Bending Performance of Concrete-Filled Steel Tube Composite Lattice Columns
Overview of the Research
This paper by Hu Jing, Zhao Renda, Zhan Yulin and Mou Tingmin addresses the flexural behaviour of concrete-filled steel tube (CFST) composite lattice columns, a structural form increasingly adopted for high piers in long-span continuous rigid frame bridges. The authors carried out scaled-model bending failure tests on a representative high pier and supplemented the experimental work with full-process finite element simulation using ANSYS. The study is funded by the National Natural Science Foundation of China and the Specialised Research Fund for the Doctoral Programme of Higher Education, which signals a solid academic pedigree. The work is published in Journal of Highway and Transportation Research and Development (2013, Vol. 30, No. 11), and the classification number U441 places it squarely in bridge engineering.
From my own experience supplying steel pipe for bridge piers and towers, I can tell you that the CFST lattice column is not a novelty item. It has been used in China for decades in various forms, but the specific configuration studied here — a composite lattice column made of multiple CFST limbs connected by lacing elements — is a modern refinement that deserves careful attention. The authors set out to answer a practical question: how does this structure actually behave when subjected to bending, and can we predict its ultimate capacity with confidence?
Key Technical Findings and Interpretation
Four-Stage Deformation Behaviour
The most significant experimental observation is that the bending deformation of the CFST composite lattice column progresses through four distinct stages:
| Stage | Deformation Characteristic | Structural Meaning |
|---|---|---|
| Elastic | Linear load-deflection relationship, no visible cracking | Serviceability limit state governs |
| Elastic-plastic | Nonlinear response, concrete micro-cracking begins, steel tube yields locally | Redistribution of internal forces between steel and concrete |
| Plastic | Significant stiffness degradation, large deflection growth, visible concrete crushing | Ultimate limit state approached |
| Descending | Load-carrying capacity drops while deflection continues to increase | Post-peak ductility and residual capacity |
This four-stage progression is textbook behaviour for a composite member, but the authors correctly emphasise that the structure exhibits good plastic deformation capacity and excellent overall performance. The failure mode under ultimate load is a typical ductile failure, not a brittle one. This is exactly what a bridge engineer wants to see in a high pier, because it provides warning before collapse and allows for emergency measures.
Validity of the Plane Section Assumption
A critical finding that has direct implications for design practice concerns the validity of the plane section assumption. The authors demonstrate that before cracking of the composite section, the plane section assumption holds reasonably well. After cracking, however, it no longer applies. This is a subtle but important point. Many simplified design methods for CFST members rely on the plane section assumption to derive strain compatibility equations. If the assumption breaks down after cracking, then those simplified methods may overestimate or underestimate the actual capacity depending on the loading regime.
From a practical standpoint, this means that for serviceability checks where the section remains uncracked, the plane section assumption is acceptable. For ultimate limit state design, however, a more rigorous approach is needed. The authors propose a finite element method that accounts for the actual nonlinear behaviour of both the steel tube and the core concrete, and they demonstrate that this method predicts the ultimate flexural capacity with good accuracy.
Parametric Study and Design Implications
The parametric analysis reveals two important trends:
- The ultimate flexural capacity increases in direct proportion to the steel tube strength and the section height-to-width ratio. This is intuitive but worth stating explicitly: the steel tube is the primary contributor to flexural capacity, and a taller section is more efficient in resisting bending.
- The ultimate flexural capacity is relatively insensitive to the strength of the infilled concrete. This is a crucial finding. It means that for a bending-dominated application, specifying a higher-grade concrete does not yield proportional benefits in capacity. The economic implication is clear: do not overspend on high-strength concrete for a member that is primarily subjected to bending. The steel tube is doing the heavy lifting.
Engineering Practice and Lessons Learned
Manufacturing Considerations for CFST Lattice Columns
When I supply steel pipe for CFST lattice columns, the key requirements are dimensional accuracy, weld quality, and material consistency. The steel tubes used as limbs in a lattice column are typically seamless or ERW pipe, and they must meet strict tolerances on outside diameter and wall thickness. If one limb is slightly larger or smaller than the others, the lacing connections will not align properly, and the eccentricity introduced will reduce the effective capacity of the column.
For the lacing elements, which are often made from smaller-diameter pipe or rolled steel sections, the weld quality is critical. The welds connecting the lacing to the limbs are typically fillet welds, and they must be sized to transfer the shear forces between the limbs. In my experience, the most common defect in lattice column fabrication is undersized fillet welds on the lacing connections. This is often a result of production pressure and inadequate weld inspection. A simple visual inspection and a dye penetrant test on a sample of the welds will catch this problem before it reaches the site.
Construction and Site Practice
On site, the CFST lattice column requires careful concrete placement. The concrete must be pumped into the steel tubes, and the tubes must be properly vented to allow air to escape. If air pockets form inside the tube, the composite action between the steel and concrete is compromised, and the column will not achieve its design capacity. This is a well-known problem in CFST construction, and it has caused failures in the past.
The authors' finding that the plane section assumption breaks down after cracking also has implications for construction quality control. If the concrete is not properly compacted and there are voids or honeycombing, the section will crack earlier and more severely than predicted, leading to premature loss of stiffness and a reduction in ductility.
Lessons from Real Projects
I recall a project where a CFST lattice column was used for a bridge pier, and the contractor encountered significant difficulties with concrete placement. The tubes were long and slender, and the concrete had to be pumped from the bottom up. Despite using a high-slump mix, there were several blockages, and the contractor had to resort to vibration from the top of the tube. This worked, but it introduced a risk of segregation. The final structure performed well, but the experience highlighted the importance of proper concrete mix design and placement procedures for CFST members.
Another lesson relates to the steel tube itself. In one case, the specified steel grade was not available from the local mill, and the contractor substituted a higher grade. This was acceptable, but it changed the failure mode from a ductile steel-controlled failure to a more brittle concrete-controlled failure. The structure still met the design requirements, but the ductility was reduced, and the safety margin was lower than intended. This is why the authors' finding that the flexural capacity is insensitive to concrete strength but sensitive to steel strength is so important: it tells us that the steel tube is the critical component, and we must not compromise on its quality.
Reflections and Study Insights
This paper is a valuable contribution to the understanding of CFST composite lattice columns, but it also raises several questions that deserve further attention.
First, the experimental work is based on scaled models, and the authors do not discuss the size effect in detail. In my experience, the behaviour of large-scale CFST members can differ from that of small-scale models due to the influence of concrete shrinkage and creep, which are time-dependent phenomena that do not scale linearly. The authors mention that they used a full-process finite element simulation, but they do not address the long-term behaviour of the column under sustained load. This is a gap that should be filled by future research.
Second, the authors' finding that the plane section assumption breaks down after cracking has implications for the design of other composite members, such as CFST beams and beam-columns. If the assumption is invalid for lattice columns, it may also be invalid for other structural forms, and this should be investigated systematically.
Third, the paper does not address the effect of construction defects on the bending performance of CFST lattice columns. As I discussed above, voids in the concrete, misaligned lacing connections, and weld defects can all reduce the capacity and ductility of the column. Future research should include a parametric study of these defects to quantify their impact and to develop acceptance criteria for construction quality.
Conclusion
The bending performance of CFST composite lattice columns is a complex topic that requires careful consideration of the interaction between the steel tube and the core concrete. This paper provides valuable experimental and analytical evidence that the structure exhibits four-stage deformation behaviour, good ductility, and a predictable ultimate capacity. The finding that the flexural capacity is primarily governed by the steel tube strength and the section height-to-width ratio, rather than the concrete strength, is a practical insight that can guide material selection and design decisions. The authors' finite element method for predicting ultimate flexural capacity is a useful tool, but it should be validated against a broader range of experimental data before it is adopted in design codes. For practising engineers, the key takeaways are: do not compromise on steel tube quality, ensure proper concrete placement, and be aware that the plane section assumption is only valid before cracking. These lessons are hard-won, and they are worth remembering on every CFST project.
Zhuojin Pipe Fitting Co., Ltd