CHAPTER 8 TRANSVERSE VIBRATIONS-IV: MULTI-DOFs ROTOR SYSTEMS Transverse vibrations have been considered previously in great detail for mainly single mass rotor systems. The thin disc and long rigid rotors were considered with various complexities at supports; for example, the rigid disc mounted on flexible mass less shaft with rigid bearings (e.g., the simply supported, overhung, etc.), the flexible bearings (anisotropic and cross-coupled stiffness and damping properties), and flexible foundations. Higher order effects of the gyroscopic moment on the rotor, for most general case of motion, was also described in detail. However, in the actual case, as we have seen in previous two chapters for torsional vibrations, the rotor system can have several masses (e.g., turbine blades, propellers, flywheels, gears, etc.) or distributed mass and stiffness properties, and multiple supports, and other such components like coupling, seals, etc. While dealing with torsional vibrations, we did consider multi-DOF rotor systems. Mainly three methods were dealt for multi-DOF systems, that is, the Newton’s second law of motion (or the D’Alembert principle), the transfer matrix method, and the finite element method. We will be extending the idea of these methods from torsional vibrations to transverse vibrations along with some additional methods, which are suitable for the analysis of multi-DOF rotor system transverse vibrations. In the present chapter, we will consider the analysis of multi-DOF rotor systems by the influence coefficient method, transfer matrix method, and Dunkerley’s method. The main focus of these methods would be to estimate the rotor system natural frequencies, mode shapes, and forced responses. The relative merit and demerit of these methods are discussed. The continuous system and finite element transverse vibration analysis of multi-DOF rotor systems will be treated in subsequent chapters. Conversional methods of vibrations like the modal analysis, Rayleigh-Ritz method, weighted sum approach, collocation method, mechanical impedance (or receptance) method, dynamic stiffness method, etc. are not covered exclusively in the present text book, since it is readily available elsewhere (Meirovitch, 1986; Thomson and Dahleh, 1998). However, the basic concepts of these we will be using it directly whenever it will be required with proper references. 8.1 Influence Coefficient Method In transverse vibrations due to coupling of the linear and angular (tilting) displacements the analysis becomes more complex as compared to torsional vibrations. A force in a shaft can produce the linear as well as angular displacements; similarly a moment can produce the angular as well as linear displacements. Influence coefficients could be used to relate these parameters (the force, the moment, and the linear and angular displacements) relatively easily. In the present section, the influence
85
Embed
CHAPTER 8 TRANSVERSE VIBRATIONS-IV: MULTI-DOFs ROTOR SYSTEMS - iitg.ac.in · CHAPTER 8 TRANSVERSE VIBRATIONS-IV: MULTI-DOFs ROTOR SYSTEMS Transverse vibrations have been considered
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
CHAPTER 8
TRANSVERSE VIBRATIONS-IV: MULTI-DOFs ROTOR SYSTEMS
Transverse vibrations have been considered previously in great detail for mainly single mass rotor
systems. The thin disc and long rigid rotors were considered with various complexities at supports; for
example, the rigid disc mounted on flexible mass less shaft with rigid bearings (e.g., the simply
supported, overhung, etc.), the flexible bearings (anisotropic and cross-coupled stiffness and damping
properties), and flexible foundations. Higher order effects of the gyroscopic moment on the rotor, for
most general case of motion, was also described in detail. However, in the actual case, as we have
seen in previous two chapters for torsional vibrations, the rotor system can have several masses (e.g.,
turbine blades, propellers, flywheels, gears, etc.) or distributed mass and stiffness properties, and
multiple supports, and other such components like coupling, seals, etc. While dealing with torsional
vibrations, we did consider multi-DOF rotor systems. Mainly three methods were dealt for multi-DOF
systems, that is, the Newton’s second law of motion (or the D’Alembert principle), the transfer matrix
method, and the finite element method. We will be extending the idea of these methods from torsional
vibrations to transverse vibrations along with some additional methods, which are suitable for the
analysis of multi-DOF rotor system transverse vibrations. In the present chapter, we will consider the
analysis of multi-DOF rotor systems by the influence coefficient method, transfer matrix method, and
Dunkerley’s method. The main focus of these methods would be to estimate the rotor system natural
frequencies, mode shapes, and forced responses. The relative merit and demerit of these methods are
discussed. The continuous system and finite element transverse vibration analysis of multi-DOF rotor
systems will be treated in subsequent chapters. Conversional methods of vibrations like the modal
analysis, Rayleigh-Ritz method, weighted sum approach, collocation method, mechanical impedance
(or receptance) method, dynamic stiffness method, etc. are not covered exclusively in the present text
book, since it is readily available elsewhere (Meirovitch, 1986; Thomson and Dahleh, 1998).
However, the basic concepts of these we will be using it directly whenever it will be required with
proper references.
8.1 Influence Coefficient Method
In transverse vibrations due to coupling of the linear and angular (tilting) displacements the analysis
becomes more complex as compared to torsional vibrations. A force in a shaft can produce the linear
as well as angular displacements; similarly a moment can produce the angular as well as linear
displacements. Influence coefficients could be used to relate these parameters (the force, the moment,
and the linear and angular displacements) relatively easily. In the present section, the influence
421
coefficient method is used to calculate natural frequencies and forced responses of rotating machines.
Up to three-DOF rotor systems the hand calculation is feasible, however, for more than three-DOF the
help of computer routines (e.g., MATLAB, etc.) are necessary. The method is described for multi-
DOF i.e., n number of discs mounted on a flexible shaft (Figure 8.1) and supported by rigid bearings;
which can be extended for the multi-DOF rotor system with flexible supports.
Figure 8.1 A multi-DOF rotor system mounted on rigid bearings
8.1.1 The static case: Let f1, f2 , …, and fn are static forces on discs 1, 2, …, and n respectively, and
x1, x2, …, xn are the corresponding shaft deflections at discs. The reference for the measurement of the
shaft displacement is from the static equilibrium and the system under consideration is linear, so that
gravity effect will not appear in equilibrium equations. If a force f is applied to the disc of mass m1,
then deflection of m1 will be proportional to f, i.e.
or (8.1)
where α11 is a constant, which depends upon the elastic properties of the shaft and support conditions.
It should be noted that we will have deflections at other disc locations (i.e., 2, 3, …, n) as well due to
force at disc 1 due to the coupling. Now the same force f is applied to the disc of mass m2 instead of
mass m1, then the deflection of m1 will still be proportional to the force, i.e.
or (8.2)
where, α12 is another constant (the first subscript represents displacement position and the second
subscript represent the force location). In general we have α12 ≠ α11. Similarly, if force f is applied to
the disc of mass mn, then the deflection at m1 will be
1 1nx fα= (8.3)
1x f∝ x fα=1 11
1x f∝ fx 121 α=
422
where, α1n is a constant. If forces f1, f2, ..., and fn are applied at the locations of all the masses
simultaneously, then the total deflection at m1, will be summation of all the three displacements
obtained above by the use of superposition theorem, as
1 11 1 12 2 1n nx f f fα α α= + + +� (8.4)
In the equation, it has been assumed that displacements are small so that a linear relation exists
between the force applied and corresponding displacement produced. Similarly, we can write
displacement at other disc locations as
2 21 1 22 2 2n nx f f fα α α= + + +� (8.5)
and
1 1 2 2n n n nn nx f f fα α α= + + +� (8.6)
Here α2j, …, αnj, where j =1, 2, …, n are another sets of constants and can be defined as described for
α1i above. Hence, in general αij is defined as a displacement at ith station due to a unit external force at
station jth and keeping all other external forces to zero. Equations (8.4) to (8.6) can be combined in a
matrix form as
1 11 12 1 1
2 21 22 2 2
1 2
n
n
n n n nn n
x f
x f
x f
α α α
α α α
α α α
=
�
�
� � � � � �
�
(8.7)
It should be noted that due to the transverse force actually both the linear and angular displacements
take place, i.e., a coupling exists between the linear and angular displacements. We have already seen
such coupling in Chapter 2 due to bending of the shaft and due to gyroscopic couples, respectively.
Moreover, we have seen in Chapters 4 and 5 couplings between the horizontal and vertical plane
linear motions (x and y) due to dynamic properties of fluid-film bearings and between the horizontal
and vertical plane angular motions (ϕx and ϕy), respectively.
Similarly, a moment gives the angular displacement as well as the linear displacement. The method
can be extended to account for the angular displacement (tilting), ϕy, of the disc, and for the
application of the point moment, M, at various disc locations along the shaft a part of loading on discs.
Then, the equation will take the following form
(8.8) { } [ ]{ }d fα=
423
with
{ }1
1
n
n
y
y
x
xd
ϕ
ϕ
=
�
�
; { }
1
1
n
n
f
ff
M
M
=
�
�
and
11 12 1(2 )
21 22 2(2 )
(2 )1 (2 )2 (2 )(2 )
n
n
n n n n
α α α
α α α
α α α
�
�
� � � �
�
(8.9)
which gives
(8.10)
where αij, with i, j = 1, 2, …,n are influence coefficients. The first subscript defines the linear (or
angular) displacement location and the second subscript defines the force (or moment) location. The
analysis so far has referred only to static loads applied to the shaft. When the displacement of the disc
is changing rapidly with time, the applied force has to overcome the disc inertia as well as to deform
the shaft.
8.1.2 The dynamic case: In Figure 8.2 free body diagrams of a disc and the shaft is shown. Let
and (not shown in free body diagrams for brevity) be the external force and moment on the disc
whereas and are the reaction force and moment transmitted to the shaft (which is equal
and opposite to the reaction force and moment of the shaft on the disc). From the force and moment
balance of disc 1, we have
and 1 11 1 d y
M M I ϕ′ − = �� (8.11)
where Id is the diametral mass moment of inertia the disc.
Figure 8.2 Free body diagrams of (a) a disc and (b) the shaft
1{ } [ ] { }f dα −=
1f ′
1M ′
1m 1f 1M
1111 xmff ��=−′
424
Similarly at other disc locations, we can write
and 2 22 2 d y
M M I ϕ′ − = �� (8.12)
and
n n n nf f m x′ − = �� and n nn n d y
M M I ϕ′ − = �� (8.13)
Substituting for 1 2 1 2, , , , , ,nf f f M M� �
and
nM from equations (8.11)-(8.13) and remembering that
for the simple harmonic motion of discs 2
x xω= −�� and
2
y yϕ ω ϕ= −�� (when no external excitation is
present then ω is the natural frequency of the system ωnf , and when there is an external excitation
then it is equal to the excitation frequency, ω), equation (8.8) gives
{ } [ ]1 1
2
1 1 1
2
2
1
2
n
n n n
d y
n d y
f m x
f m xd
M I
M I
ω
ωα
ω ϕ
ω ϕ
′ + ′ +
= ′ +
′ +
�
�
(8.14)
which can be expanded as
{ } [ ]{ } [ ]1
1 1
2
1
n
n n
nfd
d n
m x
m xd f
I
I
α ω αϕ
ϕ
′= +
�
�
with { }
1
1
n
n
f
ff
M
M
′ ′
′ = ′
′
�
�
(8.15)
In view of equation (8.9), equation (8.15) can be rearranged as
{ } [ ]{ } { }
1
1
1
1
11 1 1 1( 1) 1(2 )
1 1 ( 1) (2 )2
( 1)1 1 ( 1) ( 1)( 1) ( 1)(2 )
(2 )1 1 (2 ) (2 )( 1) (2 )(2 )
n
n
n
n
n n n d n d
n nn n n n d n n d
nf
n n n n n n d n n d
n n n n n n d n n d
m m I I
m m I Id f d
m m I I
m m I I
α α α α
α α α αα ω
α α α α
α α α α
+
+
+ + + + +
+
′= +
� �
� � � � � �
� �
� �
� � � � � �
� �
(8.16)
2222 xmff ��=−′
425
which can be written in more compact form as
[ ] [ ] { } [ ]{ }2 2
1 1A I d fα
ω ω
′− = −
(8.17)
with
[ ] { }
1
1
1
1
11 1 1 1( 1) 1(2 )
1 1 ( 1) (2 )
( 1)1 1 ( 1) ( 1)( 1) ( 1)(2 )
(2 )1 1 (2 ) (2 )( 1) (2 )(2 )
n
n
n
n
n n n d n d
n nn n n n d n n d
n n n n n n d n n d
n n n n n n d n n d
m m I I
m m I IA d
m m I I
m m I I
α α α α
α α α α
α α α α
α α α α
+
+
+ + + + +
+
=
� �
� � � � � �
� �
� �
� � � � � �
� �
(8.18)
Disc displacements x and ϕy can be calculated for known applied loads (e.g., the unbalance forces and
moments) as
(8.19)
with
[ ] [ ] [ ] [ ]1
2 2
1 1R A I α
ω ω
−
= − −
(8.20)
where R represents the receptance matrix and for the present case it contains only real elements. For
free vibrations the right hand side of equation (8.17) will be zero, i.e.
[ ] [ ] { } { }2
10A I d
ω
− =
(8.21)
which only satisfy when
[ ] [ ] { }2
10A I
ω− = (8.22)
and it will give the frequency equation and system natural frequencies could be calculated from this.
Alternatively, through eigen value analysis of matrix [A] system natural frequencies and mode shapes
could be obtained directly. In general, the receptance matrix, [R], may contain complex elements
when damping forces also act upon the shaft, in which case applied forces and disc displacements will
not all be in phase with one another. Hence, a more general form of equation (8.19) would be that
{ } [ ]{ }d R f ′=
426
which indicates both the real and imaginary parts of x, and R. In such case the real and imaginary
parts of each equations need to be separated and then these can be assembled again to into a matrix
form, which will have double the size that with complex quantities. Some of the steps are described
below
{ } { } [ ] [ ]( ) { } { }( )j j jr i r i r id d R R f f′ ′+ = + +
where r and i refer to the real and imaginary parts, respectively. Above equation can be expanded as
{ } { } [ ]{ } [ ]{ }( ) [ ]{ } [ ]{ }( )j jr i r r i i i r r id d R f R f R f R f′ ′ ′ ′+ = − + +
Now on equating the real and imaginary parts on both sides of equations, we get
{ } [ ]{ } [ ]{ }r r r i id R f R f′ ′= − and { } [ ]{ } [ ]{ }i i r r id R f R f′ ′= +
Above equations can be combined again as
[ ] [ ][ ] [ ]
r r i r
i i r i
d R R f
d R R f
′ − =
′
(8.23)
It can be observed that now the size of the matrix and vectors are double as that of equation (8.19).
Now through simple numerical examples (for the two or more DOFs) some of the basic concepts of
the present method will be illustrated.
f ′
427
Example 8.1 Obtain transverse natural frequencies of a rotor system as shown in Figure 8.3. Take the
mass of the disc, m = 10 kg and the diametral mass moment of inertia, Id = 0.02 kg-m2. The disc is
placed at 0.25 m from the right support. The shaft has a diameter of 10 mm and a span length of 1 m.
The shaft is assumed to be massless. Take the Young’s modulus E = 2.1 × 1011
N/m2
of the shaft.
Consider a single plane motion only.
Figure 8.3 A rotor system
Solution: Influence coefficients for a simply supported shaft are defined as
11 12
21 22
y
yzx
fy
M
α α
ϕ α α
=
with
and
It should be noted that subscript 1 represents corresponding to a force or a linear displacement, and
subscript 2 represents a moment or an angular displacement. To obtain natural frequencies of the rotor
system having a single disc, from equation (8.21), we have
[ ] [ ]11 122
2
21 22 2
1
1
1
d
nf
nf
d
nf
m I
A I
m I
α αω
ωα α
ω
−
− = −
(a)
Hence, the determinant of the above matrix would give the frequency equation as
Sy 1366381.12 195881901.58 1290696326.5 5308048574.2 27791417.25 27791417.25
471
Table 8.4(g) Relative values of state vectors at various stations for the seventh natural frequency
State vectors at seventh natural frequency, 10776 rad/s.
S0 S1 S2 S3 S4 S5
y 0 0.01552 0.2102 -0.2306 0.038953 0
ϕx 1.00 -1.6894 11.200 3.7243 -0.976793 0.035605
Myz 0 254747.85 -3468821.24 640227.57 22263.86 0
Sy -394294.80 6818067.54 128888160.07 -31812064.91 -148425.72 -148425.72
Table 8.4(h) Relative values of state vectors at various stations for the eighth natural frequency
State vectors at eighth natural frequency, 13282 rad/s.
S0 S1 S2 S3 S4 S5
y 0 -0.03468 0.02929 -0.001166 -0.0007505 0
ϕx 1.0 -2.69371 -1.06703 0.05671 -0.1437 8.9769
Myz 0 679094.86 16463.86 3645.74 200574.14 0
Sy -541528.33 -25017431.96 824200.49 -410329.07 -1337160.94 -1337160.94
Fig. 8.31(a) The fundamental mode shape of linear displacements
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.450
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
Axial position on the shaft
Dis
pla
cem
ent
ampli
tud
e
472
Fig. 8.31(b) The fundamental mode shape of angular (slope) displacements
Fig. 8.31(c) Higher mode shapes (2nd
to 5th) of linear displacements
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45-1.5
-1
-0.5
0
0.5
1
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45-0.08
-0.06
-0.04
-0.02
0
0.02
0.04
0.06
0.08
0.1
2
3
4 5
Axial position on the shaft
Dis
pla
cem
ent
ampli
tud
e
Axial position on the shaft
Dis
pla
cem
ent
ampli
tud
e
473
Fig. 8.31(d) Higher mode shapes (2nd
to 5th) of angular (slope) displacements
Example 8.8 Obtain bending critical speeds of a rotor system as shown in Figure 8.32. Take the mass
of the disc, kg and its diametral mass moment of inertia, kg-m2. The length of the
shaft segments are a = 0.3 m and b = 0.7 m; and the diameter of the shaft is 10 mm. Neglect the
gyroscopic effects. N/m2.
Figure 8.32 An overhang rotor system
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45-2
-1.5
-1
-0.5
0
0.5
1
1.5
2
5=m 0.02dI =
112.1 10E = ×
2
3
4
5
Axial position on the shaft
Dis
pla
cem
ent
ampli
tud
e
474
Solution: Figure 8.33 shows the station numbering and free body diagram of various segments and
supports.
Figure 8.33 Free body diagrams of rotor segments
For shaft segment (1) as shown in Figure 8.33(b), the state vector can be related as
(a)
with
For second branch, the state vector can be written as
(b)
with
End conditions for the overhang rotor as shown in Figure 8.33(a) can be written as
{ } 1 1 0 1 01[ ] [ ] { } [ ] { }
LS F P S U S= =
[ ] 1 11[ ] [ ]U F P=
{ } { } { }2 22 1 1[ ] [ ]
L R RS F S U S= =
2 2[ ] [ ]U F=
(a) Rotor system (b) FBD of segment (1)
(c) FBD of segment (2) (d) Free body diagram of a small shaft
segment at support A
475
At node 0: (c)
At node 1: 1 1 1 1 10; 0 and 0
L R L R ay M M S S R= = ≠ − + = (d)
where, Ra is support reaction at bearing A.
At node 2: (e)
From the first shaft segment, i.e., equation (a), on application of end conditions of equation (c), we
have
(f)
with
[ ]
1 211 12 13 14 6
21 22 23 24
2131 32 33 34
2
41 42 43 44 1 1
1 0 0 01 0.5
0 1 0 00 1 0.5
0 1 00 0 1
0 0 10 0 0 1
nf d
nf
u u u u l l l
u u u u lU
Iu u u u l
mu u u u
α α
α α
ω
ω
= = −
1 12 2 2 2
6 6
2 2
2 2
2
1
1 0.5 0.5
0.5 1 0.5
1
0 0 1
nf d nf
nf d nf
nf d nf
nf
m l l I l l l
m l I l
m l I l
m
ω α ω α α α
ω α ω α α α
ω ω
ω
+ −
− = −
(g)
and
(h)
From equation (f), the first set of equation gives
(i)
On using equation (i) into equation (f), we get state vectors in the left of station 1 as
0 00S M= =
2 2 20 and 0
R Ly M M= = =
11 12 13 14
21 22 23 24
31 32 33 34
41 42 43 441 01
0
0
0L
u u u u y
u u u u
u u u uM
u u u uS
ϕ ϕ
−
=
l
EIα =
1111 0 12 0 0 0
12
0 u
u y u yu
ϕ ϕ= − + ⇒ =
476
(j)
Now noting equations (d) and (j) we can have state vectors in the right of station 1 as
22 111 1 21 0
12
R L
u uu y
uϕ ϕ
= = − +
32 111 1 31 0
12
R L
u uM M u y
u
= = − +
42 111 1 41 0
12
R L a a
u uS S R u y R
u
= + = − + +
(k)
Equation (k) can be written as
021
31
411
0 0
0
0
1
a
R
y
yu
RuM
uS
ϕ
− =
(l)
with
Equation (l) can be written in a standard form as
(m)
with
10
Ly =
11 22 111 21 0 22 0 21 0
12 12
L
u u uu y u y u y
u uϕ
= − + = − +
32 11111 31 0 32 0 31 0
12 12
L
u uuM u y u y u y
u u
= − + = − +
11 42 111 41 0 42 0 41 0
12 12
L
u u uS u y u y u y
u u
= − + = − +
1 1 0R L
y y= =
32 1122 11 42 1121 21 31 31 41 41
12 12 12
; andu uu u u u
u u u u u uu u u
= − = − = −
{ } [ ]{ }111 SUSR =
477
{ }
0
21
1 131
41
0 0 0 0
00 0 0; and
00 0 0
0 0 1 a
y
uU S
u
Ru
− = =
For second branch, i.e., equation (b) and noting equation (m), we have
{ }
1 206
21
2 1 1231
412
0 0 0 01 0.5
00 0 00 1 0.5[ ][ ]{ }
00 0 00 0 1
0 0 10 0 0 1
L
a
yl l l
ulS U U S
ul
Ru
α α
α α
−
= =
(n)
On simplifying equation (n), we get
( )( )
( )
.
. .
( )
x
yz
y aL
y lu lu l u l y
u u lu l
M u lu l
S Ru
α α α
ϕ α α α
+ + −
+ += +
1 12 2
21 31 41 06 6
21 31 41
31 41
2 41 2
0 5 0 0
00 5 0 0 0 5
00 0
0 0 1
(o)
On expanding equation (o), we have
( )2 2 21 2 31 2 2 41 0 2 20.5 ( ) 0.5R L au u l u y l Rϕ ϕ α α α= = + + − +
2 2 31 2 41 0 20 ( )( )R L aM M u l u y l R= = = + − +
(p)
From equation (p) the first equation, we have
(q)
( )1 12 2
2 2 2 21 2 2 31 2 2 41 0 2 26 60 0.5 ( )R L ay y l u l u l u y l Rα α α= = = + + − +
2 2 41 0( )R L b a
S S R u y R= = = − +
( )1 2
0 2 21 2 2 31 2 2 416
1 2
2 26
0.5a
y l u l u l uR
l
α α
α
+ +=
478
On substituting equation (q) in the third equation of set of equations (p), and noting that since ,
we get
( )1 2
2 2 21 2 2 31 2 2 416
31 2 41 1 2
2 26
0.5( ) 0
l l u l u l uu l u
l
α α
α
+ +− + + = (r)
On simplification of equation (r), we get
1
21 2 3130u uα+ = (s)
which is the frequency equation. On substituting variables defined in equation (l) into equation (s), we
have
32 1122 1121 2 31
12 12
10
3
u uu uu u
u uα
− + − =
(t)
In view of variables defined in equations (g) and (h), the frequency equation (t) becomes
( )( )
3 32 2 2 21 1 1
22 21 2 2
12 22 21 1
1 1
1 1 16 6
02 3 3
2 2
nf nf d nf d nf
nf nf
nf d nf d
l l lm I I m
EI EI EIl l lm m l
EI EI EIl ll I l I
EI EI
ω ω ω ω
ω ω
ω ω
+ − − +
− + − =
− −
which can be simplified as
( ) ( )3 4 3 2 2 2
1 1 2 1 1 2 1 23 4 6 2 6 2 2 36( ) 0d nf d d nfmI l l l EI ml I l I l ml l EIω ω+ − + + + + = (u)
For the present problem, we have
; d = 0.01 m
On substituting in equation (u), we get
00y ≠
4; 1 20.02 m 0.3 m; 0.7 m; 5 kgdI l l m= = = =
4 10 4 6 2(0.01) 4.909 10 m ; 1.031 10 Nm
64I EI
π −= = × = ×
479
4 4 2 75.97 10 3.83 10 0nf nfω ω− × + × =
which gives
125.47nfω = rad/sec and
2nfω = 243.00 rad/sec
The mode shapes can be obtained by using transfer matrices between various intermediate stations
derived in equations (q), (p), (o), (m), (b) and (a). For a reference y0 = 1 may be chosen. This is left to
readers as an exercise (Fig. 8.34).
Alternative method for the above example is explained now. A transformation matrix to transform the
state vector from the left of a support to the right of the support can be developed as follows (refer
equation (d)):
1 1
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
0 0 0 1
1 0 0 0 0 1 1
A
R L
y y
M M
S R S
ϕ ϕ
− − =
or { } { }* * *
1 1AspR LS U S = (v)
where now all vectors and matrices are modified to accommodate the reaction force from the support.
Modifying all other transformation in equations (a) and (b), we get the modified overall
transformation as
{ } { } { }* * * * * * *
2 12 0 0spLS U U U S T S = = (w)
which can be expanded as
11 12 13 14 15
21 22 23 24 25
31 32 33 34 35
41 42 43 44
2 01 0 0 0 0 1 1
A
A
A
A
L
y t t t t t R y
t t t t t R
M t t t t t R M
S t t t t R S
ϕ ϕ
− − =
(x)
Boundary conditions given in equations (c) and (e), can be applied in equation (x) to get
480
11 12 13 14 15
21 22 23 24 25
31 32 33 34 35
41 42 43 44
2 0
0
0 0
0
1 0 0 0 0 1 1
A
A
A
A
L
t t t t t R y
t t t t t R
t t t t t R
S t t t t R
ϕ ϕ
− =
(y)
which gives
11 12 15
31 32 35
0
0A
t t tyR
t t tϕ
− = +
and
2521 22
4541 422 0
A
L
tt t yR
tt tS
ϕ
ϕ
− = +
(z)
The first of equation (z) can be rearranged as
11 12 15
31 32 35
0
0A
yt t t
t t tR
ϕ
−
=
(a1)
or
2 2
11 31 11 31 11 12 31 32 11 15 31 35 0
11 12 15 2 2
12 32 11 12 31 32 12 32 15 12 35 32 0
31 32 35 2 2
15 35 11 15 31 35 15 12 35 32 15 35
0
0A A
t t y t t t t t t t t t t yt t t
t t t t t t t t t t t tt t t
t t R t t t t t t t t t t R
ϕ ϕ
− + + + − = = + + + + + +
(b1)
For the non-trial solution of equation (b1), we should have
2 2
11 31 11 12 31 32 11 15 31 35
2 2
11 12 31 32 12 32 15 12 35 32
2 2
11 15 31 35 15 12 35 32 15 35
0
t t t t t t t t t t
t t t t t t t t t t
t t t t t t t t t t
+ + +
+ + + =
+ + +
It can be verified that equations (b1) and (u) will be identical. It should be noted that the
transformation between the right and left side of the support B can be written as
2 2
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
0 0 0 1
1 0 0 0 0 1 1
B
R L
y y
M M
S R S
ϕ ϕ
− − =
or { } { }* * *
2 2BspR LS U S = (a1)
which will be used for getting the state vector in the right of support B.
481
Example 8.9 Obtain the variation of the transverse natural frequency with the shaft speed (i.e.,
obtained the Campbell diagram) of an overhang rotor system as shown in Figure 8.35. From such
Campbell diagram obtain critical speeds. The end B1 of the shaft is having fixed end conditions.
Length of the shaft is 0.2 m and diameter is 0.01 m. The disc is thin and has 1 kg of mass and the
radius of the disc is 3.0 cm. Consider gyroscopic effects, however, neglect the mass of the shaft. Take
the range of the shaft speed such that it covers at least two critical speeds in the Campbell diagram.
Use the TMM.
Figure 8.35
Solution: Because of gyroscopic effect now the motion in the two orthogonal planes will be coupled.
This requires considering the modified field and point transfer matrices as given by equations (8.34)
and (8.45). In additional, the gyroscopic effect will introduce the following new terms in the point
matrix: * * * *
3,14 7,10 11,6 15,2 pP P P P I ων= − = − = = , where subscripts represent the row and column,
respectively, in the modified point matrix and ω is the spin speed of the shaft. Let station 0 be the
fixed end and station 1 be the free end. For free vibrations, then the overall transformation can be
written as
{ } { }* * *
1 0RS T S = (a)
with
* * *
1 1T P F = (b)
sym
skew
sym
skew
[ ] 0 0 [ ] 0
0 [ ] [ ] 0 0
[ ] ;0 [ ] [ ] 0 0
[ ] 0 0 [ ] 0
0 0 0 0 1
i
P G
P G
P G P
G P
∗
=
1
[ ] 0 0 0 0
0 [ ] 0 0 0
[ ] 0 0 [ ] 0 0
0 0 0 [ ] 0
0 0 0 0 1i
F
F
F F
F
∗
=
(c)
2
2
1 0 0 0
0 1 0 0[ ]
0 1 0
0 0 1
d
PI
m
ν
ν
= −
2 3
2
12 6
0 1 [ ] 2
0 0 1
0 0 0 1
l llEI EI
l lF EI EI
l
=
(d)
482
0 0 0 0
0 0 0 0[ ]
0 0 0
0 0 0 0
sym
p
GI ων
=
,
0 0 0 0
0 0 0 0[ ]
0 0 0
0 0 0 0
skew
p
GI ων
= −
(e)
{ }
{ }
{ } ;{ }
{ }
1
r
j
r
j
h
h
v
v
S
S
S S
S
∗
=
{ } ;y
h
xz
x
x
SM
S
ϕ
−
=
{ } ;x
v
yz
y
y
SM
S
ϕ
−
=
0
0{ }
0
0
u
=
(f)
sym
skew
* * *
sym
skew
1
[ ][ ] 0 0 [ ] [ ] 0
0 [ ][ ] [ ] [ ] 0 0
0 [ ] [ ] [ ][ ] 0 0
[ ] [ ] 0 0 [ ][ ] 0
0 0 0 0 1
P F G F
P F G F
T P F G F P F
G F P F
= =
(g)
( )
2 3
2 3
2
2
2 2 22 2
2
2 2 2 32 2
12 6
11 0 0 0 2 60 1
20 1 0 0 0 1[ ][ ] 20 1 0 0 1
0 0 1 20 0 1
0 0 0 11
2 6
d dd d
l llEI EIl ll
EI EI l lEI EIl l
P F EI EI I l I lI I ll EI EI
m
m l m lm m lEI EI
ν νν ν
νν νν ν
= = − − − − +
(h)
2 3
2
2sym
0 0 0 010 0 0 0 2 60 0 0 0
0 0 0 0 0 1[ ] [ ] 20 0 0 00 0 1 20 0 0 0
0 0 0 00 0 0 1
p pp p
l llEI EI
l lG F EI EI I l I lI Il EI EI
ων ωνων ων
= =
(i)
2 3
2
2skew
0 0 0 010 0 0 0 2 60 0 0 0
0 0 0 0 0 1[ ] [ ] 20 0 0 00 0 1 20 0 0 0
0 0 0 00 0 0 1
p pp p
l llEI EI
l lG F EI EI I l I lI Il EI EI
ων ωνων ων
= = − − − −
(j)
483
Boundary conditions for the present case are that all the linear and angular displacements at station 0
are zero and all the moments and shear forces are zero at right of station 1. Hence, the state vector at
station 0 and 1 have the following form
{ } { }*
00 0 0 0 0 0 0 0 1
r r i i r r i i
T
xz x xz x yz y yz yS M S M S M S M S=
and
{ } { }*
10 0 0 0 0 0 0 0 1
r r r i
T
r y i y r x i xRS x x y yϕ ϕ ϕ ϕ= − − − −
Following rows: 3, 4, 7, 8, 11, 12, 15, 16 will give the eigen value problem of the following form
484
( )
( )
22 2 2
2 2 2 3
22 2 2
2 2 2 3
2 2 2 2
1 0 0 0 02 2
1 0 0 0 0 0 02 6
0 0 1 0 02 2
0 0 1 0 0 0 02 6
0 0 12 2
p pd d
p pd d
p p d d
I l I lI l I ll
EI EI EI EI
m l m lEI EI
I l I lI l I ll
EI EI EI EI
m l m lEI EI
I l I l I l I ll
EI EI EI EI
ων ωνν ν
ν ν
ων ωνν ν
ν ν
ων ων ν ν
− −
+
− − − −
+
− −
( )
( )
2 2 2 3
2 2 2 2
2 2 2 3
0 0
0 0 0 0 1 0 02 6
0 0 0 0 12 2
0 0 0 0 0 0 12 6
r
r
i
i
r
r
i
i
xz
xz
xz
xz
yz
yz
yz
yzp p d d
M
S
M
S
M
S
m l m l MEI EI
SI l I l I l I l
lEI EI EI EI
m l m lEI EI
ν ν
ων ων ν ν
ν ν
+
− − − −
+
0
0
0
0
0
0
0
0
=
(k)
We need to find ν for which the determinant of the above matrix is zero. It should be noted that the above matrix is function of spin speed of the shaft, hence
these solutions have to be obtained for a particular speed at a time. For different operating speed the solutions will help in plotting the Campbell diagram (Fig.
8.36).
485
8.3 Dunkerley’s Formula
Dunkerley’s formula can be used to calculate the fundamental transverse natural frequency without
the help of the numerical methods. This method gives very rough estimation of natural frequency.
From the influence coefficient method, the natural frequency of the system is obtained by the
following conditions
0A = (8.66)
For two degrees of freedom system by considering only the linear displacements, the above equation
will be of the following form
1 11 2 122
1 21 2 22 2
1
01
nf
nf
m m
m m
α αω
α αω
−
=
−
1 11 2 22 1 2 12 212 2
1 10
nf nf
m m m mα α α αω ω
⇒ − − − =
which can be rearranged as
( )2
1 11 2 22 1 2 12 212 2
1 10
nf nf
m m m mα α α αω ω
− + − =
(8.67)
but for a polynomial whose first coefficient is unity, the second coefficient is equal to minus of the
sum of the roots of the equation (Scarborough, 1966)
( )1 2
1 11 2 222 2
1 1
nf nf
m mα αω ω
+ = + (8.68)
Now define
1 112
11
1m α
ω= and
2 222
22
1m α
ω=
(8.69)
where ω11 and ω22 are natural frequencies of single-DOF system, respectively, when mass m1 alone
is present and when mass m2 alone is present. Hence, we have
486
1 2
2 2 2 2
11 22
1 1 1 1
nf nfω ω ω ω+ = + (8.70)
It can be proved that in general for multi-DOF systems, we can write
1 2
2 2 2 2 2 2
11 22
1 1 1 1 1 1
nf nf nfN NNω ω ω ω ω ω+ + + = + + +� �
(8.71)
where N is the total DOFs of the rotor system. In most cases the fundamental frequency 1nf
ω will be
much lower than the all other higher natural frequencies, so the above equation may be written as
1
2 21
1 1N
inf ii
εω ω=
+ =∑
(8.72)
where ε is a small positive quantity. The above equation can be rearranged as
1
2
21
1
1nf N
i ii
ω
εω=
=
− +
∑
(8.73)
On neglecting ε, we can find the fundamental natural frequency of the rotor system, as
1
2
21
1
1nf N
i ii
ω
ω=
= ∑
(8.74)
It should be noted that since ε is a positive quantity, the we will get a lower bound of the fundamental
frequency from the present method as opposed to FEM in which we always get higher bounds of
natural frequencies.
1
2 2 2 2
11 22
1 1 1 1
nf NNω ω ω ω≅ + + +� (8.75)
Dunkerley first suggested this. Equation (8.75) always gives a value for fundamental frequency,
which is slightly lower than the true value, by virtue of the approximation involved. There are other
487
approximate methods are available that can be used to obtain bounds of natural frequencies of the
rotor system for example the Rayleigh's quotient (Meirovitch, 1986).
Example 8.10 Find transverse natural frequencies and mode shapes of a rotor system shown in Figure
8.37. B is a fixed end, and D1 and D2 are rigid discs. The shaft is made of the steel with the Young’s
modulus E = 2.1×1011
N/m2 and a uniform diameter d = 10 mm. Shaft lengths are: BD2 = 50 mm, and
D1D2 = 75 mm. The mass of discs are: m1 = 2 kg and m2 = 5 kg. Consider the shaft as massless and
neglect the diametral mass moment of inertia of discs.
Figure 8.37
Solution: For Figure 8.37, we have
N-m2, l1 =0.125 m, l2 =0.05 m,
11 3 3
1
3 3 103.1
0.125
EIk
l
×= = = 0.158×10
6 N/m,
22 3 3
2
3 3 103.1
0.05
EIk
l
×= = = 2.47×10
6 N/m,
62 61111
1
0.158 100.079 10
2
k
mω
×= = = × ,
62 62222
2
2.47 100.494 10
5
k
mω
×= = = ×
(a) Case 1 (b) Case 2
Figure 8.38 Overhang rotor systems with a single disc
The system natural frequency from the Dunkerley’s formula is given as
103.1EI =
488
1
6 6
2 2 2
11 22
1 1 1 1 110 14.68 10
0.079 0.494nfω ω ω− −
= + = + × = ×
or 1nf
ω = 260.10 rad/sec
Using the TMM the value of the fundamental natural frequency was 1
266.67nf
ω = rad/sec. Hence, the
Dunkerley’s formula estimates reasonably good estimate of the fundamental natural frequency, and it
gives the lower bound.
Example 8.11 Find fundamental transverse natural frequency of the rotor system shown in Figure
8.38. B1 and B2 are bearings, which provide simply supported end condition and D1, D2, D3 and D4 are
rigid discs. The shaft is made of the steel with the Young’s modulus E = 2.1 (10)11
N/m2 and uniform
diameter d = 20 mm. Various shaft lengths are as follows: B1D1 = 150 mm, D1D2 = 50 mm, D2D3 = 50
mm, D3D4 = 50 mm and D4B2 = 150 mm. The mass of discs are: m1 = 4 kg, m2 = 5 kg, m3 = 6 kg and
m4 = 7 kg. Consider the shaft as massless. Consider the discs as point masses.
Figure 8.38 A multi-disc rotor system
Solution: The influence coefficient for a simply support shaft with a disc is given as
2 2
3
a b
EIlα = with l = a + b = 0.45 m (a)
where l is the span of the shaft, and a and b are the disc position from the left and right bearings. The
natural frequency of a single-DOF rotor system can be obtained as
2 1
i i
ii
i y fmω
α= , 1,2,3,4i = (b)
We have d = 0.02 m, l = 0.45 m, EI = 1649.34 N-m2. Hence, EIl
3 =450.89 N-m
5. Table 8.5
summarises the calculation of the fundamental natural frequency.
489
Table 8.5 Calculation procedure of the fundamental natural frequency using the Dunkerley’s formula
S.N. ai
(m)
bi
(m)
Influence
coefficient,
αii (m/N)
Mass of
the disc,
mi, (kg)
For a single-DOF,
21 /i iii i y fmω α=
The fundamental
natural frequency,
( )1
21 / 1 /nf iiω ω= ∑ ,
(rad/sec)
1 0.150 0.300 4.49×10-6
4 1.80×10-5
95.18
2 0.200 0.250 5.54×10-6
5 2.77×10-5
3 0.250 0.200 5.54×10-6
6 3.32×10-5
4 0.300 0.150 4.49×10-6
7 3.14×10-5
Σ21/ iiω =10.93×10
-5
Exercise 8.12 Find transverse natural frequencies and mode shapes of the rotor system shown in
Figure 8.39. B is a fixed bearing, which provide fixed support end condition; and D1, D2, D3 and D4
are rigid discs. The shaft is made of the steel with the modulus of rigidity E = 2.1 (10)11
N/m2 and the
uniform diameter d = 20 mm. Various shaft lengths are as follows: D1D2 = 50 mm, D2D3 = 50 mm,
D3D4 = 50 mm and D4B2 = 150 mm. The mass of discs are: m1 = 4 kg, m2 = 5 kg, m3 = 6 kg and m4 = 7
kg. Consider the shaft as massless. Consider the disc as point masses, i.e., neglect the diametral and
polar mass moment of inertia of all discs.
Figure 8.39 A multi-disc overhung rotor
Solution: The influence coefficient for a cantilever shaft with a disc at free end is given as
3
3yf
L
EIα = (a)
where L is the span of the shaft. The natural frequency of a single-DOF rotor system can be obtained
as
2 1
i i
ii
i y fmω
α= , 1,2,3,4i = (b)
490
We have d = 0.02 m, and EI = 1649.34 N-m2. Table 8.6 summarises the calculation of the
fundamental natural frequency.
Table 8.6 Calculation procedure of the fundamental natural frequency using the Dunkerley’s formula
S.N. Li
(m)
Influence
coefficient,
αii (m/N)
Mass of
the disc,
mi, (kg)
For a single-DOF,
21 /
i iii i y fmω α=
The fundamental
natural frequency,
( )1
21 / 1 /nf iiω ω= ∑
, (rad/sec)
1 0.30 5.46×10-6
4 2.183×10-5
138.55
2 0.25 3.16×10-6
5 1.579×10-5
3 0.20 1.62×10-6
6 9.701×10-6
4 0.15 6.82×10-7
7 4.774×10-6
Σ21/ iiω = 5.209×10
-5
Concluding Remarks:
To summarise, in the present chapter we studied methods of calculation of natural frequencies and
forced responses. Two main methods namely, the influence coefficients and transfer matrix methods,
are given detailed treatment. The Durkerley’s formula for the approximate estimation of the
fundamental frequency is presented. All three methods are illustrated with simple examples keeping
calculation complexity to a minimum, while retaining various basic features of the solution method
for a variety of cases. The application of these methods for larger system is then straight forward;
however, it requires help of computer. The influence coefficient method is simple in application;
however, it requires calculations of influence coefficients with the help of load-deflection relations,
which are different for different systems. Moreover, with number of DOF of the system the matrix
size increases, so it requires higher computational time for large rotor systems to solve the eigen value
problem. The transfer matrix method is quite systematic and effective even for multi-DOF systems. It
does not require any a prior calculation of the system matrix element as in the influence coefficient
method. The overall size of the matrix remains the same, and it does not increase with the DOF of the
system. However, for the calculation of natural frequencies, this method requires roots searching
numerical method, which is time consuming and there is risk of missing some roots. The dynamic
matrix method (not described here) is similar to the TMM in that it relates to the state vectors at
different stations, however, while assembling the components equations the size of the matrix no
longer remains the same and it increases with the DOF of the system. In fact, the finite element
method is the improvised version of this method as we will see in the next chapter.
491
Appendix 8.1 Load deflection relations for various boundary conditions of the shaft
S.N
.
Boundary conditions
Angular
displacement
Linear displacement at any location of the
shaft
Maximum linear
displacement (δmax)
1
A cantilever shaft with a concentrated load P at
free end
2
2x l
Pl
EIϕ
== ( )
2
36
Pxy l x
EI= −
3
3x l
Ply
EI==
2
A cantilever shaft with a concentrated load P at
x = a
2
2x l
Pa
EIϕ
== ( )
2
36
Pxy l x
EI= − for 0 x a< <
( )2
36
Pay x a
EI= − for a x l< <
( )2
36x l
Pay l a
EI== −
3
A cantilever shaft with a uniformly distributed
load q(x) = q0 over the whole span
3
0
2x l
q l
EIϕ
== ( )
22 20 6 4
24
q xy x l lx
EI= + −
4
0
8x l
q ly
EI==
492
4
A cantilever shaft with a linearly varying
distributed load q(x) over the whole span
3
0
24x l
q l
EIϕ
== ( )
22 2 2 30 10 10 5
120
q xy l l x lx x
lEI= − + −
4
0
30x l
q ly
EI==
5
A cantilever shaft with a concentrated moment
M at the free-end
x l
Ml
EIϕ
==
2
2
Mxy
EI=
2
2x l
Mly
EI==
6
A simply supported shaft with a concentrated
load P at the mid-span
2
1 216
Pl
EIϕ ϕ= =
223
12 4
Px ly x
EI
= −
for 0
2
lx< <
3
0.5 48x l
Ply
EI==
493
7
A simply supported shaft with a concentrated
load P at any point
( )2 2
16
Pb l b
lEIϕ
−=
( )1
2
6
Pab l b
lEIϕ
−=
( ) [ ]2 2 2 for 0
6
Pbxy l x b x a
lEI= − − < <
( ) ( )
[ ]
3 2 2 3
6
for
Pb ly x a l b x x
lEI b
a x l
= − + − −
< <
( )2 2
3/22 2
3 9 3l b
x
Pb l by
lEI−
=
−=
and at center, if a b>
( )2 2
0.53 4
48x l
Pby l b
EI== −
8
A simply supported shaft with a uniformly
distributed load q(x) = q0 over the whole span
3
1 224
ql
EIϕ ϕ= = ( )3 2 3
224
qxy l lx x
EI= − +
4
0.5
5
384x l
qly
EI==
9
A simply supported shaft with a concentrated
moment M at one end point
16
Ml
EIϕ =
23
Ml
EIϕ =
2
21
6
Mlx xy
EI l
= −
2
/ 3
5
9 3x l
Mly
EI=
=
and at centre
2
0.5 16x l
Mly
EI==
494
10
3
01
7
360
q l
EIϕ =
3
02
45
q l
EIϕ =
( )4 2 2 40 7 10 3360
q xy l l x x
lEI= − +
4
0
0.5190.00652
x l
q ly
EI==
and at center 4
0
0.50.00651
x l
q ly
EI==
495
Exercise Problems
Use both the influence coefficient and transfer matrix methods for the entire problem unless otherwise
stated in the problem. The Durkerley’s formula could be used to rough estimate of the fundamental
natural frequency.
Exercise 8.1 Obtain the transverse natural frequencies of a rotor as shown in Figure E8.1. The rotor is
assumed to be fixed supported at one end and free at the other. Take mass of the disc m = 2 kg and its
diametral mass moment of inertia kg-m2. The shaft is assumed to be massless, and its
length and diameter are 0.2 m and 0.01 m, respectively. Take the Young’s modulus
N/m2 for the shaft material.
Figure E8.1
Exercise 8.2 Find the transverse natural frequencies and mode shapes of the rotor system shown in
Figure E8.2 by the influence coefficient method. Take EI = 2 MNm2 for the shaft and the diametral
mass moment of inertia of the disc is negligible.
Figure E8.2
Exercise 8.3 For Exercise 8.2 when the left and right discs have respectively diametral mass moment
of inertias as kg-m2 and kg-m
2, obtain the transverse natural frequencies and
mode shapes of the rotor system. [Hint: Use the eigen value formulation for the present case]
Exercise 8.4 Obtain the transverse natural frequency of a rotor system as shown in Figure E8.4. The
mass of the disc m, is 5 kg and the diametral mass moment of inertia, Id, is 0.02 kg-m2. Shaft lengths
are a = 0.3 m and b = 0.7 m. The diameter of the shaft is 10 mm.
0.05d
I =
112.1 10E = ×
.dI =1
0 05 .dI =2
0 06
496
Figure E8.4 An overhang rotor system
Exercise 8.5 Obtain transverse natural frequencies of a rotor system as shown in Figure E8.5. The
mass of the disc is kg and the diametral mass moment of inertia is Id = 0.02 kg-m2. Shaft
lengths are a = 0.3 m and b = 0.7 m, the diameter of the shaft is 10 mm and the modulus of elasticity
of the shaft is E = 2.1 × 1011
N/m2. Consider two different cases, i.e. when bearing A is (i) a simple
support and (ii) a flexible support, which provides a bending stiffness equal two times the bending
stiffness of a cantilevered shaft having length a. Bearing B is a fixed bearing.
Figure E8.5 An overhung rotor system
Exercise 8.6 Find the transverse natural frequencies and mode shapes of the rotor system shown in
Figure E8.6. B1 and B2 are bearings, which provide simply supported end condition, and D1 and D2
are rigid discs. The shaft is made of steel with the Young’s modulus E = 2.1 (10)11
N/m2 and uniform
diameter d = 10 mm. Various shaft lengths are as follows: B1D1 = 50 mm, D1D2 = 75 mm, and D2B2 =
50 mm. The mass of discs are: m1 = 4 kg and m2 = 6 kg. Consider the shaft as massless. Consider the
following cases (i) neglect the diametral mass moment of inertia of both discs and (ii) take
kg-m2 and kg-m
2. [Hint: Use the eigen value formulation for the second case]
Figure E8.6
5m =
.dI =1
0 05
.dI =2
0 06
497
Exercise 8.7 Find transverse natural frequencies and mode shapes of the rotor system shown in Figure
E8.7. B1 and B2 are bearings, which provide simply supported end condition and D1, D2, D3, D4 and
D5 are rigid discs. The shaft is made of the steel with the Young’s modulus E = 2.1 (10)11
N/m2 and
uniform diameter d = 20 mm. Various shaft lengths are as follows: B1D1 = 150 mm, D1D2 = 50 mm,
D2D3 = 50 mm, D3D4 = 50 mm, D4D5 = 50 mm, and D5B2 = 150 mm. The mass of discs are: m1 = 4
kg, m2 = 5 kg, m3 = 6 kg, m4 = 7 kg, and m5 = 8 kg. Consider the shaft as massless. Consider the
following cases (i) consider the disc as point masses, i.e., neglect the diametral and polar mass
moment of inertia of all discs, and (ii) consider discs as thin and take diameter of discs as d =1 8 cm,
d =2 10 cm, d =3 12 cm, d =4 14 cm, and d =5 16 however, neglect the gyroscopic effects.
Figure E8.7 A multi-disc rotor system with simply supported end conditions
For a Jeffcott rotor with an off-set disc, the following influence coefficients are valid:
11 12
21 22
x
y zx
x f
M
α α
ϕ α α
=
with
2 3 22 2
11 12
2 221 22
3 2
3 3
3 3( )
3 3
a l a ala b
EIl EIl
al a lab b a
EIl EIl
α α
α α
− − −
=
− − −−
where l is the span length, and a and b are the distance of the disc from left and right supports,
respectively.
Exercise 8.8 Find all the transverse natural frequencies and draw corresponding mode shapes of the
rotor system shown in Figure E8.8. B1 is fixed support (i.e., with the transverse linear and angular
(slope) displacements equal to zero), and B2 and B3 are bearings with simply support condition (i.e.,
with the transverse linear displacement equal to zero). Shaft segments have the following dimensions:
B1D1 = 50 mm, D1B2 = 50 mm, B2D2 = 25 mm, D2B3 = 25 mm, and B3D3 = 30 mm. The shaft is made
498
of steel with Young’s modulus E = 2.1×1011
N/m2. The mass of discs are: m1 = 1 kg, m2 = 1.5 kg and
m3 = 0.75 kg. Compare the order of magnitude of the torsional and the bending natural frequencies so
obtained for the same system. Consider two cases (i) the shaft as massless and discs as rigid lumped
masses. (ii) consider discs as thin and take diameter of discs as d =1 12 cm, d =2 6 cm, and
d =3 12
cm, however, neglect the gyroscopic effects.
Figure E8.8
Exercise 8.9 Find the unbalance response and critical speeds of a rotor system shown in Figure E8.9.
B is a bearing with fixed end conditions, and D1 and D2 are rigid discs. The shaft is made of steel with
the Young’s modulus E = 2.1(10)11
N/m2 and a uniform diameter d =10 mm. Various shaft lengths are
as follows: BD2 = 50 mm, and D2D1 = 75 mm. The diametral mass moments of inertia of discs are:
= 0.04 kg-m2 and = 0.1 kg-m
2. Take an unbalance of 2 gm at a radius of 5 cm at a convenient
location and orientation. Neglect the mass of the shaft.
Figure E8.9 A two-disc overhang rotor
Exercise 8.10 Obtain the transverse natural frequencies of an overhung rotor system as shown in
Figure E8.10. The end B1 of the shaft is having fixed end conditions and the other end is free. The
length of the shaft is 0.4 m and the diameter is 0.1 m. The disc is thin and has 1 kg of mass, 0.04 kg-
m2 of the polar mass moment of inertia, and 0.02 kg-m
2 of diametral mass moment of inertia. Neglect
the mass of the shaft and consider the gyroscopic effects. Take the shaft speed of 10,000 rpm. Obtain
the Campbell diagram and show critical on that.
Figure E8.10 A overhung rotor
1dI
2dI
499
Exercise 8.11 Obtain transverse natural frequencies of the rotor system shown in Figure E8.11 for
following parameters: (i) discs are point masees with m1 = 5 kg, m2 = 8 kg,, l = 1 m, d = 0.02 m, ρ =
7800 kg/m3 and E = 2.1×10
11 N/m
2, and (ii) for the case disc is thin with radii of 10 cm and 15 cm
with other parameters as of case (i).
Figure E8.11 Two disc rotor system
Exercise 8.12 Find transverse natural frequencies and mode shapes of the rotor system a shown in
Figure E8.12. B1 and B2 are fixed supports, and D1 and D2 are rigid discs. The shaft is made of the
steel with the Young’s modulus of E = 2.1 (10)11
N/m2, and has uniform diameter of d = 10 mm.
Different shaft lengths are as follows: B1D1 = 50 mm, D1D2 = 75 mm, and D2B2 = 50 mm. Mass of
thin discs are: m1 = 2 kg and m2 = 3 kg and radius are: r1 = 5 cm and r2 = 8 cm. Consider the shaft as
massless.
Figure E8.12
Exercise 8.13 Find the fundamental transverse natural frequency of the rotor system shown in Figure
E8.13 by using the Durkerley’s formula. Take EI = 2 MN-m2 for the shaft, and the mass moment of
inertia of the disc is negligible. [Answer: 1nf
ω = 49.01 rad/sec]
Figure E8.13 An overhang rotor system
B1 B2
D1 D2
500
Exercise 8.14 Obtain transverse natural frequencies of a rotor system as shown in Figure E8.14. The
mass of discs are m1 = 5 kg, and m2 = 8 kg. Shaft length is such that l = 0.3 m, the diameter of the
shaft is 15 mm, and the modulus of elasticity of the shaft is E = 2.1 × 1011
N/m2. Both the bearings are
roller supports. [Hint: 3
118
l
EIα = ,
3
2248
l
EIα = and
3
12 2132
l
EIα α= = ].
Figure E8.14 An overhung rotor system
Exercise 8.15 Obtain transverse natural frequencies of a rotor system as shown in Figure E8.15. The
mass of discs are m1 = 5 kg, and m2 = 8 kg, and radius are r1 = 5 cm and r2 = 8 cm. Shaft length is
such that l = 0.3 m, the diameter of the shaft is 15 mm, and the modulus of elasticity of the shaft is E =
2.1 × 1011
N/m2. Both the bearings are roller supports. [Hint: Need to consider the diametral mass
moment of inertia].
Figure E8.15 An overhung rotor system
Exercise 8.16 Find transverse natural frequencies and mode shapes of the rotor system shown in
Figure E8.16. B is a fixed bearing, which provide fixed support end condition; and D1, D2, D3, D4 and
D5 are rigid discs. The shaft is made of the steel with the Young’s modulus E = 2.1 (10)11
N/m2 and
the uniform diameter d = 20 mm. Various shaft lengths are as follows: D1D2 = 50 mm, D2D3 = 50
mm, D3D4 = 50 mm, D4D5 = 50 mm, and D5B2 = 50 mm. The mass of discs are: m1 = 4 kg, m2 = 5 kg,
m3 = 6 kg, m4 = 7 kg, and m4 = 8 kg. Consider the shaft as massless. Two cases to be considered (i)
Consider the disc as point masses, i.e., neglect the diametral and polar mass moment of inertia of all
501
discs; (ii) consider discs as thin and take diameter of discs as d =1 12 cm, d =2 6 cm, and d =3 12 cm,
d =4 14 cm, and d =5 16 , however, neglect the gyroscopic effects.
Figure E8.16 A multi-disc overhung rotor
Exercise 8.17 Consider a rotor system as shown in Figure E8.17 for the transverse natural frequency.
Two flexible massless shafts are connected by a coupling (i.e., a pin joined). A thin disc of mass 3 kg
is attached to one of the shaft (let us take toward the left side shaft) and it is not interfering the relative
motion between the two shafts. Other ends of shafts have fixed conditions. Take the length of each of
the shaft as 0.5 m and the diameter as 0.05m. Young’s modulus E = 2.1 (10)11
N/m2. Use TMM.
Figure E8.17
Exercise 8.18 Consider a rotor system as shown in Figure E8.18 for the transverse natural frequency.
Two flexible massless shafts are connected by a coupling (i.e., a pin joined). A thin disc of mass 3 kg
is attached to one of the shaft (let us take toward the left side shaft) and it is not interfering the relative
motion between the two shafts. Other ends of shafts have fixed conditions. Take the length of the
shaft as 0.6 m (left side), 0.4 m (right side), and the diameter as 0.05m. Young’s modulus E = 2.1
(10)11
N/m2. Use the influence coefficient method and the TMM.
Figure E8.18
502
Exercise 8.19 Obtain the transverse natural frequency of co-axial shafts rotor system which is
modelled as shown in Fig. E8.19. Discs have the mass of 3 kg and 2 kg, respectively, on shaft A and
shaft B. The shaft A and B are respectively 2 cm and 1.5 cm diameters with a length of 40 cm each.
Neglect the inertia of shafts. The bearing between two co-axial shafts provides an effective transverse
stiffness of 100 MN/m. Take E = 2.1×1011
N/m2. Use TMM.
Figure E8.19
Exercise 8.20 Objective questions with multiple choice answer. Select a single answer only.
(i) The Durkerley’s formula gives lower bound of the fundamental frequency.
(a) True (b) False
(ii) There is an approximation involved in the formulation of the influence coefficient method for
multi-DOF rotor system.
(a) True (b) False
(iii) There is an approximation involved in the solution procedure by the influence coefficient method
for multi-DOF rotor system.
(a) True (b) False
(iv) There is an approximation involved in the solution procedure by the transfer matrix method for
multi-DOF rotor system.
(a) True (b) False
(v) There is an approximation involved in the formulation of the transfer matrix method for multi-
DOF rotor system.
(a) True (b) False
503
(vi) For a N-disc rotor with negligible moments of inertia the number of transverse natural frequency
would be
(a) N (b) 2N (c) 4N (d) N2
(vii) For a N-disc rotor with appreciable diametral moment of inertia, however, without gyroscopic
effects, the number of transverse natural frequency would be
(a) N (b) 2N (c) 4N (d) N2
(viii) For a N-disc rotor with appreciable diametral moment of inertia and with gyroscopic effects, the
number of transverse natural frequency would be
(a) N (b) 2N (c) 4N (d) N2
(ix) A point matrix relates
(a) State vectors at either sides of a disc (b) State vectors at either sides of a shaft
(c) State vectors from a shaft to another shaft (d) State vectors at a disc to another disc
(x) A field matrix relates
(b) State vectors at either sides of a disc (b) State vectors at either sides of a shaft
(c) State vectors from a shaft to another shaft (d) State vectors at a disc to another disc
(x) A transfer matrix relates
(a) State vectors from a shaft to another shaft (b) State vectors at a disc to another disc
(c) State vectors from one end of the rotor system to other end (d) either (a) or (b)
(xi) An overall transfer matrix relates
(a) State vectors from a shaft to another shaft (b) State vectors at a disc to another disc
(c) State vectors from one end of the rotor system to other end (d) either (a) or (b)
504
References:
Meirovitch, L., 1986, Elements of Vibration Analysis, McGraw Hill Book Co., NY.