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

Steel Tube Confined Reactive Powder Concrete Short Columns: Bearing Capacity, Failure Modes and What the Shop Floor Should Know

1. Literature Overview and Why This Paper Matters

This paper by Wang Qiuwei and colleagues from Xi'an University of Architecture and Technology, published in the Chinese Journal of Applied Mechanics in 2020, tackles a problem I have seen bite hard on real projects: reactive powder concrete (RPC) is a phenomenal material for compressive strength but it is brittle as a dry bone. The authors set out to prove that a steel tube can do more than just contain the concrete — it can fundamentally change the failure mode and buy you ductility that the concrete alone will never deliver. The test matrix is compact but meaningful: 20 short columns, 133 mm outer diameter, with tube wall thicknesses of 4.5, 6, 8 and 10 mm, and two loading regimes — full cross-section loading and core-concrete-only loading. That is a smart experimental design because it separates the contribution of the steel tube from the contribution of the confined core, and it gives you a direct read on the composite action.

2. Core Technical Points and Test Observations

2.1 Failure Modes Are Governed by the Confinement Coefficient

The paper defines a confinement coefficient (套箍系数) as the ratio of the steel tube's axial yield force to the concrete core's axial capacity. The results are unambiguous: when this coefficient exceeds 0.9, the specimen fails by local bulging of the steel tube; below that threshold, the failure is a classic shear plane through the core. This is a critical insight for anyone specifying tube dimensions, because it tells you that the tube is not just a passive shell — it is an active confining element whose contribution to the failure mechanism is decided by its wall thickness relative to the core diameter.

Tube Wall Thickness (mm) Diameter-to-Thickness Ratio (D/t) Confinement Coefficient Observed Failure Mode
4.5 29.6 < 0.9 Shear failure
6.0 22.2 < 0.9 Shear failure
8.0 16.6 > 0.9 Local bulging
10.0 13.3 > 0.9 Local bulging

2.2 Load-Displacement Behaviour Depends on Tube Thickness

The load-displacement curves tell a story that anyone who has tested composite columns will recognise immediately. With 4.5 mm and 6 mm tubes, the curves show a clear descending branch after peak load — the column reaches its capacity and then sheds load as the concrete crushes and the tube yields. With 8 mm and 10 mm tubes, the curves either plateau or continue to rise, because the thicker tube provides enough confinement to keep the core from dilating and the composite section from losing stiffness. This is the classic transition from a brittle-confined system to a ductile-confined system, and it is precisely the behaviour you want to design for in seismic zones or in any application where post-peak deformation capacity matters.

2.3 The Composite Action Exceeds Simple Superposition

The paper reports that the measured axial capacity of the composite column exceeds the simple sum of the steel tube capacity plus the concrete core capacity by 12% under full-section loading and by 24% under core-only loading. That 24% figure is particularly telling because it isolates the confinement effect — when you load only the core, the tube is not carrying any axial load, so all of its contribution comes from the passive confinement it provides as the core tries to expand laterally. The fact that this confinement contribution is twice as large as the superposition effect under full-section loading tells you that the tube's role is far more sophisticated than simply adding its own axial capacity.

3. Bearing Capacity Calculation and the Twin-Shear Unified Strength Theory

The authors propose a bearing capacity formula based on the twin-shear unified strength theory, which is a generalised yield criterion that can account for the intermediate principal stress effect — something that the traditional Tresca and von Mises criteria ignore. The formula is calibrated against the test data and the reported error is less than 10%, which is respectable for a composite column model given the scatter you normally see in concrete tests. The practical value of this formula is that it gives you a design tool that is simple enough for hand calculation but sophisticated enough to capture the confinement mechanism. For a practising engineer, that is exactly what you want — a formula you can use in a spreadsheet without having to run a full three-dimensional finite element model every time you change a diameter or a wall thickness.

4. Integration with Manufacturing and Delivery Practice

4.1 Tube Manufacturing Implications

The 133 mm outer diameter with wall thicknesses from 4.5 to 10 mm covers a D/t range of 13.3 to 29.6. In my experience, this is squarely in the territory of seamless cold-drawn or hot-finished tube, and the dimensional tolerances matter enormously. If you are supplying tube for a confined RPC column, the inside diameter tolerance controls the volume of concrete you can place, and the wall thickness tolerance controls the confinement coefficient. A 0.5 mm variation on a 4.5 mm wall is an 11% change in the confinement parameter — that is enough to move a specimen from the shear failure regime to the bulging regime. When I sign off on a mill order for this kind of application, I insist on tighter-than-standard tolerances on both outside diameter and wall thickness, and I require the mill to certify the yield strength on the actual heat, not the nominal grade.

4.2 Welding and Fabrication Considerations

For the 8 mm and 10 mm wall tubes, the welding procedure for any attachment plates or stiffeners needs to be qualified for the actual wall thickness, and the heat input must be controlled to avoid excessive distortion. I have seen too many cases where a fabricator welds a stiffener to a thin-walled tube and the localised heat input creates a visible dent that then becomes a stress concentration under axial load. For RPC-filled columns, the tube is the only thing keeping the concrete from bursting, so any local imperfection in the tube wall is a potential initiation point for a bulging failure.

4.3 Concrete Placement and Curing

The paper does not discuss placement, but I will tell you from experience that filling a 133 mm diameter tube with RPC is not trivial. RPC has a very low water-to-binder ratio and high viscosity, so it does not flow like normal concrete. You need either a high-frequency vibrator or a pressure grouting method, and you need to verify that the tube is completely filled — a void at the top of the column is a guaranteed premature failure. For the core-only loading tests in this paper, the concrete had to be placed and compacted with great care to ensure uniform contact with the tube wall, because any gap between the concrete and the tube would eliminate the confinement effect entirely.

5. Key Questions and Reflections

The paper raises one question that I think deserves more attention: what happens at the ends of the column? All 20 specimens are short columns, and the loading is applied directly to the composite section or to the core. In a real structure, the column ends will be connected to beams or base plates, and the confinement at the ends will be different from the middle. I have seen columns fail at the ends because the tube was not properly welded to the base plate and the concrete spalled out through the gap. The paper's formula is calibrated on short columns with uniform confinement, but the real-world application needs a connection detail that maintains the confinement at the ends.

Another question is the long-term behaviour. RPC has a high cement content and low water-to-binder ratio, which means it has a significant autogenous shrinkage. If the shrinkage occurs after the concrete has set and bonded to the tube wall, it can create a gap between the concrete and the tube, reducing the confinement effect. The paper does not address this, but any engineer specifying RPC-filled tubes should be aware of it and should consider using shrinkage-compensating admixtures or post-tensioning the core.

6. Study Insights and Practical Implications

This paper is a solid contribution to the understanding of steel tube confined RPC, and the key takeaway for me is that the confinement coefficient is the single most important design parameter. If you keep it above 0.9, you get a ductile failure with a rising or plateauing load-displacement curve; if you drop below that, you get a brittle shear failure. The 12% and 24% enhancement figures are also valuable because they give you a direct measure of the composite action that you can use in design checks. For the shop floor, the message is clear: the tube is not just a formwork — it is a structural element that must be manufactured with the same care as any other load-bearing component, and the tolerances on diameter, wall thickness and yield strength must be controlled to ensure the confinement coefficient stays in the desired range. I would recommend this paper to any engineer designing composite columns with high-strength concrete, and I would pair it with a careful review of the connection details and the concrete placement procedure before committing to a full-scale application.