6.4 Tuned mass dampers

Tuned Mass Dampers: Concept and Applications

Fundamental Principles of Tuned Mass Dampers

A tuned mass damper (TMD) is a passive vibration control device attached to a primary structure to reduce its dynamic response. It consists of a secondary mass, spring, and damper, and it works by "absorbing" vibrational energy that would otherwise cause large-amplitude motion in the primary structure.

The core idea relies on resonance: the TMD's natural frequency is tuned to match (or nearly match) the problematic frequency of the primary structure. When the primary structure begins to vibrate at that frequency, the TMD oscillates out of phase, generating inertial forces that counteract the structure's motion.

TMD effectiveness depends on three key parameters:

• Mass ratio $\mu$: the TMD mass divided by the primary structure's effective modal mass, typically ranging from 1% to 10%. Larger mass ratios give better performance but add more weight.
• Frequency ratio: the ratio of the TMD's natural frequency to the target frequency of the primary structure. This is usually set slightly below 1.0 (see Den Hartog's formula below).
• Damping ratio: optimized based on the mass and frequency ratios. It controls how broad and how deep the vibration reduction is across the frequency range.

Applications and Design Variations

TMDs are most useful for structures susceptible to sustained or periodic vibrations:

• Tall buildings experiencing wind-induced sway
• Bridges subject to wind or pedestrian-induced oscillations
• Other structures vulnerable to seismic or machinery-induced vibrations

Design variations accommodate different vibration types. Translational TMDs handle linear motion (e.g., lateral sway of a building), while rotational TMDs address angular or torsional motion. For structures with multiple problematic modes, engineers may install several TMDs, each tuned to a different frequency.

Two well-known implementations:

• Taipei 101 uses a 728-ton steel pendulum suspended near the top of the tower to counteract wind-induced sway.
• London's Millennium Bridge was retrofitted with multiple viscous dampers and TMDs after pedestrian footfall caused unexpected lateral oscillations on opening day.

Designing Tuned Mass Dampers for Vibration Control

Optimal Parameter Selection

The design process follows a logical sequence:

1. Choose the mass ratio $\mu$. A larger TMD mass gives better vibration reduction but increases structural load. Values of 1–5% are common; up to 10% is used in demanding applications.
2. Calculate the optimal frequency ratio using Den Hartog's classical formula (for an undamped primary structure under harmonic forcing):

$f_{opt} = \frac{1}{1 + \mu}$

where $\mu$ is the mass ratio. This means the TMD is tuned slightly below the primary structure's natural frequency.

1. Determine the TMD stiffness from the desired natural frequency: $k_d = m_d \cdot \omega_d^2$, where $m_d$ is the TMD mass and $\omega_d = f_{opt} \cdot \omega_n$ (with $\omega_n$ being the primary structure's natural frequency).
2. Optimize the damping ratio. For the Den Hartog case, the optimal damping ratio of the TMD is:

$\zeta_{opt} = \sqrt{\frac{3\mu}{8(1+\mu)^3}}$

Too little damping produces a sharp but narrow reduction; too much damping broadens the effect but reduces the peak attenuation. The optimum balances both.

The frequency ratio is deliberately set below 1.0 partly to account for potential frequency shifts from environmental changes, aging, or added mass over the structure's lifetime.

Practical Design Considerations

• Space constraints influence whether a pendulum, sliding mass, or other TMD configuration is feasible. Pendulum TMDs need vertical clearance; sliding-mass designs need horizontal travel room.
• Weight limitations are especially critical when retrofitting existing structures that weren't designed for the additional load.
• Maintenance and retuning should be planned from the start. Material properties and structural characteristics can drift over time, degrading TMD performance if left unchecked.

For situations where conditions change significantly, semi-active TMDs can adjust their stiffness or damping in real time (e.g., using magnetorheological fluids). These improve performance across a wider range of operating conditions compared to purely passive designs.

Multiple TMD configurations are used when a single TMD can't adequately cover all critical modes, or when robustness against frequency mistuning is a priority. Distributing several smaller TMDs across slightly different frequencies can be more robust than relying on one large unit.

Optimizing Tuned Mass Dampers for System Requirements

Analytical Optimization Techniques

For simple systems (undamped primary structure, single harmonic excitation), Den Hartog's closed-form solutions for $f_{opt}$ and $\zeta_{opt}$ provide an excellent starting point. When the primary structure itself has significant damping, or when the loading is random (e.g., wind or seismic), these formulas no longer give exact optima. In those cases, numerical optimization is required.

Sensitivity analysis is a critical step: you vary each TMD parameter (mass, stiffness, damping) around its nominal value and observe how performance changes. A good design should be relatively insensitive to small parameter variations, since real-world conditions never match the model perfectly.

Advanced Optimization Approaches

Real TMD design often involves multi-objective optimization, balancing:

• Performance: maximum vibration reduction at the target frequency
• Cost: material, fabrication, and installation expenses
• Practicality: maintenance burden, physical size, and expected service life

Nonlinear effects may also need attention. If the primary structure behaves nonlinearly at large amplitudes (e.g., geometric nonlinearity in a slender tower), or if the TMD itself exhibits nonlinear restoring forces, the linear optimization results can be inaccurate. Adjustments or fully nonlinear simulations become necessary.

For complex problems involving multiple TMDs, constrained design spaces, or nonlinear behavior, engineers turn to computational optimization algorithms:

• Genetic algorithms explore the design space by evolving a population of candidate solutions
• Particle swarm optimization iteratively improves candidate solutions based on collective search behavior

These methods are well-suited to problems where the objective function is non-smooth or has many local optima.

Evaluating Tuned Mass Dampers for Vibration Reduction

Performance Metrics and Analysis Methods

TMD performance is quantified through several metrics:

• Peak displacement reduction of the primary structure at the target frequency
• Acceleration amplitude decrease, which directly relates to occupant comfort in buildings
• Stress reduction in critical structural elements, improving fatigue life

Frequency response function (FRF) analysis is the primary tool for assessing TMD effectiveness. The FRF relates the output (displacement, velocity, or acceleration) to the input (applied force) as a function of frequency:

$H(\omega) = \frac{X(\omega)}{F(\omega)}$

Without a TMD, the FRF shows a single sharp resonance peak. With a properly tuned TMD, that peak "splits" into two smaller peaks on either side of the original resonance. The height of these split peaks relative to the original peak directly indicates how much vibration reduction the TMD provides.

Time-domain simulations complement frequency-domain analysis by evaluating performance under realistic loading:

• Transient response to impulse or step inputs (e.g., sudden wind gusts)
• Random vibration analysis under stochastic excitations like wind turbulence or earthquake ground motion

Experimental Validation and Long-term Monitoring

Validation typically proceeds in stages:

1. Scaled model testing allows rapid prototyping and parameter tuning at reduced cost.
2. Full-scale prototype testing in a controlled laboratory environment validates design assumptions before installation.
3. In-situ measurements on the actual structure verify that the TMD performs as predicted under real-world conditions.

Long-term monitoring programs track TMD effectiveness over months or years. Sensors collect vibration data continuously, helping engineers identify performance degradation or the need for retuning. This data also feeds back into improved design methods for future projects.

Performance comparisons between different TMD configurations, or between TMDs and alternative strategies (active control systems, semi-active systems, base isolation), help engineers select the most appropriate solution for each application.