Phased Array Ultrasonic Testing of Overlay Weld Cracks in Hydrogenation Reactor Bosses

Literature Overview

This paper, authored by Chen Yongqiang, Xiao Xiong, Yu Jian, Wang Yulin, and Chen Liyi from Jinling Petrochemical Company's Pressure Vessel Inspection Center and affiliated testing organizations, was published in Nondestructive Testing (Volume 36, Issue 12, 2014, pages 12-14). It reports the successful application of Phased Array (PA) ultrasonic testing technology for detecting cracks in the overlay weld layer of hydrogenation reactor bosses (raised lugs or mounting bosses).

Core Technical Problem

Hydrogenation reactors in petrochemical plants operate under high temperature and high pressure conditions with hydrogen-rich environments, making them susceptible to hydrogen damage mechanisms such as hydrogen blistering, hydrogen cracking, and sulfide stress cracking. The overlay weld layers applied to reactor bosses for corrosion protection can develop cracks due to:

• Thermal cycling during startup and shutdown operations
• Residual stresses from the overlay welding process
• Hydrogen attack in the heat-affected zone and weld metal
• Fatigue from pressure cycling

Detecting these cracks is critical for ensuring the structural integrity and safety of the reactor. Conventional methods such as magnetic particle testing (MT) and dye penetrant testing (PT) can only detect surface-breaking cracks and cannot assess crack depth. Radiographic testing is impractical due to the geometry of the boss and the thickness of the overlay layer.

PA Ultrasonic Testing Methodology

The authors developed a PA ultrasonic testing procedure that achieved the following performance metrics:

Technical Implementation

The PA technique was implemented using a linear phased array probe that provided electronic beam steering and focusing. The key advantages over conventional ultrasonic testing include:

• Electronic beam focusing at specific depths within the overlay layer
• Beam steering to optimize the angle of incidence for different crack orientations
• Real-time A-scan, B-scan, and C-scan imaging for defect visualization
• Improved signal-to-noise ratio through electronic focusing

The scanning procedure involved coupling the linear array probe to the overlay surface using a water or gel couplant, with the probe rolling along the weld axis. The PA software was configured to produce B-scan images that showed the crack profile as a function of depth, allowing direct measurement of crack height.

Engineering Practice Integration

In my experience with hydrogenation reactor inspection, the overlay weld layers on bosses are among the most challenging areas to inspect reliably. The boss geometry creates complex stress concentrations, and the overlay layer itself may have varying thickness due to the welding process. The PA technique's ability to electronically focus at different depths is particularly valuable for this application, as it allows the inspector to optimize the inspection sensitivity for the specific thickness of the overlay layer at each location.

A practical consideration is the need for surface preparation. The overlay weld surface must be ground smooth to ensure consistent acoustic coupling. Any surface irregularities, such as weld ripples or grinding marks, can produce clutter signals that mask real defect indications. The authors' achievement of a 3 mm minimum detectable crack height on actual workpieces is impressive, especially considering the complex geometry of the boss.

Key Questions and Reflections

The reported minimum detectable crack height of 3 mm on actual workpieces raises the question of whether smaller cracks might be missed. In hydrogen service, even small cracks can grow rapidly under the combined action of hydrogen attack and cyclic loading. The acceptance criteria for such cracks must be established based on fitness-for-service assessment rather than a simple dimensional limit.

Another reflection concerns the orientation sensitivity of the PA technique. Linear phased array probes are most effective for detecting cracks that are perpendicular to the probe axis. Cracks running parallel to the probe axis may produce weaker diffraction signals. For a comprehensive inspection, the scan should be performed in at least two orthogonal directions, or a dual-probe approach should be used.

The paper also highlights the importance of calibration blocks. The test block with a 2 mm detectable crack height provides a benchmark for the technique's sensitivity, but the actual workpiece's different geometry and material properties result in a slightly reduced sensitivity. This difference must be accounted for when establishing acceptance criteria.

Study Insights and Implications

This paper demonstrates that PA ultrasonic testing is a powerful tool for detecting cracks in overlay weld layers on complex geometries such as hydrogenation reactor bosses. The key insight is that the electronic focusing and beam steering capabilities of PA technology overcome the limitations of conventional ultrasonic testing for this application. The technique provides not only detection but also sizing capability, which is essential for fitness-for-service assessment.

For engineers responsible for in-service inspection of hydrogenation reactors, this work provides a practical methodology for overlay weld inspection that can be integrated into regular inspection programs. The technique's ability to detect cracks as small as 3 mm with a measurement error of less than 1 mm makes it suitable for establishing crack growth monitoring programs.

The paper also underscores the importance of method validation. The authors' systematic approach of testing on both test blocks and actual workpieces, and reporting the performance differences, is a model for how NDT techniques should be qualified for specific applications. This level of rigor is essential for ensuring reliable inspection results in safety-critical applications.