Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
ISO 25239-5:2020 – FSW Inspection and Testing: Complete Quality Guide
Non-destructive and destructive testing for friction stir weld quality assurance
ISO 25239-5:2020 specifies inspection and testing requirements for friction stir welded joints in aluminium alloys. The standard covers quality classification, non-destructive testing (NDT) methods, destructive testing procedures, and acceptance criteria. Effective inspection is essential for FSW because many defects — such as kissing bonds and wormholes — are subsurface and not visible during production.
FSW presents unique inspection challenges. Unlike fusion welds where surface appearance often correlates with internal quality, FSW welds can appear visually perfect while containing significant subsurface defects. This makes systematic NDT selection and interpretation critical for quality assurance.
Quality Classification and Acceptance Levels
The standard defines three quality levels for FSW joints: Class D (general quality, suitable for non-structural applications where cosmetic appearance matters), Class C (intermediate quality, for general structural applications with moderate fatigue loading), and Class B (stringent quality, for safety-critical and high-fatigue applications). Each class has different acceptance criteria for both surface and internal defects.
For surface defects, Class B requires no cracks, no incomplete fill, maximum flash height of 0.5 mm, and maximum underfill of 0.2 mm. Class D permits flash up to 2 mm and underfill up to 0.6 mm. For internal defects, Class B limits void area to 3% of the weld cross-section, while Class D allows up to 10%. The standard also defines acceptance criteria for lack of penetration, kissing bonds, and oxide entrapment.
| Quality Class | Application | Max Surface Flaw | Max Internal Void Area | NDT Required |
|---|---|---|---|---|
| Class B (Stringent) | Aerospace, pressure vessels, safety-critical | 0.2 mm underfill, 0.5 mm flash | 3% | 100% NDT + macro |
| Class C (Intermediate) | Automotive, rail, general structural | 0.4 mm underfill, 1.0 mm flash | 6% | Sampling NDT + macro |
| Class D (General) | Non-structural, decorative, prototype | 0.6 mm underfill, 2.0 mm flash | 10% | Visual only |
Kissing bonds (also called “lazy S” or “joint line remnant”) are among the most dangerous FSW defects — they are nearly invisible in standard macro sections and radiography. If your application involves cyclic loading or corrosive environments, consider ultrasonic phased array inspection specifically calibrated to detect planar discontinuities at the nugget boundary.
Non-Destructive and Destructive Testing Methods
The standard specifies applicable NDT methods. Visual testing (VT) is mandatory for all quality classes and covers surface defects including flash, underfill, surface porosity, and discoloration. Ultrasonic testing (UT) using phased array is recommended for Class B and Class C welds to detect volumetric defects and planar discontinuities. Radiographic testing (RT) is effective for volumetric defects (voids, inclusions) but less sensitive to tight planar defects such as kissing bonds.
Destructive testing includes transverse tensile testing, bend testing (face bend, root bend, and side bend for thicker sections), and macro/microscopic examination. The standard specifies sampling locations — cross-sections must be taken at representative positions including the start, middle, and end of the weld, as these regions have different thermal histories and defect probability.
Develop a defect-specific inspection matrix for your FSW application. For example: visual examination for surface defects (every weld), phased array UT for volumetric defects (sampling plan), macro sections for structural verification (at WPS qualification and periodic intervals), and fatigue testing for design validation (prototype phase only). Matching inspection methods to probable defect types optimizes detection capability without excessive cost.
Frequently Asked Questions
Q: How does FSW inspection differ from fusion weld inspection?
A: FSW defects are fundamentally different — lack of penetration appears as a discontinuous bond line rather than a notch, wormholes are three-dimensional cavities caused by insufficient material flow, and kissing bonds are planar oxide remnants. Standard fusion weld acceptance criteria and NDT techniques must be adapted for FSW-specific defect types.
A: FSW defects are fundamentally different — lack of penetration appears as a discontinuous bond line rather than a notch, wormholes are three-dimensional cavities caused by insufficient material flow, and kissing bonds are planar oxide remnants. Standard fusion weld acceptance criteria and NDT techniques must be adapted for FSW-specific defect types.
Q: Is bend testing sufficient for FSW procedure qualification?
A: Bend testing is necessary but not sufficient. While bend tests confirm ductility and bond integrity, they may not reveal kissing bonds or fine oxide dispersions. Always combine bend testing with macroscopic examination at 10-50x magnification to assess the full weld zone structure.
A: Bend testing is necessary but not sufficient. While bend tests confirm ductility and bond integrity, they may not reveal kissing bonds or fine oxide dispersions. Always combine bend testing with macroscopic examination at 10-50x magnification to assess the full weld zone structure.
Q: What is the recommended sampling frequency for production NDT?
A: For Class B, 100% of production welds should receive UT or RT. For Class C, representative sampling (e.g., 1 in 20 welds or per production shift) is typical. The sampling frequency should be increased if process parameters drift, tool wear is detected, or quality trends indicate deterioration.
A: For Class B, 100% of production welds should receive UT or RT. For Class C, representative sampling (e.g., 1 in 20 welds or per production shift) is typical. The sampling frequency should be increased if process parameters drift, tool wear is detected, or quality trends indicate deterioration.
Q: Can FSW defects be repaired by re-welding?
A: Yes, but with limitations. Surface defects (flash, underfill) can often be re-machined. Internal defects may be repaired by welding a second pass over the same joint, but this requires requalification of the repair procedure. The repair WPS must specify the modified parameters (typically reduced traverse speed, increased axial force, or modified tool geometry).
A: Yes, but with limitations. Surface defects (flash, underfill) can often be re-machined. Internal defects may be repaired by welding a second pass over the same joint, but this requires requalification of the repair procedure. The repair WPS must specify the modified parameters (typically reduced traverse speed, increased axial force, or modified tool geometry).
