## Subscribe \$4.99/month

Un-lock Verified Step-by-Step Experts Answers.

## Textbooks & Solution Manuals

Find the Source, Textbook, Solution Manual that you are looking for in 1 click.

## Tip our Team

Our Website is free to use.
To help us grow, you can support our team with a Small Tip.

## Holooly Tables

All the data tables that you may search for.

## Holooly Help Desk

Need Help? We got you covered.

## Holooly Arabia

For Arabic Users, find a teacher/tutor in your City or country in the Middle East.

Products

## Textbooks & Solution Manuals

Find the Source, Textbook, Solution Manual that you are looking for in 1 click.

## Holooly Arabia

For Arabic Users, find a teacher/tutor in your City or country in the Middle East.

## Holooly Help Desk

Need Help? We got you covered.

## Q. 10.1

Consider a three-DOF spring mass system shown in Figure 10.5. Determine its frequencies and mode shapes. Given: $m_{1}$ = 2 m, $m_{2}$ = 3 m, $m_{3}$ = 2 m; $k_{1}$ = k, $k_{2}$ = 2 k, $k_{3}$ = 2 k.

## Verified Solution

The equations of motion can be written by considering the dynamic equilibrium of each mass:

$m_{1}\ddot{x} _{1}+k_{1}(x_{1}-x_{2})=0$

$m_{2}\ddot{x} _{2}+k_{2}(x_{2}-x_{3})+k_{1}(x_{2}-x_{1})=0$               (i)

$m_{3}\ddot{x} _{3}+k_{3}x_{3}+k_{2}(x_{3}-x_{2})=0$

It may be noted that the DOF may be numbered either from top or from bottom. In the present case, these have been numbered from top. The equations will change accordingly. The mass matrix can be written as follows:

It is given that: $m_{1}$ = 2 m, $m_{2}$ = 3 m, and $m_{3}$ = 2 m

$[M]=\left[\begin{matrix} 2m & 0 & 0 \\ 0 & 3m & 0 \\ 0 & 0 & 2m \end{matrix} \right]$

The stiffness matrix can be written as follows by rearranging Equation (i) in matrix form:

$[K]\left[\begin{matrix} k_{1} & -k_{1} & 0 \\ -k_{1} & k_{1}+k_{2} & -k_{2} \\ 0 & -k_{2} & k_{2}+k_{3} \end{matrix} \right]$

Given $k_{1}$ = k, $k_{2}$ = 2 k, and $k_{3}$ = 2 k.

On substitution, we get

$[K]=k\left[\begin{matrix} 1 & -1 & 0 \\-1 & 3 & -2 \\ 0 & -2 & 4 \end{matrix} \right]$

$det\left|K-m\omega ^{2}\right| =\left|\begin{matrix} k-2m\omega ^{2} & -k & 0 \\ -k &3k-3m\omega^{2} & -2k \\ 0 & -2k & 4k-2m\omega ^{2} \end{matrix} \right|$

Let k/m = β

The expression for determinant can be written as follows:

$det\left|K-m\omega ^{2}\right| =\left|\begin{matrix} \beta -2\omega ^{2} & -\beta & 0\\ -\beta & 3\beta -3\omega ^{2}& -2\beta \\ 0 & -2\beta & 4\beta -2\omega ^{2} \end{matrix} \right|$

Or, it can be expanded as follows:

$6(\omega ^{2})^{3}-21\beta (\omega ^{2})^{2}+16\beta ^{2}(\omega ^{2})-2\beta ^{3}=0$

Its three positive roots are given as follows:

$\left\{\omega ^{2}\right\} =\left\{\begin{matrix} 2.477\beta \\ 0.866\beta \\ 0.155\beta \end{matrix} \right\}$ or $\left\{\omega \right\} =\left\{\begin{matrix} 1.574 \\ 0.93 \\ 0.394 \end{matrix} \right\} \sqrt{\frac{k}{m} }$

Now mode shapes can be determined for each frequency using the following expression:

$[K-m\omega ^{2}]\left\{X\right\} =\left\{0\right\}$

$\left[\begin{matrix}\beta -2\omega ^{2} & -\beta & 0 \\ -\beta & 3\beta -3\omega ^{2} & -2\beta \\ 0 &-2\beta & 4\beta -2\omega ^{2} \end{matrix} \right] \left\{\begin{matrix} x_{1}\\ x_{2} \\ x_{3} \end{matrix} \right\} =\left\{\begin{matrix} 0 \\ 0 \\ 0 \end{matrix} \right\}$     (ii)

Let us obtain the eigenvector corresponding to $\lambda _{3}$ = $\omega ^{2}_{3}$ = 2.477 β. Its value is substituted in Equation (ii), we get

$\left[\begin{matrix} -3.954 & -1 & 0 \\ -1 &-4.431 & -2 \\ -1 & a-2& -0.954\end{matrix} \right] \left\{\begin{matrix} x_{13} \\ x_{23} \\ x_{33} \end{matrix} \right\} =\left\{\begin{matrix} 0 \\ 0 \\ 0 \end{matrix} \right\}$

Upon simplification,

$-3.954 x_{13}-x_{23}+0=0$    (iii)

$-x_{13}-4.431 x_{23}-2 x_{33}=0$     (iv)

$-2 x_{23}-0.954 x_{33}=0$     (v)

Equation (iii) and Equation (v) give,

$\frac{x_{13}}{x_{23}} =\frac{-1}{3.954}$ and $\frac{x_{23}}{x_{33}} =\frac{-0.954}{2} =\frac{-0.477}{1}$

If the value of $x_{33}$ in terms of $x_{23}$ is substituted in Equation (iv), it gives the same ratio of $x_{13}$ and $x_{23}$ as before.

∴ In such cases, only the ratio of the elements is determined. A numerical value for each element of the eigenvector {$x_{3}$} may be obtained by arbitrarily assigning any one element as unity. For example, if $x_{13}$ is assigned a unity, then

$x_{23}$ = −3.954 and $x_{33}$ = 8.289

This procedure is repeated for other eigenvalues as well.
For first mode, $\omega _{1}^{2}$ = 0.155 β, and $x_{11}$ = 1

$\left\{\begin{matrix} x_{11} \\ x_{21} \\ x_{31} \end{matrix} \right\} =\left\{\begin{matrix} 1.000 \\ 0.689 \\ 0.374 \end{matrix} \right\}$

For second mode, $\omega _{2}^{2}$ = 0.866 β , and $x_{12}$ = 1

$\left\{\begin{matrix} x_{12} \\ x_{22} \\ x_{32} \end{matrix} \right\} =\left\{\begin{matrix} 1.364\\ -1 \\ -0.882 \end{matrix} \right\}$

For third mode $\omega _{3}^{2}$ = 2.477 β, and $x_{13}$ = 1

$\left\{\begin{matrix} x_{13} \\ x_{23} \\ x_{33} \end{matrix} \right\} =\left\{\begin{matrix}1.0\\ -3.954 \\ 8.289 \end{matrix} \right\}$

Alternatively, it can be normalized with respect to $x_{33}$ = 1 instead of $x_{13}$ = 1

or,    $\left\{\begin{matrix} x_{13} \\ x_{23} \\ x_{33} \end{matrix} \right\} =\left\{\begin{matrix}0.1206\\ -0.477 \\ 1.0 \end{matrix} \right\}$

The mode shape matrix can be written as follows by arranging them column-wise or row-wise. In this case they have been arranged column-wise.

$[\Phi ]=\left[\begin{matrix} \phi _{11} &\phi _{12} & \phi _{13} \\ \phi _{21} & \phi _{22} & \phi _{23} \\ \phi _{31} & \phi _{32} & \phi _{33} \end{matrix} \right]$

where in $\phi _{ij}$ i represents DOF and j represents mode number

$[\Phi ]=\left[\begin{matrix} 1.000 & 1.364 & 0.121 \\ 0.689 & -1 & -0.477 \\ 0.374 & -0.882 & a1.0\end{matrix} \right]$

The mode shapes are shown in Figure 10.6.

Let us verify the results using MATLAB.
The command eig(K, M) gives two matrices V and D. Matrix V is for eigenvectors
arranged column-wise and matrix D is a diagonal matrix consisting of eigenvalues on the diagonal. The frequencies are obtained as follows:

λ = ω²

>>[V, D] = eig(K, M)

V =

$\begin{matrix} -0.5195 & -0.4741& 0.0734\\ a-0.&3582 0.3476 & −0.2902 \\ −0.1942 & 0.3066 & 0.6069 \end{matrix}$

D =

$\begin{matrix} 0.1552 & 0 & 0 \\ 0 & 0.8665 & 0 \\ 0 & 0 & 2.4782 \end{matrix}$

∴                          $\omega ^{2}=\left\{\begin{matrix} 0.1552 \\ 0.8665 \\2.4782 \end{matrix} \right\} \frac{k}{m}$ or $\omega =\left\{\begin{matrix}0.394\\ 0.930 \\ 1.574 \end{matrix} \right\} \sqrt{\frac{k}{m} }$

In this option, the eigenvalues are obtained in increasing order.
Alternatively, the following command can be given in MATLAB.

## Script Files

>> [V, D] = eig(M, K)

It again generates two matrices V and D. Matrix V is for eigenvectors arranged column-wise and matrix D is a diagonal matrix consisting of eigenvalues on the diagonal. The frequencies are obtained as follows:

λ = 1 / ω²

>>[V, D] = eig(M, K)

V =

$\begin{matrix} −0.0466 & 0.5093 & 1.3185 \\ 0.1844 & −0.3734 & 0.9092 \\ −0.3855 & −0.3294 & 0.4928 \end{matrix}$

D =

$\begin{matrix} 0.4035 & 0 & 0 \\ 0 & 1.1540 & 0 \\ 0 & 0 & 6.4425 \end{matrix}$

$\omega ^{2}=\left\{\begin{matrix} \frac{1}{0.4035} \\ \\ \frac{1}{1.1540} \\ \\ \frac{1}{6.4425} \end{matrix} \right\} \frac{k}{m}$ or $\omega =\left\{\begin{matrix} 1.574 \\0.93 \\ 0.394 \end{matrix} \right\} \sqrt{\frac{k}{m} }$

In this option, the eigenvalues are obtained in decreasing order.