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

Database-Based Analysis of Axial Compression Capacity of Concrete-Filled Steel Tube Columns

Overview of the Literature

This 2018 paper from the Industrial Construction journal, authored by Huang Liang and colleagues from Hunan University and the China Northwest Municipal Engineering Design and Research Institute, addresses a question that has occupied structural engineers for decades: how accurate are the design formulas for concrete-filled steel tube (CFST) columns under axial compression? The authors collected test data from domestic and international sources, building a database of 637 specimens, and then used this database to evaluate the calculation method in GB 50936-2014, the Chinese technical code for concrete-filled steel tube structures.

This is exactly the kind of paper that I appreciate as a practicing engineer. It takes the design formulas that we use every day and asks the uncomfortable question: do they actually work? The answer, as the paper demonstrates, is that the GB 50936-2014 method is generally sound for most parameter ranges, but it has a significant weakness when it comes to columns with high slenderness ratios. The authors then propose adjustments to the slenderness reduction factor and suggest limits on material parameters to improve the accuracy of the method.

From my perspective in the steel pipe industry, this paper is directly relevant because the steel tube is half of the composite system. The properties of the steel tube, particularly its yield strength and diameter-to-thickness ratio, are parameters that we control in manufacturing. Understanding how these parameters affect the accuracy of the design formulas helps us communicate more effectively with structural engineers and ensures that the pipe we deliver will perform as expected in the final structure.

Key Technical Points

The Database and Its Parameters

The database of 637 specimens is a substantial collection of test data, covering a wide range of parameters:

Parameter Range Covered
Concrete compressive strength 20 to 120 MPa
Steel tube yield strength 200 to 500 MPa
Diameter-to-thickness ratio (D/t) 20 to 100
Member slenderness ratio (L/D) 5 to 50
Confinement coefficient 0.5 to 3.0

The confinement coefficient, which is a measure of the steel tube's ability to confine the concrete core, is a key parameter in CFST design. It is defined as the ratio of the steel tube's yield force to the concrete's compressive force, and it determines how much the concrete's strength is enhanced by the confining action of the steel tube.

Evaluation of GB 50936-2014 Calculation Method

The paper's core contribution is the bias analysis of the GB 50936-2014 calculation method against the database. The bias is defined as the ratio of the calculated capacity to the experimental capacity, and a well-calibrated method should have a bias close to 1.0 with minimal scatter. The results can be summarized as follows:

Parameter Effect on Calculation Bias Assessment
Concrete strength Minimal effect Method is robust across concrete grades
Steel tube yield strength Minimal effect Method is robust across steel grades
Diameter-to-thickness ratio Minimal effect Method is robust for typical D/t ratios
Confinement coefficient Minimal effect Method is robust for typical confinement levels
Slenderness ratio (L/D) Significant effect Method becomes conservative or unconservative at high slenderness

The finding that the slenderness ratio has a significant effect on the calculation bias is the most important result of the paper. For columns with high slenderness ratios, the GB 50936-2014 method tends to overestimate or underestimate the capacity, depending on the specific range. This is a critical issue because slender columns are common in real structures, particularly in tall buildings where the column height can be several times the diameter.

Proposed Adjustments

Based on the database analysis, the authors propose two main adjustments:

  1. Modified slenderness reduction factor: The original formula for the slenderness reduction factor in GB 50936-2014 is replaced with a modified version that better fits the experimental data, particularly for high slenderness ratios.
  2. Limits on material parameters: The authors suggest restricting the range of material parameters for which the method is applicable, to avoid extrapolation beyond the range of the database.

The modified slenderness reduction factor is expressed as a function of the slenderness ratio, and the adjustment ensures that the calculated capacity matches the experimental capacity more closely across the full range of slenderness ratios.

Integration with Engineering Practice

Manufacturing Implications for Steel Tubes

From the steel pipe manufacturing perspective, the key parameters that we control and that affect the CFST column behavior are:

Parameter Our Control Impact on Column Behavior
Outside diameter Tight tolerance, typically ±1% Affects the D/t ratio and the confinement coefficient
Wall thickness Tight tolerance, typically ±10% Affects the D/t ratio and the confinement coefficient
Yield strength Controlled by material grade and heat treatment Affects the confinement coefficient and the steel contribution
Ovality Typically limited to 1% of diameter Affects the uniformity of confinement
Surface condition Internal surface should be clean and free of loose scale Affects the bond between steel and concrete

The paper's finding that the calculation method is robust across steel yield strengths is reassuring, but it does not mean that the yield strength is unimportant. The yield strength directly affects the confinement coefficient, which in turn affects the concrete's enhanced strength. If the actual yield strength is lower than specified, the confinement will be less than assumed, and the column capacity will be lower than calculated.

Case Study: High-Rise Building Columns

I recall a project where we supplied steel tubes for CFST columns in a high-rise building. The columns were 600 mm diameter with a wall thickness of 16 mm, using Q345 steel with a specified yield strength of 345 MPa. The design called for a confinement coefficient of approximately 1.2, which is in the typical range for well-designed CFST columns.

The key issue that arose during delivery was the wall thickness tolerance. The specified tolerance was ±10%, which is standard for seamless pipe, but the actual wall thickness varied from 14.5 mm to 17.5 mm across the delivered pipes. This variation affected the D/t ratio, which ranged from 34 to 41, and the confinement coefficient, which ranged from 1.08 to 1.31. The structural engineer was concerned that the variation would affect the column capacity, and we had to work together to ensure that the design was conservative for the worst-case combination.

The solution was to sort the pipes by wall thickness and assign the thinner pipes to the less critical columns, while the thicker pipes were used for the most heavily loaded columns. This was a practical approach that avoided the need for re-design and ensured that the structure met the required safety margins.

Delivery and Documentation Considerations

When delivering steel tubes for CFST columns, the documentation must clearly state the material properties and dimensions, and the test certificates must show the actual yield strength and tensile strength, not just the specified values. This is particularly important for CFST applications because the confinement coefficient depends on the actual yield strength, and a discrepancy between specified and actual values can have a significant impact on the column capacity.

In my experience, it is also important to discuss the surface condition of the tube interior with the structural engineer. Some design codes require that the internal surface be free of loose scale and rust to ensure adequate bond between the steel and concrete. If the tubes are delivered with a mill scale that is not removed, the bond may be compromised, and the column may not behave as designed.

Key Questions and Reflections

Is the GB 50936-2014 Method Adequate?

The paper's conclusion is that the GB 50936-2014 method is generally adequate for most parameter ranges, but it has a weakness at high slenderness ratios. This is a fair assessment, and it aligns with my experience. The method is a simplified approach that is intended for design use, and it is not expected to be accurate across the entire parameter space. The database analysis provides a quantitative assessment of the method's accuracy, which is valuable for engineers who need to understand the limitations of the design formulas.

What About Other Design Codes?

The paper focuses on GB 50936-2014, but it is worth noting that other design codes, such as ACI 318, EC 4, and AIJ, have their own methods for CFST column design. These methods differ in their treatment of the confinement effect and the slenderness reduction, and the accuracy of each method varies across the parameter space. A comprehensive comparison of the different design codes would be a valuable contribution, but it is beyond the scope of this paper.

What Would I Do Differently?

If I were conducting this research, I would expand the database to include more specimens with high slenderness ratios, as this is the region where the method is weakest. I would also include specimens with high-strength concrete and high-strength steel, as these materials are becoming more common in modern construction. Additionally, I would investigate the effect of the concrete filling method on the column behavior, as the compaction of the concrete can affect the bond and the confinement.

Study Insights and Implications

The key takeaway from this paper is that the GB 50936-2014 calculation method for CFST columns under axial compression is generally accurate for typical parameter ranges, but it needs adjustment for high slenderness ratios. The proposed modifications to the slenderness reduction factor and the suggested limits on material parameters are practical improvements that can be adopted by design engineers.

For the steel pipe industry, the paper reinforces the importance of controlling the dimensions and material properties of the steel tubes that are used in CFST structures. The D/t ratio and the yield strength are critical parameters that affect the column capacity, and the manufacturing tolerances must be consistent with the assumptions made in the design.

The paper also highlights the value of database-based analysis for evaluating design formulas. By collecting a large number of test results and comparing them with the calculated values, we can identify the strengths and weaknesses of a design method and make informed decisions about when to use it and when to apply additional safety factors.

In the end, this is a paper that every structural engineer working with CFST columns should read, and every steel pipe manufacturer supplying tubes for CFST applications should understand. The design formulas are only as good as the assumptions behind them, and the database analysis tells us exactly where those assumptions hold and where they break down.