Académique Documents
Professionnel Documents
Culture Documents
MACHINERY
B.TECH. DEGREE COURSE
SCHEME AND SYLLABUS
DYNAMICS OF MACHINERY
Module 1
Balancing: - Balancing of rotating masses, static balancing and dynamic balancing,
Balancing of several masses rotating in same plane, Balancing of several masses
rotating in several planes, Balancing machines.
Balancing of reciprocating masses: - The effect of inertia force of the reciprocating
mass on the engine. Partial primary balance. Partial balancing of locomotive,
Hammer blow, Variation of tractive effort, Swaying couple. Coupled locomotives,
Balancing of multi cylinder inline engines, v-engines, Radial engines, Direct and
Reverse cranks
Module 2
Vibrations: - Definitions, simple harmonic motion. Single degree freedom systems:
Undamped free vibrations: - Equations of motion Natural frequency, Energy
method, Equilibrium methods, Rayleigh’s methods, Equivalent stiffness of spring
combinations.
Damped free vibrations: - Viscous damping, Free vibrations with viscous damping,
over-damped system, critically damped system, under-damped system, Logarithmic
decrement, viscous dampers, coulomb damping.
Forced Vibrations: - Forced harmonic excitation Rotating unbalance, Reciprocating
unbalance. Energy dissipated by damping, vibration isolation and Transmissibility.
Vibration measuring instruments.
Module 3
Two degree freedom systems: - Principal modes of vibration, Rectilinear and
angular modes, systems with damping, vibration absorbers, centrifugal pendulum
damper, dry friction damper, untuned viscous damper.
Multi-degree of freedom system: - Free vibrations, equations of motion, Influence
coefficients method, lumped mass and distributed mass systems, Stodola method,
Dunkerly’s method, Holzer’s method, Matrix iteration method.
Torsional Vibrations: - Torsionally equivalent shaft, torsional vibration of two-
rotor, three-rotor, and geared systems.
Module 4
Critical speeds of shafts: - Critical speed of a light shaft having a single disc
without damping. Critical speeds of a light cantilever shaft with a large heavy disc at
its end.
Transient vibration: - Laplace transformation, response to an impulsive input,
response to a step input, response to a pulse input, phase plane method, shock
spectrum.
Non-linear vibrations: - Phase plane, undamped free vibration with non-linear
spring forces, hard spring, soft spring, Perturbation method, Forced vibration with
nonlinear forces, Duffings equation, self excited vibrations.
Module 5
Noise control: - Sound propagation, decibels, acceptance noise levels, Air columns,
Doppler effect, acoustic measurements, microphones and loud speakers, Recording
and reproduction of sound, fourier’s theorem and musical scale, Acoustics of
buildings, Acoustic impedence filters and human ear.
MODULE 1
Introduction
Static and dynamic unbalance: A rotor can in general have two types of unbalance
viz., “static” and “dynamic”. It is of course to be appreciated that practical systems
will all have dynamic unbalance only and considering it as static unbalance is a
“good-enough” approximation for some cases.
If the rotor is thin enough (longitudinally) as shown in Fig. 1.1 the unbalance force
can be assumed to be confined to one plane (the plane of the disc). Such a case is
known as “static” unbalance. Such a system when mounted on a knife-edge as shown
in Fig. 1 will always come to rest in one position only – where the centre of gravity
comes vertically below the knife-edge point. Thus in order to “balance out”, all we
need to do is to attach an appropriate “balancing mass” exactly 1800 opposite to this
position.
Fig 1.2 Thin Rotor on a knife edge - Illustration of Static Unbalance
Thus we first mount the disc on a knife edge and allow it to freely oscillate. Mark the
position when it comes to rest. Choose a radial location (1800 opposite to this
position) where we can conveniently attach a balancing mass. By trial and error the
balancing mass can be found out. When perfectly balanced, the disc will exhibit no
particular preferred position of rest. Also when the disc is driven to rotate by a motor
etc., there will be no centrifugal forces felt on the system (for example, at the
bearings). Thus the condition for static balance is simply that the effective centre of
gravity lie on the axis.
Consider the rotor shown in Fig. 1.3. It is easily observed that mass distribution
cannot be approximately confined to just one plane. So unbalance masses and hence
unbalance forces are in general present all along the length of the rotor. Such a case
is known as “dynamic unbalance”.
However, when mounted in bearings and driven by a motor etc., it could still wobble
due to the unbalanced moments of these forces as shown in Fig1.4. This becomes
apparent only when the rotor is driven to rotate and hence the name “dynamic
unbalance”. Thus it is not, in general, sufficient to do just static balance but
achieving good dynamic balance is more difficult. We will discuss one important
method of achieving dynamic balance in the next lecture.
Consider the turbo-machine rotor that was discussed earlier wherein each stage
contains several blades around the circumference of a disk. Eventhough typically
each stage is balanced in itself to the extent possible, it has a likely net unbalance.
When the rotor is set to spin, it will cause dynamic forces and moments on the
bearings that support the shaft. Therefore it is of interest to achieve “good balance”
of this shaft so that the fluctuating forces on the bearings are reduced. Conceptually
our strategy can be simply stated as follows:
Step 1: Consider the shaft supported on its bearings. For each unbalance mass, there
will be a centrifugal force set-up when the rotor spins at some speed . This would
cause some reactions at the supports. Estimate these support reactions that would
come onto the bearings.
Step 2: Estimate the balancing mass that needs to be placed in the plane of bearings,
to counter this reaction force due to unbalance mass.
Repeat steps 1 and 2 for each unbalance mass in the system and each time add the
balancing masses obtained in step 2 vectorially to determine the resultant balancing
mass required.
Let us now understand the details of the technique mentioned earlier. Firstly we
choose to place “balancing or correcting” masses on the shaft (rotating along with the
shaft) to counter-act the unbalance forces. We understand that this is to be done on
the rotor on site, perhaps during a maintenance period. From the point of view of
accessibility, we therefore choose the balancing masses to be kept near the bearings.
Now we need to repeat the calculations for ALL the unbalance masses mi (i =
1,2,3,…..) and find the resultant equivalent force in the balancing plane as shown in
blue in Fig. 2.3.1. This resultant force is balanced out by placing a suitable balancing
mass creating an equal and opposite force (shown in red).
Since all the masses are rotating at the same speed along with the shaft, we can
drop in our calculations – i.e., a rotor balanced at one speed will remain balanced
at all speeds or in other words, our technique of balancing is independent of speed.
We will review this towards the end of the lecture.
Sr. No
Cos( Sin( Cos( Sin(
) ) ) )
1
2
3
....
…..
Balancing 0
Plane 1
Balancing L
Plane 2
TOTAL FORCES 0 0 0 0
It is observed in Table 1 that the balancing masses and their locations (radial as well
as angular) are unknowns while the location of the balancing plane itself is treated as
a known (any accessible location near the bearings etc). The resultant total forces and
moments must sum up to ZERO and therefore we have four equations but six
unknowns. Thus any two of the six unknowns can be freely chosen and the other four
determined from the computations given in the table. This method of balancing is
known as the “two-plane balancing technique” since balancing masses are kept in
two planes.
We know that the inertia forces are given by mass times acceleration and we shall
now estimate the inertia forces (shaking forces and moments) due to the moving
parts on the frame (cylinder block).
CONNECTING ROD
One end of the connecting rod is circling while the other end is reciprocating and any
point in between moves in an ellipse. It is conceivable that we derive a general
expression for the acceleration of any point on the connecting rod and hence estimate
the inertia forces due to an elemental mass associated with that point. Integration
over the whole length of the connecting rod yields the total inertia force due to the
entire connecting rod. Instead we try to arrive at a simplified model of the connecting
rod by replacing it with a “dynamically equivalent link” as shown in Fig 1.7
In order that the two links are dynamically equivalent, it is necessary that:
Distribution of the mass be also same i.e., location of CG must be same and the
mass moment of inertia also must be same.
= (GB)/L
= (AG)/L
Thus the connecting rod is replaced by two masses at either end (pin joints A and B)
of the original rod. rotates along with the crank while purely translates along
with the piston. It is for this reason that we proposed use of crank's effective rotating
mass located at pin A, which can now be simply added up to part of connecting rod's
mass.
Based on our discussion thus far, we can arrive at a simplified model of the crank-
slider mechanism for the purpose of our dynamic analysis as shown in Fig. 1.8 Thus
we have either purely rotating masses or purely translating masses and these are
given by:
where the first term in the rotating masses is due to the effective crank mass at pin A
and the second term is due to the part of equivalent connecting rod mass located at
pin A. Similarly the first term in reciprocating masses is due to the mass of the piston
and the second is due to the part of equivalent connecting rod mass located at pin B.
There are inertia forces due to .
The locomotives, usually, have two cylinders with cranks placed at right
angles to each other in order to have uniformity in turning moment diagram. The two
cylinder locomotives may be classified as :
1. Inside cylinder locomotives ; and
In the inside cylinder locomotives, the two cylinders are placed in between
the planes of two driving wheels whereas in the outside cylinder locomotives, the
two cylinders are placed outside the driving wheels, one on each side of the driving
wheel. The locomotives may be
We have discussed in the previous article that the reciprocating parts are
only partially lanced. Due to this partial balancing of the reciprocating parts, there is
an unbalanced primary force along the line of stroke and also an unbalanced primary
force perpendicular to the line of stroke. The effect of an unbalanced primary force
along the line of stroke is to produce;
The resultant unbalanced force due to the two cylinders, along the line of
stroke, is known as tractive force. Let the crank for the first cylinder be inclined at an
angle with the line of stroke. Since the crank for the second cylinder is at right
angle to the first crank, therefore the angle of inclination for the second crank will be
(90° + ).
Let m = Mass of the reciprocating parts per cylinder, and
We know that unbalanced force along the line of stroke for cylinder 1
1 c m.2 .r cos
1 c m.2 .r cos 90o
As per definition, the tractive force,
1 c m.2 .r cos
1 c m.2 .r cos 90o
1 c m.2 .r cos sin
d
cos sin 0 or sin cos 0 or sin cos
d
1 c m.2 .r cos135o sin135o 2 1 c m.2.r
Swaying Couple
The unbalanced forces along the line of stroke for the two cylinders
constitute a couple.This couple has swaying effect about a vertical axis, and tends to
sway the engine alternate in clockwise and anticlockwise directions. Hence the
couple is known as swaying couple.
Swaying couple
a
1 c m.2 .r cos
2
1 c m.2 .r cos 90o a2
a
1 c m.2 .r cos sin
2
d
cos sin 0 or sin cos 0 or sin cos
d
1 c m.2 .r
a
2
cos 45o sin 45o
a
2
1 c m.2.r
We know that the unbalanced force along the perpendicular to the line of
stroke due to the balancing mass B, at a radius b, in order to balance reciprocating
parts only is B. 2 .b sin . This force will be maximum when sin is unity, i.e.
when = 90° or 270°.
The effect of hammer blow is to cause the variation in pressure between the
wheel and the rail.
Let P be the downward pressure on the rails (or static wheel load).
P B.2 .b
If P B. .b
2
is negative, then the wheel will be lifted from the rails.
Therefore the limiting condition in order that the wheel does not lift from the rails is
given by
P B.2 .b
P
Bb
It may be noted that the variation of pressure between the wheel and the rail
(i.e., hammer blow) may be reduced by equal distribution of balanced mass (B)
between the driving, leading and trailing wheels respectively.
The multi-cylinder engines with the cylinder centre lines in the same plane
and on the same side of the centre line of the crankshaft, are known as In-line
engines. The following two conditions must be satisfied in order to give the primary
balance of the reciprocating parts of a multi-cylinder engine:
2. The algebraic sum of the couples about any point in the plane of the
primary forces must be equal to zero. In other words, the primary
couple polygon must close.
We have already discussed, that the primary unbalanced force due to the
reciprocating masses is equal to the component, parallel to the line of stroke, of the
centrifugal force produced by the equal mass placed at the crankpin and revolving
with it. Therefore, in order to give the primary balance of the reciprocating parts of a
multi-cylinder engine, it is convenient to imagine the reciprocating masses to be
transferred to their respective crankpins and to treat the problem as one of revolving
masses.
Notes: 1. For a two cylinder engine with cranks 180o, condition (1) may be
satisfied, but this will result in an unbalanced couple. Thus the above method of
primary balancing cannot be applied in this case.
2. For a three cylinder engine with cranks at 120o and if the reciprocating
masses per cylinder are same, then condition (1) will be satisfied because the forces
may be represented by the sides of an equilateral triangle. However, by taking a
reference plan through one of the cylinder centre lines, two couples with non-parallel
axes will remain and these cannot vanish vectorially. Hence the above method of
balancing fails in this case also.
3. For a four cylinder engine, similar reasoning will show that complete
primary balance is possible and it follows that
When the connecting rod is not too long (i.e. when the obliquity of the
connecting rod is considered), then the secondary disturbing force due to the
reciprocating mass arises.
cos 2
FS m.2 .r
n
r
FS m. 2
2
cos 2
4n
As in case of primary forces, the secondary forces may be considered to be
equivalent to be component, parallel to the line of stroke, of the centrifugal force
produced by an equal mass placed at the imaginary crank of length r/4n and
revolving at twice the speed of the actual crank.Thus, in multi-cylinder in-line
engines, each imaginary secondary crank with a mass attached to the crankpin s
inclined to the line of stroke at twice the angle of the actual crank. The values of the
secondary forces and couples nay be obtained by considering the revolving mass.
This is done in the similar way as discussed for primary forces, the following two
conditions must be satisfied in order to give a complete secondary balance of an
engine :
1. The algebraic sum of the secondary forces must be equal to zero. In other
words, die secondary force polygon must close, and
2. The algebraic sum of the couples about any point in the plane of the
secondary forces must be equal to zero. In other words, the secondary
couple polygon must close.
It is assumed that m/2 is fixed at the direct crank (termed as primary direct
crank) pin and m/2 at the reverse crank (termed as primary reverse crank) pin . We
know that the centrifugal force acting on the primary direct and reverse crank
m 2
.r
2
m 2
.r cos
2
and, the component of the centrifugal force acting on the primary reverse crank
m
2 .r cos
2
m
2 2 .r cos m.2 .r cos = Primary force, FP
2
Hence, for primary effects, the mass m of the reciprocating parts may be
replaced by two masses each of magnitude m/2.
MODULE 2
Vibration (Oscillation)
Any motion which repeats itself after an interval of time is called vibration.
Causes of vibration
Effect of Vibration
Produces unwanted noise, high stresses, wear, poor reliability and premature
failure of one or more of the parts.
Elimination of Vibrations
Definitions
Frequency
Number of cycles per unit time.
Expressed in Hz or rad/sec.
Amplitude
The maximum displacement of a vibrating body from its equilibrium
position.
Resonance
When the frequency of external excitation is equal to the natural frequency
of a vibrating body, the amplitude of vibration becomes excessively large. This
concept is known as resonance.
Periodic Motion
A motion which repeats itself after equal intervals of time.
Time Period
Time taken to complete one cycle.
The fundamental mode of vibration of a system is the mode having the lowest natural
frequency.
Degree of Freedom
The minimum number of independent co-ordinates required to specify the
motion of system at any instant is known as degrees of freedom.
Eg: Springs supported Rigid mass. (It can move in the direction of springs
and also have angular motion in one plane)
After disturbing the system the external excitation is removed, then the
system vibrates on its own. This type of vibration is known as free vibrations.
3 types:-
a) Longitudinal vibrations.
When the particles of the shaft or disc move parallel to the axis of the shaft,
then the vibrations are known as longitudinal vibrations. In this case the shaft is
elongated and shortened alternately and thus the tensile and compressive stresses are
induced alternately in the shaft.
b) Transverse vibrations
When the particles of the shaft move approximately perpendicular to the
axis of the shaft, then the vibrations are known as transverse vibrations. In this case
the shaft is straight and bent alternately and bending stresses are induced in the shaft.
c) Torsional vibrations
When the particles of the shaft move in a circle about the axis of the shaft,
then the vibrations are known as torsional vibrations.
In this case the shaft is twisted and untwisted alternately and torsional shear
stresses are induced in the shaft to.
2. Forced Vibration
Eg: Machine tools, Electric bells.
The vibration which is under the influences of external force is called forced
vibration.
The external force applied to the body is periodic disturbing force created by
unbalance.
The vibrations have the same frequency as the applied force. Due to the
application of external forces the amplitude of these vibrations is maintained almost
constant.
3. Damped vibration
When there is a reduction in amplitude over every cycle of vibration, the
motion is said to be damped vibration.
That is if the vibrators system has a damper. The motion of the system will
be opposed by it and the energy of the system will be dissipated in friction.
4. Undamped vibration.
There is no damper. There is no loss of energy due to friction.
5. Deterministic vibration
If in the vibratory system the amount of external excitation is known in
magnitude it is deterministic vibration.
6. Random vibration
Non deterministic vibrations
8. Transient vibrations
In real systems, the amplitude of vibration decays continuously because of
damping and vanishes finally. Such vibration is real system is called transient
vibration.
9. Linear vibration
A vibratory system basically consists of there elements:
Mass
Spring
Pamper
Fig.
Linear vibration becomes non linear for very large amplitude of vibration.
a) Periodic
b) Impulsive
c) Random
Harmonic motion :
where A is the amplitude of motion, t is the time instant and T is the period of
motion.
where T is the period (sec) and f is the frequency (cycle/sec) of the harmonic motion.
FREE VIBRATIONS
A spring mass system as shown in Figure 2.2 is considered. For simplicity at present
the damping is not considered.
Figure 2.2
or
From Figure 2.2(b), we have (i.e. spring force due to static deflection is
equal to weight of the suspended mass), so the above equation becomes
The choice of the static equilibrium position as reference for x axis datum has
eliminated the force due to the gravity. Equation can be written as
or
where is the natural frequency (in rads/sec).This Equation satisfies the simple
harmonic motion condition.
The undamped free vibration executes the simple harmonic motion as shown in
Figure 2.3.
Since sine & cosine functions repeat after 2 radians (i.e. Frequency Time
period = 2 ), we have
Here T , f , are dependent upon mass & stiffness of the system, which are
properties of the system.
Above analysis is valid for all kind of SDOF system including beam or torsional
members. For torsional vibrations the mass may be replaced by the mass moment of
inertia and stiffness by stiffness of torsional spring. For stepped shaft an equivalent
stiffness can be taken or for distributed mass an equivalent lumped mass can be
taken.
For this system having springs connected in series or parallel, this equation is still
valid with the equivalent stiffness as shown in Figures 2.4 and 2.5.
Figure 2.4
Figure 2.5
Energy method :
In a conservative system (i.e. with no damping) the total energy is constant, and
differential equation of motion can also be established by the principle of
conservation of energy.
Our interest is to find natural frequency of the system, writing this equation for two
positions
Let 1 represents a static equilibrium position (choosing this as the reference point of
potential energy, here U1=0 ) and 2 represents the position corresponding to
maximum displacement of mass and at this position velocity of mass will be zero and
hence T2 = 0.
Generally mathematical model of such damping is quite complicated and not suitable
for vibration analysis.
(1)
(2)
From the condition that equation (2) is a solution for all values of t , above equation
gives a characteristic equation (Frequency equation) as
(3)
(4)
Hence the general solution of equation (1) from equations (2) and (4) is given by the
equation
(5)
where A and B are integration constants to be determined from initial conditions.
(6)
The term outside the bracket in RHS is an exponentially decaying function. The term
we can write
(7)
where
(8)
(9)
Response corresponding to the critical damping case is shown in Figure 2.9 for
various initial conditions.
(10)
(11)
with
Depending upon value of damping ratio we can have the following cases
, overdamped condition
, underdamped condition
, critical damping
, undamped system
(12)
(3.30)
(13)
x(0)=C
and
; which gives
(14)
Equation (14) indicates that the frequency of damped system is equal to,
(15)
It should be noted that for small ( which is the case of most engineering systems)
Two roots remain real with one increases and another decreases.
(16)
so that
x(0)=A + B
and
or
which gives
and
3) Critically damped systems :
(17)
(18)
so that,
x(0)=A
and
The necessary and sufficient conditions for crossing once can be obtained as
(19)
or
is a necessary condition for crossing time axis once but sufficient conditions
is given by equation (19) as shown in Figure 2.9.
Logarithmic Decrement :
since
Figure 2.10
The above illustration shows for two successive amplitude. But in case, the
amplitude are recorded after "n" cycles, the formula is modified as
Taking log,
The steady state response of the system can be determined by solving the above
equation in many different ways. Here a simpler graphical method is used which will
give physical understanding to this dynamic problem. From solution of differential
equations it is known that the steady state solution (particular integral) will be of the
form
As each term of equation (4.1) represents a forcing term viz., first, second and third
terms, represent the inertia force, spring force, and the damping forces. The term in
the right hand side of equation is the applied force. One may draw a close polygon as
shown in figure 2.12 considering the equilibrium of the system under the action of
these forces. Considering a reference line these forces can be presented as follows.
From equation the resultant of the spring force, damping force and the inertia force will
be the applied force, which is clearly shown in figure 2.12
It may be noted that till now, we don't know about the magnitude of X and which can
be easily computed from Figure 2. Drawing a line CD parallel to AB, from the triangle
OCD of Figure 2,
or
Figure 2.13 shows the magnification factor frequency ratio and phase angle
frequency ratio plot.
Following observation can be made from these plots.
system .
But for underdamped systems the maximum amplitude of excitation has a
Figure 2.13 : (a) Magnification factor ~ frequency ratio for different values of
damping ratio.
Figure 2.13 : (b) Phase angle ~frequency ratio for different values of damping ratio.
For a underdamped system the total response of the system which is the combination
of transient response and steady state response can be given by
It may be noted that as , the first part of equation tends to zero and second part
remains.
Rotating Unbalance:
One may find many rotating systems in industrial applications. The unbalanced force
in such a system can be represented by a mass m with eccentricity e , which is
rotating with angular velocity as shown in Figure 4.1.
Let x be the displacement of the nonrotating mass (M-m) from the static equilibrium
position, then the displacement of the rotating mass m is
From the freebody diagram of the system shown in figure 2.15, the equation of
motion is
or
or
or
Figure 2.16: Force polygon
or
and
From figure following observations may be made for a rotating unbalanced system.
For very low value of frequency ratio (say ) ), the response of the
system is very small.
For frequency ratio between 0.5 and1, there is a sharp increase in system
response with increase in frequency of excitation of the system.
At frequency ratio equal to 1, the phase angle is 900.
Maximum response amplitude occurs at a frequency slightly greater than
.
With increase in damping, the response of the system decreases.
For higher value of (say >2), the response amplitude approaches and
phase angle approaches 1800
In many industrial applications, one may find the vibrating machine transmit forces
to ground which in turn vibrate the neighbouring machines. So in that contest it is
necessary to calculate how much force is transmitted to ground from the machine or
from the ground to the machine.
This force will be transmitted to the ground only by the spring and damper.
Substituting these equations and defining the transmissibility TR as the ratio of the
force transmitted Force to the disturbing force one obtains
Replacing
To reduce the amplitude X of the isolated mass m without changing TR, m is often
mounted on a large mass M. The stiffness K must then be increased to keep ratio
K/(m+M) constant. The amplitude X is, however reduced, because K appears in the
denominator of the expression
Figure 2.20 shows the variation transmissibility with frequency ratio and it can be
noted that vibration will be isolated when the system operates at a frequency ratio
higher than
Equivalent Viscous Damping :
In the previous sections, it is assumed that the energy dissipation takes place due to
viscous type of damping where the damping force is proportional to velocity. But
there are systems where the damping takes place in many other ways. For example,
one may take surface to surface contact in vibrating systems and take Coulomb
friction into account. Also in many cases energy is dissipated in joints also, which is
a form of structural damping. In these cases one may still use the derived equations
by considering an equivalent viscous damping. This can be achieved by equating the
energy dissipated in the original and the equivalent system.
The primary influence of damping on the oscillatory systems is that of limiting the
amplitude at resonance. Damping has little influence on the response in the
frequency regions away from resonance. In case of viscous damping, the amplitude
at resonance is
Structural Damping :
When materials are cyclically stressed, energy is dissipated internally within the
material itself. Experiments by several investigators indicate that for most structural
metals such as steel and aluminum, the energy dissipated per cycle is independent of
the frequency over a wide frequency range and proportional to the square of the
amplitude of vibration. Internal damping fitting this classification is called solid
damping or structural damping. With the energy dissipation per cycle proportional to
the square of the vibration amplitude, the loss coefficient is a constant and the shape
of the hysteresis curve remains unchanged with amplitude and independent of the
strain rate. Energy dissipated by structural damping can be written as
Where is a constant with units of force displacement.
or
Coulomb Damping :
Summary
Some important features of steady state response for harmonically excited systems
are as follows-
In this chapter response due to rotating unbalance, support motion, whirling of shaft
and equivalent damping are also discussed.
The ratio of the maximum displacement of the forced vibration (x max) to the
static deflection under the static force F0 (xo) is known as Magnification factor.
Denoted by M.F.
x max
i.e, M.F =
x0
F0 / s
We have x max
2
2
2
1 2
n n
x0
x max
2
2
2
1 2
n n
1
M.F
2
2
2
1 2
n n
From this equation, it is clear that the magnification factor depends upon
-The ratio of circular frequencies
n
w d Fd .d x
dx
where damping force, Fd = C.
dt
Energy dissipated/cycle, Wd c x 2
Example: Accelerometers
b) Piezoelectric accelerometer
c) Servo accelerometer (force balance accelerometer)
Certain crystals exhibit the property that they generate a charge across their
faces when a stress is applied to them. This property is made use of in piezoelectric
accelerometer.
d → Piezoelectric constant
When the device is subjected to acceleration the mass exerts a variable force
on the Piezoelectric disc, which is proportional to the acceleration. The charge
developed across the disc is in turn proportional to the acceleration of the mass.
MODULE 3
TWO-DEGREE-OF-FREEDOM-SYSTEMS
A single degree of freedom system has only one natural frequency and
requires only one independent co-ordinate to define the system completely.
But a Two Degree freedom system has two natural frequencies and the free
vibration of any point in the system, in general, is a combination of two harmonies of
these two natural frequencies respectively. Under certain conditions, any point in the
system may execute harmonic vibrations, at any of the two natural frequencies, and
these are known as the principal moles of vibrations.
The vibrating systems, which require two coordinates to describe its motion,
are called two-degrees-of -freedom systems.
These coordinates are called generalized coordinates when they are
independent of each other and equal in number to the degrees of freedom of
the system.
Unlike single degree of freedom system, where only one co-ordinate and
hence one equation of motion is required to express the vibration of the
system, in two-dof systems minimum two co-ordinates and hence two
equations of motion are required to represent the motion of the system. For a
conservative natural system, these equations can be written by using mass
and stiffness matrices.
Figure 3.1(a) shows two masses m1 and m2 with three springs having spring stiffness
k1, k2 and k3 free to move on the horizontal surface. Let x1 and x2 be the displacement
of mass respectively.
Figure 3.1(a)
As described in the previous lectures one may easily derive the equation of motion
by using d'Alembert principle or the energy principle (Lagrange principle or
Hamilton 's principle)
Using d'Alembert principle for mass m1 from the free body diagram shown in figure
3.1(b)
and similarly for mass m2
Now depending on the position of point C, few cases can are studied below.
Case 1 : Considering , i.e., point C and G coincides, the equation of motion can
be written as
Figure 3.2
Hence in this case the system is dynamically coupled but statically uncoupled.
Case 3: If we choose , i.e. point C coincide with the left end, the equation of
motion will become
Here the system is both statically and dynamically coupled.
Again considering the problem of the spring-mass system in figure 6.1.1 with
, , , the equation of motion can be written as
We define a normal mode oscillation as one in which each mass undergoes harmonic
motion of the same frequency, passing simultaneously through the equilibrium
position. For such motion, we let
Hence,
Hence, and
So, the natural frequecies of the system are and
Now it may be observed that for these frequencies, as both the equations are not
independent, one can not get unique value of and . So one should find a
normalized value. One may normalize the response by finding the ratio of to .
From the first equation. the normalized value can be given by
and from the second equation the normalized value can be given by
It may be noted
VIBRATION ABSORBER
(1)
which will be maximum when Now to absorb this vibration, one may add a
secondary spring and mass system as shown in figure .
The equation of motion for this system can be given by
(2)
As we know for steady state vibration, the system will vibrate with a frequency of
the external excitation; we can assume the solution to be
(3)
(4)
Or, (5)
(6)
(7)
where
Now
Here are the roots of the characteristic equation . One may note
that these roots are the normal mode frequency for this two-degrees of freedom
system. These free-vibration frequencies can be given by
Hence, if a system called the primary system with a stiffness mass is subjected
to an exciting force or base motion to vibrate, it is possible to completely eliminate
the vibration of the primary system by suitably designing an attached spring-mass
system (secondary system) with stiffness and mass such that the natural
From equation (1) it may be noted that the primary system will have resonance when
the natural frequency of the primary system coincide with that of the excitation
frequency.
Hence to reduce the vibration at resonance of the primary system one should design
the secondary system such that the natural frequency of both the components
coincides.
For this condition
or,
For, ,
and
To keep the displacement of secondary mass small, the stiffness of the secondary
spring should be very large. To have this the secondary mass should also be large
which is not desirable from practical point of view.
Hence a compromise is usually made between the amplitude and the mass ratio. The
mass ratio is usually kept between 0.05 and 0.25.
The centrifugal pendulum vibration absorber was devised and patented in France
about 1935 and at the same time it was independently conceived and put into practice
by E. S. Taylor. Its purpose was to overcome serious torsional vibration problem
inherent in geared radial aircraft-engine �propeller system. Later it was modified
and incorporated into automobile IC engines in order to reduce the torsional
vibrations of the crankshaft. This was done by integrating the absorber mass with
crankshaft counter balance mass.
The tuned vibration absorber is only effective when the frequency of external
excitation equals to the natural frequency of the secondary spring and mass system.
But in many cases, for example in case of an automobile engine, the exciting torques
are proportional to the rotational speed �n' which may vary over a wide range. For
the absorber to be effective, its natural frequency must also be proportional to the
speed. The characteristics of the centrifugal pendulum are ideally suited for this
purpose.
Placing the coordinates through point O', parallel and normal to r, the line r rotates
with angular velocity ( ).
The acceleration of mass
Assuming to be small, , so
t may be noted that the same pendulum in a gravity field would have a natural
frequency of . So it may be noted that for the centrifugal pendulum the gravity
field is replaced by the centrifugal field .
Torque exerted by the pendulum on the wheel
With the component of equal to zero, the pendulum force is a tension along ,
given by times the component of .
Now substituting
which can be at its natural frequency. This possesses some difficulties in the
design of the pendulum. For example to suppress a disturbing torque of frequency
equal to four times the natural speed n , the pendulum must meet the requirement
. Hence, as the length of the pendulum becomes very
small it will be difficult to design it. To avoid this one may go for Chilton bifilar
design.
It must be appreciated that any real life system is actually a continuous or distributed
parameter system (i.e. infinitely many d.o.f). Hence to derive its equation of motion
we need to consider a small (i.e., differential) element and draw the free body
diagram and apply Newton 's second Law. The resulting equations of motion are
partial differential equations and more complex than the simple ordinary differential
equations we have been dealing with so far. Thus we are interested in modeling the
real life system using lumped parameter models and ordinary differential equations.
The accuracy of such models (i.e. how well they can model the behavior of the
infinitely many d.o.f. real life system) improves as we increase the number of d.o.f.
Thus we would like to develop mult-d.o.f lumped parameter models which still yield
ordinary differential equations of motion – as many equations as the d.o.f . We would
discuss these aspects in this lecture.
Or in compact form,
There are “n” equations of motion for an “n” d.o.f system. Correspondingly the mass
and stiffness matrices ([M] and [K] respectively) are square matrices of size (n x n).
ie
For a non-trivial solution to exist, we have the condition that the determinant of the
coefficient matrix must vanish. Thus, we can write,
We can substitute the values of in eqn and derive a relation between the
amplitudes of various masses yielding us the corresponding normal mode shape.
Typical mode shapes are schematically depicted in Figure for a d.o.f system.
We observed in the previous lecture that determination of all the natural frequencies
of a typical multi d.o.f. system is quite complex. Several approximate methods such
as Dunkerly's method enable us to get a reasonably good estimate of the fundamental
frequency of a multi d.o.f. system. Basic idea of Dunkerly's method
Thus,
As this is a two d.o.f. system, it is expected to have two natural frequencies viz
and ., Thus we can write Equation as:
It would appear that these two equations can be solved exactly for and . While this is
true for this simpleexample, we can't practically implement such a scheme for an n-
d.o.f system, as it would mean similar computational effort as solving the original
problem itself. However, we could get an approximate estimate for the fundamental
frequency. If >> , then we can approximately write
These can be readily verified to be the reciprocal of the equivalent stiffness values
for the modified systems.
Thus, we can write:
Holzer's Method.
This method is an iterative method and can be used to determine any number of
frequencies for a multi-d.o.f system. Consider a typical multi-rotor system as shown
in Fig. 3.7
The equations of motion for free vibration can be readily written as follows:
Thus:
Summing up all the equations of motion, we get:
Step 3: Compute the other d.o.f. using the equations of motion as follows:
Step 4: Sum up and verify if this equation is satisfied to the prescribed degree of
accuracy.
If Yes, the trial frequency is a natural frequency of the system. If not, redo the steps
with a different trial frequency.
In order to reduce the computations, therefore one needs to start with a good trial
frequency and have a good method of choosing the next trial frequency to converge
fast. Two trial frequencies are found by trial and error such that is a
small positive and negative number respectively than the mean of these two trial
frequencies(i.e. bisection method) will give a good estimate of for which
.
TORSIONAL VIBRATIONS
When the particles of a shaft or disc move in a circle about the axis of a
shaft, then the vibrations are known as torsional vibrations. In this case the shaft is
twisted and untwisted alternately and torsional shear stresses are introduced in the
shaft.
Torsional vibrations may result in shafts from following forcings:
For machines having massive rotors and flexible shafts (where system natural
frequencies of torsional vibrations may be close to, or within, the source frequency
range during normal operation) torsional vibrations constitute a potential design
problem area.
In such cases designers should ensure the accurate prediction of machine torsional
frequencies and frequencies of any of the torsional load fluctuations should not
coincide with torsional natural frequencies.
Simple systems with a single disc mass: Consider a rotor system as shown in
Figure 4.1. The shaft is considered as massless and it provides the stiffness only. The
disc is considered as rigid and it has no flexibility. If we give a small initial
disturbance to the disc in the torsional mode and allow it to oscillate its own, it will
execute free vibrations. The oscillation will be simple harmonic motion (SHM) with
a unique frequency, which is called natural frequency of the rotor system.
or
The free (or natural) vibration has the simple harmonic motion (SHM).
so that
where is the amplitude of the torsional vibration and is natural frequency of the
torsional vibration.
On substitution we get
or
Hence, the torsional natural frequency is given by square root of the ratio of torsional
stiffness to the polar mass moment of inertia.
In a two-discs torsional system as shown in Figure 4.3, whole of the rotor is free to
rotate i.e. the shaft being mounted on frictionless bearings.
From free body diagrams of discs in Figure 4.4, we can write
and
or and
For SHM,
and
On substitution we get
and
|k| = 0
and
From equation it can be concluded that, the first root of equation represents the case
when both discs simply rolls together in phase with each other as shown in Figure
4.5 i.e. the rigid body mode, which is of a little practical significance.
or
The second mode represents the case when both masses vibrate in anti-phase with
one another. Figure 4.6 shows the second mode shape of two-rotor system, showing
two discs vibrating in opposite directions.
From the second mode shape, i.e. from Figure 4.6 and noting equation we have
Since both masses are always moving in the opposite direction, so there must be a
point on the shaft where the torsional vibration is not taking place i.e. a torsional
node. The location of the node may be established by treating each end of the real
rotor system as a separate single-disc cantilever system as shown in Figure 4.5. The
torsional node being treated as the point where the shaft is rigidly fixed.
Since the natural frequency of the system is known and the frequency of oscillation
of each of the single-disc system must be same, hence we write
and
and
The equivalent shaft must have the same torsional stiffness as the real shaft, since for
the present case torsional springs are connected in series. The equivalent torsional
spring can be written as
we have
which gives
where
where , and are equivalent lengths of shaft segments having equivalent shaft
diameter d3 and le is the total equivalent length of unstepped shaft having diameter d3
as shown in Figure 4.7(b).
From Figure 4.7(b) and noting equations in the equivalent shaft the node
location can be obtained as
with and
From equations the node position a & b can be obtained in the equivalent shaft
length.
Now the node location in real shaft system can be obtained as follows:
we have
Since above equation is for shaft segment in which node is assumed to be present, we
can write
and
It can be combined as
So once a & b are obtained the location of the node in the actual shaft can be
obtained i.e. the final location of the node on the shaft in real system is given in the
same proportion along the length of shaft in the equivalent system in which the node
occurs.
MODULE 4
Consider a typical shaft, carrying a rotor (disk) mounted between two bearings as
shown in Figure.Let us assume that the overall mass of the shaft is negligible
compared to that of the rotor (disk) and hence we can consider it as a simple
torsional spring. The rotor (disk) section has a geometric centre i.e., the centre of the
circular cross-section and the mass centre due to the material distribution. These two
may or may not coincide in general, leading to eccentricity. The eccentricity could be
due to internal material defects, manufacturing errors etc. As the shaft rotates, the
eccentricity implies that the mass of the rotor rotating with some eccentricity will
cause in-plane centrifugal force. Due to the flexibility of the shaft, the shaft will be
pulled away from its central line as indicated in the figure. Let us assume that the air-
friction damping force is negligible. The centrifugal force for a given speed is thus
balanced by the internal resistance force in the shaft-spring and the system comes to
an equilibrium position with the shaft in a bent configuration as indicated in the
figure. Thus the shaft is rotating about its own axis and the plane containing the bent
shaft and the line of bearings rotates about an axis coinciding with the line of
bearings. We consider here only the case, wherein these two rotational speeds are
identical, called the synchronous whirl.
Centrifugal force
shaft resistance force
wherein, the shaft stiffness k is the lateral stiffness of a shaft in its bearings i.e., considering the rotor a
span, this is the force required to cause a unit lateral displacement at mid-span of the simply supported
Thus
Where E is the Young's modulus, I is the second moment of area, and L is the length between the suppo
ie
where we have used to represent the natural frequency of the lateral vibration of the sp
shaft-rotor system. Thus when the rotational speed of the system coincides with the natural frequen
lateral vibrations, the shaft tends to bow out with a large amplitude. This speed is known as the critical
and it is necessary that such a resonance situation is avoided in actual practice. As discussed earlier
case of resonance, it takes some time for the amplitude to build up to a large value. Some of the tu
rotors whose operating speeds go beyond the critical speed are able to use this fact and rush-throug
critical speed. It is necessary to observe that, in synchronous whirl. the heavier side remains all the tim
the outer side. Thus when the shaft bends, an inner fibre is under compressive stress and outer fibre is
tensible stress but there is NO REVERSAL of stress.
Rayleighs Method
Note that the amounts of kinetic and potential energy in the system may change with
time but their sum must remain constant. Thus if and are energies at time and
and are energies at time , then
For a shaft as shown in Fig.4.2 the potential energy is zero at the specific instant of
time when the mass is passing through its static equilibrium position and kinetic
energy is at its maximum . Similarly at the instant when the mass is at its
extreme position the kinetic energy is zero and the potential energy is at its
maximum . Thus we have the following relationship.
When a shaft carries multiple disks it is always efficient to use this method.
as where k is the lateral stiffness of the shaft in its bearings and m is the mass
of the disk.
If a light shaft having two end supports has a central disc then the system
has been shown to have one critical speed. Even if the disc is not central, the system
will have one critical speed as long as we assume the mass of the disc to be
concentrated. If, however, the disc has mass as well as moment of inertia, and is not
central, then the system will have two critical speeds. The treatment given below is
for a light cantilever shaft having a disc which has mass as well as moment of inertia.
Since the critical speed is numerically equal to the natural frequency of lateral
vibrations, we will find the later for this system.
Further let
a22 = slope at the free end of the beam per unit moment acting on
the CG of the disc, in the plane of the paper = l/EI
The inertia force and the inertia torque on the disc in the displaced position
are shown in Fig.8.6.1, along with their directions. These are as follows:
Then the deflection at the CG of the disc and the rotation of the disc in the
plane of the paper are given by
2
g1,22 (h 1) (h 1) 2 h (6)
h
Figure (2) is a plot of the above equation and shows the variation of the two
3r 2
natural frequencies of the system with the change in h 2 ; it may be recalled
l
that r is radius of gyration of the disc about an axis passing through its CG and
perpendicular to the axis of the disc.
3r 2
change in h 2 .
l
Transient Vibrations
INTRODUCTION
The use of Laplace transform is introduced in this chapter for the analysis of
systems subjected to shock pulses. The usual differential equations method or the so-
called classical method becomes very lengthy and cumbersome with transient
excitations of different shapes.
LAPLACE TRANSFORMATION
F(s) f (t)d st dt
0
L [f(t)] = F(s)
F̂ mdv
where dv is the change in velocity of the mass due to the impulse F̂ . If the system is
initially at rest, impulse gives it a starting velocity of
F̂
dv
m
The initial displacement of the mass form the equilibrium position is zero
because of the extremely small duration of the impulse.
x(0) 0
F̂ (1)
x(0)
m
mx cx kx 0 (2)
The forcing function on the right has been taken to be zero since the impulse
effectively gives only the initial conditions obtained in Eq. (1).
x 2n x n 2 x 0 (3)
F̂ 1
X(s) 2 (4)
m s 2n s n 2
In order to obtain the inverse transformation fro the above equation, the
expression on the right has to be re-arranged in one of the forms corresponding to the
transform pairs , for direct inversion. If 1 , the above equation is re-written in the
following form
F 1 2 n
X(s) (5)
2
m 1 2 n (s ) 2 1 2
n n
F̂
x(t) en t sin 1 2 n t (6)
m 1 n
2
mx cx kx F0 u(t)
F0
or x 2n x n 2 x u(t) (1)
m
F0 1
[s2 X(s) s x(0) x(0)] 2n [s X(s) x(0)] n 2X(s) .
m s
A second order system subjected to a finite step cannot have any initial
velocity or displacement. So, putting all initial conditions zero in the above equation,
and re-arranging, we have
F0 1
X(s) 2 2
(2)
m s(s 2n s n )
F0 1 1 s 2n
X(s)
m n 2 s s 2 2n s n 2
The right had expression in the bracket above is still not invertable directly.
Assuming an underdamped system, i.e. 1 , the above equation is written as
follows:
. 1 2 n
F 1 1 (s n ) 1 2
X(s) 0 2 (3)
m n s
2 2
(s n ) 2 1 2 n (s n )
2
1 n
2
Inverse the transform of Eq. (3) can now be obtained directly from the table
and is given below
F0
x(t) 1 e n t cos 1 2 n t e n t sin 1 2 n t
m2n 1 2
For an undamped case, response equation can be written from the above
equation by putting 0 , or
F0
x(t) [1 cos n t] (5)
k
Rectangular pulse
F0 F
x(t) [1 cos n t]u(t) 0 [1 cos n (t )]u(t ) (1)
k k
F0
x(t) [1 cos n t] for 0 t
k
(2)
F0
[cos n (t ) cos n t] for t
k
A spring mass system with initial conditions X0 and V0, has its differential
equation written as
x n 2 x 0 (1)
x Asin(n t )
V0 2
where A X02
n 2
X
and tan 1 n 0
V0
x
or A cos(n t ) (2)
n
It may be noted that the centre of the phase trajectory always lies on the x-
axis at a distance equal to the static equilibrium displacement of the system. In the
case discussed the static equilibrium displacement was zero and therefore the centre
of the circle was located at the origin. In case of a step force input F0, the static
equilibrium position suddenly changes through a distance F0/k. Thus the phase plane
plot for such a motion will be a circle whose centre lies F0/k above the centre. The
radius of this circle will be F0/k so that the trajectory starts from the origin
corresponding to zero initial conditions.
The use of the phase plane method is illustrated by the following examples
for systems subjected to multiple steps.
SHOCK SPECTRUM
Non-Linear Vibrations
INTRODUCTION
mx cx kx F(t) (1)
In this equation which is for a linear system, the inertia force, the damping
force and the spring force are linear functions of x, x and x respectively. This is not
so in the case of non-linear systems. A general equation for a non-linear system is
in which the damping force and the spring force are not linear functions of x and x.
There are quite some physical systems which have non-linear spring and damping
characteristics. Rubber springs and other similar isolators have spring stiffness which
increases with amplitude. Cast iron and concrete have spring stiffness which
decreases with amplitude. Examples of non-linear damping are dry friction damping
and material damping. Even so called linear systems tend to become non-linear with
larger amplitudes of vibration. The analysis of non-linear systems is comparatively
difficult. In certain cases there is no exact solution.
One major difference between the linear and non-linear systems is that the
law of superposition does not hold good for non-linear systems. Mathematically
speaking, if x1 is a solution of
mx cx kx F1 (t)
and x2 is a solution of
mx cx kx F2 (t)
This is not so in the case of non-linear systems. Even for the case of free
vibration any two known solutions of the non-linear system cannot be superimposed
to obtain a general solution.
PHASE PLANE
Phase plane was introduced in Sec. 9.6 for the case of linear systems. Here
we extend it for the case of non-linear systems.
mx f (x) 0 (1)
dv
xv
dx
dv
mv f (x) 0
dx
or mv dv = –f(x) dx (2)
v x
mv 2 mV0 2
or [F(x) F(X 0 )]
2 2
mv 2 mV0 2
F(x) F(X 0 ) (3)
2 2
which states that the total energy of the system at any instant is equal to the total
initial energy of the system. Curve in x – v plane can be drawn from Eq. (3) and this
will be a curve of constant energy. A set of such curves can be drawn, each for
different total energy. These curves are known as Energy Curves or Integral Curves
in phase plane.
We had taken the phase with x along the ordinate and v along the abscissa,
the trajectory was always counter-clockwise. Here, for convenience we will take x
along the abscissa and v along the ordinate. The trajectories here will be clockwise.
Consider the linear case when f(x) = kx. Equation (2a) then integrates to
mv 2 mV02 kx 2 kX02
2 2 2 2
The phase plane trajectories are clearly a set of ellipses with the origin as the
centre. The right hand side of Eq. (4) is the total initial energy of the system. As the
value of this initial energy (depending upon initial conditions) increases, the size of
the ellipse also increases.
In general a point P(x, v) in the phase plane, called the representative point
of the system, represents the state of the system at any instant t ad the trajectory
traced gives the history of the system. Through any and every point of the phase
plane passes one and only one trajectory and thus the trajectories in the plane do not
intersect one another.
With the passage of time the representative point moves along the trajectory
in a clockwise direction with what is known as the phase velocity given by
u x 2 v2 (5)
Consider the general case of a system with non-linear damping and non-
linear spring. Let the differential equation of motion be
Letting x v , the above equation can be written down in the form of the
following two equations.
dx
v
dt
(7)
dv (v) f (x)
dt m
dv (v) f (x)
(8)
dx mv
Equation (8) gives the slope of the trajectory at any point (x, v) in the phase
plane and is useful for constructing the plots. The slope of the trajectory is directly
obtainable at all points except the ones where v = 0 and (v) f (x) 0 . At these
points the slope becomes indeterminate and these points are called singular points.
From Eq. (7), these points correspond to the conditions v = 0 and dv/dt = 0, i.e. the
points of equilibrium. Singular points exist on x-axis (v = 0) wheredv/dt = 0. Origin
is always a singular point. Other singular points may or may not be there for the
system. At singular points the phase velocity is zero.
UNDAMPED FREE VIBRATION WITH NON-LINEAR SPRING FORCES
mx f (x) 0 (1)
dv
mv f (x) 0 (2)
dx
mv 2
F(x) E (3)
2
where F(x) is the integral of f(x) and so represents the potential energy of the system,
and E is the total energy of the system and depends upon the initial conditions.From
2
Eq. (3), we have v [E F(x)] (4)
m
f (x) x x 3 ( 0) (5)
x 2 x 4
then, F(x) (6)
2 4
mv 2 x 2 x 4
E (7)
2 2 4
The above equation gives the plot in the phase plane for different values of
E. These are closed curves when 0 for any amplitude. For the case when 1 ,
the phase plane plots are closed curves upto a certain amplitude and beyond that they
are unstable. For the case of closed curves, the system has periodic motion. Let
x max a be the amplitude of vibration. When x x max a , v = 0; Eq. (7) becomes
a 4 a 2
E0 (8)
4 2
2 4E
giving a
2
(9)
In the above equation only positive sign before the radical is applicable
whether or 0 .
dx
Writing v as Eq. (4) is written as
dt
dx m dx
dt (10)
2 2 [E F(x)]
[E F(x)]
m
For the case of periodic motion of amplitude a, the time period per cycle of
vibration is given by integrating the above equation over a quarter of a cycle and
multiplying it by 4. Thus,
a
m dx
4
2 0 E F(x)
(11)
Substituting for F(x) from Eq. (6) in the above equation, we have
a
m dx
4
2 0 E (x 2 / 2) (x 4 / 4)
(12)
The quadratic expression in the radical sign of the above equation can easily
be factorized since comparing it with Eq. (8) it shows that (a2 – x2) is one of the
factors of this quadratic function.
Therefore, let
x 2 x 4 2 2 x
4
E (a x ) b
2
(13)
2 4 4
E a 2 b2
a 2 (14)
b
2
2 4
x 2
b2 (15)
4 2
Equation (12) can now be written with the help of Eq. (13) and (15) as
a
m dx
2 0
4
a 2 x 2
(a 2 x 2 )
4 2 4
a
m dx
or 8
2 0 (a 2 x 2 )(a 2 2 x 2 )
(16)
PERTURBATION METHOD
where 0 is the natural frequency of the linear system. Assuming that the solution
can be written in the form of a Taylor series in terms of the parameter (known as
perturbation parameter), we can write
where all x’s are functions of time t. The only restriction in the above way of writing
is that be a small quantity.
(x 0 2 x 0 ) (x1 2 x1 1x 0 x 03 )
Since the above equation must hold good for any small value of , it means
that each of the terms in the parenthesis must individually be zero, therefore.
x 0 2 x 0 0
x1 2 x1 1x 0 x 03 0
(6)
x 2 x 2 1x1 2 x 0 3x 0 x1
2 2
... ... ... ... ... ... ... ...
x a
at t = 0
x 0
Again, since these equations must be satisfied for any small value of , we
have
x 0 (0) a x 0 (0) 0
x1 (0) 0 x1 (0) 0
(7)
x 2 (0) 0 x 2 (0) 0
... ... ... ...
With the first set of initial conditions in Eq. (7), the solution of the first
differential equation in Eq. (6), is
x 0 a cos t (8)
Substituting the above in the right side of the second differential equation in
Eq. (6), we get
3 1
cos3 t cos t cos3t
4 4
3 1
x1 2 x1 (1a a 3 )cos t a 3 cos3t (10)
4 4
3
In the above equation the forcing function (1a a 3 )cos t will cause
4
resonance to the system since the left hand side shows the natural frequency of the
system as , the same as that of the first part of excitation. In order to avoid this
absurdity, we must have
3
1a a 3 0 (11)
4
1
Therefore, x1 2 x1 a 3 cos3t
4
a3
x1 (A1 cos t A 2 sin t) cos3t (12)
322
Applying the zero initial conditions from Eq. (7), we get the constants as
a3
A1
322
A2 = 0
a3
x1 [cos t cos3t] (13)
322
Substituting Eq. (8) and (13) in the first two terms of Eq. (2), the solution
upto first order correction is obtained as
a 3
x x 0 x1 a cos t [cos t cos3t] (14)
322
3
2 0 2 1 02 a 2 (15)
4
Considering only small values of and F0, it is known from experience that
the frequency of steady state vibration will be the same as that of excitation plus
some higher harmonics. So, the first approximate solution can be written as
x1 a cos t (3)
Substituting the above approximate solution in the right hand side of Eq.
(2), we have
x 2 n 2a cos t a 3 cos3t F0 cos t (4)
The double integration of the above equation will give x 2 which will be a
better approximation than x1 given in Eq. (3). Using the relationship
3 1
cos3 t cos t cos3t
4 4
3 1
x 2 (F0 n 2a a 3 )cos t a 3 cos3t (5)
4 4
Integrating the above equation twice and dropping out the constants of
integration to ensure periodic motion, we have
1 3 a3
x2 {F0 n
2
a } a 3
cos t cos3t (6)
2 4 362
1 3
a (n 2a a 3 F0 ) (7)
2
4
3 F
or 2 n 2 a 2 0 (7a)
4 a
a 3
x(t) x 2 a cos t cos3t (8)
362
Equation (7) is a cubic in a and therefore for any value of there are in
general three values of a; one root is always real, the other may be real or complex
conjugate.
When damping is present in the non-linear systems, the skewed peaks wind
up at a certain stage.When the frequency of excitation is gradually increasing from
zero, the response varies along the points there being a sudden change in amplitude
from 2 to 3 at the corresponding frequency. The portion of the response curve is
never traced.This phenomenon of sudden change in amplitude from 2 to 3 while the
frequency is gradually increasing and the sudden change from 5 to 6 when the
frequency is gradually decreasing is known as Jump Phenomenon.
The self excited vibrations differ from forced vibrations in that the
fluctuating force that sustains the motion is controlled by some part of the motion
itself. The exciting force may be a function of a displacement, velocity or
acceleration of the motion. When motion is stopped by some means, the fluctuating
force also disappears. The forcing function is thus dependent motion itself unlike
forced vibrations. Tool chatter and aeroplane wing flutter are some of the common
examples of self excited vibration.
s1 = a1
s2 = a2 + jb2
s3 = a2 – jb2
When mass is stationary, the friction coefficient between the mass and the
belt is a . When mass is moving towards the right, the relative velocity decreases
and the coefficient of friction increases. On the other hand when the mass is moving
towards the left, the relative increases and therefore the coefficient of friction
decreases. Since the friction force on the mass is always towards right, the helping
friction force when mass moves towards right is always greater than the opposing
friction force when mass moves towards left. That means a certain net energy is put
into the system in each cycle. The amplitude continues to increase. If however, the
mass is brought to rest in the equilibrium position, it stays in that position. The least
disturbance will set it vibrating with increasing amplitude. This is a case of self
excited vibration. The frequency of vibration in the cases of self excited system is
approximately equal to the natural frequency of the system provided damping is not
large.
This problem can also be tackled analytically. At any instant when the
displacement of mass is x and its velocity x , the relative velocity between the mass
and the belt is (V x) . At this instant coefficient of friction is (a x) . The normal
reaction on the mass is mg. The differential equation of motion is then written as
mx kx mg(a x)
mx mgx kx mg a (1)
Equation (1) gives an effective negative damping to the system which sends
it into large increasing amplitudes. The static displacement is mg a / k .
In case there is viscous damping between the mass and the belt, the slope of
the equivalent friction line is no longer negative and there will not be any self excited
vibrations. There are numerous other examples of self excited vibrations caused by
dry friction, a few of which are:-
MODULE 5
PROPOGATION OF SOUND
2. Only the wave travels forward whereas the particles of the medium
vibrate about their mean positions.
4. The velocity of the wave is different from the velocity with which the
particles of the medium are vibrating about their mean positions. The
wave travels with a uniform velocity whereas the velocity of the
particles is different at different positions. It is maximum at the mean
position and zero at the extreme positions of the particles. There are
two types of wave motions.
a. Transverse wave.
b. Longitudinal wave
Sound wave are longitudinal waves and light waves are transverse waves.
Figure 1 shows the formation and propagation of transverse and longitudinal wave.
l1
L K log( )
l2
l1
L 10log( )
l2
102
L 10log( 12 ) 120db
10
The following table gives the approximate value of some sound measured in
decibels.
AIR COLUMNS
1. The diameter of the pipe is small compared with the length of the pipe
and with the wave length of sound.
The organ pipes are classified into two groups: Flute or Flue pipes and Reed
pipes.
DOPPLER EFFECT
It is commonly observed that the pitch of a note apparently changes when
either the source or the observer are in motion relative to each other. When the
source approaches the observer or when the observer approaches the source or when
both approach each other the apparent pitch is higher than the actual pitch of the
sound produced by the source. Similarly when the source moves away from the
observer or when the observer moves away from the source or when both move away
from each other, the apparent pitch is lower than the actual pitch of the sound
produced by the source.
This apparent change in pitch due to relative motion between the source and
the observer is called Doppler effect.
Doppler effect in sound is asymmetric, when the source move towards the
observer with a certain velocity, the apparent pitch is different to the case when the
observer is moving towards the source with the same velocity. But it is not so in the
case of light. Doppler effect in light is symmetric.
- wavelength
v - velocity of sound
n - apparent pitch
Case 1: When the source moves towards the stationary observer with a
velocity ‘a’
v
n n
va
Case 2: When the source moves away from the stationary observer with a
velocity ‘a’
v
n n
va
vb
n n
v
Case 5: When the source moves towards the observer and the observer
moves away from the source.
vb
n n
va
Case 6: When the source and the observer move towards each other
vb
n n
va
Case 7: When the source and the observer move away from each other.
vb
n n
va
Case 8: Source moving away from the observer and the observer moving
towards the source.
vb
n n
va
MUSICAL SCALES:
HARMONY: When the two notes sounded together produce concord their
combination is called Harmony.
MELODY: If the two notes are sounded one after the other, combination
is called melody.
C D E F G A B C
sa re ga ma pa dha ni (sa)'
All these notes are arranged in increasing order of frequencies so that they
present a regular graduation in pitch and their vibration frequencies are represented
by
9 5 4 3 5 15
l 2
8 4 3 2 3 8
i.e., if the frequencies of first note C called the tonic or the key note be taken
as 24, the relative frequencies (ratio of successive frequencies) of the various notes
of the diatonic scale will be
9 10 16 9 10 9 16
8 9 15 8 9 8 15
If the frequencies of the note C be taken as 256 and 264 respectively, the
various notes of the scale will be denoted by
MICROPHONE
1
V V Ka
i l sin t
R Ka sin t R R
V Ka K 2a 2 2
l sin t 2
sin t .......
R R R
The first term indicates a steady current when the diaphragm is at rest, the
second an alternating current of the same frequency as is impressed on the diaphragm
and the rest of the terms denote its other harmonics. The current is thus modulated.
This modulated electrical current passes through P, the primary of a transformer and
produces by induction a corresponding varying current in S, its secondary. This
amplified current passes through the telephone T and excites its diaphragm. The
movements of the latter set the air in corresponding vibrations reproducing the
original sound.Other commonly used types of microphones are
THE LOUDSPEAKER:
The moving part of the apparatus consists of a small coil called the 'voice
coil' wound on a cylindrical strip to which the variable current output of the
microphone is fed. The voice coil is free to move in the annular gap between the
central S and the ‘peripheral’ pole piece N, of a 'pot' magnet designed to produce a
strong radial magnetic field in it. It is usually magnetized by a steady (DC) current
flowing in the coil wound round it. The coil is attached to a conical diaphragm made
of parchment with circular corrugation and supported round the periphery by a
flexible annular strip of leather or rubber. When the variable current passes through
the coil in the magnetic field, it causes varying movement of the coil along the axis
with the frequency of the current variations. The diaphragm is thus set into vibrations
which are communicated to the external air and the sound is reproduced. The greater
the energy supplied to the voice coil, the louder will be the sound emitted by the
diaphragm. Completely surrounding the cone and attached to it by silk threads is the
'baffle'. It prevents the air vibrating behind the cone from flowing round to its front.
The relation between the current and driving force is linear and force is
independent of the position of the coil in the gap for considerable movements. When
suitably designed a fairly uniform response of 80cps to 1000cps is secured. It is
capable of radiating large power without appreciable asymmetric distortion.
ACOUSTICAL MEASUREMENTS:
The most commonly used instrument for sound measurement is the sound
level meter. This actually made up of a number of interconnected components. The
sound pressure is transduced to a voltage by means of a microphone. Microphones
generally employ a thin diaphragm to convert pressure to motion. Microphone often
have an arrangement so that it will not respond to constant and slowly varying
responses. This is necessary because the hour-to-hour and day-to-day changes in
atmospheric pressure are much greater than the sound pressure fluctuations to which
the microphone must respond. The motion is then converted to voltage by some
suitable transducer, usually a capacitance, piezo-electric or moving coil type.
The output voltage of the microphone generally is quite small and at a high
impedance level. So an amplifier of high input impedance and gain is used at the
output of the microphone. This can be a relatively simple ac amplifier, since response
to static or slowly varying voltages is not required.
Following the first amplifier are the weighting networks. They are electrical
filters whose frequency response is tailored to approximate the frequency response of
the average human ear. Readings taken with a weighting network are called sound
level rather than sound pressure level.
The output of the weighting network is further amplified and an output jack
provided to lead this signal to an oscilloscope or to a spectrum analyzer. If only the
overall sound magnitude is desired, the rms value of e3 must be found. The average
value of e3 is determined by rectifying and filtering and then the meter scale is
calibrated to read rms values. This procedure is exact for pure sine waves since there
is a precise relation between the average value and rms value of a sine wave. For non
sinusoidal wave this is not true, but the error is generally small enough to be
acceptable for relatively unsophisticated work.
ACOUSTICS OF BUILDINGS
Reverberation
1. direct waves
The quality of the note received by the listener will be the combined effect
of these two sets of waves. There is also a time gap between the direct wave received
by the listener and the direct waves received by the listener and the waves received
by successive reflection. Due to this, the sound persists for sometime even after the
source has stopped. This persistence of sound is termed as reverberation. The time
gap between the initial direct note and the reflected note up to the minimum
audibility level is called reverberation time. The reverberation time will depend on
the size of the room or the auditorium, the nature of the reflecting material on the
wall and the ceiling and the area of the reflecting surfaces.
Loudness:
Focusing:
Echelon Effect
The extraneous noise may be due to (1) sound received from outside the
room (2) the sound produced by fans etc., inside the auditorium. The external sound
cannot be completely eliminated but can be minimized by using double or triple
windows and doors. Proper attention must also be paid to maximum permissible
speed of fans and rate of air circulation in the room. The air conditioning pipes
should be covered with cork and insulated acoustically from the main building.
Resonance
5. The curved walls and comers bounded by two walls should be avoided.
This is done to avoid 1. Concentration of sound 2. Dead spaces.
7. the walls and the ceiling should be covered with the materials having
high absorption coefficient i.e., with perforated cardboards, felt
asbestos fibre glass etc.
FOURIER THEOREM
The theorem states that "any single valued periodic function can be
expressed as a sum of a number of simple harmonic terms which are multiples of the
given function." The theorem is generally referred in relation to the study of
transverse vibration of strings. However, the theorem has a wider scope.
Fourier Series
m
y f (t) Am cos(mt m )
m 1
frequency . A1, A2, ......., Am are amplitudes of the components of the 4 simple
2
harmonic vibrations 1, 2 ,........, m,........, represent their respective initial phases.
m
y f (t) Am sin mt Bm cos mt)
m 0
Here A0 = 0
m m
y f (t) B0 A m sin mt Bm cos mt
m 1 m 1
1
l
T 0
B0 ydt
1
2
T 0
Am ysin(mt)dt
1
2
Bm ycos(mt)dt
T0
Where T is time period of the function
Fourier series has been employed for the analysis of complex periodic
curves with a view to finding the various harmonic components of which they may
be built up together with their amplitudes and relative phases. It can be shown in
various ways that such components have an objective physical existence and are not
a mere mathematical fiction. A trained musical ear can easily resolve complex
musical sound hearing each simple harmonic component as a separate simple tone
and thus serve as a natural Fourier analyzer.
An acoustic filter is a device which has been extensively used for analyzing
the quality of complex sound waves. It is so designed that it transmits certain
selected ranges of frequency with negligible attenuation, and suppresses other
frequencies almost entirely. The analogy between acoustic filters and electrical filters
used in ac circuits for a similar purpose is so close that the considerations and
equations operating in the functioning of the latter have helped a good deal in
designing the former.
d2 y dy
m 2
ky Fsin t ................ 1
dt dt
d 2q dq q
L 2
R E sin t ................ 2
dt dt c
2. High pass filter: These are made with a straight tube for transmission
and short side tubes 0.5cm long and 0.28cm diameter opening through a hole with
conductivity 0.08 into a tube 10cm long and 1 cm diameter. Six sections of such a
filter would transmit about 90 percent of sounds above 800hz but would refuse
transmission to sounds of lower frequency.