Comparative Analysis of Creep Coefficient Models for Concrete-Filled Steel Tube Arch Bridges
Literature Overview and Core Findings
This paper by Lai Xiuying and Chen Baochun from Fuzhou University addresses a critical issue in the design of concrete-filled steel tube (CFST) arch bridges: the selection of an appropriate creep coefficient model. The authors compared two widely used models, ACI 209R-92 and CEB-FIP MC90, and analyzed their application to 11 actual CFST arch bridges from the closure of the empty steel tube to 10 years after bridge completion.
The key finding is that the difference between the two models' results is generally less than 5%. This is a significant conclusion because it suggests that for practical design purposes, either model can be used with acceptable accuracy. However, the authors recommend the CEB-FIP MC90 model for adoption in the national specification "Technical Specification for Concrete-Filled Steel Tube Arch Bridges" because it is more familiar to practicing engineers.
Technical Interpretation and Engineering Significance
Understanding Creep in CFST Arch Bridges
Creep is the time-dependent deformation of concrete under sustained load. In CFST arch bridges, the core concrete is subjected to sustained compressive stresses from the arch action, making creep a primary concern for long-term deflection and stress redistribution.
The two models compared in this paper have different theoretical foundations:
| Aspect | ACI 209R-92 | CEB-FIP MC90 |
|---|---|---|
| Theoretical basis | Empirical regression from test data | Semi-empirical with physical basis |
| Input parameters | Slump, fine aggregate percentage, air content, curing method | Concrete compressive strength, cement type, relative humidity, member size |
| Time functions | Hyperbolic and parabolic functions | Modified exponential and logarithmic functions |
| Applicability | General purpose, primarily North American practice | General purpose, primarily European practice |
| Familiarity in China | Less familiar | More familiar to Chinese engineers |
Practical Implications for Bridge Design
The 5% difference between the two models is within the typical accuracy of creep prediction methods. However, this does not mean that creep effects can be ignored. In CFST arch bridges, creep can cause:
- Increased deflection at midspan, potentially affecting the bridge's serviceability and aesthetics.
- Redistribution of stresses between the steel tube and concrete core, potentially increasing steel tube stresses.
- Long-term changes in the arch's geometry, affecting the bridge's structural behavior.
From my experience in steel pipe supply for bridge projects, I can confirm that the steel tube dimensions and wall thickness are often governed by the long-term stress state, not just the short-term ultimate limit state. The creep-induced stress redistribution can increase steel tube stresses by 10-20%, which must be accounted for in the design.
Integration with Manufacturing and Delivery Practice
Steel Tube Requirements for Arch Bridges
The steel tubes used in CFST arch bridges are typically large-diameter, thick-walled pipes. From a manufacturing perspective, these tubes require:
- High-quality steel with controlled chemical composition, particularly for weldability.
- Precise dimensional control to ensure proper fit-up and welding of connections.
- Surface preparation to ensure adequate bond with the core concrete.
- Non-destructive testing of all welds, including ultrasonic and radiographic inspection.
The delivery of these tubes to the bridge site is a logistics challenge in itself. For large-diameter pipes, the transportation method must be carefully planned to avoid damage and ensure timely delivery. In one project I was involved with, the delivery of 800mm diameter tubes for an arch bridge required special permits and route planning due to the oversized dimensions.
Construction Sequence and Creep Effects
The paper's analysis from empty tube closure to 10 years after completion highlights the importance of the construction sequence. The creep behavior is significantly affected by:
- The time between concrete placement and the application of additional loads.
- The age of concrete when the arch is loaded.
- The construction method, whether it is in-place casting or prefabricated segments.
In winter construction, as discussed in Topic 4, the concrete's early-age behavior is affected by temperature, which can influence the creep development. The paper's recommendation to use CEB-FIP MC90, which accounts for temperature effects, is particularly relevant for projects in cold climates.
Key Questions and Reflections
The paper's conclusion that the two models differ by less than 5% is reassuring, but I would caution against over-reliance on this finding. The 11 bridges analyzed may not represent the full range of conditions encountered in practice. Factors such as:
- High-strength concrete (C60 and above) may exhibit different creep behavior than the normal-strength concrete used in the calibration of the models.
- The effect of the steel tube's confinement on concrete creep is not explicitly considered in either model.
- The long-term behavior over periods exceeding 10 years may show greater divergence between the models.
In my experience, the creep of high-strength concrete in CFST structures can be significantly different from the predictions of standard models. The confinement provided by the steel tube reduces the creep strain, but the magnitude of this reduction depends on the confinement ratio and the concrete's stress level.
Study Insights and Implications
This paper provides valuable guidance for the design of CFST arch bridges. The key takeaways are:
- Both ACI 209R-92 and CEB-FIP MC90 models give similar results for typical CFST arch bridges, with differences generally below 5%.
- The CEB-FIP MC90 model is recommended for adoption in the national specification due to its familiarity among Chinese engineers.
- Creep effects must be considered in the design of CFST arch bridges, particularly for long-term deflection and stress redistribution.
- The construction sequence and timing of concrete placement significantly affect the creep behavior.
From a practical standpoint, I would recommend that engineers:
- Use the CEB-FIP MC90 model as the primary design tool, but validate the results with the ACI 209R-92 model for critical projects.
- Specify concrete with controlled creep characteristics, particularly for high-strength concrete.
- Monitor the actual creep behavior of the bridge during construction and the first few years of service.
- Consider the effects of temperature and humidity on creep, especially for bridges in extreme climates.
The paper's contribution is practical and directly applicable to bridge design. However, engineers should always remember that creep models are empirical or semi-empirical, and their accuracy depends on the range of data used for calibration. For critical projects, a more detailed analysis using actual material properties and environmental conditions is recommended.
Zhuojin Pipe Fitting Co., Ltd