Limit Load Analysis of Large-Opening Welded Reducers by Finite Element Method

Literature Overview

This paper by Zhang Xuewei (2013), published in Chemical Equipment and Piping, presents a finite element analysis (FEA) study of the limit load behavior of large-opening welded reducers (also referred to as large-opening welded tees or branch reducers). The author employs an idealized elastoplastic constitutive model in ANSYS to simulate the internal pressure response of the reducer geometry and determines the ultimate (limit) internal pressure using three distinct criteria: the zero-curvature criterion, the double-tangent-intersection criterion, and the twice-elastic-slope criterion. The study also examines how the limit pressure varies with different wall thickness combinations of the main pipe and branch pipe, providing design reference data for large-opening reducer applications.

Geometric Configuration and Loading Conditions

A large-opening welded reducer is a pipe fitting in which the branch opening diameter is a significant fraction of the main pipe diameter, typically exceeding the threshold defined in ASME B31.3 and B31.4 as the boundary between "small" and "large" openings. The conventional threshold is a branch diameter to main pipe diameter ratio (d/D) exceeding approximately 0.5, beyond which the stress distribution at the branch intersection deviates significantly from the solutions provided by classical small-hole pressure vessel theory.

The key geometric parameters considered in the FEA model include:

The loading condition is internal pressure applied uniformly to the inner surfaces of both the main pipe and the branch pipe. The model employs axisymmetric or 3D geometry depending on the branch orientation, with appropriate boundary conditions to simulate the continuity of the pipe run.

Limit Load Determination Criteria

The determination of the limit (collapse) load from an elastoplastic FEA load-displacement curve is not unique, as the curve typically exhibits a nonlinear transition from elastic to fully plastic behavior without a sharp discontinuity. Three criteria are employed in this study:

1. Zero-Curvature Criterion: The limit pressure is identified at the point where the curvature of the load-displacement curve changes sign, indicating the transition from strain hardening to softening behavior. This criterion is sensitive to the mesh density and material model near the yield point.
2. Double-Tangent-Intersection Criterion: Two tangent lines are drawn to the load-displacement curve — one representing the initial elastic slope and another representing the asymptotic plastic slope. The intersection of these tangents defines the limit pressure. This criterion provides a conservative estimate as it captures the onset of significant plastic deformation.
3. Twice-Elastic-Slope Criterion: The limit pressure is defined as the pressure at which the displacement reaches twice the value predicted by the linear elastic solution at the same pressure. This criterion is widely used in pressure vessel design codes and provides a practical, code-consistent definition of the collapse load.

The comparison of results from these three criteria reveals that the twice-elastic-slope criterion consistently yields the lowest limit pressure, making it the most conservative. The zero-curvature criterion produces the highest values, while the double-tangent method falls in between. For design purposes, the twice-elastic-slope criterion is recommended as it aligns with the safety philosophy of ASME and ISO pressure equipment codes.

Wall Thickness Effects and Design Implications

The parametric study reveals several important design trends:

• Increasing the branch pipe wall thickness has a more pronounced effect on the limit pressure than increasing the main pipe wall thickness, because the branch intersection is the primary location of plastic collapse initiation.
• The optimal wall thickness ratio (t_branch/t_main) for maximizing limit pressure is approximately 1.2 to 1.5, beyond which additional branch thickness provides diminishing returns.
• For d/D ratios exceeding 0.6, the limit pressure decreases significantly, and the stress concentration at the branch intersection becomes the dominant failure mechanism.
• The weld reinforcement at the branch intersection provides a beneficial stress redistribution, increasing the limit pressure by approximately 10-15% compared to a flush-welded configuration.

Engineering Practice and Code Compliance

The findings of this study have direct implications for the design and qualification of large-opening reducers in pressure piping systems. The ASME B31.3 stress evaluation method for branch connections (Appendix I) provides a simplified analytical approach that may be inadequate for large openings. The FEA-based limit load analysis provides a more accurate and reliable basis for design, particularly when:

• The d/D ratio exceeds 0.5
• The wall thickness ratio is significantly different from 1.0
• The service includes cyclic loading or thermal cycling
• The material exhibits significant strain hardening

The study also highlights the importance of weld geometry in the structural integrity of large-opening reducers. The fillet weld profile at the branch intersection acts as a stress relief feature, and its dimensions should be optimized to maximize the stress distribution benefit. Weld reinforcement beyond the minimum specified by the applicable code (typically ASME B31.3 or EN 13480) can provide additional structural benefit but must be balanced against the risk of excessive residual stress.

Study Insights and Reflections

This paper represents a valuable contribution to the understanding of large-opening reducer behavior under internal pressure loading. The use of multiple limit load criteria provides a robust framework for evaluating the results and selecting the most appropriate criterion for design purposes. The parametric study of wall thickness effects offers practical guidance for engineers who must specify wall thicknesses for large-opening reducers without relying solely on code-based simplified methods. The key insight is that the branch intersection is the critical location, and the wall thickness of the branch pipe should be the primary design variable. For engineering practice, this study supports the use of FEA as a supplementary design tool for large-opening reducers, particularly when code-based methods are either not applicable or produce overly conservative results that lead to unnecessary material cost. The idealized elastoplastic material model used in this study is a reasonable simplification for limit load analysis, though for fatigue assessment or cyclic loading evaluation, a more sophisticated material model incorporating cyclic hardening and softening behavior would be required.