3.4 Dynamic load effects and fatigue considerations

Dynamic Loads on Bridges

Characteristics of Dynamic Loads

Dynamic loads are forces that change over time, causing oscillations and vibrations in a bridge structure. This contrasts with static loads (like the bridge's own dead weight), which remain constant. Understanding dynamic loads matters because they can amplify stresses well beyond what a simple static analysis would predict.

A bridge's response to dynamic loading depends on three properties:

• Natural frequency — how fast the structure naturally wants to vibrate, determined by its mass, stiffness, and geometry
• Damping ratio — how quickly vibrations die out due to energy dissipation
• Mode shapes — the specific patterns of deformation the structure takes on when vibrating

The dynamic amplification factor (DAF) quantifies how much larger the structural response is under dynamic loading compared to the equivalent static load. A DAF of 1.3, for example, means the dynamic response is 30% greater than the static case.

The main types of dynamic loads on bridges include:

• Traffic-induced vibrations from moving vehicles
• Wind loads, especially on long-span structures
• Seismic forces from earthquake ground motion
• Impact loads such as vessel collisions with piers

If left unaddressed, these loads can cause fatigue damage, excessive deflections, or resonance (which occurs when the loading frequency matches the structure's natural frequency, causing vibrations to grow dramatically).

Structural Dynamics Concepts

The simplest dynamic model is the single degree-of-freedom (SDOF) system, which represents a bridge as a mass-spring-damper. This works for basic analysis, but real bridges vibrate in multiple patterns simultaneously. Multi-degree-of-freedom (MDOF) systems capture these multiple vibration modes and give a more realistic picture of structural behavior.

Two main types of dynamic analysis apply to bridges:

• Forced vibration analysis examines the structure's response to known external loads, producing displacement, velocity, and acceleration time histories.
• Random vibration analysis handles stochastic loads (like turbulent wind) using probabilistic methods to predict the range of likely structural responses.

The natural frequency of a simple system is calculated as:

$f_n = \frac{1}{2\pi}\sqrt{\frac{k}{m}}$

where $k$ is the stiffness and $m$ is the mass. Getting this value right is critical for checking whether any expected loading frequency could trigger resonance.

Bridge Response to Dynamic Loads

Wind-Induced Effects

Wind creates several distinct types of dynamic loading on bridges, each with different mechanisms:

• Vortex shedding — When wind flows around bluff bodies like deck sections or cables, it sheds vortices alternately from each side. This creates periodic lateral forces that can lock in to the structure's natural frequency.
• Flutter — A coupled torsional and vertical oscillation where the wind feeds energy into the structure faster than damping can remove it. Flutter is self-exciting and can lead to catastrophic failure. The 1940 Tacoma Narrows Bridge collapse is the classic example.
• Buffeting — Random vibrations caused by turbulent wind flow. Unlike vortex shedding, buffeting forces are irregular and broadband.
• Galloping — Large-amplitude, low-frequency oscillations that primarily affect slender elements like cables and hangers, often triggered by ice accretion changing their aerodynamic profile.

For long-span bridges, aerodynamic stability analysis is essential. Engineers use wind tunnel testing on scaled models and computational fluid dynamics (CFD) simulations to evaluate susceptibility to these phenomena before construction.

Seismic Response Analysis

Earthquake loading enters a bridge through ground acceleration, which is characterized by acceleration time histories recorded from past earthquakes or synthetically generated to match a target hazard level.

Response spectra are a key design tool. They plot the peak response (acceleration, velocity, or displacement) of SDOF systems across a range of natural periods, giving engineers a quick way to estimate how strongly a bridge will respond to a given earthquake.

The seismic analysis process typically involves several steps:

1. Site-specific seismic hazard analysis — Evaluate local geology and fault proximity to determine design ground motions for the bridge location.
2. Modal analysis — Perform eigenvalue analysis to find the bridge's natural frequencies and mode shapes.
3. Response analysis — Either run a full time-history analysis (solving equations of motion via numerical integration) or use response spectrum analysis, which combines modal responses using statistical methods like SRSS (square root of sum of squares) or CQC (complete quadratic combination).
4. Soil-structure interaction — Account for how the foundation soil modifies ground motions and changes the overall system dynamics. Soft soils, for instance, can amplify certain frequency components.

Mitigation Strategies

Engineers have several tools to reduce dynamic response in bridges:

Energy dissipation devices absorb kinetic energy and convert it to heat:

• Viscous dampers force fluid through orifices to resist motion
• Friction dampers dissipate energy through controlled sliding at interfaces

Base isolation systems decouple the superstructure from ground motion:

• Lead-rubber bearings provide both flexibility and energy dissipation in a single unit
• Friction pendulum systems use curved sliding surfaces so the structure's period shifts away from dominant earthquake frequencies

Tuned mass dampers (TMDs) are large masses attached to the structure and tuned to vibrate out of phase with the problematic mode, absorbing vibrational energy. These are common on pedestrian bridges and long-span structures susceptible to wind excitation.

All of these strategies rely on computational methods for design and verification. Finite element analysis models complex structural behavior under dynamic loads, while numerical integration techniques (such as Newmark's method) solve the time-dependent equations of motion.

Fatigue in Bridge Components

Fatigue Mechanisms and Influencing Factors

Fatigue is the progressive accumulation of damage under repeated loading. Even when individual stress cycles are well below the material's yield strength, thousands or millions of repetitions can initiate and grow cracks until a component fails. For bridges, the primary sources of cyclic loading are traffic, wind, and thermal fluctuations.

Several factors control how quickly fatigue damage accumulates:

Material properties:

• Fatigue strength (or endurance limit) defines the stress level below which a material can theoretically survive infinite cycles
• Crack growth rate determines how fast an existing flaw propagates under cyclic loading

Stress concentration factors amplify local stresses at geometric discontinuities. These are especially problematic at welded connections, where the toe of the weld and weld terminations create sharp transitions that act as stress risers. Holes, notches, and cutouts have the same effect.

Environmental factors compound the problem:

• Corrosion reduces the effective cross-sectional area and creates surface pitting that acts as crack initiation sites
• Temperature fluctuations induce thermal stresses that add to the cyclic load count

Applied stress characteristics that most influence fatigue life include the stress range ($\Delta\sigma$, the difference between maximum and minimum stress in a cycle), the mean stress level, and the frequency of loading cycles. Of these, stress range is typically the dominant variable.

Fabrication and Design Considerations

Fabrication processes introduce residual stresses that affect fatigue performance before a bridge ever sees traffic. Welding, for example, creates tensile residual stresses near the weld that effectively raise the mean stress and reduce fatigue life. Cold-forming operations create their own residual stress gradients.

Initial flaws are equally important. Material imperfections like inclusions and voids, or fabrication defects like incomplete weld fusion and surface scratches, act as pre-existing crack initiation sites. A component with a small initial flaw can have a dramatically shorter fatigue life than an identical defect-free component.

Design detailing has a major impact on fatigue resistance:

• Avoid abrupt changes in geometry that create stress concentrations
• Use smooth transitions and generous radii at connections
• Minimize the number of welded attachments in high-stress regions

On material selection: higher-strength steels allow thinner sections, but thinner sections experience larger stress ranges under the same load. This means high-strength steels can actually be more susceptible to fatigue cracking than lower-strength alternatives, despite their superior static strength. Fatigue-resistant alloys are available for critical components where this trade-off matters.

Fatigue Analysis for Bridge Performance

Fatigue Life Prediction Methods

Three main approaches are used to predict fatigue life in bridge components:

1. S-N Curve Approach (Stress-Life)

This method relates stress amplitude to the number of cycles to failure and is the standard for high-cycle fatigue (typically greater than $10^5$ cycles). The relationship follows:

$N = A(\Delta\sigma)^{-m}$

where $N$ is cycles to failure, $\Delta\sigma$ is the stress range, and $A$ and $m$ are material/detail constants. On a log-log plot, this produces a straight line, making it straightforward to look up expected life for a given stress range.

2. Fracture Mechanics Approach (Crack Growth)

When a crack already exists or is assumed to exist, Linear Elastic Fracture Mechanics (LEFM) predicts how fast it will grow. Paris' law describes the crack growth rate:

$\frac{da}{dN} = C(\Delta K)^m$

where $a$ is crack length, $N$ is the number of cycles, $\Delta K$ is the stress intensity factor range, and $C$ and $m$ are material constants. Engineers integrate this equation from an initial flaw size to a critical crack length to estimate remaining life.

3. Miner's Rule (Cumulative Damage)

Real bridges experience variable-amplitude loading, not constant stress cycles. Miner's rule sums the fractional damage from each stress level:

$\sum \frac{n_i}{N_i} = 1$

where $n_i$ is the number of applied cycles at stress level $i$ and $N_i$ is the number of cycles to failure at that stress level from the S-N curve. When the sum reaches 1.0, failure is predicted. This is a simplification (it ignores load sequence effects), but it's widely used in practice.

Probabilistic fatigue analysis adds another layer by accounting for uncertainties in loading, material properties, and initial flaw sizes. Monte Carlo simulation generates thousands of fatigue life scenarios to build a statistical picture, and the resulting reliability index quantifies the probability of fatigue failure over the design life.

Design and Assessment Tools

Fatigue detail categories are the primary design tool in practice. The AASHTO Bridge Design Specifications classify structural details into categories ranging from A (best fatigue resistance, such as plain rolled steel) to E' (worst, such as certain welded attachments). Each category has its own S-N curve, so the designer selects the appropriate category for each connection detail and checks that the expected stress range stays below the allowable value for the required number of cycles.

Finite element analysis (FEA) determines local stress concentrations at fatigue-critical locations. Submodeling techniques allow engineers to build a coarse global model of the entire bridge, then create a refined local model of a specific connection to extract accurate stress histories for fatigue assessment.

Non-destructive testing (NDT) methods detect and monitor fatigue cracks in existing bridges:

• Ultrasonic testing sends sound waves through the material to identify internal flaws
• Magnetic particle inspection reveals surface and near-surface cracks by applying magnetic fields and iron particles to the area of interest

Structural health monitoring (SHM) systems provide ongoing, real-time fatigue assessment. Strain gauges installed at critical locations measure actual stress cycles during service. Data analysis algorithms then use these measured histories, combined with S-N curves or fracture mechanics models, to predict remaining fatigue life and flag components that need inspection or repair.