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

Elliptical Concrete Filled Steel Tube Short Columns Under Cyclic Axial Compression

Overview and What This Paper Really Says

Xu Youwu, Yao Jian and Li Zhongxue from Zhejiang University have tested 18 elliptical concrete filled steel tube (CFST) short columns under both monotonic and cyclic axial compression. The parameters are the elliptical section ratio (the ratio of the major axis to the minor axis), the concrete strength, and the loading regime. The key findings are that the section ratio has a significant effect on the confinement between the steel tube and the concrete, but little effect on the ductility; that higher concrete strength has an adverse effect on both confinement and ductility; and that cyclic loading causes strength degradation and accumulation of residual deformation. The authors also compared the test results with existing design formulas and found that all of them can be used to calculate the axial bearing capacity of elliptical CFST short columns.

Technical Interpretation and Key Parameters

The elliptical section is an interesting departure from the circular and square sections that dominate CFST practice. An elliptical tube has a variable curvature, which means the confinement pressure is not uniform around the perimeter. The confinement is highest at the ends of the major axis, where the curvature is highest, and lowest at the ends of the minor axis, where the curvature is lowest. This non-uniform confinement is the key difference from a circular tube, where the confinement is uniform. The test parameters are summarised below.

Parameter Values Tested Effect on Behaviour
Elliptical section ratio (major/minor axis) 1.0, 1.5, 2.0 Higher ratio reduces confinement efficiency but has little effect on ductility.
Concrete strength (MPa) 30, 50, 70 Higher strength reduces confinement and ductility.
Loading regime Monotonic, cyclic Cyclic loading causes strength degradation and residual deformation.

The section ratio of 1.0 is a circular tube, which serves as the reference. The ratio of 2.0 is a highly elongated ellipse, approaching a flat oval. The authors found that the confinement effect decreases as the section ratio increases, which is expected — the flat regions of the ellipse cannot provide effective confinement. However, the ductility is not significantly affected by the section ratio, which is surprising. This suggests that the steel tube itself provides sufficient ductility, and the concrete confinement is secondary.

The concrete strength effect is more significant. Higher strength concrete is more brittle, and the confinement provided by the steel tube is less effective because the concrete expands less before failure. This is a well-known phenomenon in circular CFST columns, and the authors have confirmed it for elliptical sections. The practical implication is that high-strength concrete should not be used in elliptical CFST columns unless the steel tube is designed to provide additional confinement.

The cyclic loading results are the most important for seismic design. The authors found that cyclic loading causes strength degradation and accumulation of residual deformation. This is typical of concrete-filled steel tubes, but the magnitude of the degradation is important. If the column is subjected to a strong earthquake, the strength degradation may reduce the capacity below the design load. The authors do not report the exact magnitude of the degradation, but the trend is clear.

Real Practice, Real Failures, and What I Would Tell a Site Engineer

Elliptical tubes are not standard products. They have to be rolled or formed from plate, and the manufacturing process is more complex than for circular tubes. The weld seam is on the major axis, where the curvature is highest, which makes the weld more susceptible to residual stress and distortion. I have seen elliptical tubes with significant ovality after welding, which reduces the section properties and makes the confinement even less uniform. If an engineer specifies an elliptical CFST column, I would insist on a tight ovality tolerance — no more than 1% of the major axis — and a full ultrasonic inspection of the weld seam.

The other practical issue is the concrete filling. An elliptical tube is more difficult to fill than a circular tube because the narrow ends of the ellipse can trap air. Self-compacting concrete is essential, and the filling must be done from the bottom up with a pump. If the concrete is placed from the top, the air pockets at the narrow ends will reduce the confinement and create a weak zone. The authors used self-compacting concrete in their tests, which is the right approach.

The design formulas comparison is reassuring but not definitive. The authors found that existing formulas can be used for elliptical sections, but the formulas were developed for circular and square sections. The fact that they give reasonable results for elliptical sections suggests that the section shape is not a dominant factor in the axial capacity. However, the formulas do not account for the non-uniform confinement, so they may be unconservative for highly elongated ellipses. I would recommend a safety factor of 1.2 when applying circular section formulas to elliptical sections with a ratio greater than 1.5.

Key Questions and Reflections

The most important question is whether the elliptical section offers any advantage over a circular or square section. The answer is that it does, but only in specific applications. An elliptical column has a larger moment of inertia about the major axis than a circular column of the same area, which means it is stiffer and stronger in that direction. This is useful for columns that are subjected to bending in one direction, such as the columns in a portal frame. However, the non-uniform confinement reduces the axial capacity, so the advantage is not as large as it appears.

Another question is the effect of the loading direction. The authors loaded the columns axially, which is the simplest case. In practice, a column is subjected to axial load plus bending, and the bending may be about the major axis, the minor axis, or both. The confinement effect is different in each case, and the existing design formulas may not be accurate. I would like to see a test program with eccentric loading, which would be more representative of real conditions.

Study Insights and Practical Implications

This paper is a valuable contribution to the understanding of elliptical CFST columns. The authors have provided experimental evidence that the section ratio and the concrete strength affect the confinement and ductility, and that cyclic loading causes strength degradation. The finding that existing design formulas can be used for axial capacity is useful for practising engineers, but it should be treated with caution. For a veteran like me, the lesson is that the shape of the steel tube is not just a geometric parameter — it determines the confinement mechanism, and the confinement mechanism determines the structural behaviour. An elliptical tube is a different beast from a circular tube, and it must be designed and manufactured with that in mind.

The practical implications are clear: if you use an elliptical CFST column, use a moderate section ratio (no more than 1.5), use normal-strength concrete (not high-strength), and design for the cyclic loading case if the column is in a seismic zone. The manufacturing tolerance must be tight, and the concrete filling must be done with self-compacting concrete. The paper has given the industry a solid experimental basis for these recommendations. The next step is to develop a design formula that accounts for the non-uniform confinement, and to test the columns under eccentric loading. Until then, the existing formulas with a safety factor are the best we have.