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

Effect of Initial Imperfections on Post-Buckling Strength of Rectangular CFST Column Wall Plates — A Manufacturing and Design Perspective

1. Literature Overview and Why It Matters

This paper by Lu Xun and colleagues from Chang'an University tackles a problem that I have wrestled with on the shop floor for decades: how much do welding residual stress and geometric imperfections really reduce the load-carrying capacity of a rectangular steel tube? The authors used ABAQUS to model the local buckling and post-buckling behaviour of the wall plates of rectangular concrete-filled steel tube (CFST) columns. They varied the wall plate yield strength, width-to-thickness ratio, the magnitude of geometric initial imperfection, whether the imperfection was applied, and whether welding residual stress was included. They also studied the coupling between the two. The output is a modified effective width formula that accounts for both effects.

This is not a theoretical nicety. Every rectangular tube we make — whether it is a cold-formed section or a welded box — has residual stress from forming and welding, and it has geometric imperfections from the manufacturing process. The question is how much those imperfections cost us in terms of load capacity. The paper gives us a quantified answer, and it tells us when we can ignore the effects and when we cannot.

2. Core Findings and Technical Interpretation

The paper's key findings are summarised below.

Parameter Finding Practical implication
Welding residual stress Reduces post-buckling strength; effect is significant Cannot be ignored in design. The heat-affected zone and the residual stress field around the welds reduce the effective width of the plate.
Geometric initial imperfection Reduces post-buckling strength; effect is smaller than residual stress Important, but secondary. Manufacturing tolerances matter, but they are not the dominant factor.
Steel yield strength Effect applies across all strength grades Even high-strength steel (Q460 and above) is affected.
Width-to-thickness ratio For yield strength > 460 MPa and width-to-thickness ratio > 65, geometric imperfection can be ignored This is a key practical threshold. For slender plates in high-strength steel, the geometric imperfection is less critical.
Coupling of residual stress and geometric imperfection Both effects are present and interact Design must consider the combined effect, not just one or the other.

The proposed effective width formula is the main deliverable. It is shown to be accurate and conservative when compared with test data. This means that a designer using the formula will not overestimate the capacity of the wall plate, which is exactly what we want in a safety-critical application.

From a manufacturing standpoint, the finding on welding residual stress is the one that hits home. When we weld a box section, the heat input from the submerged arc or gas-shielded arc welds creates a complex residual stress field. The longitudinal welds at the corners of the box create tensile residual stress at the weld and compressive residual stress in the middle of the plate. That compressive residual stress makes the plate more susceptible to local buckling under axial load. The paper confirms that this effect is significant and must be accounted for.

3. Engineering Practice Cases and Lessons Learned

I have seen the consequences of ignoring residual stress in rectangular tubes. In one case, a series of rectangular CFST columns were fabricated with high heat input welds to speed up production. The welds looked fine — full penetration, good appearance — but the residual stress field was severe. When the columns were tested in the lab, the local buckling load was significantly lower than the design prediction. The paper's formula would have predicted that loss. The fix was to reduce heat input and to use a welding sequence that balanced the residual stress distribution.

On the geometric imperfection side, I have seen tubes that were within tolerance for straightness and out-of-roundness but still showed premature local buckling. The reason was that the imperfection was concentrated in a short wavelength, which is more damaging than a long-wavelength bow. The paper's treatment of geometric imperfection as a parameter that can be varied in magnitude is helpful, but in practice, we need to control the shape of the imperfection, not just its magnitude.

The threshold for high-strength steel is particularly useful. When we specify Q460 or higher for a rectangular tube, and the width-to-thickness ratio is greater than 65, we can relax our geometric tolerance requirements without compromising strength. That gives the fabricator more room to work and can reduce cost. However, the residual stress effect still applies, so we cannot relax the welding quality requirements.

4. Key Questions and Reflections

One question that the paper does not fully address is the effect of the welding sequence. In a rectangular box, the order in which the four corner welds are made affects the residual stress distribution. If you weld all four corners in the same direction, the residual stress pattern is different from welding them in a balanced sequence. My experience is that a balanced sequence — welding opposite corners alternately — reduces the residual stress and improves the post-buckling strength. The paper's model uses a simplified residual stress distribution, which is fine for parametric studies but may not capture the full complexity of real fabrication.

Another question is the effect of the concrete core on the wall plate behaviour. In a CFST column, the concrete restrains the inward buckling of the wall plate. The paper models this restraint, but the degree of restraint depends on the concrete's stiffness and its bond with the steel tube. In practice, if the concrete shrinks away from the tube, the restraint is reduced. This is a construction quality issue that can undermine the assumptions in the model.

Finally, the effective width formula is derived for rectangular tubes. What about square tubes with a different aspect ratio, or tubes with stiffeners? The paper gives a good starting point, but more work is needed to cover the full range of practical sections.

5. Study Insights and Implications for Practice

This paper is a valuable reference for both designers and fabricators. The key takeaways are:

  1. Welding residual stress is a first-order effect on the post-buckling strength of rectangular CFST wall plates. It must be considered in design and controlled in fabrication.
  2. Geometric initial imperfection is a second-order effect, but it cannot be ignored except for high-strength steel with slender plates.
  3. The proposed effective width formula is conservative and accurate, making it suitable for design use.
  4. Fabricators should pay attention to welding heat input and sequence to minimise residual stress.

For those of us who make and use rectangular CFST columns, this paper is a reminder that the "perfect" tube we think we are delivering is not perfect. It has residual stress and it has imperfections. The paper tells us how much those imperfections cost, and that is information we need to design safely and fabricate efficiently.