Lecture 9

Free vibrations of systems with two

degrees of freedom (I)

9.1 Examples of systems with two degree of freedom

9.2 Equation of motion of an undamped system for free
vibration
9.3 Solution of the equation of motion
9.4 Torsional system

1
9.1 Examples of systems with two degree of freedom
Systems that require two independent coordinates to describe their motion are called two-degree-of-
freedom systems.

2
3
4
9.2 Equation of motion of an undamped system for free
vibration
The general rule for the computation of the number of degrees of freedom can be stated as follows:

The number of equations of motion is equal to the number of degrees of freedom of a system.

k1 x1 k x2 k
2 3
m1 m2 A two-degree-of-freedom
spring-mass-damper system

𝑥𝑥1 , 𝑥𝑥̈ 1 𝑥𝑥2 , 𝑥𝑥̈ 2

Free-body
m1 m2 diagrams
k1 x1 k2(x2-x1) -k3x2
Spring k1 under tension for +𝑥𝑥1 Spring k2 under tension for + 𝑥𝑥2 − 𝑥𝑥1 Spring k2 under compression for +𝑥𝑥2 5
9.3 Solution of the equation of motion
Analyzing free-body diagrams and using Newton’s second law of motion, we can write

𝑚𝑚1 𝑥𝑥̈ 1 = −k1𝑥𝑥1 + k2 𝑥𝑥2 − 𝑥𝑥1 𝑚𝑚1 𝑥𝑥̈ 1 + k1 + k2 𝑥𝑥1 − k2𝑥𝑥2 = 0

� � (∗)
𝑚𝑚2 𝑥𝑥̈ 2 = −k2 𝑥𝑥2 − 𝑥𝑥1 + −k3𝑥𝑥2 𝑚𝑚2 𝑥𝑥̈ 2 − k2𝑥𝑥1 + k2 + k3 𝑥𝑥2 = 0

Assuming that the harmonic motion of 𝑚𝑚1 and 𝑚𝑚2 has the same frequency 𝜔𝜔 and the same phase
angle 𝜙𝜙, then, the solutions of the above equations are:

𝑥𝑥1 (𝑡𝑡) = 𝑋𝑋1 cos ω𝑡𝑡 + φ

𝑥𝑥2 (𝑡𝑡) = 𝑋𝑋2 cos ω𝑡𝑡 + φ (∗∗)

where 𝑋𝑋1 and 𝑋𝑋2 are constants that denote the maximum amplitudes of 𝑥𝑥1 (𝑡𝑡) and 𝑥𝑥2 (𝑡𝑡)

Substituting Eqs. (**) into Eqs. (*) and differentiating twice 𝑥𝑥1 (𝑡𝑡), 𝑥𝑥2 𝑡𝑡 , we obtain

(−𝑚𝑚1 𝜔𝜔2 + k1 + k2)𝑋𝑋1 − k2𝑋𝑋2 cos ω𝑡𝑡 + φ = 0

�
−k2𝑋𝑋1 + −𝑚𝑚2 𝜔𝜔2 + k2 + k3 𝑋𝑋2 cos ω𝑡𝑡 + φ = 0

Since the above equation must be satisfied for all values of the time t, the terms in the squared
brackets must equal zero. Hence, 6
 −𝑚𝑚1 𝜔𝜔2 + k1 + k2 𝑋𝑋1 − k2𝑋𝑋2 = 0
� (∗∗∗)
−k2𝑋𝑋1 + −𝑚𝑚2 𝜔𝜔2 + k2 + k3 𝑋𝑋2 = 0

Eqs. (***) is satisfied by the trivial solution 𝑋𝑋1 = 𝑋𝑋2 = 0, which implies that there is no vibration.

For a nontrivial solution: a system of linear homogeneous eqs. has a solution different from the
trivial solution if and only the determinant 𝐷𝐷 = 0

−𝑚𝑚1 𝜔𝜔2 + k1 + k2 −k2

𝐷𝐷 = det =0
−k2 −𝑚𝑚2 𝜔𝜔2 + k2 + k3

from linear algebra

Then,

𝑚𝑚1 𝑚𝑚2 𝜔𝜔4 − k1 + k2 𝑚𝑚2 + k2 + k3 𝑚𝑚1 𝜔𝜔2 + k1 + k2 k2 + k3 − 𝑘𝑘22 = 0

frequency (characteristic) equation

It can be seen that the frequency equation is a quadratic equation for 𝜔𝜔2 . Thus, the roots are: 7
 1 k1 + k2 𝑚𝑚2 + k2 + k3 𝑚𝑚1 −𝑏𝑏± 𝑏𝑏2 − 4𝑎𝑎𝑎𝑎
𝜔𝜔12 , 𝜔𝜔22 = ∓ 𝑥𝑥 2 =
2 𝑚𝑚1 𝑚𝑚2 2𝑎𝑎

1
2 2
1 k1 + k2 𝑚𝑚2 + k2 + k3 𝑚𝑚1 k1 + k2 k2 + k3 − 𝑘𝑘22
∓ −4
2 𝑚𝑚1 𝑚𝑚2 𝑚𝑚1 𝑚𝑚2

where 𝜔𝜔1 and 𝜔𝜔2 - the natural frequencies of the system.

Since amplitudes 𝑋𝑋1 and 𝑋𝑋2 are functions of frequencies 𝜔𝜔1 and 𝜔𝜔2 , let’s denote the values of
(1) (1) (2)
𝑋𝑋1 and 𝑋𝑋2 corresponding to 𝜔𝜔1 as 𝑋𝑋1 and 𝑋𝑋2 . Then, those corresponding to 𝜔𝜔2 are 𝑋𝑋1
(2)
and 𝑋𝑋2 .

Since Eqs. (***) are homogenous, only the ratios 𝑟𝑟1 and 𝑟𝑟2
(2)
(1)
𝑋𝑋2 𝑋𝑋2
𝑟𝑟1 = for ω12 , and 𝑟𝑟2 = (2) for ω22 can be found.
(1)
𝑋𝑋1 𝑋𝑋1

𝑋𝑋1 and 𝑋𝑋2 cannot be specified uniquely because of the singular matrix system. Only the ratio can be
specified for each 𝜔𝜔1 and 𝜔𝜔2 .
8
 −𝑚𝑚1 𝜔𝜔2 + k1 + k2 𝑋𝑋1 − k2𝑋𝑋2 = 0
From Eqs. (***), � (∗∗∗)
−k2𝑋𝑋1 + −𝑚𝑚2 𝜔𝜔2 + k2 + k3 𝑋𝑋2 = 0
(1)
𝑋𝑋2 −𝑚𝑚1 𝜔𝜔12 + k1 + k2 k2
𝑟𝑟1 = = =
(1)
𝑋𝑋1 k2 −𝑚𝑚2 𝜔𝜔12 + k2 + k3

(2)
𝑋𝑋2 −𝑚𝑚1 𝜔𝜔22 + k1 + k2 k2
𝑟𝑟2 = = =
(2)
𝑋𝑋1 k2 −𝑚𝑚2 𝜔𝜔22 + k2 + k3

The amplitudes of free-vibration of the multi-degree-of-freedom system at one of the natural

frequencies is called a natural mode (normal mode) of vibration.

Vibration of two-degrees-of-freedom system has:

• two natural frequencies (𝜔𝜔1 and 𝜔𝜔2 )
• two natural modes of vibration (𝑋𝑋 (1) and 𝑋𝑋 (2) )

The natural modes of vibration 𝑋𝑋 (1) and 𝑋𝑋 (2) corresponding to 𝜔𝜔12 and 𝜔𝜔22 can be expressed as:

← ω 12
(1) (1)
𝑋𝑋1 𝑋𝑋1
𝑋𝑋 (1) = (1)
= (1)
𝑋𝑋2 𝑟𝑟1 𝑋𝑋1 The vectors 𝑋𝑋 (1) and 𝑋𝑋 (2) are
(2)
𝑋𝑋1 𝑋𝑋1
(2) the modal vectors of the system.
𝑋𝑋 (2) = =
(2)
𝑋𝑋2
(2)
𝑟𝑟2 𝑋𝑋1 ← ω 22 9
The free-vibration solution

(1) (1)
𝑥𝑥1 (𝑡𝑡) 𝑋𝑋1 cos 𝜔𝜔1 𝑡𝑡 + 𝜙𝜙1
𝑥𝑥⃗ (1) (𝑡𝑡) = (1)
= (1) First mode of vibration
𝑥𝑥2 (𝑡𝑡) 𝑟𝑟1 𝑋𝑋1 cos 𝜔𝜔1 𝑡𝑡 + 𝜙𝜙1

(2) (2)
𝑥𝑥1 (𝑡𝑡) 𝑋𝑋1 cos 𝜔𝜔2 𝑡𝑡 + 𝜙𝜙2
𝑥𝑥⃗ (2) (𝑡𝑡) = (2)
= (2)
Second mode of vibration
𝑥𝑥2 (𝑡𝑡) 𝑟𝑟2 𝑋𝑋1 cos 𝜔𝜔2 𝑡𝑡 + 𝜙𝜙2

(1) (2)
where the constants 𝑋𝑋1 , 𝑋𝑋1 , 𝜙𝜙1 , 𝜙𝜙2 can be determined by initial conditions.

The resulting free vibration motion can be obtained by a linear superposition of the two normal
(natural) modes of vibration
1 2
𝑥𝑥⃗ 𝑡𝑡 = 𝐶𝐶1 𝑥𝑥⃗ 𝑡𝑡 + 𝐶𝐶2 𝑥𝑥⃗ 𝑡𝑡
where 𝐶𝐶1 and 𝐶𝐶2 are the constants.

(1) (2)
Since 𝑥𝑥⃗ (1) (𝑡𝑡) and 𝑥𝑥⃗ (2) (𝑡𝑡) already have the unknown constants 𝑋𝑋1 , 𝑋𝑋1 , we can set 𝐶𝐶1 = 𝐶𝐶2 = 1.

Then, the components of the vector 𝑥𝑥⃗ 𝑡𝑡 can be expressed as

10
 1 2 (1) (2)
𝑥𝑥1 𝑡𝑡 = 𝑥𝑥1 𝑡𝑡 + 𝑥𝑥1 𝑡𝑡 = 𝑋𝑋1 cos 𝜔𝜔1 𝑡𝑡 + 𝜙𝜙1 + 𝑋𝑋1 cos 𝜔𝜔2 𝑡𝑡 + 𝜙𝜙2
(IV)
1 2 (1) (2)
𝑥𝑥2 𝑡𝑡 = 𝑥𝑥2 𝑡𝑡 + 𝑥𝑥2 𝑡𝑡 = 𝑟𝑟1 𝑋𝑋1 cos 𝜔𝜔1 𝑡𝑡 + 𝜙𝜙1 + 𝑟𝑟2 𝑋𝑋1 cos 𝜔𝜔2 𝑡𝑡 + 𝜙𝜙2

Initial conditions (two initial conditions for each mass):

𝑥𝑥1 𝑡𝑡 = 0 = 𝑥𝑥1 0 , 𝑥𝑥̇ 1 𝑡𝑡 = 0 = 𝑥𝑥̇ 1 0

(V)
𝑥𝑥2 𝑡𝑡 = 0 = 𝑥𝑥2 0 , 𝑥𝑥̇ 2 𝑡𝑡 = 0 = 𝑥𝑥̇ 2 0

After substitution of Eqs. (V) into Eqs. (IV), we obtain

(1) (2)
1 𝑥𝑥1 0 = 𝑋𝑋1 cos 𝜙𝜙1 + 𝑋𝑋1 cos 𝜙𝜙2
(1) (2)
2 𝑥𝑥̇ 1 0 = −𝜔𝜔1 𝑋𝑋1 sin 𝜙𝜙1 − 𝜔𝜔2 𝑋𝑋1 sin 𝜙𝜙2 (VI)
(1) (2)
3 𝑥𝑥2 0 = 𝑟𝑟1 𝑋𝑋1 cos 𝜙𝜙1 + 𝑟𝑟2 𝑋𝑋1 cos 𝜙𝜙2
(1) (2)
4 𝑥𝑥̇ 2 0 = −𝜔𝜔1 𝑟𝑟1 𝑋𝑋1 sin 𝜙𝜙1 − 𝜔𝜔2 𝑟𝑟2 𝑋𝑋1 sin 𝜙𝜙2
11
Eq. (VI) can be regarded as a system of algebraic equations with 4 unknowns:
(1) (2) (1) (2)
𝑋𝑋1 cos 𝜙𝜙1 , 𝑋𝑋1 cos 𝜙𝜙2 , 𝑋𝑋1 sin 𝜙𝜙1 , and 𝑋𝑋1 sin 𝜙𝜙2

The solution of Eq. (VI) can be found by solving 4 Eqs. (VI) in pairs: (1st and 3rd) (2nd and 4th)

(1)
(2) (2) 𝑥𝑥2 0 − 𝑟𝑟1 𝑋𝑋1 cos 𝜙𝜙1
For example, expressing 𝑋𝑋1 cos 𝜙𝜙2 from the 3rd equation 𝑋𝑋1 cos 𝜙𝜙2 =
𝑟𝑟2
and substituting it into the 1st, we get
(1)
(1) 𝑥𝑥2 0 − 𝑟𝑟1 𝑋𝑋1 cos 𝜙𝜙1
𝑥𝑥1 0 = 𝑋𝑋1 cos 𝜙𝜙1 +
𝑟𝑟2
(1)
Then, solving it with respect to 𝑋𝑋1 cos 𝜙𝜙1 , we obtain
(1)
(1) 𝑥𝑥2 0 𝑟𝑟1 𝑋𝑋1 cos 𝜙𝜙1
𝑥𝑥1 0 = 𝑋𝑋1 cos 𝜙𝜙1 + −
𝑟𝑟2 𝑟𝑟2
𝑥𝑥2 0
𝑥𝑥1 0 − 𝑟𝑟2
(1)
𝑋𝑋1 cos 𝜙𝜙1 = 𝑟𝑟
1 − 𝑟𝑟1
2

(1) 𝑟𝑟2 𝑥𝑥1 0 − 𝑥𝑥2 0

𝑋𝑋1 cos 𝜙𝜙1 =
𝑟𝑟2 − 𝑟𝑟1 (2)
to obtain 𝑋𝑋1 cos 𝜙𝜙2
12
 (2) −𝑟𝑟1 𝑥𝑥1 0 + 𝑥𝑥2 0
𝑋𝑋1 cos 𝜙𝜙2 =
𝑟𝑟2 − 𝑟𝑟1

Similarly, solving 2nd and 4th eqs. together:

(1) −𝑟𝑟2 𝑥𝑥̇ 1 0 + 𝑥𝑥̇ 2 0 (2) 𝑟𝑟1 𝑥𝑥̇ 1 0 − 𝑥𝑥̇ 2 0

𝑋𝑋1 sin 𝜙𝜙1 = 𝑋𝑋1 sin 𝜙𝜙2 = (VII)
𝜔𝜔1 𝑟𝑟2 − 𝑟𝑟1 𝜔𝜔2 𝑟𝑟2 − 𝑟𝑟1

Then, the desired constants:

(1)
2 2 𝑋𝑋1
(1) (1) (1)
𝑋𝑋1 = 𝑋𝑋1 cos 𝜙𝜙1 + 𝑋𝑋1 sin 𝜙𝜙1

𝑋𝑋1 sin 𝜙𝜙1

𝜙𝜙1

(1)
(2) (2) 2 (2) 2
𝑋𝑋1 = 𝑋𝑋1 cos 𝜙𝜙2 + 𝑋𝑋1 sin 𝜙𝜙2
(1)
𝑋𝑋1 cos 𝜙𝜙1
(1)
𝑋𝑋1 sin 𝜙𝜙1 (VIII)
𝜙𝜙1 = tan−1 (1)
𝑋𝑋1 cos 𝜙𝜙1
(2)
𝑋𝑋1 sin 𝜙𝜙2
𝜙𝜙2 = tan−1 (2)
𝑋𝑋1 cos 𝜙𝜙2
13
Finally, substituting Eqs. (VII) into eqs. (VIII), we obtain

2
(1) 1 2 −𝑟𝑟2 𝑥𝑥̇ 1 0 + 𝑥𝑥̇ 2 0
𝑋𝑋1 = 𝑟𝑟2 𝑥𝑥1 0 − 𝑥𝑥2 0 +
𝑟𝑟2 − 𝑟𝑟1 𝜔𝜔1

2
(2) 1 2 𝑟𝑟1 𝑥𝑥̇ 1 0 − 𝑥𝑥̇ 2 0
𝑋𝑋1 = −𝑟𝑟1 𝑥𝑥1 0 + 𝑥𝑥2 0 +
𝑟𝑟2 − 𝑟𝑟1 𝜔𝜔2

−𝑟𝑟2 𝑥𝑥̇ 1 0 + 𝑥𝑥̇ 2 0

𝜙𝜙1 = tan−1
𝜔𝜔1 𝑟𝑟2 𝑥𝑥1 0 − 𝑥𝑥2 0

𝑟𝑟1 𝑥𝑥̇ 1 0 − 𝑥𝑥̇ 2 0

𝜙𝜙2 = tan−1
𝜔𝜔2 −𝑟𝑟1 𝑥𝑥1 0 + 𝑥𝑥2 0

14
 9.4 Torsional system
A torsional system undergoes forced vibration. The system consists of two discs of mass moments of
inertia 𝐽𝐽1 and 𝐽𝐽2 mounted on a shaft, where 𝑀𝑀𝑡𝑡𝑡 and 𝑀𝑀𝑡𝑡2 - the applied torques; 𝑘𝑘𝑡𝑡𝑡 , 𝑘𝑘𝑡𝑡2 , 𝑘𝑘𝑡𝑡3 - the
rotational spring constants of 3 segments of the shaft; θ1 and θ2 - the rotational degrees of freedom
of the torsional system.

The differential equations of rotational motion can

be derived as:

𝐽𝐽1 θ̈ 1 = −𝑘𝑘𝑡𝑡𝑡 θ1 + 𝑘𝑘𝑡𝑡𝑡 θ2 − θ1 + 𝑀𝑀𝑡𝑡𝑡

�
𝐽𝐽2 θ̈ 2 = −𝑘𝑘𝑡𝑡𝑡 θ2 − θ1 − 𝑘𝑘𝑡𝑡𝑡 θ2 + 𝑀𝑀𝑡𝑡𝑡

from free-body diagram

Torsional system with 2 discs mounted on a shaft *

15
* S.S. Rao. Mechanical Vibrations, 5th Edition, Pearson Education, 2011.
Rearranging the last expression, we obtain:

𝐽𝐽1 θ̈ 1 + 𝑘𝑘𝑡𝑡𝑡 + 𝑘𝑘𝑡𝑡𝑡 θ1 − 𝑘𝑘𝑡𝑡𝑡 θ2 = 𝑀𝑀𝑡𝑡𝑡

�
𝐽𝐽2 θ̈ 2 − 𝑘𝑘𝑡𝑡𝑡 θ1 + 𝑘𝑘𝑡𝑡2 + 𝑘𝑘𝑡𝑡3 θ2 = 𝑀𝑀𝑡𝑡𝑡

In the case of free-vibration:

similarity
𝐽𝐽1 θ̈ 1 + 𝑘𝑘𝑡𝑡𝑡 + 𝑘𝑘𝑡𝑡𝑡 θ1 − 𝑘𝑘𝑡𝑡𝑡 θ2 = 0 𝑚𝑚1 𝑥𝑥̈ 1 + k1 + k2 𝑥𝑥1 − k2𝑥𝑥2 = 0
� �
𝐽𝐽2 θ̈ 2 − 𝑘𝑘𝑡𝑡𝑡 θ1 + 𝑘𝑘𝑡𝑡2 + 𝑘𝑘𝑡𝑡3 θ2 = 0 𝑚𝑚2 𝑥𝑥̈ 2 − k2𝑥𝑥1 + k2 + k3 𝑥𝑥2 = 0
to a translational
motion system

Thus, the further analysis can be done by substituting θ1 , θ2 , 𝐽𝐽1 , 𝐽𝐽2 , 𝑘𝑘𝑡𝑡𝑡 , 𝑘𝑘𝑡𝑡𝑡 , and 𝑘𝑘𝑡𝑡𝑡 for 𝑥𝑥1 , 𝑥𝑥2 , 𝑚𝑚1 ,
𝑚𝑚2 , k1, k2, and k3, respectively.

16