1 1 2 2 2

Dr Asif Israr - Institute of Space Technology

 Coordinate Coupling
6

or
mx  (k1  k2 ) x  (k2l2  k1l1 )  0


J  (k l  k l ) x  (k l 2  k l 2 )  0
2 2 11 11 2 2

 In matrix form

m 0    x
  (k1  k2 ) (k2l2  k1l1 )  
x
     2  
 
 0 J    
 (k2l2  k1l1 ) (k1l1  k2l2 ) 
2

 If k1l1  k2l2 , the coupling disappears, and we obtain uncoupled x
and θ vibrations

Dr Asif Israr - Institute of Space Technology

 Coordinate Coupling
7

 Dynamic coupling:
 There is some point P along the slab where a force applied normal to the slab
produces pure translation as shown in the figure.
 Here y(t) and θ(t) are used as generalized coordinates.
 The equations of motion can be written from the FBD
my  k1( y  l1 )  k2 ( y  l2 )  m  (k1  k2 ) y  (k2l2  k1l1)  m
J p  k1( y  l1 )l1  k2 ( y  l2 )l2  my  (k2l2  k1l1) y  (k1l12  k2l22 )  my

In matrix form


m m   y
  (k1  k2 ) (k2l2  k1l1)    y
  0

      
2  
  
m J p            
0
2
( k
   2 2 1 1 l k l ) ( k l
11 k l )
2 2   

Dr Asif Israr - Institute of Space Technology

 Coordinate Coupling
8

 Static and Dynamic Coupling: If we choose x = x1 at the end of

the slab as shown in figure, the equation of motion become
mx1  k1x1  k2{x1  (l1  l2 ) }  ml1  (k1  k2 ) x1  k2 (l1  l2 )  ml1
J1  k2{x1  (l1  l2 ) }(l1  l2 )  ml1x1  k2 (l1  l2 ) x1  k2 (l1  l2 )2  ml1x1
 In matrix form
 m ml1  
x1 
  (k1  k2 ) k2 (l1  l2 )   x1 
  0
           
 1
ml J 
1  
  k (l 
  2 1 2 l ) k (l
2 1  l2 
) 2

   
  0 

 Both static and dynamic coupling are now

present
Dr Asif Israr - Institute of Space Technology
 Problem 1
9

 Determine the normal modes of vibration of an automobile

simulated by the simplified 2-DOF system with the following
numerical values
W = 3220 Ib l1 = 4.5 ft k1 = 2400 Ib/ft
Jc = Wr2/g l2 = 5.5 ft k2 = 2600 Ib/ft
r = 4 ft l = 10 ft

Dr Asif Israr - Institute of Space Technology

 Problem 1
10

 The EoM of this system is

mx  k1 ( x  l1 )  k2 ( x  l2 )  0  mx  (k1  k2 ) x  (k2l2  k1l1 )  0
J c  k1 ( x  l1 )l1  k2 ( x  l2 )l2  0  J c  (k2l2  k1l1) x  (k1l12  k2l22 )  0
 Indicate static coupling. Let the system be harmonic
x  X sin t and    sin t
mX 2 sin t  (k1  k2 ) X sin t  (k2l2  k1l1 ) sin t  0
 J c 2 sin t  (k2l2  k1l1 ) X sin t  (k1l12  k2l22 ) sin t  0

 or (k1  k2  m 2 ) X  (k2l2  k1l1 )  0

(k2l2  k1l1 ) X  (k1l12  k2l22  J c 2 )  0
(k1  k2  m 2 ) (k2l2  k1l1 )  X   0
 In matrix form       
 ( k l
2 2  k l
11 ) ( k l
11
2
 k l
2 2
2
 J c 2
) 
    
  0 
Dr Asif Israr - Institute of Space Technology
 Problem 1
11

 From the determinant of the matrix

equation, the two natural frequencies are
 ω1 = 6.90 rad/s = 1.10 cps

 ω2 = 9.06 rad/s = 1.44 cps

 The amplitude ratios for the two

frequencies are

X  (k2l2  k1l1 )
    14.6 ft / rad  3.06 in / deg
  1 (k1  k2  m )
2

X  (k2l2  k1l1 )
    1.09 ft / rad  0.288in / deg
  2 (k1  k2  m )
2

Dr Asif Israr - Institute of Space Technology

 Problem 2
12

 For the 2 DOF shown in the figure, determine

 The differential equations of motion using the coordinates x1 and x2
for m and 2m
 The normal mode frequencies
 If the coordinate x1 and θ at m are used. Find the type of coupling
that exists

Dr Asif Israr - Institute of Space Technology

 Problem 2
13

Solution:
(a) The FBD is shown in the previous slide. The equation of motion
are:
mx1  2mx2  k (2 x1  x2 )  k (2 x2  x1 )
L 2L Eqn. (1)
J G  k (2 x2  x1 )( L  )  k (2 x1  x2 )( L  )
3 3
x1  x2 2L 2 L 2
 Since  and J G  m( )  2m( )2  mL2
L 3 3 3
 Substituting in Eq. (1), we get
mx1  2mx2  kx1  kx2
2   
2 x1  x2
  4L   5L 
mL    k (2 x2  x1 )
  3   k (2 x1  x2 ) 3 
3  L     
Dr Asif Israr - Institute of Space Technology
 Problem 2
14

 Further simplification gives:

mx1  2mx2  kx1  kx2

Eqn. (2)
2mx1  2mx2  14kx1  13kx2

 Eqn. (2) can be written in matrix form

m 2m   x1 
  k k   x1 
  0

          
2m  2m 
 x2   14k  13k    
 x2  0


Dr Asif Israr - Institute of Space Technology

 Problem 2
15

(b) From the determinant of the matrix equation, the two natural
frequencies are

k k
1  0.811 and 2  2.62
m m

 The amplitude ratios for the two frequencies are

 X1  2 m  2
k
    0.921
X  k  m 2
 2 1

 X1  2 m  2
k
    2.17
X  k  m 2
 2 2

Dr Asif Israr - Institute of Space Technology

 Problem 2
16

(c) If the coordinates x1 and θ at m are used, the EoMs are given by

mx1  2m( x1  L)  k ( x1  L )  k ( x1  2 L )

 4L 5L
J G  k ( x1  2 L )  k ( x1  L )
3 3
 We can observe both static and dynamic coupling from the above
set of equations

Dr Asif Israr - Institute of Space Technology

 Forced Harmonic Vibration of 2-DOF System
17

 For forced vibration without damping, the EoMs are

m1x1  k1x1  k2 ( x2  x1 )  Fo sin t
m2 x2  k2 ( x2  x1 )

 or m1x1  ( k1  k2 ) x1  k2 x2  Fo sin t
Eqn. 1
m2 x2  k2 x1  k2 x2  0

 It can be expressed in matrix form

m1 0   x1 
 (k1  k2 ) k2   x1  Fo sin t 
  
          
 0 m2  x 
  2    k 2 k 2   x 
 2    0 


Dr Asif Israr - Institute of Space Technology

 Forced Harmonic Vibration of 2-DOF System
18

 Let the system be harmonic, the solution can be assumed as


 x1 
   X1 

    sin t


 x2 
  X 2 


 Substituting this into the differential equation i.e. Eqn. 1, we

obtain
m1 X1 2 sin t  (k1  k2 ) X1 sin t  k2 X 2 sin t  Fo sin t
 m2 X 2 2 sin t  k2 X1 sin t  k2 X 2 sin t  0

(k1  k2  m1 2 ) X1  k2 X 2  Fo
or
 k2 X1  (k2  m2 2 ) X 2  0
Dr Asif Israr - Institute of Space Technology
 Forced Harmonic Vibration of 2-DOF System
19

(k1  k2  m1 2 )  k2  X1 

  Fo 

 2 
 
  
  k2 (k2  m2 )   
X 2  0 

 Solving by Cramer’s rule, the amplitudes of the force vibration

are Fo  k2
0 (k2  m2 2 ) Fo (k2  m2 2 )
X1  
(k1  k2  m1 2 )  k2 (k1  k2  m1 2 )( k2  m2 2 )  k22
 k2 (k2  m2 2 )

(k1  k2  m1 2 ) Fo
 k2 0 Fo k2
X2  
(k1  k2  m1 2 )  k2 (k1  k2  m1 2 )( k2  m2 2 )  k22
 k2 (k2  m2 2 )
Dr Asif Israr - Institute of Space Technology
 Problem: Forced Harmonic Vibration
20

 A block of mass, m resting on a frictionless horizontal plane, is

connected through a spring of constant, k to a homogeneous
uniform rod of mass, M and length, L as shown in figure.
Determine the steady state response of the block

Free Body Diagram

Dr Asif Israr - Institute of Space Technology

 Problem: Forced Harmonic Vibration
21

(k  m 2 ) kL X   Fo 


      
  kL (kL  Mga  J ) 
2 2

     
0 
 Solving above, we get
Fo kL
0 (kL2  Mga  J 2 ) Fo (kL2  Mga  J 2 )
X  
(k  m 2 )  kL (k  m 2 )( kL2  Mga  J 2 )  (kL)2
 kL (kL2  Mga  J 2 )

 Steady state response of the block is

Fo (kL2  Mga  J 2 )
x(t )  X sin t  sin t
(k  m )( kL  Mga  J )  (kL )
2 2 2 2

Dr Asif Israr - Institute of Space Technology

 Problem: Forced Harmonic Vibration
22

 Applying the force equation  F  ma to the block, and the

moment equation  M  J to the pivoted point O, we have
mx  k ( x  L )  Fo sin t  mx  kx  kL  Fo sin t
J  k ( x  L ) L  Mga  J  ( Mga  kL2 )  kLx  0
where moment of inertia of the rod w.r.t. to point O is J = ML2/3
 Let x  X sin t ,    sin t ; then
m 2 X  kX  kL  Fo  (k  m 2 ) X  kL  F0
 J 2  (kL2  Mga)  kLX  0  (kL2  Mga  J 2 )  kLX  0

Dr Asif Israr - Institute of Space Technology

 Damped Free Vibrations
23

 Consider the free vibration response of the viscously damped two

degree of freedom system as shown in figure

Dr Asif Israr - Institute of Space Technology

 Damped Free Vibrations
24

 For convenience, x1(t) and x2(t) are assumed positive, with x2 > x1.
 Referring to the Free Body Diagram, and applying Newton’s
second law for each mass, we obtain the two equations of motion
as

m1x1  k1 x1  c1 x1  k2 ( x2  x1 )  c2 ( x2  x1 )

Eqn. 1
m2 x2   k2 ( x2  x1 )  c2 ( x2  x1 )
 Or, rewriting:
m1x1  (c1  c2 ) x1  (k1  k2 ) x1  k2 x2  c2 x2  0
Eqn. 2
m2 x2  c2 x2  k2 x2  k2 x1  c2 x1  0
Dr Asif Israr - Institute of Space Technology
 Damped Free Vibrations
25

 Eqn. (2) is statically (elastically) and dynamically coupled

through the stiffness and damping terms
 Eqn. (2) can now be written in the compact form as

[m]{x}  [c]{x}  [k]{x}  0

where, the mass, damping and stiffness matrices are:

m1 0  c1  c2 c2  k1  k2  k2 

[ m]    [c ]    [k ]   
 0 m2
   c2 c2    k2 k2 

Dr Asif Israr - Institute of Space Technology

 Damped Free Vibrations
26

 And the displacement matrix or displacement vector is


 x1 (t ) 

{x(t )}   

 2 
x (t ) 
 Since the components of vibration for a damped system are non-
periodic or oscillatory with diminishing amplitudes, we assume a
solution of the form

 A

{x(t )}   est

B 


Dr Asif Israr - Institute of Space Technology

 Damped Free Vibrations
27

 Differentiating

 
  st 
sA  2 
 A st
s

{x(t )}   e {x(t )}   2 e

sB 
s B 
 Substituting into the differential equation (2) and dividing through
by est, we get:

[m1s 2  (c1  c2 )s  (k1  k2 )] A  (c2 s  k2 ) B  0

[m2s 2  c2s  k2 ]B  (c2 s  k2 ) A  0

Dr Asif Israr - Institute of Space Technology

 Damped Free Vibrations
28

 Non-trivial solutions exist only if

[m1s 2  (c1  c2 ) s  (k1  k2 )] (c2 s  k2 )

0
 (c2 s  k2 ) m2 s  c2 s  k2
2

 Expending the determinant, we obtain

[m1s2  (c1  c2 )s  (k1  k2 )][ m2s2  c2s  k2 ]  (c2s  k2 )2  0

 Which is known as the characteristic equation of the system, and

will yield four values for ‘s’. Hence the complete solution of the
system can be written as:
Dr Asif Israr - Institute of Space Technology
 Damped Free Vibrations
29

x1 (t )  A1es t  A2es t  A3es t  A4es t

1 2 3 4

x2 (t )  B1es t  B2es t  B3es t  B4es t

1 2 3 4

where the four unknown coefficients A1, A2, A3 and A4 (B1 = λ1A1,
B2 = λ2A2, B3 = λ3A3 and B4 = λ4A4) are to be determined from the
four initial conditions of the system. The four initial conditions
are:
x1 (0), x2 (0), x1 (0), x2 (0)
 The amplitude ratios are found from the algebraic equations with
coefficients A and B:
Ai c2 si  k2 1
  i  1,2,3,4
Bi m1si  (c1  c2 ) si  (k1  k2 ) i
2

Dr Asif Israr - Institute of Space Technology

 Damped Free Vibrations
30

 Assuming the roots of the characteristic equation are complex,

then there must be the corresponding complex conjugate roots,
i.e.,
s1  (r  id ) and s2  (r  id )

and hence
A1e ( r  id )t  A2e ( r id )t  e rt[ A1eidt  A2eidt ]
 e rt[ A1 cos dt  iA1 sin dt  A2 cos dt  iA2 sin dt]
 Ce rt sin(dt   )
 Similarly,
B1e s1t  B2e s 2 t  De  rt sin(dt   )
Dr Asif Israr - Institute of Space Technology
 Damped Free Vibrations
31

 Therefore, the general solution is

x1 (t )  Ce rt sin(dt  )  A3es t  A4es t

3 4

x2 (t )  De rt sin(dt  )  B3es t  B4es t 3 4

 Which are oscillatory with diminishing amplitude and aperiodic

Dr Asif Israr - Institute of Space Technology

 Forced Vibration with Damping
32

 Consider a system having forced vibration with damping

Dr Asif Israr - Institute of Space Technology

 Forced Vibration with Damping
33

 The EoMs for forced damped vibration are:

m1x1  k1x1  c1x1  k2 ( x2  x1 )  c2 ( x2  x1 )  Fo sin t
m2 x2  k2 ( x2  x1 )  c2 ( x2  x1 )

Or,

m1x1  (c1  c2 ) x1  (k1  k2 ) x1  k2 x2  c2 x2  Fo sin t
m2 x2  c2 x2  k2 x2  k2 x1  c2 x1  0
 Substituting: Fo sin t  Foeit
x1  X 1eit
x2  X 2eit
Dr Asif Israr - Institute of Space Technology
 Forced Vibration with Damping
34

 Rearranging and dividing through by eiωt, the EoMs become

(k  k )  m  2

 i(c1  c2 ) X1  (k2  ic2) X 2  Fo
 
1 2 1

 (k2  ic2) X1  k2  m2 2  ic2 X 2  0

Dr Asif Israr - Institute of Space Technology

 Forced Vibration with Damping
35

 Solving by Cramer’s rule

Fo (k2  ic2 )
0 (k2  m2 2  ic2 )
X1 
[( k1  k2 )  m1 2  i(c1  c2 ) ][ k2  m2 2  ic2 ]  (k2  ic2 )2

and,
[( k1  k2 )  m1 2  i (c1  c2 ) ] Fo
 (k2  ic2 ) 0
X2 
[( k1  k2 )  m1 2  i(c1  c2 ) ][ k2  m2 2  ic2 ]  (k2  ic2 ) 2

Dr Asif Israr - Institute of Space Technology

 Forced Vibration with Damping
36

 Further simplification gives

Fo2 (k2  m2 2 )2  c22 2
X1 
(m1m2 4  m1k2 2  m2k1 2  c1c2 2  k1k2 )2  (k1c2  k2c1  m1c2 3  m2c1 3  m2c2 3 )2

Fo2 (k22  c22 2 )

X2 
(m1m2 4  m1k2 2  m2k1 2  c1c2 2  k1k2 )2  (k1c2  k2c1  m1c2 3  m2c1 3  m2c2 3 )2

 Here, the forcing function is

Fo sin t  Im( Fo eit )
and x1  Im( X 1eit )  Im( X 1ei (t 1 ) )  X 1 sin(t  1 )
x2  Im( X 2 eit )  Im( X 2ei (t 2 ) )  X 2 sin(t  2 )
Dr Asif Israr - Institute of Space Technology
 Forced Vibration with Damping
37

 We can find the equations for φ1 and φ2 as follows:

i 1
X 1e  X 1 (cos1  i sin 1 )  E ( A  iB)
i 2
X 2e  X 2 (cos2  i sin 2 )  E (C  iD)

B 1  D 
1  tan   and 2  tan  
1

 A C

Dr Asif Israr - Institute of Space Technology