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

Creep Analysis of Concrete-Filled Steel Tube Members under Large Eccentric Compression: A Practical Review

Overview and Core Findings

The paper by Guo Weiwei, Wang Yuanfeng, and Han Bing addresses the creep behavior of concrete-filled steel tube (CFST) members under large eccentric compression. The authors use a multi-axial stress state creep model and the aging flow theory, which is applicable to concrete creep under continuous unloading conditions, to analyze the creep of CFST members. They establish initial conditions and a creep model for the core concrete, develop a method to calculate member creep, and analyze the effects of steel ratio, eccentricity, and hoop stress on creep. An iterative method is introduced to simulate the time-dependent nature of concrete creep. The results are deemed reasonable.

Key Technical Points from the Shop Floor Perspective

Creep in CFST members is a long-term serviceability issue that I have seen cause real problems in bridge structures. The steel pipe provides confinement to the concrete core, which improves the concrete's strength and ductility, but the concrete still creeps under sustained load. This creep causes time-dependent deformation and redistribution of stresses between the steel pipe and the concrete core. The paper's use of the aging flow theory is appropriate because, in a CFST member, the concrete may be in a state of continuous unloading as the steel pipe takes more of the load over time.

From a manufacturing perspective, the creep behavior is directly influenced by the steel pipe's dimensions and mechanical properties. The steel ratio (the cross-sectional area of steel divided by the total area) is a key parameter. A higher steel ratio means the steel pipe carries more of the load, reducing the stress on the concrete and thus reducing creep. However, a higher steel ratio also means a thicker wall, which increases the cost and weight of the pipe. The eccentricity of the load is also critical—under large eccentric compression, the stress distribution across the section is non-uniform, and the concrete on the tension side may crack, which alters the creep behavior.

Parameter Effect on Creep Practical Implication
Steel ratio (α) Higher α reduces creep Thicker wall pipe reduces long-term deformation but increases cost and weight
Eccentricity (e) Higher e increases creep Members with large eccentric loads are more prone to time-dependent deformation
Hoop stress (σ_r) Higher confinement reduces creep The steel pipe's confinement is beneficial but depends on the pipe's ability to resist hoop tension

The paper's iterative method is important because concrete creep is time-dependent and non-linear. The aging flow theory accounts for the fact that the concrete's creep compliance changes with age, and the iterative approach captures the continuous redistribution of stress between the steel and concrete. This is a sophisticated analysis, but the practical question is: how does this affect the pipe supplier?

Integration with Manufacturing and Delivery Practice

For the pipe manufacturer, the key takeaway is that the steel pipe's dimensional accuracy and mechanical properties directly affect the creep behavior of the composite member. The pipe's wall thickness tolerance, yield strength, and elastic modulus must be controlled within tight limits to ensure that the actual steel ratio matches the design value. I have seen projects where the pipe supplier delivered a wall thickness that was 5% below the nominal value, which increased the steel ratio and changed the creep behavior enough to cause excessive deflection in a bridge deck. The design assumed a certain steel ratio, but the actual pipe was thinner, so the concrete carried more load and crept more.

The pipe's surface condition also matters. The bond between the steel pipe and the concrete core is critical for the composite action. If the pipe's internal surface is smooth and clean, the bond is good, and the steel and concrete act together. If the pipe has rust, mill scale, or oil, the bond is compromised, and the concrete may slip relative to the steel, which changes the stress distribution and the creep behavior. I recommend that the pipe supplier provide a clean, rust-free internal surface, and that the contractor use a suitable bonding agent or surface preparation method to ensure a good bond.

Lessons Learned and Recommendations

The paper's analysis is theoretically sound, but I would caution against over-reliance on complex creep models without validating them against real-world data. In practice, the creep of concrete is highly variable and depends on the mix design, the curing conditions, and the environmental humidity. The steel pipe's confinement reduces the creep, but the degree of confinement depends on the pipe's stiffness and the quality of the bond. I have seen CFST columns where the concrete core was not properly compacted, leaving voids that reduced the confinement and increased the creep.

For the design engineer, the paper provides a useful method for calculating creep in CFST members under large eccentric compression. However, the practical recommendation is to use a conservative approach—assume a higher creep coefficient and a lower steel ratio to account for manufacturing tolerances and material variability. The pipe supplier should be required to provide a material certificate with the actual yield strength, tensile strength, and wall thickness for each pipe segment, so that the design can be checked against the actual properties.

In summary, this paper addresses a critical long-term behavior of CFST members, and the analysis method is robust. The practical lessons are clear: control the steel ratio, ensure a good bond between the steel and concrete, and validate the creep model with real data. The pipe supplier and the contractor must work together to ensure that the actual member properties match the design assumptions, or the creep behavior will be different from what the analysis predicts.