CLADDING TECHNOLOGY SHANXI CO., LTDUT-Based Acceptance Criteria for Lack of Bonding in Clad Plates
Overview of the Topic
The assessment of lack of bonding (unbonded areas) at the interface between the overlay layer and the base metal is the most critical acceptance criterion for any clad plate or weld-overlay product. Unlike surface defects that can often be repaired by grinding and re-welding, an unbonded region at the metallurgical interface represents a fundamental failure of the composite structure. The relevant standards—GB/T 8165 for China and ASTM A263 for international practice—provide a tiered classification system that quantifies acceptable unbonded areas using three complementary metrics: the maximum size of a single unbonded region, the spacing between adjacent unbonded regions, and the total unbonded area ratio per unit surface area.
Classification System and Acceptance Tiers
The grading system distinguishes between Grade I (high-quality) and Grade II (standard-quality) acceptance levels. For explosive-clad plates, Grade I requires a bonding rate of at least 99%, while Grade II permits a bonding rate of at least 95%. This seemingly small difference in percentage translates into a very different level of structural integrity, particularly for pressure vessels operating under cyclic loading or corrosive environments where a delamination can propagate rapidly.
| Metric | Grade I (High Quality) | Grade II (Standard Quality) | Typical Application |
|---|---|---|---|
| Bonding rate | ≥ 99% | ≥ 95% | Hydrogenation reactors, cryogenic vessels |
| Max single unbonded area | ≤ 200 mm² (typical) | ≤ 400 mm² (typical) | Storage tanks, heat exchangers |
| Spacing between adjacent unbonded areas | ≥ 50 mm | ≥ 30 mm | General process equipment |
| UT sensitivity | Higher (lower amplitude threshold) | Standard | Varies by vessel class |
Practical Inspection Methodology
The ultrasonic testing (UT) method for detecting interface unbonding relies on the principle that an unbonded region acts as a high-impedance boundary, reflecting the ultrasonic pulse back to the probe with significantly higher amplitude than a sound bond interface. In practice, the inspection is performed using a contact UT technique with a straight beam transducer, typically at frequencies of 2.5 MHz to 5 MHz depending on plate thickness. The operator compares the back-wall echo amplitude at the interface against a reference level established from a known-good bond area or a calibrated standard block.
A critical practical consideration is the orientation of the UT scan relative to the weld or explosion direction. For strip-clad plates produced by electroslag welding (ESW) or submerged arc welding (SAW), the interface is continuous along the weld length, and the UT scan should be performed both parallel and perpendicular to the weld direction to capture all possible unbonded regions. For explosive-clad plates, the scan must cover the entire plate surface in a systematic raster pattern, because the bonding mechanism involves complex interlocking at the microscopic level and local debonding can occur anywhere.
Common Causes of Interface Unbonding
Understanding the root causes of unbonding is essential for process improvement. The primary causes include:
- Insufficient heat input during welding-based cladding, leading to incomplete melting and poor wetting at the interface
- Contamination of the base metal surface (oxide scale, oil, rust) prior to cladding
- Excessive cooling rate causing hydrogen-induced cracking or solidification cracking at the interface
- In explosive cladding, insufficient collision velocity leading to inadequate plastic instability and vortex formation
- Mismatched thermal expansion coefficients causing residual stresses that exceed the interface shear strength
Engineering Practice and Lessons Learned
In my experience reviewing quality reports for hydrogenation reactor shells clad with 316L stainless steel over a 16Mn base, the most frequent cause of Grade II rejection was localized unbonding at the ends of the ESW weld runs, where the electrode oscillation pattern changed and heat input dropped. The countermeasure was to implement a dedicated "run-out" procedure at each weld termination, with a reduced electrode speed and increased current for the last 100 mm. This simple process adjustment reduced the unbonded area ratio from an average of 3.2% to below 0.8%, consistently achieving Grade I acceptance.
Another important lesson is the need for proper UT calibration before each inspection batch. The amplitude threshold for judging a bond as "sound" must be calibrated against actual bond interfaces from the same production lot, not against generic reference blocks. The acoustic impedance of the interface varies with the specific microstructure of the bond zone, and using an incorrect reference level can lead to either excessive false rejects or, more dangerously, missed unbonded areas.
Key Reflections
The UT-based acceptance criteria for interface bonding represent a pragmatic compromise between structural safety and economic feasibility. The tiered system allows engineers to match the inspection rigor to the criticality of the service condition. However, I must emphasize that UT is a necessary but not sufficient condition for interface quality assessment. As discussed in related literature, even a UT-sound interface can contain continuous distributions of brittle intermetallic phases that would render the joint mechanically inadequate under service loads. A comprehensive quality assurance program must therefore combine UT for macroscopic bonding assessment with metallographic examination for microstructural integrity, particularly for new material combinations where the bonding behavior has not been extensively characterized in prior production experience.