Round-Ended Concrete-Filled Steel Tube Short Columns Under Eccentric Compression: Failure Mechanisms and Design Implications
Literature Overview and Core Findings
This paper by Wang Fengqin, Wang Jingfeng, and Shen Qihan from Hefei University of Technology investigates the behaviour of round-ended concrete-filled steel tube (CFST) short columns under eccentric compression. The round-ended section, which consists of a rectangular middle portion with two semicircular ends, is increasingly used in bridge piers and building columns where aesthetic considerations or directional stiffness requirements are important. The paper uses ABAQUS finite element analysis to develop a numerical model, validates it against experimental results, and then conducts a parametric study to examine the effects of eccentricity, concrete strength, steel strength, cross-sectional area, and diameter-to-thickness ratio on the column's performance.
The key findings are that the failure modes under eccentric compression are primarily local indentation combined with end bulging, and excessive rotation due to the eccentric load. Within a certain range of eccentricity, the axial compressive capacity and the maximum bending moment increase with increasing cross-sectional area, steel strength, and concrete strength, but decrease with increasing eccentricity and diameter-to-thickness ratio. The paper also fits an axial force–bending moment interaction curve, which is essential for design.
Technical Interpretation and Practical Implications
The Round-Ended Section: Why It Matters
The round-ended section is a hybrid between a circular and a rectangular section. The semicircular ends provide excellent confinement to the concrete, similar to a circular tube, while the flat middle portion offers a convenient surface for connecting to other structural elements. This makes the section ideal for bridge piers, where the pier cap needs to be supported by a flat surface, and for building columns, where the flat face can accommodate beams or walls.
However, the round-ended section also introduces complexity. The confinement effect is not uniform around the perimeter. The semicircular ends are well confined, but the flat middle portion is less effective at restraining the concrete. This non-uniform confinement leads to a complex stress distribution that cannot be captured by simple analytical models. The paper's finite element analysis is therefore a valuable tool for understanding the behaviour.
Failure Modes and Their Practical Significance
The paper identifies two main failure modes: local indentation with end bulging, and excessive rotation. The first mode is a classic local buckling failure, where the steel tube's wall buckles inward at the point of maximum compression, while the ends bulge outward due to the concrete's lateral expansion. This mode is more likely in columns with high diameter-to-thickness ratios, where the tube wall is relatively thin.
The second mode, excessive rotation, is unique to eccentric loading. When the load is applied at a significant eccentricity, the column tends to rotate as a rigid body, and the failure is governed by the loss of stability rather than by material failure. This is a critical distinction because it means that the column's capacity is limited by its geometric stability, not by the strength of the steel or concrete.
For design engineers, the failure mode determines the appropriate design method. If the failure is by local buckling, then the design must check the local buckling resistance of the steel tube. If the failure is by excessive rotation, then the design must consider the second-order effects and the column's stiffness.
Parametric Effects and Design Guidance
The paper's parametric study provides clear guidance on the effects of various parameters:
| Parameter | Effect on Axial Capacity | Effect on Maximum Moment | Practical Implication |
|---|---|---|---|
| Eccentricity | Decreases | Decreases | Larger eccentricity reduces both capacity and moment, so the design must account for the actual load path. |
| Cross-sectional area | Increases | Increases | A larger section provides more material to resist both axial and bending loads. |
| Steel strength | Increases | Increases | Higher steel strength improves the confinement and the bending resistance. |
| Concrete strength | Increases | Increases | Higher concrete strength improves the core's stiffness and capacity, but the effect is less pronounced than that of steel strength. |
| Diameter-to-thickness ratio | Decreases | Decreases | A thinner wall is more prone to local buckling, reducing the capacity. |
The interaction curve fitted in the paper is a practical tool for design. It allows the engineer to determine the allowable combination of axial load and bending moment for a given section. This is similar to the interaction curves used in reinforced concrete design, but the curve for CFST columns is different because of the confinement effect.
Integration with Manufacturing and Delivery Practice
Tube Fabrication and Dimensional Tolerances
The round-ended section is typically formed by cold-forming a flat steel plate into the desired shape and then welding the longitudinal seam. The manufacturing process is similar to that of rectangular or square tubes, but the die design is more complex because of the semicircular ends. The tube's dimensional accuracy is critical for the structural performance. If the flat middle portion is not perfectly flat, or if the semicircular ends are not truly circular, the confinement effect will be compromised.
From a manufacturing perspective, the diameter-to-thickness ratio is the most important parameter. The paper shows that a high D/t ratio reduces the capacity due to local buckling. In practice, we often specify a maximum D/t ratio of 40 for CFST columns, but the paper suggests that this limit may need to be lower for round-ended sections because of the non-uniform confinement.
Welding and Quality Control
The longitudinal weld in a round-ended tube is typically located on one of the flat faces or at the transition between the flat and curved portions. The weld quality is critical because it is the weakest link in the tube. A weld defect, such as lack of fusion or a slag inclusion, can initiate a crack that propagates under the complex stress state of eccentric loading.
The paper's finite element model assumes a perfect weld, but in practice, the weld should be inspected using ultrasonic testing (UT) or radiographic testing (RT) to ensure its integrity. The heat-affected zone (HAZ) of the weld may have different mechanical properties, which could affect the local buckling behaviour. This is a subtle but important consideration.
Site Handling and Concrete Placement
Round-ended CFST columns are typically filled with concrete after the tube is erected. The concrete placement is more challenging than in circular tubes because the flat middle portion may cause the concrete to segregate or form voids. The use of self-compacting concrete (SCC) is recommended to ensure that the concrete fills the entire section without voids.
The tube must also be able to withstand the hydrostatic pressure of the wet concrete during placement. The flat middle portion is more susceptible to bulging under this pressure than the semicircular ends. The paper's analysis of the confining pressure development is relevant here, although the loading conditions are different.
Key Questions and Reflections
- How does the round-ended section compare with a circular or rectangular section of the same area? The round-ended section offers a compromise between the excellent confinement of a circular section and the practical advantages of a rectangular section. The paper does not provide a direct comparison, but the interaction curve can be used to compare the capacities of different sections.
- What is the effect of the flat middle portion's width on the confinement? The paper does not explicitly address this, but intuitively, a wider flat portion will reduce the confinement effect because a larger area of the concrete is not effectively restrained. This is a topic that warrants further research.
- How does the method account for the steel tube's initial imperfections? Real tubes are not perfectly round-ended. They may have ovality, wall thickness variation, and residual stresses from manufacturing. These imperfections can affect the confinement behaviour. The paper assumes a perfect tube, but in practice, the tube's geometric imperfections should be within the tolerances specified in standards such as EN 10219.
- What is the effect of the concrete's shrinkage and creep? The paper does not consider time-dependent effects. In practice, concrete shrinkage can reduce the confinement effect by creating a gap between the concrete and the steel tube. This is a real concern for long-term performance.
Study Insights and Implications
This paper is a valuable contribution to the design of round-ended CFST columns. The finite element model is well validated, and the parametric study provides clear guidance on the effects of various parameters. The interaction curve is a practical tool that can be used in design.
From a manufacturing perspective, the paper highlights the importance of dimensional accuracy and weld quality. The round-ended section is more complex to manufacture than a circular section, but it offers significant advantages in certain applications. Manufacturers who can produce high-quality round-ended tubes will have a competitive advantage in the bridge and building construction market.
The paper also underscores the need for a thorough understanding of the failure modes. The distinction between local buckling and excessive rotation is critical for design. Engineers must be able to identify the governing failure mode and design accordingly.
Reference Value and Outlook
The paper's findings are directly applicable to the design of round-ended CFST columns in bridges and buildings. However, further research is needed to address the effects of long-term loading, fire exposure, and seismic loading. The interaction curve should also be verified against full-scale experimental tests.
For the steel pipe industry, the paper reinforces the need for high-quality tubes with tight dimensional tolerances. The growing use of round-ended CFST columns in bridge piers means that the demand for reliable, high-strength steel tubes with this specific cross-section will continue to increase.
In conclusion, this paper is a valuable reference for structural engineers and steel pipe manufacturers alike. It provides a rational basis for the design of round-ended CFST columns and highlights the importance of considering the individual parameters that govern the composite behaviour. The interaction curve is a practical tool that can be used in everyday design and quality control. As an industry veteran, I would recommend that every engineer involved in CFST design read this paper and incorporate its insights into their practice.
Zhuojin Pipe Fitting Co., Ltd