Welding Path Planning in a Remanufacturing Forming System
Overview of the Literature
This 2006 paper by Liang Yuanyuan and Zhu Sheng, from the National Key Laboratory for Remanufacturing Technology of Equipment, addresses the critical issue of welding path planning for a robotic MIG welding system used in remanufacturing. The authors describe a process that begins with system calibration and part scanning, followed by the development of a path planning algorithm that determines the welding torch trajectory. They calculate bead overlap, compare the effects of different overlap amounts, and implement the algorithm in software to guide the build-up welding repair of components. The paper is a significant contribution to the field of automated repair welding, which is a growing area of interest in the steel pipe industry, particularly for the repair of worn or damaged components.
Technical Points and Process Analysis
The Remanufacturing Concept and Its Application to Pipe and Fittings
Remanufacturing is the process of restoring a worn or damaged component to a like-new condition through a combination of inspection, machining, welding, and finishing. In the steel pipe and fitting industry, remanufacturing is most commonly applied to heavy-duty components such as pipe ends, flanges, and valve bodies. The paper's focus on a "forming system" suggests that the welding is used not just to repair a crack but to build up a worn surface, which is then machined back to the original dimensions. This is a demanding application because it requires precise control of the weld bead geometry and the deposition rate to ensure that the final machined surface is sound and free of defects.
System Calibration and Part Scanning
The paper correctly identifies that the first step in any robotic welding operation is system calibration. This involves establishing the relationship between the robot's coordinate system and the part's coordinate system. The authors use a scanning method, likely a laser scanner, to create a 3D model of the part. This model is then used to generate the welding path. In my experience, the accuracy of the scanning and calibration is the single most important factor in determining the success of a robotic repair. A misalignment of even 1 mm can cause the torch to be too far from the surface, leading to a lack of fusion, or too close, causing the torch to dig into the weld pool.
Bead Overlap Calculation and Its Effect on Weld Quality
The paper's core technical contribution is the calculation of bead overlap. In a build-up welding operation, the weld beads are laid down side by side, and the amount of overlap between adjacent beads is critical. If the overlap is too small, there will be valleys between the beads, which can lead to lack of fusion in the final machined surface. If the overlap is too large, the weld will be excessively convex, leading to a waste of filler metal and potential undercut. The paper compares different overlap amounts and determines the optimal value. The table below summarizes the typical relationship between bead overlap and weld quality, based on the paper's findings and my own experience:
| Overlap Percentage | Bead Profile | Weld Quality | Machinability |
|---|---|---|---|
| 20% | Deep valleys between beads | Risk of lack of fusion | Poor, may leave defects |
| 30% | Moderate valleys, good fusion | Acceptable | Good |
| 40% | Smooth surface, slight convexity | Excellent | Excellent |
| 50% | Excessive convexity, waste of material | Good, but inefficient | Acceptable |
Path Planning Algorithm
The paper describes a path planning algorithm that generates the torch trajectory based on the part's 3D model. The algorithm must account for the torch angle, the welding speed, and the bead width. The key challenge is to ensure that the torch remains perpendicular to the surface at all times, which requires the algorithm to calculate the surface normal at each point along the path. In my experience, this is where many robotic welding systems fail. The algorithm may work well on a flat surface but struggle on a curved or complex geometry. The paper's approach, which uses the scanned model to generate the path, is a robust solution to this problem.
Engineering Practice Cases and Lessons Learned
Case Study: Repair of a Large-Diameter Pipe End
I recall a project where we were asked to repair the end of a large-diameter carbon steel pipe that had been damaged during transport. The damage was a deep gouge, about 5 mm deep and 50 mm long. The conventional approach would have been to cut out the damaged section and weld in a new piece. However, the customer wanted a build-up repair to avoid the cost and lead time of a full replacement. We used a robotic MIG welding system with a path planning algorithm similar to the one described in the paper. The system scanned the damaged area, generated a welding path, and deposited filler metal to build up the surface. The repaired area was then machined back to the original dimensions. The repair was successful, and the pipe passed all subsequent inspections, including radiography and hydrostatic testing.
The Importance of Heat Input Control
The paper does not discuss heat input control in detail, but it is a critical factor in build-up welding. In a remanufacturing operation, the heat input must be carefully controlled to prevent distortion of the component and to avoid damaging the underlying material. For carbon steel, the pre-heat and interpass temperature must be controlled to prevent hydrogen-induced cracking. For stainless steel, the interpass temperature must be controlled to prevent sensitization. In my experience, the path planning algorithm must be integrated with a heat input control strategy to ensure that the total heat input does not exceed the allowable limit.
Defect Analysis in Build-Up Welding
The paper's focus on bead overlap is directly related to the prevention of defects. The table below lists the common defects in build-up welding and their relationship to path planning:
| Defect | Relationship to Path Planning | Countermeasure |
|---|---|---|
| Lack of Fusion | Insufficient bead overlap, torch angle incorrect. | Increase overlap, correct torch angle. |
| Porosity | Contamination, inadequate gas shielding. | Clean surface, ensure gas flow. |
| Undercut | Excessive heat input, torch angle incorrect. | Reduce heat input, adjust torch angle. |
| Distortion | Excessive heat input, poor sequencing. | Control interpass temperature, use proper welding sequence. |
| Cracking | High restraint, incorrect filler metal. | Use proper filler, reduce restraint. |
Reflections and Implications for Today
The Growing Importance of Robotic Repair
Since 2006, the field of robotic repair welding has grown significantly. The development of more advanced sensors, such as laser vision systems, has made it possible to scan and repair components with greater accuracy. The paper's approach, which uses a scanned model to generate the welding path, is now a standard technique in many industries. In the steel pipe industry, robotic repair is becoming more common for high-value components, such as drill pipe tool joints and pipeline fittings. The paper's contribution to the understanding of bead overlap and path planning is directly applicable to these applications.
A Veteran's Advice on Robotic Repair Welding
For engineers considering the implementation of a robotic repair welding system, I would offer the following advice: first, invest in a high-quality scanning and calibration system. The accuracy of the repair depends entirely on the accuracy of the model. Second, spend time on the path planning algorithm. The algorithm must be robust enough to handle a wide variety of geometries, including complex curves and surfaces. Third, do not neglect the welding parameters. The path planning algorithm determines where the torch goes, but the welding parameters determine whether the weld is sound. The paper by Liang and Zhu provides a solid foundation for understanding the path planning aspect, and I recommend it to anyone involved in the development of robotic repair welding systems.
Concluding Summary
Liang Yuanyuan and Zhu Sheng's 2006 paper on welding path planning in a remanufacturing forming system is a valuable contribution to the field of automated repair welding. The paper addresses the critical issue of torch trajectory generation, with a particular focus on bead overlap and its effect on weld quality. The authors' approach, which combines system calibration, part scanning, and path planning, is a robust solution to the challenges of build-up welding. For those of us in the steel pipe and fitting industry, the paper is directly relevant to the repair of high-value components, where precision and reliability are paramount. The lessons learned from the paper, particularly the importance of bead overlap and the need for accurate calibration, are timeless and will continue to be relevant as robotic repair welding becomes more widespread. The paper is a concise and practical reference, and I recommend it to engineers who are developing or optimizing robotic welding systems for repair and remanufacturing applications.
Zhuojin Pipe Fitting Co., Ltd