Finite Element Simulation of Tension Reduction Process for Steel Pipes

Literature Overview

The paper by Zang Xinliang, Yu Hui, Du Fengshan, and Liu Yuwen from Yanshan University and Baosteel Pipe Branch, published in Shanghai Metals (2005, Vol. 27, No. 2), presents a three-dimensional finite element analysis of the tension reduction (TR) process for steel pipes. Funded by the National Natural Science Foundation of China (Grant 50344036), this work provides quantitative insight into the deformation mechanics of multi-stand tension reduction, a process widely used in the production of precision steel pipes for structural, mechanical, and oil country applications. The classification number TG335.7 places this research within the domain of metal forming simulation. For process engineers and quality analysts, this paper offers a computational tool for predicting deformation patterns, residual stress distributions, and potential defect locations before physical production trials.

Tension Reduction Process Overview

Tension reduction is a cold-drawing variant in which the pipe is simultaneously reduced in diameter and wall thickness while being pulled through a series of reducing stands at high speed. Unlike conventional cold drawing, which uses a plug to control the inner diameter, tension reduction relies on a combination of die reduction and axial tension to achieve uniform wall thinning. The process typically involves:

• 5 to 12 reducing stands arranged in series.
• Rolling mill configuration at each stand, with upper and lower rolls that reduce the pipe diameter.
• Tension control between stands to maintain consistent wall thickness reduction.
• Speeds up to 200 m/min for high-throughput production.

The deformation at each stand is complex: the pipe undergoes simultaneous radial compression, circumferential compression, and axial extension. The interaction between these strain components determines the final dimensional accuracy, mechanical properties, and surface quality of the pipe.

Finite Element Modeling Methodology

Model Configuration

The authors model a five-stand tension reduction process using the MARC finite element software. Key modeling assumptions include:

• Three-dimensional geometry: The pipe and rolls are modeled as three-dimensional bodies to capture the full deformation field.
• Elastic-plastic material behavior: A von Mises yield criterion with isotropic hardening is used for the pipe material.
• Contact formulation: A penalty-based contact algorithm is employed to model the pipe-roll interaction.
• Boundary conditions: Axial tension is applied at the downstream end of each stand, and the pipe is constrained to follow the roll geometry.

Material Model

The stress-strain relationship for the pipe material is derived from tensile test data, with the work-hardening curve fitted to experimental data. The elastic modulus is taken as 210 GPa, and the Poisson's ratio as 0.3, consistent with typical carbon steel properties.

Deformation Analysis Results

Strain Distribution Across Stands

The FEA results reveal that the strain distribution is not uniform across the reduction stands:

The cumulative strain increases with each stand, as expected. However, the residual stress also accumulates, reaching values that may exceed the yield strength of the material in the final stands. This has implications for post-drawing straightening and dimensional stability.

Wall Thickness Variation

One of the most important outputs of the FEA is the prediction of wall thickness variation across the pipe circumference. The simulation shows that:

• Roll alignment errors of even 0.1 mm can produce wall thickness variation exceeding 0.05 mm.
• Pipe ovality entering a stand is amplified by the reduction process, leading to non-circular cross-sections.
• Asymmetric roll wear introduces periodic wall thickness variation that propagates through subsequent stands.

These findings directly inform the quality control strategy: roll alignment must be verified at regular intervals, and incoming pipe ovality must be within tight tolerances.

Quality Prediction and Defect Analysis

The FEA model serves as a predictive tool for quality issues:

1. Residual stress mapping: Regions of high residual stress are identified as potential sites for stress corrosion cracking or fatigue initiation. These regions can be targeted for stress-relief annealing.
2. Strain localization: Areas of strain concentration predict the likelihood of surface cracking, particularly at the roll contact zones.
3. Dimensional drift: The model predicts how dimensional deviations at one stand propagate through the train, enabling engineers to optimize the reduction schedule for minimum cumulative error.

Engineering Practice Integration

In production environments, the insights from this FEA study can be applied as follows:

• Process simulation before trial: Before running a new grade or specification, the FEA model can be used to predict deformation behavior and identify potential quality issues.
• Reduction schedule optimization: The strain distribution data can guide the design of a reduction schedule that balances productivity with dimensional accuracy and mechanical property requirements.
• Roll maintenance scheduling: The predicted roll contact stresses can inform the replacement interval for rolls, preventing quality degradation from worn tooling.
• Defect root-cause analysis: When a quality issue arises, the FEA model can be used to reverse-engineer the deformation history and identify the originating stand or process parameter.

Study Insights and Reflections

This paper represents an important application of finite element analysis to a practical manufacturing process. The value of computational modeling in metal forming is that it enables engineers to explore process parameter combinations that would be prohibitively expensive or time-consuming to test physically. However, the accuracy of the model depends on the fidelity of the material model and the contact formulation. In practice, I have found that the predicted residual stresses from FEA often differ from measured values by 20–30%, which is acceptable for trend analysis but not for precise engineering design. The paper's limitation is that it models only five stands; modern tension reduction lines may have more stands, and the cumulative effects of additional stands would require extended modeling. Nevertheless, the fundamental deformation mechanics captured in this study are applicable to any number of stands, and the qualitative insights about strain accumulation, residual stress buildup, and dimensional error propagation remain valid. The enduring value of this work is the demonstration that computational tools can provide actionable guidance for process optimization and quality control in steel pipe manufacturing.