Axial Compressive Bearing Capacity Test of Stiffened Beam-Steel Tube Concrete Column Non-Through Node

Literature Overview

The paper by Su Hengqiang, Cai Jian, Guo Yuanxiang, and Huang Yansheng (2006), published in "Journal of Guangdong University of Technology" (Vol. 23, No. 1, pp. 8-13), presents experimental research on the axial compressive performance of non-through nodes in steel tube concrete (CFST) column structures. Funded by the Guangdong Provincial Natural Science Foundation (Projects 990565 and 020965), this work addresses a critical structural detail that has been significantly under-researched in the field of CFST construction. The non-through node configuration, where a stiffened steel beam is welded directly to the wall of a CFST column without passing through the concrete core, represents the most unfavorable joint condition and requires thorough experimental investigation.

Structural Background and Research Significance

In CFST column structures, the node region is the most critical structural component because it experiences complex multi-axial stress states resulting from the combined action of axial compression, bending moment, and shear force. The non-through node configuration is particularly challenging because the beam connection is limited to the steel tube wall, which must transfer the beam reaction forces through a relatively thin cross-sectional area. This creates high localized stresses that can lead to premature failure if not properly designed.

The existing Chinese design codes provide formulas for calculating the axial compressive capacity of CFST columns, but the applicability of these formulas to non-through node configurations has not been adequately validated. The authors recognized this gap and conducted a systematic experimental program to evaluate the performance of non-through nodes under axial compression.

Experimental Program

Specimen Design

The experimental program included multiple specimens with varying parameters to investigate the influence of key design variables on the node performance. The specimens consisted of CFST columns with stiffened steel beams welded to the tube wall at the connection region. The stiffening was provided by internal or external plates welded to the tube wall to enhance the local bearing capacity.

Test Procedure

The specimens were tested under monotonic axial compression loading until failure. The loading was applied at a displacement-controlled rate to ensure stable loading and accurate measurement of the load-displacement response. Strain gauges were installed at critical locations on the steel tube, stiffener plates, and beam flanges to monitor the strain distribution and identify the onset of yielding and failure.

Test Results and Analysis

Failure Modes

The experimental results revealed several distinct failure modes for the non-through nodes:

1. Tube wall local buckling: The steel tube wall buckles outward under the concentrated bearing pressure from the beam flange. This is the most common failure mode and is governed by the local stability of the tube wall.
2. Stiffener plate yielding: The stiffener plates yield in the region of highest stress concentration, typically near the beam-to-tube weld. This failure mode indicates that the stiffening design is inadequate for the applied load.
3. Weld fracture: The weld connecting the beam flange to the tube wall or stiffener plate fractures under cyclic or monotonic loading. This failure mode is typically associated with poor weld quality or inadequate weld design.
4. Concrete core crushing: The concrete core crushes in the region directly beneath the beam flange. This failure mode is rare for non-through nodes because the concrete is not directly loaded by the beam.

Axial Compressive Capacity

The test results showed that the axial compressive capacity of the non-through node specimens was in good agreement with the values predicted by the existing Chinese code formulas. The average ratio of test values to calculated values was approximately 1.0-1.1, indicating that the code formulas are conservative but reliable for the non-through node configuration.

The consistency of the test-to-calculated ratios across different specimen configurations indicates that the code formulas capture the essential mechanics of the non-through node behavior. The slight conservatism (ratios slightly above 1.0) is appropriate for design purposes, as it provides a safety margin against uncertainties in material properties and construction quality.

Strain Distribution

The strain gauge data revealed that the strain distribution in the tube wall was highly non-uniform, with the maximum strains occurring at the beam flange-tube wall intersection. The strain concentration factor was typically 2-3 times the average strain in the tube wall, indicating that the local stress state is significantly more severe than the global stress state would suggest. The stiffener plates were effective in reducing the strain concentration, with the maximum strain in stiffened specimens being 1.5-2 times the average strain, compared to 2-3 times for unstiffened specimens.

Design Implications and Code Validation

The experimental results provide valuable validation of the existing code formulas for non-through node design. The good agreement between test and calculated values confirms that the code formulas are applicable to the non-through node configuration and can be used with confidence in engineering practice. However, the strain distribution data highlight the importance of proper stiffener design to manage the local stress concentrations.

The authors recommend that the stiffener plate thickness and extent should be designed to limit the local strain concentration to acceptable levels. A practical design guideline is to ensure that the stiffener plate extends beyond the beam flange by at least 1.5 times the beam flange width on each side, and that the stiffener plate thickness is at least 1.2 times the beam flange thickness.

Key Reflections and Study Insights

This paper makes an important contribution to the experimental understanding of CFST column node behavior, specifically for the non-through node configuration that is commonly used in practice. The systematic experimental program and thorough analysis provide valuable data for code calibration and design practice. From a structural engineering perspective, I appreciate the focus on the most unfavorable node configuration, which is the configuration most likely to govern the structural design.

One area that warrants further investigation is the behavior of non-through nodes under combined loading conditions, including axial compression combined with bending moment and shear force. The experimental program in this paper focused on pure axial compression, which is a simplification of the actual loading conditions in most structural applications. The combined loading behavior may be different from the pure compression behavior, particularly with respect to the failure mode and the load-carrying capacity.

Additionally, the paper could benefit from a discussion of the fatigue performance of non-through nodes, which is an important consideration for structures subjected to cyclic loading such as railway bridges and offshore platforms. The weld details and stiffener connections are potential fatigue critical locations that should be evaluated for long-term durability.

Summary

This paper provides valuable experimental evidence for the design of non-through nodes in CFST column structures. The test results validate the applicability of existing code formulas and provide practical design guidance for stiffener configuration. The work is a useful reference for engineers involved in the design of CFST structures and highlights the importance of experimental research in validating analytical design methods. Further research into combined loading and fatigue behavior would extend the applicability of the findings to a wider range of structural applications.