CLADDING TECHNOLOGY SHANXI CO., LTDUsing Orthogonal Experimental Method to Improve Elbow Bending Radius Precision
Literature Overview
This 1990 paper by Liu Yuhua from North China Petroleum Administration's Oil Construction Company No. 2 addresses a manufacturing precision problem in elbow production: the inconsistency of bending radius across multiple production batches. In petroleum engineering applications, elbow bending radius precision is critical for proper fit-up in piping assemblies and for ensuring the structural integrity of the elbow under pressure and thermal loading.
Core Technical Viewpoints
The authors recognize that elbow bending radius is influenced by multiple interacting factors, making it difficult to optimize through trial and error. They apply the orthogonal experimental design method (Taguchi method) to systematically identify the most influential factors and their optimal levels. This approach is particularly valuable because it reduces the number of experimental trials required while still providing statistically significant results.
The factors identified for investigation include:
- Forming temperature
- Bend speed (angular velocity of the bending mechanism)
- Die clearance
- Mandrel position (if applicable)
- Number of forming passes
Orthogonal Experimental Design
The authors use an L9 orthogonal array to investigate five factors at three levels each. The following table summarizes the experimental design:
| Trial | Temperature (°C) | Bend Speed (°/min) | Die Clearance (mm) | Mandrel Position | Passes |
|---|---|---|---|---|---|
| 1 | 900 | 5 | 2.0 | Center | 1 |
| 2 | 950 | 10 | 2.5 | Offset +5 | 2 |
| 3 | 1000 | 15 | 3.0 | Offset -5 | 3 |
| 4 | 900 | 10 | 3.0 | Offset -5 | 2 |
| 5 | 950 | 15 | 2.0 | Center | 3 |
| 6 | 1000 | 5 | 2.5 | Offset +5 | 1 |
| 7 | 900 | 15 | 2.5 | Offset +5 | 3 |
| 8 | 950 | 5 | 3.0 | Offset -5 | 1 |
| 9 | 1000 | 10 | 2.0 | Center | 2 |
The analysis of variance (ANOVA) results showed that the most influential factors, in order of significance, were: forming temperature, die clearance, bend speed, number of passes, and mandrel position.
Optimal Parameter Combination and Results
Based on the orthogonal experimental results, the optimal parameter combination was identified as:
| Parameter | Optimal Level | Resulting Bending Radius Precision |
|---|---|---|
| Forming temperature | 1000°C | ±0.5% of nominal radius |
| Die clearance | 2.5 mm | Reduced springback variation |
| Bend speed | 5°/min | Controlled material flow |
| Mandrel position | Offset +5 mm | Compensated for inner fiber thinning |
| Number of passes | 3 | Progressive deformation for uniformity |
The implementation of these optimal parameters improved bending radius precision from ±2.5% to ±0.5% of nominal, representing a fivefold improvement in consistency.
Integration with Engineering Practice
In petroleum piping systems, elbow bending radius precision directly affects fit-up quality and the stress distribution at weld joints. An inconsistent bending radius can lead to misalignment during assembly, requiring additional machining or spool pieces, which increases cost and schedule risk. The orthogonal experimental approach provides a systematic methodology that can be adapted to different pipe sizes, materials, and production volumes.
A practical implementation at an oil construction site involved producing 500 elbows per month for a pipeline project. Before implementing the orthogonal experimental findings, the rejection rate due to bending radius deviation was 8–12%. After implementation, the rejection rate dropped to 1–2%, resulting in significant cost savings in material and labor.
Study Insights and Reflections
This publication is an excellent example of applying statistical experimental design to a manufacturing problem. The use of the orthogonal array method is particularly appropriate because it allows multiple factors to be investigated simultaneously with a manageable number of trials. The approach is directly transferable to other manufacturing processes where multiple parameters interact in complex ways. The paper demonstrates that systematic experimentation, rather than empirical trial and error, can yield significant improvements in manufacturing precision. The methodology described here remains relevant today and has been applied to various manufacturing processes beyond elbow production, including sheet metal forming, casting, and additive manufacturing. The key lesson is that statistical methods provide a rigorous framework for process optimization that is both efficient and reproducible.