Introduction to Vibrating Mechanical Systems
Airplanes, bridges, ships, machines, cars, etc. are vibrating mechanical systems.
The simplest mechanical system is the mass–spring system.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 1 / 29
Mass–Spring System with Damping and Forcing
O O O
A B e B e
(a) m x
(b) m Damper
P
F (t)
(c)
Mass–spring system: natural, equilibrium, and displaced positions
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 2 / 29
Oscillations of a Spring
Consider a spring OA suspended vertically from a fixed support at O. Let a body
of mass m be attached to its lower end at A.
Let e(= AB) be the elongation produced by the mass m hanging in equilibrium.
Then B is called the position of static equilibrium, and e is called the static
extension.
If k is the stiffness of the spring (i.e., the restoring force per unit stretch), then at
equilibrium position B,
mg = T = ke. (1)
Now let the mass be set in motion from equilibrium position. At any time t, let
the mass be at point P where BP = x.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 3 / 29
Forces Acting on the Mass
The mass experiences the following forces:
1 Gravitational force mg (weight of the mass), acting downward.
2 Restoring force of the spring k(e + x) due to displacement (elongation) of
the spring, acting upward.
3 Damping (frictional or resistance) force of the medium,
dx
c ,
dt
opposing the motion (acting upward).
4 External force F (t), considering downward direction as positive.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 4 / 29
Differential Equation of Motion
By Newton’s second law, the equation of motion is
d 2x dx
m = mg − k(e + x) − c + F (t).
dt 2 dt
Using the equilibrium condition mg = ke from (1), we get
d 2x dx
m 2
= −kx − c + F (t).
dt dt
Hence,
d 2x dx
m 2
+c + kx = F (t).
dt dt
Dividing throughout by m,
d 2x c dx k 1
2
+ + x = F (t). (2)
dt m dt m m
Its solution gives the displacement x of the mass at any instant t.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 5 / 29
Hooke’s Law and Systems of Units
Hooke’s Law states that the magnitude of the force needed to produce a certain
elongation of a spring is directly proportional to the elongation.
In symbols,
F = ke,
where
F is the magnitude of force,
e is the amount of elongation,
k is the constant of proportionality.
The constant k is called the spring constant.
Systems of Units:
1 F.P.S. system: force in pound (lb), time in second (sec).
If mass is m in pounds and acceleration is a in ft/sec2 ,
then force is F = ma in poundals.
2 C.G.S. system: length in cm, time in sec.
If mass is m in grams and acceleration is a in cm/sec2 ,
then force is F = ma in dyne.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 6 / 29
1 M.K.S. system: length in meter (m), mass in kilogram (kg), time in second
(sec).
If mass is m in kg and acceleration is a in m/sec2 ,
then force is F = ma in newton (N).
Note: g is the acceleration due to gravity and is the weight of the body in
gravitational units.
2 2 2
g = 32.2 ft/sec = 980 cm/sec = 9.8 m/sec (approx.)
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 7 / 29
Case 1: Free Oscillations (Undamped)
If the external force F (t) is absent and the damping force is negligible, then
equation (2) reduces to
d 2x k
2
+ x = 0.
dt m
Writing
k
ω2 = ,
m
it becomes
d 2x
+ ω 2 x = 0. (3)
dt 2
The auxiliary equation of (3) is
D 2 + ω 2 = 0.
Hence,
D = ±iω.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 8 / 29
Therefore, the solution of equation (3) is
x(t) = c1 cos(ωt) + c2 sin(ωt).
This can also be written as
x(t) = A cos(ωt + B), (4)
where c1 , c2 (or A, B) are arbitrary constants determined by the initial conditions.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 9 / 29
Period and Frequency of Free Oscillations
Equation (4) represents the simple harmonic motion of the mass m.
The period (time for one complete oscillation) is given by
2π
T = .
ω
Since
k
ω2 = ,
m
we have r
2π m
T =p = 2π .
k/m k
Also, since mg = ke, we obtain
r
e
T = 2π .
g
The frequency (number of oscillations per second) is
r
1 ω 1 g
f = = = .
T 2π 2π e
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 10 / 29
Case 2: Damped Free Oscillations
If the external force F (t) is absent and damping is present, then equation (2)
reduces to
d 2x c dx k
+ + x = 0.
dt 2 m dt m
Writing
c k
= 2p and = ω2,
m m
it becomes
d 2x dx
2
+ 2p + ω 2 x = 0. (5)
dt dt
This is a linear differential equation with constant coefficients.
Its auxiliary equation is
D 2 + 2pD + ω 2 = 0.
The roots are p
D = −p ± p2 − ω2.
Thus the nature of the solution depends upon the relative values of p and ω.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 11 / 29
Subcase 1: Overdamped Motion (p > ω)
When p > ω, the roots of the auxiliary equation are real and distinct:
p
D = −p ± p 2 − ω 2 .
Hence the solution of equation (5) is
√ √
−p+ p 2 −ω 2 t −p− p 2 −ω 2 t
x(t) = c1 e + c2 e .
Equivalently, h √2 2 √ 2 2 i
x(t) = e −pt c1 e p −ω t + c2 e − p −ω t . (6)
Observation:
x(t) decreases exponentially.
Motion is non-oscillatory.
As t → ∞, x(t) → 0.
This motion is called overdamped (dead-beat motion).
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 12 / 29
Subcase 2: Underdamped Motion (p < ω)
When p < ω, the roots are complex:
p
D = −p ± i ω2 − p2 .
Hence the solution of equation (5) is
h p p i
x(t) = e −pt c1 cos ω 2 − p 2 t + c2 sin ω2 − p2 t .
This can also be written as
p
x(t) = Ae −pt cos ω2 − p2 t + B . (7)
The motion is oscillatory because of the trigonometric terms.
The periodic time of oscillation is
2π
T = p ,
ω2 − p2
which is greater than the time period of free oscillations (without damping),
2π
T0 = .
ω
Thus damping increases the time period and the oscillations gradually die out.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 13 / 29
Subcase 3: Critically Damped Motion (p = ω)
When p = ω, the roots of the auxiliary equation are equal:
D = −p.
Hence the solution of equation (5) is
x(t) = (c1 + c2 t)e −pt . (8)
Observation:
Motion is non-oscillatory.
The displacement approaches zero as t → ∞.
This motion is called critically damped motion, because it separates the
non-oscillatory motion (overdamped case) from the oscillatory motion
(underdamped case).
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 14 / 29
Summary of Damped Free Oscillations
The damped free oscillation equation is
d 2x dx
+ 2p + ω 2 x = 0.
dt 2 dt
Three Cases
1 Overdamped: p > ω (non-oscillatory)
√ 2 2 √ 2 2
x(t) = c1 e (−p+ p −ω )t + c2 e (−p− p −ω )t
2 Underdamped: p < ω (oscillatory)
p
x(t) = Ae −pt cos ω2 − p2 t + B
3 Critical damping: p = ω (non-oscillatory)
x(t) = (c1 + c2 t)e −pt
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 15 / 29
Comparison of Damped Motions
x(t)
p>ω
p=ω
t
0
p<ω
Note: Damping causes the amplitude to decrease with time.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 16 / 29
Case 3: Forced Oscillations (Without Damping)
If an external periodic force
F (t) = Q cos(nt)
is applied to the spring and damping force is negligible, then equation (2) reduces
to
d 2x k Q
+ x = cos(nt). (9)
dt 2 m m
Writing
k Q
ω2 = and E = ,
m m
it becomes
d 2x
+ ω 2 x = E cos(nt).
dt 2
This is a linear differential equation with constant coefficients.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 17 / 29
Solution of Forced Undamped Oscillation
The auxiliary equation of
d 2x
+ ω 2 x = E cos(nt)
dt 2
is
D 2 + ω2 = 0 ⇒ D = ±iω.
Hence the complementary function is
xc = c1 cos(ωt) + c2 sin(ωt).
To obtain the particular integral, we consider two cases:
ω 6= n and ω = n.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 18 / 29
Subcase 1: When ω 6= n
For ω 6= n, we take
xp = A cos(nt).
Substituting in the equation,
−An2 cos(nt) + ω 2 A cos(nt) = E cos(nt),
which gives
E
A= .
ω 2 − n2
Hence,
E
cos(nt).
xp =
− n2 ω2
Therefore, the complete solution of (9) is
E
x(t) = c1 cos(ωt) + c2 sin(ωt) + cos(nt). (10)
ω2 − n2
This may also be written as
E
x(t) = A cos(ωt + B) + cos(nt).
ω 2 − n2
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 19 / 29
Subcase 2: Resonance Case (ω = n)
When ω = n, the previous trial solution fails. We take
xp = At sin(ωt).
Differentiating and substituting, we obtain
E
A= .
2ω
Thus,
E
xp = t sin(ωt).
2ω
Hence the complete solution becomes
E
x(t) = c1 cos(ωt) + c2 sin(ωt) + t sin(ωt). (11)
2ω
Since the amplitude increases with time, this phenomenon is called resonance.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 20 / 29
Remark on Resonance
In the resonance case (ω = n), the solution contains the term
E
t sin(ωt),
2ω
whose amplitude increases as t increases.
Thus the oscillations become very large when the forcing frequency equals the
natural frequency.
Note: In practice, resonance is controlled because damping is always present in
real systems.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 21 / 29
Graph: Forced Oscillation and Resonance
x(t)
Resonance
t
0
Amplitude increases with time when ω = n.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 22 / 29
Summary: Forced Undamped Oscillations
The forced undamped equation is
d 2x
+ ω 2 x = E cos(nt).
dt 2
Case 1: ω 6= n
E
x(t) = c1 cos(ωt) + c2 sin(ωt) + cos(nt).
ω 2 − n2
Case 2: ω = n (Resonance)
E
x(t) = c1 cos(ωt) + c2 sin(ωt) + t sin(ωt).
2ω
Amplitude increases with time.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 23 / 29
Forced Oscillations (With Damping)
If an external force
F (t) = Q cos(nt)
is applied to the system and damping force is present, then equation (2) reduces to
d 2x c dx k Q
+ + x = cos(nt).
dt 2 m dt m m
Writing
c k Q
= 2p, = ω2, = E,
m m m
it becomes
d 2x dx
+ 2p + ω 2 x = E cos(nt). (13)
dt 2 dt
This is called the forced damped oscillation equation.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 24 / 29
Complementary Function
The auxiliary equation of
d 2x dx
2
+ 2p + ω 2 x = E cos(nt)
dt dt
is
D 2 + 2pD + ω 2 = 0.
Hence, p
D = −p ± p2 − ω2.
Therefore, the complementary function is
h √ √ i
2 2 2 2
xc = e −pt c1 e p −ω t + c2 e − p −ω t .
This represents the free oscillations of the system which die out as t → ∞ due to
the factor e −pt .
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 25 / 29
Particular Integral
The particular integral is given by
E
xp = cos(nt).
D2 + 2pD + ω 2
Using standard results, we obtain
2
(ω − n2 ) cos(nt) + 2pn sin(nt)
xp = E .
(ω 2 − n2 )2 + 4p 2 n2
Let
p 2pn
h= (ω 2 − n2 )2 + 4p 2 n2 , φ = tan−1 .
ω 2 − n2
Then
E
xp = cos(nt − φ).
h
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 26 / 29
Complete Solution of Forced Damped Oscillation
Thus the complete solution of equation (13) is
x(t) = xc + xp ,
i.e., h √2 2 √ 2 2 i E
x(t) = e −pt c1 e p −ω t + c2 e − p −ω t + cos(nt − φ).
h
where
p 2pn
h = (ω 2 − n2 )2 + 4p 2 n2 , φ = tan −1
.
ω 2 − n2
As t → ∞, the complementary part dies out and the motion approaches the
steady forced oscillation.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 27 / 29
Amplitude and Period of Forced Damped Oscillations
The particular integral represents the forced oscillations of the system.
The steady-state solution is
E
x(t) = cos(nt − φ),
h
where p
h= (ω 2 − n2 )2 + 4p 2 n2 .
Hence the amplitude of forced oscillations is
E E
A= = p .
h (ω − n2 )2 + 4p 2 n2
2
The period of forced oscillations is
2π
T = ,
n
which is the same as that of the imposed force.
Thus as t increases, free oscillations die out and forced oscillations remain, giving
the steady state motion.
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 28 / 29
Forced Damped Oscillation: Transient and Steady State
x(t)
Transient part dies out
Steady-state oscillation
Note: As t → ∞, transient oscillations vanish and the system oscillates with
Dr. Bandaru Ravikumar (VIT-AP University) Applications of Linear Differential Equations 29 / 29