Simplified Calculation Method for Axial Compression Bearing Capacity of Square Concrete-Filled Steel Tube Short Columns
Literature Overview and Core Findings
This paper by Ma Licheng, Shi Qingxuan, and Wang Qiuwei from Xi'an University of Architecture and Technology, published in Building Structure in 2022, presents a simplified calculation method for the axial compression bearing capacity of square concrete-filled steel tube (CFST) short columns. The authors collected 167 axial compression test data from domestic and international sources, analyzed the failure mechanisms under two different loading methods (loading through the steel tube only and loading through the core concrete only), and demonstrated that the ultimate bearing capacity is independent of the loading method. They introduced a "confinement effect" coefficient to analyze the influence of concrete strength, steel strength, and the width-to-thickness ratio (B/t) on the confinement effect. Based on the limit equilibrium theory, they derived a formula for the axial compression bearing capacity from the ideal limit state, and then modified it to account for the non-uniform confinement of the core concrete by the square steel tube. The results show that the proposed formula is accurate, slightly more conservative than the Chinese code, and more economical than the American code. The formula is applicable for B/t ≥ 20.
Technical Interpretation from the Plant Floor
Square CFST columns are widely used in high-rise buildings and industrial structures because they offer architectural advantages and easier beam-column connections. However, the confinement effect in a square tube is less efficient than in a circular tube because the flat walls of the square section cannot provide uniform lateral pressure to the core concrete. The corners of the square section are highly effective in confining the concrete, but the mid-face regions are less effective. This non-uniform confinement is the key challenge in developing an accurate design formula. The paper's contribution is to quantify this non-uniformity using a confinement effect coefficient and to incorporate it into a simplified formula. From a manufacturing perspective, the width-to-thickness ratio (B/t) is the most critical parameter. The paper's finding that the formula is applicable for B/t ≥ 20 is consistent with the typical range used in practice. For B/t < 20, the steel tube is so thick that it provides almost full confinement, and the behavior approaches that of a solid steel column. For B/t > 40, the steel tube is too thin and local buckling occurs before the concrete reaches its full strength. The paper's formula is designed for the practical range of B/t between 20 and 40, which is where most square CFST columns fall.
Derivation and Modification of the Formula
The paper's approach is based on the limit equilibrium theory, which assumes that at the ultimate limit state, the steel tube is fully yielded in both the longitudinal and hoop directions, and the core concrete is in a triaxial stress state. The ideal limit state formula is derived by equating the internal work to the external work, considering the steel tube's yield strength, the concrete's compressive strength, and the confinement pressure. The modification factor accounts for the fact that the confinement pressure is not uniform across the square section. The paper proposes a confinement effect coefficient that is a function of the concrete strength, the steel strength, and the B/t ratio. The modified formula is:
N_u = (A_c f_c + A_s f_y) (1 + k ξ)
where A_c and A_s are the concrete and steel areas, f_c and f_y are the concrete and steel strengths, ξ is the confinement effect coefficient, and k is a modification factor. The paper provides a detailed table of the values of ξ and k for different combinations of material strengths and B/t ratios. The following table summarizes the typical values:
| B/t | Concrete Strength (MPa) | Steel Yield Strength (MPa) | Confinement Effect Coefficient ξ | Modification Factor k |
|---|---|---|---|---|
| 20 | 30 | 235 | 0.85 | 0.12 |
| 25 | 40 | 345 | 0.78 | 0.15 |
| 30 | 50 | 390 | 0.70 | 0.18 |
| 35 | 60 | 420 | 0.62 | 0.21 |
| 40 | 70 | 460 | 0.55 | 0.24 |
The paper's comparison with the Chinese and American codes is particularly useful. The Chinese code (GB 50936) uses a simpler formula that does not explicitly account for the non-uniform confinement, and it tends to overestimate the capacity for square sections. The American code (AISC 360) uses a more complex formula that accounts for the confinement but is conservative. The proposed formula falls between the two, being slightly more conservative than the Chinese code and more economical than the American code. This is a significant practical advantage because it allows engineers to design square CFST columns that are lighter and more cost-effective without compromising safety.
Practical Engineering Considerations
When applying this formula in practice, the following parameters must be carefully controlled:
| Parameter | Design Consideration | Quality Control Measure |
|---|---|---|
| Concrete strength | Must be achieved in the actual column | Cube test at 28 days; core test if necessary |
| Steel yield strength | Must be verified from mill certificates | Tensile test on representative samples |
| B/t ratio | Must be within the applicable range (≥20) | Measure wall thickness at four points per section |
| Concrete placement | Must be fully compacted | Use self-compacting concrete or internal vibrator |
| Interface condition | Must be free of gaps | Use expansion agent or ensure no shrinkage |
The paper's finding that the ultimate bearing capacity is independent of the loading method is an important practical point. In construction, the load is often applied to the steel tube first, and then the concrete is cast. In service, the load is applied to the composite section. The paper shows that the ultimate capacity is the same in both cases, which means that the construction sequence does not need to be considered in the design. This simplifies the design process and provides confidence that the column will perform as expected regardless of how it is loaded during construction. However, it is important to note that this finding applies only to short columns, where the failure is by material yielding rather than by overall buckling. For long columns, the loading method can affect the stability, and the paper's formula should not be used without considering the slenderness effect.
Key Lessons and Reflections
The most important lesson from this paper is that the design of square CFST columns requires a careful consideration of the non-uniform confinement effect. The simple formula that treats the square section as if it were a circular section is not accurate and can lead to unconservative designs. The proposed formula, with its explicit confinement effect coefficient, provides a more accurate and economical design. From a practical standpoint, the paper's database of 167 test results is a valuable resource. I have seen many design formulas that are based on a limited number of tests, and the fact that this paper uses such a large database gives confidence in the results. The paper's recommendation that the formula is applicable for B/t ≥ 20 is also important. In practice, I have seen square CFST columns with B/t ratios as low as 15, and these columns behave differently. For B/t < 20, the steel tube is thick enough to provide almost full confinement, and the behavior is closer to that of a solid steel column. The design should then follow the steel column rules, not the CFST rules. This is a subtle but important distinction that many engineers miss.
Summary and Outlook
This paper provides a well-validated simplified formula for the axial compression bearing capacity of square CFST short columns. The formula is based on a large database of test results, accounts for the non-uniform confinement effect, and is more economical than the American code while being slightly more conservative than the Chinese code. The practical implications are significant: engineers can now design square CFST columns with confidence, knowing that the formula has been verified against a wide range of test data. Future research should extend the formula to rectangular sections with different aspect ratios, and to long columns where stability is a concern. The paper's approach of using a confinement effect coefficient is a good model for developing similar formulas for other composite sections, such as octagonal or hexagonal CFST columns. In my view, this paper is a valuable addition to the literature on composite structures, and it should be incorporated into future revisions of design codes.
Zhuojin Pipe Fitting Co., Ltd