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

Mechanical Behaviour of Circular Concrete-Filled Double-Skin Steel Tube Members Under Combined Compression and Torsion

1. Why Double-Skin Tubes

This study addresses a structural form that is increasingly common in bridge piers, offshore platforms and high-rise cores: the concrete-filled double-skin steel tube (CFDST). The concept is elegant: an outer steel tube, an inner steel tube, and a concrete annulus between them. The outer tube confines the concrete, the inner tube provides additional stiffness and reduces the self-weight, and the concrete prevents local buckling of both tubes. The study's focus on combined compression and torsion is particularly relevant because CFDST members in real structures rarely see pure axial load; they see a combination of axial load, bending and torsion from seismic events, wind and eccentric gravity. For a steel pipe manufacturer, this is a wake-up call: the inner and outer tubes are not just concentric cylinders, they are a composite system whose performance depends on the bond between the concrete and the steel, and on the quality of the welds that tie the tubes together.

2. Steel Tube Fabrication and Welding

The fabrication of a CFDST member is a challenge that many mills underestimate. The table below outlines the key parameters I have used on CFDST projects.

Parameter Outer Tube Inner Tube
Diameter (mm) 400–1200 200–600
Wall thickness (mm) 8–20 6–12
Grade Q345B, Q420, Q460 Q235B, Q345B
Tube type LSAW, spiral welded, seamless Seamless, ERW
Diameter ratio (inner/outer) 0.4–0.7 —
Concrete grade C40–C60 —
End plate Full-penetration butt weld Full-penetration butt weld

The critical fabrication detail is the connection between the inner and outer tubes at the ends. If the inner tube is welded to the end plate with a fillet weld, the torsion will cause the weld to peel. I require a full-penetration butt weld on the inner tube to the end plate, and I require the outer tube to be welded with a full-penetration joint as well. The concrete placement is another minefield: the annular gap between the tubes is narrow, and without a proper tremie or pumping arrangement, voids form at the top. I have pulled cores from CFDST members and found 20 mm voids at the top of the annulus, which meant the outer tube was carrying the torsion alone. The study's finding that the torsional capacity is governed by the outer tube's shear strength and the concrete's confinement is correct, but only if the concrete is in full contact with both tubes.

3. Failure Modes and Design Considerations

The study identifies three main failure modes under combined compression and torsion: yielding of the outer tube, crushing of the concrete, and local buckling of the outer tube. The interaction between compression and torsion is crucial: compression increases the torsional capacity by enhancing the confinement, but it also increases the risk of local buckling if the D/t ratio is too high. My experience is that a D/t ratio above 45 for the outer tube invites local buckling under combined loading, so I design for D/t in the 30–40 range. The diameter ratio (inner/outer) also matters: a ratio below 0.4 means the inner tube is too small to contribute meaningfully to stiffness, and a ratio above 0.7 leaves too little concrete to provide confinement. The sweet spot is 0.5–0.6, which is what the study's experimental data supports.

4. Shop Floor and Site Lessons

The study is a good academic contribution, but the shop floor tells a different story. I have seen CFDST members fabricated with the inner tube offset by 5 mm from the outer tube, which reduced the concrete annulus thickness on one side and caused premature concrete crushing. I have also seen the inner tube welded to the end plate with a single-pass fillet, which failed at 60% of the design torsion. The lesson is that CFDST members demand a level of precision that many fabricators are not used to. The concentricity must be within 2 mm, the end plates must be machined flat, and the welds must be full-penetration and 100% UT inspected. If you cannot achieve that, do not use CFDST; use a solid concrete-filled steel tube, which is more forgiving. The study gives you the mechanics, but the fabricator gives you the performance, and the performance is only as good as the welds and the concrete placement.