Shop-Fabricated Segments for Bridge Lifecycle Performance

Fatigue Design for Traffic Loads

A bridge is a dynamically loaded structure. Every vehicle that crosses it applies a moving point load that cycles through the structure's members and connections. Over a 50-100 year design life, a highway bridge may experience 50-200 million truck load cycles — a fatigue loading regime that no building structure encounters.

Fatigue design for steel bridges is governed by the stress range at fatigue-critical details and the number of cycles. The critical details are not the primary members themselves — a well-designed chord or diagonal is sized for strength, and its fatigue life is usually adequate. The critical details are the connections: the welds attaching gusset plates to chords, the welds connecting transverse stiffeners to webs, the bolt groups at field splices, and the welds at bearing stiffeners. These are the locations where stress concentrations amplify the nominal stress range, and where fatigue cracks initiate.

Our fatigue design approach for steel structure bridges follows:

• Detail classification: Every welded and bolted detail is classified per the fatigue category tables in the applicable code (AASHTO LRFD Table 6.6.1.5.3-1, Eurocode 3 Table 8.1-N, or GB 50017). The category determines the allowable stress range for a given cycle count.
• Stress range calculation: For each fatigue-critical detail, the stress range is calculated from the structural analysis under the fatigue truck loading (a design truck with specified axle spacing and weight, applied at multiple positions to maximize the stress range at the detail).
• Miner's rule cumulative damage: The fatigue damage from trucks of varying weights is accumulated using Miner's rule, which sums the damage ratio (actual cycles / allowable cycles) across all truck weight categories. The cumulative damage must be less than 1.0 at the design life.
• Detail improvement: Where a detail's fatigue capacity is insufficient, the detail is re-engineered: welds are reoriented to reduce stress concentration, grinding of weld toes reduces the stress concentration factor, or bolted splices replace welded splices at high-stress-range locations.

Seismic Isolation Bearings

In seismic zones, the bridge's superstructure (the steel truss or girder) must be isolated from the substructure (the piers and abutments) to prevent the transfer of damaging seismic forces. The isolation is achieved through seismic isolation bearings — devices placed between the superstructure and the substructure that decouple the horizontal motion of the ground from the horizontal motion of the bridge deck.

We engineer three types of seismic isolation bearings, selected based on the bridge's seismic hazard, span, and functional requirements:

• Elastomeric bearings: Layers of rubber interleaved with steel shim plates. The rubber provides horizontal flexibility (allowing the bridge to move relative to the piers), while the steel plates provide vertical stiffness (supporting the bridge's dead and live loads). Suitable for low to moderate seismic zones.
• Lead-rubber bearings (LRB): A lead core is inserted into an elastomeric bearing. The lead core yields under seismic displacement, dissipating energy through hysteretic damping. Suitable for moderate to high seismic zones. The LRB reduces the seismic force transferred to the substructure by 40-60% compared to a non-isolated bridge.
• Friction pendulum bearings (FPB): A sliding bearing with a curved concave surface. The bridge deck slides on the curved surface during an earthquake, with the curvature providing a restoring force that returns the bridge to its centered position. The friction between the sliding surfaces provides energy dissipation. Suitable for high seismic zones and long-span bridges where large displacement capacity is required.

The bearing design is not just a device specification — it is integrated into the structural analysis. The bearings' stiffness and damping properties are modeled in the bridge's seismic analysis, and the reduced seismic forces in the superstructure and substructure are used to optimize member sizes. A bridge with seismic isolation bearings has lighter superstructure members and smaller substructure elements than a non-isolated bridge of the same span — the isolation system pays for itself through structural savings.

The 748 m Covered Bridge: A Building-Bridge Hybrid

Our 748 m covered bridge project represents a structural type that defies conventional categorization. It is a bridge — it carries pedestrians across a river or valley — but it is also a building, with a roof, walls, and enclosed interior space. The structure must resist not only bridge loads (traffic, wind on the deck) but also building loads (wind on the walls and roof, snow on the roof, seismic mass of the envelope).

The engineering challenge of a covered bridge is the interaction between the bridge structure and the building envelope. The bridge trusses carry the deck load and resist the horizontal forces from the envelope. The envelope — roof, walls, end gables — adds wind area that the bridge trusses were not designed for in an open-deck configuration. The truss top chord, which in a conventional bridge is a compression member carrying deck load, becomes a combined compression-bending member carrying both deck load and wind load from the wall above.

The 748 m length of the bridge means thermal expansion is a major design consideration. A 748 m steel bridge experiencing a 60°C temperature change expands by approximately 540 mm at the free end — accommodated by expansion joints at each span break. The covered envelope, however, moderates the temperature range: the enclosed interior experiences less temperature variation than an exposed bridge, reducing the thermal movement demand. This is one of the structural benefits of the covered bridge typology — the envelope that adds wind load also reduces thermal stress.

For the covered bridge, we deployed a modular truss system: the bridge is divided into structural modules of 40-60m, each module shop-fabricated, transported, and bolted together at site. The modular breaks coincide with expansion joint locations, so each module is thermally independent. The bolted field splices are designed as pinned connections (transferring shear but not moment), allowing each module to articulate independently under thermal movement and differential settlement.