Creep Deformation of Concrete-Filled Square Steel Tube Axially Loaded Columns
Test Program and Analytical Model
The paper by Han Linhai, Tao Zhong, Liu Wei, and Chen Baochun presents one of the earliest systematic studies on the long-term deformation behavior of concrete-filled square steel tube (CFST) columns under sustained axial load. The test program included four specimens with different axial load ratios, which is a critical parameter because it determines the level of concrete stress and thus the magnitude of creep. The load ratios ranged from low to high, covering the practical range for building columns. The specimens were instrumented to measure both the total deformation and the individual contributions of the steel tube and the concrete core.
The analytical model is based on the ACI 1992 method for concrete creep, which uses a creep coefficient that depends on the age at loading and the duration of loading. The authors modified this approach to account for the confinement effect of the steel tube on the concrete core. In a CFST column, the steel tube restrains the lateral expansion of the concrete under axial load, which reduces the creep strain compared to an unconfined concrete member. The model also accounts for the stress redistribution between the steel and the concrete over time. As the concrete creeps, it transfers load to the steel tube, which means the steel stress increases while the concrete stress decreases.
From my perspective in the steel pipe industry, this paper is important because it quantifies the long-term stress state in the steel tube. When I supply square steel tubes for CFST columns, the design engineer needs to know the maximum stress the tube will experience over the life of the structure. The creep-induced stress redistribution can increase the steel stress by 10% to 20% above the initial elastic value, and this must be accounted for in the wall thickness design. I have seen cases where the tube was designed for the initial elastic stress only, and the column developed excessive deformation over time because the steel yielded under the increased stress.
Creep Mechanism and Load Transfer
The creep of the concrete core is the driving mechanism. Under sustained load, the concrete undergoes time-dependent deformation, and because the steel tube is elastic and has a much lower creep rate, the concrete tends to shed load to the steel. This is a classic example of stress redistribution in a composite member. The paper shows that the axial load ratio has a significant effect on the creep behavior. At high load ratios, the concrete stress is higher, which increases the creep rate and the amount of load transfer to the steel. At low load ratios, the creep is smaller, and the stress redistribution is less pronounced.
The paper also analyzes the effects of loading age and load duration. The loading age is important because concrete creep is higher when the concrete is loaded at an early age. In practice, this means that if the column is loaded soon after casting, the creep will be larger, and the steel tube will carry more load over time. The load duration is obviously important because creep continues for years, with a significant portion occurring in the first year after loading. The model captures this time-dependent behavior and provides a method to calculate the deformation at any point in time.
The practical implication for steel tube manufacturers is that the tube must have adequate ductility and yield strength to accommodate the increased stress without yielding or buckling. The tube wall thickness must be sufficient to resist both the initial elastic stress and the additional stress from creep. The tube also must have a high-quality weld if it is a welded square tube, because the weld is the weak point in the section. I have seen welded square tubes with defective welds that failed under the combined stress of axial load and creep-induced stress redistribution.
Manufacturing and Delivery Implications
For the steel pipe industry, the key parameters are the tube dimensions, wall thickness, and material grade. The square tube is typically made from hot-rolled coil or plate, formed and welded by ERW or SAW processes. The corner radius is important because it affects the confinement behavior. A sharp corner provides better confinement but is more difficult to form without cracking. A large corner radius is easier to form but provides less confinement. The design engineer must specify the corner radius based on the structural requirements, and the manufacturer must be able to achieve it consistently.
The tube surface condition is also critical. The bond between the steel tube and the concrete core depends on the surface roughness and cleanliness. Mill scale, rust, or oil on the inner surface will reduce the bond and can cause slip at the interface, which reduces the composite action. In my experience, the inner surface of the tube should be cleaned by shot blasting or wire brushing before concrete placement. Some manufacturers skip this step, and the result is a column with reduced capacity and increased deformation under long-term loading.
The tube ends must be properly prepared for welding to the base plate or the adjacent column section. The end preparation must be square and clean, and the weld must be full penetration to ensure that the load is transferred through the joint. I have seen CFST columns where the tube-to-base plate weld was incomplete, and the column developed a hinge at the base under sustained load. The cost of repairing such a defect is enormous, and it can be avoided by proper welding procedure qualification and inspection.
Standards and Engineering Judgment
The paper uses the ACI 1992 method as the basis for the creep model, which is a reasonable approach for design purposes. However, the ACI method was developed for reinforced concrete, and the modification for the steel tube confinement is an approximation. The authors validated the model against their test data and found good agreement, which gives confidence in the approach. For design in China, the relevant standard is GB 50936, which provides a method for calculating the long-term deformation of CFST columns. The paper's model is more detailed and can be used for a more accurate assessment.
The engineer must decide whether to use the simplified code method or the more detailed model from the paper. For most practical design, the code method is adequate. For critical structures, such as high-rise buildings or long-span bridges, the more detailed model is justified. The engineer must also consider the creep of the concrete in the design of the steel tube. The tube must be designed for the maximum stress that will occur over the life of the structure, which includes the initial elastic stress plus the creep-induced stress increase.
Key Questions and Reflections
The paper raises several questions that are worth considering. The first is the effect of the concrete shrinkage on the long-term behavior. Shrinkage is a time-dependent deformation that occurs independently of the applied load, and it can cause the concrete to separate from the steel tube if the bond is not adequate. The paper does not address shrinkage, but it is a significant factor in practice. The second is the effect of the loading history. Real structures are loaded and unloaded many times over their life, and the creep behavior under cyclic loading is different from that under sustained loading. The paper only considers sustained loading, so the results are conservative for cyclic loading.
The third question is the effect of the tube shape. The paper considers square tubes, but round tubes are also common in practice. The confinement behavior is different for round tubes because the hoop stress in the tube is uniform, while in a square tube, the hoop stress is concentrated at the corners. The creep behavior is therefore different, and the model from the paper may need to be modified for round tubes. This is an area where further research is needed.
Study Insights and Implications
This paper is a classic in the field of concrete-filled steel tube structures, and it provides a solid foundation for understanding the long-term behavior of these columns. The key insight is that creep is not just a concrete problem; it is a composite problem that involves stress redistribution between the steel and the concrete. The steel tube must be designed for the maximum stress that will occur over the life of the structure, not just the initial elastic stress.
For my fellow engineers in the steel pipe industry, this paper is a reminder that the tubes we supply are not just static members; they are part of a time-dependent system that changes over decades. The tube must have adequate strength, ductility, and weld quality to accommodate the stress redistribution caused by concrete creep. We must work closely with the structural engineers to understand the long-term stress state and ensure that our products are designed and manufactured to meet those requirements.
The practical lesson is that the design of CFST columns requires a multidisciplinary approach. The structural engineer must understand the creep behavior of the concrete and the stress redistribution in the composite section. The steel pipe manufacturer must understand the structural requirements and provide a tube with the right dimensions, material, and weld quality. The contractor must ensure that the concrete is properly placed and that the tube is clean and free of defects. If any of these links fails, the column will not perform as intended, and the consequences can be severe.
Zhuojin Pipe Fitting Co., Ltd