Design Stiffness Value for Dumbbell-Shaped Steel Tube Concrete Arch Ribs

Literature Overview

This paper by Wei Jiangang and colleagues from Fuzhou University, published in the Journal of Fuzhou University (Natural Science) (2007, Vol. 35, No. 4, pp. 582-587), addresses a fundamental but often overlooked aspect of steel tube concrete (STC) arch bridge design: the determination of the design stiffness value for dumbbell-shaped arch ribs. The research is supported by the Fujian Provincial Major Science and Technology Project (2003F007) and is based on measured data from an actual STC dumbbell-shaped arch bridge. The paper highlights the inconsistency in stiffness definitions across different national codes and proposes practical recommendations for stiffness value selection in engineering practice.

Core Technical Content

The authors identify that different design codes—whether Chinese, European, or American—define the effective stiffness of STC members differently. Some codes use the full gross section stiffness (EI of the uncracked composite section), while others introduce reduction factors to account for cracking, creep, and the nonlinear interaction between steel and concrete. For dumbbell-shaped arch ribs, which consist of two parallel steel tubes with a concrete core or web connecting them, the stiffness determination is further complicated by the geometry and the load path through the composite section.

The paper compares several stiffness calculation methods:

1. Full gross section method — using the complete composite section without any reduction factor. This tends to overestimate the actual stiffness, particularly under serviceability limit state conditions where cracking may have occurred.
2. Reduced stiffness method — applying a reduction factor (typically 0.4 to 0.8) to the gross section stiffness to account for cracking and long-term deformation.
3. Effective section method — using an effective moment of inertia that accounts for the cracked section properties, as prescribed in some codes for reinforced concrete members.

The authors propose that the design stiffness value should be selected based on the specific loading condition and the limit state being checked. For strength design (ultimate limit state), the full gross section stiffness may be acceptable, while for serviceability checks (deflection, vibration), a reduced stiffness value is more appropriate.

Material and Manufacturing Quality Considerations

The dumbbell-shaped STC arch rib is a complex composite member that requires careful attention to both material properties and fabrication quality. The two parallel steel tubes are typically made from high-strength steel pipe conforming to standards such as GB/T 8163, ASTM A53 Grade B, or API 5L Gr. B. The concrete infill must be of sufficient quality to ensure bond and composite action with the steel tubes.

The fabrication of the dumbbell section involves welding the two steel tubes together with transverse stiffeners or webs, and then infilling the cavity with concrete. The welding quality is critical, as any defect in the welded joints could compromise the load transfer between the two tubes and reduce the effective composite action. Non-destructive testing (NDT) of all welded joints, including ultrasonic testing (UT) for volumetric defects and magnetic particle testing (MT) for surface-breaking defects, should be performed in accordance with ASME Section V or GB/T 3323.

Code Comparison and Design Recommendations

The paper's central contribution is the comparison of stiffness definitions across different codes and the proposal of a practical approach for stiffness value selection. The key insight is that the choice of stiffness value has a direct impact on the calculated internal forces, deformations, and stability of the arch rib, and therefore on the overall safety and serviceability of the bridge.

For the specific case of the dumbbell-shaped STC arch bridge studied by the authors, the measured deflection data were compared with analytical predictions using different stiffness values. The results showed that the full gross section stiffness overestimated the actual stiffness by approximately 20-30%, while a reduced stiffness factor of 0.7-0.8 provided a reasonable match with the measured data.

Based on these findings, the authors recommend the following approach for design:

• Use a reduced stiffness factor of 0.7-0.8 for serviceability limit state checks (deflection, vibration).
• Use the full gross section stiffness for ultimate limit state checks (strength, stability), as the composite section remains largely intact at failure.
• Apply additional caution for long-term deflection predictions, considering the effects of concrete creep and shrinkage.

Engineering Practice and Reflections

In reviewing this paper, I am struck by the practical importance of the stiffness value selection problem, which is often treated as a minor detail in design practice. The authors' approach of using measured data from an actual bridge to calibrate the stiffness value is a commendable methodology that bridges the gap between theoretical analysis and real-world performance.

However, I would note that the paper does not fully address the time-dependent behavior of the STC arch rib, including the progressive loss of composite action due to creep, shrinkage, and potential corrosion of the steel tubes. In long-span arch bridges, these time-dependent effects can lead to significant deflection over the service life, and the initial stiffness value may not be representative of the long-term behavior.

Furthermore, the paper does not discuss the impact of construction sequence on the stiffness and behavior of the arch rib. The erection method—whether by cantilever construction, incremental launching, or cable-stayed erection—can introduce residual stresses and deformations that affect the initial stiffness and the subsequent behavior under load.

In conclusion, this paper makes a valuable contribution to the understanding of STC arch rib stiffness by highlighting the inconsistencies in code provisions and proposing a practical approach for stiffness value selection based on measured data. Engineers involved in the design of STC arch bridges should carefully consider the appropriate stiffness value for each limit state check and should not rely solely on the full gross section stiffness for all calculations.