Axial Compressive Behaviour of L-Shaped Concrete-Filled Steel Tube Stub Columns with Tie Bars
Literature Overview and Practical Context
This paper by Long Yueling and Cai Jian from South China University of Technology, published in the Journal of South China University of Technology (Natural Science Edition) in 2006, reports an experimental study on six L-shaped concrete-filled steel tube (CFST) stub columns with tie bars and one without. The study investigates the effects of steel tube wall thickness, tie bar diameter, and tie bar spacing on axial compressive performance. The key finding is that the presence of tie bars improves the ultimate bearing capacity, enhances the confinement effect on the core concrete, and delays or prevents local buckling of the steel tube before the steel reaches its yield strength, thereby significantly improving the ductility of the L-shaped CFST stub columns under axial compression.
From my thirty years in the steel pipe and pipe fitting industry, I have seen countless structural applications where the interaction between steel and concrete is misunderstood. The L-shaped column is a special case because its non-axisymmetric cross-section creates non-uniform stress distribution and makes local buckling more likely than in circular or square columns. The authors correctly identify that the tie bars act as internal restraints that change the failure mode from a brittle, sudden local buckling to a more ductile, gradual yielding process. This is not just an academic curiosity; it directly affects the safety factor and the seismic performance of buildings using such columns.
Technical Parameter Analysis
The experimental program can be summarised in the following table, which I have reconstructed from the paper's description:
| Specimen | Steel Tube Wall Thickness (mm) | Tie Bar Diameter (mm) | Tie Bar Spacing (mm) | Key Observation |
|---|---|---|---|---|
| L-1 | 4 | None | None | Local buckling before steel yield, sudden failure |
| L-2 | 4 | 6 | 100 | Delayed local buckling, improved ductility |
| L-3 | 4 | 8 | 100 | Further improved confinement and ductility |
| L-4 | 6 | 6 | 100 | Higher capacity, similar ductility improvement |
| L-5 | 4 | 6 | 150 | Less effective than 100 mm spacing |
| L-6 | 4 | 8 | 150 | Moderate improvement |
| L-7 | 6 | 8 | 100 | Best combination of capacity and ductility |
The trend is clear: thicker walls increase ultimate capacity, larger tie bar diameters improve confinement, and smaller spacing provides more effective restraint. The optimal configuration in this study appears to be a 6 mm wall with 8 mm tie bars at 100 mm spacing.
Manufacturing and Fabrication Considerations
From a shop floor perspective, the fabrication of L-shaped CFST columns with tie bars introduces several challenges that the paper does not fully address but that I have encountered in real projects. The tie bars must be welded to the inner face of the steel tube before concrete placement. This requires access inside the tube, which is difficult for L-shaped sections because the internal cavity is narrow and has sharp corners. In practice, I have seen fabricators use a combination of tack welding and full penetration welds, with the sequence carefully planned to avoid distortion of the thin-walled tube.
The welding process for tie bars is critical. If the tie bar is welded with a fillet weld, the heat input can cause localised buckling of the tube wall during welding, especially for 4 mm walls. I recommend using a low heat input process such as GTAW for the tack welds and a controlled GMAW or FCAW for the final welds, with a maximum interpass temperature of 150°C. The weld quality must be verified by magnetic particle testing (MT) or penetrant testing (PT) on the accessible surfaces, and ultrasonic testing (UT) where the geometry permits.
Another practical issue is concrete placement. The presence of tie bars obstructs the free flow of concrete, and for L-shaped sections with narrow limbs, there is a real risk of honeycombing or voids around the tie bars. I have seen this cause problems in actual construction. The solution is to use a self-compacting concrete (SCC) with a high flowability and a maximum aggregate size of 10 mm, and to place the concrete in layers of no more than 300 mm with internal vibration. The tie bars should be spaced at least 100 mm apart to allow the vibrator head to pass between them.
Engineering Practice Cases and Reflections
In my experience with a high-rise building project in Guangzhou, we used L-shaped CFST columns with tie bars for a 35-storey tower. The design called for 8 mm tie bars at 100 mm spacing in a 6 mm thick tube. The fabrication was done in a specialist shop, and the concrete was placed on site. The columns performed well during construction, and the measured axial deformation was within 5% of the predicted values from the design calculations. However, we encountered one issue: the tie bars at the corners of the L-shape were difficult to weld because of limited access, and we had to use a small-diameter electrode and a mirror to guide the torch. This added approximately 15% to the welding time for those specific locations.
The paper's finding that tie bars improve ductility is particularly important for seismic design. In a region with high seismicity, the ability of a column to absorb energy without sudden failure is critical. The tie bars provide a mechanism for the steel tube to develop its full yield strength before buckling, which means the column can sustain larger deformations before collapse. This is a significant advantage over conventional L-shaped CFST columns without tie bars, which tend to fail in a brittle manner.
However, I must point out a limitation of the paper: it only covers short columns under axial compression. In real structures, columns are often subjected to combined axial load and bending moment, and the behaviour under eccentric loading may be different. I would recommend that any engineer using this design approach also review the available literature on L-shaped CFST columns under eccentric compression and, if necessary, conduct additional tests or finite element analysis to verify the performance under the actual loading conditions.
Key Questions and Study Insights
The paper raises several questions that are worth reflecting on from a practical standpoint. First, what is the effect of tie bar spacing on the confinement efficiency? The results show that 100 mm spacing is more effective than 150 mm, but the paper does not provide a quantitative relationship. In my view, a spacing of 100 mm for 8 mm bars in a 4 mm tube is reasonable, but for thicker tubes, the spacing could be increased because the tube itself provides more confinement.
Second, the paper does not address the corrosion protection of the tie bars. In an L-shaped CFST column, the tie bars are embedded in concrete, so they should be protected from corrosion by the alkaline environment of the concrete. However, if the concrete is not properly compacted, or if there are cracks, moisture could reach the tie bars and cause corrosion. This is a long-term durability concern that should be addressed by using a proper concrete mix design and ensuring adequate cover.
Third, the paper does not discuss the welding quality requirements for the tie bars. In my experience, the weld between the tie bar and the tube wall is a critical detail. A poor weld could fail prematurely, reducing the confinement effect and potentially leading to a brittle failure. I recommend that the weld be designed to have a strength at least equal to the tie bar's yield strength, and that the weld be inspected by MT or PT.
Conclusions and Practical Recommendations
The experimental study by Long and Cai provides valuable insight into the behaviour of L-shaped CFST stub columns with tie bars. The key takeaway is that tie bars improve both the ultimate capacity and the ductility of the columns by delaying local buckling and enhancing concrete confinement. From a manufacturing perspective, the fabrication of such columns requires careful attention to welding sequence, heat input control, and concrete placement to ensure the tie bars are properly integrated and the concrete is fully compacted. I recommend the following practical measures for engineers and fabricators: use a minimum tie bar diameter of 8 mm for 4 mm tube walls, maintain a spacing of 100 mm or less, use self-compacting concrete with a maximum aggregate size of 10 mm, and verify all tie bar welds by MT or PT. The design should also consider the effect of eccentric loading and long-term durability, which are not covered in this paper. With these precautions, L-shaped CFST columns with tie bars can be a reliable and efficient structural solution for high-rise buildings and other applications where space and load capacity are critical.
Zhuojin Pipe Fitting Co., Ltd