Numerical Simulation of Erosion Effects on Tees Under Asymmetric Dual-Inlet Conditions in Natural Gas Pipelines

Literature Overview

The paper by Fu Wencong (2025), published in Petroleum and Natural Gas Journal (Vol. 47, No. 4, pp. 653–669), addresses a critical but under-explored scenario in natural gas pipeline integrity management: the erosion behavior of tee fittings under "two-in-one-out" operating conditions where the inlet parameters of the main pipe and branch pipe differ significantly. The study employs a mixture multiphase flow model coupled with the discrete phase model (DPM) to simulate gas-liquid-solid three-phase flow within the tee, combined with the E/CRC erosion model to quantify erosion patterns and rates. This work is particularly relevant to engineers dealing with natural gas gathering systems where multiple well streams converge at tee junctions with inherently different flow conditions.

Core Technical Points and Interpretation

The study identifies three key variables that influence tee erosion under asymmetric inlet conditions: inlet velocity difference, inlet liquid volume fraction difference, and pipe diameter difference between the main and branch inlets. Each variable alters the internal flow field distribution, which in turn governs where erosion occurs, what shape it takes, and how rapidly material is removed.

The E/CRC model used in this study is a well-established erosion prediction framework that calculates erosion rate based on particle impact velocity, impact angle, particle size, and material hardness. The model assumes that erosion rate is proportional to the kinetic energy of impacting particles and follows a power-law relationship with impact velocity. For tee fittings, the most vulnerable locations are typically the downstream wall opposite the branch pipe inlet (the "impact zone") and the downstream wall of the branch pipe outlet, where high-velocity jets impinge upon the pipe wall.

The finding that increasing branch pipe diameter leads to a "larger erosion area but lower erosion rate" inverse relationship is particularly noteworthy. This occurs because a larger branch diameter reduces the jet velocity at the branch inlet, which decreases particle impact energy per unit area, but simultaneously allows more particles to be dispersed over a wider area of the downstream wall. From a pipeline integrity perspective, this means that larger-diameter branches may experience more widespread but shallower erosion, which could be missed by conventional point-based ultrasonic thickness measurements but detected by full-bore scanning or phased array ultrasonic testing (PAUT).

Connection with Engineering Practice

In natural gas gathering systems, tee junctions are frequently subjected to sand-laden gas streams, especially in high-permeability formations such as shale gas reservoirs. The China Petroleum and Chemical Corporation (Sinopec) and CNPC standards require periodic inspection of tee fittings in high-erosion-risk sections, typically using magnetic flux leakage (MFL) or ultrasonic testing. The findings of this study suggest that inspection strategies should be adapted based on the specific inlet conditions at each tee location.

For field application, I recommend the following engineering practices derived from this study:

1. Pre-installation assessment: When designing tee junctions for natural gas gathering, the expected inlet conditions (velocity, liquid content, particle loading) should be documented and used to predict erosion-prone zones.
2. Inspection interval optimization: Tees with high branch liquid volume fractions should have shorter inspection intervals due to the accelerated erosion rate and the tendency for erosion to expand from elliptical to bell-shaped patterns.
3. Material selection: For tees subjected to severe asymmetric erosion conditions, consider using corrosion-resistant alloy (CRA) materials such as 13Cr or 22Cr duplex stainless steel, which offer superior erosion-corrosion resistance compared to standard carbon steel or low-alloy steel grades specified in API 5L or ASTM A234.
4. Flow conditioning: Installing flow conditioners or cyclone separators upstream of tee junctions can reduce liquid carry-under and particle concentration, thereby mitigating erosion at the tee itself.

The study also highlights an important limitation: the numerical simulation, while providing valuable qualitative and semi-quantitative insights, does not account for the coupled effects of erosion and corrosion, which are often synergistic in real natural gas environments containing H₂S and CO₂. Future work should integrate electrochemical corrosion modeling with the erosion prediction framework to provide more comprehensive lifetime assessments.

Key Questions and Reflections

Several questions arise from this study that warrant further investigation. First, the study assumes steady-state flow conditions, but in real natural gas gathering systems, flow rates fluctuate significantly due to well production variations, compressor cycling, and pipeline shutdowns. How do transient flow conditions affect erosion accumulation at tee junctions? Second, the study does not address the effect of operating temperature on erosion rate, which is significant in deep-well natural gas production where temperatures can exceed 100°C at the wellhead and decrease to ambient at the gathering station. Third, the E/CRC model parameters used in the simulation should be validated against field data from actual tee inspections to ensure predictive accuracy.

From a metallurgical perspective, the erosion damage morphology described in the study—elliptical expanding to bell-shaped—is consistent with the "erosion pit" formation mechanism observed in metallographic studies of eroded pipeline components. The initial elliptical pit forms at the point of maximum particle impact angle (typically 30°–45° for ductile materials), and as material is removed, the flow field reconfigures, causing the erosion zone to expand laterally. This progressive damage mechanism means that early-stage erosion may be difficult to detect, and by the time it is visible, significant material loss may have already occurred.

Study Insights and Implications

The most valuable contribution of this study is the systematic identification of how asymmetric inlet conditions alter tee erosion behavior, providing engineers with a framework for risk assessment at tee junctions in natural gas gathering systems. The inverse relationship between erosion area and erosion rate under varying branch diameters is a particularly important finding that challenges the conventional assumption that larger fittings are inherently less prone to erosion damage. In practice, this means that inspection protocols must consider both the severity and the extent of erosion, not just the depth of material loss.

For my own practice in pipeline integrity management, I will incorporate the following lessons: when evaluating tee fittings for remaining life, always consider the specific inlet conditions at that location rather than applying generic erosion allowance values; when selecting inspection methods, prioritize area-survey techniques such as PAUT or MFL over point measurements for tees in high-erosion-risk locations; and when specifying materials for new tee installations, request erosion-corrosion testing data under conditions that replicate the expected asymmetric flow environment rather than relying solely on standard corrosion resistance data from ASTM G102 or NACE MR0175.

The study also underscores the importance of computational fluid dynamics (CFD) as a tool for pipeline integrity engineering. While CFD cannot replace physical testing or field inspection, it provides invaluable guidance for identifying high-risk locations, optimizing inspection strategies, and evaluating the effectiveness of mitigation measures such as flow conditioners or material upgrades. As computational resources become more accessible, I expect CFD-based erosion prediction to become a standard part of pipeline integrity management programs, particularly for critical tee junctions in high-risk environments.