Failure Modes of Rectangular Concrete-Filled Steel Tube Continuous Composite Truss Beams
Overview of the Study
This paper by Ma Guogang, Pu Beichen, Jiang Lei, Liu Yongjian, and Liu Bin from Chang'an University and the China Municipal Engineering Northwest Design and Research Institute tackles a problem that is becoming increasingly common in bridge engineering: how does a rectangular concrete-filled steel tube (CFST) continuous composite truss beam actually fail, and what design parameters control the failure mode. The team used finite element analysis validated against experimental results to trace the full failure process of the main truss, the concrete bridge deck, the deck reinforcement, the core concrete, the yield progression of the members, and the plastic deformation of the members. They then proposed a reasonable failure mechanism based on a section plastic hinge model and studied the reasonable value ranges for a series of design parameters including deck reinforcement ratio, deck thickness, web-to-chord wall thickness ratio, and chord core concrete strength grade.
The study is particularly timely because continuous composite truss bridges are being used more frequently in urban and highway applications, and the design provisions have not always kept pace with the structural behavior. The authors are to be commended for addressing the failure mechanism directly rather than relying solely on elastic analysis.
Core Technical Points
The Failure Process and Plastic Hinge Model
The finite element analysis revealed a clear progression of failure. The process begins with yielding of the chord members at the intermediate support, where the negative bending moment is highest. This is followed by yielding of the web members adjacent to the support, then yielding of the chord at midspan where the positive bending moment is highest. The final failure occurs when a sufficient number of plastic hinges have formed to create a mechanism.
The authors proposed a reasonable failure mechanism based on a section plastic hinge model, which assumes that the composite section can develop a plastic hinge at the locations of maximum moment. This is a significant departure from the elastic design approach, which assumes that the structure remains elastic under service load and that failure occurs at the first yield. The plastic hinge model allows the structure to redistribute moments, which means the ultimate capacity can be higher than the elastic capacity.
Recommended Parameter Ranges
| Parameter | Recommended Range | Effect on Behavior |
|---|---|---|
| Deck reinforcement ratio | 1.39% to 2.31% | Controls negative moment capacity at supports |
| Deck thickness | Not specified | More significant effect on negative moment capacity |
| Web-to-chord wall thickness ratio α | 0.667 to 1.0 | Controls web buckling and shear capacity |
| Chord core concrete strength | Higher grade at intermediate support and midspan | More significant effect on stiffness and capacity |
The deck reinforcement ratio range of 1.39 to 2.31 percent is interesting because it is relatively narrow. Below 1.39 percent, the deck is likely to crack excessively under negative moment, and the stiffness will be compromised. Above 2.31 percent, the reinforcement is likely to yield before the concrete reaches its ultimate strain, which means the section is over-reinforced and the failure will be brittle rather than ductile. This is a classic balanced reinforcement ratio concept, and the authors have identified the practical range for this specific structural configuration.
The web-to-chord wall thickness ratio α of 0.667 to 1.0 is also significant. If the web is too thin relative to the chord, the web will buckle before the chord reaches its yield strength, and the shear capacity will be limited. If the web is too thick, the web will be over-designed and the additional weight will not contribute to the capacity. The range of 0.667 to 1.0 represents the practical band where the web and chord can develop their full strengths simultaneously.
The Effect of Concrete Strength Grade
The study found that increasing the concrete strength grade in the chord at the intermediate support and at midspan had a more significant effect on stiffness and capacity than increasing the concrete strength elsewhere. This makes sense because these are the locations of maximum moment, and the concrete is most effective at resisting compression in these regions. The concrete at the intermediate support is in compression on the bottom side of the chord, and the concrete at midspan is in compression on the top side of the chord.
From a practical standpoint, this means that the designer can specify a higher strength concrete for the chords at these critical locations without having to use the higher strength concrete throughout the entire structure. This can result in significant cost savings, especially if the higher strength concrete requires special mix design or placement techniques.
Manufacturing and Fabrication Considerations
Welding the Web-to-Chord Connections
The web-to-chord wall thickness ratio has a direct impact on the weldability of the connection. When the web is thinner than the chord, the weld size is governed by the web thickness, and the fillet weld can be designed to develop the full strength of the web. When the web and chord have the same thickness, the weld can be a full penetration weld or a large fillet weld, depending on the design.
From a shop floor perspective, the most challenging aspect is the joint preparation for the web-to-chord connection. The web members are typically cut at an angle to fit against the chord, and the fit-up must be tight to ensure adequate weld penetration. If the web is cut with a slight gap, the weld can have a lack of fusion at the root, which is a serious defect in a truss member that is primarily loaded in axial compression or tension.
I have seen projects where the web members were cut with a plasma cutter and the cut surface was not cleaned properly, resulting in a weld with porosity and slag inclusions. The solution is to use a machine-cut edge with a smooth surface and to clean the cut surface with a grinder or wire brush before welding. The weld should be inspected with ultrasonic testing or radiography, especially for the web-to-chord connections at the intermediate support, where the stresses are highest.
Concrete Placement in the Chords
The chord members are typically filled with concrete to increase their stiffness and capacity. The concrete placement in a horizontal chord is relatively straightforward, but the concrete placement in an inclined chord or a chord with a variable cross-section can be more challenging. The concrete must be placed in layers with proper vibration to ensure that it fills the entire cross-section and that there are no voids.
One issue I have seen in practice is the formation of a void at the top of the chord when the concrete is placed from one end and the air cannot escape. This is particularly problematic when the chord is long and the concrete has a low slump. The solution is to place the concrete from both ends or to use a pump with a high pressure to force the concrete through the entire length of the chord. Some projects use self-compacting concrete to address this issue, and that is a sensible approach when the chord geometry is complex.
Engineering Practice and Design Implications
The Importance of the Failure Mechanism
The most important contribution of this study is the identification of the reasonable failure mechanism. In bridge design, the failure mechanism determines the ductility of the structure and its ability to redistribute moments. A structure that fails by forming a plastic hinge at the intermediate support and then redistributing moment to the midspan is ductile and can absorb significant energy before failure. A structure that fails by brittle fracture or by buckling of a single member is not ductile and can fail suddenly.
The practical implication is that the design should ensure that the failure mechanism is the plastic hinge mechanism, not a local buckling or fracture mechanism. This means that the web members should be designed to resist the shear forces without buckling, and the chord members should be designed to develop their full plastic moment capacity. The wall thickness ratios recommended in the study are intended to achieve this.
What the Results Mean for Bridge Design Practice
The recommended parameter ranges provide a practical starting point for the design of rectangular CFST continuous composite truss beams. However, I would caution against treating these ranges as absolute limits. The study is based on a specific set of geometric and material parameters, and the recommended ranges may need to be adjusted for different span lengths, truss depths, and loading conditions.
The deck reinforcement ratio range of 1.39 to 2.31 percent is a good starting point, but the designer should verify the actual reinforcement ratio against the cracking moment and the ultimate moment capacity of the deck. The web-to-chord wall thickness ratio range of 0.667 to 1.0 is a good starting point, but the designer should also check the shear capacity of the web against the actual shear forces at the intermediate support.
Lessons from Real Bridge Projects
I have been involved in several bridge projects that used CFST truss girders, and the lessons from those projects align with the findings of this study. The most common issue is the deck cracking at the intermediate support, which is caused by the negative moment exceeding the cracking moment of the deck. The solution is to provide adequate reinforcement in the deck at the support, and the recommended range of 1.39 to 2.31 percent is consistent with what I have seen in practice.
Another common issue is the web buckling at the intermediate support, which is caused by the shear force exceeding the buckling capacity of the web. The solution is to increase the web thickness or to provide stiffeners at the support. The recommended web-to-chord wall thickness ratio range of 0.667 to 1.0 is consistent with what I have seen in practice, and I would recommend staying at the upper end of this range for bridges with high shear forces.
Concluding Reflection
This study provides a comprehensive analysis of the failure modes of rectangular CFST continuous composite truss beams and identifies the key design parameters that control the failure mechanism. The plastic hinge model is a significant contribution because it allows the designer to consider the ultimate capacity of the structure rather than just the elastic capacity. The recommended parameter ranges for deck reinforcement ratio, deck thickness, web-to-chord wall thickness ratio, and chord concrete strength grade are practical and can be used as a starting point for design. From my perspective, the most important takeaway is that the failure mechanism must be a plastic hinge mechanism, not a local buckling or fracture mechanism. This requires careful attention to the web-to-chord wall thickness ratio and the concrete strength grade in the chords. The study is a valuable addition to the literature, and I would like to see future work that extends the analysis to different span lengths, truss depths, and loading conditions.
Zhuojin Pipe Fitting Co., Ltd