Full-Scale Model Experimental Design for Bending Resistance of Rectangular Steel Pipe Grid Arch Support Members in Tunnels

Engineering Context and Problem Statement

In tunnel engineering, particularly in urban metro and road tunnel construction, steel pipe grid arch supports are widely used as temporary and permanent support systems in complex geological conditions. Rectangular cross-section steel pipes are increasingly favored over circular pipes for their superior bending resistance, better spatial utilization, and easier integration with concrete lining systems. However, the bending performance of rectangular steel pipe grid arch members under complex loading conditions—particularly when subjected to asymmetric ground pressure—is not well documented in existing standards. This full-scale model experimental study provides essential design data for the safe application of rectangular steel pipe grid arch supports in tunnel engineering.

Specimen Design and Material Specifications

The full-scale specimens are designed to replicate actual tunnel support conditions with realistic dimensions and loading configurations.

The rectangular cross-section provides a bending stiffness (EI) that is approximately 1.5–2.0 times greater than an equivalent-area circular pipe, making it particularly advantageous for grid arch applications where bending moments dominate the design.

Experimental Methodology

The full-scale testing program employs a three-point and four-point bending test setup with instrumented loading frames capable of applying controlled displacements up to 300 mm. The test procedure follows a systematic approach:

1. Initial elastic loading to verify the theoretical stiffness predictions.
2. Incremental loading in 5 kN steps up to 50% of the estimated yield load.
3. Displacement-controlled loading at 2 mm/min beyond the yield point.
4. Continuation to failure or to a maximum displacement of 0.05 times the span.

Instrumentation includes strain gauges at critical locations (mid-span bottom, quarter-span top, and weld seams), displacement transducers at loading points and supports, and acoustic emission sensors to detect crack initiation and propagation.

Results and Design Formula Development

The bending resistance of rectangular steel pipe grid arch members follows a predictable pattern that can be captured by modified design formulas.

The ductility ratio (ultimate moment divided by yield moment) remains remarkably consistent at approximately 1.20–1.21 across all specimens, indicating that the rectangular cross-section provides good post-yield behavior. This is attributed to the presence of two flanges that continue to carry load even after the web yields, providing a stable plastic hinge mechanism.

Failure Modes and Defect Analysis

The primary failure modes observed in the full-scale tests include:

• Local buckling of the compressed flange: occurs when the width-to-thickness ratio exceeds 50 for Q345 steel or 45 for Q390 steel.
• Web crippling: develops when concentrated loads are applied near supports without adequate stiffening.
• Overall lateral-torsional buckling: occurs for slender members with insufficient lateral restraint.
• Weld seam failure: initiates at the longitudinal weld where residual stresses combine with applied bending stresses.

The longitudinal weld is the weakest link in welded rectangular pipes. Residual stresses from the welding process can reach 0.7–0.8 times the yield strength in the heat-affected zone, significantly reducing the local buckling resistance of the compressed flange near the weld. Post-weld stress relief (PWSR) at 600–620 °C for 2 hours per 25 mm of wall thickness is strongly recommended for critical applications.

Design Recommendations and Practical Guidelines

Based on the full-scale experimental results, the following design recommendations are provided for rectangular steel pipe grid arch support members:

• The bending design moment should be taken as M_d = W_p × f_y × 0.95, where W_p is the plastic section modulus and the 0.95 factor accounts for residual stress effects.
• The width-to-thickness ratio of the flange should be limited to 50 for Q345 and 45 for Q390 grades to prevent local buckling.
• Lateral bracing should be provided at intervals not exceeding 20 times the smaller cross-sectional dimension to prevent lateral-torsional buckling.
• For concrete-filled rectangular steel pipe grid arches, the composite bending capacity can be calculated using the superposition method with a confinement factor of 1.1–1.2 for the concrete contribution.
• Weld quality inspection should include 100% MT of all longitudinal and circumferential welds, with UT of critical joints per GB/T 11345 Level B.

The full-scale model experimental data presented in this study fills a significant gap in the existing design standards for rectangular steel pipe tunnel support systems. Engineers applying these findings should note that the test specimens represent idealized conditions, and field factors such as uneven ground pressure, construction tolerances, and long-term creep should be accounted for through appropriate safety factors of 1.3–1.5 in the design calculations. The consistent ductility ratios observed across all test specimens provide confidence that rectangular steel pipe grid arch members can be reliably designed using plastic analysis methods with appropriate reduction factors for residual stresses and imperfections.