Dynamics: Multi-Degree-of-Freedom Vibration Introduction

Moving beyond the simple oscillators you’ve mastered, real-world engineering systems—from vehicle suspensions and aircraft wings to multi-story buildings—involve multiple parts moving in complex, interdependent ways. Analyzing these systems requires a robust, matrix-based framework that reveals not just if something will vibrate, but how it will vibrate through its distinct patterns of motion, known as mode shapes. This introduction to multi-degree-of-freedom (MDOF) systems provides the mathematical foundation for predicting these behaviors, essential for designing structures and machines that are safe, efficient, and reliable.

From Single to Multiple Degrees of Freedom

A degree of freedom (DOF) is defined as an independent coordinate needed to completely specify a system’s configuration. A simple mass on a spring has one DOF: its vertical displacement. Now, imagine two masses connected by springs to each other and to fixed walls. To describe this system’s position at any instant, you need two independent coordinates: the displacement of mass one, $x_1(t)$, and the displacement of mass two, $x_2(t)$. This is a classic two-DOF system, the simplest entry point into MDOF analysis. The core challenge shifts from solving a single differential equation to solving a set of coupled differential equations, where the motion of one mass directly influences the motion of the other through the connecting spring.

The power of the MDOF approach lies in its systematic methodology. Instead of tackling coupled equations directly, we use matrix notation to organize the system's physical properties. This leads to an eigenvalue problem, the solution of which yields the system's intrinsic dynamic properties: its natural frequencies and mode shapes. These concepts are fundamental to understanding resonance, performing vibration isolation, and executing more advanced techniques like modal analysis.

Deriving the Equations of Motion & Matrix Formulation

Using Newton’s second law or Lagrange’s equations for the two-mass, three-spring system, you derive two second-order differential equations. For free, undamped vibration, they take the form:

$$m_1{\ddot{x}}_1+(k_1+k_2)x_1-k_2{x}_2=0$$ $$m_2{\ddot{x}}_2-k_2{x}_1+(k_2+k_3)x_2=0$$

The critical step is recognizing the pattern and expressing this system compactly in matrix form:

$$\left[\begin{array}{cc}{m}_1& 0\\ 0& m_2\end{array}\right]\left\{\begin{array}{c}{\ddot{x}}_1\\ {\ddot{x}}_2\end{array}\right\}+\left[\begin{array}{cc}(k_1+k_2)& -k_2\\ -k_2& (k_2+k_3)\end{array}\right]\left\{\begin{array}{c}{x}_1\\ x_2\end{array}\right\}=\left\{\begin{array}{c}0\\ 0\end{array}\right\}$$

This is written universally as:

$$\mathbf{M}\ddot{\mathbf{x}}+\mathbf{Kx}=0$$

Here, $\mathbf{M}$ is the mass matrix and $\mathbf{K}$ is the stiffness matrix. For this simple system, the mass matrix is diagonal because the coordinates are associated directly with discrete masses. The stiffness matrix is symmetric and its off-diagonal terms, $-k_2$, represent the coupling between degrees of freedom. These coupling terms are what make the equations interdependent; if they were zero, the system would behave as two independent single-DOF oscillators.

The Eigenvalue Problem: Solving for Natural Frequencies and Mode Shapes

To find the system's natural vibrations, we assume a harmonic solution of the form $\mathbf{x}(t)=\mathbf{u}{e}^{i\omega t}$, where $\mathbf{u}$ is a constant vector describing the shape of the vibration and $\omega$ is its frequency. Substituting this into the matrix equation of motion leads to the generalized eigenvalue problem:

$$\left(\mathbf{K}-{\omega }^2\mathbf{M}\right)\mathbf{u}=0$$

For a non-trivial solution ($\mathbf{u}\ne 0$), the determinant of the coefficient matrix must be zero:

$$\det \left(\mathbf{K}-{\omega }^2\mathbf{M}\right)=0$$

This is known as the characteristic equation. For a two-DOF system, it yields a quadratic equation in ${\omega }^2$. The two positive roots, ${\omega }_1$ and ${\omega }_2$ (where ${\omega }_1<{\omega }_2$), are the system's natural frequencies (in rad/s). The lower one, ${\omega }_1$, is called the first or fundamental frequency.

Each natural frequency, ${\omega }_r$, has an associated mode shape, ${\mathbf{u}}^{(r)}$, found by substituting ${\omega }_r$ back into the eigenvalue equation $(\mathbf{K}-{\omega }_r^2\mathbf{M}){\mathbf{u}}^{(r)}=0$. Since the determinant is zero, the equations are not independent; you can only solve for the ratio of the displacements. For a two-DOF system, you typically express the mode shape relative to the first coordinate: ${\mathbf{u}}^{(r)}=\left\{\begin{array}{c}1\\ u_2^{(r)}/u_1^{(r)}\end{array}\right\}$. The mode shape is a vector describing the relative amplitude and phase of motion of each mass when vibrating purely at that specific natural frequency.

Physical Interpretation of Modes and Introduction to Modal Analysis

Mode shapes reveal the inherent vibration patterns of the system. In the first mode (at ${\omega }_1$), both masses typically move in the same direction; they are "in phase." In the second mode (at ${\omega }_2$), they move in opposite directions; they are "out of phase." A node, a point of zero relative displacement, may appear between the masses. These shapes are intrinsic properties, dependent only on $\mathbf{M}$ and $\mathbf{K}$, much like the shape of a drumhead depends on its boundary conditions.

This leads directly to the core idea of modal analysis. The mode shapes possess a powerful orthogonality property with respect to the mass and stiffness matrices. This allows you to transform the coupled equations of motion in the physical coordinates ($\mathbf{x}$) into a set of uncoupled equations in a new coordinate system called modal coordinates or principal coordinates. Each uncoupled equation looks exactly like that of a single-DOF oscillator. You solve these simple equations independently and then superimpose (sum) the results to reconstruct the total physical response. This is the essence of modal analysis: decomposing complex, coupled motion into a sum of simple, independent modal motions.

Understanding and Identifying Coupling

Coupling between degrees of freedom is a central theme. It means the motion in one coordinate induces forces in another. In our example, coupling is caused by the off-diagonal stiffness term $-k_2$. There are two primary types:

1. Static or Elastic Coupling: Present when the stiffness matrix is non-diagonal.
2. Dynamic or Inertial Coupling: Present when the mass matrix is non-diagonal.

The form of coupling depends entirely on the choice of coordinates. A clever choice of coordinates can sometimes decouple the equations from the start, but for general systems, the modal transformation via the eigenvalue problem is the systematic method to achieve decoupling. Recognizing coupling is crucial for understanding how vibrations transmit through a system and for identifying which components will be most affected by a disturbance at a specific location.

Common Pitfalls

1. Misinterpreting Mode Shape Values: A mode shape vector $\left\{\begin{array}{c}1\\ -2\end{array}\right\}$ does not mean mass two moves twice as far with twice the energy. It indicates mass two moves with twice the amplitude but in the opposite direction to mass one. The absolute magnitude is arbitrary; only the relative ratios between entries are physically meaningful.
2. Confusing Forced and Free Vibration Solutions: The eigenvalue problem solves only for free vibration properties (natural frequencies and mode shapes). It tells you nothing directly about the amplitude of vibration due to a specific force. Forced vibration analysis requires solving the full inhomogeneous equation, often using modal analysis as a tool.
3. Ignoring the Assumptions of the Model: The foundational $\mathbf{M}\ddot{\mathbf{x}}+\mathbf{Kx}=0$ model assumes the system is undamped, linear, and has constant mass/stiffness properties. Introducing damping ($\mathbf{C}\dot{\mathbf{x}}$) complicates the eigenvalue problem significantly, and nonlinearities (like large deformations or contact) render this entire linear analysis invalid.
4. Incorrectly Forming the Stiffness Matrix: A frequent error in setting up $\mathbf{K}$ is misassigning the signs of off-diagonal coupling terms. A reliable method is to ensure the stiffness matrix is symmetric and positive definite for a stable system. The diagonal term $k_{ii}$ represents the total stiffness directly resisting motion at coordinate $i$, while $k_{ij}$ (where $i\ne j$) is the negative of the stiffness coupling coordinate $i$ to coordinate $j$.

Summary

• Multi-degree-of-freedom systems are described by coupled matrix equations of motion: $\mathbf{M}\ddot{\mathbf{x}}+\mathbf{Kx}=0$.
• The system's intrinsic dynamic properties are found by solving the eigenvalue problem $(\mathbf{K}-{\omega }^2\mathbf{M})\mathbf{u}=0$, yielding natural frequencies (${\omega }_r$) and mode shapes (${\mathbf{u}}^{(r)}$).
• Mode shapes are vectors defining the relative displacement pattern of all masses when vibrating at a specific natural frequency; their absolute scale is arbitrary.
• Modal analysis is a powerful technique that uses the orthogonality of mode shapes to transform coupled equations into a set of independent single-DOF equations, which are solved and then superimposed.
• Coupling between degrees of freedom, represented by off-diagonal terms in the mass or stiffness matrices, means the motion in one coordinate influences another. The modal transformation systematically decouples these interactions.