CLADDING TECHNOLOGY SHANXI CO., LTDCNC Tracking System for Cladding of Pressure Vessel Heads
Literature Overview
This 2002 publication from Yanshan University, authored by Yu Zhonghai and published in Manufacturing Technology & Machine Tools, addresses a critical challenge in pressure vessel fabrication: the automated cladding of hemispherical and elliptical heads. Heads are among the most difficult components to clad because of their three-dimensional curvature, which makes manual weld tracking unreliable and inconsistent. The paper describes the design and implementation of a computerized numerical control (CNC) tracking system specifically configured for weld overlay operations on vessel heads, representing an important early contribution to the automation of overlay welding in China.
Core Technical Approach
The fundamental problem addressed is maintaining a constant standoff distance and precise travel path along a curved surface during multi-pass overlay welding. The system described integrates several key subsystems:
- Sensor subsystem: Infrared or capacitive sensors detect the surface contour of the head in real time, feeding positional data to the controller.
- Servo drive subsystem: Linear and rotary servo motors drive the welding torch along both the circumferential and meridional directions of the head.
- Interpolation controller: The CNC controller performs real-time coordinate transformation from the flat coordinate system of the weld program to the curved coordinate system of the head surface.
- Weld parameter control: Arc length, travel speed, and torch angle are modulated dynamically to compensate for the changing surface geometry.
The tracking algorithm relies on continuous surface profiling, where the sensor scans ahead of the torch tip and the controller adjusts the servo positions to maintain the programmed weld geometry. This approach is analogous to adaptive control in machining, but adapted for the high-heat-input environment of overlay welding.
Key Technical Parameters and Process Considerations
| Parameter | Typical Range | Purpose |
|---|---|---|
| Torch standoff distance | 3–5 mm | Arc stability and penetration control |
| Travel speed | 150–350 mm/min | Heat input and dilution management |
| Sensor sampling frequency | 100–500 Hz | Surface contour resolution |
| Servo positioning accuracy | ±0.1–0.3 mm | Weld bead alignment |
| Interpass temperature limit | Below 150°C (for stainless steel overlay) | Prevent sensitization and cracking |
The most challenging aspect of head cladding is the accumulation of geometric error as the weld progresses from the equator toward the pole, where the curvature radius decreases rapidly. The CNC system must compensate for this by adjusting the travel speed and torch angle continuously. Additionally, the number of passes required to achieve a specified overlay thickness (typically 3–5 mm for stainless steel on carbon steel) means that each subsequent pass must be precisely positioned on the previous one, demanding excellent repeatability from the tracking system.
Engineering Practice Implications
In modern pressure vessel fabrication, head cladding remains one of the most labor-intensive and quality-critical operations. The concepts presented in this paper—sensor-based adaptive tracking, real-time coordinate interpolation, and closed-loop torch positioning—are still the foundation of contemporary automated overlay welding systems. Today's implementations incorporate more sophisticated sensors such as laser triangulation and machine vision, but the fundamental architecture remains unchanged.
For engineers working on clad heads today, several lessons from this early work remain relevant:
- Weld sequence planning is critical: The starting point, direction of travel, and pass sequence must be optimized to minimize residual stress and distortion. A spiral pattern from pole to equator is generally preferred over circumferential passes.
- Preheating strategy must be uniform: Non-uniform preheat on a curved surface leads to differential thermal expansion and potential cracking. Induction preheating with programmable power distribution is now preferred over torch preheating.
- Post-weld inspection is non-negotiable: Every pass must be inspected for lack of fusion, undercut, and porosity before the next pass is deposited. Ultrasonic testing (UT) or magnetic particle testing (MT) is typically applied after every 2–3 passes.
Study Insights and Reflections
This paper is significant as an early attempt to bring manufacturing automation to a process that had been almost entirely manual in the Chinese pressure vessel industry. The approach of using real-time surface sensing with servo-driven torch positioning was ahead of its time, and the engineering challenges identified—coordinate transformation on curved surfaces, sensor reliability in high-temperature environments, and multi-pass alignment accuracy—are still active areas of research.
Reflecting on this work from a contemporary perspective, I note that the main limitation of the described system is the reliance on point sensors, which provide limited spatial resolution. Modern systems benefit from line sensors and 3D scanning, which provide full surface maps before welding begins. However, the fundamental challenge of maintaining weld quality on complex geometries remains unsolved in a fully automated sense, and human intervention for quality verification is still required in most production environments.
The paper also implicitly highlights a broader issue in cladding technology: the gap between process knowledge and automation capability. Experienced welders intuitively adjust their technique based on visual and auditory cues that are difficult to quantify and program. Bridging this gap requires not only better sensors and controllers but also a deeper understanding of the metallurgical and thermal phenomena governing weld quality.