Laser Vision Detection System for TIG Welding Rapid Manufacturing Research

Literature Overview

This 2007 publication in the journal Laser Technology, authored by Luo Yong, Zhang Hua, Li Yuehua, Xiao Min, and Xu Jianning from Jiangxi University of Science and Technology and Nanchang University, presents a laser vision detection system designed for real-time monitoring and feedback control of TIG welding in rapid manufacturing applications. The research addresses the critical challenge of ensuring dimensional accuracy and surface quality in additive manufacturing processes that employ TIG welding as the deposition method.

System Architecture and Principles

The laser vision detection system employs a structured light approach, projecting a laser line or pattern onto the weld surface and capturing the reflected image with a CCD or CMOS camera. By analyzing the deformation of the projected pattern due to surface topography, the system reconstructs the three-dimensional geometry of the weld bead in real time.

System Components

Component	Specification	Function
Laser source	635 nm, 5-10 mW, line pattern	Structured light projection
Camera	CCD/CMOS, 640×480 or higher	Image acquisition
Lens	8-25 mm focal length	Field of view control
Image processor	Real-time processing unit	Feature extraction
Controller	PLC or industrial PC	Feedback control
Mounting system	Vibration-isolated, adjustable	Stable alignment

Detection Capabilities

The system monitors several critical weld characteristics in real time:

Weld bead width: Extracted from the cross-sectional profile of the deposited layer, with accuracy of ±0.1 mm.
Weld bead height (reinforcement): Measured from the peak of the weld profile, with accuracy of ±0.05 mm.
Weld bead position: Tracked relative to the programmed path, with accuracy of ±0.1 mm.
Surface roughness: Estimated from the high-frequency components of the surface profile.
Defect detection: Identification of undercut, overlap, porosity, and lack of fusion from characteristic profile anomalies.

Process Control Strategy

The detection system feeds real-time geometric data to a controller that adjusts welding parameters to maintain the target bead dimensions:

Feedback Control Logic

Measured Parameter	Deviation	Control Action
Bead width too wide	> +0.5 mm	Reduce current or increase travel speed
Bead width too narrow	< -0.5 mm	Increase current or decrease travel speed
Bead height too high	> +0.3 mm	Reduce current or increase wire feed rate
Bead height too low	< -0.3 mm	Increase current or decrease wire feed rate
Position offset	> ±0.2 mm	Adjust torch position via servo motor

Control Algorithm

The system implements a PID (Proportional-Integral-Derivative) control algorithm:

Proportional term: Provides immediate correction proportional to the current deviation.
Integral term: Eliminates steady-state error by accumulating past deviations.
Derivative term: Anticipates future deviations based on the rate of change, preventing overshoot.

The control loop operates at a frequency of 10-50 Hz, sufficient to track the relatively slow dynamics of the TIG welding process.

Application to Overlay Welding and Cladding

The laser vision detection system has significant potential for overlay welding applications, particularly for:

Multi-pass overlay thickness control: Ensuring each pass deposits the target thickness (typically 1.5-3.0 mm) without excessive overlap or insufficient coverage.
Edge tracking for clad plate fabrication: Maintaining the overlay within specified boundaries to prevent excessive coverage of the base metal or insufficient coverage of the target area.
Surface quality monitoring: Detecting surface defects (porosity, lack of fusion) that would require rework before the next pass is deposited.
Dimensional accuracy for pressure vessel cladding: Ensuring the overlay layer meets the specified thickness uniformity requirements (typically ±0.5 mm across the vessel surface).

Integration with Quality Assurance

The detection system can be integrated into a comprehensive quality assurance framework:

Quality Requirement	Detection Method	Acceptance Criteria
Overlay thickness	Bead height measurement	Within ±0.5 mm of target
Surface continuity	Profile anomaly detection	No gaps or discontinuities
Bead width uniformity	Width measurement across length	Within ±10% of average
Edge alignment	Position tracking	Within ±1.0 mm of programmed path
Surface defects	High-frequency profile analysis	No porosity or lack of fusion indicators

Defect Detection Capabilities

The system can identify several weld defects from the surface profile:

Porosity: Appears as localized depressions in the weld surface, typically 0.1-0.5 mm in depth and 0.5-2.0 mm in diameter.
Lack of fusion: Manifests as irregular profile transitions between adjacent passes, with sharp depth changes exceeding 0.5 mm.
Undercut: Detected as a groove along the weld edge, with depth exceeding 0.1 mm.
Overlap: Identified as excessive bead width with insufficient fusion at the edges, characterized by a smooth but wide profile.
Spatter: Detected as discrete protrusions on the weld surface, typically 0.1-0.5 mm in height.

Study Insights

The laser vision detection system represents a significant advancement in process monitoring for TIG welding, particularly for applications requiring high dimensional accuracy and surface quality. For the cladding and bimetal pressure vessel industry, this technology enables closed-loop control of overlay thickness—a critical parameter that directly affects the corrosion resistance and service life of clad pressure vessels. The real-time feedback capability allows immediate correction of process deviations, reducing the need for post-weld inspection and rework. Engineers developing welding procedure specifications should consider integrating such monitoring systems into their quality assurance protocols, particularly for critical applications such as hydrogenation reactor cladding and nuclear pressure vessel overlay where weld quality directly impacts safety and reliability. The system's ability to detect defects during the welding process—rather than after completion—represents a paradigm shift from reactive to proactive quality control.