Nonlinear Aerostatic Stability of a Long-Span Concrete-Filled Steel Tubular Arch Bridge
1. Overview of the Study and Its Engineering Significance
This paper from South China University of Technology addresses a problem that has caused real headaches in bridge engineering: the aerostatic stability of a long-span CFST arch bridge under wind loading. The authors analysed the Maocaojie Bridge, a 368 m main-span CFST arch, using an incremental-iterative method that simultaneously considers geometric nonlinearity and the nonlinearity of the static wind load. They examined the effects of wind load nonlinearity, initial wind attack angle, rise-to-span ratio, and cross-brace arrangement on the lateral stability of the arch.
For a veteran who has supplied steel pipe for dozens of arch bridges, this paper is not academic abstraction. The CFST arch is one of the most elegant and efficient bridge forms, but it is also one of the most wind-sensitive. The arch is a curved, slender structure. Under lateral wind, it experiences a complex combination of bending, torsion, and axial compression. If the wind load is treated as a constant linear force, the analysis can be dangerously optimistic. The paper's key contribution is to show that the nonlinearity of the wind load itself — the fact that the wind pressure changes as the structure deforms — can significantly reduce the critical buckling load.
2. Interpretation of the Technical Analysis
The authors used an incremental-iterative method, which is essentially a load-controlled or displacement-controlled Newton-Raphson approach. They tracked the load-displacement path until the tangent stiffness matrix became singular, indicating the limit point of the aerostatic instability. The parameters they varied are summarised below.
| Parameter | Range Studied | Effect on Lateral Stability | Practical Implication |
|---|---|---|---|
| Wind load nonlinearity | Linear vs. nonlinear | Nonlinear wind load reduces critical wind speed by up to 15-20% | Ignoring this can lead to unconservative design |
| Initial wind attack angle | 0° to ±10° | Positive attack angle (wind blowing upward) reduces stability; negative angle increases it | The worst case is not always the horizontal wind |
| Rise-to-span ratio (f/L) | 1/4 to 1/8 | Lower f/L (shallower arch) reduces lateral stability | Shallow arches need more lateral bracing |
| Cross-brace arrangement | Various | Dense cross-bracing significantly improves lateral stability | Cross-braces are cheap compared to the cost of a collapsed bridge |
The most striking finding is the effect of the initial wind attack angle. Many designers assume that the wind acts horizontally, but in reality, the wind can have a vertical component, especially in mountainous terrain or near coastal areas. A positive attack angle (wind blowing upward) reduces the vertical load on the arch, which reduces the axial compression and thus the stability. But it also creates an upward lift that can reduce the contact between the arch and the deck, potentially causing a different failure mode. This is a subtle but critical point.
3. Connection to Steel Pipe Manufacturing and Delivery
Now let me bring this down to the shop floor. The main tubes of a CFST arch bridge are typically large-diameter LSAW or spiral welded pipes, with diameters ranging from 800 mm to 1500 mm and wall thicknesses from 16 mm to 30 mm. The steel grade is usually Q345, Q390, or Q420. The pipe is filled with high-performance concrete after erection. The manufacturing challenges are enormous:
- The pipe must be manufactured with tight dimensional tolerances because the arch geometry is sensitive to any deviation. A 1% error in the pipe diameter can cause a 3% error in the section stiffness, which can shift the buckling load by a similar amount.
- The longitudinal welds of LSAW pipes are the weak points. Under lateral wind, the arch experiences bending about its weak axis, which puts the longitudinal welds in tension. I have seen weld defects in LSAW pipes that were not detected by the initial UT but were found by TOFD during re-inspection after the pipe had been bent into the arch shape. The bending introduces residual stresses that can open up lack-of-fusion defects.
- The cross-braces, which the paper identifies as critical for lateral stability, are usually made from smaller-diameter seamless pipes. These are welded to the main chords using full-penetration butt welds or fillet welds. The weld quality is crucial because the cross-braces carry the lateral load. I have seen cross-brace welds fail during erection because the welder did not preheat the joint properly, leading to hydrogen-induced cracking in the heat-affected zone.
On the delivery side, the pipes for an arch bridge are usually shipped in long segments, sometimes 12 m or more. They must be supported at multiple points to prevent sagging, and the ends must be protected with caps to prevent debris from entering. I once had a shipment of 1400 mm diameter pipes arrive with one end dented because the truck driver had used a chain sling without a spreader bar. That dent was in the middle of a chord segment, and we had to cut out 2 m of pipe and weld in a new piece. That cost us three days of erection time and a lot of money. The lesson is simple: you cannot treat a large-diameter pipe like a piece of rebar. It is a precision structural element.
4. Reflections and Recommendations
This paper reminds me that bridge engineering is a discipline where the analysis is only as good as the assumptions, and the assumptions are only as good as the data. The authors have done a thorough job of including geometric nonlinearity and wind load nonlinearity, but they have not addressed the effect of manufacturing tolerances on the actual buckling load. In practice, the pipe is not perfectly circular, the wall thickness is not perfectly uniform, and the residual stresses from welding are not perfectly symmetric. These imperfections can reduce the buckling load by 5-10% compared to the ideal analysis.
My recommendations for fellow engineers working on CFST arch bridges are:
- Always perform a nonlinear aerostatic stability analysis that includes both geometric and wind load nonlinearity. Do not rely on a linear eigenvalue analysis.
- Consider the worst-case wind attack angle, which may not be horizontal. Consult the local wind climate data and use a range of ±10°.
- Specify tight dimensional tolerances for the steel pipe: diameter tolerance of ±0.5%, wall thickness tolerance of ±1%, and out-of-roundness of less than 1%.
- Require 100% UT and TOFD inspection of the longitudinal welds, and 100% MT of the cross-brace welds after erection.
- During transport and erection, use proper lifting points and supports. Never use a chain sling directly on a large-diameter pipe without a spreader bar.
In conclusion, the Maocaojie Bridge study is a valuable reference for the design of long-span CFST arch bridges, but the real message is that the bridge is only as stable as the pipe that makes it. A perfect analysis on an imperfect pipe is still a perfect analysis on a flawed structure. We must bridge the gap between the computer model and the physical reality by controlling the manufacturing quality and the erection process with the same rigour that we apply to the analysis.
Zhuojin Pipe Fitting Co., Ltd