10.

Harmonic oscillator
10.1. Simple harmonic motion
Harmonic oscillator is an example of periodic motion, where the displacement of
a particle from the origin is a following function of time:
x  t   A cos  t    A – amplitude
  t    - phase
2 φ – phase constant
  2 f  ω – angular frequency
T
T – period (time for one cycle), unit: s
f – frequency (No of oscillations per second), unit: hertz (Hz)

Examples of harmonic oscillators:

A
• spring-mass system,
• simple pendulum,
• physical pendulum,
• torsion pendulum
• electrical circuit with capacitance and inductance
1
 Example
Oscillations of a particle are given by equation:
 
x  6,0 cos  3t   where: [x] = m, [t] = s
 3

What is:
a) the displacement of a particle at t = 2.0 s
  
x  2   6,0 cos 6    6,0 cos  3,0 m
 3 3
b) the velocity at t = 2.0 s
dx  
v  3  6,0 sin 3t  
dt  3
Figure from HRW,2
   m
v  3  18 sin 6    18 sin  9 3
 3 3 s (a) The displacemnt x(t),
c) the period of oscillations (b) velocity v(t) = dx/dt and
(c) acceleration a(t) = dv/dt
2 2 2 in a simple harmonic motion.
T   s
 3 3 Phase angle φ = 0.
d) the frequency of oscillations
3
f  1 / T  Hz
2 2
 10.2. Spring – mass oscillator
The spring oscillator consists of a spring with spring constant k and an attached
mass m.
An alternative definition of a simple harmonic motion:
stretched
spring
This is the motion of a particle under the influence of
a force proportional the displacement but opposite
in sign.
The force can be written as
compressed F  kx (10.1)
spring
According to the second Newton’s law
 kx  ma (10.2)
and taking into account that a  d 2 x / dt 2 one gets
Mass m attached to a spring d 2x
moves on a frictionless m 2  kx  0 (10.3)
dt
horizontal floor.

In this way we obtained the differential equation of motion of a simple harmonic

oscillator: d 2x k
 x0 (10.4)
dt 2 m
3
 Spring oscillator, cont.
Solution of eq.(10.4) can be easily guessed if one takes into account that the
second derivative of x(t) must be equal to x(t) with adequate coefficient and the
reverse sign. This property is characteristic of the function:
x  A cos   t    (10.5)
In order to prove this we calculate
dx
  A sin  t   
(10.5a)
dt
and next
d 2x (10.5b)
2
  2 A cos  t   
dt
Substituting (10.5) and (10.5b) into (10.4) we get
k (10.6)
  2 A cos  t     A cos  t     0
m
Therefore if  2  k / m then function (10.5) is a solution of eq.(10.4). Thus the
magnitudes of m and k determine the oscillator frequency   k / m or the period:
2 m
T   2
 k (10.7)
The constants A and φ depend on the initial conditions.
4
 10.3. The simple pendulum
A simple pendulum consists of a particle of mass m suspended from one end of an
unstretchable , masless string of length L.
There are two forces acting on m: gravitational Q and tension R
forces. The net force is
Q  Q sin   mg sin  (10.8)
s

which plays the role of a restoring force (acts opposite the

displacement to bring the mass back toward its equilibrium
position). For small angles θ we make an approximation
x
sin    
L
The equation of motion for the simple pendulum can then be
The string makes an
written as
mg d 2 x mg angle θ with the
ma   x  m 2  x0 (10.9) vertical. Q = mg is
L dt L
resolved into radial
Comparing (10.9) with (10.4) we see that the constant mg / L is Qr and tangent Qs
equivalent to k and according to (10.7) one obtains components.
m L
T  2  2
mg g 5
L
 10.4. The physical pendulum
The physical pendulum is a rigid body suspended from a fixed
pivot point O and oscillating under the influence of gravity.
The analysis of motion is similar to that for a simple pendulum
but instead of a mass point we have here the moment of inertia
I of a body and instead of restoring force – restoring torque τ.
The restoring torque for small angles can be approximated as
  mgh sin   mgh    
(10.10)
The equation of motion for angle displacements (Chapter 7),
Figure from HRW,2
taking into account (10.10), can be written as
d 2 The gravity force is
I 2   
dt (10.11) applied to the center
After some ordering eq.(10.11) takes the form of mass C placed at
a distance h from a
d 2 
  0 pivot point O.
dt 2 I (10.12)
Taking into account the formal similarity of eqs(10.12) and (10.4) one can write
expression for the period of a physical pendulum at small amplitudes, in analogy
to eq.(10.7): T  2
I
 2
I
 mgh 6
 10.5. Energy of an oscillator
The energy of the oscillating object transfers back and forth between kinetic and
potential energies.
Kinetic energy
The kinetic energy of an oscillator is associated entirely with a mass of a moving
body, then 2
1 1  dx  1
Ek  mv  m   mA cos t    
2 2

2 2  dt  2

Ek 
1
2
1

m 2 A2 cos 2  t     m 2 A2  A2 sin2  t   
2

1

Ek  m 2 A2  x 2
2
 (10.13)

where
x  A sin t   

Potential energy
The change in oscillator potential energy is associated with the work of a
conservative force (a spring force) and as was discussed in Chapter 4 is
kx 2 1 2
U  kA sin 2  t   
2 2 (10.14)

7
 Energy of an oscillator, cont.
Total energy
The total (mechanical) energy of an oscillator is a sum of kinetic and potential
energies
E  Ek  U  m   A  x  
1 kx 2
2
2 2 2

2
(10.15)
As
k
2   k  m 2
m

one obtains from (10.15)

E
1
2
  1
2
1
m  2 A 2  x 2  m 2 x 2  m 2 A 2
2
(10.16)

The total energy of an oscillator is constant,

independent of time and is proportional
to the square of an amplitude.

Every oscillating system contains an element

of springiness storing potential energy and an
element of inertia storing kinetic energy.
Potential, kinetic and total
energies of a linear oscillator 8
as functions of displacement x.
 Energy of an oscillator, cont.
Average values of energy
The average value of time dependent function x(t) is defined as follows:
t
1
x  lim x  t  dt (10.17)
t  t 
0

For the periodic function one gets

T
1
x   x  t  dt
T 0
(10.18)
The average energy of an oscillator is then equal:
T T
1
0 Ek  t dt 2 m A 0 cos  t    dt
2 2 2

Ek  
T T
2
1
m A 2  cos 2  t    d  t 
2
Ek  0
2

Taking into account that
 2 2
2
2
1 1 1
cos y  1  cos 2y  one obtains  cos y dy   dy   cos 2y dy  0 
2 2

2 0
2 0 2 0 2

9
 Average values of energy
Finally one obtains for the average kinetic energy
1
mA2 
1
Ek  2  m2 A2 (10.19)
2 4

The average potential energy is
T 2
kA2 kA2
sin  t    dt  sin t    d t 
2 0
2 2

2
U   0
T 2 / 
Taking into account that
2 2
1
sin 2 y  1  cos 2 y  one obtains
 sin ydy   cos 2 ydy  
2

2 0 0
Accordingly the average potential energy
kA2
 kA2 1
U  2   m 2 A 2
2 /  4 4 (10.20)
Thus, for the average total energy one gets
1 (10.21)
E  Ek  U  m 2 A 2  E
2
what was expected because the total energy is constant. The average potential and
kinetic energies are equal. This is not true for the damped and anharmonic
(nonlinear) oscillators. 10
 10.6. Damped harmonic oscillator
Real oscillators are always damped. The damped oscillator
shown in the figure consists of a mass m, a spring of
constant k and a vane submarged in a liquid. The liquid
exerts a damping force which in many cases is proportional
to the velocity (with opposite sign):
dx
Fb  b
dt b – damping constant (10.22)
In this case the equation of motion can be written as
dx
ma  kx  b
dt (10.23)
After rearrangement we have
d 2x dx (10.24) Figure from HRW,2
m 2 b  kx  0
dt dt
k b
Introducing the substitutions:  0
2
 2 one gets
m m
d 2x dx (10.25)
 2    0
2
x0
dt 2 dt
The solution of (10.25) for a small damping is:
x  Ae   t cos   t    (10.26)
11
2
where    0  2
 Damped harmonic oscillator, cont.
Solution (10.26) can be regarded as a cosine function with
a time dependent amplitude A t   Ae  t . Time t = τ, after
which the amplitude decreases e1/2 times is called the
average lifetime of oscillations or the time of relaxation.
A t  Ae  t
1
 1
  e  e2 for  
A t    Ae    t  
2
The angular frequency ω of the damped oscillator is less
than that of undamped oscillator ωo. For the small
damping, i.e. for ωo>> β, solution (10.26) can be The amplitude for the
x  Ae   t cos  
approximated by 0t  damped oscillator
decreases exponentially
(10.27) with time.

For the oscillator with a small damping one obtains for the average energy:
Energy losses
1 for the damped oscillator
2
E  m0 A2e  t /  (10.28)
2
The average power of losses is:
d E 11 2 2  
t
(10.29)
P t       m0 A e 
dt   2 
Therefore the average power of losses is related to the average energy as
Et  (10.29a)
Pt   12

 Damped harmonic oscillator, cont.
For the oscillating system with damping one introduces the dimentionless factor,
called quality factor Q, defined as follows:
energy stored E E (10.30)
Q  2  2 
energy lost in one cycle PT P

For low damping one obtains from the last equation substituting for power of
losses from (10.29a)
E0
Q  0
E/ (10.31)
Applying the last result to the electrical oscillator L,R,C, and introducing the
analogous quantities
1 1 L
0   
LC 2 R

one obtains for the quality factor

 0L (10.32)
Q
R

Examples of quality factors Q

resonance radio circuit several hundreds
violin string 103
microwave resonator 104 13
excited atom 107
 10.7. The driven oscillator with damping
The damped oscillator responds to a periodic driving force. In this case on the right
side of eq.(10.24) we introduce the driving force F(t)
m
d 2x
b
dx
 kx  F  t 
(10.33)
dt 2 dt
Moving
Substituting support
b k
m
 2 and m
 0
2
one obtains from (10.33)
d 2x dx F t 
2
 2
2
 0 x  (10.34)
dt dt m

Let the driving force be F(t)  F0 sin t .

Then the solution of eq.(10.34) takes the form
x  x 0 sin t    (10.35)
The system is driven by a
where the amplitude x0 is given by moving support that oscillates
F0 at an arbitrary angular
x0 

m  2  0
2 2
  4  2 2
(10.35a)
frequency ω. The natural
frequency of a freely
and the phase angle φ: oscillating system is ω0.
 2  
  arctg  2 
2  (10.35b) 14
   0 
 The driven oscillator with damping, cont.
Analysis of solution (10.35).
a) driving frequency much lower than the
natural frequency ω0 (ω << ω0)
In this case:
F0 F0
x0  
m 0
2
k

b) resonance – the condition at which

the amplitude of a displacement (or velocity,
or power) of oscillations is maximum. Amplitude of the driven oscillations
For ω = ω0 the amplitude x0 is not generally as a function of the driving force
x0 
F0 F
 0
frequency for varying damping
maximum: 2  m0 b0 (b1<b2<b3). Smaller damping gives

x0  max 
f      2  0 
2 2
 4  2 2
taller and narrower resonance peak.
when is minimum.

Thus the condition for minimum is:

df   
d
 2

 2  2  0 2  8 2  0 
2
 2  0  2  2  0

what yields: 2
rez  0  2  2  rez   0
15
 The driven oscillator with damping, cont.
The velocity of the driven damped oscillations can be calculated by differentiation
of eq.( 10.35): V  dx  x  cos t   
0
dt (10.36)
The amplitude of a velocity is then given by:
F0 F0
v 0  x0   

m   0 2

2 2
 4 
2 2 m

 2
 0 
2 2
 4 
2 2

F0
 ( 10.37 )
2
 
2

m    0   4 2

  

From (10.37) it follows that the amplitude of

a velocity is maximum exactly for ω = ω0 .
From the analogy between oscillating mechanical
system and electrical circuit: Amplitude of a current in a
x  q resonance electrical circuit for a
dx dq varying supply voltage frequency.
V 
dt
 I
dt
one obtains that the maximum
m  L
1
of electrical current amplitude
k  (resonance) is for ω = ω0.
C
b  R 16