Double Pendulum - Dif Eq
Double Pendulum - Dif Eq
Double Pendulum - Dif Eq
211
Equations of Motion Assuming:
(3.122)
(3.123)
212
Steps for obtaining the correct differential equations of
motion:
(3.124)
214
Figure 3.48 a. Two-mass, linear vibration system with motion
of the left-hand support. b. Free-body diagram for assumed
motion .
(3.125)
215
In matrix notation
(3.126)
216
Connection with Dampers
(3.128)
217
In Matrix format
(3.129)
218
Base Excitation via Right-hand wall motion
219
Matrix Statement
Base excitation causes the additional forcing functions on the
right. The stiffness and damping matrices should always be
symmetric. If they are not, you have made a mistake.
(3.131)
220
Developing the Equations of Motion for a Double Pendulum
(3.131)
221
Simple -pendulum equations of motion,
(3.82)
(3.132)
(3.133b)
222
(3.133c)
(3.134)
(3.135)
223
This the second of the two required differential equations. In
matrix format the model is
(3.137)
224
We have symmetricized the inertia matrix and now have a
diagonal stiffness matrix. The inertia matrix couples these two
degrees of freedom.
225
Lecture 15. EIGENANALYSIS FOR 2DOF VIBRATION
EXAMPLES
226 227
(3.145)
(3.142)
(3.144)
228 229
two unknowns. Setting gives: Figure 3.52 illustrates the two eigenvectors.
(3.146)
(3.148b)
The matrix of eigenvectors is
(3.147)
where the right vector is the vector of modal coordinates.
230 231
Substituting from Eqs.(3.148) into Eq.(3.124) gives The modal mass matrix is diagonal, with the first and
(3.149) second modal masses defined by .
Premultiplying Eq.(3.149) by the transpose of gives We want to normalize the eigenvectors with respect to the
mass matrix such that the modal mass matrix reduces to
(3.150) the identity matrix [ I ]. The modal-mass coefficient for the jth
mode is defined by . Dividing the jth
We can now show by substitution (for this example problem)
eigenvector by will yield an eigenvector with a modal
that
mass equal to 1, yielding
(3.148) (3.150)
(3.151)
(3.149) You may want to repeat calculations for this set of eigenvectors
to confirm that the modal mass matrix is now the identity
matrix. Proceeding with this normalized version of the
eigenvector matrix to verify that the modal stiffness matrix is
diagonal yields
232 233
Modal Units
Given that , and ,
the units for an entry in normalized eigenvector matrix is
. Hence, for the SI system, the eigenvector units are
; for the USA standard unit system, the units are
. From the coordinate transformation ,
The normalized matrix of eigenvectors yields a diagonalized the units for a modal coordinate is . For the SI
modal stiffness matrix [ Kq ]; moreover, the diagonal entries are
and USA standard systems, the appropriate units are,
the eigenvalues defined in Eq.(3.142); i.e.,
respectively, and . Looking at the first of
Eq.(3,154), a dimensional analysis yields
(3.152)
234 235
Lessons:
a. Vibration problems can have multiple degrees of
freedom.
236
Lecture 16. SOLVING FOR TRANSIENT MOTION
USING MODAL COORDINATES
Free Motion
With normalized eigenvectors, the matrix version of the modal
Eq.(3.125) is Adding particular solutions corresponding to
specific right-hand forcing functions, yield the complete
solution
(3.153)
(3.154)
The homogeneous version of Eq.(3.154) Hence, the modal-coordinate initial conditions are defined by
237 238
For the prior example
(3.144)
(3.151)
239 240
shapes (or eigenvectors).
Figure 3.53 Solution from Eq.(3.157) for and for
Example. With , Eqs. (3.155)
gives , and:
1 1
0.5 0.5
q_1
q_2
0 0
-0.5 -0.5
-1
-1
0 10 20 30 40
0 10 20 30 40
t
t
x_1
0
x_2
0
-1
-0.5
-2
-1
0 10 20 30 40
0 10 20 30 40
(3.157) t
t
241 242
Modal Transient Example Problem 1. Free Undamped
Motion
( ii )
yielding
( ii )
(i)
Solve for and the reaction forces
.
Once contact is established, the system looks like the model of
frame b and is governed by the matrix differential equations of
motion
243 244
Solution. Following the procedures of the preceding examples, conditions are zero. From , the modal-
the eigenvalues and natural frequencies are: velocity initial conditions are
(iv)
(v)
245 246
q_1 versus time x_1 (m) versus time
10
0.8
8 0.7
0.6
6
0.5
0.4
q_1
x_1
0.3
2 0.2
0.1
0 0
0 0.1 0.2 0.3 0.4 0.5 0.6 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6
-2
-0.2
t (seconds)
t (seconds)
0 0.25
-0.2 0 0.1 0.2 0.3 0.4 0.5 0.6 0.2
x_2
-0.4 0.15
-0.6 0.1
-0.8 0.05
-1 0
t (seconds) -0.05 0 0.1 0.2 0.3 0.4 0.5 0.6
-0.1
t (seconds)
247 248
Do the numbers seem right? We can do a quick calculation to
f_s1=k_1 (x_1 - x_2 ) versus time see if the peak force and deflection seem to be reasonable.
7000
Suppose both bodies are combined, so that one body has a mass
of . Then a conservation of
6000
energy equation to find the peak deflection is
f_s1 = k_1 (x_1 - x_2 ) (N)
5000
4000
3000
2000
1000
0
0 0.1 0.2 0.3 0.4 0.5 0.6
-1000 The predicted peak deflection for is 0.35 m , which is on
t (seconds)
the order of magnitude for the estimate, but lower. We would
expect the correct number to be lower, because two masses with
f_s2 = k_2 x_2 versus time a spring between them will produce a lower collision force than
6000 a single rigid body with an equivalent mass.
5000
Note that the spring-mass system loses contact with the wall
4000 when changes sign at about . After this
f_s2 = k_2 x_2 (N)
3000 time, the right spring is disengaged, and the model becomes
2000
1000
0
0 0.1 0.2 0.3 0.4 0.5 0.6
-1000
time (seconds)
249 250
Damping and Modal Damping Factors where A1 and A2 are complex coefficients. The solution can be
The SDOF harmonic oscillator equation of motion is stated
(3.24)
For a damping matrix of this particular (and very unlikely)
form, the modal damping matrix is defined by
is called the damped natural frequency. The
homogeneous solution looks like
(3.25)
251 252
where and are the identity matrix and the diagonal Transient Modal Example Problem 2, Free-Motion with
matrix of eigenvectors, respectively. With this damping-matrix Modal Damping
format, an n-degree-of-freedom vibration problem will have
modal differential equations of the form:
253 254
10% for each mode; i.e., . Hence
(3.27)
The transformation to obtain the physical coordinates remains
unchanged as
For the initial conditions , the constants A,
B are solved from , and from
The solution is
255 256
q_1 versus time
x_1 (m) versus time
9
8 0.7
7
0.6
6
0.5
5
0.4
q_1
x_1
3 0.3
2 0.2
1 0.1
0
0
-1 0 0.1 0.2 0.3 0.4 0.5 0.6
0 0.1 0.2 0.3 0.4 0.5 0.6
-0.1
t (seconds)
t (seconds)
x_2
-0.2 0.15
-0.4 0.1
-0.6 0.05
0
-0.8
0 0.1 0.2 0.3 0.4 0.5 0.6
t (seconds) -0.05
t (seconds)
257 258
4820 N.
f_s1 = k_1 (x_1 - x_2) versus time
5000
f_s1 = k_1 ( x_1 - x_2) (N)
4000
3000
2000
1000
0
0 0.1 0.2 0.3 0.4 0.5 0.6
-1000
t (seconds)
6000
5000
f_s2 = k_2 x_2 (N)
4000
3000
2000
1000
0
0 0.1 0.2 0.3 0.4 0.5 0.6
-1000
t (seconds)
Ad
ding damping reduces the peak force in spring 2 from 5363 N to
259 260
Lecture 17. MORE TRANSIENT MOTION USING c. Solve for two cycles of motion for the lowest natural
MODAL COORDINATES frequency.
Example
Equation of Motion from Free-body diagrams:
(1)
Matrix Format:
Engineering-analysis tasks:
a. Draw free-body diagrams and derive the equations of motion.
261 262
Component modal differential equations:
(3)
263 264
A previous undamped model had the physical parameters: From,
(3)
yielding
265 266
Substituting into,
Further
nets
where
Similarly, the complete solution for is
267 268
The complete solution for satisfying the initial conditions
is
269
Lecture 18. FORCED HARMONIC MOTION FOR 2DOF
EXAMPLES.
(3.159)
with a harmonic-excitation force vector on the right hand side.
270 271
40
35
30
(3.160)
25
abs q_1
20
15
10
0
0 0.5 1 1.5 2 2.5 3
omega
50
0
0 0.5 1 1.5 2 2.5 3
-50
phase q_1
-100
-150
-200
omega
272 273
50
30
40
30
abs x_1
25 20
10
20 0
-10 0 0.5 1 1.5 2 2.5 3
abs q_2
15 om e ga
50
10
0
0 -150
0 0.5 1 1.5 2 2.5 3
-200
omega om ega
50
50
0 40
30
abs x_2
0 0.5 1 1.5 2 2.5 3
20
phase q_2
-50 10
0
-10 0 0.5 1 1.5 2 2.5 3
-100
om e ga
50
-150
0
-200 -100
omega
-150
Figur -200
om ega
274 275
(i) its contribution is primarily static; i.e., and
Amplitude for harmonic motion; =0.05
Fo sin(t) or Fo cos(t) , and .
12 =0.1
Amplitude factor (G)
10 =0.2
(ii) the static contribution (being proportional to ) is itself
8 =0.5
small.
6
(3.38)
where
278 279
The steady-state solutions are 10 1.4
9
1.2
8
7 1
abs q_2
abs q_1
0.8
5
0.6
4
3 0.4
2
0.2
1
0 0
0 1 2 3 4 0 1 2 3 4
om ega om ega
0 0
-20 0 1 2 3 4 0 1 2 3 4
-20
-40
where , , and
-40
-60
-60
-80
phase q_1
-80
phase q_2
-100
-100
-120
-120
-140
-140
-160
-160
-180
-180
-200
-200
om ega
om ega
280 281
The physical solution is given below. Note the dramatic Direct (non Modal) Harmonic Solution
reduction in amplitudes due to 10% damping.
Substituting the assumed steady-state solution,
4.5 , into Eq.(3.158) yields
4
3.5
3
abs x_1
2.5
2
1.5 (3.161)
1
0.5
0
0 0.5 1 1.5 2 2.5 3 3.5
omega
5
Solving for the unknowns using Cramers rule gives
4
abs x_2
1
(3.162)
0
0 0.5 1 1.5 2 2.5 3 3.5
-1
omega
282 283
Forced Harmonic Response with Damping in Multi-Degree-
of-Freedom Systems.
(3.163)
Combining these equations gives,
where is an n-dimensional vector of physical coordinates,
and is an n-dimensional vector of physical forces. (3.165)
Assumed solution
a single matrix equation in the 2-n unknowns .
Note that the phase angle defines the phase of the response with
respect to the input excitation vector that is driving the system at
. In general, with damping the eigenvalues and
Equating coefficients of on both sides of this eigenvectors are complex, and solutions based on eigen analysis
equation gives are significantly more complicated.
284 285
Lessons excitation frequency.