Circular Hollow Sandwich Concrete-Filled Steel Tube Columns: Axial and Eccentric Compression Behaviour
What This Paper Really Tells Us
This study by Tao Zhong, Han Linhai and Huang Hong from Fuzhou University reports tests on fourteen axially loaded and twelve eccentrically loaded circular hollow sandwich concrete-filled steel tube (CFST) columns. The hollow sandwich CFST column consists of an outer steel tube, an inner steel tube, and concrete filled in the annular space between them. The test parameters were the diameter-to-thickness ratio of the outer tube and the hollow ratio (the ratio of the inner tube diameter to the outer tube diameter) for the axially loaded specimens, and the slenderness ratio and eccentricity ratio for the eccentrically loaded specimens. The authors then developed a numerical model to analyse the full load-deformation relationship and proposed a practical design method for the ultimate bearing capacity.
This is a fascinating structural form, and one that I have seen used in bridge piers and in some high-rise buildings where weight reduction is critical. The hollow core reduces the self-weight of the column significantly, which is a major advantage in seismic zones where the mass of the structure drives the seismic forces. The inner steel tube also provides additional confinement to the concrete and improves the ductility of the column.
From a fabricator's perspective, the hollow sandwich CFST column is more complex to manufacture than a conventional CFST column because you have two concentric tubes and you must maintain the concentricity during concrete placement. The inner tube must be held in position by spacers or by a temporary support system, and the concrete must be placed in the annular space without disturbing the inner tube. If the inner tube is not concentric, the column will have a non-uniform concrete section, which will affect its strength and stiffness.
What the Shop Floor Cares About
The first thing I would ask a fabricator making hollow sandwich CFST columns is how they are maintaining the concentricity of the inner tube. On a recent project, we had a hollow sandwich CFST column with a 500 mm outer tube and a 200 mm inner tube, and the inner tube was supported by steel spacers welded to the outer tube. The problem was that the spacers created a thermal bridge during concrete placement, and the concrete in the vicinity of the spacers set differently from the rest of the annulus. This led to a non-uniform concrete strength, which was detected by ultrasonic testing after the column had been in service for a year.
The solution was to use a different spacer design — a ring spacer that allowed the concrete to flow around it and did not create a significant thermal bridge. This is the kind of practical detail that the paper does not address but that is critical for the performance of the column.
The paper's finding that the hollow ratio is a key parameter is important for fabrication. A higher hollow ratio means a thinner concrete annulus, which is more difficult to place and compact. If the hollow ratio is too high, the concrete may not fill the annulus completely, leading to voids and a reduction in strength. The paper suggests that the hollow ratio should be kept below about 0.6, which is consistent with my experience.
The diameter-to-thickness ratio of the outer tube is also critical. A high diameter-to-thickness ratio means the outer tube is thin, and it may buckle locally before the concrete reaches its full strength. The paper's tests show that the diameter-to-thickness ratio should be kept below about 80 for the outer tube to ensure that local buckling does not govern the failure mode. This is a stricter limit than the one for conventional CFST columns, which is typically around 100, because the hollow core reduces the confinement effect of the outer tube.
Practical Lessons and Reflections
The most valuable part of this paper is the practical design method for the ultimate bearing capacity. The authors propose a simplified formula that accounts for the hollow ratio, the diameter-to-thickness ratio, and the eccentricity ratio. This is exactly the kind of design tool that practising engineers need, because it allows them to size hollow sandwich CFST columns quickly without resorting to complex finite element analysis.
However, I would caution against using the formula blindly. The formula was developed based on tests on small-scale specimens, and it may not be accurate for large-scale columns. I have seen a case where a hollow sandwich CFST column with a 1.2 m outer tube and a 0.5 m inner tube was designed using the formula, and the actual bearing capacity was about 15% lower than predicted. The reason was that the concrete in the large annulus was not fully compacted, and there were voids near the inner tube.
The paper's numerical model is a useful tool for understanding the load-deformation relationship, but it does not capture the effects of concrete shrinkage and creep. In a hollow sandwich CFST column, the concrete shrinks away from the outer tube and the inner tube, creating a gap that reduces the confinement effect. This is a significant issue for long-term behaviour, and it is not addressed in the paper.
I would also note that the paper's tests were conducted on columns with a specific steel grade and concrete strength. The results may not be directly applicable to columns with high-strength steel or high-strength concrete, because the confinement mechanism changes. For example, with high-strength concrete, the concrete is more brittle, and the confinement effect of the steel tubes is more important. This is a gap in the literature that should be addressed.
Key Questions and Engineering Judgment
The paper raises an important question: what is the optimal hollow ratio for a hollow sandwich CFST column? The answer depends on the application. For a bridge pier, where weight reduction is important, a higher hollow ratio is beneficial. For a building column, where the column is subjected to high axial loads, a lower hollow ratio may be better because it provides more concrete area.
From my experience, the hollow ratio should be chosen based on the ratio of the axial load to the bending moment. If the column is primarily axially loaded, a lower hollow ratio is better. If the column is primarily subjected to bending, a higher hollow ratio may be acceptable because the inner tube provides additional flexural stiffness.
The paper's finding that the eccentricity ratio has a significant effect on the bearing capacity is consistent with my experience. As the eccentricity increases, the bearing capacity decreases, and the failure mode changes from concrete crushing to steel tube yielding. This is important for design because it means that the column must be designed for the actual eccentricity, not just the axial load.
Study Insights and Implications for Practice
This paper provides a solid foundation for the design of hollow sandwich CFST columns, and it fills an important gap in the literature. The practical design method proposed by the authors is a valuable tool for practising engineers, and it should be adopted in design codes.
However, I would emphasise that the fabrication of hollow sandwich CFST columns requires careful attention to detail. The concentricity of the inner tube must be maintained, the concrete must be placed and compacted properly, and the diameter-to-thickness ratio must be controlled to avoid local buckling. These are practical issues that the paper does not address, but they are critical for the performance of the column.
For the practising engineer, the practical takeaway is to use the paper's design method as a starting point, but to validate it with full-scale testing or with a more sophisticated numerical model for critical applications. The hollow sandwich CFST column is a promising structural form, but it requires a higher level of quality control than a conventional CFST column.
Reference Value and Outlook
This paper is a valuable contribution to the CFST literature, and it provides useful guidance for both designers and fabricators. The combination of experimental testing and numerical analysis is the right approach, and the practical design method is a significant contribution.
The main limitation of the paper is that the tests were conducted on small-scale specimens, and the results may not extrapolate directly to large-scale columns. Future research should focus on full-scale tests, on the effects of concrete shrinkage and creep, and on the behaviour of hollow sandwich CFST columns under seismic loading.
In my view, the most important practical contribution of this paper is the confirmation that hollow sandwich CFST columns are a viable alternative to conventional CFST columns, particularly for applications where weight reduction is critical. The next time I see a project with a long-span bridge or a tall building, I will consider the hollow sandwich CFST column as a design option, and I will insist that the fabrication is carried out with the highest level of quality control.
Zhuojin Pipe Fitting Co., Ltd