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

Design Method for CFST Arch Bridges with Cosine-Varying Rib Height

Background and Why This Paper Matters

I have spent thirty years watching steel pipe go into bridges, and the concrete-filled steel tube (CFST) arch is one of the most elegant uses of our product. The rib is the spine of the bridge, and its cross-section height drives both the steel tonnage and the erection sequence. For decades, the standard approach to variable cross-section design has been the Li Te formula, which ties the section height to the arch axis geometry in a way that is mathematically sound but not always fabrication-friendly. This paper by Zhou Shuixing and colleagues offers a fresh design method that defines the rib height as the crown height divided by the cosine of the arch axis horizontal inclination angle raised to a power, with that power uniquely determined by the crown height, the springing height, and the arch axis equation. The authors then validate the method against two real bridges through dynamic characteristic and elastic stability analysis.

What caught my attention is the practical framing. The paper explicitly targets the common "equal width, variable height" section form used in large-span CFST arches. In my experience, the Li Te formula often produces sections that are difficult to fabricate because the height variation is not smooth enough, leading to awkward transitions in the web members and chord splices. The cosine-based method is inherently smoother, which means the steel fabricator can plan the chord member splices and web member layout with far less rework. The reported differences in chord axial force between the new method and the design documents are 1.7% to 7.0%, and the differences in dynamic characteristics, elastic stability, and web member stress are all within 5%. For a bridge designer, that is an acceptable margin. For a pipe supplier, it means the steel section schedule is predictable and the fabrication tolerances are achievable.

Technical Interpretation of the Cosine Method

The core of the method is the definition of the rib height h(x) as a function of the arch axis inclination angle theta(x):

Parameter Definition Practical Significance
h(x) Rib height at position x along the arch axis Controls the section modulus and thus the bending capacity
h_crown Rib height at the crown The minimum height, typically set by the need for the concrete fill and the web member spacing
h_springing Rib height at the springing The maximum height, driven by the bending moment at the springing
theta(x) Horizontal inclination angle of the arch axis Derived from the arch axis equation, typically a parabola or catenary
n Power exponent, determined by h_crown, h_springing, and the arch axis equation The unique solution that satisfies the boundary conditions

The power exponent n is the key innovation. In the Li Te formula, the exponent is often assumed or fitted to a specific arch geometry, which can lead to inconsistencies when the arch axis changes. The new method solves for n exactly, given the crown and springing heights and the arch axis equation. This eliminates the need for iteration and makes the design process more deterministic. The authors also provide a recommended range for n based on the common rise-to-span ratios and arch axis coefficients used in CFST arch bridges, which is a practical gift to designers who want to apply the method without re-deriving the mathematics.

From a manufacturing standpoint, the cosine-based height variation means the steel tube chord members will have a smoothly varying section along the length. In practice, most CFST arches use constant diameter chords with the height variation achieved through the web members. The new method does not change that fundamental fabrication approach, but it does make the web member layout more rational. The web members, typically K or N shaped, will have more uniform stress distribution because the chord axial forces are more consistent along the length. This is a real benefit for the welding shop, because it reduces the number of different web member sizes and simplifies the fit-up.

Comparison with the Li Te Formula and Engineering Practice

I have sat in design reviews where the Li Te formula produced a rib height profile that required the web members to be fabricated in five or six different lengths, with some of them differing by only a few millimeters. That is a nightmare for the pipe fabricator, because each different length means a separate cutting setup, a separate fit-up jig, and a separate welding procedure qualification if the joint geometry changes. The cosine method, by virtue of its smooth variation, tends to produce a more regular web member pattern. The authors' validation against two actual bridges shows that the axial force differences are within 7%, which is well within the typical design safety factor of 1.5 to 2.0 for CFST arches.

The dynamic and stability results are also reassuring. The differences in natural frequencies and elastic buckling loads are within 5%, which means the new method does not introduce any unexpected structural behavior. For the pipe supplier, this is important because it means the steel section schedule derived from the cosine method will not require any special high-strength or thick-wall pipe that would be outside the standard product range. The method is compatible with the common grades of Q345, Q370, and Q420 steel pipe that are used in CFST arch bridges.

One practical lesson I would share with younger engineers is that the fabrication tolerances for the rib height are often more critical than the design calculations themselves. If the steel fabricator cannot hold the rib height within the specified tolerance, the actual section modulus will deviate from the design value, and the bridge will not perform as predicted. The cosine method does not solve this fabrication problem, but it does make it easier to manage because the height variation is smooth and the tolerance requirements are more uniform along the length.

Key Questions and Reflections

The paper raises an important question that is not fully addressed: how does the cosine method handle the transition from the rib to the springing and the foundation? In my experience, the springing region is where the steel tube is most heavily loaded, and the connection to the concrete foundation is a critical detail. The cosine method defines the rib height at the springing as an input, but it does not provide guidance on the transition geometry. This is where the designer must apply judgment, and where the pipe fabricator must coordinate closely with the civil contractor.

Another question is the effect of the concrete fill on the effective stiffness of the rib. The paper treats the CFST section as a composite, but the actual behavior depends on the bond between the steel tube and the concrete core. If the concrete shrinks or if the bond is compromised, the composite action is reduced, and the rib height becomes less effective. This is a known issue in CFST bridges, and the cosine method does not address it directly. The designer must still apply the appropriate reduction factors for the concrete contribution, which are typically specified in the relevant design codes.

From a delivery perspective, the cosine method has a subtle but important implication for the steel pipe procurement. The rib height variation means that the chord members will have varying bending moments along the length, but the axial forces are the dominant load. The pipe wall thickness is usually governed by the axial force and the local buckling limit. The cosine method produces a more uniform axial force distribution, which means the wall thickness can be more consistent along the length. This is a benefit for the pipe mill, because it reduces the number of different wall thicknesses that need to be rolled, and it simplifies the welding of the chord splices.

Study Insights and Implications

This paper is a solid contribution to the design literature for CFST arch bridges. It offers a mathematically rigorous and practically applicable method for determining the rib height variation, and it validates the method against real bridge projects. The reported differences are within acceptable engineering tolerances, and the method is compatible with standard steel pipe grades and fabrication practices.

For the steel pipe industry, the main takeaway is that the cosine method simplifies the fabrication planning for CFST arch ribs. The smoother height variation reduces the number of different web member sizes, and the more uniform axial force distribution allows for more consistent wall thickness selection. This translates into fewer change orders, less rework, and a more predictable delivery schedule. I would recommend that designers consider this method for future CFST arch bridges, particularly those with large spans where the rib height variation is significant.

The paper also reminds us that the design method is only one part of the equation. The fabrication quality, the concrete filling procedure, and the erection sequence are equally important. A well-designed rib with poor fabrication will fail, and a poorly designed rib with excellent fabrication will still perform below expectations. The cosine method is a tool, not a solution. The solution comes from the integration of design, fabrication, and construction, and that is where the industry veteran's experience matters most.