Diagnosis of TIG Welding Process Based on Ultraviolet Radiation Monitoring

Literature Overview

This paper by Li Zhiyong, Gu Xiaoyan, and Wang Bao from the Welding Technology Research Center of the North University of China, published in 2009 in China Welding, presents a novel approach to online welding process monitoring using ultraviolet (UV) radiation from the TIG welding arc. The research, supported by the National Natural Science Foundation of China (Grant No. 50505048), investigates the UV spectral characteristics of the welding arc and demonstrates their application for real-time diagnosis of welding conditions and disturbance factors.

Core Technical Content

Arc Spectrum Analysis

The TIG welding arc is a rich source of electromagnetic radiation spanning from the infrared through visible to the ultraviolet spectrum. The paper focuses on the UV region (typically 200-400 nm), which is particularly sensitive to changes in arc conditions such as electrode composition, shielding gas purity, arc length, and current level.

The spectral radiation theory of arc light explains the UV emission through several mechanisms:

Atomic line emission: Excited atoms in the arc column emit discrete spectral lines at characteristic wavelengths. For argon, the prominent lines include Ar I at 696.5 nm and 750.4 nm in the visible region, but numerous lines exist in the UV region as well.
Molecular band emission: Molecular species formed in the arc, such as N₂ and OH radicals (from atmospheric contamination), produce band emission in the UV region.
Continuous emission: Free-free and bound-free transitions contribute a continuous background spectrum, particularly in the vacuum UV region below 200 nm.

UV Radiation Variation with Welding Parameters

The study systematically examines how UV radiation intensity varies with key welding parameters:

Parameter	Effect on UV Intensity	Mechanism
Welding current increase	UV intensity increases	Higher electron density, more excitation events
Arc length increase	UV intensity decreases	Arc column expansion, reduced emission density
Shielding gas purity decrease	UV intensity changes (specific lines increase)	Introduction of contaminant spectral lines
Electrode material change	UV spectral distribution shifts	Different electrode vaporization products
Travel speed increase	UV intensity decreases	Reduced interaction time per unit length

Disturbance Factor Analysis

The paper identifies and analyzes several disturbance factors that affect UV radiation monitoring:

Electrode wear: As the tungsten electrode erodes, its geometry changes, altering the arc shape and radiation characteristics.
Gas contamination: Atmospheric inclusions of oxygen, nitrogen, and water vapor introduce new spectral lines and modify the overall UV spectrum.
Weld pool conditions: Changes in weld pool composition, temperature, and surface conditions affect the radiation reflected from the pool surface.
Spatter and oxide formation: Oxide particles ejected from the weld pool scatter and absorb UV radiation, attenuating the signal.

Technical Implementation

UV Sensor Configuration

The diagnostic system employs a UV-sensitive photodetector positioned to capture radiation from the welding arc. The key design considerations include:

Spectral range: The sensor should be sensitive to the UV region (200-400 nm) while rejecting visible and infrared radiation to minimize interference from ambient light.
Spatial resolution: The sensor field of view should be focused on the arc region to maximize signal-to-noise ratio.
Temporal resolution: The detection frequency should be sufficient to capture transient arc behavior, typically requiring sampling rates above 1 kHz.

Signal Processing and Diagnosis

The integral intensity signal of UV radiation is used as the primary diagnostic parameter. The signal processing approach includes:

Baseline calibration: Establish reference UV intensity values for known-good welding conditions.
Real-time comparison: Continuously compare measured UV intensity with the reference baseline.
Deviation detection: Identify significant deviations that indicate process disturbances.
Classification: Categorize detected disturbances based on the nature and magnitude of the UV signal change.

Engineering Practice Application

Online Weld Quality Monitoring

The UV radiation monitoring approach has significant practical implications for welding quality assurance:

Early defect detection: Changes in UV radiation can indicate developing problems such as gas contamination, electrode wear, or arc instability before visible defects appear.
Process parameter verification: The UV signal can confirm that welding parameters are within the specified range, providing an additional layer of quality assurance beyond parameter setting.
Automated process control: The UV signal can be fed back to the welding power source to implement adaptive control, automatically adjusting parameters to maintain optimal welding conditions.

Integration with Welding Equipment

For practical implementation in pipe welding applications, the UV monitoring system should be integrated with:

Orbital welding machines: The UV sensor can be mounted on the welding head, providing continuous monitoring during circumferential welds on pipe joints.
Robotic welding cells: Integration with robotic systems enables automatic response to detected disturbances, such as adjusting travel speed or gas flow rate.
Welding procedure qualification: UV monitoring data can be recorded as part of the welding procedure qualification documentation, demonstrating process control and quality assurance.

FMEA Application to UV Monitoring System

Potential Failure Mode	Cause	Effect	Detection Method
False positive alarm	Ambient UV source (sunlight, arc from adjacent weld)	Unnecessary process interruption	Spatial filtering, signal threshold adjustment
Sensor degradation	UV window fouling, photodetector aging	Reduced sensitivity, missed detections	Periodic calibration, signal trend monitoring
Signal noise	Electrical interference from power source	Unreliable readings	Shielded cables, signal filtering
Electrode geometry change	Tungsten erosion during long weld	Gradual UV signal drift	Adaptive baseline, periodic recalibration

Study Insights and Implications

This research demonstrates the potential of optical monitoring as a powerful tool for welding process diagnosis. The UV radiation approach is particularly attractive because it is non-contact, non-intrusive, and provides real-time information about the arc conditions. Unlike conventional monitoring methods that measure electrical parameters (voltage, current), UV monitoring provides direct information about the physical state of the arc plasma.

The key insight from this work is that the UV spectral signal contains rich information about the welding process that can be extracted and utilized for quality assurance. The challenge lies in developing robust signal processing algorithms that can distinguish between meaningful process changes and noise or interference. Future work should focus on developing multi-parameter monitoring systems that combine UV radiation with electrical measurements, arc sound analysis, and thermal imaging for comprehensive process diagnosis.

For pipe welding applications, where weld quality is critical for structural integrity and leak tightness, the integration of UV monitoring into standard welding procedures could significantly improve quality assurance. The approach is particularly relevant for automated and robotic welding systems where human visual monitoring is not feasible.