You are on page 1of 7

CEE490b

REVIEW OF SDOF CONCEPTS


F(t) y(t)

c k

Fig. 1 SDOF Oscillator

Equation of motion describing “dynamic” equilibrium:

My&& + cy& + ky = F (t ) (1)

it is useful to examine several limiting cases:

A) FREE UNDAMPED VIBRATIONS (HOMOGENEOUS EQUATION)

i.e. F(t) = 0 and c=0

My&& + ky = 0 (2)

and assuming the motion of the mass can be described by the sum of
sinusoidal components:

y (t ) = C1 sin ω o t + C 2 cos ω o t (3)

k
where ω o = the “undamped” natural frequency
M

Constants C1 and C2 are determined from initial conditions (e.g. if the mass is
held at a given initial displacement and released, C1 = 0 and C2 = yo )

the motion can also be written in terms of an initial displacement and a phase-
shift:

y (t ) = y o sin(ω o t + φ ) (4)

Page 1 of 7
CEE490b

where the initial displacement y o , is:

y o = (C12 + C 22 ) (5a)

and the phase-shift φ , is:

φ = tan −1 (C 2 / C1 ) (5b)

B) FREE DAMPED VIBRATIONS (HOMOGENEOUS EQUATIONS)

My&& + cy& + ky = 0 divide by M

c k
y&& + y& + y = 0 (6)
M M

and introducing; ω o2 = k / M and 2α = c / M

y&& + 2αy& + ω o2 y = 0 (7)

a Particular Solution to this differential equation is:

y = Ce pt so y& = Cpe pt and y&& = Cp 2 e pt

substituting, we now get a quadratic equation in the variable, p:

p 2 + 2αp + ω 2 = 0
− 2α ± 4α 2 − 4ω 2
p1,2 = (8abc)
2
p1,2 = −α ± α 2 − ω 2

defining the damping ratio as: D = α / ω = ς , then:

p1,2 = −α ± ω o D 2 − 1 (9)

therefore the solution (magnitude of the motion), depends on the magnitude


of the damping D, relative to 1. There are 3 possible situations:

D is less than 1 (termed subcritical damping)


D is equal to 1 (termed critical damping), or
D is greater than 1 (termed supercritical damping)

Page 2 of 7
CEE490b

1
Sub-critical
0.8 Critical
0.6 Super-critical
Peak
0.4

0.2
y(t)

0
0 2 4 6 8 10
-0.2

-0.4
-0.6
-0.8

-1
time, t

Fig. 2. Free Damped Oscillations (f = 1.0 Hz, D = 0.032, critical and


supercritical)

Only the Subcritical case with D < 1, results in oscilliatory response. The other two
cases are not relevant to civil engineering.

For D < 1

The solution of free vibrations becomes:

y (t ) = y oe −αt sin(ω o' t + φo ) (10)

where ω o′ = ω o 1 − D 2 , the damped natural frequency


(recall that the undamped natural frequency, ω o = k / M )

For most civil engineering applications, the structural damping is much less than 1,
typically of the order of 0.001 (for a bridge cable) to 0.05 (for cracked concrete) so
for practical purposes, ω o = ω o′

There are two important cases that arise with free damped vibrations:
Page 3 of 7
CEE490b

i) when the initial velocity is zero, but with a finite initial amplitude
ii) when the initial amplitude is zero, but with a finite initial velocity

i) y o = A and y& o = 0

y (t ) = Ae −αt cos ω o′ t
(11)
≈ Ae −αt cos ω o t

The structural damping of a system (e.g. chimney, flagpole, bridge) can be


estimated by utilizing this relationship. Often, a structure can be given an initial
displacement, A, then released and the subsequent decay with amplitude recorded.

The decrease in amplitude between successive cycles of vibration is dependent on


the damping (recall that D = α / ω ). The amplitude y(t) for multiples of the period of
vibration N* (2π / ω o ) at the same point in the oscillation, will yield a straight line when
AN +1 / A is plotted vs. N

12

10
N = 5.00 Ln(A(N)/A(0))
8
N

0
1 10
A(N)/A(0)

Fig. 3. Fit to Decay in Peak Amplitude Ratio vs. Cycle Number


⎛A ⎞ 1
From the above fit, the structural damping would be: δ = − ln⎜⎜ N +1 ⎟⎟ ≅ 2πD = = 0.2
⎝ AN ⎠ 5
0 .2
(logarithmic decrement) or D= = 0.032 = 3.2%

Page 4 of 7
CEE490b

ii) y o = 0 and y& o = V

V −αt V −αt
y (t ) = e sin ω o′ t ≅ e sin ω o t (12)
ω o′ ωo

An example of this type of situation would be with a body initially at rest and
experience an impact with another body.

C) RESPONSE TO HARMONIC EXCITATION

My&& + cy& + ky = Fo cos ωt (13)

The steady-state response (once the influence of the initial conditions is lost) is:

Fo
η cos(ωt + φ )
y (t ) =
K
(14)
F
where o = y st , the static deflection under force, Fo
K

η= 1
1/ 2
, the dynamic amplification factor, or
the mechanical admittance
⎡⎛ ω 2 ⎞ 2 ω2 ⎤
⎢⎜⎜1 − 2 ⎟⎟ + 4D 2 2 ⎥ (15a)
⎢⎣⎝ ω o ⎠ ωo ⎥

⎡ 2Dω / ωo ⎤ , the phase angle between the excitation and


φ = tan −1
⎢ 2⎥
the response (15b)
⎣⎢ (1 − ω / ω o ) ⎦⎥

The complete solution is the summation of the steady-state response and the
solution of the “Homogeneous” equation ( My&& + cy& + ky = 0 ), when evaluated with
specific initial conditions.

y (t ) = y stη cos(ωt + φ ) + y o e −αt sin(ω o' t + φ o ) (16)

y(t) = Steady-state Solution + Transient Solution

the transient solution “damps” out or disappears when the initial disturbance
is past.

Page 5 of 7
CEE490b

STEADY-STATE RESONANT VIBRATIONS

At resonance, the excitation frequency, ω is equal to the natural frequency, ω o , and


the dynamic amplification factor becomes:

1/ 2
⎛ 1 ⎞
η=⎜ 2⎟ = 1
⎜ ⎟ (17a)
⎝ 4D ⎠ 2D

The phase angle becomes:

⎡ 2D ⎤ π
φ = tan −1 o
⎢ ⎥ = −90 = − 2 (17b)
⎣ 0 ⎦

The response becomes:

⎛ 1 ⎞ π
y (t ) = y st ⎜ ⎟ cos(ωt − ) (18)
⎝ 2D ⎠ 2
The maximum response is:

1
y (t )max = y st (19)
2D

Page 6 of 7
CEE490b

a) Constant Force Excitation


P (t ) = Po cos ωt and y (t ) =
Po
k
η cos(ωt + φ )

ω2 y
b) Quadratic Excitation, where η ′ =η 2 =
ω o me e / m
m eω 2 m e ω2
P (t ) = me eω 2 cos ωt and y (t ) = e
k
η cos(ωt + φ ) = e
m ωo2
η cos(ωt + φ )
Fig. 4. Dimensionless response to Harmonic Loads
(Dynamic Amplitudes and Phase Shift)

Page 7 of 7

You might also like