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

Dumbbell-Shaped CFST Arch Bridges: Design Parameters and Practical Construction Insights

Literature Overview and Core Findings

The paper by Chen Lirong and Chen Baochun, published in the Journal of Guangxi University (Natural Science Edition) in 2016, presents a comprehensive statistical analysis of concrete-filled steel tube (CFST) dumbbell-shaped section arch bridges based on a large collection of completed bridge data. This is a valuable reference for both designers and fabricators because the dumbbell section — two parallel steel tubes connected by steel webs — is one of the most common cross-sectional forms used in CFST arch bridges in China, yet its design parameters have often been selected by intuition rather than systematic analysis.

The study addresses three critical questions: what bridge span structures have been built, what are the force characteristics of the dumbbell-shaped arch rib, and what construction methods are commonly used. The authors then statistically analyse the main design parameters, including arch axis form, rise-span ratio, width-span ratio, rib span-height ratio, rib section height-width ratio, material properties, and the arrangement of cross-bracing and hangers.

Key Technical Parameters and Statistical Findings

The statistical analysis yields several important parameter ranges that should guide future design work. The table below summarises the key findings.

Parameter Statistical Range Dominant Value Practical Significance
Rise-span ratio (f/L) 0.2–0.25 0.2 (about 50% of bridges) The 0.2 value dominates because it balances arch thrust against rib slenderness and provides a favourable aesthetic profile.
Rib span-height ratio (L/h) 30–50 Increases with span Longer spans require deeper ribs to control deflection and maintain stiffness.
Section height-width ratio (h/d) 2–3 (over 90% of bridges) About 2.5 The dumbbell shape is inherently taller than wide, and this ratio is relatively insensitive to span length.
Arch axis form — Parabolic or catenary These forms minimise bending moments under dead load.
Cross-bracing arrangement — Varies with span Transverse stiffness is critical for out-of-plane stability, especially during construction.

The finding that the height-width ratio concentrates around 2.5 regardless of span length is particularly useful. It suggests that the dumbbell section has a natural geometric efficiency that designers have converged upon through experience, and it provides a starting point for preliminary sizing.

Interpretation from a Manufacturing and Construction Perspective

From my years on the shop floor and on site, I can add several practical observations that complement the paper's findings.

First, the dumbbell section presents a significant fabrication challenge. The two main steel tubes must be connected by steel webs, and the welding sequence is critical to control distortion. In my experience, the webs should be welded symmetrically and with controlled heat input to prevent the tubes from bowing. The weld quality at the web-to-tube junctions is often the weak point in these structures, and I have seen cases where lack of fusion or incomplete penetration at these joints led to premature cracking during concrete filling.

Second, the concrete filling process for dumbbell sections deserves special attention. The two tubes and the connecting web region form a complex cavity, and ensuring complete filling without voids requires careful design of the pouring sequence and venting. Self-compacting concrete is often used, but even with SCC, the web region can trap air if the vent holes are not properly positioned. I have witnessed a bridge where the web region was only partially filled, and the resulting stiffness reduction was not detected until the bridge was loaded.

Third, the rise-span ratio of 0.2 is not merely an aesthetic choice. It has direct implications for the horizontal thrust at the springing, which must be resisted by the abutment or tie system. A lower rise-span ratio increases the thrust, which can drive up foundation costs. The statistical dominance of 0.2 suggests that this ratio provides a good balance between rib weight and foundation cost for typical spans.

Practical Recommendations and Lessons Learned

Based on my experience and the insights from this paper, I offer the following recommendations for engineers designing dumbbell-shaped CFST arch bridges.

  1. For preliminary design, adopt a rise-span ratio of 0.2 for spans up to 200 m, and consider increasing to 0.25 for longer spans where rib stiffness becomes more critical.
  2. Use a section height-width ratio of 2.5 as a starting point, but verify that the resulting section provides adequate out-of-plane stiffness during construction, when the arch is often unsupported.
  3. Pay particular attention to the welding of the web-to-tube connections. Use full-penetration welds where feasible, and perform 100% ultrasonic testing on these joints.
  4. During concrete filling, ensure that vent holes are provided at the highest points of the web region, and use a visual inspection method such as borescope examination to verify complete filling.
  5. Consider the construction sequence carefully. The arch is most vulnerable during the concrete filling stage, before the concrete has gained sufficient strength. Temporary bracing or cable stays may be required.

Study Insights and Implications

This paper is a valuable contribution because it moves CFST arch bridge design from an art to a more systematic science. The statistical analysis of existing bridges provides a solid basis for parameter selection, and the identified ranges can be used as a sanity check for new designs.

However, I would caution against over-reliance on statistical ranges. These parameters are derived from bridges that were built successfully, but they do not necessarily represent optimal values for all conditions. Site-specific factors such as foundation conditions, seismic risk, and construction method availability must always take precedence over statistical averages.

The paper also highlights a gap in the literature: there is limited information on the long-term performance of dumbbell-shaped CFST arch bridges, particularly regarding the effects of concrete creep and shrinkage on the composite action between the steel tube and the concrete core. This is an area where more research is needed, and where field monitoring of existing bridges would provide valuable data.

Reference Value and Outlook

For engineers involved in the design and construction of CFST arch bridges, this paper provides a practical reference for preliminary parameter selection. The statistical ranges are particularly useful for feasibility studies and concept design, where detailed analysis is not yet warranted.

Looking forward, I expect to see continued evolution in dumbbell-shaped CFST arch bridge design, driven by advances in high-performance concrete, improved welding techniques, and more sophisticated analysis tools. The integration of monitoring systems into these bridges will also provide valuable feedback on actual performance, which can be used to refine design parameters in the future.

In conclusion, this paper is a solid reference that I would recommend to any engineer working on CFST arch bridges. It provides a clear statistical basis for parameter selection, and it highlights several practical considerations that are often overlooked in design. The lessons learned from the analysis, combined with my own experience on the shop floor and on site, reinforce the importance of careful attention to fabrication details and construction sequence in ensuring the long-term performance of these structures.