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

Orthogonal Experimental Design for Concrete-Filled Steel Tube Mix Proportions

Overview of the Literature

This paper by Zhang Hong, Fang Hua, Huang Peng and Ye Fen, published in Journal of Highway and Transportation Research and Development in 2007, addresses a practical problem that I have seen derail more than one bridge project: getting high-performance concrete to flow properly through a steel tube without segregation, while still achieving the specified compressive strength. The authors applied orthogonal experimental design theory to the mix proportion design of high-performance concrete-filled steel tubes (CFST), using the Gao'an Yunzhou Bridge as their case study. The key insight is that with only a small number of tests, they could systematically evaluate the influence of multiple admixture factors at different levels, and identify which factors significantly affect flowability and strength.

Core Technical Points and Interpretation

The orthogonal design method is not new to statistics, but its application to CFST concrete is where the practical value lies. In a CFST column, the concrete is pumped from the bottom up through the steel tube, and it must be able to flow around internal stiffeners, through diaphragm plates, and fill every corner without leaving voids. The paper identifies the main admixture factors as water-reducing agent, fly ash, and micro-silica fume, each tested at different dosage levels. The variance analysis (ANOVA) revealed that the water-reducing agent had the most significant effect on both slump flow and compressive strength, while fly ash primarily improved flowability at the expense of early strength, and micro-silica fume contributed to strength gain but reduced workability if over-dosed.

From my years on the shop floor and on site, I can tell you that the single most common failure mode in CFST construction is not the steel tube — it is the concrete. I have seen concrete arrive at site with a 200 mm slump that would not pump 15 metres up a tube because the aggregate gradation was wrong or the admixture dosage had been adjusted by a laboratory that had never seen a CFST pour. This paper's contribution is that it gives engineers a systematic way to optimise the mix before the pour, rather than discovering the problem halfway up the tube.

Integration with Engineering Practice

The practical implications of this study extend beyond the laboratory. When I supply steel pipe for CFST applications, I always ask the contractor about their concrete mix design, because the pipe dimensions and the concrete properties are interdependent. A mix with high flowability and low shrinkage is essential to avoid voids at the steel-concrete interface, which directly affects the composite action that the structural engineer has assumed in their design. The paper's finding that the water-reducing agent is the dominant factor aligns with my experience: I have seen projects where a change in the brand of superplasticiser, without any change in the mix proportions, caused the concrete to segregate and block the pump line, resulting in a partially filled column that had to be core-drilled and grouted.

The orthogonal design approach also has a cost benefit. Instead of running a full factorial experiment with, say, four factors at three levels each (81 combinations), the orthogonal array reduces this to nine or eighteen tests, which is a significant saving in time and materials. For a project like the Gao'an Yunzhou Bridge, where the CFST is a major structural element, this means the mix can be optimised within the project schedule rather than delaying the pour.

Key Questions and Reflections

One question that this paper raises but does not fully answer is the interaction between the concrete mix and the internal geometry of the steel tube. In my experience, the presence of internal stiffener rings or diaphragm plates creates local flow restrictions that can cause the concrete to bridge, even with a well-designed mix. The orthogonal design method can be extended to include geometric factors, but this would require a much larger test matrix and is rarely done in practice. Another practical point is the temperature sensitivity of the concrete. On a hot summer day in the Middle East or southern China, the concrete temperature can rise above 30°C, which reduces flowability and accelerates setting. The paper does not address this, but it is a critical factor for site engineers to consider.

Study Insights and Implications

The lasting value of this paper is its demonstration that statistical design of experiments is not just an academic exercise but a practical tool for solving real construction problems. The mix proportion that was developed for the Gao'an Yunzhou Bridge met both the flowability and strength requirements, and the methodology can be adapted to other CFST projects with different materials and environmental conditions. For engineers involved in CFST design and construction, the key takeaway is that the concrete mix must be treated as a critical design parameter, not an afterthought. The steel tube is only half of the composite system; the concrete is the other half, and getting it right requires a systematic approach to mix optimisation, not trial and error.

This paper remains a useful reference for any engineer who is specifying CFST concrete, particularly for bridge applications where the concrete must be pumped over long distances and through complex internal geometries. The orthogonal design method, combined with proper quality control during batching and placement, is the foundation for a successful CFST pour. I have seen too many projects where the concrete was treated as a commodity, and the result was a column with voids, a delayed schedule, and a costly repair. This study is a reminder that the mix design deserves the same attention as the steel pipe specification, and that a little statistical rigour up front can save a lot of trouble on site.