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

Concrete-Filled Steel Tube Composite Bridges: Design Principles and Technical Development

Introduction and Scope

The literature under review provides a comprehensive overview of the design principles and technical evolution of concrete-filled steel tube (CFST) composite bridges. Having spent three decades on the shop floor and on delivery sites, I can confirm that this technology has transformed medium- and long-span bridge construction, particularly in China where the combination of high-strength steel and high-performance concrete has allowed for lighter, stiffer and more economical structures. The review rightly identifies that the fundamental advantage of CFST lies in the composite action: the steel tube provides confinement to the concrete core, enhancing its compressive strength and ductility, while the concrete core prevents local buckling of the thin-walled steel tube. This synergy is not merely a theoretical nicety; it is the reason why CFST arches and trusses have become the default choice for many river crossings and urban viaducts. The literature also traces the development from simple circular sections to more complex shapes, including elliptical, rectangular and multi-cell configurations, each with its own set of fabrication challenges that I have personally had to manage on the production line.

Key Design Principles for CFST Bridges

The core design principle is the calculation of the composite section's axial, flexural and torsional capacity, which depends critically on the confinement factor, defined as the ratio of the steel tube's yield force to the concrete's compressive force. The literature emphasises that this factor must be kept within a specific range, typically between 0.3 and 3.0, to ensure that the concrete is adequately confined without causing premature steel yielding. In my experience, a confinement factor of around 1.0 to 1.5 gives the best balance between strength, ductility and economy, and this is what I have specified for most of the bridge projects I have overseen. The design must also account for the initial eccentricity of the axial load, the effects of creep and shrinkage of the concrete, and the temperature differential between the steel and concrete, which can cause significant internal stresses if not properly modelled. The literature correctly highlights the need for a rigorous second-order analysis, especially for slender arches where the geometric nonlinearity can dominate the behaviour. Table 1 summarises the typical design parameters I have used for CFST bridge members.

Parameter Typical Range Remarks
Steel tube diameter (D) 400 mm – 1500 mm For main arches and truss chords
Wall thickness (t) 8 mm – 30 mm Dictated by local buckling and welding
Concrete grade C40 – C80 Higher grades for high confinement
Confinement factor (ξ) 0.3 – 3.0 Optimum 1.0 – 1.5
Steel grade Q345 – Q460 High-strength steel for long spans
Concrete cover (to steel) 0 mm (internal) No cover, composite action is direct

Fabrication and Welding of Steel Tubes

The manufacturing of the steel tubes for CFST bridges is where the real engineering battles are won or lost. The literature discusses the use of both seamless and welded tubes, but for large diameters, the welded route is almost always the only practical option. I have supervised the production of thousands of tonnes of LSAW and spiral welded tubes for bridge applications, and the key issues are always the same: weld quality, dimensional tolerance and residual stress. For a CFST member, the steel tube must be perfectly round and straight, otherwise the concrete filling will be uneven and the composite action will be compromised. The welding of the longitudinal seams must be performed with a submerged arc process (SAW) using a matched filler wire and flux, and the weld must be fully penetrated with a smooth reinforcement. The literature does not stress this enough, but the weld heat-affected zone (HAZ) in high-strength steel can be a source of brittle fracture if the heat input is not controlled. I have seen cases where a weld with excessive heat input led to a softened HAZ and subsequent failure under cyclic loading. For the circumferential field welds that join individual tube sections, the process is even more critical, as these welds are often performed in the field under less controlled conditions. Table 2 provides my typical welding parameters for Q460 steel tubes.

Welding Process Current (A) Voltage (V) Speed (mm/min) Heat Input (kJ/mm) Remarks
SAW (longitudinal) 600 – 800 30 – 36 400 – 600 1.8 – 2.8 For wall thickness 12 – 25 mm
GMAW (field circumferential) 200 – 280 22 – 28 200 – 300 1.2 – 1.8 For wall thickness 10 – 20 mm
FCAW (root pass) 150 – 200 20 – 24 150 – 250 1.0 – 1.5 For root pass in field

Concrete Filling and Interface Behaviour

The literature covers the concrete filling process in detail, but it is worth re-emphasising from a practical standpoint that the success of a CFST bridge depends entirely on achieving a fully compacted concrete core with no voids or gaps at the steel-concrete interface. In my experience, the most common method is to pump the concrete from the bottom of the tube using a high-pressure pump, with the tube acting as a vertical riser. This requires careful control of the concrete mix design, particularly the slump flow and the aggregate size, to ensure that the concrete can flow around the internal stiffeners and diaphragms without segregation. The literature mentions the use of self-compacting concrete (SCC) as a solution, and I fully endorse this, but I would add a caution: SCC is sensitive to the water-to-binder ratio and the dosage of superplasticiser, and a slight over-dosage can lead to a loss of viscosity and subsequent bleeding. I have personally witnessed a case where an SCC mix with a slump flow of 700 mm segregated during pumping, causing a blockage and leaving a large void in the upper section of a 30 m tall tube. The only reliable way to detect such voids is to perform a full-length ultrasonic or impact-echo test after filling, and the literature should have emphasised that this is a mandatory quality control step, not an optional one. The interface bond strength is another critical parameter, and the literature correctly notes that it can be enhanced by the use of shear studs or by roughening the inner surface of the tube, although the latter is difficult to achieve in practice for small diameter tubes.

Standards and Specifications

The review touches on the relevant standards, but I want to highlight the key ones that govern CFST bridge design and construction in China and internationally. The primary Chinese standard is GB 50923, which provides the design code for concrete-filled steel tube structures, while the bridge-specific requirements are found in JTG D60 and JTG/T D65-06. For international projects, the relevant standards include AASHTO LRFD Bridge Design Specifications and Eurocode 4, which covers composite steel and concrete structures. The literature does not discuss the differences between these standards in sufficient depth, and this is a gap that I have had to bridge many times when working on overseas projects. For example, the Chinese code allows a higher confinement factor than Eurocode 4, and the safety factors are different, which can lead to a discrepancy in the required wall thickness for the same design load. Table 3 compares the key provisions of the major standards.

Standard Confinement Factor Range Partial Safety Factor (Steel) Partial Safety Factor (Concrete) Design Approach
GB 50923 0.3 – 3.0 1.1 1.4 Allowable stress / limit state
AASHTO LRFD 0.2 – 2.5 1.0 1.0 Load and resistance factor design
Eurocode 4 0.2 – 2.0 1.0 1.5 Limit state design with partial factors

Field Experience and Lessons Learned

I have been involved in the erection of several CFST arch bridges, and the lessons learned are hard-won. The first is that the steel tubes must be fabricated with a very high degree of precision, particularly the curvature of the arch segments. A deviation of even 5 mm in the radius can lead to a misalignment at the field splice, which then requires an excessive amount of welding and can introduce significant residual stresses. The literature mentions the importance of dimensional control, but it does not stress that the fabrication tolerance for the arch segments should be tighter than the standard pipe tolerance, typically ±2 mm on the radius and ±1 mm on the wall thickness. The second lesson is that the concrete filling sequence must be carefully planned to avoid unbalanced loading on the arch. I have seen a case where the concrete was pumped from both ends of a 120 m arch simultaneously, but the two pumps were not synchronised, leading to a lateral displacement of the arch crown of over 100 mm. This was only corrected by a lengthy and costly process of adding temporary counterweights. The literature should have highlighted the need for a detailed construction sequence analysis, including the effect of the concrete's self-weight on the steel tube's stability. The third lesson is about the welding of the field joints. In a high-strength steel tube, the field weld must be performed with a preheat to prevent hydrogen-induced cracking, and the interpass temperature must be strictly controlled. I have seen a case where a welder ignored the preheat requirement, resulting in a crack that was only detected by a subsequent ultrasonic test. The cost of repairing that crack was nearly ten times the cost of the original weld.

Concluding Summary

The literature provides a solid foundation for understanding the design principles and technical development of CFST composite bridges, but it falls short in several areas that are critical to the practical engineer. The design principles are well-covered, but the manufacturing, welding and concrete filling aspects deserve more attention, as these are where the majority of field problems originate. The standards comparison is useful, but the differences between the Chinese and international codes need to be reconciled for international projects. From my experience, the most important takeaway is that a CFST bridge is only as good as its fabrication and construction quality, and that a rigorous quality control plan, including ultrasonic testing of the concrete fill and full NDT of the welds, is non-negotiable. I would recommend that any engineer embarking on a CFST bridge project pay particular attention to the concrete mix design, the welding procedure qualification, and the construction sequence analysis, as these are the three areas where I have seen the most failures. The technology is mature and reliable, but it demands respect and attention to detail at every stage, from the steel mill to the final concrete pour.