Beruflich Dokumente
Kultur Dokumente
Volume 5
Movement Equations 5
Georges Vénizélos
Abdelkhalak El Hami
First published 2019 in Great Britain and the United States by ISTE Ltd and John Wiley & Sons, Inc.
Apart from any fair dealing for the purposes of research or private study, or criticism or review, as
permitted under the Copyright, Designs and Patents Act 1988, this publication may only be reproduced,
stored or transmitted, in any form or by any means, with the prior permission in writing of the publishers,
or in the case of reprographic reproduction in accordance with the terms and licenses issued by the
CLA. Enquiries concerning reproduction outside these terms should be sent to the publishers at the
undermentioned address:
www.iste.co.uk www.wiley.com
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Table of Notations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
Preface
This book is the fifth and last volume in the Non-deformable Solid
Mechanics series. As with the first four volumes, its content is
inspired by the mechanics teachings offered at the Conservatoire
National des Arts et Métiers (CNAM) in France.
Georges VÉNIZÉLOS
Abdelkhalak EL HAMI
August 2019
Table of Notations
M material point
t time
m( S ) mass of a solid ( S )
δ ij Kronecker symbol
ε ijk alternate symbol of 3rd order
V vector
( λ ) ≡ ( xλ yλ zλ ) basis
λ ≡ Oλ xλ yλ zλ frame
scalar product of the two vectors V
V ⋅W
and W
vector product of the two vectors V
V ∧W
and W
polar unit vector in cylindro-polar
u(α )
coordinates
polar unit vector in spherical
k (α,β )
coordinates
vector rotation of angle α around
Ru , α ≡ u α
the axis defined by the vector u
OM vectorial bipoint or vector
situation vector of the point O S in
Oλ O S
the reference frame λ
{ S ,α }O
λ
s
= αδ ( λ , S ) α d O S partial distributing torsor relative to
the variable Qα
...
...
Let:
– “upstream of (Si)” refer to the subset of solids (Sj) such that
1≤ j≤ i ;
The solid (S1), with which the frame 1 is associated, is located
with respect to the reference frame g using k1(k1 < 6) parameters
1 2 k
q , q ,..., q 1 forming the group ( q1 ) ≡ q1,q 2 ,...,q k1 .
1 1 1 1 1 1
Let us begin by recalling the two basic principles that are applied
when studying chains of solids.
= +
= {1 → 1}
1= ⋃2
{1 → 1} = { ⋃2 → 1} = { → 1} + {2 → 1}
= {1 → 1} = { → 1} + {2 → 1}
= {∆ → } + {∆ → } + {ℒ → } + {ℒ → }
= {2 → 2}
6 Movement Equations 5
{2 → 2} = { ⋃1 → 2} = { → 2} + {1 → 2}
= {2 → 2} = { → 2} + {1 → 2}
= {∆ → } + {∆ → } + {ℒ → } + {ℒ → }
={ → },
where = 1⋃2.
{ → }={ → 1⋃2} = { → 1} + { → 2}
⟹ ={ → 1} + { → 2}
= {∆ → } + {∆ → } + {ℒ → } + {ℒ → }
Set of Solids with Neither Loops Nor Branches 7
+ ={ → 1} + {2 → 1} + { → 2} + {1 → 2}
Consequently
= + ={ → 1} + {2 → 1} + { → 2} + {1 → 2}
With, in addition:
={ → 1} + { → 2}
{2 → 1} + {2 → 1} = {0}
{∆ → } + {∆ → } + {ℒ → } + {ℒ → } = {0}
{∆ → } + {∆ → } = {0}
{ℒ → } + {ℒ → } = {0}
8 Movement Equations 5
18+p-12=6+p equations.
In the specific scenario where one of the three link torsors is null,
the deficit is reduced to p equations.
( )⋂( ) = ∅
( ) = ( )⋃( )⋃( ) where ( )⋂( ) = ∅
( )⋂( ) = ∅
∗( )⇒ = {1 → 1} = { → 1} + {2 → 1}
Set of Solids with Neither Loops Nor Branches 9
∗( )⇒ = {2 → 2} = { → 2} + {1 → 2} + {3 → 2}
∗( )⇒ = {3 → 3} = { → 3} + {2 → 3}
The three solids are disjoint: given the composition of the chain
and applying what was shown in section 1.2, according to the theorem
of mutual actions, we can write that
{2 → 1} + {1 → 2} = {0}
{3 → 2} + {2 → 3} = {0}
= + + ={ → 1} + { → 2} + { → 3}
The external and the link efforts acting on each solid come under
the various laws of forces already mentioned, making it possible to
separate the right-hand terms of the torsional equations according to
these laws, applying the fundamental principle of dynamics to each
solid, giving:
= {∆ → } + {ℒ → } − {∆ → } − {ℒ → }
= {∆ → } + {∆ → } + {ℒ → } − {∆ → } − {ℒ → }
= {∆ → } + {∆ → } + {ℒ → }
where the link of ( ) with ( ) leads to the link with < >.
= ( ), = 1, … , ∈( )
1 2.
1 1
= ( ), = 1, … , ∈( )
2 2.
2 2
= ( ), = 1, … , ∈( )
3 3.
1 1
= + = +
2 2 1 1. 2.
1 1
⇒ = + =
2. 2. 1. 2.
1
⇒ = + =
2. 2. 1. 1.
hence
= +
2 2. 2.
2 1
= + +
3 3 2 1
2 1
= + +
3. 2. 1.
Set of Solids with Neither Loops Nor Branches 11
= + +
3 3. 3. 3.
( ) = ( )⋃( )⋃( )
( ): =⋯
1.
… ⨂[{∆ → } + {ℒ → } − {∆ → } − {ℒ → }]
1.
( ): =⋯
2.
… ⨂[{∆ → } + {∆ → } + {ℒ → } − {∆ → } − {ℒ → }]
2.
( ): =⋯
3.
… ⨂[{∆ → } + {∆ → } + {ℒ → }]
3.
= =
1. 2. 3.
⇒ ⨂ + + = =⋯
1. 1.
Set of Solids with Neither Loops Nor Branches 13
.... ⨂[{∆ → } + {ℒ → } + {∆ → } + {∆ → }]
1.
( ) ( ) ( ) ( )
− + − …
( ) ( )
…+ − …
( )
…= ∏ (∆ → ) + ∏ (ℒ → ) + ∏ (∆ → ) + ∏ (∆ → )
( )= ( )+ ( )+ ( )
( ) ( )
− =∏ ( → )
( )
where the only link that appears is that of ( ) with < >. If this link
is perfect, that is, if ∏ (ℒ → ) = 0, the equations above constitute
a first set of movement equations.
( ): =⋯
1.
… ⨂[{∆ → } + {ℒ → } − {∆ → } − {ℒ → }]
1.
14 Movement Equations 5
λ
(S ): =⋯
2. α
λ
… ⨂[{∆ → } + {∆ → } + {ℒ → } − {∆ → } − {ℒ → }]
2. α
λ
(S ): =⋯
3. α
λ
… ⨂[{∆ → } + {∆ → } + {ℒ → }]
3. α
where
= {0} and =
1. 2. 3.
⨂ + = ⋃ =⋯
2. 2.
.... ⨂[{∆ → } + {∆ → } + {ℒ → } + {∆ → }]
2.
( ) ( ) ( ) ( )
− + − …
( )
…= ∏ (∆ → ) + ∏ (∆ → )+ {∆ → }+
2.
{ℒ → }
2.
( ) ( )
= 0 and =0
( ) ( )
−
{∆ → }+ {ℒ → }
2. 2.
( ) (∆
→ ) = 1 {∆ → }
2.
1
= {∆ → } =∏ (∆ → )
2.
1
As = {0} and =
1. 2. 2.
we obtain
(∆ → )= {∆ → }= (∆ → )
2.
( ) ( )
− …
16 Movement Equations 5
…= ∏ (∆ → ) + ∏ (∆ → ) + ∏ (∆ → ) + ∏ (ℒ → )
(ℒ → )= (ℒ → )=0
( ): =⋯
1.
… ⨂[{∆ → } + {ℒ → } − {∆ → } − {ℒ → }]
1.
( ): =⋯
2.
… ⨂[{∆ → } + {∆ → } + {ℒ → } − {∆ → } − {ℒ → }]
2.
( ): =⋯
3.
… ⨂[{∆ → } + {∆ → } + {ℒ → }]
3.
where:
= = {0}
1. 2.
making it possible to write:
( ) ( )
− …
Set of Solids with Neither Loops Nor Branches 17
…= ∏ (∆ → )+ {∆ → }+ {ℒ → }
3. 3.
( ) ( ) ( ) ( )
− = −
{∆ → }+ {ℒ → }
3. 3.
We obtain
2
{∆ → }= {∆ → }= (∆ → )= (∆ → )
3. 3.
18 Movement Equations 5
( ) ( )
− = ( → )
with the only link terms that they contain being those relating
to {ℒ → }. If the latter is perfect, that is if ∏ (ℒ → ) = 0, these
relations are counted among the movement equations.
F
Figure 1.4. Po
oint contact of two solids
1
ℳ
2
1
.ℳ =0
2
and redduces the num
mber of degrrees of freedo
om of ( ).
20
0 Movement Equations
E 5
{ℒ } = [ ⋮
{ } = { Δ} + { ℒ }
= ( . ) + ⋀( ⋀ )
where:
1 { } 1
ℒ = ⨂ ℒ = .ℳ
2 2
1
ℒ = [( . ) + ⋀( ⋀ )]. ℳ
2
1
where . ℳ = 0 as we consider the contact to be persistent given
2
that there is no displacement of point M along the normal, hence:
1 1
ℒ = [ ⋀( ⋀ )]. ℳ = .ℳ
2 2
The configuration of this type of point contact can present several
aspects, which are explained below.
1.4.1.1. Configuration 1
When the solid ( ) is not bonded to ( ), the relation
. >0
indicates that, with the two vectors forming an acute angle, the normal
contact action exerted by ( ) on ( ) is directed from ( )
toward ( ).
1.4.1.2. Configuration 2
When the sliding velocity of ( ) on ( ) at M is null, that is,
1
ℳ =0
2
the tangential component of the link efforts is small compared to the
normal component, which is reflected by the relation
<
where:
– =| . |= .
– = [ ⋀( ⋀ )] = ( ⋀ )
– . < ( ⋀ )
1
As ℒ = , the power developed with respect to the
.ℳ
2
Galilean frame < 1 > by the link associated with the contact actions
that ( ) exerts on ( ) is null. This relation makes it possible to
determine the contact action . This then involves verifying the two
inequalities
. >0
. < ( ⋀ )
Set of Solids with Neither Loops Nor Branches 23
. = ( ⋀ )
1.4.1.3. Configuration 3
Let us now consider the case where
1
ℳ ≠0
2
a condition that leads to the three relations
(1) =
(2) 1
⋀ℳ =0
2
(3) 1
.ℳ <0
2
According to relation (2) above, the tangential contact action and
the sliding velocity at M of ( ) on ( ) are colinear. This relation is
written as
1 [
ℳ ⋀ ⋀( ⋀ )] = 0
2
1 ( 1
ℳ . ⋀ ) − ℳ . ( ⋀ )=0
2 2
1
As we have the condition ℳ . = , we deduce from the
2
preceding relation that
1 (
ℳ . ⋀ )=0
2
According to relation (3), the tangential contact action and the
sliding velocity at M of ( ) on ( ) are opposite. This relation is
written as
24 Movement Equations 5
1
⋀( ⋀ ). ℳ <0
2
1
⇒ [ − ( . ) ]. ℳ <0
2
With the same condition as before, we deduce that
1
.ℳ <0
2
That is to say that the power developed with respect to the Galilean
frame < 1 > by the link torsor associated with the contact actions at
M of ( ) and ( ) is negative.
1
Thus, when we have ℳ ≠ 0, we have the three relations
2
. <
( ⋀ )
( ⋀ ). ℳ 1 = 0
2
1
.ℳ <0
2
The first two provide the two missing scalar equations to reduce
the equation deficit to zero. Having then determined the contact
action , it remains for the inequality . > 0 to be verified.
= where >0
1.4.2.1
1. Arbitrary line contactt
wo solids occurs along the
Thee contact bettween the tw t curve’s aarc
(C).
Associated withh point M off the contactt curve (C), whose vicinnity
on this curve (Figuure 1.6) is deefined by thee elementaryy arc ℓ( ),, is
the elem
mentary forcce
( ) ℓ( )
Thee link torsorr {ℒ } between the two solids in contact has tthe
followiing reductionn elements att Q:
{ℒ } = ( ) ℓ(( )
∈( )
ℳ {ℒ } = ⋀ ( ) ℓ( )
∈( )
Figure
e 1.7. Straightt line contact
1 ( )( )+
.ℳ =0 ⇒ . ⋀ =0
2
1 ( )( )+
.ℳ =0 ⇒ . ⋀ =0
2
From the term-by-term
m difference for the last two
t expressiions, we
obbtain
. ⋀ =0
By writing = , the
t relation
. ⋀ =0
Set of Solids with Neither Loops Nor Branches 27
. ⋀ =0 ⇒ ( ⋀ ). =0 ⇒ . =0
1
.ℳ =0
2
. =0
{ℒ } = ( )
∈( )
ℳ {ℒ } = ⋀ ( ) = ⋀ ( )
∈( ) ∈( )
{ℒ } = + +
ℳ {ℒ } = +
( )= ( )
28 Movement Equations 5
{ℒ } = ( )
∈( )
ℳ {ℒ } = − ( )
∈( )
The link torsor involves only two unknowns and its reduction
elements at Q are
{ℒ } =
ℳ {ℒ } =
= 0; = 0; =0
( ) ( )
{ℒ } = ( ) ( )
∈( )
ℳ {ℒ } = ⋀ ( ) ( )
∈( )
1.4.3.2
2. Plane surrface contacct
1
.ℳ =0
2
1
.ℳ =0
2
which are summarized in a single equation:
. ⋀ =0
. ⋀( + ) =0
. [( + )⋀ ] = 0
.( − )=0
. =0
. =0
1
.ℳ =0
2
. =0
. =0
{ℒ } = ( , )
∈( )
ℳ {ℒ } = ( + )⋀ ( , )
∈( )
( , )= ( , )
{ℒ } = ( , )
∈( )
ℳ {ℒ } = ( , ) − ( , )
∈( ) ∈( )
The link torsor involves only three scalar unknowns and its
reduction elements at Q lead to:
{ℒ } =
ℳ {ℒ } = +
= 0; = 0; =0
{ℒ } = + +
ℳ {ℒ } = + +
=0
1
= 0⋮ ′
2
is accompanied by the partial distributor
1
= 0⋮
2.
Set of Solids with Neither Loops Nor Branches 33
+ =0
1
= + + ′ ⋮0
2
1 1 1
= ⋮0; = ⋮0; = ⋮0
2. 2. 2.
= = =0
1
=[ +⋮ ′ + ′ ]
2
1 1 1
= ⋮0 ; = ⋮ ; = 0⋮
2. 2. 2.
=0 and = =0
Vibration Mechanics
of Systems of Solids
( ) = ( )⋃ … ⋃ ⋃ … ⋃( )
( ) ⇔ |⃗ ⃗ ⃗
...
⇔ ⃗ ⃗ ⃗
...
( ) ⇔ |⃗ ⃗ ⃗
– mass ;
– center of inertia ;
− −
– inertia matrix ( ) = − − .
/( ⃗ ⃗ ⃗)
− −
The links between the solids are all assumed to be configurable and
holonomic, which amounts to saying that the reduction elements of
each link torsor are expressed as a function of the only k situation
parameters of the set of solids in the Galilean frame 〈 〉; these links
are expressed as a function neither of time nor of the derivatives of the
k parameters.
= +
When all of the links between the solids are thus identified, we see
the appearance of k parameters , … , … , to situate the chain of
n solids, with ≤ 6 .
= ⃗⋮ ( ⃗)
( ), ( ) ,…, ( ), ( )
ℎ
= + +⋯ +
ℎ
ℎ
, = , + , +⋯ + ℎ,
,
38 Movement Equations 5
( ) = ( ⋃ ⋃… ⋃ ⋃. . . ⋃ )⋃ = ⋃
i.e.
This expression can be applied to all the solids of the set in the
following summary form:
→ = {ℒ( → )} + {Δ( → )}
+{ℒ( → )} + {Δ( → )}
where
= {ℒ( → )} + {Δ( → )}
+{ℒ( → )} + {Δ( → )}
( )
There are 6n scalar equations for +6 unknowns, hence a
deficit of + 3 ( − 1) relations.
⨂ = ⨂ { ℒ( → )} + ⨂ℒ → …
, , ,
≠
… ⨂∑ ≠ {Δ( → )}+ ⨂Δ →
, ,
( ) ( ) ( ) ( )
− = {ℒ( → )} + {ℒ( → )} …
( )
( ) ( )
…+ {Δ( → )} + {Δ( → )}
( ) = (1) + (2) + ⋯ + ( ) + ⋯ + ( ) = ( )
( ) ( )
−
( )
= ⨂ {ℒ( → )} …
,
+ ⨂ {ℒ( → )} …
,
+ ⨂ {Δ( → )} + ⨂{Δ( → )}
, ,
( )
⨂{ℒ( → )} = {ℒ( → )}
,
Vibration Mechanics of Systems of Solids 41
( )
⨂{Δ( → )} = {Δ( → )}
,
⨂ {ℒ( → )} ⨂ {Δ( → )}
, ,
Taking the first term as an example, we can see that the following
expression appears as
⨂{ℒ( → )} + ⨂{ℒ( → )} …
, ,
= ⨂{ℒ( → )}
,
⨂ {ℒ( → )} = ⨂{ℒ( → )}
, ,
…+∑ ∑ , ⨂{ ( → )} i = 1,…, k
When the links are perfect, the partial power terms of the link
efforts disappear; we then obtain equations where the only unknowns
42 Movement Equations 5
are the k degrees of freedom of the set ( ). These are the movement
equations of this set.
( )
⨂ = =⋯
= ⨂ {ℒ( → )} + ⨂{ℒ( → )} …
+ ⨂ {Δ( → )} + ⨂{Δ( → )}
provide, particularly in the case where the links are perfect and
consequently, the powers that they put in play are null, subsidiary
movement equations, which can advantageously replace the Lagrange
or dynamic equations.
Figure
e 2.1. Solid ( ), a compone
ent of the chaiin
– spprings of natural
n lengthh ℓ and stifffness .
Figure
e 2.2. Spring Ri , a compone
ent of the chaiin
Fig
gure 2.3. Osciillatory system
m of n solids
We set: ⃗= ⃗
= 0⃗ ⋮ ′⃗
2. =
where ⃗. ⃗ = 0. The power applied by the gravity torsor is null for the
movement considered ( ) → =0;
– the torsors of the actions of the two springs that frame each solid
ℓ
→ = 1− ⃗ ⋮ 0⃗ = →
where
⃗= ⃗+ ⃗+ ⃗+ ⃗
= − + + ⃗
and
= = − −2 because − >
2
⇒ → = − − −2 −ℓ ⃗ ⋮ 0⃗
ℓ
→ = 1− ⃗ ⋮ 0⃗
= →
Vibration Mechanics of Systems of Solids 45
where
⃗= ⃗+ ⃗+ ⃗+ ⃗
= − − + − ⃗
and
= = − −2 because − >2
⇒ → = − −2 −ℓ ⃗ ⋮ 0⃗ ;
– the prismatic link torsor that assures the sliding of the solids on
the ( ⋮ ⃗) axis and that is perfect
( )
ℒ→ = 0.
=− − −2 −ℓ …
…+ − −2 −ℓ
We will now examine the case of the two end springs of the set,
and .
ℓ
{ → } = 1− ⃗ ⋮ 0⃗ = { → }
⇒{ → } = − ( − − ℓ ) ⃗ ⋮ 0⃗
ℓ ⃗′ ⋮ 0⃗ = {
{ → } = 1− → }
= ( − − −ℓ ) ⃗ ⋮ 0⃗
(1) = −( + ) + + ( −2 )
+ ℓ − ℓ
(2) = −( + ) + +2 ( − )
+ ℓ − ℓ
………………………………………………………………………
( )
= −( + )
+ +2 ( − )
+ ℓ − ℓ
……………………………………………………………………
( − 1)
= −( + )
+ +2 ( − )+ ℓ − ℓ
( )
= −( + )
+ + (2 − )
+ ℓ − ℓ
= 0, = 0, = and = 1, … ,
hence the equilibrium equations resulting from the above relations are
given as
(1 ) −( + ) + =− ( +ℓ )+ (2 + ℓ )
Vibration Mechanics of Systems of Solids 47
(2 ) −( + ) + =− (2 + ℓ ) + (2
+ℓ )
…………………………………………………………….
( ) −( + )
+ =− (2 + ℓ ) + (2
+ℓ )
…………………………………………………………….
( −1 ) −( + )
+ =− (2 + ℓ )+ (2 + ℓ )
( ) −( + )
=− − (2 + ℓ ) + (2 + ℓ )
= + ( ); = ; =
= ( , , )
= ( , , )+ ( , , )
= −( + ) +
(1 ) +( + ) − =0
(2 ) − +( + ) − =0
……………………………………………………….
( ) − + + − =0
…………………………………………………….....
( −1 ) − +( + ) −
=0
( ) − +( + ) =0
= =⋯= =⋯= = =
= =⋯= =⋯= = =
with, moreover, = .
(1 ) +2 − =0
(2 ) − +2 − =0
Vibration Mechanics of Systems of Solids 49
……………………………………….
( ) − +2 − =0
……………………………………….
( −1 ) − +2 − =0
( ) − +2 =0
ℰ = ( + )
=− ( + ); ℰ =− ( + )=−
(1) (2 − ) − =0
(2) − + (2 − ) − =0
…………………………………………..
( ) − + (2 − ) − =0
……………………………………………
( − 1) − + (2 − ) − =0
( ) − + (2 − ) =0
(1) − =0
(2) − + − =0
50 Movement Equations 5
…………………….
( ) − + − =0
……………………..
( − 1) − + − =0
( ) − + =0
= , = + ⇒ = −1= −1
By also setting
We deduce therefrom
2 cos sin sin 2
= = 2 cos = =
sin sin
= − = − 2 = 8cos − 4 cos
4sin cos cos 2
= 4 cos (2 cos − 1) =
sin
2sin 2 cos 2 sin 4
= =
sin sin
We then have at our disposal the first elements of a seemingly
recurrent sequence with
( )= = 1; ( )=
( )= ; ( )=
sin( − 1)
( )=
sin
Continuing this reasoning by recurrence according to the law
( )= ( )− ( )
we have
sin(p − 1) sin( − 2)
= 2 cos −
sin sin
sin
=
sin
sin sin( − 1)
( ) ⇒ − = 2 cos −
sin sin
= sin( + 1) = 0
( + 1) = ⇒ =
+1
with = 1, … , ; ≠2; ≠ 0 and ( + 1)
( ) ( ) sin
= cos[ ( + )] = cos[ ( + )]
sin
( )
which introduces 3n unknowns: , , where = 1, … .
=2 sin for = 1, …
( )
Vibration Mechanics of Systems of Solids 53
( )
The remaining 2n unknowns, and , for = 1, … , are
determined from the 2n initial conditions of the problem.
( ∗) = − + (2 + )
( ∗)
− =− (2 + )
( ∗)
− + =0
""" − + =0
( ∗) − +( − ) =0
""" − ( − 1) =0
⇒ ( − 1) − =0
sin sin( − 1)
(2 cos − 1) − =0
sin sin
⇒ (2 cos − 1) sin − sin( − 1) = 0
1
sin siin = cos cos ⇒ cos( + ) = 0
2 2 2
1 (2
2 − 1)
+ = (2 − 1) ⇒ =
2 2 2 +1
heence, =2 sin .
= ; = , …, = ,…
F
Figure 2.4. Co
onfiguration fo
or passing to the
t limit
Vibration Mechanics of Systems of Solids 55
( + 1)
( )− ( )= , − ,
+1 +1
""" = + , − ,
+1
Likewise, if we consider
( − 1)
( )− ( )= , − ,
+1 +1
""" = − , − ,
+1
when →0
( + , )= ( , )+ ( , )+ ( , )[1 + ( , , )]
2
where ( , , ) → 0 when →0
( − , )= ( , )− ( , )+ ( , )[1 + (− , , )]
where (− , , ) → 0 when → 0.
56 Movement Equations 5
Consequently,
( )− ( )+ ( )− ( )…
= , 2+ , , + − , ,
2
2 2
2 2+ , , + − , ,
⇒ = 0 2 2
,
( + 1) 2
= ; =
[ ]= , [ ]= , =
[ ]= , [ ]= = ,⇒ [ ]=
( )
Moreover, = = ⇒ [ ]=
=
Vibration Mechanics of Systems of Solids 57
with = = = [ ] ⇒ = ∼
velocity.
− =0
= − + +
= ( ) ( ) ⇒ = ( ) ( ), = ( )
As = ⇒ =
= = ⇒ − = 0, =− =0
58 Movement Equations 5
+ =0 ⇒ = cos + sin
where:
=2 sin with =
( )
+1
=2 sin
2( + 1)
For the transition from the discrete set to the continuous set, we
have → ∞, consequently
→0 ⇒ sin →
2( + 1) 2( + 1) 2( + 1)
⇒ ∼ when →∞
=
Vibration Mechanics of Systems of Solids 59
( , )= − + +
−
( , )= cos + …
2
+
+ sin + …
2
+
+ cos − …
2
−
− sin − …
2
yet
⇒ Dsin =0
sin =0 ⇒ = =
– At instant t = 0, we set
( , 0) = ( ) and ( , 0) = ( )
as
so we can identify
( )= sin
( )= sin
1
= 2sin sin
2
1
= cos(k − n) −cos(k + n)
2
Vibration Mechanics of Systems of Solids 61
1 sin(k − n) sin(k + n)
I= − =0
2 (k − n) (k + n)
1 (k + n) 1 sin(k + n)
= 1− = − =
2 L 2 (k + n) 2
= sin sin =
2
= =
2 2
""" = =
2 2
2
= ( )sin
62 Movement Equations 5
2
= ( )sin
2
( , )= sin cos ( )sin …
2
+ sin ( )sin
2.4. Exercises
O
(R)
We use (S) to refer to the solid consisting of the set of the disk, two
rods and two axes
We set: ⃗ = 2ℓ ⃗, ⃗=ℓ ⃗
( ⃗, ⃗) = ( ⃗, ⃗) =
{ → } =[ ⃗ ⋮ ⃗ ].
⃗= ⃗ ( > 0)
Vibration Mechanics of Systems of Solids 65
0 0
Its inertia matrix is [ ]( ⃗, ⃗, ⃗) = 0 − .
0 −
We set: ( ) = ( ⃗, ⃗) measured on ⃗.
2) The link efforts between the disk and the solid ( ) are
represented by:
{ → } ={ ⃗+ ⃗ ⋮ ⃗+ ⃗}
={ ⃗ ⋮ ⃗ }
⃗ = ⃗
⃗ = 2ℓγ'cos ⃗
⃗ ∧⃗ ⃗ ∧ 2ℓ cos ⃗
⃗= = = 2ℓ cos ⃗
(‖ ⃗ ‖)
⃗= ⃗+ ⃗ = 2ℓ(sin ⃗ + cos ⃗) = 2ℓ ⃗ −
2
The “basis” of the movement here is a circle of center O and radius
2ℓ.
Vibration Mechanics of Systems of Solids 67
= ⃗ + ℓ(sin 2 ⃗ + cos2 ⃗)
or by ⃗ = ℓ(sin 2 ⃗ + cos2 ⃗)
v⃗ ( ) = v⃗ ( ) = 2ℓ ⃗(− )
68 Movement Equations 5
Figure 2.7. Representation of the basis and mobile part of the motion of (S)
1) a) ⃗ = ℓ(sin ⃗ + cos ⃗) = ℓ ⃗ −
⃗{ }=( + 2 )J⃗ ( )
Vibration Mechanics of Systems of Solids 69
⋮ ⃗, ⃗ =0; ⋮ ⃗, ⃗ = 0; = ⃗ = +2 ℓ
ℳ⃗ { }= ⃗ + ⃗ ⋀ ( ⃗ )=( +2 ℓ ) ⃗
2
Hence the dynamic torsor:
⃗{ ̅ → } =Y ⃗ + ⃗- (M+2m)g ⃗ − 2 (ℓ sin − ℓ ) ⃗
ℳ ⃗{ → }= ⃗∧ ⃗ = ℓ sin ⃗
ℳ ⃗{ → }= ⃗∧ ⃗ = −ℓ cos ⃗
ℳ ⃗{ → } = 0⃗
ℳ ⃗{ → } = ⃗ ∧ [−2 (ℓ sin − ℓ )] ⃗
= −2 ℓ(ℓ sin − ℓ ) cos ⃗
hence:
{ ̅ → } = {[ − 2 (ℓ sin − ℓ)] ⃗
+ [ − (M + 2m)g] ⃗
/[ℓ Ysin − ℓ cos − 2 ℓ(ℓ sin − ℓ )cos ] ⃗}
70 Movement Equations 5
{ }={ ̅→ }
We obtain:
= −( + 2 )ℓ + ( + 2 ) ℓ sin
− 4 ℓ(ℓ sin − ℓ ) cos
i.e.
( ) = ( + 2 ) ⃗( ) + ⃗ . (⃗ )
i.e.
1
( ) = [( + 2 )ℓ + ]
2
Vibration Mechanics of Systems of Solids 71
{ → }= { → }=0
{ → }=0
{ → } = −( + 2 ) ⃗. ⃗ = ( +2 ) ℓ sin
{ → } = [−2 (ℓ sin − ℓ )] ⃗. ⃗
= −4 ℓ (ℓ sin − ℓ ) cos
( )
= { ̅→ }
i.e.
[ +( + 2 )ℓ ]
= [( + 2 ) ℓ sin − 4 ℓ(ℓ sin − ℓ ) cos ]
O
Figure 2.8. Configuration of the two solids
72 Movement Equations 5
⃗= ⃗+ ⃗+ ⃗
= ℓ⃗ − + ⃗( ) + ⃗( )
2
⃗ = ℓ ⃗(− ) + ⃗ + + ⃗ +
2 2
– the acceleration of in (T):
⃗ = ℓ ⃗(− ) + ⃗ + + ⃗ +
2 2
−ℓ ⃗ − − ⃗( ) − ⃗( )
2
Hence the components in the basis (2) of the dynamic resultant:
⃗ = ⃗ [− − cos( − ) + sin( − )
+ℓ cos( + ) − ℓ sin( + )]
+⃗ [ + sin( − ) + cos( − )
−ℓ sin( + ) − ℓ cos ( + )]
Let us set:
⃗ = [ ⃗ + ⃗ ]
ℳ ⃗{ }= ⃗∧ ⃗ + ⃗ + ⃗ ∧ (⃗ )
with ⃗ = ⃗
Vibration Mechanics of Systems of Solids 73
⃗∧ ⃗ = ⃗∧( ⃗+ )⃗ = ⃗
= [ + sin( − )
+ cos( − ) + ℓ sin( + )
− ℓ cos ( + )] ⃗
⃗ = (− ⃗+ ⃗)
⃗ ∧ ( ⃗ )= ⃗
hence:
ℳ ⃗{ }= ⃗− ⃗
+ ( + )
+ [ cos( − ) − ℓ sin( + )
+ sin( − ) − ℓ cos ( + )] ⃗
With a view to projecting the resultant onto the basis (x ⃗,y ⃗,z ⃗)
and the moment at G of this torsor equality onto the basis (2), let us
calculate:
ℳ ⃗{ → } = ℳ⃗ + ⃗∧ ⃗= ⃗ + ⃗ + ⃗ ∧( ⃗ + ⃗ )
= ⃗ + ⃗ + [ ( − )+ ( − )] ⃗
[3′] 0 = 0
[4′] 0 = L
[5′] 0 = M
={ → }+{ → }
ℳ ⃗{ → } = ℳ ⃗{ → }+ ⃗⋀ ⃗{ → }
= 0⃗ − ⃗⋀ ⃗=− ⃗
[3′′] 0 = 0
[4′′] =
[5′′] − =
Equations [1′] and [2′] provide the values of the link efforts at
equilibrium:
ℓ√2
X=2 −ℓ , =( +2 + )
2
ℓ√2
4 −ℓ = +2 + (1 − )
2 ℓ
⃗ = ( )⃗
⃗ = ⃗= ⃗ (b
b being a positive constannt)
{ → } = (ℓ + − ) ⃗ ⋮ 0⃗ = { → }
0 0
= 0 0
0 0
( ) is in perfect pivot link of axis ( ⋮ ⃗) with respect to ( ).
We set:
( ) = ( ⃗, ⃗) = ( ⃗, ⃗).
{ → } = 0⃗ ⋮ −Γα ⃗ = −{ → }
We associate:
– the link of the plane ( ⋮ ⃗ , ⃗) with ( ) the torsor:
{ → } ={ ⃗ ⋮ ⃗+ ⃗}
Question 1:
Using their components on the basis ( ⃗, ⃗, ⃗), determine the
reduction elements at of the torsors:
a) Kinematics: , ,
b) Dynamics: ,
We set: ⃗ = a ⃗ = a ⃗( )
= ⃗ /⃗ = ⃗/ ′⃗
= ⃗ /⃗ −⃗ = { ′ ⃗/ ⃗ }
= + ={ ′ ⃗/ ′ ⃗ }
b) The sum (or resultant) of the dynamic torsor of the solid (S1) is:
⃗ = ⃗= "⃗
ℳ⃗ = ( )( ⃗)+ ⃗^ ( )( ⃗) = 0⃗
as ⃗ = 0⃗
⃗ = ⃗= − ⃗( ) + " ⃗( + ) + ′′ ⃗
ℳ⃗ = ⃗∧ ⃗+ ( )( " ⃗) + ′ ⃗ ∧ ( )( ′ ⃗)
= (m y′′ cos + C ′′) ⃗
Vibration Mechanics of Systems of Solids 79
Question 2:
Write, in tensorial form, the fundamental principle of dynamics
applied to (S1), and then to (S2):
={ → }+{ → }-{ → }
hence ℳ ⃗{ → }= ⃗ ∧ (- ⃗) = sin ⃗
Question 3:
Write the scalar consequences on ( ⃗, ⃗, ⃗) of the fundamental
principle applied to (S1), and then to (S2).
[1] 0 = X+ - ;
[2] "= +2k(ℓ+b-y) ;
[3] 0 = ;
[4] 0 = ;
[5] 0 = + ;
[6] 0 = N+Γ ;
[8] - ²
+ " + "=- ;
[9] 0=- ;
[10] 0 = - ;
80 Movement Equations 5
[11] 0 = - ;
[12] ′′ '' = -Γ + .
Question 4:
Deduce therefrom the two movement equations (equations not
involving an unknown link).
Question 5:
– Verify that = 0 is an equilibrium position and determine :
the value of y at equilibrium.
– For what value of Γ does another equilibrium position
∈ 0, also exist?
[13]eq: 0 = 2k(+b - )
[12]eq: 0 = −Γ + because ′= ′′ = ′′ = 0
[14] = sin
Vibration Mec
chanics of Syste
ems of Solids 81
Figure 2.1
10. Plot of func
ctions
Questiion 6:
Wriite the moveement equatiions linearizzed around thhe equilibriuum
positionn: y = , = 0
By writing:
w y = + (t) and = (t), we connsider andd
as infinnitely small first-order
f vaalues.
At the
t first orderr, cos = 1 and
a sin = .
82 Movement Equations 5
[13]lin: ( + ) ′′+ 2k + ′′ = 0
The presence of the term ′′ in the first equation and the term
′′ in the second equation indicates that there is a first-order
coupling between the variations of the movement parameters of the
two solids.
Question 7:
²
In the case where: = = , = = , G = 2ma², Γ =
²
3ma²
1
''+ω²0 ε1 + ε2 " = 0
2
′′ + 2 " + 2 ² = 0
d1 = 1; d2 = ½; d3 = 2; d4 = 2
² ²
− − =0
2
² ² ²
− +2 − =0
( ² ²) ²
[15] = ²
= ²)
(
− =
− = ⟹ ⇒
− =−
² ²
=
² ²
=2
(t) = cos + - √2 +
84 Movement Equations 5
Question 8:
If the initial conditions are:
(0) = , (0) = 0
(0) = 0, (0) = 0
= +
0= −
0=− − √2
⇒ sin =sin =0
0=− + √2
= +
⇒ = =
0= − 2
Hence: ( )= (cos + √2 )
( )= (cos − √2 )
Vibration Mechanics of Systems of Solids 85
⃗= (ℓ + + ) ⃗ + a( ⃗ + ⃗)= a ⃗+ (ℓ + b+ cos t) ⃗
⃗= a ⃗ = a (cos ⃗+sin ⃗)
⃗= a( ⃗ + ⃗) = a ⃗ + (cos t - cos√2 t) ⃗
(
– The solid (S1), to which we attach the frame O1 x1 , y1 , z1 , is )
( )
in pivot link of axis O1 z1 with respect to (g). We set: O1 = Og
⎯⎯
→
γ 1 = ( x g , x1 ) = ( y g , y1 ) measured on z1 = z g , O1G 1 = d1 z1 ,
(
– The solid (S2), to which we attach the frame O 2 x 2 , y 2 , z 2 , is )
in sliding link of axis ( O 2 z 2 ) with respect to solid (S1). We set:
⎯⎯
→
O1O 2 = z 2 z 2 ; x 2 = x1 , y 2 = y1 , z 2 = z1
⎯⎯ →
O 2 G 2 = b 2 y 2 , b2: positive constant.
A 0 0
2
I
O2 ( )
S2 = 0
B2 −D2
0 −D2 C2
Vibration Mechanics of Systems of Solids 87
(
– The solid (S3), to which we attach the frame O 3 x 3 , y3 , z 3 , is )
(
in sliding link of axis O 3 z 3 = O 2 y 2 ) ( ) with respect to solid (S2).
We set:
⎯⎯ →
O 2 O 3 = z 3z 3 x 3 = z 2 , y 3 = x 2 , z 3 = y 2
⎯⎯→
O 3G 3 = a 3 x 3 + d 3 z 3 , a3 and d3: positive constants.
A 0 −E 3
3
I
O3 ( )
S3 =
0 B3 0
−E 3 0 C3
Given efforts:
{Δ }
g →1 O
1
{
}
= 0 N1z1 , {Δ1→ 2 }
O2
{ }{
= Z2 z 2 0 , Δ2→ 3 }
O3
{
= Z3 z 3 0 }
Link efforts: we have here three perfect links
Fig
gure 2.11. Ro
obot (one rotattion + two tran
nslations)
Prreliminary question:
q
Write the basis
b change matrices: [ g,1
g ] and [ 2,3] .
Q
Questions:
1) Using thheir componnents in the basis
b (x2 , y2 , z2 ) , determ
mine the
reeduction elem
ments of thee following kinematic torsors, at thhe point
inndicated:
g
a) at O1 and att O3
1
Vibration Mechanics of Systems of Solids 89
g
b) at O2 and at G2
2
g
c) at O3 and at G3
3
4) Write, for the set (D), the Lagrange equations relating to the
parameters: γ 1 , z 2 , z3 which provide movement equations in this
study, where the links are perfect.
Preliminary question:
The basis change matrices between two bases of this exercise are:
− 0 0 1 0
[ , 1]= 0; [2,3] = 0 0 1
0 0 1 1 0 0
90 Movement Equations 5
Solutions:
1) a) The kinematic torsor of the solid ( ) in pivot link with the
frame has as its reduction elements
at : = ′ ⃗ 0⃗
1
We have:
ℳ⃗ = ℳ⃗ = ℳ⃗ = 0⃗
1 1 1
and ℳ⃗ = ℳ⃗ + ω⃗ ∧ ⃗= ⃗ ∧ ⃗=- ⃗
1 1
b) The solid ( ) being in sliding link with the solid ( ) of axis
( / ⃗ ) = ( / ⃗ ), its kinematic torsor is:
1 1
= = 0⃗ ⃗
2 2
1
= + ={ ′⃗ | ⃗ }
2 2 1
ℳ⃗ = ′ z⃗ + γ′ ⃗ ∧ ⃗ = - γ′ ⃗ + ⃗
2
This vector represents the velocity of the point in its movement
in (g): ⃗ ( ).
2
= + ={ ′⃗ | ⃗+ ′⃗ }
3 3 2
ℳ⃗ = ′ ⃗+ ⃗+ ′ ⃗∧ ⃗
3
=- ′ ′ ⃗+ ⃗+ ′ ⃗
ℳ⃗ = ℳ⃗ + ⃗∧ ⃗
3 3
= ℳ⃗ + ′ ⃗ ∧( ⃗ + ⃗ )
3
= - (z d ) ′ x⃗ + ⃗+ ′ ⃗= ⃗ ( )
= [⃗ ( )] + ⃗ .( ⃗ ∧ ⃗ ( )) +
⃗ . ( )( ⃗ )
we can calculate:
b) = [⃗ ( )] + ⃗ .( ⃗ ∧ ⃗ ( )) +
⃗ . ( )( ⃗ )
= + γ′ ⃗ . ( ⃗ ∧ ⃗)+
c) = [⃗ ( )] + (⃗ ∧ ⃗ ). ⃗ ( )
+ ⃗ . ( ) ⃗ = ( + + )
+ ⃗ ∧( ⃗+ ⃗ )]. ⃗ ⃗ ( )+
= ( + + )+ (− ′ ⃗ ).(- ′ ⃗ +
1
′ ⃗+ ⃗) +
2
= [( +2 ) + + ]+ the kinetic
energy of ( ).
(D) = 1∪2∪3:
T= + + = [ ( +2 )+ + + ]+
( + ) +
{ → }= { → }=- ⃗ . ⃗ ⃗ ( ), i = 1, 2 or 3
{ → 2} = − ⃗ .. − ⃗ + ⃗ =− z′
{ → 3} = − ⃗ . .[-( ) ′ ⃗ + ′ ⃗ + ⃗]=
− z′
Vibration Mechanics of Systems of Solids 93
{Δ → 1} = {Δ → 1} = ⃗ . ⃗ = ′
1
{Δ → 2} = 1 ⃗ 0⃗ =
2
{Δ → 3} = 2 ⃗ 0⃗ =
3
( ): − =Π { → }
Π { → } = 1, { → } = ⃗ . ℳ⃗ {∆ → 1} =
with = 0 and = ′ [ ( +2 )+ + + ]
− = [ ( +2 )+ + + ]+2 ′
( ′ + ′ )
hence:
( )∶ [ ( +2 )+ + + ]+2 ′
′ ( + )=
94 Movement Equations 5
( ): − = Π {2 ∪ 3 → 2 ∪ 3}
with
Π {2 ∪ 3 → 2 ∪ 3} = 2, {2 ∪ 3 → 2 ∪ 3} =
⃗ . ⃗ {2 ∪ 3 → 2 ∪ 3}= - ( + )g +
= 0,
=( + ) ′ , =( + ) ′′
hence:
( ):( + ) ′′ = −( + ) +
( )∶ − = Π {3 → 3}
with Π {3 → 3} = 3, {3 → 3} = ⃗ . ⃗{3 → 3} =
= ( + )
= z ′, = z ′′
hence:
( )∶ z ′′ − ( + )=
we have: ′′ = − + independent of t
then ′ = (− + ) + ′
= − + + ′ +
+ 2
(t) = + +ℓ −
We deduce from this that the solid ( ) and the solid ( ) that it
carries, oscillate along ⃗ with the pulsation = around
the equilibrium position defined by = ℓ − .
3.1. Introduction
f 1 (t ) f 2 (t )
k1 k2 k3
m1 m2
x
c1 c2 c3
x1 x2
The case where the mass matrix is not diagonal but simply
symmetric will be treated as an example, as the procedure for solving
the movement equations is the same. A system with 2 DOF will serve
as an example when the detailing of the process with explanation of
the results is sought.
([ K ] − ω 2
[ M ]) {X} eiω t = { 0} , ∀t
a 2 N ω 2 N + a 2 N − 2 ω 2 N − 2 + ... +a 0 = 0
relation: ([ K ] − ω 2
r [ M ]) {X r } = { 0}
Corresponding to a particular mode, of pulsation ωr , there is a
form vector { Xr } , which translates the mode shape (normal mode).
The set of pulsations ω r and form vectors { Xr } (r = 1, 2, ..., N)
constitutes the modal model of the structure.
k1 + k 2 − m1ω 2 − k2
dét =0
− k2 k 2 + k 3 − m 2ω 2
(k 1 )( )
+ k 2 − m1ω 2 k 2 + k 3 − m 2ω 2 − k 22 = 0
i.e.
m1 m 2ω 4 − [m1 (k 2 + k 3 ) + m 2 (k1 + k 2 )] ω 2 + k1 k 2 + k 2 k 3 + k1 k 3 = 0
We can conclude that this equation always admits two real positive
roots: ω1 and ω2, two eigen pulsations of the conservative system.
X 1r
The vector {X r}= corresponding to ω r , r ∈ {1, 2}, of
X 2r
arbitrary magnitude, is a solution vector of the system ( Σ 2 ) obtained
from one or the other of these two dependent equations:
k1 + k 2 − m1ω r2 − k2 X 1r 0
=
− k2 k 2 + k 3 − m2ω r2 X 2 r 0
x1 − iω1t X 11 −iω 2 t X 12
(iω1t
x = B1e + C1e ) iω 2 t
(
X + B2 e + C 2 e X )
2 21 22
x1 X 11 X 12
x = (D1 cos ω1t + E1 sin ω1t ) X + (D2 cos ω 2 t + E 2 sin ω 2 t ) X
2 21 22
([ K ] − ω 2
r [ M ]) {X r } = { 0}
The form vectors of two different modes, r and s, corresponding to
ωr and ωs, are not orthogonal and their scalar product is non-zero:
{Xs }T {X r } ≠ 0 .
On the other hand, by multiplying, on the left-hand side, the
definition relation concerning the rth mode, with the transposed
T
matrix of a form vector {X s } that is different (r ≠ s) , we obtain a
zero scalar product:
[M ] T
[ ]
= M , we obtain by subtracting, member by member, the
relations above:
(ω 2
s )
−ω r2 {X s } [M ]{X r } = 0
T
{X s }T [M]{X r } = 0
{X s }T [K]{X r } = 0
kr k r : generalized stiffness
2
r = of the rth mode.
r r : generalized mass
{ X r } [ K ] {X r }
T
The ratio ω = 2
is the Rayleigh quotient.
{X r } [ M ] {X r }
T r
Vibrations with N Degrees of Freedom 105
0
– [ Q] [ K ] [ Q] =
T the generalized stiffness
kr = Kg
0
matrix, with [Q] the matrix, the rth column of which is composed of
the vector {X r } :
0 0
μr and kr are diagonal matrices of
0 0
elements, respectively, μr and kr at the rth row and rth column.
−1
We deduce that the product [ μ ] K g gives:
−1
0 0 0
= = [Ω]
μr kr ωr
2
0 0 0
Thus, the space model of the system described by [M] and [K] is
0
converted into a modal model described by [ Ω] = ωr2 and
0
[Q] the basis change matrix leading to the modal basis.
[M ]{ x ′′} + [K ]{ x} = {0}
– multiply the left-hand side by [ Q ] .
T
Vibrations with N Degrees of Freedom 107
We thus obtain:
[ Q ] [ M ][ Q ]{y ′′ }+ [ Q ] [ K ][ Q ]{ y }= { 0 }
T T
[μ ]{ y′′} +
[K ]{ y} = {0}
g
0
(σ ) {y ′′ } + y = 0
ω { } { }
2
r
0
y ′r′ +ω r2 y r = 0
admits as solution:
where:
1
⋮ = is the rth column of [Q]
2δ n ω n y n′ (t )
y n (t ) = An e −δ n ω nt sin (Ω n t + ϕ n )
2
where Ωn = ωn 1 − δ n
y n ( 0 ) = y n0 y n′ (0) = y n′ 0
( ) (
y n′ + δ nω n y n 2 + y n Ω n 2
An = 0
) 1/ 2
0
2
0
Ωn
y′n 0 + δn ωn y n 0
cos φn = 1/ 2
,
(
y′ + δ ω y
) ( )
2 2
+ yn0 Ωn
n 0 n n n0
y n0 Ωn
sin φn = 1/ 2
(
y′ + δ ω y
) ( )
2 2
+ yn0 Ωn
n 0 n n n0
[C] = β [ M] + γ [ K]
where β and γ are real constants. This form of damping is known as
proportional damping.
m k c
det m k c =0
0 k c
−1
and by multiplying, on the left-hand side, by [ μ ] , we obtain the
system
( )
y ′n′ + β + γ ω n2 y n′ +ω n2 y n = 0
n = 1, 2, …, N
If we set: β + γ ω n2 = 2 δ n ω n
β γ ωn
i.e. δn = + < 1 for a lightly damped motion,
2ωn 2
Thus, the modal equations of the system (σ ) take the same form
yn′′ + 2 δ n ω n yn′ + ω n2 yn = 0 as those obtained by adding a modal
damping to the uncoupled equations of the conservative system.
y n (t ) = An e −δ n ωnt sin(Ω n t + ϕ n )
{x} = [Q ]{y}
Hence the displacement of the ith coordinate:
xi = X in yn = X in An e −δ nωnt sin(Ω n t + ϕ n )
n n
[C] = β [ M] + γ [ K]
of which each mass, mi , is excited by a force, fi (i = 1,2, … , N).
( Σ ) : [ μ ]{ y′′} + ( β [ μ ] + γ K g ) { y′} + K g { y} = [Q ]T { f }
(
yn′′ + β + γ ω n2 ) yn′ + ωn2 yn = Fn n = 1, 2, …, N
e − δ nω n t t
yn = d n e −δ nωn t
sin (Ω n t + φn ) + F (u )e
n
−δ ωn nu
sin Ω n (t − u )du
Ωn 0
x = X 11 y1 + X 12 y 2
⎯→ 1
⎯
x 2 = X 21 y1 + X 22 y 2
3.4. Exercises
vertical with its center of mass, G2, below the center of the pivot link,
O2, and where the springs have their natural length.
The cart (1), with center of mass G1 and mass m1, is in translation
along B1 x .
The pendulum (2), to which we attach the frame (O 2 x 2 , y 2 , z ) , is
⎯⎯→
of mass m2, center of mass G2, defined by G 2 O 2 = y 2 , and moment
of inertia m2 r 2 with respect to G 2 z , where r is the radius of gyration.
The pendulum is in pivot link of axis O 2 z with respect to the frame.
α = (x , x 2 ) = (y , y 2 ) (the angle between the rod of the pendulum (2)
and the vertical).
m1 = 100 kg m2 = 10 kg
Solution to Exercise 1
1) The dynamic resultant for the cart (1) in translation along x is
written as m 1 x 1′′
The efforts applied to the cart along x are:
– the force of spring B1B2: −k1x1
The moment of inertia along the O2z axis of the pendulum (2) is
given as
(
IO 2 Z (2) = IG 2 Z (2) + m22 = m2 r 2 + 2 )
Then its dynamic moment about the O2 z axis is written as
( )
m 2 r 2 + 2 α ′′
i.e. at the first order: − k 2 a ( aα − x1 ) z
Thus, the moment equation along O 2 z is written as
( )
m 2 r 2 + 2 α ′′ = − m 2 gα − k 2 a (aα − x 1 )
m1x1′′ + (k1 + k 2 ) x1 − k 2 x 2 = 0
2
m m g
x ′′ − k 2 x1 + 22 + k 2 x 2 = 0
2 a 2 2 a
m1 0
100 0 (kg)
[ M] = 2 =
0 m2 2
0 40
a
k1 + k 2 −k 2
[ K ] = m 2 g = 32, 400 −24,000 (N/m)
−k 2 + k 2 −24,000 24, 400
a 2
k1 + k 2
Ω1 = = 18 rad/s, under the action of two springs.
m1
ℓ
= (
= 24.57 rad/s, under the action of gravity and
ℓ )
of the spring A1A2.
f1 (t ) f 2 (t )
k 20 = 12 (N/m)
f1 (t ) = 5sin t (N ), f 2 (t ) = 0
Solution to Exercise 2
1) The equation of the dynamic resultant along x for each of the
solids of mass m1 and m2 gives the movement equations:
m 1x 1′′ = − k 10 x 1 − k 12 (x 1 − x 2 ) + f1 (t )
m 2 x ′2′ = − k 12 (x 2 − x 1 ) − k 20 x 2 + f 2 (t )
1 0 7 −4
[M] = and [ K ] =
0 4 −4 16
det −ω 2 [ M ] + [ K ] = 0
7 −ω2 −4
=0
−4 16 − 4ω 2
(
4ω 4 − 44ω 2 + 96 = 4 ω 2 − 3 ω 2 − 8 = 0)( )
2
ω1 2 = 3 and ω 2 = 8
Vibrations with N Degrees of Freedom 121
X1j
The form vectors X j = { } are determined from:
X 2 j
[− ω 1
2
[M ] + [K ]]
X 11 0
= X 11 = X 21 { X 1 } =
1
X 21 0 1
[− ω 2
2
[M ] + [K ]]
X 12 0
= − 4 X 22 = X 21 { X 2 } =
− 4
X 22 0 1
1 1 1 0 1 − 4 5 0
[μ ] = [Q ]T [M ][Q ] = 0 4 1 1 = 0 20
− 4 1
[K ] = [Q] [K ][Q]
g
T
1 1 7 − 4 1 − 4 15 0
= =
− 4 1 − 4 16 1 1 0 160
3 0
We verify that [μ ] K g =
−1
[ ]
0 8
122 Movement Equations 5
x1 = y1 − 4 y 2
{x} = [Q ]{y} gives
x2 = y1 + y 2
x1 = A1 sin ( )
3t + ϕ1 − 4 A2 sin( 8t + ϕ ) 2
x2 = A sin (
1 3t + ϕ ) + A sin ( 8t + ϕ )
1 2 2
0= sin − 4 sin
sin =4 sin = 0.02
0.025 = sin + sin
0 = 3 A1 cos ϕ1 − 4 8 A2 cos ϕ 2
A1 cos ϕ1 = A2 cos ϕ 2 = 0
0 = 3 A1 cos ϕ1 + 8 A2 cos ϕ 2
X1
0 ,0 5
0 ,0 4
0 ,0 3
0 ,0 2
0 ,0 1
- 0 ,0 1
0 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3 1 4 1 5 1 6 1 7 1 8 1 9 2 0 t
- 0 ,0 2
- 0 ,0 3
- 0 ,0 4
- 0 ,0 5
Figure 3.4. Variations of x1 with respect to time for the first 20 sec
x2
0.03
0.02
0.01
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
-0.01
-0.02
-0.03
(σ ) : {y ′′} + [Ω ]{y} = [μ ] −1 [Q ]T { f (t )} = { F (t )}
5 sin t F1
Thus, the modal forces are: [μ ]
−1
[Q ]T =
0 F2
1
5 0 1 1 5 sin t sin t F1
1 − 4 1 0 = − sin t = F
0 2
20
124 Movement Equations 5
y1′′ + 3 y1 = sin t
y 2′′ + 8 y 2 = − sin t
Given that the system is excited from rest (equilibrium of the set),
the solution of each equation of (σ ) is of the form:
1 t
yn = Fn ( u ) sin ωn ( t − u ) du (see Volume 3 of this set)
ωn 0
1 t
i.e. explicitly: y1 = sin u sin 3 ( t − u ) du …
3 0
…=
2
1
3 0
{cos[(
t
) ] [( )
3 + 1 u − 3 t − cos 1 − 3 u + 3 t du … ]}
…= [ ]
1 sin ( 3 + 1) u − 3 t sin (1 − 3 ) u + 3 t
−
[ ] t
…
2 3 3 +1 1− 3 0
…
1 sin t sin t sin 3 t sin 3 t …
= + + +
2 3 3 +1 3 −1 3 +1 1− 3
1 1
y1 = sin t − sin 3 t
2 2 3
1 t
y2 = − sin u sin 8 ( t − u ) du …
8 0
Vibrations with N Degrees of Freedom 125
1 1
y2 = − sin t + sin 8 t
7 7 8
Thus, the response in the time domain of the excited system is:
15 1 2
x1 = y1 − 4 y 2 = sin t − sin 3 t − sin 8 t
14 2 3 7
5 1 1
x2 = y1 + y 2 = sin t − sin 3 t + sin 8 t
14 2 3 7 8
4 0
mass [M] = (in kg)
0 9
12 −6
stiffness [K] = (in N/m)
−6 27
42 14
Ω1 = rad/s and Ω 2 = rad/s
5 2
126 Movement Equations 5
Solution to Exercise 3
12 − 4ω 2 −6
Thus, in this case we have: =0
−6 27 − 9ω 2
(
i.e. 12 − 4ω 2 )( 27 − 9ω ) − 36 = 0
2
(3 − ω ) ( )( 2 − ω ) = 0
2
2
or 36 − 1 = 36 4 − ω 2 2
2) The purpose of the tests is to deduce the dampings from the real
structure.
2 1 Ω n2
Then from: δ n = 1 −
2 ωn2
Vibrations with N Degrees of Freedom 127
42
1 2 √2
δ 12 = 1 − 25 = 1 = = 0.282
2 2 25 5
14
1 1 1
δ 2 = 1 − 4 =
2
2 = = 0.25
2 4 16 4
b) For a proportional damping, the damping matrix is a linear
combination of the mass and stiffness matrices:
[C ] = β [M ] + γ [K ]
For a system with 2 DOF, the coefficients β and γ can be
determined from the relation:
β + γωn2 = 2δ nωn
2 4
which, for the first mode, is written as β + 2γ = 2 x x 2=
5 5
1
β + 4γ = 2x x2 = 1
4
From these two equations, we deduce:
1 3
γ = and β =
10 5
Thus,
3 1 1 36 −6
[C] = [ M ] + [ K ] = (Ns/m)
5 10 10 −6 81
128 Movement Equations 5
f1 (t ) f 2 (t )
k1 k2
m1 m2 x
c1 c2
x1 x2
Solution to Exercise 4
1) The equation of the dynamic resultant for each solid, projected
along the x axis, gives the movement equations of the coupled
system:
m 1 x 1′′ = − k 1 x 1 − k 2 (x 1 − x 2 ) − c1 x 1′ − c 2 (x 1′ − x ′2 ) + f1 (t )
m 2 x ′2′ = k 2 (x 1 − x 2 ) + c 2 (x 1′ − x ′2 ) + f 2 (t )
m 0 x1′′ c1 + c 2 − c 2 x 1′ k1 + k 2 − k 2 x 1 f1
(S) : 1 + + =
0 m 2 x ′2′ − c 2 c 2 x ′2 − k 2 k 2 x 2 f 2
9 0
has the mass matrix [M] = and the stiffness matrix
0 1
− 3
[K] =
27
− 3 3 .
[K ] − ω 2 [I ] =0
27 − 9ω 2 −3
= (
= 9 3−ω2 ) 2
( )(
−9 = 9 4 −ω2 2 −ω2 )
−3 3−ω2
X 1n
The (eigen) vectors of form { X n}= , which compose the
X 2n
basis change matrix to modal basis, are determined from the first
( )
equation (or from the second) of: [K ] − ω 2 n [M ] { X n } = { 0}
9 − 3 X 11 0 1
− 3 1 X = 0 → 3X11 = X 21 → { X1} = 3
21
− 9 − 3 X 12 0
− 3 − 1 X = 0
22
1
→ −3 X 12 = X 22 → { X 2} =
−3
Vibrations with N Degrees of Freedom 131
[Q] =
1 1
.
3 − 3
1 1 y1 y1 + y 2
c) By setting { x} = [Q ]{ y} =
= , in
3 − 3 y 2 3 y1 − 3 y 2
the system (S1) and by multiplying by [Q ] on the left-hand side, we
T
1 3 9 0 1 1 18 0
[μ ] = [Q ]T [M ][Q ] = =
1 −3 0 1 3 −3 0 18
Then, by multiplying the left-hand side of this relation by
1 1 0 ,
[ μ ]−1 = we obtain the system (σ1 ) : { y′′} +
18 0 1
[Ω]{ y} = { 0} , which is composed of the equations:
y1′′ + 2 y1 = 0
y 2′′ + 4 y 2 = 0
the solutions of which are:
y1 = A1 sin ( 2 t + ϕ1 )
y 2 = A2 sin (2t + ϕ 2 )
132 Movement Equations 5
[
x 2 = 3 y1 − 3 y 2 = 3 A1 sin ( )
2 t + ϕ1 − A2 sin (2t + ϕ 2 ) ]
3) The damped system (S 2 ) : [M ]{x ′′}+ [C ]{x ′}+ [K ]{x } = {0} has the
9 − 1
damping matrix: [C] =
− 1 1
1
b) From the relation β + γ ωn2 = 2δ n ωn = ωn2 with
3
ω1 = 2
[Ω] =
2 0 1
, we deduce: δ1 = , which is
2
0 4 2δ1ω1 =
3 2
3
ω2 = 2 1
the damping factor of the first mode, and δ2 = ,
4 3
2δ 2ω2 = 3
which is the damping factor of the second mode.
1 ′
[ M ]{x ′′} + [ K ] {x } + {x} = {0} .
3
′′ 2 ′
y1 + 3 y1 + 2 y1 = 0
4
y2′′ + y2′ + 4 y2 = 0
3
of the form:
y n′′ + 2δ nω n y ′n + ω n2 y n = 0
√17
Ω = 1− = = 1.37 rad/s
3
4√2
Ω = 1− = = 1.88 rad/s
3
Thus, the solution of (σ 2 ) is:
−
t
17
y1 = A1e 3 sin t + ϕ1
3
−
2t
4 2
y 2 = A2 e 3
sin t + ϕ 2
3
and the solution of (S2) is:
−
t
17 −
2t
4 2
x1 = y1 + y 2 = A1e 3 sin t + ϕ1 + A2 e 3 sin t + ϕ
3 3 2
−
t
17 −
2t
4 2
x2 = 3 y1 − 3 y 2 = 3 A1e 3
sin t + ϕ1 − 3 A2 e 3 sin t + ϕ 2
3 3
134 Movement Equations 5
with { F } = [μ ] −1 [Q ] T { f (t )}
1 1 0 1 3 3 sin t
=
18 0 1 1 − 3 sin t
1 sin t
=
3 0
′′ 2 ′ 1
y1 + 3 y1 + 2 y1 = 3 sin t
4
y ′2′ + y ′2′ + 4 y 2 = 0
3
Note that only the modal coordinate, y1, is excited.
Thus, as the system is excited from rest, the zero initial conditions:
x 1 (0 ) = x 1′ (0 ) = x 2 (0 ) = x ′2 (0 ) = 0
y1 p = B1 sin t + B2 cos t
2
− B1 sin t − B2 cos t + (B1 cos t − B2 sin t ) + 2(B1 sin t + B2 cos t ) = 1 sin t
3 3
2 1
(of sin t): − B1 − B2 + 2 B1 =
3 3
3 2
we deduce: B1 = and B2 = −
13 13
− 17
t
3 2
y1 = A1e sin
3
t + ϕ1 + sin t − cos t
3 13 13
2
y1 (0 ) = 0 = A1 sin ϕ1 −
13
1 17 3
y1′ (0) = 0 = − A1 sin ϕ1 + A1 cos ϕ1 +
3 3 13
136 Movement Equations 5
2 7
we obtain: 𝐴1 sin 𝜑1 = 13 and 𝐴1 cos 𝜑1 = −
13 17
1 −𝑡 17 7 17 3 2
𝑦1 = 𝑒 3 2 cos 𝑡− sin 𝑡 + sin 𝑡 − cos 𝑡
13 3 17 3 13 13
𝑖𝑛 𝑡𝑟𝑎𝑛𝑠𝑖𝑒𝑛𝑡 𝑠𝑡𝑎𝑡𝑒
steady state
It, therefore, becomes negligible with respect to the terms sint and
cost (time period 6.28 sec) that dominate the displacements of the
system of two solids (steady state).
x2
Transient Steady state
m1 m2 x (t ) = x 0 sin ω t
Figure 3.8. Schematic diagram of a system with 2 DOF excited by the base
x1′′ x1′ x1
2) We set: [ M ] + [ C] + [ K ] = { f ( t )}
x ′′2 x ′2 x2
3) Numerical application:
2
m1 = 2 (kg) c10 = 2 (Ns/m) k10 = 4 (N/m) x0 = (m)
7
F = A sin (ω t + θ ) .
Solution to Exercise 5
1) The movement equations of the solids of mass m1 and m2,
respectively, are:
m 2 x ′2′ = − k 12 (x 2 − x 1 ) − k 23 (x 2 − x ) − c12 (x ′2 − x 1′ ) − c 23 (x ′2 − x ′)
[M ]{ x ′′}+ [C ]{ x ′}+ [K ]{ x } = {f (t )}
we have:
m 0 c + c12 −c12
[ M ] = 01 m 2
, [C] = 10−c c12 + c23
,
12
0
and { f ( t )} = , which represents the
k 23 x 0 sin ω t + c23 x 0ω cos ω t
system’s excitation force.
Vibrations with N Degrees of Freedom 139
0
{f (t )} =
0
=
2 (4 sin 7t + 11cos 7 t ) f 2
[K ] − ω 2 [M ] =0
8 − 2ω 2 −4
( ) ( )( )
2
i.e. = 16 4 − ω 2 − 16 = 16 5 − ω 2 3 − ω 2 = 0 ,
2
−4 32 − 8ω
hence:
1 = √3 = 1.732
(in rad/s)
2 = √5 = 2.236
The eigen vectors are determined from one or other equation of the
relations:
{ X1} =
2
[K − 3M ]{ X1} = {0} ⎯
⎯→ X 11 = 2 X 21 ⎯
⎯→
1
2
[K − 5M ]{ X 2 } = {0} ⎯
⎯→ X 12 = −2 X 22 ⎯→ { X 2 } =
⎯
− 1
[Q ] =
22
1 − 1
140 Movement Equations 5
1 1 5 5
2 3δ1 = + x3 = , hence 1 = = 0.36
2 4 4 8√ 3
1 1 7 7
2 5δ 2 = + x5 = , hence 2 = = 0.39
2 4 4 8√5
25 √167
= 1− = √3 1 − = = 1.615 rad/s
64 × 3 8
49 √271
= 1− = √5 1 − = = 2.057 rad/s
64 × 5 8
y n′′ + 2δ n ω n y ′n + ω n2 y n = Fn
2 1 2 0 2 2 16 0
[μ ] = [Q ]T [M ][Q ] = 2 =
−1 0 8 1 −1 0 16
Vibrations with N Degrees of Freedom 141
F1 1 1 0 2 1 0 f 2 1
F = [μ ] [Q ] { f (t )} = 16 0 1 2 − 1 f = 16 − 1
−1 T
2 2
with
f 2 = 2 (4 sin 7t + 11 cos 7t )
hence, explicitly:
′′ 5 ′ f2
y1 + 4 y1 + 3 y1 = 16
(σ ) 7 f
y2′′ + y′2 + 5 y2 = − 2
4 16
f2 2
e) The modal excitation force F = = ( 4sin 7t + 11cos7t )
16 16
can also be written as
with:
4 4
cosθ = =
2
4 + 11 2 137
11
and sin θ = , θ = 70° or 1.22 rad
137
274
and A =
16
√
then = sin(7 + 1.22)
142 Movement Equations 5
y n = Bn sin(ω t + θ + ϕ n ) , n = 1 or 2
where:
A / ω n2
Bn = ,
Dn
ω2 ω
1− − 2δ n
ω n2 ωn
cos ϕ n = and sinϕ n =
Dn Dn
7 5 7
D = 1− + 2 = 1− + 2
3 8√3 √3
46.82
=
3
7 7 7
D = 1− + 2 = 1− + 2
5 8√5 √5
45.67
=
5
and the response:
√
= sin(7 + + ) = 0.0221sin (7 + + ) (m)
.
√
and =
.
sin(7 + + ) = −0.0226sin (7 + + ) (m)
because:
7
1−
= 3 = −46 = −0.982
D 46.82
and
sin ϕ 1 < 0
= 2( + ) = 2 0.0221 sin(7 + + )−
0.0226 sin(7 + + ) (m)
with:
we have:
h)
xi(t)
0,05
0,04 x2(t)
0,03
0,02 x1(t)
0,01
0 t
-0,01 1 1,5 2 2,5 3 3,5 4 4,5 5 5,5 6
-0,02
-0,03
-0,04
-0,05
Modal Analysis of
N Degrees of Freedom
4.1. Introduction
[M ]{x′′(t )} + [K ]{x(t )} = { f (t )}
where [M ] is the mass matrix and [K ] is the stiffness matrix. In the
linear modeling of a structure admitting N degrees of freedom
(N ≥ 2 ) , these two matrices are square, symmetric and positive
defined, of constant and real elements. { x} is the column matrix
representing the displacements (N physical coordinates) of the
structure.
Let {f ( t ) }= {F }e ω i t
be the matrix representing the excitation
force vector.
{x ( t ) }= {X }e ω
i t
[M ]{ x ′′}+ [K ]{ x } = {f (t )}
Modal Analysis of N Degrees of Freedom 147
([ K ] − ω [ M ] ){X }= {F }
2
{X } = ( [ K ] − ω [ M ] ) { F }
2 −1
{ X} = [α (ω )]{ F}
with [α ( ω ) ] = ([ K ]− ω [ M ] )
2 −1
of =
Fk
α (ω ) Xj
X = F if F = 0, ∀m ≠ k, therefore
m ∈ l, … , k − 1, k + 1, … , N
148 Movement Equations 5
We obtain
0
[ ]
With μ −1 / 2 =
1
the rth column of [P ] is
μr
0
P1r q 1r
. .
1 1
. = . = {X r }
. μ r . μ r
P q
Nr Nr
0
and that [P] [K ][P] = [μ ]
T −1 / 2
[K ][μ ]
g
−1 / 2
=
ω 2
r
0
0
we have
= [ P ] [α ( ω )] [P ]
T −1
ω r2 −ω 2
0
Modal Analysis of N Degrees of Freedom 149
−1
0
[α ( ω ) ] = [ P ] ω r2 − ω 2 [P ] T
0
NOTE.–
The general element αjr of the product [P][ ] is: α = ∑ P
but = ( : the Kronecker symbol)
1
If [Ω ] is a diagonal matrix of elements Ω r = ,
ω 2
r −ω 2
then = Ω .
Thus, the general element
Xj X
Here: α jk = = α kj = k
F
Fk Fm = 0, ∀m≠k j Fm = 0, ∀m ≠ j
This indicates that we have the same response from the structure
by measuring X j and by exciting at the coordinate k, as by
measuring X k with the same excitation applied at the coordinate j.
N Pjr Pkr N A jk
Thus: α jk (ω ) = = r
r =1 ω −ω
2
r
2
r =1 ω −ω2
2
r
150 Movement Equations 5
where r A jk is the modal constant, which, for the rth mode, links the j
and k coordinates.
A jk A jk A jk
α jk (ω ) = 1
+ 2
+ ... + N
ω −ω
2
1
2
ω −ω
2
2
2
ω −ω2
2
N
X j = α j1F1 + α j2 F2 + . . . + α jN FN
N p jr p1r N p jr p Nr
Xj = F1 + . . . + FN
r =1 ω −ω
2
r
2
r =1 ωr2 − ω 2
[C] = β [M ] + γ [K ]
Modal Analysis of N Degrees of Freedom 151
([M ] −1
[K ])([M ]−1[C]) = ([M ]−1[C])([M ]−1[K ])
As a result, the similar matrix of [C] in the modal basis will be a
diagonal matrix.
( ( ) )
i.e. (1 + iγ ω ) [K] + iβ ω − ω 2 [M] {X} = {F}
0 0
[Q] [C][Q] = β μ r + γ k r = [Cg ]
T
0 0
0
= [C ]
[P] [C][P] = β [I ] + γ ω r
T 2
n
0
152 Movement Equations 5
Thus:
[α ]−1 = [K ] + iω [C] − ω 2 [M ]
becomes:
0
[P]T [α ]−1 [P] = (iβω − ω )[I] + (1 + iγω ) ω r
2 2
0
0
= ω r − ω + i(β + γω r )ω
2 2 2
0
0
1 T
[α ] = [P] [P]
ω r − ω + 2iδ rω rω
2 2
0
N Pjr Pkr
α jk (ω ) =
r =1 ω −ω + 2iδ rω rω
2
r
2
ω = iδ r ω r ± ω r 1 − δ r2
Pjr Pkr 1
= Pjr Pkr
ω − ω + 2iδ rω rω (ω − ω 2 ) + (2δ rω rω )
2 2 2
2 2
r
r
[D ] = β [M ] + γ [K ]
By also using here the normalized form vectors of the undamped
system, composing the matrix [P ], we have:
154 Movement Equations 5
0
[P] [D][P] = β [I] + γ ω r
T 2
0
Thus,
0
= ω r − ω + i (β + γω r )
2 2 2
0
0
= ω r2 − ω 2 + iηrω r2
0
β
by setting ηr = γ + : damping loss factor.
ω r2
N Pjr Pkr
α jk (ω ) =
r =1 ω − ω 2 + iηrω r2
2
r
4.5. Exercises
x1 x2
k1 k2 k3
x
m1 m2
f1 ( t ) f2 ( t )
The links are perfect and the displacements x1 and x2 along x are
measured from equilibrium. The spatial characteristics of the system
are:
x 20 = 1 cm x 20 = −1 cm x 20 = 0
a) 2 x103 rad / s
b) √0.4x10 rad/s
Solution to Exercise 1
I) 1) The free movement equations is:
m1 0 x1′′ k1 + k 2 − k 2 x 1 0
+ =
0
m 2 x ′2′ − k 2 k 2 + k 3 x 2 0
k1 + k 2 − m1ω 2 − k2
=0
− k2 k 2 + k 3 − m 2ω 2
{ X r } =
q1r
b) The form vectors , r = 1 or 2, are deduced
q 2 r
from the first relation of:
( )
i.e.: k1 + k 2 − ω r2 m1 q1r = k 2 q 2 r
158 Movement Equations 5
1
Hence, for r = 1: q11 = q 21 and { X1} =
1
1
and for r = 2: − q12 = q 22 and { X 2 } =
− 1
1 1
Thus, [Q] = is the basis change matrix to the modal basis.
1 − 1
[Q] T [K ][Q] = K g =
0,8 0 6
[ ] 10 (N/m)
0 4
x1 y1 x1 = y1 + y 2
d) By setting: x = [Q] y , that is to say,
2 2 x 2 = y1 − y 2
the modal-variable equations are:
y1′′ + ω12 y1 = 0
y′2′ + ω 22 y 2 = 0
with:
= √2 × 10 = 1,414 rad/s
y1 = A1 sin(ω1t + ϕ1 )
y 2 = A 2 sin(ω 2 t + ϕ 2 )
Modal Analysis of N Degrees of Freedom 159
x1 = A1 sin(ω1t + ϕ1 ) + A2 sin(ω2 t + ϕ 2 )
x 2 = A1 sin(ω1t + ϕ1 ) − A2 sin(ω2 t + ϕ 2 )
k1 k2 k3
m1 m2
k1 k2 k3
m1 m2
we have: x1 (t ) =
1
(cosω1t + cosω2 t ) = cos ω1 + ω 2 t cos ω 2 − ω1 t
2 2 2
(in cm)
and x 2 (t ) =
1
(cos ω1t − cos ω 2 t ) = sin ω1 + ω 2 t sin ω 2 − ω1 t (in cm)
2 2 2
2π 2π
T1 = = = 6 ms
Ω1 ω1 + ω 2
2
2π 2π
and T2 = = = 16 ms
Ω2 ω 2 − ω1
2
T2
but the mass m2 is offset by with respect to the mass m1.
4
( )
we have: k1 + k 2 − ω 2 m1 X1 − k 2 X 2 = F1
( )
− k 2 X1 + k 2 + k3 − ω 2 m2 X 2 = F2
k1 + k 2 − ω 2 m1 − k2 X1 F1
=
2 F
− k2 k 2 + k 3 − ω m 2 X 2 2
X F
= [ ( )]
X F
1 k +k − m k F
=
det[K − M] k k +k − m F
Thus:
k2 + k3 − ω 2m2 X1
α11 = = if F2 = 0
F1
K − ω 2M
k1 + k 2 − ω 2 m1 X1
α 22 = = if F2 = 0
F1
K − ω 2M
k2
α12 = = α 21
K − ω 2M
X X
= if F = 0, or = if F = 0
F F
162 Movement Equations 5
|K − M| = 1.2 x 10 − − 0.8 x 10
−0.8 x 10 1.2 x 10 −
= − 2.4 x 10 + 0.8 x 10
and the receptance matrix elements:
1.2 x 10 −
=
− 2.4 x 10 + 0.8 x 10
−0.5 −0.5
= + =
− 0.4 x 10 − 2 x 10
(for this example)
0.8 x 10
=
− 2.4 x 10 + 0.8 x 10
−0.5 −0.5
= + =
− 0.4 x 10 − 2 x 10
(always).
NOTE.– We could obtain these results directly from the relation:
Pj1Pk1 Pj2 Pk 2
α jk = 2 2
+
ω −ω
1 ω 22 − ω 2
−1/2 1 1 1
where Pjr is the element of the matrix [ P ] = [ Q][ μ ] = of
2 1 −1
change to the normalized modal basis.
α11 = α 22
0.0002
0.00015
0.0001
0.00005
ω (rad / s )
0
0 200 400 600 800 1000 1200 1400
-0.00005
-0.0001
α12 = α 21
0.0002
0.00015
0.0001
0.00005
0
0 200 400 600 800 1000 1200 1400
ω (rad / s )
-0.00005
-0.0001
0.5(F + F ) 0.5(F − F )
X = F + F =− +
− 0.4 x 10 − 2 x 10
. ( )
4) When ω is close to √0.4 x 10 rad/s: X ≅ X ≅
.
f1 (t) f2 (t)
k10 k12
m1 m2
c20 = 1 (Ns / m)
Write the mass [M], stiffness [K] and damping [C] matrix of the
system.
Modal Analysis of N Degrees of Freedom 165
4) Determine the generalized mass [μ ] and stiffness K g
matrices. Calculate [μ ]
-1
[K ].
g
Solution to Exercise 2
1) The movement equations are written in the matrix form:
2) Numerically, we have:
22 − 10
[M ] =
2 0 2.6 −1
, [C] = , [K ] =
0 5 −1 2 − 10 10
2,6 − 1 1 2 0 1 22 − 10
Note that [C] = = + = β [M ] + γ [K ]
− 1 2 5 0 5 10 − 10 10
That is,
1 1
= and =
5 10
3) The eigen pulsations of the undamped structure are the positive
roots of:
22 − 2ω 2
[K] − ω 2 [M ] =
− 10
(
= 10 ω 4 − 13ω 2 + 12 )
− 10 10 − 5ω 2
{ X r } =
q1r
The corresponding form vectors are determined
q 2 r
from:
{X1} =
r =1 1
⎯⎯→
⎯ 2q11 = q 21
2
[K − ω M ]{X } = {0}
2
r r
{X 2 } =
r =2 5
⎯⎯⎯→ q12 = −5q 22
− 1
1 5
Thus, the basis change matrix to the modal basis is [Q] =
2 − 1
22 0
4) [μ ] = [Q] T [M ][Q] =
0 55
0
[K ] = [Q] [K ][Q] = 220
g
T
660
1
22 0 22 0 1 0 ω 2 0
= = 1
1 0 660 0 12 0 ω 2
0 2
55
Modal Analysis of N Degrees of Freedom 167
1 = 0.15 (15%)
= + = +
2 2 10 20 √
= 0.202(20.2%)
[P] = [Q][μ ] −1 / 2
1 1 5
0
1 5 22
= = 22 55 = p11 p12
2 − 1 0 1 2 1 p 21 p 22
−
55 22 55
p11 p 21 p12 p 22
α12 = +
ω − ω + i0,3ω1ω ω 2 − ω 2 + i 7 ω ω
2
1
2
2 2
5 12
1 1
−
= 11 + 11
1 − ω 2 + 0,3iω 12 − ω 2 + 1,4iω = α 21
= +
− + 0.3 7
− +
5√12
2 1
= 11 + 55
1− + 0.3i 12 − + 1.4i
As the receptance matrix connects the excitation to the response of
the structure:
X1 α11 α12 F1
X = α
2 12 α 22 F2
α11 0.2
0.15
0.1
0.05
0
0 1 2 3 4 5 6 7 8
Figure 4.8. Bode plot of the modulus of the frequency response function α11
ℑm
0 ℜe
-0.05 -0.025 0 0.025 0.05 0.075 0.1 0.125
-0.02
-0.04
-0.06
nd
2 mode
-0.08
1st mode
-0.1
ω R2
-0.12
-0.14
ωR1
-0.16
{ f (t )} =
1
sin 3t (in N)
− 1
Solution to Exercise 3
5 − 2ω 2 − 4 −ω2
1) [K] − ω [M ]
2
= (
= 5 − 2ω 2 ) − (4 + ω ) = (9 − ω )(1 − 3ω ) = 0
2 2 2 2 2
− 4 −ω2 5 − 2ω 2
1
Hence the natural pulsations: ω1 = = 0.577 (rad/s) and
3
= 3 (rad/s)
{ X1} =
1
( [K] − ω [M] ){X } = {0}
2
1 1 ⎯
⎯→ q11 = q 21 ⎯⎯→
1
8 2
= =
13 13
Hence the damping matrix:
8 2 1 2 5 − 4 2 0
[C] = β [M ] + γ [K ] = + =
13 1 2 13 − 4 5 0 2
1 1
4) With [Q ] = , we have:
1 − 1
[μ ] = [Q] T [M ][Q] =
6 0
0 2
0
[C ] = [Q] [C][Q] = 40
g
T
4
[C ] = β [μ ] + γ [K ]
g g
1 1
6 2
5) [P] = [Q][μ ] −1 / 2 =
1 1
−
6 2
2 Pjr Pkr
α jk =
r =1 ω − ω 2 + i2δ rω rω
2
r
{ x} = [Q]{ y}.
a) By multiplying by [Q] T on the left-hand side, we obtain the
uncoupled equations with modal variables y1 and y2:
are diagonal
Note that only the second mode is excited, and the excitation has
the same pulsation as the structure’s second natural pulsation.
f0 1
y2 = sin (Ω t + ϕ ) ,
k 2 1/ 2
2 2
Ω Ω
1 − 2 + 2δ 2
ω2 ω 2
with Ω = ω 2 = 3 (rad/s)
π
and ϕ = − : the excitation and the response are in quadrature
2
π
(phase-lag response of ), we deduce:
2
1 1 π 1
y2 = sin 3t − = − cos 3t
1
9 2 x x1 2 6
3
x1 y1
x = [Q] y → x1 = − x 2 = y 2
2 2
1
x1 = − cos 3t
6
1
x 2 = cos 3t
6
B, C, D inertia matrix, 36
initial conditions, 54
boundary conditions, 59
conservative system, 100
K, L, M
damper, 97
damping factor, 152 kinetic energy, 11
degree of freedom, 36 Lagrange equations, 40
link, 14
E, F mass matrix, 99
modal
effort
constant, 150
generator, 3
synthesis, 106
receiver, 4
eigen
N, O, P
mode, 48
pulsation, 101 normal mode, 145
energy-power equation, 42 oscillation, 42
equilibrium, 40 poles, 150
form vector, 101 power, 21
friction coefficient, 20 proportional damping, 109
fundamental principle of
dynamics, 39 R, S, V
receptance matrix, 147
G, H, I
residuals, 150
generalized coordinates, 106 resonance, 145
harmonic excitation, 151 rolling, 28
hysteretic damping, 153 set of solids, 1
sliding velocity, 20
solidifying parameters, 9
spatial model, 105
steady state, 145
stiffness, 97
velocity distributor, 37
vibration, 97
Other titles from
in
Mechanical Engineering and Solid Mechanics
2019
EL HAMI ABDELKHALAK, POUGNET PHILIPPE
Embedded Mechatronic Systems 1– Second Edition Revised and Updated
BAYLE Franck
Reliability of Maintained Systems Subjected to Wear Failure Mechanisms:
Theory and Applications
(Reliability of Multiphysical Systems Set – Volume 8)
LE VAN Anh, BOUZIDI Rabah
Lagrangian Mechanics: An Advanced Analytical Approach
MICHELITSCH Thomas, PÉREZ RIASCOS Alejandro, COLLET Bernard,
NOWAKOWSKI Andrzej, NICOLLEAU Franck
Fractional Dynamics on Networks and Lattices
SALENÇON Jean
Virtual Work Approach to Mechanical Modeling
Viscoelastic Modeling for Structural Analysis
2018
BOREL Michel, VÉNIZÉLOS Georges
Movement Equations 4: Equilibriums and Small Movements
(Non-deformable Solid Mechanics Set – Volume 4)
FROSSARD Etienne
Granular Geomaterials Dissipative Mechanics: Theory and Applications in
Civil Engineering
RADI Bouchaib, EL HAMI Abdelkhalak
Advanced Numerical Methods with Matlab® 1: Function Approximation
and System Resolution
(Mathematical and Mechanical Engineering SET – Volume 6)
Advanced Numerical Methods with Matlab® 2: Resolution of Nonlinear,
Differential and Partial Differential Equations
(Mathematical and Mechanical Engineering SET – Volume 7)
2017
BOREL Michel, VÉNIZÉLOS Georges
Movement Equations 2: Mathematical and Methodological Supplements
(Non-deformable Solid Mechanics Set – Volume 2)
Movement Equations 3: Dynamics and Fundamental Principle
(Non-deformable Solid Mechanics Set – Volume 3)
BOUVET Christophe
Mechanics of Aeronautical Solids, Materials and Structures
Mechanics of Aeronautical Composite Materials
BRANCHERIE Delphine, FEISSEL Pierre, BOUVIER Salima,
IBRAHIMBEGOVIC Adnan
From Microstructure Investigations to Multiscale Modeling:
Bridging the Gap
CHEBEL-MORELLO Brigitte, NICOD Jean-Marc, VARNIER Christophe
From Prognostics and Health Systems Management to Predictive
Maintenance 2: Knowledge, Traceability and Decision
(Reliability of Multiphysical Systems Set – Volume 7)
EL HAMI Abdelkhalak, RADI Bouchaib
Dynamics of Large Structures and Inverse Problems
(Mathematical and Mechanical Engineering Set – Volume 5)
Fluid-Structure Interactions and Uncertainties: Ansys and Fluent Tools
(Reliability of Multiphysical Systems Set – Volume 6)
KHARMANDA Ghias, EL HAMI Abdelkhalak
Biomechanics: Optimization, Uncertainties and Reliability
(Reliability of Multiphysical Systems Set – Volume 5)
LEDOUX Michel, EL HAMI Abdelkhalak
Compressible Flow Propulsion and Digital Approaches in Fluid Mechanics
(Mathematical and Mechanical Engineering Set – Volume 4)
Fluid Mechanics: Analytical Methods
(Mathematical and Mechanical Engineering Set – Volume 3)
MORI Yvon
Mechanical Vibrations: Applications to Equipment
2016
BOREL Michel, VÉNIZÉLOS Georges
Movement Equations 1: Location, Kinematics and Kinetics
(Non-deformable Solid Mechanics Set – Volume 1)
BOYARD Nicolas
Heat Transfer in Polymer Composite Materials
CARDON Alain, ITMI Mhamed
New Autonomous Systems
(Reliability of Multiphysical Systems Set – Volume 1)
DAHOO Pierre Richard, POUGNET Philippe, EL HAMI Abdelkhalak
Nanometer-scale Defect Detection Using Polarized Light
(Reliability of Multiphysical Systems Set – Volume 2)
DE SAXCÉ Géry, VALLÉE Claude
Galilean Mechanics and Thermodynamics of Continua
DORMIEUX Luc, KONDO Djimédo
Micromechanics of Fracture and Damage
(Micromechanics Set – Volume 1)
EL HAMI Abdelkhalak, RADI Bouchaib
Stochastic Dynamics of Structures
(Mathematical and Mechanical Engineering Set – Volume 2)
GOURIVEAU Rafael, MEDJAHER Kamal, ZERHOUNI Noureddine
From Prognostics and Health Systems Management to Predictive
Maintenance 1: Monitoring and Prognostics
(Reliability of Multiphysical Systems Set – Volume 4)
KHARMANDA Ghias, EL HAMI Abdelkhalak
Reliability in Biomechanics
(Reliability of Multiphysical Systems Set –Volume 3)
MOLIMARD Jérôme
Experimental Mechanics of Solids and Structures
RADI Bouchaib, EL HAMI Abdelkhalak
Material Forming Processes: Simulation, Drawing, Hydroforming and
Additive Manufacturing
(Mathematical and Mechanical Engineering Set – Volume 1)
2015
KARLIČIĆ Danilo, MURMU Tony, ADHIKARI Sondipon, MCCARTHY Michael
Non-local Structural Mechanics
SAB Karam, LEBÉE Arthur
Homogenization of Heterogeneous Thin and Thick Plates
2014
ATANACKOVIC M. Teodor, PILIPOVIC Stevan, STANKOVIC Bogoljub,
ZORICA Dusan
Fractional Calculus with Applications in Mechanics: Vibrations and
Diffusion Processes
Fractional Calculus with Applications in Mechanics: Wave Propagation,
Impact and Variational Principles
CIBLAC Thierry, MOREL Jean-Claude
Sustainable Masonry: Stability and Behavior of Structures
ILANKO Sinniah, MONTERRUBIO Luis E., MOCHIDA Yusuke
The Rayleigh−Ritz Method for Structural Analysis
LALANNE Christian
Mechanical Vibration and Shock Analysis – 5-volume series – 3rd edition
Sinusoidal Vibration – Volume 1
Mechanical Shock – Volume 2
Random Vibration – Volume 3
Fatigue Damage – Volume 4
Specification Development – Volume 5
LEMAIRE Maurice
Uncertainty and Mechanics
2013
ADHIKARI Sondipon
Structural Dynamic Analysis with Generalized Damping Models: Analysis
ADHIKARI Sondipon
Structural Dynamic Analysis with Generalized Damping Models:
Identification
BAILLY Patrice
Materials and Structures under Shock and Impact
BASTIEN Jérôme, BERNARDIN Frédéric, LAMARQUE Claude-Henri
Non-smooth Deterministic or Stochastic Discrete Dynamical Systems:
Applications to Models with Friction or Impact
EL HAMI Abdelkhalak, RADI Bouchaib
Uncertainty and Optimization in Structural Mechanics
KIRILLOV Oleg N., PELINOVSKY Dmitry E.
Nonlinear Physical Systems: Spectral Analysis, Stability and Bifurcations
LUONGO Angelo, ZULLI Daniele
Mathematical Models of Beams and Cables
SALENÇON Jean
Yield Design
2012
DAVIM J. Paulo
Mechanical Engineering Education
DUPEUX Michel, BRACCINI Muriel
Mechanics of Solid Interfaces
ELISHAKOFF Isaac et al.
Carbon Nanotubes and Nanosensors: Vibration, Buckling
and Ballistic Impact
GRÉDIAC Michel, HILD François
Full-Field Measurements and Identification in Solid Mechanics
GROUS Ammar
Fracture Mechanics – 3-volume series
Analysis of Reliability and Quality Control – Volume 1
Applied Reliability – Volume 2
Applied Quality Control – Volume 3
RECHO Naman
Fracture Mechanics and Crack Growth
2011
KRYSINSKI Tomasz, MALBURET François
Mechanical Instability
SOUSTELLE Michel
An Introduction to Chemical Kinetics
2010
BREITKOPF Piotr, FILOMENO COELHO Rajan
Multidisciplinary Design Optimization in Computational Mechanics
DAVIM J. Paulo
Biotribolgy
PAULTRE Patrick
Dynamics of Structures
SOUSTELLE Michel
Handbook of Heterogenous Kinetics
2009
BERLIOZ Alain, TROMPETTE Philippe
Solid Mechanics using the Finite Element Method
LEMAIRE Maurice
Structural Reliability
2007
GIRARD Alain, ROY Nicolas
Structural Dynamics in Industry
GUINEBRETIÈRE René
X-ray Diffraction by Polycrystalline Materials
KRYSINSKI Tomasz, MALBURET François
Mechanical Vibrations
KUNDU Tribikram
Advanced Ultrasonic Methods for Material and Structure Inspection
SIH George C. et al.
Particle and Continuum Aspects of Mesomechanics