Wall Thickness Analysis and Mathematical Modeling of Thickened Ends at Pipe Head and Tail

Literature Overview

This paper by Wang Yong, Zhang Min, and Long Gongming from the Technology Center of Hengyang Valin Steel Pipe Co., Ltd. was published in the journal Steel Pipe (Vol. 40, No. 3, 2011, pp. 22–26). The study addresses a persistent and practically significant problem in seamless steel pipe production: the unavoidable wall thickening that occurs at the head and tail ends of pipes during the piercing and rolling process. The authors conducted systematic measurements of wall thickness at these thickened ends, performed regression analysis on the distribution patterns, and developed both linear and nonlinear mathematical models to describe the wall thickness variation. The comparative evaluation of fitting accuracy between the two modeling approaches provides a theoretical basis for predicting the cut-off length of head and tail in actual production.

Core Technical Points

Physical Mechanism of End Thickening

During the hot piercing and rolling of seamless steel pipe, the plug or mandrel does not travel the full length of the pipe simultaneously with the rolling reduction. At the entry and exit ends, the reduction ratio is incomplete, resulting in a localized increase in wall thickness. This thickened region must be trimmed off to meet the dimensional tolerance requirements specified in standards such as GB/T 8162 or API 5CT. The length of this waste region directly affects material yield, which is a critical economic parameter in seamless pipe manufacturing.

Mathematical Modeling Approach

The authors established two classes of models:

The linear model uses reduction rate and average tension coefficient as the primary influencing factors. While it achieves acceptable accuracy for predicting the wall thickening rate, it falls short in predicting the actual length of the thickened end. The nonlinear model, by contrast, captures the complex interactions between process parameters and delivers high fitting accuracy for both the wall thickening rate and the thickened end length.

Key Process Parameters

Engineering Practice Integration

In production practice, the thickened ends represent a direct material loss. For a typical seamless pipe mill producing 108 mm × 4.5 mm pipe, the head and tail thickening may account for 300–500 mm of waste per pipe. When multiplied across thousands of pipes per production campaign, this waste becomes economically significant. The nonlinear model proposed in this paper enables process engineers to predict the thickened end length with high accuracy based on known process parameters, allowing for optimized cutting decisions.

Practical Application Scenarios

1. Cut-off length prediction: By inputting the actual reduction rate and tension coefficient of a given production batch into the nonlinear model, the operator can determine the minimum cut-off length required to ensure the remaining pipe meets dimensional specifications.
2. Yield optimization: The model allows comparison of different piercing and rolling parameter combinations to identify the combination that minimizes end thickening while maintaining acceptable pipe quality.
3. Quality control feedback: When measured thickened end lengths deviate from model predictions, this signals process instability that requires investigation—possibly indicating plug wear, temperature deviation, or tension control issues.

Key Questions and Reflections

One critical question that arises from this study is the generalizability of the nonlinear model. The model was developed based on data from a specific seamless pipe production line at Hengyang Valin. The geometric and process parameters of different mills—such as different plug designs, rolling mill configurations, and temperature control strategies—may alter the functional relationships between parameters. Engineers applying this model to a different production environment should validate the model against their own measurement data before relying on it for production decisions.

Another reflection concerns the boundary conditions of the model. The study focuses on wall thickening at the head and tail ends, but in practice, other dimensional defects such as ellipticity, diameter deviation, and surface irregularities may also be more pronounced at the pipe ends. A comprehensive end quality model would ideally integrate all dimensional and surface quality parameters, not just wall thickness.

Study Insights and Implications

The value of this paper lies in its systematic approach to a seemingly simple but economically important problem. The demonstration that nonlinear modeling significantly outperforms linear modeling for thickened end length prediction is a valuable lesson for process engineers. In seamless pipe manufacturing, the relationships between process parameters and product quality are inherently nonlinear due to the complex plastic deformation behavior of steel at high temperatures. Linear approximations may be adequate for certain quality parameters where the response is relatively monotonic, but for parameters that exhibit threshold behavior or interaction effects—such as thickened end length—nonlinear models are essential.

This work also highlights the importance of quantitative measurement and data-driven process optimization in steel pipe manufacturing. Rather than relying on empirical rules of thumb for cut-off length determination, the adoption of validated mathematical models enables more precise and consistent production decisions, ultimately improving material yield and reducing waste. The methodology can be extended to other geometric features of seamless pipe production, such as the prediction of surface defects, diameter uniformity, and residual stress distribution along the pipe length.