Experimental Study and Discussion on Stress Intensification Factors for Tees

Literature Overview

This paper by He Zhimin and Dong Bangping, published in Electric Power Construction in 1994, presents an experimental investigation of stress intensification factors (SIF) for tees, specifically focusing on ultra-thick main pipe 45-degree welded oblique tees. The study is significant because it bridges the gap between theoretical SIF calculations and experimental reality, proposing a general calculation formula and introducing a probabilistic approach to SIF determination. The paper is classified under strength of materials (TH49) and is directly relevant to pressure vessel and piping design, particularly in the power generation industry.

Core Technical Points

The stress intensification factor is a critical parameter in piping design that accounts for the geometric discontinuity at a tee junction. According to ASME B31.3 and related codes, the SIF is used to calculate the stress at the intersection of a branch pipe and a run pipe, where stress concentrations arise due to the change in cross-sectional area and the resulting load redistribution. The conventional approach to SIF determination relies on empirical formulas derived from a limited set of test data, which may not accurately represent the behavior of all tee geometries, particularly those with non-standard angles or unusual wall thickness ratios.

The authors conducted fatigue tests on ultra-thick main pipe 45-degree welded oblique tees, which represent a challenging configuration because the oblique angle introduces additional stress concentration beyond that of a standard 90-degree tee. The test results were analyzed to derive a general calculation formula for the SIF of this class of tees. The formula likely incorporates parameters such as the diameter ratio (d/D), the wall thickness ratio (t/d), the branch angle, and the geometric ratio (C) defined in ASME B31.3.

A key innovation in this paper is the application of probabilistic methods to analyze the fatigue test data. The authors recognize that SIF is not a deterministic quantity but rather a statistical one, subject to scatter due to manufacturing variations, material inhomogeneities, and test specimen geometry differences. By applying probability methods, the authors propose the concept of SIF at a given reliability level, which allows engineers to select a design SIF that corresponds to the required structural reliability. This is a significant departure from the deterministic approach used in most code provisions, where a single SIF value is specified for a given geometry.

Standards and Code Context

The SIF concept is embedded in multiple piping codes and standards. ASME B31.3 provides SIF values for tees in its piping stress analysis provisions, while ASME B31.4 and B31.8 address similar requirements for liquid and gas transmission pipelines. The API 5L and ASME B31.3 interaction between these standards must be carefully managed in practice, as different standards may prescribe different SIF values for the same geometry. The authors' work is particularly relevant to the power industry, where the Electric Power Construction journal is published, and where piping systems operate under demanding conditions of high temperature, high pressure, and cyclic loading.

The experimental approach used by the authors is consistent with the methodology recommended by ASME B31.3 for determining SIF values. The code recognizes that SIF values are derived from experimental data and provides a framework for extrapolating from test data to design conditions. The authors' contribution extends this framework by proposing a probabilistic extension that accounts for the inherent variability in both the test data and the actual field conditions.

Engineering Practice Implications

For practicing engineers, the key takeaway from this paper is that SIF values should not be treated as fixed constants but should be selected based on the required reliability level and the specific geometry of the tee. In critical applications, such as power plant steam lines or nuclear piping, where the consequences of failure are severe, a higher reliability level may justify the use of a more conservative SIF value. Conversely, in less critical applications, a lower SIF value may be acceptable if the reliability requirement is less stringent.

The paper also highlights the importance of considering the branch angle in SIF calculations. A 45-degree oblique tee, as studied by the authors, will generally have a higher SIF than a 90-degree tee of the same diameter and thickness ratios. Engineers must ensure that the SIF used in design calculations corresponds to the actual branch angle of the fitting, rather than assuming a standard 90-degree value. This is particularly important in power plant applications, where oblique tees are commonly used to optimize pipe routing and reduce the number of elbows in the system.

Reflections and Outlook

The probabilistic approach proposed by the authors is ahead of its time and anticipates the modern trend toward risk-based inspection and reliability-based design in the engineering community. While the paper was published in 1994, its methodology remains relevant today, particularly in the context of structural integrity assessment and life extension of aging piping systems. The concept of SIF at a given reliability level can be integrated into modern probabilistic risk assessment frameworks, providing a more nuanced basis for inspection and maintenance decisions. Engineers should consider adopting this probabilistic perspective in their design and assessment work, as it leads to more rational and economically efficient engineering decisions.