Finite Element Analysis of Plastic Limit Load for Internally Pressurized Welded Pipe Tees

Literature Overview

This paper, published by Xuan Fuzhen, Liu Changjun, and Li Peining from the Chemical Machinery Institute of East China University of Science and Technology in 2001, presents a systematic finite element study on the plastic limit pressure of internally pressurized welded pipe tees. The research employs an ideal elastoplastic material model combined with a small deformation assumption to evaluate the limit load behavior of tees with diameter ratios d/D ranging from 0.5 to 1.0. The study also investigates the sensitivity of numerical results to mesh density, element type, and boundary condition formulations, providing valuable guidance for engineering applications where tee integrity under internal pressure is critical.

Core Technical Approach

The authors adopt an ideal elastoplastic constitutive model, which simplifies the material behavior to elastic response followed by perfectly plastic yielding without strain hardening. This assumption is justified for limit load analysis, where the objective is to determine the maximum load the structure can sustain before collapse, rather than to predict the load-displacement behavior during progressive deformation. The small deformation assumption is another key simplification, which eliminates the need for geometric nonlinearity in the formulation. While this may seem restrictive for structures undergoing significant plastic deformation, the authors demonstrate that for the range of diameter ratios studied (d/D ≥ 0.5), the small deformation results converge to values that are acceptable for engineering purposes.

The finite element model incorporates both the parent pipe and the branch pipe, with the weld region treated as part of the parent pipe material. The mesh density study reveals that refinement beyond a certain threshold yields diminishing returns in terms of solution accuracy, which is an important practical finding for engineers who must balance computational cost against result precision. The element type comparison likely involves both shell elements and solid elements, with the latter providing more accurate through-thickness stress distributions but at greater computational expense.

Key Findings and Their Engineering Significance

The most notable finding is that equal-diameter tees (d/D = 1.0) exhibit superior limit load capacity compared to unequal-diameter tees with d/D = 0.83 and d/D = 0.65, even when the parent pipe diameter-to-thickness ratio D/T is held constant. This result is particularly pronounced at larger D/T values, which correspond to thinner-walled parent pipes. From a design perspective, this implies that the transition from equal-diameter to unequal-diameter configurations introduces a disproportionate loss in structural capacity, and this loss is amplified in thin-wall applications.

The finding that small deformation analysis produces engineering-acceptable results is particularly valuable, as it means that engineers can avoid the computational complexity of large deformation formulations without sacrificing meaningful accuracy for limit load estimation. This is consistent with the general principle that limit load analysis, being concerned with the onset of collapse rather than post-yield deformation paths, is less sensitive to geometric nonlinearity than stress-strain analysis.

Reflections and Practical Implications

From a practical standpoint, this study reinforces the importance of understanding how geometry affects structural capacity in pipe fittings. In piping system design, unequal-diameter tees are frequently used to connect branches of different sizes to a main header, and the results here suggest that the design should not simply scale down from equal-diameter tee data. Engineers should be particularly cautious when designing thin-walled headers with small branch connections, as the combination of high D/T and low d/D creates a doubly unfavorable condition. The mesh and element sensitivity studies also serve as a useful reminder that numerical results are only as reliable as the computational model that produces them, and that appropriate validation studies should accompany any finite element analysis used for design decisions.