Page 230 - Dynamics of Mechanical Systems
P. 230

0593_C07_fm  Page 211  Monday, May 6, 2002  2:42 PM





                       Inertia, Second Moment Vectors, Moments and Products of Inertia, Inertia Dyadics  211


                       Interestingly, these coefficients may be shown (see Reference 7.4) to be independent of the
                       choice of unit vectors  n  in which  I is expressed.  As a consequence,  I ,  I , and  I  are
                                                                                          II
                                             i
                                                                                                 III
                                                                                       I
                       sometimes called invariants of I.
                        Equation (7.7.12) is a cubic equation for λ which in general has three distinct roots: λ ,
                                                                                                     1
                       λ , and λ . This implies that there are at least three solutions for each at the a , a , and a 3
                                                                                              2
                        2
                                                                                            1
                               3
                       of Eq. (7.7.10). To obtain these solutions, we can select one of the roots (say, λ ) and then
                                                                                            1
                       substitute λ  for λ in Eq. (7.7.10). This produces three equations that are dependent because
                                 1
                       λ  is a solution of Eq. (7.7.11). Hence, only two of the three equations are independent.
                        1
                       However, we can obtain a third independent equation by recalling that n  is a unit vector
                                                                                       a
                       and thus a , a , and a  also satisfy the equation:
                                1
                                         3
                                   2
                                                        a + a + a =  1                         (7.7.16)
                                                         2
                                                             2
                                                                 2
                                                         1   2  3
                       Then, by selecting any two equations from Eq. (7.7.10) together with Eq. (7.7.16), we can
                       solve the resulting system of three equations and determine the values of a , a , and a .
                                                                                             2
                                                                                                   3
                                                                                          1
                        Next, by letting λ be λ , we can repeat the process and find a second set of components
                                            2
                       (a , a , and a ) of a principal unit vector n . Similarly, for λ = λ , we obtain a third principal
                                                                             3
                           2
                                                           a
                        1
                                 3
                       unit vector.
                        By following this procedure we obtain the components of three principal unit vectors
                       defining three principal directions of inertia of the system. The values λ , λ , and λ  are
                                                                                           2
                                                                                                  3
                                                                                        1
                       then the corresponding principal moments of inertia. This procedure and these concepts
                       are illustrated in the example in the following section.
                       7.8  Principal Axes, Principal Moments of Inertia: Example
                       Consider a system whose central inertia matrix relative to mutually perpendicular unit
                       vectors n , n , and n  is:
                               1  2      3
                                              I   I    I     92     −3 4   3 3   4
                                               11  12   13                      
                                         I =  I    I    I   =  −3 4    39 8     3 8             (7.8.1)
                                          ij []  21  22  23                     
                                              I   I    I     33 4    3 8    37 8  
                                              31   32   33                      
                       To determine the principal values, the principal unit vectors, and the principal directions,
                       we need to form Eq. (7.7.10). That is,
                                              ( 92 − ) +− (  3 4)a 2 +  3 3 4a 3  =  0
                                                   λ a
                                                      1
                                               − (  34) +( 39 8 − ) (  3 8) a  =  0             (7.8.2)
                                                                 +
                                                   a
                                                             λ a
                                                    1           2        3
                                              ( 33 4) +(  3 8) a 2 ( 37 8 − ) λ a 3  =  0
                                                                +
                                                     a
                                                      1
                       Because these equations are linear and homogeneous (right sides are zero), there is a
                       nontrivial solution for the a , a , and a  only if the determinant of the coefficients is zero
                                               1  2      3
                       (Eq. (7.7.11)). Hence,
   225   226   227   228   229   230   231   232   233   234   235