Zhuojin Pipe Fitting Co., LtdUltrasonic Detection of Interlayer Cracks in P91 Steel Pipe Butt Welds
Literature Overview
The paper by Wang Zhiyong, Li Shujun, Hao Xiaojun, Niu Xiaoguang, Wang Yongxin, and Wang Ying (2011), published in Nondestructive Testing (Vol. 33, No. 5, pp. 54-56), addresses a critical quality assurance challenge in power plant construction involving P91 steel pipe welds. P91 (ASTM A335 P91) is a 9Cr-1Mo-V-Nb ferritic-martensitic alloy widely used in supercritical and ultra-supercritical boiler components, including reheater headers, main steam lines, and balance-of-plant piping. The study focuses specifically on interlayer cracks—a welding defect that forms between weld passes—and demonstrates the limitations of conventional radiographic testing and the effectiveness of TOFD (Time of Flight Diffraction) as a supplementary detection method.
Defect Formation Mechanism
Interlayer cracks in P91 welds form due to the combined effects of high carbon content, low hydrogen diffusion rates in martensitic microstructures, and the specific thermal cycling of multi-pass welding. The defect characteristics are:
| Defect Characteristic | Typical Range |
|---|---|
| Crack length | 2-15 mm |
| Crack height | 0.1-1.5 mm |
| Crack orientation | Parallel to weld surface |
| Location | Between weld passes (interpass) |
| Root cause | Hydrogen embrittlement, restraint stress |
| Detection difficulty | High (low aspect ratio, surface-parallel) |
The martensitic microstructure of P91, with its high hardness (typically 280-320 HB in the HAZ), is particularly susceptible to hydrogen-induced cracking. The low diffusivity of hydrogen in the fine-grained martensite means that hydrogen atoms accumulate at interpass interfaces during multi-pass welding, creating localized tensile stresses that exceed the local fracture toughness.
Limitations of Conventional Detection Methods
The study demonstrates two fundamental limitations of standard NDT approaches for interlayer cracks:
- Radiographic Testing (RT): Interlayer cracks oriented parallel to the weld surface produce minimal density contrast on film or digital detectors. The thin crack width relative to the radiographic source-to-film distance results in negligible attenuation differences, leading to high probability of non-detection (PND).
- Conventional Pulse-Echo Ultrasonic Testing (UT): The reflection wave amplitude from interlayer cracks is significantly lower than from volumetric defects of equivalent area because the crack faces are not perpendicular to the incident beam. Standard UT acceptance criteria, calibrated for volumetric indications, either miss these cracks or generate excessive false indications.
TOFD Detection Methodology
Time of Flight Diffraction exploits the diffraction of ultrasonic waves at crack tips rather than reflection from crack faces. This approach is inherently insensitive to crack orientation and provides reliable detection of surface-parallel defects. The study proposes a three-step detection protocol:
| Step | Method | Purpose |
|---|---|---|
| 1 | Radiographic Testing (RT) | Detect volumetric defects and provide baseline |
| 2 | Conventional UT | Screen for indications and identify suspect areas |
| 3 | TOFD on suspect areas | Confirm and size interlayer cracks |
The TOFD technique uses a pair of 5 MHz transducers (one transmitting, one receiving) positioned on either side of the weld, with the crack tips generating diffracted signals that arrive at the receiver after the direct and back-wall signals. The time difference between diffracted signals from the top and bottom crack tips directly yields crack height, while the scan position gives crack location.
Engineering Practice and Quality Assurance
For P91 piping systems in power generation applications, the following quality assurance measures should be implemented:
- Preheat temperature of 200-250°C is mandatory to reduce hydrogen-induced cracking susceptibility.
- Interpass temperature control between 250-350°C prevents excessive cooling rates that promote martensite formation.
- Post-weld heat treatment (PWHT) at 740-770°C for a minimum of 1 hour per 25 mm of wall thickness is essential to relieve residual stresses and convert retained austenite.
- Hydrogen bake-out at 200°C immediately after welding, before PWHT, reduces trapped hydrogen levels.
- Mandatory TOFD supplementary examination for all suspect indications, not limited to random sampling.
The field dissection cases presented in the study provide valuable validation data. The actual crack dimensions recovered from sectioned welds correlated well with TOFD sizing estimates, confirming the technique's quantitative reliability for interlayer defect characterization.
Study Insights and Implications
This paper represents an important contribution to the NDT methodology for advanced alloy piping systems. The practical recommendation of combining RT with UT and TOFD for suspect areas provides a cost-effective and technically robust quality assurance approach. Engineers involved in power plant piping fabrication should recognize that P91 welds demand more rigorous inspection protocols than conventional carbon steel or low-alloy steel, particularly in applications where leak-tight integrity is critical. The interlayer crack detection challenge should be incorporated into weld procedure qualification records and inspector training programs to ensure consistent detection capability across projects.