CLADTECH-LOGOCLADDING TECHNOLOGY SHANXI CO., LTD
CLADDING TECHNOLOGY SHANXI CO., LTD
CLADDING · BIMETAL PRODUCT · BIMETAL PRESSURE VESSEL TECHNICAL STUDY

Visual Detection of Molten Pool Shape Parameters in Continuous Current TIG Butt Welding

Literature Overview

This 2005 publication by Liu Zhiyong et al. addresses the visual detection of molten pool shape parameters in continuous current TIG butt welding. The work, involving researchers from Shandong Yankuang Group, Shandong Zibo Chenyang Precision Casting, Shengli Oilfield, and Shandong University, represents an early but significant contribution to weld pool monitoring technology. The ability to detect and characterize the molten pool shape in real time is fundamental to closed-loop welding control, which is essential for achieving consistent weld quality in automated and robotic welding applications.

Core Technical Content

The molten pool is the heart of the welding process—its shape, size, and dynamics determine the weld geometry, microstructure, and defect susceptibility. In continuous current TIG welding, the molten pool parameters that can be visually detected include:

Parameter Definition Typical Range (mm) Measurement Method
Pool width Maximum lateral extent 3–8 Image processing
Pool length Maximum longitudinal extent 2–6 Image processing
Pool area Total projected area 8–40 Image processing
Pool depth Penetration depth 1–5 Indirect estimation
Pool eccentricity Deviation from centerline ±0.5–2.0 Image processing
Pool aspect ratio Length/width ratio 0.8–1.5 Image processing

The visual detection system typically employs a high-speed camera positioned at a specific angle to the weld pool, with image processing algorithms extracting the pool boundary from the captured frames. The key challenge is distinguishing the molten pool from the surrounding solid metal and slag, which requires sophisticated image segmentation techniques.

Image Processing Methodology

The visual detection system described in this work employs the following processing pipeline:

  1. Image acquisition: High-speed camera (50–200 fps) captures the weld pool region with appropriate lighting and filtering.
  2. Preprocessing: Noise reduction, contrast enhancement, and geometric correction to compensate for camera perspective.
  3. Segmentation: Thresholding or edge detection algorithms identify the pool boundary.
  4. Feature extraction: Geometric parameters (width, length, area, eccentricity) are calculated from the segmented boundary.
  5. Output: Parameters are transmitted to the welding control system for real-time adjustment.

Relationship Between Pool Parameters and Weld Quality

The molten pool shape parameters are directly correlated with weld quality:

Pool Parameter Weld Quality Impact Acceptable Range
Pool width Bead width, reinforcement 3–8 mm
Pool length Penetration depth, undercut 2–6 mm
Pool area Heat input, dilution 8–40 mm²
Pool eccentricity Weld centering, misalignment ±0.5 mm
Pool aspect ratio Weld geometry, porosity risk 0.8–1.5

An elongated pool (high aspect ratio) indicates excessive heat input and may lead to porosity, while an excessively wide pool suggests high current and increased dilution. Pool eccentricity is a direct indicator of torch misalignment, which can cause asymmetric weld profiles and potential lack of fusion on one side.

Engineering Applications and Standards Compliance

The visual detection of molten pool parameters has direct applications in pressure vessel and piping fabrication:

The following table summarizes the relationship between pool parameters and common weld defects:

Defect Pool Parameter Indicator Detection Threshold
Porosity Pool area > 40 mm², high aspect ratio Area > 40 mm²
Undercut Pool length > 6 mm, high eccentricity Length > 6 mm
Lack of fusion Pool width < 3 mm, low eccentricity Width < 3 mm
Excessive reinforcement Pool area > 35 mm² Area > 35 mm²
Centerline cracking Pool length > 5 mm, high aspect ratio Length > 5 mm

Integration with Automated Welding Systems

The visual detection system described in this work can be integrated into automated and robotic welding systems to create closed-loop control. The control logic follows a PDCA (Plan-Do-Check-Act) cycle:

  1. Plan: Define target pool parameters based on the welding procedure specification (WPS).
  2. Do: Execute the welding process with the defined parameters.
  3. Check: Monitor pool parameters in real time and compare with targets.
  4. Act: Adjust welding parameters (current, voltage, travel speed) to bring pool parameters back within acceptable ranges.

This closed-loop approach significantly improves weld consistency and reduces defect rates, which is particularly valuable for high-volume production of pressure vessels, heat exchangers, and piping systems where quality consistency is paramount.

Key Reflections

This 2005 work represents an important milestone in the development of visual weld pool monitoring technology. While the image processing techniques have advanced significantly since then, the fundamental concept—using visual information to characterize the molten pool and control the welding process—remains unchanged. For engineers in the cladding and bimetal field, this work highlights the importance of process monitoring as a quality assurance tool.

In cladding applications, where the interface between the overlay layer and base metal is critical, pool monitoring can be particularly valuable. The pool shape and dynamics directly influence the bonding quality at the interface, which determines the service performance of the cladded component. Engineers should recognize that pool monitoring is not merely a research tool but a practical quality assurance instrument that can be integrated into production welding systems.

The visual detection approach described in this work also has implications for the qualification and certification of welding procedures. As codes and standards evolve to incorporate real-time monitoring data, the ability to provide pool parameter records will become increasingly important for demonstrating process control and weld quality.