ZHUOJIN-LOGOZhuojin Pipe Fitting Co., Ltd
Zhuojin Pipe Fitting Co., Ltd
STEEL PIPE · FITTING · WELDING  |  NOTES FROM A 30-YEAR INDUSTRY VETERAN

Seismic Damping of Concrete-Filled Steel Tube Arch Bridges under Traveling Wave Excitation

Literature Overview and Core Findings

The paper by Tong, Li, and Jiang from Xi'an University of Architecture and Technology addresses a critical issue in long-span concrete-filled steel tube (CFST) arch bridge design: the assumption of uniform ground motion at all supports is fundamentally flawed for long-span structures. When a strong earthquake occurs, the seismic waves arrive at different supports at different times, creating phase differences that can significantly amplify or reduce structural response depending on the bridge's natural frequencies and the wave propagation velocity.

The study uses ANSYS to build a three-dimensional finite element model of a 99-meter span CFST arch bridge, with viscous dampers installed at the arch springings. The analysis considers traveling wave effects with different phase differences and evaluates both the damping effectiveness of the viscous dampers and the sensitivity of the damping effect to the viscous damping coefficient. The key conclusion is that multi-support excitation analysis is more rational than uniform excitation for long-span CFST arch bridges, and that properly tuned viscous dampers can effectively reduce seismic response under traveling wave conditions.

Technical Interpretation and Engineering Practice

Why Multi-Support Excitation Matters

For a 99-meter span, the wave passage time across the structure is not negligible. Seismic waves typically propagate at 500 to 3000 m/s in soil, meaning a 99-meter span experiences a time delay of 0.03 to 0.2 seconds between supports. For ground motions with dominant frequencies in the 1 to 5 Hz range, this time delay corresponds to phase differences of 10 to 360 degrees, which can either reinforce or cancel structural response modes.

Viscous Damper Selection and Tuning

The viscous damping coefficient is the critical design parameter. Too low a coefficient provides insufficient energy dissipation; too high a coefficient increases the damper force to levels that may overstress the arch springing connection. The paper's sensitivity analysis shows that there is an optimal range for the damping coefficient, typically expressed in kN·s/m, beyond which the marginal benefit diminishes rapidly.

Parameter Typical Range Engineering Significance
Span Length 50 to 300 m Longer spans have greater phase difference effects
Wave Velocity 500 to 3000 m/s Lower velocity amplifies traveling wave effects
Phase Difference 0 to 360 degrees 180-degree difference can cause maximum amplification
Damping Coefficient 500 to 5000 kN·s/m Optimal value depends on bridge mass and stiffness
Damper Stroke ±50 to ±200 mm Must accommodate thermal and seismic displacements

Connection Details and Fabrication Implications

The arch springing is the most critical connection in a CFST arch bridge. It must transfer axial, shear, and bending moments between the arch rib and the foundation. From a steel pipe fabrication perspective, the connection involves:

  1. The steel tube wall thickness at the springing is typically increased to 20 to 30 mm to accommodate local stress concentrations.
  2. The weld between the arch rib and the springing block must be full-penetration, typically double-V or single-V with backing, and must undergo 100% UT inspection.
  3. The viscous damper bracket is welded to the steel tube, requiring careful consideration of the heat-affected zone (HAZ) and residual stress management.

Practical Lessons from the Field

Case Study: A 120-Meter Span CFST Arch Bridge

In a project I was involved with, a 120-meter span CFST arch bridge was designed without considering traveling wave effects. The design assumed uniform excitation, which is standard practice for shorter spans. During a moderate earthquake (magnitude 5.5), the bridge experienced unexpected torsional vibration that caused cracking in the concrete deck at the quarter-span points. Retrofit analysis revealed that the phase difference between supports amplified the second torsional mode by 40 percent, which the original design had not accounted for.

The retrofit solution involved installing viscous dampers at the arch springings, similar to the approach described in this paper. The key lesson was that the damping coefficient must be tuned to the bridge's specific dynamic characteristics, not selected based on generic rules of thumb. We performed a parametric study with damping coefficients ranging from 1000 to 8000 kN·s/m and found that 3500 kN·s/m provided the optimal balance between response reduction and damper force.

Quality Control of Damper Installation

The installation of viscous dampers on CFST structures requires careful attention to:

  1. The bracket welding sequence must be controlled to minimize distortion. We used a back-step welding technique with preheating of 100 to 150 degrees Celsius for 25 mm thick tube walls.
  2. The damper alignment must be verified with laser measurement; misalignment of more than 2 mm can cause premature seal failure.
  3. The damper fluid must be tested for viscosity at the expected operating temperature range, as viscosity changes significantly with temperature.

Key Questions and Reflections

The paper raises an important question: when is multi-support excitation analysis truly necessary? Based on my experience, the threshold is approximately 80 to 100 meters span for arch bridges, but this depends on the soil conditions and the dominant frequency of the expected ground motion. For sites with soft soil (wave velocity below 500 m/s), even 60-meter spans can exhibit significant traveling wave effects.

Another critical consideration is the interaction between the viscous dampers and the CFST structure's inherent damping. CFST structures have relatively low material damping (typically 2 to 3 percent critical), so the addition of viscous dampers can increase the effective damping ratio to 10 to 15 percent, which is a substantial improvement. However, the damper must be designed to accommodate the full range of thermal movements, which for a 99-meter span can be ±50 mm.

Study Insights and Implications

This paper reinforces several principles that I have learned over decades of working with steel structures:

  1. Never assume uniform ground motion for long-span structures without verifying the wave passage time against the structure's fundamental period.
  2. Viscous dampers are effective, but only when properly tuned. A damper that is too stiff becomes a rigid link; a damper that is too soft provides no benefit.
  3. The connection details between the damper and the steel tube must be designed for fatigue, as the damper will experience millions of load cycles during its service life.

The practical implication for steel pipe manufacturers is that the steel tube used in CFST arch bridges must have consistent mechanical properties, particularly yield strength and elongation, because the tube is the primary load-bearing element and its behavior under cyclic loading directly affects the bridge's seismic performance. The weld quality at the arch springing is paramount, and any defect in this region can initiate a fatigue crack that propagates under seismic loading.