TOFD Inspection of Butt Welds Between Clad Cylindrical Shells and Nozzles

Literature Overview and Technical Challenge

The detection of volumetric defects in butt welds joining clad cylindrical shells to nozzles represents one of the most challenging non-destructive testing (NDT) problems in pressure vessel fabrication. The presence of the cladding layer introduces complex acoustic impedance mismatches, multiple reflection paths, and signal attenuation that can mask or mimic defect indications. This study focuses on the application of Time of Flight Diffraction (TOFD) techniques to inspect such welds, addressing the unique challenges posed by the bimetallic construction and proposing practical solutions for reliable defect detection and characterization.

Fundamental Principles of TOFD in Bimetallic Welds

TOFD relies on the diffraction of ultrasonic waves at the tips of planar defects such as cracks, lack of fusion, and planar inclusions. In a clad shell-to-nozzle butt weld, the ultrasonic signal must traverse through the cladding layer, the base metal, and the weld metal before reaching the inspection zone. This multi-layered acoustic path introduces several complications:

Challenge	Description	Impact on Inspection
Cladding layer attenuation	High attenuation in stainless steel or nickel-based cladding reduces signal amplitude	Reduced signal-to-noise ratio and decreased detection sensitivity
Interface reflections	Acoustic impedance mismatch at cladding-base metal interface creates clutter	Increased background noise and potential false indications
Beam refraction	Sound velocity differences between layers cause beam refraction at interfaces	Misplacement of defect indications and inaccurate depth measurement
Near-surface dead zone	Cladding layer thickness creates a region inaccessible to conventional TOFD probes	Defects near the cladding interface may be missed

The study recommends the use of high-frequency dual-element probes (typically 5 MHz to 10 MHz) with a center-to-center separation of 15 mm to 25 mm, depending on the cladding thickness and the expected defect size. The higher frequency provides better resolution for detecting small defects, while the wider probe separation accommodates the larger inspection volume required by the clad geometry.

Inspection Procedure and Technique Optimization

The inspection procedure developed in the study involves several critical steps that go beyond conventional TOFD practice for single-material welds:

Surface preparation: The cladding surface must be ground to a smooth finish (Ra less than 2.5 μm) to ensure consistent acoustic coupling and minimize surface clutter. Any surface irregularities from the cladding process must be removed before inspection.
Probe selection and calibration: Calibration should be performed on a reference block that replicates the actual geometry of the clad shell-to-nozzle weld, including the cladding thickness, base metal thickness, and weld reinforcement profile. The use of a through-thickness calibration block with known reflectors at different depths allows for accurate time-gain compensation (TGC) curve development.
Scan strategy: A dual-probe scan configuration with both probes on the same side of the weld is typically employed. For clad welds, the probes should be positioned to maximize the coverage of the weld root and the cladding interface, which are the most critical regions for defect detection. The scan should include both axial and transverse orientations to detect defects at various angles.
Signal interpretation: The study emphasizes the importance of distinguishing between true defect diffraction signals and interface reflections or clutter. The use of a gate to isolate the region of interest, combined with the characteristic hyperbolic shape of diffraction signals, aids in this discrimination. The amplitude of the diffraction signal should be evaluated against the relevant acceptance criteria specified in standards such as GB/T 24731 or EN ISO 17638.
Defect characterization: Once a defect indication is identified, its depth, length, and orientation should be determined. For planar defects such as cracks, the TOFD technique provides accurate depth measurement from the probe face, which can be converted to the actual defect location within the weld using the known geometry of the clad assembly.

Case Study and Practical Experience

The study presents a case study involving the inspection of a hydrogenation reactor shell-to-nozzle weld with a 304 stainless steel cladding layer on a 16MnR carbon steel base. The weld was fabricated using a submerged arc welding (SAW) process for the base metal and a gas metal arc welding (GMAW) process for the cladding layer. The TOFD inspection was performed using a 5 MHz dual-element probe with a 20 mm center-to-center separation and a 0° wedge angle.

The inspection revealed a lack of fusion defect at the weld root, approximately 2 mm from the cladding-base metal interface. The defect was confirmed by phased array ultrasonic testing (PAUT) and subsequently verified by radiographic testing (RT). The defect length was determined to be 12 mm, which exceeded the acceptance criteria specified in GB/T 150. The weld was rejected and repaired using a gouging and rewelding procedure, followed by re-inspection.

This case illustrates several important points for engineering practice:

TOFD is capable of detecting and characterizing defects in clad welds, but the technique requires careful setup and interpretation to avoid false negatives or false positives.
The cladding layer does not prevent effective TOFD inspection, but it does require adjustments to the standard procedure, including higher frequency probes, careful calibration, and skilled signal interpretation.
The combination of TOFD with other NDT methods such as PAUT and RT provides a more comprehensive defect detection capability and should be considered for critical pressure vessel welds.

Standards Compliance and Acceptance Criteria

The acceptance criteria for TOFD inspection of clad welds are typically derived from the relevant pressure vessel codes and standards. The following table summarizes the key requirements from several major standards:

Standard	Applicable Scope	TOFD Requirements	Acceptance Criteria Reference
GB/T 24731	TOFD testing of welded joints	Probe frequency, scan coverage, signal evaluation	GB/T 150 or specific product standards
EN ISO 17638	UT by TOFD of welds	Equipment calibration, scan technique, defect assessment	EN 1435 or product-specific criteria
ASME V Art. 5	UT of welds	General UT requirements applicable to TOFD	ASME VIII Div.1 or Div.2
NB/T 47013.10	TOFD for pressure equipment	Specific requirements for pressure vessel applications	NB/T 47002 or GB/T 150

Engineers should note that while TOFD is widely accepted for volumetric defect detection, it is not a substitute for surface inspection methods such as magnetic particle testing (MT) or penetrant testing (PT), which are required for detecting surface-breaking defects. A comprehensive inspection program should include both volumetric and surface NDT methods.

Study Insights and Recommendations

The study reinforces the conclusion that TOFD is a viable and effective technique for inspecting clad shell-to-nozzle butt welds, provided that the unique challenges of the bimetallic geometry are properly addressed. The key to successful inspection lies in thorough preparation, appropriate probe selection, careful calibration on representative reference blocks, and skilled signal interpretation. Engineers should invest in training and qualification of NDT personnel specifically for clad weld inspection, as the skills required differ significantly from those needed for single-material welds. Furthermore, the integration of TOFD with automated scanning systems and digital signal processing tools can enhance inspection efficiency and repeatability, making it a practical choice for high-volume production environments. The continued development of phased array techniques offers additional capabilities for complex geometries and should be considered as a complementary approach for critical applications.